(James Robertson, Claude Roux, Kenneth G. Wiggins) (B-Ok - Xyz)

Forensic Examination
of Fibres
International Forensic Science and Investigation Series
Series Editor: Max Houck
Forensic Examination of Hair

James Robertson
ISBN 9780748405671 · 1999
Forensic Examination of Fibres, 2nd edition

J. Robertson and M. Grieve
ISBN 9780748408160 · 1999
Forensic Examination of Glass and Paint: Analysis and Interpretation

B. Caddy
ISBN 9780748405794 · 2001
Forensic Speaker Identification

P. Rose
ISBN 9780415271827 · 2002
Bitemark Evidence
B. J. Dorion
ISBN 9780824754143 · 2004
The Practice of Crime Scene Investigation

J. Horswell
ISBN 9780748406098 · 2004
Fire Investigation
N. Nic Daéid
ISBN 9780415248914 · 2004
Forensic Computer Crime Investigation

Thomas A. Johnson
ISBN 9780824724351 · 2005
Analytical and Practical Aspects of Drug Testing in Hair

Pascal Kintz
ISBN 9780849364501 · 2006
Nonhuman DNA Typing: Theory and Casework Applications

Heather M. Coyle
ISBN 9780824725938 · 2007
Chemical Analysis of Firearms, Ammunition, and Gunshot Residue James Smyth Wallace
ISBN 9781420069662 · 2008
Forensic Science in Wildlife Investigations

Adrian Linacre
ISBN 9780849304101 · 2009
Scientific Method: Applications in Failure Investigation and Forensic Science Randall
K. Noon
ISBN 9781420092806 · 2009
Forensic Epidemiology
Steven A. Koehler and Peggy A. Brown
ISBN 9781420063271 · 2009
Ethics and the Practice of Forensic Science

Robin T. Bowen
ISBN 9781420088939 · 2009
Introduction to Data Analysis with R for Forensic Scientists

James Michael Curran
ISBN: 9781420088267 · 2010
Forensic Investigation of Explosions, Second Edition

A. Beveridge
ISBN 9781420087253 · 2011
Firearms, the Law, and Forensic Ballistics, Third Edition

Tom Warlow
ISBN 9781439818275 · 2011
The Neuroscience of Handwriting: Applications for Forensic Document Examination

Michael P. Caligiuri and Linton A. Mohammed
ISBN 9781439871409 · 2012
Forensic Metrology: Scientific Measurement and Inference for Lawyers, Judges, and
Criminalists
Ted Vosk and Ashley F. Emery
ISBN 9781439826195 · 2015
Fingerprints and Other Ridge Skin Impressions, Second Edition

C. Champod, C. Lennard, P. Margot, and M. Stoilovic
ISBN 9781498728935 · 2016
Forensic Examination of Fibres, Third Edition

James Robertson, Claude Roux, and Kenneth G. Wiggins
ISBN 9781439828649 · 2017
Forensic Examination
of Fibres
Third Edition
Edited by
James Robertson, Claude Roux and
Kenneth G Wiggins
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
© 2018 by Taylor & Francis Group, LLC

CRC Press is an imprint of Taylor & Francis Group, an Informa business
No claim to original U.S. Government works
Printed on acid-free paper
International Standard Book Number-13: 978-1-4398-2864-9 (Hardback)
This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been made to
publish reliable data and information, but the author and publisher cannot assume responsibility for the validity of all materials
or the consequences of their use. The authors and publishers have attempted to trace the copyright holders of all material
reproduced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained. If
any copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any
form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming,
and recording, or in any information storage or retrieval system, without written permission from the publishers.
For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.
copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-
8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that
have been granted a photocopy license by the CCC, a separate system of payment has been arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for
identification and explanation without intent to infringe.
Library of Congress Cataloging-in-Publication Data
Names: Robertson, J. (James), 1950- editor. | Roux, Claude (Forensic scientist) editor. | Wiggins, Kenneth
G., editor.
Title: Forensic examination of fibres / edited by James R. Robertson, Claude Roux, and Kenneth G. Wiggins.
Description: Third edition. | Boca Raton : CRC Press, [2018] | Includes bibliographical references and index.
Identifiers: LCCN 2017021297| ISBN 9781439828649 (hardback : alk. paper) | ISBN 9781315156583 (ebook)
Subjects: LCSH: Criminal investigation. | Textile fibers–Identification.
Classification: LCC HV8077.5.C6 F67 2018 | DDC 363.25/62–dc23
LC record available at https://lccn.loc.gov/2017021297
Visit the Taylor & Francis Web site at

http://www.taylorandfrancis.com
and the CRC Press Web site at

http://www.crcpress.com
Contents
Preface to second edition ix

Preface to the third edition xi
Editors xiii
Contributors xv
1 Fibres, Yarns and Fabrics: An Introduction to Production,
Structure and Properties 1
STANLEY MACARTHUR FERGUSSON AND JANE HEMMINGS
2 Textile Damage Interpretation 61

JANE HEMMINGS, DEBRA CARR AND JAMES ROBERTSON
3 Ropes and Cordages 89

KENNETH G WIGGINS
4 From Crime Scene to Laboratory 99

JAMES ROBERTSON AND CLAUDE ROUX
5 Microscopic Examination of Fibres 145

SAMUEL J PALENIK
6 Colour Analysis of Fibres 179

THOMAS W BIERMANN AND KENNETH G WIGGINS
7 Infrared Microspectroscopy of Fibres 245

K PAUL KIRKBRIDE
8 Raman Spectroscopy of Fibres 289

JANE HEMMINGS
9 Other Instrumental Approaches to Fibre Examination 309

CLAUDE ROUX, RACHEL MORISON AND PHILIP MAYNARD
10 Interpretation of Fibre Evidence 345

MICHAEL GRIEVE, CLAUDE ROUX, KENNETH G WIGGINS,
CHRISTOPHE CHAMPOD AND FRANCO TARONI
vii
viii Contents
11 Future Trends for Forensic Fibre Examination 427

KORNELIA NEHSE, SANDRA KOCH, MICHAEL GRIEVE, MAX M
HOUCK AND JAMES ROBERTSON
Index 455
Preface to second edition
By the time this volume is published, it will have been seven years since the first edition
of Forensic Examination of Fibres appeared in print. This new volume contains material
based on some of the content of that edition, and contributions from many new authors.
The subject material covered has been considerably expanded, and all material, at the very
minimum, has been revised and brought up to date. The past seven years have seen the
introduction of several new fibres into the marketplace, yet the fibre types likely to be seen
in the routine life of a forensic scientist remain relatively few. The content of this volume
reflects an increased understanding of the importance of the interpretation of fibre find-
ings and of the critical content-type of quality assurance in ensuring the reliability of the
technical observations and results.
It is our hope that this volume will remain relevant for another seven years. To our
contributing authors, our sincere and warm thanks for their efforts. To our various helpers
in bringing this project to fruition and to our Commissioning Editor, Dilys Alam, thanks
for your patience and understanding.
James Robertson
Canberra
Mike Grieve
Wiesbaden
ix
Preface to the third edition
In the preface to the second edition my co-editor, Mike Grieve, and I hoped the second
edition would remain relevant for another seven years. In fact, it will be closer to 17 years
between editions with this third edition.
Sadly, the passage of time has seen the sad loss of Mike Grieve and Ken Wiggins, both
giants in the forensic fibre world. The content of this new edition sees some new authors,
again reflecting the passage of time, but we very much want to preserve the contributions
by Mike and Ken for at least one more edition.
We hope that this edition continues to be of value to students and forensic practitio-
ners around the world whether new to the field or experienced forensic scientists.
This will be the final edition for one of us (JR) as it is time to hand on the baton to a
younger generation.
Finally, to our authors, your patience with us has been nothing short of remarkable! To
Taylor & Francis Group, once again, our thanks for not losing faith.
James Robertson
Canberra
Claude Roux
Sydney
xi
Editors
James Robertson of the National Centre for Forensic Studies (NCFS) at the University
of Canberra graduated with a BSc (Hons) in 1972 and PhD in 1976 from the University
of Glasgow. He lectured in forensic science at the University of Strathclyde from 1976 to
1985 before moving to Australia. After five years at the State Forensic Science Centre in
Adelaide, he completed twenty years as the head of the forensic group with the Australian
Federal Police (AFP), in the process developing a world class and respected forensic group
from modest beginnings. He returned to an academic role as a Professorial Fellow on his
retirement from the AFP in 2010. He is currently also the Director of the NCFS.
He is also an Honorary or Adjunct Professor at several Australian and overseas uni-
versities. He has published well over 100 scientific papers and book chapters, edited or
coedited four textbooks, is the Editor of the Australian Journal of Forensic Sciences and is
a member of the editorial boards of several journals.
He has held senior positions on all forensic advisory groups and organisations in
Australia, including being the current President of the Australian Academy of Forensic
Sciences (AAFS), Vice-President of the Australian and New Zealand Forensic Science
Society (ANZFSS), Chair of the Advisory Forum to the National Institute of Forensic
Sciences (NIFS), Chair of Standards Australia (SA) Forensic Committee and a past Chair
of the Senior Managers Australian and New Zealand Forensic Laboratories (SMANZFL).
His contributions to the forensic sciences and education have been recognised with the
awards of a Doctor of the University by the University of Canberra, the Australian Public
Service Medal (PSM) and a Member of the Order of Australia (AM).
James continues his active interest in, and passion for, the forensic aspects of trace
evidence, especially fibres and hairs.
Claude Roux completed his undergraduate and PhD studies in forensic science at the
University of Lausanne, Switzerland, then migrated to Australia in 1996, where he was
appointed as a lecturer at the University of Technology Sydney (UTS). His PhD thesis
focused on fibre evidence interpretation. Since then, Claude has been pivotal to the devel-
opment of forensic science in his adopted country over the past 21 years by developing and
leading the first undergraduate degree and PhD programs in forensic science in Australia.
In 2002, he became the founding Director of the UTS Centre for Forensic Science, being
Professor of Forensic Science since 2004. His professional motivation has been largely
driven by his vision of forensic science as a distinctive academic and holistic research-
based discipline.
Throughout his career, Claude has published more than 160 refereed papers and
25 book chapters and a large number of conference presentations. Although his research
activities cover a broad spectrum of forensic science across social and experimental sci-
ences, he remains active in the area of fibres, being a member of the European Fibres Group
xiii
xiv Editors
(now ENFSI Textile and Hair Group) since its inception in 1993. Claude is a member of the
editorial board of six scientific journals and of a number of working and advisory groups.
He is the immediate past President of the Australian and New Zealand Forensic Science
Society (ANZFSS), the current Vice-President of the Australian Academy of Forensic
Sciences and a Fellow of the Royal Society of New South Wales. Between 2014 and 2016,
he also served on the Scientific Advisory Board of the International Criminal Court, The
Hague, the Netherlands.
Kenneth G Wiggins joined the then Metropolitan Police Forensic Science Laboratory
(MPFSL) in London in 1970. Moving from the Biochemistry section, Ken specialised in
the forensic examination of fibres. With the merger of the MPFSL into the Forensic Science
Service (FSS) in 1996, he moved to the Analytical Services group before a move to the
Research and Development Division as an advisor and consultant dealing with fibres.
Ken published a number of key papers especially in the areas of fibre dye analysis, and he
was the author of the chapter on ropes and cordage in the second edition of the Forensic
Examination of Fibres which has been retained in this third edition. Ken formed a close
professional relationship with Mike Grieve that saw them work on a number of major stra-
tegic projects such as the establishment of the European Fibres Group (EFG), with Mike
and Ken taking on the executive roles in this group and developing it into one of the most
successful ENSFI (European Network of Forensic Institutes ) working groups. Ken was to
be a co-editor of this third edition, but sadly he passed away in 2009 from motor neurone
disease. As with the contributions to the second edition from Mike Grieve, the contribu-
tions from Ken have also been retained and updated, where appropriate, to recognise the
substantial contribution he made to the forensic examination of fibres over a forty-year
time span in forensic science.
Contributors
Thomas W Biermann is currently with the Forensic Science Institute in Wiesbaden,

Germany. Thomas graduated from the University of Frankfurt with a BSc in Biology in
1983. In 1988 he gained his PhD after completing research dealing with the impact of UV-B
radiation on lipid metabolism and on fatty acid patterns in synchronously growing marine
diatoms. Thomas then joined the Forensic Science Institute in the Bundeskriminalamt
and specialized in the examination of fibres and textiles. His special interests included the
development of software for databases and for the evaluation of fibre and textile frequen-
cies in clothing. From 1999 to 2006 Thomas was the head of the Fibres Section. He lectured
in Europe and the United States of America and he has been the author and co-author of
many publications concerning fibre and colour frequencies. He was awarded the PW Allen
Award of the Forensic Science Society (now The Chartered Society of Forensic Sciences)
for the most meritorious paper published in Science & Justice in 1997 and 2001 with Mike
Grieve and in 2007 as a sole author. In 2006, Thomas won the Michael C. Grieve Award for
the best presentation given on the Meeting of the European Fibres Group at Rome. Thomas
was a member of the steering committee of the European Fibres Group and together with
Ken Wiggins a member of the Scientific Working Group for Materials (SWGMAT) Fibers
Section in the United States. After almost 20 years of reporting in court and research deal-
ing with the evidential value of fibres and colours, since November 2006, he has been the
head of the Drugs Section of the Forensic Science Institute at the Bundeskriminalamt.
From 2013 on, Thomas has been the head of Biometrics Department and vice president of
the Forensic Science Institute at the Bundeskriminalamt.
Debra Carr has a BSc(Hons) in Materials Science and a PhD in Engineering. Debra is
a Reader in Impact and Armour at Cranfield University at The Defence Academy of the
UK. Debra is a Chartered Engineer, a Fellow of The Institute of Materials, Minerals and
Mining and a Professional Member of The Chartered Society of Forensic Science. In 2012,
Debra was a Winston Churchill Travelling Fellow in Forensic Textile Science. Debra has
published over 60 articles in peer-reviewed journals and over 50 conference papers. Debra's
research interests include: forensic textile science, body armour and helmet optimisation,
wound ballistics and clothing and textile sciences.
Christophe Champod received his MSc and PhD (summa cum laude), both in Forensic
Science, from the University of Lausanne, in 1990 and 1995, respectively. He remained in
academia until holding the position of assistant professor in forensic science. From 1999 to
2003, he led the Interpretation Research Group of the Forensic Science Service (UK), before
taking a full professorship position at the School of Criminal Justice (ESC) of the University
of Lausanne. He is in charge of education and research on identification methods and
maintains an activity as an expert witness in these areas. He is a Steering Committee mem-
ber for the International Fingerprint Research Group (IFRG), past member of SWGFAST
xv
xvi Contributors
(Scientific Working Group on Friction Analysis, Study and Technology) and, since 2014, an
invited member of Friction Ridge subcommittee (part of the Physics and Pattern Evidence
Scientific Area Committee) of the Organization for Scientific Area Committees (OSAC).
In 2015, he received the Distinguished ENFSI (European Network of Forensic science)
Scientist award for his contribution to forensic science. His research is devoted to the infer-
ential aspects associated with forensic identification techniques.
Stanley (“Mac”) MacArthur Fergusson graduated from Leeds University with a diploma
in Textile Industries in 1963. His early employment was as a dyer in both wool and syn-
thetic fibres in the United Kingdom. Upon his return to Australia in 1966, Mac worked for
Yorkshire Chemicals Pty Ltd as technical representative in both leather and textile dyeing,
covering Australia and the Asia Pacific regions. Mac also spent over 4 years in Indonesia
as CEO of PT Yorkshire Indonesia. He joined RMIT after a short time with the Melbourne
College of Textiles. In 1970, Mac was awarded chartered Textile Technologist (C.Tex.ATI)
from the Textile Institute Manchester. He is also a Chartered Colourist (C.Col ASDC)
from the Society of Dyers & Colourists, Bradford, Yorkshire. In 2009, Mac graduated with
a masters of technology (MTech) from RMIT University, where his research focused on
the effect of residual detergents on the light fastness of certain reactive dyes on cotton.
Currently, his major areas of research are in dye performance and the forensic aspects of
textile materials.
Michael Grieve. Sadly, this bio for Michael Grieve recognises the significant contribution
made by Mike to the advancement of forensic fibre examination over a close to 40-year
period before his untimely and sudden death in 2002. His contribution is more fully recog-
nised in the addendum to chapter 11, which deals with the future of forensic fibre examina-
tion. He is retained as an author in this third edition with his previous contributions being
updated where appropriate. Included in his many achievements, along with Ken Wiggins,
he established the European Fibres Group (EFG) that continues its work today as part of
the ENFSI (European Network of Forensic Institutes) Textile and Hair Group (ETHG).
Mike was the driving force behind the production of a Manual of Best Practice for the
Forensic Examination of Fibres. In 2003 he was awarded the ENSFI Distinguished Scientist
Award posthumously. Mike published the results of his research throughout his career
and unselfishly shared his knowledge. His only goal was to promote the best use of fibre
evidence. He was a friend, colleague and mentor to many, and it is a privilege to retain his
contribution to this book which he had such a strong influence in developing in the second
edition as co-editor.
Jane Hemmings is currently working in the Capability Development Office of the Australian
Federal Police and prior to this was a Senior Scientist with the Chemistry team in the
Forensics portfolio. Jane started work in the AFP Chemical Criminalistics team in 2005
after graduating with a PhD from the University of Technology Sydney, focusing on the use
of Raman spectroscopy in forensic fiber examinations. As a qualified fiber and textile dam-
age examiner, Jane has been involved in major homicide investigations that have required
scene attendance, laboratory examination and the provision of expert evidence. Following a
homicide in 2005, Jane reinvigorated textile damage examinations in the AFP and in 2008
established the Australian and New Zealand Textile Damage Working Group bringing
together experts from across jurisdictions to advance the evidence class in the region.
Contributors xvii
Max M Houck is an international forensic expert with over 25 years of experience.

Houck has experience in the private sector, academia, local government, and worked at
the Federal Bureau of Investigation Laboratory Division. He has worked as a forensic
anthropologist, a trace evidence analyst, a researcher, and has managed millions of dol-
lars in grants and awards. Most recently, he was the inaugural Director of the Department
of Forensic Sciences in Washington, D.C., overseeing 150 employees and managing the
forensic science laboratory, the public health laboratory, and crime scene sciences for
the nation’s capital. Max has worked on a number of mass casualty scenes, including
the Branch Davidian Investigation and the September 11, 2001 attack on the Pentagon. He
is one of the most published forensic experts in the world. He has dozens of peer-reviewed
journal articles and is the author and editor of numerous books. He is co-author of the
best-selling Fundamentals of Forensic Science, Science of Crime Scenes, and Success with
Expert Testimony, among others. He is the editor of the Advanced Forensic Science series
of books. Houck is also founding co-editor of Forensic Science Policy and Management
(the official journal of ASCLD), the only journal that addresses the management, policy,
and administration of forensic science.
K Paul Kirkbride is currently Professor of Forensic Science at Flinders University in

Adelaide South Australia. Before entering academia in 2013 he was Chief Scientist with
the Forensic and Data Centres portfolio of the Australian Federal Police for five years, and
prior to that, he was Manager of Business Programs at the National Institute of Forensic
Science in Melbourne. For the other 20 years of his time as a forensic scientist, Paul worked
with Forensic Science SA in Adelaide South Australia. With that organization he was a
trace evidence examiner, illicit drugs analyst, clandestine drug laboratory investigator
and, eventually, Assistant Director – Science.
He attributes any success he might have in regards to handling textile fibres under
various microscopes to a long and devoted attachment to trout fly-tying.
Sandra Koch has worked in the field of forensic trace evidence analysis for over 16 years.
She holds a BA in Anthropology and Political Science from University of Michigan, Ann
Arbor, an MFS in Forensic Science from George Washington University, an MS in Forensic
Anthropology from the Boston University School of Medicine. From 1997 until 2013,
she worked for the FBI Laboratory Trace Evidence Unit. During that time she analyzed
hairs, fibers, fabric damage, and feathers using a range of instrumentation including light
microscopy, polarized light microscopy, fluorescence microscopy, microspectrophotom-
etry and infrared spectroscopy. She has taught numerous workshops and short courses on
hairs, fibers, crime scene evidence collection and other specialized trace evidence analysis.
She is a Fellow of the American Academy of Forensic Sciences, a founding member of the
American Society of Trace Evidence Examiners and a Fellow of the American Board of
Criminalistics (Hairs and Fibers). Sandy has been a member of the Scientific Working
Group for Materials Analysis (SWGMAT) since 1999, served as the chair of the fiber sub-
group from 2009 to 2013, and is currently a member of the OSAC Materials (trace) sub-
group. She has published in the areas of forensic analysis of hairs, fibers, fabric damage,
and feather identification. Currently Sandy is pursuing a PhD in biological anthropology
at the Pennsylvania State University, drawing on her background in microscopical analy-
sis to integrate forensic and anthropological analysis of hair and has research interests in
archaeological textiles.
xviii Contributors
Philip Maynard is a senior lecturer with the Centre for Forensic Science at the University
of Technology Sydney. He teaches the Crime Scene Investigation and Fire and Explosives
Investigation subjects at UTS, as well as supporting the other forensic subjects in the foren-
sic undergraduate programme. Phil graduated from the University of Sydney in 1992 with
a PhD in chemistry, and entered the forensic world in 1995 when he was employed in the
Physical Evidence Section at the Division of Analytical Laboratories (FASS). His casework
experience includes flammable liquids, explosives, paint, fibres and glass analysis. Phil
joined the forensic science programme at UTS in 1998 and has published peer-reviewed
research in the fields of fibre evidence, glass evidence, latent fingermark detection, foot-
wear impressions and explosives analysis.
Rachel Morison of the Centre for Forensic Science is a member of the Faculty of Science
staff at the University of Technology Sydney, where she is currently a PhD candidate inves-
tigating the detection and identification of acquired characteristics on fibres. She com-
pleted a Bachelor of Forensic Science (Honours) in Applied Chemistry in 2012 with her
Honours project looking at the effect and identification of unknown chemicals in cases of
textile damage.
Kornelia Nehse has been working in the field of forensics for more than 25 years. She
received her diploma in Textile and Clothing Engineering in 1986 from the University
of Applied Sciences in Moenchengladbach, Germany. After teaching in the field of tex-
tile technology she moved to forensic science. She is currently the head of the Textiles
and Micromorphology department of the Forensic Science Institute (LKA KT) in Berlin,
Germany and furthermore the permanent representative within ENFSI (European
Network of Forensic Science Institutes) for the Forensic Science Institute in Berlin. She is
also the current chairperson of the European Textile and Hair Group (ETHG), a position
she has held since 2003.
She is a founding member of the European Fibres Group (EFG), now European
Textile and Hair Group (ETHG) and member of the Scientific Working Group Material
(SWGMAT) in the United States. She recently has also been involved in European projects,
e.g. the set up of a Reference Data Base and Fibre Type Information System (FTIS) for the
identification of textile fibres.
Kornelia has lectured at or been affiliated with numerous universities in the fields of
forensic science and textile technology including the HTW University of Applied Sciences
and the Berlin University of the Arts (Berlin, Germany), The Brandenburg University of
Technology (Cottbus, Germany), and the University of Parma (Parma, Italy). She is also an
internationally recognized and invited speaker (e.g. EFG, ETHG, ENFSI, EAFS, SWGMAT,
Trace Evidence, ANZFSS, IAFS) and has taught in numerous workshops and short courses
from crime scene to court.
Samuel J (“Skip”) Palenik has had a lifelong fascination with the microscope that started
when he received his first instrument at the age of eight. Since then he has devoted himself
to increasing his knowledge of analytical microscopy and microchemistry and applying
it to the solution of real world problems, especially those of forensic interest. He was for-
tunate in having worked closely with his mentor, Dr Walter McCrone, for over 30 years
and to have studied forensic microscopy with Dr Max Frei-Sulzer of Zurich, a disciple
of Dr Edmond Locard of Lyon. Skip has been teaching analytical microscopy to forensic
Contributors xix
scientists for more than 40 years and has published numerous scientific articles and book
chapters on the applications of chemical and forensic microscopy. His most recent contri-
bution is a chapter on the use of heavy minerals in forensic science published by Elsevier.
He has also played a significant content-type in numerous criminal investigations includ-
ing the Atlanta child murders, the Air India Bombing, the Jon Benet Ramsey case, the
Narita Airport bombing (Tokyo), the Hillside Strangler (Los Angeles) case, the Oklahoma
City bombing, the Ivan the Terrible (Jerusalem) case, the assassination of Dr Martin Luther
King (reinvestigation by U.S. House Select Committee on Assassinations), the Unabomber
case, the disappearance of Helen Brach, the ‘Kiki’ Camarena murder case and the Green
River serial murders. He established Microtrace in 1992 to provide a resource for organi-
zations and individuals in need of scientific services involving the analysis of microscopic
trace evidence. His special research interests are the identification of single small particles,
small amounts of complete unknowns and tracing dust and soil back to their origins. He is
the 2009 recipient of the Paul L Kirk Award, the highest award given by the criminalistics
section of the American Academy of Forensic Sciences, the 2013 Ernst Abbe Award for
outstanding contributions to microscopy, the 2013 Edmond Locard Award presented by
the American Society of Trace Evidence Examiners, and the Chamot Medal in chemical
microscopy in 2010. He is listed in American Men and Women of Science.
Franco Taroni is full professor of forensic statistics at the School of Criminal Justice of
the University of Lausanne. He received his PhD in forensic science in 1996 from the
University of Lausanne. He was awarded two European Community Grants for Training
and Mobility of Researchers for a collaboration with Professor Aitken of the Department
of Mathematics and Statistics of the University of Edinburgh. He spent four years as a
research project manager at the Institutes of Forensic Medicine of the Universities of
Lausanne and Zürich (Switzerland). Franco’s main areas of research, casework, and teach-
ing are forensic statistics, Bayesian networks, and Bayesian decision-theoretic data analy-
ses. He has authored and coauthored several books on these topics (i.e. Statistics and the
Evaluation of Evidence for Forensic Scientists (2004), Bayesian Networks and Probabilistic
Inference in Forensic Science (2006), Data Analysis in Forensic Science: A Bayesian Decision
Perspective (2010) and Bayesian Networks for Probabilistic Inference and Decision Analysis
in Forensic Science (2014). He is associate editor of the journal Law, Probability and Risk
(Oxford University Press).
Fibres, Yarns and Fabrics:
An Introduction to Production,
Structure and Properties 1
STANLEY MACARTHUR
FERGUSSON JANE HEMMINGS
Contents
1.1 Introduction 2
1.2 Fibres 2
1.2.1 Classification of Textile Fibres 3
1.2.2 Fibre Production 4
1.2.3 Fibre Characteristics 4
1.2.4 Properties of Fibre Forming Polymers 5
1.3 Natural Fibres – Vegetable Fibres 5
1.3.1 Seed Fibres – Cotton, Kapok and Coir 8
1.3.2 Stem Fibres – Flax, Jute, Ramie, Hemp, Kenaf 12
1.3.2.1 Flax 12
1.3.2.2 Jute 13
1.3.2.3 Ramie 13
1.3.2.4 Hemp 14
1.3.3 Leaf fibres – Sisal, Hemp (Manila and New Zealand) 14
1.3.3.1 Hemp-Abaca (or Manila Hemp) and New Zealand Hemp 14
1.3.3.2 Sisal and Henequen 15
1.4 Natural Fibres – Animal Fibres 15
1.4.1 Wool 16
1.4.1.1 Hair fibres or Speciality Wools 17
1.4.2 Silk—Silkworm 19
1.4.2.1 Spider Silk 20
1.5 Mineral Fibres (Asbestos) 20
1.6 Man-made Fibres 20
1.6.1 Fibres Regenerated from Natural Polymers 21
1.6.1.1 Viscose Rayon 21
1.6.1.2 Bamboo 23
1.6.1.3 Improved Forms of Viscose 24
1.6.1.4 Esters of Cellulose 25
1.6.1.5 Regenerated protein fibres 26
1.6.1.6 Alginate Fibres 26
1.6.2 Synthetic Fibres 26
1.6.2.1 Polyesters 28
1.6.2.2 Polyamides 30
1.6.2.3 Polyolefins 32
1
2 Forensic Examination of Fibres
1.6.2.4 Polyvinyl Fibres 34

1.6.2.5 Polyurethane Fibres 35
1.6.2.6 Aramides 36
1.6.2.7 Polyvinyl Alcohol Fibres 37
1.6.2.8 Inorganic Man-Made Fibres 37
1.7 Bicomponent Fibres 38
1.8 New Fibres 39
1.9 From Fibre to Fabric 39
1.9.1 Introduction 39
1.9.2 Yarns 40
1.9.2.1 Types of Yarns 41
1.9.3 Fabrics 41
1.9.3.1 Woven Fabrics 42
1.9.3.2 Knitted Fabrics 47
1.9.3.3 Pilling 52
1.9.4 Non-Woven Materials 53
1.9.4.1 Leather 55
1.10 Conclusions 57
Acknowledgments 58
Bibliography 58
Further Readings 58
Textbooks 58
1.1 Introduction
If forensic fibre examiners are to fully exploit fibre and textile evidence during analysis
they require not only specialised forensic knowledge but also knowledge of fibres, yarns
and fabrics. Production, structure (both chemical and physical) and the properties of these
materials is required in order to assess the value of fibre evidence. It is necessary to also
know production figures, fashion changes, sudden arrivals of new material/s, dye variabil-
ity and other factors that may have a bearing on the information obtained.
What follows is a summary of fibres and the textiles that can be constructed from
them. Information can also be obtained from catalogues, journals and texts, examples of
which are listed at the end of this chapter.
1.2 Fibres
The fundamental use of textiles by ancient man was for the purpose of protection and the
initial textile used was the animal skin. As man became more sophisticated textiles became
a way of showing status in wealth and society. Rapid population growth since World War
II and consumerism has seen a rapid expansion of textile products from synthetic sources.
Without synthetic textile products world production of natural fibres would be unable to
satisfy world consumer demand for fibres and their products.
The common misconception is that textiles relate in the main to apparel and home
furnishings. There are however many other applications. These applications can be in the
Fibres, Yarns and Fabrics 3
Table 1.1 Use of Textiles in Various Industries

Industry Uses
Medical Bandages, swabs, sutures, absorbent cloths
Transport Inflatables, ropes, canvas, sails
Mining Conveyor belts, flexible ducting, sacks
Sports Windproof fabrics, playing surfaces, Equipment
Furniture Upholstery, netting, bedding, wadding
Horticulture Netting, shade cloths, hail protection, Twine
Clothing Sewing threads, hats, garments, shoes
Water Hoses, membranes, filters
Chemical Filter fabrics, vat liners, overalls
Architecture Awnings, carpets, wall coverings, drapes
Aerospace Glider fabric, parachutes, composites
Paper Making Filtration aids, conveyors, felts
Automotive Deat belts, drive belts, interior trims, insulaion
Civil Engineering Road construction, retaining walls, soil separation,
slings
Agriculture Animal coats & collars, sacks, flexible silos
Defence Camouflage, clothing, ballistic vests
industrial, medical, aerospace and automotive industries. Fibre types and processing routes
vary depending on the industry of application. Table 1.1 lists examples of the wide variety
of applications for textile products in the ever-changing world of industry.
1.2.1 Classification of Textile Fibres

The basic component of a textile is a fibre. At the basic level all fibres fall into one of two
categories: natural fibres and man-made fibres. Natural fibres fall into three chemical
classes:
• Cellulose based fibres (from the seeds, stem and leaf of plants)
• Protein based fibres (from the hair, wool or silk of animals)
• Mineral based fibres (the only naturally occurring mineral fibre is asbestos the use
of which is banned in many countries)
Man-made fibres are manufactured and fall into three classes:
• Regenerated fibres (those formed from naturally occurring fibre-forming poly-

mers, such as viscose from cellulose)
• Synthetic fibres (those formed from non-renewable sources, such as polyester)
• Inorganic fibres (those formed from inorganic materials, such as carbon and glass)
Classification of fibres at this broad level is summarised in Figure 1.1. A more detailed
breakdown within each class will be given as each is considered throughout this chapter.
Fibres
Natural Man-made
Animal Vegetable Organic Inorganic

e.g. e.g. e.g.
Wool Cotton Carbon
Glass
Transformation of natural Synthetic polymers

e.g. e.g.
Acetate Acrylic
Lyocell Aramid
Modal Modacrylic
Figure 1.1 Broad classification of natural and man-made fibres.
1.2.2 Fibre Production

Originally all textiles were composed of natural fibres, such as cotton, wool, silk, flax and
jute. With the increasing demand for fibres that could perform a variety of functions, the
invention and production of man-made fibres emerged. The first man-made fibre intro-
duced in 1911 was rayon, which was a transformation of the naturally occurring poly-
mer cellulose. The creation of synthetic polymers in 1930 made possible the production of
organic synthetic polymer fibres. The first man-made synthetic polymer fibre was nylon in
1939. The development of new fibres continues even today (see Chapter 11) as advancing
technical innovations expand fibre and fabric manufacturing processes. These innovations
have resulted in more efficient manufacturing processes giving rise to a wider variety of
products at lower cost to the consumer.
As demand has shaped the quality of fibres manufactured so does demand influence
the quantity of fibres produced each year. Global textile fibre production for the years
2000 and 2008 is illustrated in the Table 1.2. These figures show the very large increase in
demand for polyester fibre, with relatively stable cotton production. Other natural fibres
such as wool, silk and flax show little or no growth and static production. Polypropylene
production and use, on the other hand, is growing, mainly due to its less complex manu-
facture compared to nylon and polyester.
1.2.3 Fibre Characteristics

Fibres used in textiles are fine and thin; the Textile Institute defines a fibre as:
Textile raw material, generally characterised by flexibility, fineness and high ratio of
length to thickness*
Natural fibres, with the exception of silk, exist as staple fibres† with lengths ranging
from 2 to 50 cm in length and cross sections ranging from 10 to 40 microns. The limited
length of staple fibres means they need to be spun to be converted to a yarn.
Synthetic fibres (and silks) are produced as continuous filament yarns. These yarns are
sometimes cut to form staple yarns of specified lengths depending on the end use of the fibre.
* Textile Terms and Definitions, 11th edition, Textile Institute.

† A fibre of relatively short length.
Table 1.2 Global Textile Fibre

Production, for Major Fibre Types,
for the Years 2000 and 2008
Global Textile Fibre Production
(Million Tons)
– 2000 2008
Cotton 19.7 24.4
Polyester 18.9 30.7
Polyamide 4.1 3.5
Polypropylene 6.0 5.9
Acrylics 2.7 1.9
Wool 1.3 1.2
Silk 0.1 0.2
Total 56.9 67.8
Source: Koslowski, H.-J. and Kassel, F. M.,
Fasen für Vliesstoffindustrie aktuelle
Markttrends, 2008; American Fiber
Manufacturers Association, Fibre
Organon, 80, 95–112, 2009.
The success of a fibre within a textile is significantly influenced by the polymer com-
prising the fibre.
1.2.4 Properties of Fibre Forming Polymers

Fibre forming polymers, either natural or synthetic, are a series of polymer chains held
together by strong bonding with the lengths being several hundred times greater than the
width. These polymers must also possess other general traits to make them useful as a fibre
for textile purposes such as:
1. A degree of crystallinity of the molecular chains to give longitudinal strength

along the fibre axis
2. Molecular chain flexibility to give extensibility and drape, particularly in apparel
fabrics
3. High molecular weight giving both a high melting point and low solubility in most
solvents
Figure 1.2 shows the internal molecular structure of a man-made fibre illustrating the
amorphous and crystalline regions. This polymer has all the features listed above, thereby
making it suitable as a textile fibre.
1.3 Natural Fibres – Vegetable Fibres
Vegetable fibres are divided into three groups based on the section of the plant from which
they are produced (refer to Figure 1.3).
The chemical composition of vegetable fibres can vary dramatically, however the
principle component in all the vegetable fibres is cellulose. Depending on the fibre,
other constituents present consist of: hemi-cellulose; water soluble pectins; lignin as well
as fats and waxes (refer to Table 1.3 for typical ratios of fibre constituents for some vegetable
fibres).
Crystalline
region
Amorphous
region
Figure 1.2 Morphology of a synthetic fibre. (Adapted from East, A.J., In J. E. McIntyre (Eds.),
Synthetic Fibres: Nylon, Polyester, Acrylic, Polyolefin, Cambridge, Woodhead Publishing and
CRC Press, 2005.)
Fibres
Natural Man-made
Seed Stem (bast) Leaf

Cotton Flax
Kapok Jute
Coir Ramie
Hemo
Figure 1.3 Sub-classification of vegetable fibres.
Table 1.3 Approximate (%) Chemical Composition of Cellulosic Fibres

Fibre Cellulose Hemicellulose Pectin Lignin Fat & Wax
Cotton 92–95 5.7 1.2 0 0.6
Flax 62–71 16–18 1.8–2.0 2.0–2.5 1.5
Hemp 67–75 16–18 0.8 2.9–3.3 0.7
Ramie 68–76 13–14 1.9–2.1 0.6–0.7 0.3
Jute 59–71 12–13 0.2–4.4 11.8–12.9 0.5
Sisal 66–73 12–13 0.8 9.9 0.3
Abaca 63–68 19–20 0.5 5.1–5.5 0.2
Coir 36–43 0.2 3–4 41–45
OH
OH
O HO O
HO O O
OH
OH n
Figure 1.4 Repeating unit of glucose (units joined between the 1 and 4 carbon atoms).
Cellulose is the carbohydrate part of the plant’s cell wall and is a polysaccharide macro
molecule consisting of a large number of anhydroglucose units. The number of 1,4-β–
glucose units in the cellulose chain can vary from several hundred to several thousand
depending on the degree of hydrolysis that the cellulose undergoes. The repeating unit is
illustrated in Figure 1.4.
The molecular chains in natural cellulose vary in length. Variations in chain length
results in differing solubilities in alkali and different viscosities but the same chemical
properties. The degree of polymerisation (DP) is a measure of the chain length. Bast fibres
have a DP of about 9000; wood cellulose on the other hand has a DP between 7,500 and
10,500.
Hemi-cellulose is the name applied to a group of polysaccharides found in the cell
walls of plants.
The hemi-cellulose found in flax differs in constitution to that found in jute, which
is again different to that found in hemp. Whilst hemi-cellulose is similar in structure to
cellulose, the typical hemi-cellulose, xylose, does not contain the projecting CH2OH group
found in cellulose. Because hemi-cellulose has a much shorter chain length (DP of about
120) it is readily soluble in dilute solutions of alkali.
Lignin is an amorphous polymeric material concentrated in the spaces between the
cells. Chemically, lignin differs from other natural polymers in being aromatic; destruc-
tive distillation yields derivatives of phenyl propane. Lignified tissues are stronger and are
more resistant to retting*. The lignin in plant cell walls can be almost completely removed
by treatment with chlorine compounds such as sodium hypochlorite or sodium chlorite
used in the bleaching process. The presence of large amounts of lignin in cellulosic fibres
can result in substantial changes to colour on exposure to light. Jute, for example, a high
lignin content fibre, will yellow when exposed to sunlight.
Pectins are a complex group of substances of which the principle constituent is
polygalaturonic acid in which a large proportion of the carboxyl groups have been meth-
ylated. The pectins do not contribute to the overall strength of the fibre, as once removed
by boiling in 1.0% solution of sodium hydroxide for 30 minutes in the absence of air,
the strength of the fibre and its fluidity † remain unchanged. In mature cells, more rigid
bonding between the cells is required and insoluble salts of calcium and magnesium
pectinates occur.
* Retting is the bacterial decomposition of lignan and hemicellulose.

† Fluidity – A measure of the damage to cellulose.
1.3.1 Seed Fibres – Cotton, Kapok and Coir

Of all the natural fibres used in textiles, cotton is the most important with 24.4 million
tonnes produced in 2008. It is also the most commonly encountered natural fibre in foren-
sic examinations.
Cotton is a natural vegetable fibre from the seedpod of plants in the genus Gossypium
and is grown in subtropical climates. Two species are commonly cultivated; G. hirsutum,
or upland cotton, accounts for 94% of world production and, G. barbadense, or Egyptian
or Pima type cotton, which accounts for only 3%–4% of world production (Gordon, 2009).
Cotton fibres exhibit considerable variability in their chemical and physical properties
depending on the country of origin and the conditions under which they are cultivated.
Variability can be evident in fibres even from the same seed-pod. Hence consistency in
fibre sourcing is required to ensure minimal changes in processing characteristics between
production lots. This presents a major challenge to the textile manufacturing industry.
Figure 1.5 shows the major features in the cross-section of the cotton fibre.
The cellulose content of cotton varies from a low of 88% to a high of about 96%. In the
natural harvested state, the average cotton fibre is approximately 90% pure cellulose. This
percentage rises to about 96% after industrial scouring and bleaching. Scouring removes
naturally occurring waxes, salts and pectins present in the fibre. Bleaching gives the fibre a
brighter, cleaner colour and serves to reduce imperfections in fibres from different sources.
Depending on the method of yarn manufacture, cotton seed or husk may be present in
the final yarn. Combed yarns* contain less cotton seed, as the majority of this is removed
during the combing process. Carded yarns† however contain more of the seed, which is
bleached during the bleaching process to give a more uniform fabric appearance.
Unlike other cellulosic fibres, for example flax, hemp and jute, cotton is unique in that
its strength does not decrease when wet. The strength of the cotton fibre increases when
wet due to increased hydrogen bonding. This is because water molecules can easily form
hydrogen bonds with the free cellulose hydroxyl groups. Water absorption at high humid-
ity results in the swelling of the secondary cellular wall, which has little effect on the fibre
length. This is unusual, as when swelling takes place there is normally a reduction in length
proportional to the degree of swelling.
During the scouring of cotton fabrics prior to dyeing, alkalis such as sodium hydrox-
ide and sodium carbonate are used. This scouring removes waxes and impurities from
the fabric and has an influence on the dye uptake depending on the amount of alkali
used. Treatment of cotton with higher quantities of alkali, i.e. 26% sodium hydroxide
Cellulose
Lumen
Wax
Figure 1.5 Cross-section of a cotton fibre showing the major features.
* Combing is a process that removes short fibres and some vegetable matter.
† Carding is a process that straightens the fibres.
(mercerising), has a more marked effect on the physical and chemical properties of the
cotton fibre. The average regain of the fibre increases by almost 25%–10.5% at 65% relative
humidity and 20°C. Mercerising (a permanent treatment) also results in physical changes
to the cotton fibre that give added value to the final product. If the fabric is under tension
during the sodium hydroxide treatment the fibre is prevented from shrinking during the
swelling process. Surface lustre is developed, in part due to the changes that take place
in the fibre cross-section. The fibre loses its kidney shape and becomes more circular as
shown in Figure 1.6, thus increasing the surface reflective properties. Increased hydro-
gen bonding between the molecular chains also occurs, thus giving an increase in fibre
strength of approximately 20%. Since the fibre swells dramatically during the treatment,
the fibrils in both the crystalline and non-crystalline regions become more accessible to
the penetration of moisture. Thus the relative moisture absorbency increases. This increase
in moisture absorbency increases the comfort factor of a typical cotton garment. At the
(a)
(b)
Figure 1.6 Scanning electron micrographs of cotton fibre before (a and b) and after (c and d)
treatment with 26% sodium hydroxide illustrating the change in cross-section and longitu-
dinal aspect. Note – presence of dead cotton not affected by mercerisation process (circled in
white). (Courtesy of Dr L. Arnold, RMIT University.)
(Continued)
(c)
(d)
Figure 1.6 (Continued) Scanning electron micrographs of cotton fibre before (a and b) and
after (c and d) treatment with 26% sodium hydroxide illustrating the change in cross-section
and longitudinal aspect. Note – presence of dead cotton not affected by mercerisation process
(circled in white). (Courtesy of Dr L. Arnold, RMIT University.)
same time the dye-ability of the fibre increases, so that a lower quantity of dye is required
for a given shade depth. Not all cotton fabrics are mercerised, however most cotton sewing
threads are mercerised to give increased yarn strength. Cotton fibres have a natural twist
or convolution which is present during fibre growth. The process of mercerising reduces
and in some cases completely removes the convolutions. Figure 1.6a through d show the
effect of mercerising on both the cross-section of the fibre and the longitudinal aspect of
the fibre.
Within any given cotton crop, a percentage of immature and dead fibres will always be
present. The percentage of these fibres depends to a large extent on the conditions during
the ripening period of the crop. Figure 1.7 illustrates the variability in the fibre cross sec-
tion between mature, immature and dead fibres. Immature fibres dye lighter than mature
fibres causing colour variation. Dead fibres in a yarn or fabric show as uncoloured or white
flecks lowering the appearance of a fabric and thus the final finished quality.
Cotton fibres sourced from different growing regions vary widely in fibre length, fine-
ness and colour. These variations can result in large fibre losses during spinning and wet
processing particularly when immature fibres are present. The presence of short fibres in a
yarn can result in pilling, and immature fibres can cause colour variations (due to poor or
no dye uptake). Variations between batches can also result in variable uptake of dye by the
fibre, also giving rise to colour variations.
Mature fibre with a

very thick secondary
wall. When dyed the
secondary wall is dark.
Notice the lumen is
completely collapsed.
Immature fibre
secondary wall shows
both thick and thin
places, so the fibre will
dye lighter. Dyeing
properties can be
improved by mercerising.
Dead fibre with very

little cell wall, when dyed
the fibre appears white.
Cannot be improved by
mercerising.
Figure 1.7 Mature (top), immature (middle) and dead (bottom) cotton fibres.
This is particularly so when scouring using sodium hydroxide. Partial mercerisation

of the fibre can occur. For this reason, the quantities of chemicals used in scouring and
bleaching processes are carefully monitored.
Cotton can be dyed with a wide variety of dyes; these include direct, reactive, vat, sul-
phur and azoic dyes. Today the most common dye for cotton is the reactive dye. Military
fabrics are usually dyed with vat dyes due to their very high performance to light and
washing. Many T-shirt garments are printed or dyed using organic pigments, the colour
being held on the garment by means of an acrylic resin binder.
Kapok is from the seed pods of the kapok tree Celba pentandra, which grows in Malaysia
and Indonesia. The fibre is light weight soft and hollow but brittle and is removed by hand
from the seed pods, dried and shaken to remove any seeds. Fibre length is approximately
18–24 mm with a pronounced lumen. Compared to cotton kapok has a much lower cellulose
content. The fibre is unsuitable for spinning into yarn due to its brittleness and inflexibility.
It is an extremely buoyant fibre and will support more than thirty times its own weight and
does not waterlog. Originally used in life vests and safety equipment, mattress and furniture
fillings and pillows. Today its use has declined due to the introduction of synthetic fibres such
as polyester, which is now widely used as fillings in many furniture applications.
Coir is a coarse fibre that comes from the fibre mass between the husk of the coconut and
the shell of the kernel. Coconuts are soaked in water to loosen the fibre and remove some of the
hemicelluloses and pectins. The raw fibre consists of about 40% cellulose 40% lignin, the bal-
ance being hemi-cellulose and pectins. The fibre can be recovered either by hand or machine
that tears away the non-fibrous material from the longer fibres. It is dark brown in colour.
Individual fibres are 15–30 cm in length. The fibres are combed and then spun into yarn, it can
be bleached and dyed and has good resistance to abrasion and weathering.
The principle application of coir is in ropes, twines and matting. Fishing nets can be
made from coir fibre but to a large extent this application is diminishing and being replaced
by fibres of synthetic origin. Coarser fibres are used in brushes, broom and door mats.
1.3.2 Stem Fibres – Flax, Jute, Ramie, Hemp, Kenaf

Stem or bast fibres form the fibrous bundles in the inner vascular areas of the stems
of dicotyledenous plants. They are thick-walled cells that overlap one another and are
cemented together by non-cellulose materials, principally pectin. The fibres are released
from the woody bundles by a process of natural decomposition called retting, which
could be considered to be controlled rotting. Unlike cotton, bast fibres have a lower
cellulose content being only 65%–75% of the fibre. The other components are hemi-
cellulose, pectin, and lignins. Of the bast fibres the major material is flax, the source of
linen. Other bast fibres of interest are ramie, hemp, jute and kenaf. Blends of linen with
synthetic fibres such as polyester are now common in high-class fashion shirtings and
blouses.
1.3.2.1 Flax
Flax comes from the stem of an annual plant Linum usitatissimum, which grows in both
temperate and sub-tropical climates. The seeds of the flax plant are crushed, yielding flax
oil known commercially as linseed oil. Within the inner bark there are long, thin, thick
walled cells that compose the fibre strands. Only about one quarter of the stem is fibre.
The fibres are held together by woody material and cellular tissue that must be removed to
liberate the final fibre. Retting of the woody bundles is achieved through several different
processes:
• Dam retting where the bundles of stems are immersed in ponds or dams for about
10 days.
• Dew retting where the stems are laid on the ground and left for several weeks,
where they are continually wetted with dew and rain, which results in the forma-
tion of moulds that enhance fermentation. Dew retted flax is darker than fibre that
has been dam or tank retted.
• Tank retting where the stems are mechanically stripped of seed bolls and the straw
tied in bundles and packed into a concrete tank filled with water. The water is
heated to about 30°C and retting is completed in about 3 days. A second retting
may be employed to improve the fibre quality.
• Chemical retting using alkalis such as sodium hydroxide and sodium carbonate
with soap have also been used, but this method is more costly compared to the
biological methods previously discussed.
The lightest colour and best quality flax is produced by tank retting.
The dried retted flax is then ready for breaking and scutching. In this process, the dried
retted flax is passed between fluted rollers that break the woody core without damaging
the fibres. In scotching, the broken straw is beaten on a scutching machine to separate the
individual fibres from the unwanted woody matter. After scotching, the fibres are combed
or ‘hackled’ by drawing the fibre through sets of pins, each successive set finer than the
previous set. This process separates the coarser fibres and also arranges them parallel ready
for spinning into yarn.
After boiling and bleaching, flax is almost pure cellulose with a specific gravity of 3.54.
Flax is soft with a lustrous appearance, yellowish white in colour, but can vary depending
on the retting process used. Fibre length is 6–65 mm, but on average is about 20 mm, and
has a fineness of approximately 20 microns. In cross-section the fibre is polyganol with

thick cell walls. Immature fibres are more oval in cross-section. A characteristic lumen
or canal runs through the centre of the fibre. The fibre is stronger than cotton and has
minimal extensibility. It has good resistance to alkalis and will withstand the action of
dilute acids. Because of its high moisture absorption properties it is used in table linen
and towelling. Other applications of flax are in high quality ladies’ blouses and shirts. The
application of blends of linen with synthetic fibres such as polyester greatly improves its
easy-care properties.
1.3.2.2 Jute
The jute plant Corchorus is an annual that grows in hot damp climatic regions and is culti-
vated mainly in India and Bangladesh. Only two species are grown on a commercial scale,
G, capoularis and G. olitorius; the first species produces ‘white jute’ and the second ‘Tossa
jute’. White jute is a little misleading as the natural fibre is in fact a golden yellow colour.
Retting is carried out in a similar manner to that employed for flax. Commercial jute var-
ies in colour from yellow to brown. Chemically the dry fibre consists of 71% cellulose, 13%
lignin 13% hemicelluloses, 0.2% pectin 0.5% fats and waxes and 2.3% water solubles. The
fibre strands can be up to 2 meters in length. Following retting, the fibres are hung to dry
in the sun, then compressed into bales ready for spinning. In cross section, the jute fibre is
polygonal with five to six sides. The lumen is oval in cross-section but irregular compared
to flax, which has a regular lumen. Compared to flax, jute is not as strong due to the uneven
thickness in the cell walls. The moisture absorption properties of jute are unusual, being
very hygroscopic, able to absorb up to 23% moisture in humid conditions with a normal
regain of 13.75%. Prolonged exposure to moisture causes the fibre to lose strength. Jute is
more resistant to micro-organisms than either grey cotton or flax; this is due to the pro-
tective effect of the lignin. To a large extent, many applications for jute have been replaced
by synthetic fibres such as polypropylene and polyester, however jute is still used as the
secondary backing in carpets due to its non slip properties. Other applications for jute are
ropes, cordage, linings in older style furniture, sackings and bales.
Kenaf, Guinea hemp or Mesta from the plant Hibiscus cannabinus grown in Africa,
India and Bangladesh, has similar properties to jute and can be used as a substitute but is
produced on a much smaller scale on the international market.
1.3.2.3 Ramie
Ramie or China grass comes from the plant Boehmeria nivea that belongs to the family of
stinging nettles. It is a perennial grown mainly in China. The root sends up many stalks to
a height of approximately 1.2–2.0 meters. These are harvested every two to three months
when the lower stalks turn yellow and new stalks begin to appear. The fibres are removed
from the stalks by decortication which consists of mechanically peeling or beating the bast
and bark followed by soaking in water and scraping to free the fibres from the bark. The
final fibres vary in length from 2.5–30 cm. The cellulose content of ramie rises from 75%
cellulose to 85% cellulose after decortications, and after degumming the cellulose content
on dry basis is 96%–98%. Degumming is accomplished by treatment in sodium hydroxide
solution followed by bleaching. Ramie fibres are white and lustrous and are unaffected
by the action of sunlight and can be blended with other fibres. The fibre has a high resis-
tance to bacteria, fungi and mildew. Treatment in strong 26% sodium hydroxide solution
(mercerising strength) brings about changes to the fibre similar to that achieved in the
mercerising process used on cotton. Fabrics made from ramie are very absorbent and dry
quickly and can be used for the manufacture of towels and table linen.
1.3.2.4 Hemp
True hemp derived from the plant Cannabis sativa has been in use since ancient times. The
term hemp has been used to describe a range of fibres from different species used for the
manufacture of ropes and cordages, some being derived from stalks and others from leaves
such as Manila hemp and New Zealand hemp. Cannabis sativa is an annual plant grown
from seed in temperate climates. The stalks should not exceed 1.8 meters in height if good
quality fibre is to be produced. Harvesting takes place when the lower leaves turn yellow
and the stalks are pulled from the ground. Retting and scutching are similar to that used
for flax. Under microscopic examination hemp resembles flax in appearance; this similar-
ity means that some hemp varieties are difficult to distinguish from flax.
Hemp is overall a courser fibre than flax, being darker in colour and difficult to bleach.
The cross-section of hemp is polygonal and the lumen is broader than that of flax. Dry
hemp fibres have high cellulose content, approximately 75%; lignins are lower at 3.6%,
compared to jute. Other components are hemi-cellulose about 17%, fats and waxes 0.8%,
water solubles 2.0%, pectins 1.0% and lignins 3.6%. The major applications of hemp are in
the manufacture of ropes, twines and cordages. More recently however there have been
garments manufactured from blends of hemp with cotton. High quality hemp can be used
as a substitute for flax.
1.3.3 Leaf fibres – Sisal, Hemp (Manila and New Zealand)

The leaves of certain tropical and sub-tropical monocotyledonous plants are held in shape
by fibrous cells that run in strands (vascular tissue) through the length of the leaf. These
cells help to give the fibre strength and maintain its shape. As a general rule the leaf fibres
are courser than fibres that come from dicotyledonous plants, which are the source of the
bast fibres. The leaf fibres are often referred to as ‘hard’ fibres being that they are thicker
and stiffer than either the bast or seed fibres. The major leaf fibres are sisal, henequin, New
Zealand hemp and abaca.
Synthetic and regenerated fibres from natural polymers have, to a large extent, taken
a substantial share of the market for both the bast and leaf fibres. More recently however,
with the rising cost of petroleum, there has been an increased awareness of the advantages
of these natural products, particularly in relation to their sustainable production.
1.3.3.1 Hemp-Abaca (or Manila Hemp) and New Zealand Hemp

Musa textilis a member of the banana family indigenous to the Philippines and is the
source of manila hemp for the manufacture of high quality ropes and cordages. The fibre
is in the form of strands of many fibres held together by natural gums. The length of
these strands varies from 1–3 meters and 0.05–0.3 mm in diameter. Final fibre length is
3–12 mm and 16–32 μm in diameter. The fibre has a good natural lustre and is off-white
in colour. Chemically abaca fibres contain about 77% cellulose on a dry weight basis and
are coated with about 9% lignin. The individual fibres are separated from the soft fleshy
material by drawing the strips of the leaf between a knife edge and a wooden block. The
repeated scraping in this manner removes the soft cellular matter that surrounds the fibre.
Once the fibre is cleaned it is hung up to dry. When dry it is ready for use. The fibre has a
variable cross-section that is irregularly circular or oval in shape. One of the advantages
of abaca is that it is not affected by salt water; hence it is ideally suited for use in marine
cordage.
New Zealand hemp is the perennial plant Phormium tenax native to New Zealand.
Captain Cook noted that the Maoris produced fine fabrics from the leaves of this plant.
Fibres are extracted from the leaves by mechanical scraping. Whilst the fibre is fairly
strong and has good resistance to sea water it is weaker and softer than Manila hemp. The
individual fibres are smooth with pointed ends with an average length of about 6–7 mm,
a near circular cross-section and a narrow circular lumen. Compared to other leaf fibres,
the lignin and hemicelluloses is higher and the cellulose content lower.
1.3.3.2 Sisal and Henequen

Sisal is the leaf fibre that comes from the plant Agave sisalana. The name is derived from
the Yucatan port ‘Sisal’ situated on the Gulf of Mexico. It was from this port that the
fibre was first exported. The sisal plant is grown in tropical regions and grows huge leaves
almost from ground level. The leaves are harvested when the plants are 2–4 years old. A
good plant may yield up to 400 leaves in its lifetime of seven to eight years. The fibres are
extracted from the leaves using a decortication process in a machine that scrapes the pulpy
matter from the fibres. The fibres are washed, then dried either in the sun where bleaching
takes place or in an oven. Fibre strands vary in length from 60–120 cm in length and have
an average diameter of 24 μm. The fibre is polygonal in cross-section with a broad lumen
which varies in thickness and is often packed with small granules. The fibre ends are blunt
and in some cases forked. Sisal is used for the manufacture of ropes, cordages, twines,
upholstery, carpet and custom rugs. Yarns have a tendency to shed the fibre and fabrics
quickly fade out.
Henequen fibre is produced from the plant Agava bourcroydes and is often known as
‘Yucatan sisal’ or ‘Cuban sisal’. The plant has prickly leaves that are grey green in colour.
Processing of the leaves is similar to that for conventional sisal. The fibre strands are long
and slightly finer than those of sisal. Applications include string and agricultural twine.
1.4 Natural Fibres – Animal Fibres
Natural fibres that have an animal source (refer to Figure 1.8 for sub-classification), such as
wool or silk, are protein based.
Proteins are chain like molecules formed by the union of α-amino acids joined together
by the peptide link as shown below.
R R
NH2 – CH NH2 – C – COOH
COOH H
The difference between proteins is related to differences in the side group –R pendent
to the main chain. Over 20 amino acids are known to exist because of the difference in size,
shape and chemical reactivity. The limited number of side groups is sufficient to result in
an extremely large number of polypeptides and proteins due to arrangements that are pos-
sible. Table 1.4 lists the differing ratios of amino acids in wool and silk fibres.
Fibres
Natural Man-made
Hair Silk
Wool – sheep Bombyx mori – silkworms
Goat family – mohair, cashmere
Arachnids – spider silk
Camel family – camel hair, alpaca, vicuna
Hymenotpera – bees, wasps, ants
Other – rabbit fur, possum fur
Figure 1.8 Sub-classification of animal fibres.
1.4.1 Wool
Of the protein fibres, wool is the predominant fibre, with world production being approxi-
mately 1.2 million tonnes in 2008. Raw wool after shearing contains impurities, such as
grease, swint (sweat), dirt and vegetable matter. The quantities of these impurities will vary
depending on the agricultural region where the animal has been grazing. Merino sheep
that are raised under shedded conditions show minimal amounts of vegetable matter and
dirt present in the fleece. After scouring and removal of vegetable matter under conven-
tional grazing conditions the yield will be about 65% of clean wool. It is normal for scoured
wool to retain about 0.5% residual grease as this assists in later mechanical processes such
as carding and combing to minimise fibre breakage.
In a typical wool fibre there are four regions:
1. The epicutical (or outer sheath): A thin hydrophobic membrane and the only non-
protein part of the wool fibre. This membrane is water repellent for water droplets
but will absorb water vapour through its many microscopic pores (thus absorbing
water vapour from the body and releasing into the air);
2. The scale-cell layer (comprising the enzyme resistant exocuticle and enzyme digestible
endocuticle): This layer contains the flat scale-like cells which overlap each other
(refer to Figure 1.9 for a cross section of wool illustrating the overlapping scales).
The number of scales present on a wool fibre is known to vary greatly depending
on the quality of the wool with fine wools having significantly more scales than
coarse or medium wools.
3. The cortex: This layer comprises some 90% of the wool fibre consisting of millions
of spindle shaped cells; and
4. The medulla: Many wool fibres (particularly the coarser ones) have a hollow cen-
tral space running longitudinally to the fibre length. The medulla may be either
empty or contain a loose network of open cell walls. Coarse and medium wools
are generally characterised as having a greater proportion of the fibres medullated.
Conversely, in fine wools the medulla is either absent or so fine as to not be visible.
The dimension of wool fibres varies considerably depending on the quality of the wool.
Fine wool fibres are generally about 3.8–12.7 cm long, medium wools are generally 6.3–
15.2 cm long and long wools are 12.7–38 cm long. The diameter can also vary significantly
Table 1.4 Amino Acids Present in Silk and Wool Protein Fibres
g- Amino Acid per 100 g Protein
Amino Acid Silk Fibroin Wool Keratin
Inert acids
Glycine 43.8 6.5
Alanine 26.4 4.1
Valine 3.2 5.5
Phenylalanine 1.5 1.6
Isoleucine 1.4 0.0
Leucine 0.8 9.7
Acidic
Glutamic Acid 2.0 16.0
Aspartic Acid 3.0 7.3
Basic
Histidine 0.5 0.7
Lysine 0.9 2.5
Arginine 1.1 8.6
Hydroxyl
Serine 12.6 9.5
Tyronsine 10.6 6.1
Threonine 1.5 6.6
Miscellaneous
Proline 1.5 7.2
Cystine 0.0 11.8
Tryptophan 0.0 0.7
Cysteine 0.0 0.1
Methionine 0.0 0.4
with the average diameter for a top-quality merino wool being 17 μm; a medium wool fibre
about 24–34 μm and a long wool fibre about 40 μm.
Wool fibres have a natural wave or crimp which is a unique characteristic amongst
natural fibres. The crimp is not in a single plane but has a three dimensional waviness.
This natural waviness allows fibres to hold together when made into a yarn. Once again the
quality of the wool will be reflected in the waviness of the fibre. The finest wools can have as
many as 12 waves to the centimetre whereas poorer quality wools will have 2 (or even less).
Intercellular cement holds the cuticle cells together and separates them from the
underlying cortical cells. During the processing of washable wool, i.e. shrink resistant
wool, the hydrophobic epicuticle is damaged and the scale structure is significantly
changed due to the action of chlorine and a cationic resin. This process is the ‘KROY TM’
process.
1.4.1.1 Hair fibres or Speciality Wools

Hair fibres or ‘Exotic fibres’ are those that come from other animals such as:
1. Goat family—mohair, cashmere

2. Camel family—camel hair, alpaca, vicuna
3. Rabbit fur and other fur animals such as the possum
Figure 1.9 Cross section of 28 micron wool illustrating the overlapping scales.
Mohair is from the Angora goat which originated in Turkey, is a lustrous fibre with a
staple length of 20–30 cm for a full year’s growth. The fibre’s surface has some similarities
to wool; it is covered with epidermal scale anchored much more closely to the body of the
fibre compared to wool, there being only about half the number of scales as there are on the
wool fibre. The fibre has a circular cross section with an average diameter of about 25 μm
for kid mohair and 35 μm for the adult fibre. Chemically, the main constituent is keratin,
and its chemical behaviour is very similar to wool.
Cashmere was the name originally given to the hair from the Asiatic goat Capra hircus
laniger, the Tibetan Cashmere goat, which is raised as a domestic animal in parts of China,
Persia and northern India. In Australia, New Zealand and Scotland hair from the selec-
tively bred feral goats is also called cashmere. The cashmere fibre varies in length from
5–10 cm with diameter of 14–16 μm. In cross-section the fibre is circular or slightly oval,
and in the coloured fibre the natural pigment can be clearly seen in the cortical layer.
Because of its softness, cashmere is mainly used in high-quality ladies’ dress wear and
knitwear. The fibre is more easily damaged chemically particularly by alkalis due to its
fineness and better wetting properties.
Camel hair is from the camel Camelus bactrianus. The hair is shed by the animal and
collected. The soft inner hair is the finest and is used in high grade men’s over-coatings.
Camel hair gives warmth without weight so the fibres are highly prized for quality apparel.
The coarser outer hair or guard hair length is approximately 26–30 cm and is used in the
manufacture of beltings and interlinings.
Alpaca is from the fleece of the Lama pacos from South America. The fleece is coloured
brown, fawn, black and sometimes white, with a fibre diameter of 24–26 μm and fibre
length of 20–28 cm. Alpaca has a distinctive scale structure with a natural crimp and
medullation. Dress fabrics, tropical suiting’s and blends with wool are common. Alpaca is
difficult to dye and is usually used in its natural colours. The fibre is noted for its softness,
fineness and lustre.
Vicuna is the smallest and rarest of the llama family. The undercoat hair fibre from the
vicuna is regarded as being the finest wool like fibre with a diameter of 13–14 μm. About
500 g of fibre is obtained from each animal. It is the rarest and most expensive of all fibres
and has similar uses to cashmere.
Angora fibre is obtained from the pelts of the angora rabbit. After shearing the hair
from the pelts is separated by blowing. The fine hair, 6–8 cm in length 13 μm in diameter,
is used in the manufacture of felt hats. Angora is difficult to dye and is very slippery and is
usually blended with wool and or nylon to facilitate spinning.
Other animal fibres may be encountered in textiles including fibres from domestic
animals such as cats, dogs, cows, horses, rodents, etc., as well as human body hair.
1.4.2 Silk—Silkworm
The silkworm is the caterpillar of a small white moth of the species Bombyx mori from
the class of insects Lepitoptera. This caterpillar thrives only on Mulberry leaves and dur-
ing the growth stage the worms do nothing but eat except for four periods of sleep, which
last about 24 hours. During this resting period the worms shed their skin or moult. After
the fourth moult it settles down to a feed lasting about 10 days during which period it
eats about twenty times its own weight of leaves. Approximately thirty-five days after
hatching the worm begins to spin its cocoon. The liquid silk is from two glands inside the
worm. The silk flows in two channels to a common exit point which is the spinneret, in
the worm’s head. As it exits the liquid hardens into very fine filaments that are coated and
stuck together with a gummy substance called sericin which comes from two other glands
nearby.
Silk as it emerges from the worm is therefore a twin filament held together as a single
strand by the sericin cement. Sericin is an amorphous globular protein and dissolves in hot
soap solution. Raw silk contains approximately 75% fibroin and 25% sericin. Sericin remains
on the filament surface to protect the fibre from damage during processing. De-gumming
during wet processing of both silk yarns and fabrics removes the sericin, leaving a fibre of
almost pure fibroin. Silk will dissolve in powerful hydrogen bond-breaking solvents such
as cuprammonium hydroxide (as it does not contain the disulphide links present in other
animal fibres).
Each raw filament of silk is roughly elliptical in cross-section. After de-gumming the
fibroin fibre is transparent and uniform in width being 9–12 μm with a smooth surface.
Wild silk (Tussah Silk) fibres are darker in colour and coarser, approximately 28 μm, and
less uniform in width, with pronounced longitudinal striations.
Cross-sections of different silk fibres show significant differences, cultivated Bombyx
being roughly triangular with the corners being rounded. Tussah silk on the other hand
is wedge shaped, whilst Anaphe silk is roughly triangular and the apex of the triangle is
elongated and bent.
Because of silk’s outstanding strength, toughness, high regain, soft handle, excellent
drape and appearance, it has a wide variety of uses in apparel, in particular high-qual-
ity dress goods, lingerie and handkerchiefs. As silk loses about 25% of its weight during
de-gumming, manufacturers add weight to the silk with tin salts, so called tin weighting,
stannous chloride often being used.
1.4.2.1 Spider Silk

The protein fibre spun by spiders has superior mechanical properties compared to those
of the silkworm. When compared to synthetic fibres such as polyamide and polyester, spi-
der silk is said to be superior in strength, elasticity and resistance to compression. There
are many specimens of spider silk, some exhibit elongation as high as 200% whilst oth-
ers exhibit strength similar to high-performance fibres such as high tenacity polyamide.
Spider silk has better water repellent properties compared to both types of natural silk. The
domestication of spiders has not yet been possible, therefore its use in textiles has not been
fully exploited. Being predators, the raising of spiders in dense populations is not possible.
Compared to the silkworm the amount of silk produced by the spider is small.
It has been possible to produce artificial spider silk. Currently DuPont has successfully
spun a protein analogue of spider silk using solvent spinning technology and regenerating
the fibre in a coagulation bath. Spider silk is sticky and may stick to textiles and could be
mistaken for natural silk.
1.5 Mineral Fibres (Asbestos)
Of the naturally occurring fibres asbestos is the most peculiar. Unlike other rocks that
crystallise during their formation, asbestos is derived from a rock that crystallizes in the
form of fibres that are closely packed together giving the surface a grainy appearance. The
use of asbestos has been known since ancient times when it was used as wicks in lamps.
There are three natural fibrous minerals:
• Anthophyllite: Magnesium-iron silicate but of least importance for commercial

use.
• Amphibole: There are a number of varieties of this mineral, of which Crocidolite is
the most important as it is from this mineral blue asbestos is extracted. Chemically
the long flexible fibres are iron sodium silicate and are a blue colour. The average
fibre length is 7.5–10 cm, but it is not as resistant to high temperatures as crysotile.
• Serpentine: Which occurs in two fibre forms is a hydrated silicate of magnesium.
The most important of these is crysotile. It is green to brown in colour and the
fibres extracted form the major world supply of asbestos. The crysotile fibres are
strong, but the ultimate fibrils are very fine, about 25 nm in diameter.
Asbestos has been spun into yarns and then woven into fire-resistant fabrics. Asbestos
was used in manufacture of construction materials (asbestos cement sheet) but this has
now been discontinued in Australia due to health considerations. Aramid polymer fibres
such as NomexTM, and glass fibres have replaced asbestos in fire resistant fabrics and yarns.
1.6 Man-made Fibres
Man-made fibre can be classified into two groups, those derived from natural polymers
such as cellulose (the regenerated fibres) or true synthetic fibres synthesized from simple
organic chemicals found in coal and oil. Methods of production vary with the different
polymers. Essentially there are three methods of producing fibres from polymeric materi-
als. These are;
1. Melt spinning: Applied only to thermoplastic polymers. The concentrated molten

polymer is forced through tiny holes in a spinneret. As the polymer cools, the fila-
ments are formed.
2. Dry spinning: The polymer is dissolved in a volatile organic solvent and the vis-
cous solution pumped through the spinneret into a column of air where the sol-
vent evaporates, leaving a continuous filament of fibre. The solvent is recovered for
reuse in the process.
3. Wet spinning: The viscous polymer solution, dissolved in either an aqueous or
organic solvent, is pumped through the spinneret, which is immersed in a bath
containing suitable coagulating chemicals. The polymer precipitates in the bath;
the extruded filaments are formed and drawn off ready for further processing.
Following extrusion the filaments are stretched or drawn mechanically to orient the
molecular chains along the longitudinal axis of the fibre. This stretching maximises the
molecular forces between the molecular chains, increasing polymer crystallinity and fibre
strength. The degree of crystallinity is a measure of the percentage of polymer present in
its crystalline form, the balance being in a disordered or random state. The random state of
a polymer is referred to as being an amorphous state. Within the fibre the crystalline and
amorphous regions do not have any particular order but are randomly spaced along the
fibre axis. The extent to which one or the other predominates will determine the ultimate
properties of the fibre.
The filaments can be cut into lengths (staple fibres) or remain continuous (continuous
filament fibres); they can also be crimped so that they are compatible with the natural fibre
yarn spinning system that may be used. Often synthetic fibres are blended with a natural
fibre to enhance the properties of the final fabric. A small amount of twist is inserted into
a filament yarn to aid in its further processing. Some synthetic fibre yarns are often further
processed to increase the bulk and elasticity of the yarn. One such treatment is that of false
twisting where high twist is inserted in the filaments as the yarn passes through a heating
zone. The heat applied sets the twist, which is then untwisted, giving the yarn an increase
in bulkiness in its relaxed state.
1.6.1 Fibres Regenerated from Natural Polymers

Man-made fibres regenerated from natural polymers are sub-classified (refer to Figure 1.10)
based on the source of the natural polymer (i.e. cellulose based or protein based).
1.6.1.1 Viscose Rayon

Of all the regenerated cellulosic fibres viscose is by far the most important. There are many
manufacturers throughout the world. In 2006 world production of viscose grew by about
10% to 2.7 million tons, China being the largest consumer at 1.5 million tons*. Viscose
rayon is manufactured from wood pulp, usually made from spruce wood. During the
* Source: www.worldresearchandmarkets.com.
Fibres
Natural Man-made
Transformation of natural polymers Synthetic polymers
Cellulose based Protein based

Rayon (viscose, cupro, lyocell, modal) Casein
Cellulose ester (acetate, triacetate) Soya protein fibre
Figure 1.10 Sub classification of major man-made fibres from natural polymers.
preparation of the pulp using the acid sulphite process, the lignin is removed by sulphona-
tion as a lignosulphonic acid. The cellulose chains are shortened due to the degradation
that takes place. The degree of polymerisation (DP) of Spruce is about 2400; this is too
large for fibres. After pulping, the DP is 1400 and the degree of polymerisation in the final
viscose fibre is 200–400 compared to cotton at 300–4000. At this stage the α-cellulose
content is too low and the lignin content too high so the pulp is bleached using sodium
hypochlorite. The pulp is then beaten to separate the cellulose fibres, the slurry of pulp and
water run out onto a gauze belt where the water is extracted using vacuum extraction. The
fibres in the pulp are held together by strong hydrogen bonds between the chains. The non-
crystalline (amorphous) regions have random distribution of hydrogen bonds that are not
numerous. The pulp is then treated with alkali to form alkali cellulose (see Equation 1.1)
and the mixture allowed to age to reduce the polymer chain lengths. The aging is carefully
controlled to produce chain lengths short enough to give optimum viscosity in the spin-
ning process but still long enough to impart good physical properties to the fibre product.
R = cellulose chain
OH = any one of 3 –OH groups in the glucose residue
R − OH + NaOH → R − ONa + H2O (1.1)

(alkali cellulose)
R − ONa + CS 2 → R − O.CSS Na (1.2)
2R − O.CSSNa + H2SO 4 → 2R − O.CSSH + 2Na 2SO 4 → 2R − OH + 2CS 2 (1.3)
The alkali cellulose is reacted with carbon disulphide (Equation 1.2) to form the cellu-
lose xanthate, which is completely soluble in 7% sodium hydroxide, giving a viscous liquid,
which is the spinning solution. The spinning solution is squirted through a spinneret into
a bath containing sulphuric acid (Equation 1.3), where the filaments harden, initially from
the outside. As regeneration occurs toward the inside of the fibre the fibre shrinks giving
rise to a wrinkled cross-section (refer to Figure 1.11a and b).
Viscose has a much lower crystallinity, about 35%–40%, compared to cotton 65%–
70%, which results in a much weaker fibre in both the dry and wet states. Viscose fabrics
Figure 1.11 (a and b) Cross-section of viscose fibre showing wrinkled appearance and Viscose
spinneret, diameter 26 mm with 200 holes.
have good drape but tend to feel cold compared to cotton. Viscose is amorphous in nature
with a large number of polar hydroxyl groups in the polymer, which results in a fibre with
higher moisture-absorbing properties.
Viscose has much lower resistance to acids and bleaching agents than cotton and is
discoloured and weakened when attacked by mould and fungi.
1.6.1.2 Bamboo
Bamboo fibre is a type of viscose rayon manufactured from cellulose pulp derived from
the bamboo plant. Its properties of moisture absorbency, tensile strength and extensibil-
ity are similar to those of normal viscose. It is interesting to note that it is not possible to
determine the origin of the cellulose (i.e. bamboo or spruce) once it has been through the
viscose process.
Bamboo has become increasingly popular as a fibre for garments and is marketed as
environmentally friendly due to the renewable source. The major source of bamboo for
fibre manufacture is grown in China with plants growing up to 30 m at a rate of 0.3 m per
day. There are more than 70 genera divided into about 1450 species. The cellulose content
of the bamboo plant is about 61% and lignin content approximately 32%. Although the
bamboo itself is a renewable resource, the chemical process required to produce a fibre
useable in a textile can have serious implications for the environment.
Bamboo fibre is predominantly produced using a process similar to that for mak-
ing viscose where the woody stalks are steamed and then mechanically crushed. Sodium
hydroxide solution is added to produce a pulp that is then treated with carbon disulphide
to produce the cellulose xanthate, which is then extruded into a bath of dilute acid.
There is an emergence in legislation in relation to labelling bamboo products with
two governing bodies (the Canadian Competition Bureau and the U.S. Federal Trade
Commission) stipulating that products must be labelled as rayon with the optional quali-
fier of ‘from bamboo’.
1.6.1.3 Improved Forms of Viscose

1.6.1.3.1 High Wet Modulus Rayon An improved form of Viscose was developed by
LENZING as a high wet modulus viscose under the trade name MODAL ® for many uses.
To produce this high tenacity viscose the cellulose is regenerated slowly. Slow regeneration
is accomplished by increasing the amount of Zinc Sulphate and lowering the acid concen-
tration in the coagulating bath, at the same time a higher degree of stretch is applied. The
higher stretch results in a more crystalline fibre; the filament is all skin and has no core.
This improved form of viscose has lower water absorption and swelling and has
mechanical properties that are much closer to cotton. They have a fine micro-fibrillar
structure resistant to 8% aqueous sodium hydroxide solution. Characteristically these
fibres have higher wet strength, increased resistance to swelling in alkaline solutions, a
higher degree of polymerisation DP and a more fibrillar internal structure similar to cot-
ton. Typically the DP would be 500–700 (regular viscose has a DP of 250–400) with the
fibres being 50%–60% crystalline with larger crystallites than in cotton. Because of the
higher crystallinity and the higher orientation these fibres are less likely to deform under
wet conditions compared to regular viscose.
1.6.1.3.2 Lyocell Lyocell is a regenerated cellulosic fibre made by spinning from a sol-
vent. The development of the process of dissolving cellulose in a tertiary amine N-oxide
was patented in 1989 by Akzona, Inc. (now Akzo Nobel) and the process licensed to
Courtauld’s and Lenzing. Lyocell fibres are sold under the trade name ‘TENCEL ®’. This
development has to a large extent resolved two of the major issues in relation to regener-
ated cellulose (viscose) production, that of poor wet strength and environmental pollution.
Little has been divulged regarding the manufacturing process for Tencel. What is
known of the process is that the regenerated cellulose is produced directly from a solution
of cellulose rather than from a cellulose derivative, which increases the degree of polymeri-
sation. Raw cellulose is dissolved directly in a solvent, amine oxide, the solution is filtered,
extruded into a dilute amine oxide solution and coagulated into fibres. The technology
used also allows the recovery of the solvent and the reuse of most of the process water.
Lyocell fibres have better extensibility than cotton, higher crystallinity (resulting in
greater tenacity) and a wet strength that is 2–3 times greater than regular viscose (refer to
Table 1.5 for comparison of properties of cotton, viscose and lyocell). Because the fibres
have a fibrillar structure the fibrils will separate when the fibres are wet and when under
tension. Fibrillation is the longitudinal splitting of the fibre into a bundle of micro fibres
of smaller diameter usually 1–4 μm. This occurs mainly at the fibre surface and usually
occurs during processing in rope form either in a winch or jet-dyeing machine. The degree
of fibrillation can be controlled during processing and allows the production of attractive
fabric surface effects such as peach skin with a soft pleasant handle. Tencel has been used
in a range of garment types including sportswear, mens’ shorts and trousers.
1.6.1.4 Esters of Cellulose

Schutzenberger in 1869 first prepared cellulose acetate, but it was Cross & Bevan in 1894
who described a practical method of production. In 1903 it was discovered that cellulose
acetate could be dissolved in the relatively cheap and safe solvent acetone. Early produc-
tion of cellulose acetate in Britain was carried out by the brothers Dreyfus at their factory
in Spondon, Derby. Initial production of the cellulose acetate was as a plasticised lacquer
(dope) for the wings of aircraft. At the end of the First World War the brothers Dreyfus
found themselves with a large factory and no demand for aircraft dope. By 1921 they had
ironed out most of the technical difficulties and cellulose yarn was being marketed in
Britain with the name ‘Celanese’. The development of cellulose acetate was a major break-
through in the development of alternatives to natural silk. Cellulose is reacted with acetic
anhydride and acetic acid in the presence of sulphuric acid. Initially the cellulose is fully
acetylated, all –OH groups are acetylated and cellulose triacetate or 6/6 acetate is formed.
Whilst fibres could be made from cellulose triacetate, and were by the Lustron
Company, it ceased production in 1915 because the fibres could not be dyed with existing
dyes and the solvents used (chloroform) in production of the fibre were not satisfactory.
Courtauld’s work in the 1950’s resulted in the production of ‘Courpleta’, dry spun cellulose
triacetate using methylene chloride as the solvent.
To hydrolyse to a 5/6 acetate, the triacetate is run together with the excess acetic anhy-
dride and acetic acid into water so that a 95% solution of acetic acid results. The solution is
warmed to 50°C so that the water can remove some of the acetyl groups. This acid hydro-
lysis takes place over a period of about 20 hours. At this stage the product becomes soluble
in acetone, which has a boiling point of 56.5°C. The Degree of Polymerisation is 350–400.
By acetylation the possibility of hydrogen bonding has been removed and therefore
the water absorption properties are greatly reduced. The following illustrates the effect
of hydrogen bonding on the water absorption properties of both natural and regenerated
Table 1.5 Comparison of Physical Features of Viscose, Lyocell and Cotton
Viscose Lyocell Cotton

Tenacity (cN tex-1) Dry 18–22 37–41 20–24
Wet 11–13 30–35 26–30
Moisture regain (%) at 20°C 65% RH 12–14 11–12 8.5
Extensibility (%) Dry 17–25 13 7–9
Wet 23–38 15 12–14
Melting Point (°C) Does not melt Does not melt Does not melt
Softening point (°C) 130 none none
Handle Soft and cold Soft and Silky Soft and warm
cellulosic fibres when measured at 20° C and 65% relative humidity, cotton 8.5%, viscose
13.0%, secondary acetate 6.5% and triacetate 4.5%. Acids will hydrolyse the fibres and the
acetyl group can be saponified by alkali leading to fibre yellowing. The secondary acetate
is soluble in acetone but not soluble in chloroform and the triacetate because of its higher
degree of acetylation is soluble in methylene chloride and chloroform.
Both secondary acetate (5/6 acetate) and triacetate are thermoplastic and can be heat
sett to give a permanent pleat. Triacetate is more extensively affected by heat setting than
the secondary acetate. Secondary acetate has been used extensively as linings in suitings
and in satin fabrics destined for wedding apparel. Blends of triacetate with polyester in
knitted fabrics have been popular in the past.
1.6.1.5 Regenerated protein fibres

Casein: Acid treatment of milk coagulates the protein casein into curd which is
then washed and dried. The casein fibres are spun from an alkaline solution of
the casein into an acid coagulating bath which may also contain formaldehyde.
The coagulated filaments are stretched during the coagulation process. Initially
the filaments are soft and weak and require further hardening in formaldehyde to
increase their stability to water and alkali and give adequate strength to the fibre.
Fibres were manufactured as Merinova in Italy, Fibrolane in the UK and Lanital in
France. Casein fibres have a silk-like appearance and are insoluble in cold concen-
trated hydrochloric acid which will dissolve silk.
Soya protein fibre (SPF): The development of this fibre was pioneered by the Ford
Motor Company, production commencing in 1939. The manufacturing process is
similar to that of casein fibre. However production in the United States ceased after
several years. More recently Harvest SPF Textile Co., Ltd., have been manufactur-
ing and marketing soya protein fibre. The fibre has a soft cashmere-like handle and
is ideally suited for knitted fabrics. The fibre cross-section is a dumb-bell in shape
which gives it natural lustre. The fibre is readily dyed with acid and reactive dyes
with colour yields and fastness properties similar to wool. SPF fibre has a moisture
regain of 8.6%, which is comparable to that of cotton. Acid and alkali resistance
are similar to that of wool.
1.6.1.6 Alginate Fibres

Sodium alginate solution extracted from seaweed is extruded into a coagulating bath con-
taining a calcium salt. This precipitates the fibre as the calcium alginate, which is non-
flammable, but soluble in alkaline solutions. Alginate fibres have been used as a support
yarn to produce ultra light-weight fabrics, when the alginate yarn is dissolved out in the
finishing process. Alginate fibres are used in surgical dressings which are non toxic and
readily absorbed into the blood stream. The fibre is soluble in sodium carbonate and has
high ash content.
1.6.2 Synthetic Fibres

True synthetic fibres are manufactured from two polymer types, condensation polymers
and addition polymers. Figure 1.12 shows the sub classification of synthetic fibres.
Condensation polymers are prepared from the condensation reaction of two mono-
mers having two functional groups from which a simple molecule is eliminated. Polyesters,
Fibres
Natural Man-made
Transformation of natural polymers Synthetic polymers
Polyester Polyamide Polyolefin Polyvinyl Polyurethane Aramid

-Polyethylene -Nylon -Polypropylene -Acrylic -Segemented -Kevlar
terephthalate -Polyethylene -Modacrylic -Non segmented -Nomex
(PET) -Chlorofibre (PVC)
-Polybutylene -Vinylal (PVA)
terephthalate -Fluorofibre
(PBT)
-Polytrimethylene
terephthalate
(PMT)
Figure 1.12 Sub classification of synthetic polymer fibres.
polyamides, polyurethanes and aramids are examples of linear polymers produced by this
process. The repeating units in the polymers’ chain are joined by functional groups such
as amide and ester which can undergo hydrolysis by acids and alkalis resulting in polymer
degradation.
Addition polymers are formed by the direct addition of the monomer to itself without
the elimination of any molecules. Monomers that undergo addition polymerisation must
contain double bonds. Addition polymers are more crystalline and have a much higher
degree of polymerisation than condensation polymers and because the C-C bonds along
the molecular chain are strong there are no sites for easy access to corrosive chemicals that
would cause cleavage of the chain.
Polymers formed by the addition method have the same number of atoms in the struc-
tural unit as the monomer from which the polymer has been made. Typical examples of
addition polymers are polyolefins and the polyvinyls (the acrylic polymers). Polymers
can be linear where there is one single repeating unit as in polyethylene or they may be
branched chain polymers. Some branched chain polymers may link two linear chains to
form a cross-linked structure. Internal cross-linking of polymers in cotton and viscose
improve the dimensional stability of the thereby reducing the tendency of the fabric to
shrink).
Polymers may also be classified as Atactic, Isotactic and Syndiotactic.
1. Atactic polymers possess a random distribution along the polymer chain of the
two asymmetrical carbon atoms; because they are not orderly structures they will
not crystallise hence are not suitable for fibres.
X X X
– CH2 – CH – CH2 – CH –CH2 – CH – CH2 – CH – CH2 – CH – CH2 – CH –
X X X
2. Isotactic polymers have a regular arrangement of the substituent groups, i.e. where
all the groups X are disposed in the same fashion. Isotactic polymers are of con-
siderable interest as they have higher melting points than the normal atactic mate-
rials, e.g. atactic styrene has a melting point of 85°C compared to 230°C for the
isotactic styrene.
– CH – CH2 – CH – CH2 – CH – CH2 – CH – CH2 –
X X X X
3. Syndiotactic polymers have a regular arrangement of the substituent groups.

X X
CH2 – CH – CH2 – CH – CH2 – CH – CH2 – CH – CH2 – CH – CH2 –
X X X
1.6.2.1 Polyesters
Polyester fibres, the most common being polyethylene terephthalate (PET)*, dominate the
global synthetic fibre industry. Polyester fibre represented 77% of world synthetic fibre pro-
duction in 2004 with an expected annual growth of 7% over 10 years†. There are a large
number of manufacturers of polyester, the majority being in Asia. Typical trade names are
Dacron, Terylene, Fortrel, Terigal, Teteron, Trevira, Grilene, Diolen and Tergal. Polyesters
are a very versatile fibre and have extensive end product uses from carpets, industrial fibres
and yarns for tyre cords, to car seat belts, sailcloth, and fabrics as well as apparel and
household furnishing fabrics.
The condensation polymer is synthesized from a dibasic acid (teraphthalic acid) and a
diol (ethylene glycol) this combination of monomers results in a fibre of high melting point
and good stability to chemical attack.
( n + 1) R ( OH )2 + n R ′ ( COOH )2 → HO[ROOCR ′COO ]n ROH + 2n H2O

Polyester is highly crystalline (65%–85%) and hydrophobic in character, which results
in very low moisture regain, 0.4%–0.5% at 20°C and 65% relative humidity. The presence
of the aromatic nucleus in the chain increases the melting point (249°C–250°C) and the
chemical stability. Stability to concentrated acids is excellent but the ester linkage is hydro-
lysed by alkali. Alkaline hydrolysis with sodium hydroxide of the surface is used in the
weight reduction of polyester fabrics, this process in common to polyester woven georgette.
The surface hydrolysis of the fibre improves the drape as well as giving the fibre improved
antistatic properties. The fibre is insoluble in most organic solvents, but will dissolve in
σ-chlorophenol and hot meta-cresol.
Polyester is melt spun and produced as filament yarns and staple fibre. Staple fibre
lengths vary depending on the spinning system, the shortest staple being made for the cot-
ton system. Filament yarns are textured to increase the bulk, some staple fibre is crimped,
* Other forms of polyester include polybutylene terephthalate (PBT) and polytrimethylene terephthalate.
† Journal for Asia on Textile & Apparel – June 2006.
again to increase the bulk and make it more suitable for spinning with wool. The fibre is
de-lustred with titanium dioxide, the quantity being varied depending on whether the
yarn is dull or bright. Yarn cross-sections can vary depending on the configuration of the
holes in the spinneret.
Polyester and polyester blended fabrics can be heat set to give dimensional stability,
and prevent shrinking during domestic laundering. Fabrics made from polyester are more
stable to ultraviolet light than a comparable product made from nylon. The blending of
polyester staple fibre with wool and cotton take advantage of the easy care properties of
the fibre and increase its suitability in a wide variety of furnishing and fashion fabrics.
Disperse dyes are used for the dyeing of polyester yarns and fabrics; some filament yarns
are coloured during the spinning process, pigments being added to the polymer melt to
accomplish this technique. High tenacity continuous filament yarns are used for ropes,
conveyor belts, automotive seat belts and tarpaulins. Polyesters are available that have been
modified by copolymerisation with an acidic monomer to make them dyeable with cat-
ionic (basic) dyes. Trevira CS is modified polyester that is inherently flame retardant that
has been used in home furnishings, particularly drapes and upholstery fabrics.
The introduction of polyester microfibres in the late 1980’s has enhanced the per-
formance properties of the fibre. Microfibres are generally regarded as being finer than
1 Decitex per filament [dtexpf]*. Microfibres enable fabrics to be made that have a tighter
construction, lighter weight, improved drape and greater softness similar to that avail-
able from natural fibres. The diameters of a number of common fibres are compared in
Table 1.6†.
To date the major application of microfibres has been in apparel, particularly sports
wear and active wear, where garments are required to be both windproof and permeable
to perspiration.
There have been a number of innovations in the production of polyester fibres, prin-
cipally from ‘INVISTA CORPORATION‡’ of the United States. These innovations have
principally been in the modification of the extrusion technique. These modifications pro-
duce specific variations in cross-section to enhance the performance of the fibre. The two
fibres in question are Type 704EF and Type 727. Type 704EF is sold under the trade name
COOLMAX® (refer to Figure 1.13) the registered trade name of INVISTA. This modified
cross-section improves the transport of moisture and results in a garment with improved
comfort particularly for sporting apparel. Type 727EF is marketed as THERMOLITE ®
(refer to Figure 1.14) the cross-section of this fibre has a hollow core resulting in increased
thermal insulation properties.
The world of fibre production has undergone dramatic change in the past two decades
with many long established chemical companies exiting from the fibre production industry.
These include household names such as ICI, Hoechst, Monsanto and Eastman. INVISTA
emerged from DuPont. Geographically, production has moved from North America to
Asia and other emerging economies. New fibre producers will now buy polyester polymer
as a commodity convert it into fibre and yarn, Hence, the old regime of end-to-end produc-
tion from chemicals to fibres to end products has dramatically changed forever. As a result,
* Decitex = weight in grams of 10,000 metres of yarn.

† Dr.Glover, B., October 2004. Microfibres. In SDCANZ Conference, Christchurch, New Zealand.
‡ LYCRA®, COOLMAX®, THERMOLITE ®, TACTEL ® and Antron® are registered trademarks of INVISTA.
Table 1.6 Diameters of Some Common Fibres

Fibre Diameter (micron)
Human hair 30–100
Wool 15–49
Regular polyester 15
Cotton 6–13
Silk 5–8
Microfibre 2.5–5
Super/Ultra fine Less than 2.5
Microfibre
Figure 1.13 Cross section of COOLMAX® FABRIC (Type 704EF) fibres.
new players and new fibres are emerging. Polyester will continue to evolve as it is a cheaper
polymer to manufacture than, for example, polyamide (nylon).
1.6.2.2 Polyamides
Nylon 66 is a condensation polymer made from a dicarboxylic acid (adipic acid) and a
diamine (hexamethylene diamine).
The 66 in Nylon 66 refers to the number of carbon atoms in each of the monomers.
The final polyamide contains polar amide groups along the polymer chain that allow many
hydrogen bonds to form between adjacent polymer chains. Since there are no side groups
present hydrogen bonding is maximised resulting in a highly oriented crystalline structure
with good tensile strength.
Nylon 6 is manufactured by the polymerisation of ε-caprolactam yielding the polymer
H ⎡⎣ − NH − ( CH2 )5 CO − ⎤⎦ n OH
Comparative properties of nylon 6, nylon 66 and polyester are shown in Table 1.7.
Figure 1.14 Cross section of THERMOLITE® FABRIC (Type 727) fibres.
The highly crystalline nature of the polymer reduces the tendency for the polar amide
groups in the polymer chain to attract water into the non-crystalline amorphous parts
of the chain. From Table 1.7 it can be seen that on wetting the fibre loses about 10%–20%
of its strength due to the water disrupting the hydrogen bonding between the molecular
chains. A consequence of the low moisture absorbency of nylon is its tendency to build up
static electricity. This can be overcome by the addition of hygroscopic chemicals to the fibre
surface during finishing. A novel approach to controlling static in nylon carpets was the
introduction of a fine copper thread into the carpet backing.
When exposed to sunlight over a prolonged period the fibre is degraded by oxidation
of the amide groups resulting in a substantial; lowering of strength.
The cross section of nylon can be highly varied. Some nylons are marketed with a cir-
cular cross section, while others may have a tri-lobal cross section. For example a tri-lobal
cross section is marketed as white tri-lobal nylon by INVISTA and is used extensively for
the manufacture of tufted carpets.
Cross sectional shape is determined by the shape of the spinnerets through which the
fibre forming substance is extruded (refer to Figure 1.15a and b). Nylon fibres are melt
extruded. The purpose of the different cross sectional shapes is to enhance bulk and
improve the performance of the fibre in the end products. A popular carpet fibre Antron®
from INVISTA uses a square cross section fibre with four internal holes. This shape is pro-
duced by an unusual spinneret of two ‘I’ shapes at right angles with extended bars at the
top and bottom.
Nylon can be heat sett either with dry heat or steam. The heat setting process disrupts
the hydrogen bonds which are then reformed in new positions when the fabric or yarn is
cooled. Heat setting is usually carried out under controlled stress conditions. Both nylon 6
and 66 are reasonably stable to alkalis, but are soluble in concentrated acids (formic acid)
and phenols. Nylon 6 is soluble in 4.4N Hydrochloric acid which distinguishes it from
nylon 66.
Nylon is readily dyed with acid and premetallised dyes. Pale shades can be dyed with
disperse dyes due to their very even dyeing properties. Blends of nylon with wool are com-
mon in the carpet industry, usually an 80% wool 20% nylon blend. In sportswear, blends of
nylon with elastomeric fibres such as Lycra® are common, the amount of elastomeric fibre
Table 1.7 Comparison of Features of Nylon 6, Nylon 66 and Polyester Fibres

Nylon 6 Nylon 66 Polyester
-1
Tenacity (cN tex ) Dry 39–51 40–51 35–85
Wet 36–45 35–45 35–85
Moisture regain (%) at 4–4.5 4–4.5 0.4–0.6
20°C 65% RH
Extensibility (%) 23–42 26–32 8–30
Melting Point (°C) 210–216 252–260 249–260
Specific gravity (g/cm3) 1.14 1.14 1.38
Solubility cold conc. cold conc. hot
Formic acid Formic acid meta-cresol
varying up to 20%. Nylon hosiery and support hose are common uses of the fibre. Many
swimwear manufacturers use nylon containing fabrics in their products. Tufted nylon car-
pets are a common application due to the very high abrasion resistance of the fibre.
High tenacity nylons with lower extensibility are produced by increased stretching
after extrusion. These are used in ropes, conveyor beltings and parachute fabrics.
A number of other nylon fibres noted for their abrasion resistance and high thermal
stability have been developed for use in engineering these include nylons 6.10, 7, 11 and
4.6. Dupont developed a soft silky nylon under the trade name Qiana in 1968 for the high
fashion trade. The fabric became popular in the 1970’s, it had higher moisture regain due
to the additional polar groups incorporated into the polymer. Cost may well have been the
cause of the cessation of production.
Typical trade names for Nylon 6.6 are Antron®, Rhodiastar, TACTEL ®, Ultron and for
Nylon 6 Perlon, Enkalon, Patina, Ansa and Zeftron.
1.6.2.3 Polyolefins
Polyethylene (polyethene) and polypropylene (polypropene) are manufactured by addition
polymerisation from the common petroleum products ethene and propene. The two poly-
mers are manufactured by free radical polymerisation under conditions of high tempera-
ture and pressure to low-density polyethylene (LDPE) and atactic polypropylene. Atactic
polypropylene, being an irregular structure, cannot crystallise and is therefore unsuitable
for fibres. Isotactic and syndotactic polypropylene are regular structures polymerised using
the special Natta-Ziegler catalyst which enables polymerisation to take place at a much
lower pressure (30atm) and temperature (100°C). This process allows for chain packing
and there is no possibility of side chain or branch chain formation, thus a highly crystalline
fibre, 90% crystalline in the case of the isotactic polymer, can be produced. Polyethylene
can be produced in the highly crystalline (85%) high-density form. Because of the densely
packed polymer chains the interchain forces have the effect of giving the highly drawn
fibre high tenacity and good elongation similar to that of high-tenacity nylon and polyester.
Both polyethylene and polypropylene are melt spun, into both filaments and sheet
film. When extruded into sheet the film is slit into tapes that can be handled like yarn.
Since polyolefin fibres do not absorb moisture, they are very difficult to dye and are
resistant to attack by acids and alkalis. Fibre colouration is by mass pigmentation, i.e. pig-
ment is added to the polymer melt before extrusion, however a number of polypropylene
(a) (b)
Figure 1.15 (a) spinneret for melt spinning trilobal Nylon fibres, (b) Spinneret profile for tri-
lobal Nylon fibre.
fibres have been chemically modified to make them dyeable with disperse, acid and pre-
metallised dyes. The hydrophobic nature of polyolefins results in low moisture absorbency
giving the fibre water transport properties that enables moisture to be wicked away from
the fibre surface without being trapped within the fibre itself.
Active wear when worn next to the skin takes advantage of this property. Compared
to nylon and polyester, polyethylene and polypropylene have relatively low melting points
being 135°C and 165°C respectively, which makes them unsuitable for many textile appli-
cations. Being hydrocarbons, polypropylene is flammable and burns rapidly with black
smoke.
Polypropylene when extruded as sheet film is cut into strips approximately 2–3 mm
in width. In this form the fibre can be used as a yarn and woven into an industrial weight
fabric that is used as a primary backing in tufted carpets and in sacks. Wider strips up to
50 mm wide are formed into a yarn by twisting and are used as varying types of cordage
such as ropes and fishing nets. Warp-knitted polypropylene is finding extensive use as
sun awnings, i.e. shade cloths, the polymer having been treated with a suitable ultraviolet
stabiliser to minimise degradation. There are many textile applications for polypropylene
fibres; these include carpets, furnishings, synthetic grass (particularly for tennis courts). In
apparel the very low density of the fibre makes it ideally suited for light-weight sports cloth-
ing. Thermally bonded non-woven products have been developed for medical applications,
the material is light weight, absorptive and resistant to bacteria.
Recent developments include a new fibre from BASF Performance Chemicals
MOOOTM. This polypropylene fibre is dyeable by conventional means with disperse dyes,
is light weight, and controls odour and bacterial growth. Other developments include
ultra-high-strength polyethylene fibre with the strength similar to aramides, DyneemaTM
from DSM Nederlands BV; being light weight and ultra strong it is being manufactured
into cargo nets for the airline industry resulting in fuel savings.
Trade names include Dyneema, Polysteen, Spectra and Vestolan for polyethylene, and
for polypropylene Danaklon, Asota, Downspun, Gymlene, Merkalon, MOOOTM, Novatron
and Vegon.
1.6.2.4 Polyvinyl Fibres

1.6.2.4.1 Acrylic Fibres
There are two types of acrylic fibres, both polymerised from acrylonitrile. Acrylic fibres
contain at least 85% by weight of acrylonitrile and modacrylics contain at least 35% but not
more than 85% acrylonitrile. Production of modacrylic fibre was begun by Union Carbide
in 1949.
Polyacrylonitrile had been known for some time, but for making fibres it suffered from
the defect of being insoluble in common solvents. Because its softening temperature was
close to its decomposition point it was precluded from melt spinning. It was found that
hydrotropic solvents such as calcium thiocyanate would dissolve polyacrylonitrile and
polar organic solvents such as dimethyl formamide and tetramethyl suphone were also
suitable. Since 1943 polyacrylonitrile and its copolymers have been manufactured into
fibres. These are marketed under various trade names; some of the common fibres available
have the names Orlon, Courtelle, Acrilan, Cashmilon, Dolan, Dralon, Leacril and Vonnel.
Polymerisation of acrylonitrile is an addition polymer initiated by free radicals.
CH2=CH.CN → -CH2 – CH – CH2 – CH –
CN CN n
The reaction is exothermic and large-scale polymerisations have to be carefully con-

trolled. The polymer is insoluble in the monomer and separates in an unswollen state, and
the polymerisation is normally carried out in N,N-dimethylformamide. For solution spin-
ning a molecular weight in the range 35,000–50,000 is desirable. Fibres are solution spun
either wet or dry. The polymerisation is complicated by the possibility of transfer to the
dead polymer leading to a branched chain polymer. Acrylonitrile is often polymerised in
conjunction with a comonomer such as methacrylic acid or methyl acrylate. The objective
of the comonomer is to open up the molecular structure and incorporate anionic or cat-
ionic groups to increase the dyeability of the fibre since the homopolymer polyacrylonitrile
is very difficult to dye. The polymer formed may only contain a small amount of the second
component (up to 15%) but this is sufficient to modify the properties of the fibre. Acrylic
fibres are normally dyed with modified basic (cationic) dyes yielding bright shades of excel-
lent light and wet fastness. Mass colouration of acrylic fibres is carried by the process of
gel dyeing. In this process the acrylic tow while in the gel state is passed through a dyebath
containing modified basic dyes which are absorbed very rapidly.
The tensile strength of acrylic fibres is lower than that of nylon or polyester. Typical
physical and chemical properties of acrylic and polypropylene fibres are shown in Table 1.8.
Copolymerisation with either vinyl chloride or vinylidene chloride monomers results in a
modacrylic that does not support combustion due to the presence of the chlorine contain-
ing polymers. Typical modacrylics are Kanekalon, Kanecaron, Velicron FR and Nonbur as
well as a number of manufacturers in China who promote their products as Modacrylic.
Acrylic fibres are usually blended with other fibres such as wool and cotton, in the case
of the former blend used in knitted outerwear, carpets and of the latter in furnishings.
The high bulk characteristics of acrylic make it highly suitable for the knitwear industry.
Modacrylics are often blended with other fibres to enhance their flame retardent properties
and are used in home furnishing, such as flame resistant draperies and curtains, as well as
children’s sleepwear.
Table 1.8 Properties of Acrylic and Polypropylene Fibres

Acrylic Polypropylene
-1
Tenacity (cN tex ) Dry 17.7–31.8 26.5–44.1
Wet 14.1–23.9
Moisture regain (%) at 1.0–3.0 negligible
20°C 65% RH
Extensibility (%) 20–55 15–25
Melting Point (°C) No true melting point 160–170
Specific gravity (g/cm3) 1.16–1.18 0.9–0.95
Solubility NN’dimethylformamide No known
solvent
1.6.2.4.2 Chlorofibres
Fibres manufactured from polyvinyl chloride and polyvinylidene chloride are character-
ised by their very low flammability. The fibres are either melt spun or dry spun from a
mixed solvent of acetone and carbon disulphide. Both staple fibre and filaments are avail-
able. The staple fibre can be blended with wool or cotton. Addition polymerisation of vinyl
chloride yields polyvinyl chloride which on extrusion has a disordered polymer structure,
with neither isotactic nor syndotactic stereo polymers predominating.
The fibre therefore has only a medium tenacity 20–40 cN tex-1 high elongation. At
low temperatures 100°C the fibre has a high shrinkage, up to 40%. These high shrinkage
characteristics are used to some advantage in the production of high-bulk yarns blended
with other fibres such as wool. PVC fibres do not absorb moisture so colouration by con-
ventional means is not possible. The fibres are usually coloured by mass pigmentation of
the polymer mix prior to extrusion. One of the major limitations of chlorofibres is their
solubility in chlorinated hydrocarbon and aromatic solvents, thus making them unsuitable
for dry cleaning. The inherent flame retardant properties of the fibre make it suitable for
applications in drapes and furnishings. Trade names of these fibres are Clevyl, Fibravyl,
Rhovyl and Thermovyl.
1.6.2.4.3 Fluorofibres
The most common Flurofibre is Teflon which is manufactured by the polymerisation of
tetrafloroethene gas (CF2CF2) under pressure in a special dispersion medium. The poly-
mer does not melt (CF2 – CF2) and is not soluble in all normal solvents. Spinning is from
polymer dispersion into a coagulating bath, from which the filaments are drawn and heat
treated at 385°C for several seconds then quenched in water. The PTFE polymer chain
is a densely packed orderly system with a high density of 2.3 gcm-3. The high stability to
heat, corrosive chemicals, oxidising agents and solvents is due to the inert properties of
the fluorine atom. At 300°C the fibre decomposes slowly but is stable up to 215°C. The very
low coefficient of friction, and zero moisture absorbancy make it ideal for the filtration of
liquids and gases at temperatures of 260°C.
1.6.2.5 Polyurethane Fibres

Polyurethane fibres are the basis of a group of fibres known as elastomerics, commonly
referred to as spandex. LYCRA® fibre from INVISTA is the best recognised trademark in
elastomeric fibres. Other trade names include Dorlastan, Cleerspan, Glospan and Opelon. A
chlorine-resistant elastomeric fibre for swimwear has been introduced by INVISTA under
the registered trade name XTRA LIFE LYCRA® 275B Elastane fibre. Elastomeric fibres are
characterised by their high degree of stretch and recovery, from a complex polymer system
which comprises a flexible segment for the stretch and a rigid segment giving the necessary
strength to the fibre. To a large extent the elastomeric fibre has replaced the use of natural
rubber in many apparel applications, such as the waistband of underwear.
A long polyester or polyether chain can form the flexible segments and the rigid seg-
ments are composed of a diphenyl group attached to a urethane group.
HO – (CH2 – CH2 – CH2 – CH2 – O)n – H where n = 14

The elastomers are solution-spun as continuous monofilaments that are off-white in
colour; multifilament yarns appear as single filaments fused together. The polymer net-
work is predominantly amorphous as the long flexible polyether or polyester chains are
folded on themselves. Some hydrogen bonding occurs between the urethane groups of
adjacent chains contributing to overall polymer strength. As the fibre stretches the folded
flexible segments extend up to about 500%–600% elongation, as interchain attraction is
not broken so that on release the chains recoil immediately to their original configura-
tion. Elastomeric fibres have a much higher tensile strength, 4.9–8.8 cNtex-1 compared to
natural rubber 2.2 cNtex-1 at all levels of extensibility. As the flexible segments of poly-
urethane are hydrophobic and the rigid segments are crystalline in character they do not
attract moisture so are prone to develop static electricity and are therefore difficult to dye.
Elastomeric fibres are therefore not used alone but in combination with other fibres such as
nylon, polyester or cotton. A bicomponent yarn where nylon and a polyurethane filament
are extruded together has been developed and is described in U.S. Patent 6413635.
Polyurethanes are thermoplastic and sensitive to excessive heat losing their elastic
properties. When blended with nylon, cotton and polyester in a knitted fabric heat setting
is required to stabilise the fabric structure. The main advantage of polyurethane elastomers
when used in fabrics is their stretch and recovery properties that give comfort and support
to the wearer.
1.6.2.6 Aramides
An interesting development has been the production of aromatic polyamides or aramides
(first U.S. commercial aramide fibre production was in 1961 by DuPont). In this polymer
the flexible aliphatic hydrocarbon chains are replaced by rigid benzene rings. In KevlarTM
this results in a rigid linear polymer chain with a high degree of inter-chain hydrogen
bonding.
Kevlar is made from the polymerisation of p-phenylene diamine and the acid chlo-
ride of para-phthalic acid. As the polymer has almost perfect molecular orientation as it
emerges from the spinneret, no stretching is required to increase chain alignment.
One of the outstanding attributes of Kevlar is its strength-to-weight ratio; it is five
times stronger than steel and more than ten times stronger than aluminium. It is for this
reason that it finds application as a ballistic fibre in body armour, reinforcing fibres in
sporting goods, aerospace industry, as well as protective gloves for engineering fitters.
Nomex is a condensation polymer made from m-phenylene diamine and iso-phthalic
acid. This polymer is very heat resistant and is used for making clothing for firefighters and
racing drivers.
Aramid fibres have high tensile strength, have extremely high resistance to solvent and
chemical attack and have high heat resistance, up to 300°C. The fibres do not melt but char
and decompose.
1.6.2.7 Polyvinyl Alcohol Fibres

Polyvinyl alcohol fibres are produced by the polymerisation of vinyl acetate to polyvinyl
acetate (PVA) which is then hydrolysed to the polyvinyl alcohol (PVOH) by the addition
of sodium hydroxide. The fibres are wet spun into a coagulating bath containing sodium
sulphate. The fibres are stretch drawn and heat treated to increase their mechanical prop-
erties. The fibres at this stage have poor resistance to water. Water insolubility can be
conferred by treatment with formaldehyde which introduces intermolecular and intramo-
lecular chain bonding. PVOH fibres may be made even more water resistant by the acety-
lation of the –OH groups. The treatments outlined result in fibres of varying degrees of
molecular orientation and a wide range of tenacities from 200–600cNtex-1. The presence of
the –OH groups increases the moisture regain to around 5.0% under standard conditions
but at 100% humidity this increases to 12.0%. Acetylated or formaldehyde treated fibres
are stable to cold dilute acids and alkalis, but are soluble in hot acids, hydrogen peroxide
and phenols.
Both water-soluble and water-resistant filaments are available. The water-resistant type
behaves similarly to cellulosic fibres in terms of dyeing and comfort.
Japan is the major manufacturing centre and a range of applications exist in apparel
and industry particularly as a biologically resistant substitute for cellulose in ropes and
packaging. Water-soluble PVOH fibres are used as linking threads and as support threads
in weaving, where the fibre can be removed by a simple scouring treatment.
1.6.2.8 Inorganic Man-Made Fibres

Fibres can be constructed from inorganic materials such as glass and metal for the textile
industry (refer to Figure 1.16).
Glass fibres made from sand (SiO2), sodium carbonate (Na 2CO3) and limestone
(CaCO3) is made by continuously drawing a molten stream of glass at high speed. Filament
diameter is in the range 3 –16 μm at this diameter the fibres are flexible enough to with-
stand most textile processes. Multiple filaments are processed into a yarn or roving or
as chopped strand mats. Glass fibres have high tensile strength but are brittle if folded
and can be woven into fabrics suitable for reinforcing of plastics, boat building and filters.
Fibre glass has a high ratio of surface area to weight and is ideally suited for insulation
batts. The batts are a non-woven material held together by a Novalac resin. Pink batts were
coloured using an acid dye which was dissolved in the resin bonding material. Different
types of glass are available these include ‘A’ Glass (alkali glass; window glass), ‘E’ Glass low
alkali boro-silicate glass with high resistance heat and moisture, ‘C’ Glass high resistance
to chemical attack including acids and alkalis.
Ceramic fibres are inorganic fibres developed for their very high-temperature resis-
tance. Chemically they can contain either silica or alumina or both. Boron and silicon
carbide fibres are other inorganic fibres. They can be spun from a molten material at high
speed obtaining fibres as fine as those obtained from glass. An alternative method is to
pyrolyze at high temperature the precursor fibre leaving a metallic oxide residue. These
Fibres
Natural Man-made
Carbon Glass Metal Ceramic
Figure 1.16 Sub classification of inorganic man-made fibres.
fibres can be used at very high operating temperatures over 1000°C and high chemical
stability, for hot gas filtration, thermal and acoustic insulation.
Carbon fibres are manufactured from existing man-made fibres, principally polyac-
rylonitrile. Other fibres such as rayon and aramids can be used. Manufacture involves a
three-stage heating process in which the fibre is pyrolyzed and carbonised leaving pure
carbon fibre. The final heating temperature is in the range 1500°C–3000°C.
The lustrous black carbon fibre is very resistant to heat but tends to oxide in air at
450°C. The fibre is built up of carbon atoms arranged in parallel layers orientated along
the fibre axis. Carbon fibres are very brittle and as such they are required to be embedded
in a support matrix for end–use applications. Major area of application are sporting goods
such as golf clubs and tennis racquets, where light weight and stiffness are important.
Carbon fibre resin composites are being used in the aerospace industry for skin structures
to reduce aircraft weight.
Ultra fine steel fibres can be incorporated into textile structures to dissipate static
electricity. Approximately 1 steel fibre to 100 normal yarns is the norm; under these cir-
cumstances the steel fibre is invisible. In the manufacture a bundle of fine steel fibres is
sheathed in a dissimilar alloy, spun and stretched, the sheath removed chemically to yield
the fine steel fibres.
Metallised yarns (Lurex)-polyester film is coated with a thin layer of vaporised alu-
minium. The film is slit into ribbons 0.25–1.0 mm wide and used as a decoration on gar-
ments. Coloured metallic yarns are obtained by dyeing the polyester film with disperse
dyes.
1.7 Bicomponent Fibres
Bicomponent fibres represent a special class of synthetic fibres/filaments that are made up
of two different polymers. There are a number of arrangements that can be used, but the
three most common are side-by-side, core-sheath and mixed polymer or islands in the sea
effect.
Side-by-side extrusion of two polymers of differing molecular weight or chemical con-
stitution is used to produce self bulking yarns. This type of fibre is common in the carpet
industry where one polymer has a different shrinkage temperature compared to the other.
A simple analogy is a bimetal strip where the two metals have different expansion coef-
ficients, which when heated causes the strip to curl up and twist.
Figure 1.17 Some cross-sections of bicomponent fibres (from left: hollow pie wedged, side-by-
side, trilobal tipped and segmented).
Core-sheath process can be used with a variety of different polymers in the core and
sheath to give a large number of differing surface effects. One of the most common yarns of
the core sheath type has as its core elastane, whilst the sheath may be polyamide, polyester
or other suitable polymer depending on the final effect required.
Islands in the sea effect are produced when polymers are fed together into the same
spinneret, and a fibre of mixed polymers is produced. The ratio of the two polymers may
be the same within the fibre but their distribution along the fibre length may vary. In such
a case the resultant fibre will have varying physical properties. If for example a mixture of
polyester and polyamide polymer is fed to the spinneret, then the final filaments are treated
with a solvent to dissolve the polyamide the resulting filament will in fact be a microfibre.
Bicomponent fibres can be extruded in a variety of cross-sections. Several of the cross
section shapes are illustrated in Figure 1.17.
The preceding section on fibres is only an introduction to fibres and their production/
manufacture, properties and uses. The fibre industry is complex and as a forensic fibre
examiner it is necessary to remain aware of significant changes in the industry that will
directly affect your interpretations. For more detailed information on fibres please refer to
the relevant sections in the Further Reading section of this chapter.
1.8 New Fibres
Research into new fibres to meet consumer demand is always being undertaken (for fur-
ther information on new and emerging fibres refer to Chapter 11). An example of a new
fibre is Sorona® a fibre by DupontTM. This fibre is a unique polymer based on 1,3 propane-
diol (PDO) derived from corn sugar, a renewable resource. Sorona is also known as PTT
(polytrimethylene terephthalate). The fibre has a number of advantages over both polyester
(PET) and nylon, it has a softer feel and is more easily dyeable having excellent wash fast-
ness and UV stability. It is possible to have both normal and texturised yarns made from
Sorona. Fully drawn flat filament yarns have a high degree of recovery from strain, recov-
ering 100% from 120% stretch.
1.9 From Fibre to Fabric
1.9.1 Introduction
The previous section has shown that there is a great variety of fibre types as well as indus-
tries that use them. The textile supply chain to the average individual appears complex, yet
in any manufacturing supply chain there are both simple and complex processes. Having
already discussed the origins of the fibres from natural, regenerated and synthetic sources
Spinning Weaving, Fabirc

system knitting or finishing
- Cotton tufting then
- Worsted cutting and
- Semi worsted sewing
- Woollen
Figure 1.18 Textile production chain.
it is logical therefore to now consider the more complex processes involved in the manufac-
ture of a fabric. Figure 1.18 outlines the basic processes that a fibre must go through before
it reaches the consumer.
1.9.2 Yarns
The two main types of yarn, i.e. staple yarns and continuous filament, have already been
briefly discussed in the preceding section; there are however other yarns that may be
encountered in the manufacture of textile materials. The principle of yarn manufacture is
similar for all staple fibres. The objective is to bring a series of short parallel fibres together
and by means of frictional forces created by the insertion of twist hold the bundle of fibres
together into a strong continuous length. The differences between the various spinning
systems are related to the fibre length being processed and the various mechanical opera-
tions that are required. Mechanical spinning is the drawing out of an assembly of staple
fibres to the desired thickness, inserting twist to give the necessary strength to the yarn.
Cotton System: The fibre length of cotton varies from about 2.0 cm to a maximum of
6.5 cm. The staple length variation depends on the variety of cotton and the growing
conditions. Some cotton yarns, particularly carded types (fibre that has not been
combed), contain neps. Neps are bundles of fibre that consist of immature cotton
and usually show as light spots within a dyed cotton fabric. The average diameter of
the finest cotton fibre is about 14.5–15.0 μm. Because of the fineness of cotton fibre,
very fine yarns can be manufactured. Synthetic fibres such as polyester are often
blended with cotton. In such a case the synthetic fibre must be compatible with the
cotton and the fibre length and fineness must be similar to that of the natural fibre.
Ring spun yarns are smooth where as rota spun yarns are hairier. In blends of cotton
with synthetic fibres the cotton fibre tends to lie on the outside of the yarn and the
synthetic on the inside. Blend ratios of cotton to synthetic yarns vary widely; 65%
polyester 35% cotton is a very common blend for knitted apparel, whilst 50% cotton
50% polyester is a very common blend in sheeting fabrics. Examples of other blends
that may be encountered are polyester/viscose, nylon/cotton and cotton/elastane.
Worsted System: Predominately for the spinning of wool and wool synthetic fibre
blends intended for the apparel and upholstery markets. For combing the fibre
length should be greater than 65 mm but can be as high as 176 mm. Very fine wools
35–68 mm can be combed in the most modern French combs. The combing pro-
cess increases the fibre orientation so that the fibres are more parallel. This results
in a smother, uniform yarn. Combing also removes short, damaged fibres as well
as any vegetable matter that has not been completely removed during carbonising.
(Carbonising is the process of treating the raw wool with sulphuric acid, drying,
then heating to carbonise the vegetable material in the raw wool.) A common wool
polyester blend is 55% wool 45% polyester and is common in both men’s and ladies’
suiting fabrics. A popular blend for men’s socks is wool-nylon, the nylon compo-
nent varying depending on the style and construction. The very finest yarns are
made from 14 μm wools and are intended for the fine high-quality apparel market.
Worsted spun wools intended for knitwear are usually HercosettTM treated. This
process damages the surface scales of the wool and renders the yarn non-felting.
Semi-worsted System: In many respects this spinning system in similar to the worsted
method with the exception that the combing operation is omitted. Semi-worsted
spun yarns are heavier and are used in the carpet industry as well as for hand knit-
ting and machine knitting. Average fibre length processed on this system is 100–120
mm, and for carpets fibre diameter of 25–40 μm. Yarns for hand knitting are heavier
being several singles yarns plied together to produce the required yarn count.
Woollen System: This system is different from other spinning systems in that is
involves fewer processing steps. The fibres within the yarn are less uniformly par-
allel and as such they are crossed and intertwined and held in this position when
twist is inserted. A fuzzier yarn surface with more protruding fibre ends is pro-
duced, giving rise to a higher volume per unit weight. Woollen yarns are used in
woven coatings as well as sweaters, blankets and jumpers.
1.9.2.1 Types of Yarns

The simplest of all the yarns is the singles yarn which is produced from the spinning frames.
The twist of a singles yarns is usually ‘Z’ twist. When yarns are plied together, for example
two singles yarns are doubled, the twist inserted will be in the ‘S’ direction (refer to Figure
1.19). It is not possible to twist together two Z twist yarns with Z twist as this combination
would cause the fibres in the yarns to untwist.
Synthetic fibres and regenerated fibres, as previously discussed, are produced as multi-
filament yarns and can be with or without twist. There are applications particularly in
industrial textiles where monofilament yarns are used. Fishing lines are a typical example
of a monofilament yarn. Fancy yarns such as bouclé (refer to Figure 1.20) produce a fabric
with a rough or granulated surface and may be found in ladies coats and dress-goods.
Folded yarns for hand knitted goods may be described as 2-ply, 3-ply or 5-ply indi-
cating the number of single yarns folded together. Crêpe yarns are yarns of high twist
that when woven into a suitable fabric construction produce a pebbled surface effect. The
type of yarn, degree of twist and appearance can have significant effect on the appearance,
handle and drape of a fabric.
1.9.3 Fabrics
The two most important fabric structures are woven and knit. Of the knits, weft knitting is
by far the most common. In examining fabric it is necessary to observe the yarn formation.
If there are two sets of yarns that can be deconstructed from the fabric then the fabric has
been woven. When deconstructing the fabric, if only one yarn comes out and show char-
acteristic loops then the fabric is a knit.
Figure 1.19 Z and S twist in a single yarn.
(a) (b)
(c)
Figure 1.20 Examples of different spun yarn (from top: Bouclé yarn [a], 3-fold open end yarn
[b] and single ring spun yarn [c]).
Woven fabrics are very stable and when compared to knitted fabrics are less flexible
(body hugging). Yarns from woven fabrics show a crimp across their length.
1.9.3.1 Woven Fabrics

Woven fabrics are produced on looms. The selvedge (edge of the fabric) can give valuable
information, if present, as to the type of loom on which the fabric was woven. Warp fibres
are arranged parallel to the fabric edge and the weft fibres are inserted through a variety
of methods including:
• Projectile: Weft is carried across the loom by a projectile, the yarn is cut and a small
end tucked back.
• Rapier: Weft is carried across the loom by two rapiers the yarn being passed from
one to the other, the yarn is cut leaving a fringe at the selvedge.
• Airjet: Weft is carried by a series of airjets and a fringe formed at the selvedge.
• Waterjet: This loom is very common in the weaving of 100% polyester light-weight
fabrics. The weft is carried by a jet of water leaving a selvedge.
Typical woven apparel fabrics are suitings, men’s and ladies shirtings, and skirts. A
number of woven fabrics are also found in home furnishings such as sheets, upholstery
fabrics, curtains, drapes as well as blinds and block out window dressings.
There are several basic woven fabric structures:
Plain weave is the simplest of all the woven structures and the most economical to
produce. The weave is formed by the yarns at right angles passing alternatively over and
under each other. Weft yarns interlace with the warp yarns to form the maximum number
of interlacings. The illustrations in Figure 1.21a and b show the black warp yarns interlac-
ing with the white weft yarns.
Variations in the plain weave structure are caused by varying the count (thickness)
of the warp or weft yarns. If the warp and weft yarns are the same count and the same
distance apart then the weave is considered to be balanced. If on the other hand there are
twice as many yarns in one direction than in the other an unbalanced rib plain weave fab-
ric is produced. Balanced plain weave fabrics are typical of light-weight sheer fabrics such
as georgette and chiffon made from synthetic filament yarns.
Unbalanced plain weave forms a rib appearance when the weft yarns are thicker than
the warp yarns. The thicker the weft yarns the more prominent is the rib effect. This type
of structure is more resistant to abrasion. Typical unbalanced plain weave fabrics are taf-
feta made from filament yarns, poplins and Bedford cords where the spun warp yarns are
larger than the weft yarns.
A variation on the plain weave structure is the basket weave where two or more adja-
cent warp yarns interlace with the weft yarn/s. The yarns follow the same parallel path.
The most common situation is two warp yarns interlacing with two weft yarns. The fab-
ric illustrated in Figure 1.22 is referred to as a 2/2 Hopsack, the half-basket weave is 2/1
Hopsack*.
Plain weave fabrics have no technical face or back. The face and the back appear to
be the same. If however the fabric is printed then the face and back can be distinguished.
Plain weave fabrics have a flat surface and are therefore ideal for printed designs. Because
of the large number of interlacings plain woven fabrics made from natural fibres have a
tendency to wrinkle.
Twill weave fabrics such as denim are very common. The high fibre density and excel-
lent wear resistance makes the twill construction very suitable for industrial clothing and
workwear. In this weave, two weft threads float across two warp threads with a progression
of one or two yarns either to the right or the left, resulting in a diagonal in the warp direc-
tion of the fabric. The illustrations in Figure 1.23 show a 2/1 twill to the left and to the right.
Figure 1.24 illustrates a 3/1 left handed-twill suitable for workwear.
A 2/2 twill is a very common structure for men’s and boys suiting’s. When the twill
lines appear more prominent on one side of the fabric this will be the face. Traditionally
cotton twills are left handed and wool/woollen twills are right handed. The herring bone
as shown in Figure 1.25 is a variation of the twill weave, where the direction of the twill
* RMIT University, School of fashion & Textiles, Industry Training Manual.

reverses at various intervals to form a pattern. Herringbone and twill structures are often
used in upholstery fabrics, as they are more durable and stains are less noticeable.
Satin weave fabrics as illustrated in Figure 1.26 are lustrous due to the long warp floats
on the fabric surface (referred to as warp faced). In the satin weave the warp yarn floats
over four to eight weft yarns before it passes under one weft yarn. This arrangement results
in a fabric with a smooth surface. The back of the fabric resembles twill particularly if
coarse yarns are used. It is easy to distinguish the back from the face because they are
significantly different. The combination of fine yarns together with the long floats on the
surface result in a fabric with low abrasion resistance and is prone to snagging.
Figure 1.21 Simple plain weave structures* – the white yarns are the weft and the black yarns
are the warp.
* RMIT University, School of fashion & Textiles, Industry Training Manual, Kadolph S.J., 2007. Textiles,
Upper Saddle River, NJ: Pearson Education.
Sateen weave fabrics unlike the satin weave are weft faced. The long weft floats give a
very lustrous surface to the fabric and like the satin this structure is prone to snagging par-
ticularly when fine continuous filament yarns are used. Figure 1.27a and b show example
of sateen weave on the left and both satin and sateen on the right.
Jacquard weaves are very common in high-quality furnishing fabrics, mattress-tickings
and exclusive dress designs. In jacquard weaving each individual warp thread is controlled
by a computerised mechanism that allows the production of complex designs. The brocade
fabric illustrated has both satin and twill weaves within the design area and the floats
within the design vary in length and the yarns are of differing colours.
Pile fabrics as shown in Figure 1.28a and b are three dimensional structures made by
weaving an extra set of warp or weft yarns into the ground yarns to make loops or cut ends
on the surface. Typical woven pile fabrics are towellings, velvet, chenille, and Axminster
and Wilton carpets.
Figure 1.22 Top two images (a and b) 2/1 or half-basket weave; bottom two images (c and d)
0 2/2 hopsack or basket weave.
Figure 1.23 2/1 Twill to both left and right.
Figure 1.24 3/1 Left hand twill suitable for workwear.
The pile height is usually short, approximately 10–15 mm; this enables interesting
effects to be achieved by combining cut and uncut piles, differing pile heights and com-
binations of high and low twist yarns to give interesting patterns. A common example of
an uncut pile fabric is terry towelling whereas a velvet and corduroy would be examples of
cut pile fabrics. High-quality cotton staple yarns are used for both the pile and the ground.
Polyester/cotton blends may be use for the ground where added strength is required. The
ground weave may be either plain or a twill weave.
Double cloths are two fabrics woven one above the other on the same loom, a fifth
yarn (usually in the warp) interlacing with both cloths. A true double cloth can be sepa-
rated by pulling out the yarn that holds the two layers together. Typical double cloths are
heavy coatings and some blanket type fabrics.
Tufting is a more economical way of producing a pile fabric. The yarn is carried by
a needle and inserted into a backing cloth (refer to Figure 1.29). Tufted cotton fabrics are
used as bedspreads and chenille dressing gowns. Tufted carpets are now the most common
form of floor coverings. In this case the pile yarn is inserted into a woven backing cloth
usually made from polypropylene. Following tufting, a secondary backing (woven jute fab-
ric) is attached to the primary backing using acrylic latex adhesive. This secondary backing
serves to stabilise the carpet and prevent movement of the surface after it has been laid.
During the finishing of woven fabrics chemicals such as softeners, antibacte-
rial agents, acrylic polymer resins, waterproofing and flame retardant chemicals may
be applied depending on the fabric end-use. Copper containing compounds such as
Figure 1.25 Herring bone weave.
Figure 1.26 An example of satin weave.
copper-8-hydroxyquinone and copper naphthanate have been used as rot proofing agents
on 100% cellulosic materials such as timber and canvas awnings.
1.9.3.2 Knitted Fabrics

Knitting is the formation of a fabric by the inter-looping one or more sets of yarns. There
are many advantages to knitted fabrics compared to woven fabrics. These include the abil-
ity of the knitted garment to adapt to body shape due to the elasticity of the knitted loop.
For the active individual knitted fabrics are more comfortable to wear and they retain their
appearance. A disadvantage is that because of the loop structure there is a tendency for the
fabric to sag and/or snag against sharp objects.
There are two types of knitting;
• Weft knitting where the yarn is carried back and forth or around under needles
to form a fabric. Weft knitting can therefore be divided into flat bed knitting
where the yarn moves back and forth and circular knitting where the needles are
arranged around a cylinder. The yarns move horizontally.
• Warp knitting where the yarns move vertically from a warp beam and are inter-
looped to form the fabric.
While both forms of knitting consist of loops, there are significant differences in the
performance and structure of the fabrics produced.
In woven fabrics we refer to the warp (yarns in the lengthwise direction) and the weft
yarns going across the fabric. Where knitted fabrics are concerned the loops going across
Figure 1.27 (a) Sateen weave, (b) Satin and sateen.
Pile yarns
Ground
weft yarns
Ground warp
Figure 1.28 Pile fabric construction (a – top: cut pile, b – bottom: loop pile).
the fabric are referred to as ‘courses’ and the loops in the lengthwise direction as ‘wales’ as
shown in Figure 1.30.
The following flow chart (Figure 1.31) for knitted fabrics shows the diversity of the
materials that are produced.
1.9.3.2.1 Weft Knitting There are four basic stitches used in weft knitting these are
• Plain stitch
• Tuck Stitch
• Float stitch
• Purl or reverse stitch.
Tufted surface
yarns
Primary backing
Adhesive coating
Secondary jute backing
Figure 1.29 Tufted carpet.
One course
One wale
Figure 1.30 The basic weft knit structure.
The plain stitch is the simplest of all and is the basic stitch used to produce the major-
ity of knitted fabrics. In plain the needles knit as a single set. It is possible to distinguish
between the back and the face of the fabric by the loop position as shown in the illustra-
tions Figure 1.32.
The tuck stitch is used to create a pattern in the fabric and prevent laddering. The tuck
loop is not intermeshed through the old loop but tucked in behind it on the reverse side of
the stitch.
The float or miss stitch shows the yarn floating freely on the reverse side of the fabric,
no new stitch is formed at the needle, the adjacent needle forming the new stitch. The yarn
may float over one or more needle as shown in the figure.
Purl stitches are made with courses of plain stitches alternating with courses of back
stitches. Each wale contains both plain and reverse stitches. The face and back of the fabric
appear to be the same. Two needle beds are required to produce purl stitch, the beds set at
an angle of 180°. The purl stitch is not commonly used on its own but is an essential part of
pattern construction particularly for sweaters. The most common is a 1 × 1 purl.
The simplest of all the knitted structures is single jersey. The density of the fabric is
expressed as courses and wales per cm or inch and fabric weight is expressed in g/m2.
Single jersey fabrics are characterised by their two-way stretch and if 100% cotton
are prone to distortion during laundering. Fabrics knitted on fine gauge machines have a
tighter construction and are less prone to distortion. The gauge of the machine is a measure
of the number of needles per inch, thus a 25-gauge machine would have 25 fine needles per
inch, so a finer, tighter fabric would be produced compared to a 16-gauge machine having
courser needles. Finer yarns being required when knitting on fine gauge machines. The
measurement of the number of wales per inch gives a fair indication of the machine gauge.
Weft knit Warp knit
Single knit fabric Double knit fabric Simple Complex
Single Double Tricot Lace

Jersey Interlock Tulle Power net
Jacquard jersey Rib Raschel knits
Terrycloth Jacquard knit
Fake fur Purl knit
Fleece
Figure 1.31 Fabrics and their construction types.
Figure 1.32 Loop structures for single knit.
Rib knit fabrics are the next most popular to single jersey knit fabrics. The simplest rib
knit is the 1x1 rib and is made of face wales and back wales, forming lengthwise ridges on
both sides of the fabric by pulling adjacent stitches first to the face and next to the back of
the fabric (refer to Figure 1.33a and b).
The rib knit has vertical cord appearance on the face and back of the fabric and is
thicker than a single jersey. Unlike single jersey, rib knits do not curl at the edges but they
can unravel or run. Major uses of the rib knit are socks, neck bands and cuffs of knitwear,
underwear and garment trims.
Double knits are where both sides of the fabric appear the same as the face of a jersey
knit. It is a firmer fabric but stretches in a similar manner to plain jersey. Many interlock
fabrics are either plain or printed; they do not curl and can be unravelled from only one
side. The easiest way to identify a double knit is to look at an edge of the fabric parallel to a
course. If all loops point at one direction it is a single knit, if some of the loops point toward
the front and some toward the back it is double knit. Using a pick glass or microscope it
is possible to distinguish between rib and interlock. If the loops are directly opposite each
other it is interlock, but if the loops are not directly opposite each other it is a rib. Figure
1.34 shows the face and back of three weft knitted fabrics described above.
1.9.3.2.2 Warp Knitting Warp knitted fabrics have a wide variety of applications
these include home furnishings, lingerie, medical products and technical fabrics such as
Figure 1.33 Loop structure for 1 × 1 Rib Knit (a - top; schematic, b - bottom; fabric example).
geotextiles, sun awnings and shade cloths. The loops are formed by the movement of guide
bars in conjunction with specially designed needles.
Rachel knit is one of the most common warp knit structures. This is a lace-like open
construction with a heavy textured yarn held in place by a much finer yarn. An example is
shown at Figure 1.35. Rachel knits can be made in a variety of types ranging from fragile
to coarse and usually have limited stretch. As a general rule Rachel knits are composed of
heavy yarns and large designs resulting in complex open spaces with surface effects.
Tricot knit utilises two parallel lines of yarns interloped, the loops being formed in
roughly the warp direction. Compared to weft knits such as jersey, tricot knits are dimen-
sionally stable. Tricot has a number of industrial applications, including as a base material
for coating and laminating, filtration and automotive seating. Tricot fabrics are also used in
active-wear blends of polyester or nylon together with an elastomeric fibre such as LycraTM.
Another popular application for tricot fabrics is women’s undergarments; a variety of natu-
ral and synthetic fibre types can be used. Brushed tricot has become popular for outdoor
wear, particularly jackets; a pile is produced on the surface of the fabric by brushing, giving
the garment added warmth, comfort and softness. The structure for tricot is close-knit, the
yarns running lengthwise with an interlooped yarn pattern. One side will feature fine ribs
running in a lengthwise direction, while the other side will feature ribs that run crosswise.
There has been considerable investment in warp knitting technology and the leading
manufacturers of these machines continually developing newer and more sophisticated
methods of producing textile fabrics.
1.9.3.3 Pilling
Pilling is a fault that can occur in both woven and knitted fabrics. Pills are formed by the
entanglement of fibres that migrate from the yarn (refer to Figure 1.36a and b). It is usually
brought about by some form of mechanical action. The mechanical action can be during
wear, laundering or dry cleaning. The degree of pilling is not a measure of fabric age, rather
one of performance. In some cases pilling may occur very early on in the life of a garment.
A typical example is illustrated; the fabric is a polyester wool woven suiting where severe
fibre migration from the yarns has occurred. The most probable cause is poor finishing of
the fabric. This type of fabric is usually given a singeing treatment to burn off surface fibre,
followed by brushing and shearing to further clean the fabric surface. Small beads of poly-
mer may show on the fabric from singeing under microscopic examination.
Interlock Rib knit Jersey
Fabric face
Fabric back
Figure 1.34 Showing face and back of three knitted fabrics. (Adapted from Textile Training
Through Technology, Canada, DVD.)
Figure 1.35 Example of a Rachel knit.

Other factors that contribute to pilling are yarn twist, fibre length, yarn smoothness
and yarn doubling. Fabrics made from yarns of low twist and short fibres are very prone
to pilling. Sweaters made from woollen spun yarns may pill in areas of high friction, par-
ticularly under the arms.
Weft-knitted fabrics can show signs of fibre migration forming fuzz which with con-
stant wear can produce pills which ultimately break off. The example illustrated in Figure
1.37 is from a polyester 65% cotton 35% T-shirting. The fuzzy surface developed after only
several washes and the extensive fibre migration from the yarns together with the mechan-
ical action has commenced the pill formation. Rotor spun yarns tend to be more hairy and
thus fibres migrate to the surface. As previously mentioned, in polyester cotton rotor spun
yarn the polyester has a tendency to lie in the centre of the yarn surrounded by the cotton.
As polyester is the stronger fibre, during mechanical action the polyester balls up forming
the pill which surrounds migrated cotton fibre.
1.9.4 Non-Woven Materials

The non-woven textile material sector is an increasing market segment for textile prod-
ucts. But what is a non-woven textile? INDA (the U.S. Association of Non-woven Industry)
define a nonwoven as a ‘sheet, web or batt of natural or man-made fibres or filaments,
Figure 1.36 Woven Polyester wool fabric showing fibre migration and pill formation.
excluding paper that have not been converted into yarns, and are bonded to each other by
any of several means’.
Nonwovens consist of natural or synthetic fibres deposited at random and bonded
together by various processes, and can include stitching. Some of the processes used to
create nonwoven fabrics are:
Needling is a web of fibres is passed through a needling machine which consists of a
series of barbed needles moving up and down at high speed that entangle the fibres pro-
ducing a compact mat of material.
Needle felts produced by this method are commonly used as sound insulators in auto-
mobiles and as boot liners, an example is shown at Figure 1.38. More open needled fabrics
are used as insulation in house ceilings.
Spun bonded fabrics are produced by extruding filaments of fibre onto a cooled cyl-
inder or conveyor belt; the resulting web of fibres is then calendared to achieve the neces-
sary bonding of the individual filaments (refer to Figure 1.39). The main applications are
geofabrics, filtration fabrics and some household wipes.
Stitch bonding fabrics are where layers of fibre webs are stitched together to form a
stable fabric. Blanket underlays and hygiene product covers can be produced by this pro-
cess (refer to Figure 1.40).
Chemical or adhesive bonding is a variety of different fabrics are made by this
method. The web of fibres is impregnated with a suitable resin adhesive, either by foam or
spray. Typical adhesive compounds are ethylene vinyl acetate or styrene butadiene emul-
sions. Once impregnated, the fabric is dried then passed through a heated calendar to
cure the resin and impart a design onto the fabric. Typical end uses are disposable bags as
used by airlines and shopping bags. Household wipes made from viscose rayon fibres are
another type of adhesive bonding. In this case a design containing the adhesive is printed
onto the fabric as shown in Figure 1.41a and b and then cured, resulting in a product with
high water absorption properties.
Thermal bonding is a process in which fibres of differing melting points are carded
together to form the web of fibres. For example Nylon 6 (m.pt. 215°C–217°C) in combina-
tion with polyester (m.pt.250°C–260°C), the web of fibres is then heat treated, the nylon 6
Figure 1.37 Knitted fabric showing surface fuzz and start of a pill.
binding the whole web of fibres together. Polypropylene can be used alone or in a mixture
with other fibres as it has a lower fusion temperature of 125°C. This process has been used
successfully to produce car carpets and automobile door liners.
Hydro entanglement – In this process the web of fibres is subjected to high pressure
water that physically creates fibre entanglement. A wide range of products are made by this
process the main areas of application being medical textiles, for example, medical wipes,
incontinence pads and medical swabs are just some of the products produced.
1.9.4.1 Leather
Hides and skins are treated to prevent putrefaction by tanning. A number of processes
can be used; these include vegetable tannins, chrome tannage, aluminium tannage, syn-
thetic tannins or oil tannages. Different types of leather are produced by each method.
Many different hides and skins are used; garment leathers are often made from chrome
tanned sheepskins, pigskins, kangaroo and also cattle hides. As a general rule garment
leathers are lightweight with a substance (thickness) of approximately 0.8 -1.1 mm.
Chrome tanned kangaroo leather is very light weight and extremely strong and can be
found in high class sporting shoes. Some kangaroo leathers, because of their high ten-
sile strength, are made into thongings and leather lacings. The majority of upholstery
Figure 1.38 Needle felt.
Figure 1.39 Spun laid and needled non-woven.

Figure 1.40 Dry laid stitch bonded fabric.
leather is made from cattle hides and is usually chrome tanned. Pigskin has a distinc-
tive grain structure and can be found in shoe linings as well as suede leathers.
The most common tanning method is tannage with basic chromium sulphate. Leathers
produced using this method are used extensively in shoe uppers, upholstery and garments
as well as hair on floor rugs such as sheepskins.
Tannins occur to varying degrees in nearly all plants where the amount of tannin is
high and large quantities of plant material are available for commercially exploitation,
tannins are extracted and processes into useable material. The main vegetable extracts
used are, mimosa (from wattle bark), quebracho (red-brown coloured leather), myroba-
lans (dried fruit of Indian tree) gives a light yellow-brown colour to leather. Chestnut is
favoured for the manufacture of sole leather. Harness and saddler leathers are often veg-
etable tanned.
Aluminium tannage gives a white leather and has been used for the manufacture of
cricket ball and baseball leathers.
In all cases leather is made soft by fat-liquoring, that is, the application of a sulphated
or sulphonated oil during the latter stages of wet processing usually dyeing. This oil lubri-
cates the fibre bundles and prevents hardening of the material.
Leathers can be dyed or pigmented. Acid and direct dyes are the most common dyes
used for dyeing or colouring leather. In the manufacture of white leathers titanium dioxide
is drummed into the leather during the fat-liquoring process.
Modern Chamois leather tannages are combinations of aldehydes such as formalde-
hyde or gluteraldehyde together with oxidising oil such as cod oil. The oxidation of the oil
gives rise to the yellow colouration on the leather. A small amount of an acid dyestuff may
be added to maintain colour consistency from batch to batch.
In shoes the uppers may be either full-grain or corrected-grain. In corrected-grain
leathers the grain surface is buffed off in a rotary sanding machine using emery paper to
give a more uniform surface. Acrylic emulsion resins are then applied to replace the sur-
face, a pigmented resin coat is applied and finally a lacquer topcoat based on either nitro-
cellulose emulsion or polyurethane emulsions.
Figure 1.41 Dry laid print bonded viscose household wipe (a - left; overall shot, b - right; close
up image of the structure).
1.10 Conclusions
The depth and variety of textile materials, fibres, fabrics and end uses discussed in this
chapter shows the importance of textile manufacture in our ever-changing world. In our
daily lives we invariably come into contact with a textile material, whether it be apparel,
furnishing, medical or automotive, and the textile chain will continue to dominate our
lives.
The large volume of textiles surrounding us means that they will be directly and/or
indirectly involved during the commission of crime. It is therefore necessary to under-
stand the production, structure and properties of fibres and fabrics.
This chapter has only been an introduction to fibre types and fabric constructions. In
reality forensic examiners will normally only encounter the most common fibre and fabric
types. However, knowledge of the less common varieties is critical in those instances where
they are encountered in case work and the knowledge of fabric structures is particularly
important in examinations concerning textile damage.
Acknowledgments
The authors wish to acknowledge the contribution of images by Dr Lyndon Arnold and
drawings by David Hemmings and the help and contributions of colleagues at RMIT
University School of Fashion & Textiles, Dr Lyndon Arnold, Ms Ina Konopov and Mr
Suniat Islam.
Bibliography
American Fiber Manufacturers Association, 2009. Fibre Organon, 2009, 80(6), 95–112.
East, A. J., 2005. Polyester fibres. In J. E. McIntyre (Eds.), Synthetic Fibres: Nylon, Polyester, Acrylic,
Polyolefin (Chap. 3, pp. 95–166). Cambridge: Woodhead Publishing and CRC Press.
Gordon, S., 2009. Identifying plant fibres in textiles: The case of cotton. In M. M. Houck (Eds.),
Identification of Textile Fibres (Chap. 13, pp. 239–258). Cambridge: Woodhead Publishing and
CRC Press.
Koslowski, H.-J. and Main Kassel, F., November 2008. Fasen für Vliesstoffindustrie aktuelle
Markttrends.
Further Readings
Journals and Catalogues

Chemical Fibres International
Colouration Technology. Bradford: Society of Dyers & Colourists.
International Textile Bulletin, in International Textile Services. Schlieren Zurich: ITS Publishing.
Textile Research Journal
Textile Progress. Manchester: Textile Institute.
Textile Technology Catalogue. Cambridge: Woodhead Publishing.
Textbooks
Anon., 1975. The Identification of Textile Materials (7th ed.). Manchester: The Textile Institute.
Bast, A. J., 2005. Polyester fibres. In J. E. McIntyre (Ed.), Synthetic Fibres: Nylon, Polyester, Acrylic,
Polyolefin (Chap. 3, pp. 95–166). Cambridge: Woodhead Publishing and CRC Press.
Cook, G. J., 1984. Handbook of Textile Fibres–Man-Made Fibres. Herts, UK:Merrow Publishing Company.
Cook, G. J., 1984. Handbook of Textile Fibres–Natural Fibres. Merrow Publishing Company.
Eichhorn, S. J., Hearle, J. W. S., Jaffe, M., and Kikutani, T. (Eds.), 2009. Handbook of Textile Fibre
Structure, Volume 1: Fundamentals and Manufactured Polymer Fibres. Cambridge: Woodhead
Publishing and CRC Press.
Eichhorn, S. J., Hearle, J. W. S., Jaffe, M., and Kikutani, T. (Eds.), 2009. Handbook of Textile Fibre
Structure, Volume 2: Nature, Regenerated, Inorganic and Specialist Fibres. Cambridge: Woodhead
Fan, J. and Hunter, L., 2009. Engineering Apparel Fabrics and Garments. Cambridge: Woodhead
Frank, R. R., 2005. Bast and Other Plant Fibres. Cambridge: Woodhead Publishing and CRC Press.
Hongu, T. and Phillips, G.O., 1997. New Fibres (2nd ed.). Cambridge: Woodhead Publishing and
CRC Press.
Houck, M. M. (Ed.), 2009. Identification of Textile Fibres. Cambridge: Woodhead Publishing and
CRC Press.
Jeffries, E., 1971. Merrow Monograph, Bicomponent Fibres. Watford, UK: Merrow Publishing.
Kadolph, S.J., 2007. Textiles. Upper Saddle River, NJ: Pearson Education.
Mather, R. R. and Wardman, R. H., 2011. The Chemistry of Textile Fibres. Cambridge: RSC Publishing.
McIntyre, J. E. (Ed.), 2005. Synthetic Fibres: Nylon, Polyester, Acrylic, Polyolefin. Cambridge:
Woodhead Publishing and CRC Press.
McIntyre, J. F. and Daniels, P. N., 1995. Textile Terms & Definitions. Manchester: Textile Institute.
Textile Damage Interpretation
JANE HEMMINGS
DEBRA CARR
JAMES ROBERTSON
2
Contents
2.1 Introduction 61
2.2 Recognition Level Examination 63
2.2.1 Causes of Textile Damage 64
2.2.2 Documentation 65
2.2.3 Packaging and Preservation 65
2.2.4 Briefing 67
2.3 1st Level Examination 68
2.3.1 Morphological Information to be Described at the Textile Product Level 69
2.3.2 Morphological Information to be Described at the Textile Level 70
2.3.3 Morphological Information to be Described at the Yarn Level 71
2.4 2nd Level Examination 73
2.4.1 Detailed Examinations 73
2.4.1.1 Morphological Information from the Fibre Level 74
2.4.2 Weapon/Implement Effects 75
2.4.2.1 Knives 76
2.4.2.2 Scissors 77
2.4.2.3 Blunt Implements 77
2.4.3 Textile Effects 77
2.4.4 Simulations 78
2.4.5 Limitations and Cautions 81
2.5 Testing 81
2.6 Interpretation, Reporting and Standards 82
2.7 Conclusions 83
Acknowledgements 83
Glossary 84
References 84
Further Reading 85
Appendix 2.1 Classification Scheme for Assessing Damage 86
Appendix 2.2 Parts of a Knife 87
2.1 Introduction
Textile damage examination refers to the examination of any textile that may have under-
gone damage and usually has legal implications. Another term commonly encountered
is textile severance morphology. Modern terminology favours the use of the word fabric
(then woven fabric, knitted fabric or nonwoven fabric) over the word textile (which now
61
means a woven fabric only); however, practitioners are more familiar with the word textile
and as such, the term textile is used throughout this text to encompass woven, knitted and
non-woven fabrics. For an overview of fibres, yarns, fabrics and their construction refer to
Chapter 1.
Damage to textile products may result from:
• Edged or blunt weapons

• Puncture type weapons
• Chemicals
• Heat sources
• Mechanical forces
• Animals
• Environmental factors
• Age (‘wear and tear’)
The event that causes the damage may be criminal in nature (or associated with the
commission of a crime) or it may be due to events that were of a non-criminal nature such
as accidental or legitimate damage.
The purpose of examination of textiles in a forensic context is to determine, where possible,
the cause of the damage, the mechanism of creating the damage and, where relevant, the imple-
ment that may have caused the damage. Issues faced when examining damage to textiles include:
1. There are unknown and therefore unquantifiable actions that lead to the creation
of the damage in the first place.
2. There are limited publications documenting the procedures or a standard protocol
for identifying the cause of textile damage. This is partially due to the wide vari-
ety of both textiles and weapons/implements, making it difficult to ‘standardise’ a
single specific protocol.
3. Visual assessment forms a large part of the examination of textile damage; results
involve observations that are qualitative in nature (and between examiners are
open to variation).
The examination of damage to textiles should normally not be conducted in isolation

from other evidence types and case information. For example, the collection of trace mate-
rial from possible implements as well as from the textile is important as well as the descrip-
tion (and testing) of staining. Relevant background information about the case is critical
in formulating hypothesis testing/simulation experiments. Other useful information can
include the pathologist’s report, victim and suspect statements (although it is best not to
see these until after all examinations are completed), CCTV footage and general crime
scene notes and photographs.
Increasingly in forensic laboratories, there is a shift towards triaging exhibits to man-
age evidence from exhibits. In the case of textile damage, triaging of the exhibits focuses
on what level of examination is required. The authors consider that there are four different
types of assessment and examination.
Textile Damage Interpretation 63
1. Recognition and assessment: This could be achieved by a first responder at the

scene (police officer, emergency services), first responder at hospital, crime scene
examiner or laboratory examiner. This is about recognising that damage exists
and ensuring that its integrity is preserved.
2. 1st level examination: This level is about describing the textile product and the
damage, identifying the type of damage and providing exclusionary-type inter-
pretations (e.g. the knife could not have caused the damage observed).
3. 2nd level examination: This level is about identifying complex damage and com-
paring the damage with a weapon(s) with a view to providing inclusionary-type
interpretations (e.g. the knife could have caused the damage observed).
4. Specialist testing: This is higher-level specialist examination that would fall out-
side of the expertise of the forensic examiner and would include interpretation of
specialist garments, textile testing by an accredited laboratory and biomechanical
interpretations.
At each level of examination it is important to be systematic, thorough and compre-

hensive. Whilst there is no checklist that would suit each case circumstance, informa-
tion such as that provided in the Appendix can assist examiners to take a comprehensive
approach.
The aim of this chapter is to introduce the concept of four levels of examination that
a textile can undergo as well as highlight the limitations of textile damage and point to
those instances where the examiner will need to seek outside expertise. Examiners should
recognise their level of expertise and not conduct examinations that are beyond their dem-
onstrated skill level and proven competence.
2.2 Recognition Level Examination
Recognition of the value of textile damage examinations is required early in an investiga-

tion in order to preserve it as potential evidence. Once recognised, the evidence requires
correct documentation, collection and packaging to allow interpretation by the textile
damage examiner.
Different groups of people may be involved in this recognition stage; predominantly
they will be the first responders to the crime or crime scene, such as police, paramedics,
medical staff and crime scene examiners. In instances where the crime has resulted in
death, recognition of textile damage should be relatively straightforward. However, often
the textile in question is not in a static state at a crime scene but is on a victim who requires
medical treatment either at the scene or as a walk-in at a hospital emergency department.
The primary role of the first responder in relation to textile damage should be limited to
documentation and preservation. If it is a crime scene examiner this will be through pho-
tographs, sketches, crime scene notes and packaging, but if it is medical personnel it would
ideally be through removal of clothing in a manner that doesn’t disrupt damage that is
already present (through cutting or excessive staining by blood) and noting anything that
has been done to the garment. The removal of clothing (by first responders) is the point
where there is the greatest potential for disruption or loss of the damage.
In most instances the fact that textile damage will be of interest is obvious, however
there are some circumstances where it will not be immediately obvious that examination
of textiles for damage will be of assistance to an investigation. A number of indicators can
assist with recognising potential textile damage evidence. These include:
• Crime scene and/or first responder indicators

• Visible cuts, tears or punctures to a textile that are above normal wear and tear.
• Visible thermal damage to a body and/or textiles.
• Likely exposure to the elements and/or animals.
• Presence of blood indicating a possible underlying injury beneath the textile.
• Clothing or other textiles skewed from their normal position.
• Implement still left in the body.
• Victim and/or Witness statement indicators
• Allegation/s of assault.
• Witness statement indicates textile damage (i.e. eye holes cut into a balaclava,
suspect clothing torn, etc.).
• Allegation of textile failure (such as failure of a safety harness through tearing
or abrasion).
• Intelligence indicators
• Section of textile or yarn found at primary or secondary crime scenes indicat-
ing that a textile may have been torn or cut during the commission of a crime.
• The victim was bound (i.e. potential physical fit of binding materials).
• Hit and run incident (damage may be microscopic and not visible to the human
eye such as compression of the textile or the impression of the textile in the
paint of the vehicle).
• CCTV footage shows a garment where the possible damage may make that
garment unique.
2.2.1 Causes of Textile Damage

The first responder needs to be aware not only of textiles that may have undergone dam-
age during the commission of a crime but also what may have caused this damage. This is
particularly important for those sources that might not otherwise be obviously involved.
Implements/weapons that can create textile damage include:
• Edged weapons: Predominantly knives but can include, axes, scissors, machetes,
razor blades, broken bottles or crockery, shard of glass from a broken window pane
• Impact weapons: Baseball bats, golf clubs, bits of wood, blunt knives
• Puncture type weapons: Screwdrivers, blunt knives, teeth
• Projectiles: Bullets, arrows, crossbow bolts, etc.
• Chemicals: Acids, bases, solvents, oxidisers
• Thermal sources: Irons, stoves, ignitable liquids, burning particles (embers, ciga-
rettes, sparks, electric arcing), heated metals, heaters (gas/bar, etc.)
• Mechanical forces: Tearing by hand, caught in machinery, abrasion type damage
from falling or dragging on rough surfaces, tearing by catching on edges of fur-
niture, etc.
• Animal sources: From teeth or claws of animals

• Environmental sources: Fading, degradation
The first responder should endeavour to document and preserve any observed dam-
age with the least amount of disruption. First responders may not necessarily be crime
scene examiners and it is appreciated that for emergency workers that the primary role will
not be the preservation of evidence. The following information relates specifically to the
documentation, packaging and preservation, identification of causes of textile damage and
briefings conducted by crime scene examiners.
2.2.2 Documentation
It is important to document all damage in situ through photographs (with scales), sketches
and notes prior to any movement or packaging of the textiles of interest. This is particu-
larly important for textile damage as it may provide information such as alignment of
clothing, number of layers of clothing, fit of clothing and amount of any staining. The use
of scales is to give an indication of the damage as seen but may not give a genuine measure
of the size of the damage.
As the process of examination of a textile fragment is focused on a physical fit, the
alteration or loss of parts of the textile may hinder this process. For this reason, detailed
documentation of textile fragments is required.
The crime scene may be the best opportunity for interpreting the textile damage as
changes to the textile through handling and movement may render interpretation at the
laboratory impossible. All too often, the interpretation of damage is rendered more dif-
ficult, or impossible, by leakage of blood onto areas of damage post removal of the body
and/or inadequate procedures for the removal of clothing at the morgue and subsequent
handling by crime scene officers.
2.2.3 Packaging and Preservation

Preservation of the textile is critical as even small amounts of handling can alter the dam-
age. It is important to note any alteration to the damage or subsequent damage that occurs
during the removal and packaging process.
Things that can occur that have the potential of altering the textile damage include:
• First responder assistance: This would usually be by way of cutting the textile worn
by an injured person (refer to Figure 2.1). Statements from any first responder will
greatly assist the examiner and, where possible, will form part of the briefing to
the laboratory examiner.
• Removal of an implement: Care should be taken when removing an implement
from a garment. This is because this action has the potential to disturb the fibres
and/or yarns along the damaged edge, and may increase the damage through
furher cutting. Yarns or fibres must be recorded and collected prior to any removal.
• Excessive handling during removal and packaging: The alignment of the yarns and
fibres at the edge of damage is very important for interpretation and movement
of these can occur quite readily. Care should be taken when handling textiles to
minimise disruption to the alignment of these fibres and yarns.
Figure 2.1 Example of hospital type damage (circled in red) where first responders have cut
through area of possible textile damage of interest (indicated by red line in inset image).
• The drying of staining: This can lead to changes in the damage as the garment can
contract during the drying process (Gore et al., 2006). In some instances it may be
important to examine the textile damage as soon as possible before drying of the
staining occurs.
• Ongoing chemical reactions: Initially observed damage may develop or progress
further due to chemical reactions. Where it is not feasible to inhibit the reactions,
high importance should be placed on photographing and noting damage at the
scene.
Ideally, textile damage examinations would be conducted with the garment in situ,
thus avoiding any issues of removal and packaging. However, in most circumstances, this
is not practically possible and the crime scene examiner will inevitably need to package
the textile for transportation back to the laboratory. The packaging of damaged textiles is
a delicate process, and it is essential that disruption to the physical form of the damage is
limited. In order to preserve textile damage, the following points should be adhered to as
closely as possible:
• Packaging: Ideally each garment is packaged individually.

• Folding: It is not ideal to fold the garment. If however, folding cannot be avoided
(i.e. transport from scene / morgue to laboratory) then folding should be such that
no fold marks will cross areas of textile damage. Consideration should also be
given to rolling the textile if suitable.
• Storage: The textile should be stored in an open and flat condition. In some instances
support may need to be provided by a substrate such as cardboard. Long-term stor-
age can lead to degradation from the storage packaging and consideration should
be given to the use of museum type products such as acid-free card and bags.
• Drying: If the garment requires drying due to biological staining, then this should
be done with consideration of biological evidence (including evidence harvesting,
contamination and biological hazard) and in a flat condition.
• Laundering: Under no circumstances should the textile be washed. The physical

action of washing and drying can severely distort any textile damage and it has
been shown that laundering can alter both the visible and microscopic damage
(Daroux et al., 2010).
In preserving the chemical form, all examiners in the examination chain need to be
aware that fibres, particularly those of a natural origin, such as cotton or wool, are suscep-
tible to attack by microbes and insects and steps should be taken to protect the textile in
the following ways:
• Microbial damage can occur through the growth of bacteria and fungi on the tex-
tile. These thrive in damp conditions, hence the textile should be dried thoroughly
for any long-term storage. It should be noted that drying only removes the mois-
ture and will not kill the bacteria or fungi. For long-term storage, it may be neces-
sary to apply a biocide such as exposing the textile to ethylene oxide gas or flash
freezing the textile in liquid nitrogen and allowing it to thaw.
• Insect damage to textiles, predominantly the protein fibres (e.g. wool), is through
certain moth larvae and beetles. Correct packaging, drying and cool storage will
eliminate or minimise the potential damage to the garment. In some instances, it
may be necessary to include an insecticide in the packaging for long-term storage.
If the textile cannot be removed to the laboratory in its entirety (such as a large lounge
or mattress), and must be cut, then the cut edges need to marked in such a way that they
are easily identified. The orientation of the textile damage is also required (i.e. up direction
on a piece of textile removed from a hanging curtain as well as the ‘echnical face’ of the
textile).
2.2.4 Briefing
Briefing of laboratory based staff who will undertake the more detailed examination is
particularly important in the examination of textiles for damage. The provision of doc-
umentation such as CCTV footage, pathologist/doctor’s report, and victim and suspect
statements is more important for textile damage examinations than for other types of fibre
examination as this information is generally required for formulating hypotheses for test-
ing. The effect of context bias must be considered and, whilst this is critical information for
a textile damage examination, there is debate over when in the examination process this
should be provided to the examiner to minimise bias.
The handover from the crime scene examiner to the textile damage examiner is an
important step. This should not be considered a single event but should include information
sharing as the case develops and new information/evidence comes to light. Documentation
as well as information that should be passed on includes:
• History of the garment: Washed either intentionally or inadvertently, found at the

garbage tip, retrieved from the clothesline, found in a dishevelled condition then
folded for packaging, intelligence about the age of the textile, cut or torn in the
process of removal and packaging.
• Intelligence at hand: Victim/suspect/witness statements, CCTV footage, pathol-

ogist reports, pertinent weather information (in the instance of textiles found
exposed to the elements), and relevant photographs and notes from the crime scene.
As the examination of damage to textiles is not undertaken in isolation from other

evidence types and case information, it is important that the textile damage examiner is
also aware of other evidence (associated with the textiles and implements) that is to be con-
sidered. This could include ignitable liquids that may be present on the textile or chemical
and biological trace evidence that may be present on the implement and/or textile. The act
of examining textiles for other evidence can lead to disruption to the damage of interest
and therefore it is important that the history of the garment continues to be documented
throughout the laboratory examinations.
2.3 1st Level Examination
The role of the laboratory examiner at this stage is to comprehensively describe the tex-
tile, including the construction and, of course, the textile damage of interest through case
notes, drawings and photographs. The skilled textile damage examiner at this level needs
to have a knowledge base that extends further than that of the standard fibre examiner into
yarn and garment construction. An introduction to textile construction and fibre compo-
sitions can be found in Chapter 1.
The examiner needs to be aware of other evidence types that may be present on a textile
and consider whether the particular textile requires any other type of evidence collection
or testing. During this consideration (and if necessary consultation with other examiners)
the order of examination needs to be determined, e.g. damage examination conducted
after chemical trace collection but before biological trace collection. Collection of trace
biological material is an exception. Where it may be necessary to recover trace biological
material from near an area of damage this should be done by both the biologist and the
textile damage examiner.
At this level the textile damage examiner will often be able to determine (where
possible):
• That the damage is considered above normal wear and tear

• Any physical fit of textile sections
• That the damage is cut, tear, puncture or combination
• That the damage is recent
The broad classifications of damage that can be identified at this level are:
• Cut: Damage caused by a sharp-bladed implement (refer to Figure 2.2)

• Tear: Damage caused by a pulling force (refer to Figure 2.3)
• Puncture: Damage caused by a blunt implement with a pushing force
• Abrasion: Damage caused by frictional forces
• Thermal: Damage caused by a heat source
Figure 2.2 Cut yarns in an open lace like structure.
Figure 2.3 Tear type damage to underwear.
In order to identify these damage types, it is necessary to describe the morphology of

the damaged textile, yarns and fibres and relate it to the textile construction. The majority
of the information that will help to determine the cause of the damage is observed in the
morphology of the textile, yarns and fibres at and near the textile damage of interest. The
examination of this morphology is predominantly undertaken by the naked eye and at
relatively low-level magnification (such as that obtained by stereomicroscope).
2.3.1 Morphological Information to be Described

at the Textile Product Level
This information will include:
• General condition of the textile product

• Location and orientation of the damage of interest
Figure 2.4 Fading of a garment (possibly due to excessive light exposure).
In addition to identification of the textile damage and determination of its freshness, it

is important to identify and describe the conditions and characteristics of the textile gener-
ally. This should include:
• Areas of normal wear and tear: Damage that is likely to have occurred through
normal use of the textile and includes the thinning of the textile that leads to holes
or fraying, fading (refer to Figure 2.4), unravelling of hems and seams, snags, abra-
sion and runs.
• Presence and degree of staining: It is important to be aware that the presence of
staining can hinder the identification of the likely cause of the damage as a stain
tends to affect the morphology of the damaged edge. A stain may cause fibres at
the yarn end to adhere to one another and as the stain dries the relative position of
those fibres may change further.
• Labelling information: Manufacturer’s information in relation to fibre composition.
2.3.2 Morphological Information to be Described at the Textile Level

The information that can be recorded at this level can be quite varied as it is dependent on
the damage and textile encountered. The information can be documented through a mix
of drawings, photographs and notes. It is important that a systematic and comprehensive
approach to recording textile damage is undertaken. The information below may lend itself
to the development of a checklist; Appendix 2.1 outlines other useful information.
• Shape
• Dimension/s
• The direction of the line of damage in relation to the textile (warp and weft of
woven textile, and wale and course of knitted textiles)
• Distortion of the textile immediately surrounding the damage (this would include
yarn displacement such as tightened yarns or buckling)
• Presence and degree of curl
• Loose yarns
• Dimensions of any missing section of textile
• Discontinuities or stoppages typical of scissor cuts
• Changes to the normal yarn spacing
• The relative positions of the damaged yarn ends (e.g. well aligned in the case of cut
damage or ruptured in the case of tear or puncture damage)
2.3.3 Morphological Information to be Described at the Yarn Level

• Any matting of the fibre ends within the yarn
• Fraying due to untwisting of yarn
• Melted or burnt yarn ends
• Snippets in a knitted textile
• Position of the fibre ends within the yarn (planar array or bearded yarn ends)
• Planar array is where a clean cut leaves all the fibre ends in the same plane of
the yarn
• A bearded yarn occurs during tearing where the fibres break at different posi-
tions in the yarn
The same damage mechanism (be it a cutting action or a tearing action) can produce
different damage morphology dependent on the construction of textile (and to some degree
the different chemical composition of the fibre itself). The three different types of textile
construction are woven, knitted and non-woven. Table 2.1 lists the damage characteristics
of the commonly encountered cut and tear damage mechanism for the different textile
constructions.
• Puncture damage occurs when a blunt instrument (such a screwdriver or blunt

knife tip) penetrates the textile by pushing, tearing and shearing the yarn) and
generally presents with the appearance of a ‘hole’ that has irregular yarn ends and
associated textile distortion. If this area of damage is handled (even gently) the
‘hole’ can flatten and the only indication is the frayed yarn ends.
• Impact damage occurs when a blunt instrument (such as a baseball bat or even the
edge of a knife) deforms the textile fibres without cutting or tearing. The features
of the damage will vary with both the fibre composition and the yarn and textile
construction. Generally there is a flattening of the fibres which may or may not be
permanent. If the impact is significant enough (and the yarns are synthetic) fibres
can be melded together.
• Abrasion damage occurs when a textile undergoes a friction contact with a rough-
ened surface. The damage observed macro and microscopically is: damaged sur-
face fibre (missing fibres, frayed yarn ends), may be full thickness with section of
textile missing, presence of possible dirt/particulate matter.
Table 2.1 Damage Characteristics for Woven, Knit and Nonwoven Textiles
Textile
Construction Indicator Tear Characteristics Cut Characteristics
Woven Direction Damage follows warp or weft in the No preferred direction
preferred direction of tear
Edges Devoid of planar array Significant ‘planar array’
Associated stretching Relatively featureless edges
‘Curling’ of textile edge Ability to match fibre ends or
textile design/pattern
Yarn ends ‘bearded’ Yarn ends ‘neat’
Knit Direction Damage generally follows wale or No preferred direction
course in the preferred direction of
the tear. Laboratory tests have shown
that the direction can change during
the course of a tearing action (Dann
et al., 2012)
Edges Associated stretching ‘Planar array’ – though may be
difficult to visualise as severed
yarn ends may retract into
textile
‘Curling’ of textile edge Presence of ‘snippets’
Yarn ends ‘bearded’ Yarn ends ‘neat’
Non-woven Direction No preferred direction No preferred direction
Edges Ragged or fibrous Neat
Associated stretching Pattern matching may be
possible of surface design/
colours
• Thermal damage can occur through several actions including: thermal contact
(such as an iron, or ember); thermal radiant (from close proximity to heat source);
flame; and, friction. The features observed microscopically from thermal dam-
age depend significantly on the fibre composition. For example cotton ashes when
exposed to flame whereas a synthetic fibre will tend to melt.
In identifying the textile damage it is also important to determine (where possible) if

the damage is recent (and therefore of interest to the examination at hand). Generally, if
a garment has been worn for a significant period, or laundered since the creation of the
damage, the alignment of the fibre ends in the plane of the damage will be disrupted and
possibly matted with foreign fibres present (Daroux et al., 2010). However, caution must be
exercised when determining freshness as some garment constructions are such that fibre
ends do not move significantly after wearing or laundering.
As previously discussed, the presence of body fluids can interfere with the interpreta-
tion of freshness as it tends to bind yarn ends in the damage and needs to be documented
(and reported). However, the presence of staining in relation to the damage may be used
to indicate the likelihood of the damage being associated with the staining (and therefore
likely to be associated with the incident).
It is important to emphasise that even with level 1 examinations it is not always possible to
determine a definitive conclusion. This could be because the condition and/or history of the
item may have so influenced the textile as to make any conclusion impossible. As an example
of such a situation clothing from a buried body may have significant levels of deterioration
such that it is not possible to even determine whether or not an item has been damaged.
2.4 2nd Level Examination
This level of examination moves from simply identifying the textile damage present and
providing exclusionary type statements to analysing and reporting on more complex dam-
age, providing inclusionary type statements and hypothesising on the type of implement/s
that could have caused the damage. At this level the textile damage examiner will be able
to determine (where possible):
• If cut damage is stab type, slash type or combination (refer to Figure 2.5)
• The source of complex damage (animal/insect/microbial)
• The alignment of multi-layer damage
• The minimum number of thrusts
• If staining present is consistent with the damage
• If a reference weapon could have/could not have caused the damage (through sim-
ulation experiments)
• The (relative) degree of force used
• What may have caused the damage in the absence of a hypothesis and/or suspect
weapon
2.4.1 Detailed Examinations

The documentation of the damage at this level happens in the same way as for 1st level
examinations with the same features noted at the textile and yarn level. The complexity
Figure 2.5 (from left) Slash type damage (note the some of the yarns of the fabric are still
intact), cut type damage.
arises in the more complex interpretation that is required from the features observed as
well as an understanding of how individual fibres may be severed.
2.4.1.1 Morphological Information from the Fibre Level

The examination of individual fibres depends on the linear density of the fibre. Large fibres
may be individually visualised by stereomicroscopy. Smaller fibres may require higher
magnification using compound or scanning electron microscopy. This level of exami-
nation focussed on the individual fibre fracture. References such as the Atlas of Textile
Damage (Hearle et al., 1998) are useful resources to assist in the interpretation of fibre end
characteristics.
It is important to understand the different ways a fibre can be severed/separated during
different actions and/or forces. An individual fibre may undergo:
• Pushing: The fibre is moved out of its original position within the yarn/textile
(i.e. moves out of plane). Occurs in cutting and puncturing actions.
• Cutting: The fibre is separated due to sharp transverse pressure.
• Shearing: The fibre is separated due to blunt transverse pressure.
• Tearing: The fibre is separated by pressure being applied to the fibre beyond its
extension point resulting in breakage (refer to Figure 2.6).
• Melting/burning: The fibre is heated at a temperature above its melting/combus-
tion point (refer to Figure 2.7).
• Biting: Through the teeth of animals or insects.
• Decomposition: The chemical structure of the fibre can fail through UV exposure,
action of solvent/oxidiser, etc.
Examination of the individual fibre can be undertaken (usually with the assistance of
a scanning electron microscope – SEM) and the work conducted by Hearle et al. (1998)
is an excellent reference for the examination of individual fibre failure. Pelton, in Hearle
et al. outlines research conducted on the use of SEM for forensic fibre examinations and
highlights says that whilst SEM can be a useful diagnostic tool it has several limitations
some of which are that
1. There is a general inadequate knowledge of SEM features caused by different rup-

turing sources.
2. There has to be very careful consideration associated with comparing the damage
features from the evidence and the damage features for single fibre fractures.
3. Some textile structures give similar overlapping features for sharp instruments.
4. A source can result in different failure mechanisms among fibres in a single object.
For the reasons outlined above extreme care must be taken when using the SEM and
images of individual fibre fractures within damage to interpret the possible source of the
damage of interest. The examination of individual fibres should only ever take place after a
holistic examination of the overall damage.
Figure 2.6 Tent fabric showing partial stellate tearing and mushroom fibre ends.
Figure 2.7 Polyester shirt that has incurred thermal damage.
2.4.2 Weapon/Implement Effects

The morphology of an implement such as sharpness, dimensions, width and number of
cutting edges will influence the damage they create. Most implements used as weapons in
criminal cases are mass-produced and as such produce damage that can only be associated
with class characteristics. In rare situations the implement is altered (either intentionally
or through damage) in such a way that this alteration translates as a unique characteristic
observed on a textile. The most commonly encountered implements are
• Knives which can be double edged like a dagger or single edged like a serrated
bread knife or smooth kitchen knife
• Scissors/shears which have effectively two blades each with a single cutting edge
• Blunt instruments such as screwdrivers.
2.4.2.1 Knives
Knives are of particular interest as they account for the dominant number of textile dam-
age cases examined. Variations to a knife, where significant enough, can be translated as
changes to damage to a textile. Johnson (1991) noted that the following parts of the knife
can affect the damage observed:
• The tip: A blunt tip will cause more distortion at the point of entry than a sharper
tip. This is because the yarns at this point fail due to tension rather than through a
cutting force (refer to Figure 2.8).
• The blade: A sharp blade will create a neat cut as it passes through the textile
resulting in little to no distortion. On the other hand a blunt blade will pull yarns
resulting in a less clean cut with more distortion.
• Thickness of the blade: A thick blade will push the fibres apart and can produce
a tapered damage. If the edges of the spine are sharp enough they can also act as
cutting edged producing a y-shaped damage at one end.
• The width of the blade: This will affect (to some extent) the length of the severance.
Several factors contribute to length including width of blade, depth of penetration
and any slashing action that may have occurred.
Figure 2.8 Same knife dimension but difference in tip sharpness: (left) sharp tip; (right)
blunted tip.
• Blade irregularities: The scallops in a serrated knife (if large enough) can cause
increased fraying and distortion to the damage. This is because each scallop point
is out of plane with the cutting edge and will strike yarns at a higher angle. Nicks
and other imperfections can have a similar effect to the scallops and if large enough
may even cause pulling of yarns.
• Knife guard: The knife guard can leave impact damage to the fabric.
• Stabbing angle: The angle at which a blade enters the textile can affect the overall
shape of the damage observed. Variations of curves and multidirectional damage
can result. This is also possible if the blade enters (or is withdrawn) in a twisting
motion.
• Secondary cuts: Can occur when the textile is drawn into the wound creating a fold
which is nicked. These cuts are generally in line with the main damage.
Refer to Appendix 2.2 for a definition of different knife components of a kitchen type
blade.
2.4.2.2 Scissors
In some cases it is important to be able to differentiate between a cut made with scissors
over a cut made with a knife. Cutting with scissors differs from that of a knife in that there
are two cutting blades on a pivot point for scissors versus a single blade for a knife. The
action when using scissors can be either cutting (where the two blades cut the textile from
both the top and the bottom when used in an open and close action) or slicing (where the
textile is cut through contact with the cutting edge of the blade and the textile is forced
through as the scissors are held in an open position).
When scissors are used in an open and close action it is possible that a ‘stoppage’ is
produced which is characteristic of scissor cuts. The effect of scissors is dependent on the
extensibility of the fibre type and textile construction.
2.4.2.3 Blunt Implements

Blunt instruments such as screwdrivers sever a textile by pushing, tearing and/or shearing
the individual fibres. Depending on the surface area and shape presented to the textile the
damage observed will be a combination of tearing and cutting with some level of distortion
around the puncture damage.
2.4.3 Textile Effects

The same implement can produce quite different damage depending on the textile that is
being damaged. For example a loose open knit will have fewer yarns affected by the imple-
ment than say a tight knit textile and it would be harder to penetrate the tight knit. It is
therefore extremely important to investigate the effect of the implement or force on the
textile from the garment itself or a textile as similar in constitution and construction as
possible.
It is also important to understanding that some textile types will be more informative
than others. It follows that the tighter the weave or the knit the more characteristic features
are retained (with non-woven materials tending to provide the most valuable information).
This is particularly important in the examination of damage that has occurred through
Figure 2.9 Same knife different fabrics (from left) knit, woven, nonwoven.
multiple layers. An example of the effect of textile on the observed damage is shown at
Figure 2.9 where the same knife has been used to penetrate different textile construction.
2.4.4 Simulations
Examination of the damage to the garment in conjunction with the examination of a par-
ticular weapon may allow the examiner to exclude or include that weapon as being a pos-
sible source of the damage. A simulation experiment, which is an attempt to reproduce the
damage, may be required to support or determine conclusions regarding the origins of the
damage.
The design of the simulation experiment must be considered carefully so that it is rel-
evant. Considerations for any simulation experiment are
• Hypothesis (as provided by the victim/suspect/witness)

• Textile (evidential garment or purchased and if purchased consideration should be
given to whether to launder the garment before testing)
• Weapon (provided or if general class indicated or if examination of damage indi-
cates a class of weapon)
• Backing (pork, none, other substrate, mannequin) as research has shown that the
backing can affect the evidence obtained in simulation experiments (Carr and
Wainwright, 2011)
• Examiner (is it appropriate to have someone of similar weight or height to the
victim/suspect undertake the simulations or is it suitable to utilise a machine to
reduce human variability)
• Number of replicates
The examiner needs to understand not only the construction of the garment and how
this may affect the damage characteristics but also the mechanical and physical properties
of the textile such as extensibility and elasticity. The yarn type (filament vs. continuous,
single vs. multiple) may also influence characteristics. Therefore, it is important to match
the characteristics of the incident and test textile products as accurately as possible.
The textile damage examiner may have to source second hand clothing to better reflect
a garment’s condition in instances where a well-worn garment has been damaged. It is
rarely possible or advisable to conduct simulations on the item under examination. It is

also known that changes occur in new fabrics as they are laundered. This factor needs to be
taken into account if it is not known if a damaged item is relatively new or if a new item is
used for a simulation experiment.
CASE BACKGROUND AND EXAMINATIONS

Clothing from a non-fatal stabbing was submitted for examination (Figure 2.10). The
items were received with significant blood staining and examinations were conducted
immediately to reduce the impact of drying blood on potential edge characteristics.
(a) (b)
Figure 2.10 The submitted clothing: The hooded jumper (left) and the t-shirt (right).
A knife (Figure 2.11) was also recovered and was used for simulation testing.
Damage on the garments (Figures 2.12 and 2.13) were interpreted to be a combi-
nation of cut and tear that could have been caused by a double edged blade (with both
a smooth and serrated edge) entering at an angle.
Simulation experiments were conducted using garments of similar composi-
tion and construction to the evidential garments and using the recovered implement
(knife) in various different actions (Figures 2.12 and 2.13).
87 mm 95 mm
33 mm
40 mm 30 mm
210 mm
325 mm
Figure 2.11 The suspected implement: A decorative knife with a sharp smooth curved
edge, a false edge and a serrated spine.
(Continued)
Results and Conclusions

It was possible to reproduce macroscopic and microscopic features of the damage
observed on the jumper and shirt with the knife. Therefore, it was concluded that the
knife could have been the source of the damage to the jumper and shirt.
~45 mm
~63 mm
~60 mm ~35 mm ~34 mm
(a) (b)
Figure 2.12 Evidential damage on the jumper (left) displaying the combination of cuts
from a smooth edge and a serrated edge; and simulation damage (right) using the suspected
implement, blade angled left at ~45° relative to the fabric.
~34 mm
~31 mm
(a) (b)
Figure 2.13 Evidential damage on the t-shirt (left) displaying a stab type cut with
associated tearing or serrated edge cut damage; and simulation damage (right).
2.4.5 Limitations and Cautions

Several limitations must be considered by the textile damage examiner. These include that
it is
• Rarely possible to correlate the length of damage with the dimension of the weapon.
Studies such as that produced by Costello (1980) have shown that it is not possible
to always determine blade length from the length of the damage produced and
extreme caution must be exercised if such an interpretation is to be undertaken.
• Not always possible to determine one scenario over another.
• Not possible to quantify the force used to create the damage observed on a textile.
The force is dependent on a range of factors such as age of garment, relative body
positions, etc., that are generally unknown or cannot be reproduced. An exception
to this is tensile testing of textiles that have failed and are relatively new or are
required to maintain a certain standard of robustness (e.g. the testing of a para-
chute harness).
• Not generally possible to determine the direction of thrust.
• Not always possible to determine the order of events leading to the various areas
of damage observed.
• Not possible to control/measure or reproduce all of the dynamic events that led to
the creation of the damage. The use of controlled experiments (from research) is
important in providing a baseline understanding of non-human related elements
(i.e. effect of sharpness of blade).
2.5 Testing
There is a point in the forensic examination of some textile products where the expertise
of the textile damage examiner is no longer enough to thoroughly examine the textile in
question. An example is where the force required to break a textile needs to be measured
in an instance of textile failure. In such cases clothing and textile scientists may need to be
consulted. Worldwide there are a number of University departments and certified textile
test centres that may be able to provide assistance. Expertise regarding the description
of the product including identification of fibre type(s), yarn morphology and textile type
(including, but not limited to, mass per unit area, thickness and sett for woven textiles and
stitch density for knitted textiles) to appropriate national and international standards will
assist with simulation experiments. Such properties are measured under standard envi-
ronmental conditions (ISO 139: 2005; 20 ± 2 °C and 65 ± 4 % R.H.) due to the hygroscopic
nature of textile products. Garment descriptions using product analysis including the
identification of any facing textiles and stitches and seams may also be of use. Advice on
standard laundering procedures and the effect of laundering on textiles may be of impor-
tance, particularly in simulation experiments.
Specialist testing centres will, generally, not have an understanding of forensic proce-
dures and standards and best results will be achieved through a co-operative approach to
the examination.
Such organisations also often have available testing machines that can be used and/
or adapted for recreating events without the variation that occurs during human perfor-
mance testing. Machines of use include tensile testers (for measuring strength and tear
strength) and impact machines which can be adapted for stab and blunt impact testing.
Examples include work conducted by Carr and Wainwright (2011), Daroux et al. (2010),
Mitchell et al. (2012), Dann et al. (2012), Kemp et al. (2009).
The value of using testing machines lies primarily in reducing the human variable so
that a baseline understanding of the impact of other variables (such as knife dimensions or
textile type) can be established. However, it is important to remember that damage caused
in real life involved the human factor which can never be completely reproduced in simula-
tions with or without the use of testing machines.
2.6 Interpretation, Reporting and Standards
The role of the textile damage examiner is to provide opinion to the courts based on the
examiner’s knowledge and skills through systematic and comprehensive examination. The
examiner needs to be careful of what is to be used for the identification and interpretation
of the damage and what will be presented to the courts. There is the very real possibil-
ity that jurors will undertake their own interpretation of the textile damage of interest if
they are provided with too much information (such as detailed photography). Therefore,
it is important to use case notes and detailed photography and sketches for the examiner’s
interpretation whilst the imagery provided to the courts is more generalised.
Depending on the level of examination (i.e. 1st or 2nd) report writing should include
the description of the textile, the identification of the damage type and classification of
the damage as recent or otherwise. If an implement has been submitted the report should
include the types of simulation testing conducted and whether the implement could or
could not have been a source of the damage. If an implement has not been submitted the
examiner may include the type/s of implement/s that may have caused the damage
A recent standard developed in Australia, Australian Standard 5338—Forensic analy-
sis, provides generic level guidance which would inform the development of a textile dam-
age discipline guideline or standard. No such specific standard exists for forensic textile
damage examinations. The above standard has four parts covering recognition, recording,
recovery, transplant and storage of material (Part 1), analysis and examination of material
(part 2), interpretation (part 3) and reporting (part 4).
Part 3 includes an informative appendix in which the elements are considered that
together contribute to the process of evaluation and interpretation. This recognises that
data may come from observations (qualitative data) and quantitative data. In textile dam-
age interpretation qualitative data play an important role. Whilst measurements should be
taken as appropriate it is important that the examiner fully understands the factors which
limit how useful such measurements may be in practice.
This above appendix also recognises that the forensic practitioner will use their profes-
sional judgement and that this needs to be informed through experience, qualifications,
training, continuous professional development, contemporary knowledge and demon-
strated competence.
The forensic practitioner who undertakes textile damage examinations will require a
mixture of core and relevant scientific qualifications, usually some additional more spe-
cific knowledge acquisition (for example, in textile construction), specific training relevant
to the levels of examination to be undertaken and demonstrated competence, with profi-
ciency tests relevant once again to the level of examinations.
At an organisational level, the quality system needs to include formal protocol and
method manuals, a formal training program, competency testing and appropriate authori-
sation to report identifying at what level. As there are no commercially available proficiency
tests for forensic aspects of textile damage these will need to be designed and created in
house. These tests need to be designed to test the different levels of examination.
Laboratories undertaking this type of examination should build up appropriate ref-
erence collections. These may include collections of implements and examples of typical
damage caused by such implements. The latter would also be of value in training programs
and for proficiency tests.
Laboratories should also systematically capture case studies which would again sup-
port training and proficiency testing.
Technical review is an essential component of all forensic examinations and this is
certainly true for textile damage interpretation. However, we would also strongly recom-
mend that a team approach is useful in this type of work where less experienced examiners
are able to discuss the whole holistic approach most appropriate to the particular circum-
stances of a case. Technical review in this approach is more than a final paper review of a
case file after all examinations are completed but should be incorporated into the whole
examination process.
As there is only limited ongoing research in textile damage it is particularly important
that individuals engage in continuing professional development including maintaining
contemporary knowledge.
2.7 Conclusions
Textile damage examinations are an aspect of forensic science that is significantly differ-
ent to fibre examinations as damage is not just about the textile but the implement or force
involved. As examinations of this type necessarily involve human observational data this
needs to be understood and requires the examiner to approach any examination in a sys-
tematic and comprehensive manner.
Some cases are simple and straightforward whereas other cases have a huge number
of variables that increase the case complexity. Of critical importance to all cases, whether
simple or complex, is that the end result is so dependent on the handling of the textile and
implements at the beginning of the examination. Textile damage examinations and inter-
pretation is a holistic process which starts at the crime scene.
Acknowledgements
The authors wish to thank Timothy Simpson of the Australian Federal Police for the provi-
sion of casework images, case studies and for reviewing the document.
Glossary
Recent damage Damage to a garment that has not been laundered or worn for a significant
period of time prior to examination.
Normal wear and tear Damage to a garment through its normal use such as fading, runs, small holes,
seam separation, hem wear etc
Damage of interest Textile damage that is likely to be associated with the incident in question (i.e.
not normal wear and tear and not hospital type damage)
Hospital type damage Damage (usually cutting) created by first responders during the removal of the
clothing
Nick Additional cut or notch generally at the end of the main cut damage
Planar array The neat alignment of fibre ends within a yarn that has been cut
Cut Failure of fibres due to cutting and/or shearing forces
Tear Failure of fibres due to tearing forces
Puncture Cut and/or tear damage created by a blunt implement
Seam separation The failure of the sewing thread creating the seam (generally through tearing)
Stab type cut A cut created from an implement (usually a knife) used in an ‘in and out’ thrust
motion
Slash type cut A cut created from a sharp edge running across the textile
Snippet A loop of yarn that is created when cutting a knitted textile at an angle to the
weft
Thread The yarn that is used to sew a garment together
Stoppage Created by scissors in the opening and closing action
Textile technical face The side of the textile that is meant to be seen.
Abrasion damage Damage to a textile, yarn or fibre from frictional forces (usually through contact
with roughened surface)
Thermal damage Damage to a textile, yarn or fibre from a heat source
References
Carr, D. J. and Wainwright, A., 2011. Variability of stimulants used in recreating stab events, Forensic
Sci Int., 210, 42–46.
Costello, P. A. and Lawton, M. E., 1980. Do stab-cuts reflect the weapon which made them? J Forensic
Sci Soc., 30, 89–95.
Dann, T. J., Carr, D. J., Laing, R. M., Niven, B. E., and Kieser, J., 2012. Tearing of knicker fabrics,
Forensic Sci Int., 217, 93–100.
Daroux, F. Y., Carr, D. J., Kieser, J., Niven, B. E., and Taylor, M. C., 2010. Effect of laundering on blunt
force impact damage in fabrics, Forensic Sci Int., 197, 21–29.
Gore, S. E., Laing, C. A., Carr, D. J., and Niven, B. E., 2006. Standardizing a pre-treatment cleaning
procedure and effects of application on apparel fabrics, Text Res J., 76, 455–464.
Hearle, J. W., Lomas, B., and Cooke, W. D., 1998. Atlas of Fibres and Damage to Textiles. Cambridge:
Woodhead Publishing.
Johnson, N., 1991. Physical damage to textiles. In APPTEC Conference Proceedings, Canberra.
Australia.
Kemp, S. E., Carr, D. J., Kieser, J., Niven, B. E., and Taylor M. C., 2009. Forensic evidence in apparel
fabric due to stab events, Forensic Sci Int., 191, 86–96.
Mitchell, J. L., Carr, D. J., Niven, B. E., Harrison, K., and Girvan, E., 2012. Physical and mechanical
degradation of shirting fabrics in burial conditions, Forensic Sci Int., 222, 94–101.
Pelton, W., 1998. Use of SEM in textile forensic work. In J. W. Hearle, B. Lomas, and W. D. Cooke
(Eds.), Atlas of Fibres and Damage to Textiles. Cambridge: Woodhead Publishing.
Further Reading
Boland, C. A., McDermott, S. D and Ryan, J., 2007. Clothing damage analysis in alleged sexual
assaults—The need for a systematic approach, Forensic Sci Int.,167, 110–115.
Carr, D., 2013. Fibres and textiles. In J. Kieser, M. Taylor, and D. Carr (Eds.), Forensic Biomechanics
(Chap. 6, pp. 138–157). Chichester: Wiley-Blackwell.
Carr, D. J. and Wainwright, A., 2011. Variability of simulants used in recreating stab events, Forensic
Sci Int., 210, 42–46.
Choudhry, M. Y., 1987. The use of scanning electron microscopy for identification of cuts and tears
in fabrics: Observations based upon criminal cases, Scanning Microsc., 1, 119–125.
Daied, N. N., Cassidy, M., and McHugh, S., 2008. An investigation into the correlation of knife dam-
age in clothing and the lengths of skin wounds, Forensic Sci Int., 179, 107–110.
Daly, D. J., Lee-Gorman, M., and Ryan, J., 2009. Distinguishing between damage to clothing as a
result of normal wear and tear or as a result of deliberate damage: A sexual assault case study,
J Forensic Sci., 54, 1–4.
Dann, T. J., 2011. The tearing behaviour and effect of laundering on knicker fabrics. Thesis, Master of
Science, University of Otago, Dunedin.
Green, M. A., 1978, Stab wound dynamics—A recording technique for use in medico-legal investiga-
tions, J Forensic Sci Soc., 18, 161–163.
Horsfall, I., Watson, C., Champion, S., Prosser, P., and Ringrose, T., 2005. The effect of knife handle
shape on stabbing performance, Appl Ergon., 36, 505–511.
Jones, S., Nokes, L., and Leadbeatter, S., 1994. The mechanics of stab wounding, Forensic Sci Int., 67,
59–63.
Knight, B., 1975. The dynamics of stab wounds, Forensic Sci Int., 6, 249–255.
Mamaguchi, A., Masui, M., and Wakasugi, C., 1987. Ultrastructural observation of fabrics damaged
by stabbing, Can Soc Forensic Sci J., 20, 145.
Meyer, W., Schwarz, R., and Neurand, K., 1978. The skin of domestic mammals as a model for the
human skin, with special reference to the domestic pig, Curr Probl Dermatol., 7, 39–52.
Monahan, D. L. and Harding, H. W. J., 1990. Damage to clothing-cuts and tears, J Forensic Sci., 35,
901–912.
Murphy, J., 1966. Smoke damage to textiles, J Forensic Sci Soc., 6, 45.
Pelton, W. R., 1995. Distinguishing the cause of textile fiber damage using the scanning electron
microscope (SEM), J Forensic Sci., 40, 874–882.
Pelton, W. and Ukpabi, P., 1995. Using the scanning electron microscope to identify the cause of fibre
damage part II: An exploratory study, Can Soc Forensic Sci J., 28, 189–200.
Stowell, L. I. and Card, K. A., 1990. Use of scanning electron microscopy (SEM) to identify cuts and
tears in a nylon fabric, J Forensic Sci., 35, 947–950.
Taupin, J. M., 1998. Arrow damage to textiles-analysis of clothing and bedding in two cases of cross-
bow deaths, J Forensic Sci., 43, 205–207.
Taupin, J. M., 1998. Damage to a wire security screen: Adapting the principles of clothing damage
analysis, J Forensic Sci., 43, 897–900.
Taupin, J. M., 1998. Testing conflicting scenarios-a role for simulation experiments in damage analy-
sis of clothing, J Forensic Sci., 43, 891–896.
Taupin, J. M., 1999. Comparing the alleged weapon with damage to clothing-the value of multiple
layers and fabrics, J Forensic Sci., 44, 205–207.
Taupin, J. M., 2000. Clothing damage analysis and the phenomenon of the false sexual assault, J
Forensic Sci., 45, 568–572.
Ukpabi, P. and Pelton, W., 1995. Using the scanning electron microscope to identify the cause of fibre
damage part I: A review of related literature, Can Soc Forensic Sci J., 28, 181–187.
Was-Gubala, J. and Krauss, W., 2004. Damage caused to fibres by vapour cloud explosions, Forensic
Sci Int., 141, 77–83.
Was-Gubala, J. and Krauss, W., 2004. Textile damage caused by vapour cloud explosions, Sci Justice.,
44, 209–215.
Was-Gubala, J. and Krauss, W., 2006. Damage caused to fibres by the action of two types of heat,
Forensic Sci Int., 159, 119–126.
APPENDIX 2.1 Classification Scheme for Assessing Damage
Appearance of Severance
Length The length of the cut is recorded in two ways:
Distance – measure along the line of severance from the point of entry.
Displacement – measure the severance from end to end in a straight line.
This will give an indication of the linearity of the severance.
Angle of severance Follows the line of best fit, or longest section of the cut.
The angle is measured with respect to the wale or warp (machine
direction).
Distortion Uniformity of the interlacing/looping points has been disturbed with
respect to the rest of the textile. Wales, courses, warp or weft yarns may be
permanently moved apart, stitching is strained; loops are elongated or
shortened in neighbouring stitches to accommodate the new configuration.
The location on the severance should be recorded.
Curl Curl is often a textile property resulting from a cut (especially in knitted
textiles). To examine the severance, the cut should be gently flattened and
the curl noted.
Shape Straight – less than two yarn thicknesses’ deviation.
Curved – smoothly varying segment of an arc or circle.
Forked – severance has more than one branch.
Angled – severance consists of two nearly straight segments diverging from
a common point.
Multisegmented – a cut with more than two segments and/or two or more
direction changes.
Secondary cuts Small cuts which are close to, but separate from, the primary or major
severance and which may have been formed during penetration or
withdrawal of the knife.
Body fluids Can be an indication as to the age of the cut. The presence and location
(ends or edges) of body fluids should be noted. If a cut is contaminated
with body fluid, the cut probably occurred at the time of stabbing.
However, it does not necessarily follow that the absence of body fluids
indicates a prior or post cut.
(Continued)
Edges
Unravelling These are long floats of yarn that have moved out of their original positions.
Isolated threads The thread is across the severance action but has not been cut through.
Orientation is recorded as: parallel, perpendicular or 45 degrees with
respect to the severance.
Planar array All fibre ends lie within the same plane at any angle to the yarn axis, and
the yarn retains cohesion through twist. The cross-section of the yarn is not
necessarily circular.
Steps Refers to the microscopic appearance of severance. The cut follows a prime
direction and ‘steps’ down across the wale. The steps are less than two yarns
in diameter. Yarn ends are clean cut but are not necessarily parallel to the
direction of the severance.
Jagged Refers to the macroscopic appearance of the severance. The cut crosses
more than two yarn diameters.
Ruptured ends Yarns are not in planar array. The fibres are of a variety of lengths and may,
but need not, be splayed out.
Nicks Refers to any small cut or notch in the severance. Often they are unnoticed
until the severance edges are moved apart. The location (end or edge)
should be recorded.
Snippets Small cohesive segments of yarn (between 0.5 and 2.0 mm) which have
been cut completely from the textile.
APPENDIX 2.2 Parts of a Knife
Bolster Scales
Tip Heel
Butt
Point
Edge
Tang Rivets
Ropes and Cordages
KENNETH G WIGGINS
3
Contents
3.1 Introduction 89
3.2 Terminology 90
3.3 Structure of Rope 90
3.4 Sample Handling and Identification 91
3.4.1 Comparison Microscopy 92
3.4.2 Natural Fibre Ropes and Twines 92
3.4.2.1 Ashing 93
3.4.2.2 Maceration 93
3.4.3 Man-Made Fibre Ropes and Twines 93
3.4.3.1 Polarized Light Microscopy 93
3.4.3.2 Infrared Spectroscopy 95
3.4.3.3 Melting Point Determination 95
3.4.3.4 Density 95
3.4.4 Markers and Colour Comparison 95
3.4.5 Other Points of Comparison 96
3.5 Conclusion 97
References 98
3.1 Introduction
Vegetable and animal fibres have probably been used for rope-making for many thousands
of years. Indeed, as Ford (1974) stated, cave paintings dated at 18,000 BC show a twisted
rope structure used for climbing. Ford goes on to say that the Phoenicians were using
ropes for sailing around 1200 BC and later, in 480 BC, Xerxes is known to have crossed the
Hellespont using six ropes of a mile each in length.
Due to their widespread availability, ropes are often encountered by the forensic sci-
entist who is faced with the task of comparing different lengths in order to say whether
two or more pieces could have originated from the same source. The traditional vegetable
fibres used for rope manufacture were cotton, hemp and sisal for land usage, and manila
for marine usage, due to its high resistance to rot. In recent times man-made fibre ropes
have become more prevalent, and polypropylene, polyethylene, polyester and polyamide
are usually encountered.
The usage of rope and twine is varied. In past times many ropes would have been
supplied for agricultural use, particularly with animals. As already stated, the sailing and
shipping industry have been using ropes for many years, and still do. Climbing has always
required the use of ropes and twine, and as its popularity as a leisure-time activity has
grown, so has the demand for these products. Borwick (1973) pointed out the benefit of
polyamide ropes to climbers, and stated that those with breaking loads of 5000 lbf are
89
readily available. He went on to say that a filament is as weak as the join to the next fila-
ment, and therefore the introduction of continuous filaments, made possible by the cre-
ation of man-made materials, could not be exaggerated. Horticulturists, both in business
and on a domestic basis, are big users of twine, and how many households do not have a
tow rope available in case the car breaks down? These are only a few of the more usual
places where one might find ropes or twines. However, we should remember that although
these products may be designed and produced for a specific end-use, they often end up
being used for something totally different. Regrettably, this is almost inevitably the case in
a forensic context!
3.2 Terminology
In order to discuss or to compare ropes and twines, the examiner must understand a num-
ber of basic definitions.
• Rope is formed by twisting at least three strands together. The diameter must be at
least 4 mm.
• Twine is a strong thread consisting of two or more strands. The outcome must be a
balanced twisted structure of continuous length and less than 4 mm in diameter.
• A strand contains a number of yarns twisted together.
• A core consists of a fibre or a group of fibres running lengthways through the cen-
tre of a rope or twine. These may be parallel, twisted, cabled or knitted, but are not
combined structurally with the rope or twine.
• Length of lay is one complete turn of a strand which forms part of a rope or twine
(Figure 3.1a).
• Angle of lay is the angle formed between a strand and the axis of a rope or twine
(Figure 3.1b).
• Lay of a rope or twine is the direction of the lay of the strands in a helix about the
rope or twine axis, as described by the capital letters S and Z (Figure 3.1c).
• Additional information can be found in Anon (1995).
3.3 Structure of Rope
Ford (1974) said that it is a remarkable historical feature that rope formed by laying three
strands with opposite twist directions at successive stages has been authenticated for the
period 500–300 BC. Laid ropes are formed by the hawser lay (three strands) or shroud lay
(four strands), as well as by cable lay, which is formed by three or more ropes twisted to
form a helix around the same central axis. These structures have not been significantly
improved upon since. The major change is that, in many applications, plaited and braided
ropes have superseded three-strand ropes. Cores are often encountered in both plaited and
braided constructions, the main reason being to prevent the construction collapsing. Little
additional strength results from the core fibres. In man-made fibre products, the core is
often of a different fibre type to the outer sheath.
Ropes and Cordages 91
Figure 3.1 The lay of the rope. (a) the length of lay in (above) three-strand rope and (below)
eight-strand (plaited) rope (b) angle of lay (c) Z lay and S lay ropes. (Reproduced from Wiggins,
K. G., Sci Justice, 35, 53–58, 1995. With permission.)
3.4 Sample Handling and Identification
Prior to conducting any analysis or taking any measurements, a photograph or drawing

should be made. Other evidence, which may require forensic analysis, such as blood, fibres,
hair, paint or glass, should be collected using the appropriate method. A general macro-
scopic examination can then be made making careful notes regarding the general condi-
tion, e.g. wear and any cut, broken or frayed ends. Care should be taken to preserve the
ends and any knots that appear along the length. A basic description of the rope could
now be made, e.g. a blue, Z twist, three-ply rope, 1.4 cm in diameter with a length of lay of
5.0 cm, an angle of lay of 45° and an overall length of 4.7 m. A macroscopic examination
is normally sufficient to make an initial determination of whether a rope is of natural or
man-made fibre composition. A detailed microscopic examination of the fibres will follow.
3.4.1 Comparison Microscopy

When two ropes or twines are received for comparison, fibres from all constituent parts, i.e.
body, core and markers, should be sampled. To confirm homogeneity, individual strands
should be tested. The fibres should be mounted on a microscope slide using an appropriate
mounting medium, e.g. XAM neutral improved white. A comparison of the samples should
be made using brightfield and fluorescence microscopy. Features that should be compared
include colour, diameter, delustrant (particle size and distribution), cross-sectional shape
and stress marks.
Many of the cheaper ropes are manufactured from split fibres based on tapes. These
vary in thickness. There are flat tapes and profiled tapes. Thickness and profiles are both
useful characteristics for differentiation.
Stress marks or ‘fish-eyes’ (Figure 3.2) can occasionally be seen in polyolefin film. These
are thought to be caused when undissolved polymer, pigment granules or other inorganic
compounds get caught up in the polymer flow while the film is drawn. They can be useful
for comparative purposes. Indeed, in a recent case received at the Metropolitan Laboratory
of the Forensic Science Service, these marks proved to be an important factor. A white,
three-strand twine was recovered from the wrists of a murder victim and a request was
made to compare it to a hank of twine recovered from an address used by a member of
the suspect’s family. The twine was relatively common polypropylene/polyethylene split
film, white in colour. However, when the twine was microscopically examined, a very
unusual feature was noted. In two strands, the film was seen to change dramatically across
its width. Stress marks were initially present in vast numbers, but as the film was scanned
the numbers reduced and eventually disappeared altogether. In the third strand the marks
were present in reasonable numbers which remained consistent across its width. When the
two samples were compared they proved to be identical in all respects, including the way
the stress marks were distributed across the strands. The suspect eventually admitted that
the twine from the victim’s wrists had originated from the hank of twine recovered from
the address of a family member.
3.4.2 Natural Fibre Ropes and Twines

Although natural fibres can be compared simply by removing a sample from the ropes or
twines, additional sample preparation is generally necessary to identify the fibre type by its
microscopic characteristics.
Where possible, two methods of sample preparation are used to obtain the maximum
amount of information from natural fibres, as follows.
Figure 3.2 Stress marks in polyolefin. (Reproduced from Wiggins, K. G., Sci Justice, 35, 53–58,
1995. With permission.)
3.4.2.1 Ashing
A sample of rope or twine is placed in a porcelain crucible and heated at 600°C for three
hours in a muffle furnace. After cooling, the resulting ash is mounted, with minimal dis-
turbance, on a microscope slide using a suitable mounting medium. When a furnace is
unavailable adequate results can be obtained by heating the crucible over a Bunsen burner
flame, ensuring that the crucible lid remains in place to prevent sample loss. Treatment
with hydrochloric acid can sometimes help to make crystals or stegmata easier to see
(Luniak, 1948).
3.4.2.2 Maceration
A sample of rope or twine is placed in a conical flask and covered with equal volumes of
20 volume hydrogen peroxide and glacial acetic acid. A cotton wool bung having been
placed in the neck of the flask, it is heated on a water bath for approximately six hours. Care
should be exercised to make sure that the sample does not dry out. The sample is removed
and placed in a small bottle with water, shaken vigorously, and finally the water is replaced
with ethanol. The fibre bundles which have now been broken down into their individual
cells, known as ultimates, can be mounted on a microscope slide as for the ashing prepara-
tions. Using a combination of microscopic characteristics, as listed in Table 3.1, the sample
can be identified. However, a considerable degree of experience and skill as well as a good
authenticated reference collection is necessary.
Additional information about the identification of vegetable fibres can be found in sec-
tion 4.6 of this book.
Cotton twines can be produced using waste fibre, and as such may contain up to 15%
‘contaminants’ even in a good-quality product. Both polyester and viscose have been seen
mixed with cotton in twines examined by the author.
3.4.3 Man-Made Fibre Ropes and Twines

Many of the man-made fibre ropes and twines encountered in forensic science are manu-
factured by reputable companies, and as such will generally have one of the structures
given in Table 3.2. However, others will be cheaply produced and may not conform to the
expected standards. Techniques other than comparison microscopy are usually required
when comparison and identification of man-made fibres are being undertaken. Polarized
light microscopy, infrared spectroscopy and melting point determination are the three
techniques generally used. Although they are dealt with in more detail in other chapters,
their value in relation to the identification of man-made fibres is now summarized.
3.4.3.1 Polarized Light Microscopy

Birefringence, which is the difference between refractive indices parallel and perpendicular
to the longitudinal axis of the fibre, often varies with polymer type. When man-made fibres
are placed between crossed polars on a polarizing microscope, the birefringence value can
be estimated from the interference colours they show. Quantitative work using a tilting
compensator or quartz wedge is used to establish path difference between polarized light
which has been resolved in directions parallel and perpendicular to the fibre axis. Once the
thickness of the fibre through which the polarized light has passed, is measured the bire-
fringence value is calculated from published tables. Reference to these tables or comparison
with authenticated samples is the basis for identifying most generic classes of polymer.
94
Table 3.1 Identification of Common Vegetable Fibres Found in Ropes
Fibre Colour On Ashing Crystals Lumen On Maceration Pits Cross Marks Miscellaneous
Sisal (Agave sisalana) White Acicular (black bananas) Regular Angular (slit-like) Spirals
Coir (Cocos nucifera) Red-brown Round stegmata
Hemp (Cannabis sativa) Brown Clusters in short chains and singularly Variable in Parallel to long axis Frequent Hairs
Occasionally rhombic/ cubic crystals width (slit-like)
Manila (Musa textilis) Brown Silica (stegmata) Variable in Slight angle to long axis Few but distinct –
width (slit-like)
Flax (Linum usitatissimum) Usually white None Narrow regular Very fine, difficult to see Few, faint –
Jute (Corchorus spp.) Brown Mainly rhombic and cubic in chains Constricted Bordered Few, faint Few spirals
Single cluster crystals
Kenaf/Roselle (Hibiscus Brown Cluster crystals in chains and As jute As jute As jute –
spp.) singularly Very occasionally cubic/
rhombic crystals
Cotton (Gossypium spp.) Grey-white – Convoluted – Dark cross marks –
where spiral
direction changes
Table 3.2 Structure of Man-Made Fibres used in Rope Making

Fibre Structure Description Type
Multifilament A very fine continuous fibre, circular cross-section, <50 μm in diameter Polyamide, polyester, polypropylene
Monofilament Coarser fibres, continuous, circular cross-section, <50 μm in diameter Polyethylene, polypropylene
Staple fibres Discontinuous, from either multifilaments or monofilaments cut into discrete lengths Polyamide, polyester, polypropylene
Fibrillated film* Extruded as a tape, oriented by heat treatment. Fibrillation caused by spinning (twist Polypropylene
fibrillation) or by pinned rollers contracting the film (mechanical fibrillation)
Roll embossed film* Extruded as above, but tape is profiled by passing it between rollers, one of which is grooved Polypropylene
while the other is flat
* Differentiate by cross-sectioning.
Forensic Examination of Fibres
3.4.3.2 Infrared Spectroscopy

Although polarized light microscopy is a cheap and efficient way of obtaining much of
the data required for identification of fibres, it cannot distinguish between various poly-
olefins or subtype polyamides. One way of achieving this information is to use infrared
spectroscopy.
3.4.3.3 Melting Point Determination

This technique can be used to differentiate between high- (mp 133°C) and low-density
(mp 108°C–113°C) polyethylene.
3.4.3.4 Density
Polyolefin fibres can immediately be distinguished from other man-made fibres, as their
density is less than 1.0 and they will float in water. A furthur subdivision is possible if the
fibres are immersed in isophorone (density 0.92 g cm–3), as polyethylene sinks but polypro-
pylene floats (Anon, 1975).
3.4.4 Markers and Colour Comparison

Many manufacturers include ‘house markers’ which may be used to identify their prod-
ucts. Additionally, some use colour-coded threads, tapes or yarns to allow the fibre type
used to be identified. House markers generally appear as external threads or yarns which
form part of one or more strands. The colour-coded materials are placed within the rope
or twine so as to remain recognizable despite general wear and tear. Markers are normally
found in one of two formats. Either they replace one of the yarns and are therefore of the
same size and fibre type as the constituent yarns of the rope or twine or, alternatively, a dif-
ferent fibre type and finer yarn may be inserted in addition to the rope or twine’s constitu-
ent yarns. Table 3.3 gives the combination of colour and fibre type that are now common in
good-quality products. It is important to note that many ropes and twines do not conform
to these guidelines. Samples have been examined by the author which first appeared to
originate from a particular company, but after the house marker was examined in detail
it was clear that they were intended as copies of more expensive ropes. The way the house
marker had been inserted and the fibre type used for the marker were the pointers to the
rope being a copy. This was confirmed when the rope was tested by the company making
the product that the fake was supposed to represent. The scientist should never presume
a particular fibre type has been used for rope manufacture just because the colour-coded
marker implies that it is a certain fibre type. The identity must always be confirmed by
analysis. Old rope or twine samples may still be available with markers which do not con-
form to the currently accepted combinations.
If ropes or twines are being compared, it is essential to look at the dye, pigment or
any other form of colourant present. Coloured fibres from the body, core and markers
should all be compared. Pigments are added directly either to the molten polymer or to a
molten carrier, e.g. polyethylene, before being added to the bulk polymer, e.g. polypropyl-
ene. Polyethylene is also used as a carrier for ferric oxide when it is used to colour poly-
propylene ropes and twines. Ropes manufactured from fibre types other than polyolefins
will usually be dyed. Whether a dye, pigment or other colourant is present, microspec-
trophotometry can always be used to produce absorbance (or transmittance) spectra for
Table 3.3 Fibre Type/

Colour Code Combinations
Fibre Colour
Hemp Green
Manila Black
Sisal Red
Polyamide Green
Polyester Blue
Polyethylene Orange
Polypropylene Brown
comparative purposes. If, after dye classification, a dye is found to be present, then thin-
layer chromatography can be used to analyze the coloured components present. Details
of microspectrophotometry, dye classification and thin-layer chromatography appear in
other chapters of this book. Irrespective of whether the rope is natural or man-made, the
colourants can be compared using these techniques.
3.4.5 Other Points of Comparison

Polymers used in polypropylene rope production can be homopolymers or copolymers. If
copolymers are used, the other polyolefin material will normally be less than 5% by mass.
If a polypropylene film twine is produced, additives not exceeding 10% may be added in
the melt. These additives are present to improve the mechanical properties of the finished
twine.
However, when infrared spectroscopy is performed, it would not be possible to differ-
entiate between a true copolymer where polypropylene and polyethylene were used, and a
polypropylene homopolymer where polyethylene had been used as a pigment carrier. The
other copolymer that can be found with polypropylene is polystyrene, which is used to give
the rope a sheen and to improve its handling properties by making it more pliable. There
are also occasions where it is used simply as an adulterant to add bulk to the rope.
Mechanical fits should always be considered if two pieces of rope or twine need to be
compared to determine if they could have originated from the same length. More informa-
tion on this topic can be found in the work of Adolf (1995) and Grieve (1997). If the ends
have been cut or broken, a mechanical fit may be possible, but the fraying of loose ends
often makes this difficult if not impossible.
Having decided that two pieces of rope or twine could have originated from the same
length, or having a list of parameters for a particular sample, it may be important to trace
the manufacturer. The manufacturer may be able to say how much was made, where it was
distributed and whether it could differ from batch to batch.
Tracing a manufacturer can be difficult if not impossible. At present, in the author’s
laboratory, three sources of information are used: an index of enquiries; a rope collec-
tion which can be searched for matching or similar samples; and the manufacturers
themselves. A scientist with little knowledge of the subject and the industry may find
it difficult to trace a sample. However, by contacting one or two experts in the field, a
successful outcome may be possible. Further information on industrial enquiries can be
found in Chapter 10.
3.5 Conclusion
Once the examination of the rope or twine is complete, the checklists (Figure 3.3) should
be used to ensure that all points of comparison have been considered. Separate microscopic
comparison forms should also be completed to ensure that microscopic features, particu-
larly of man-made fibres forming the various parts of the rope or twine, are considered.
This information, together with that from manufacturers (if available), needs to be con-
sidered and the findings put into perspective for the appropriate legal system. The type
of rope, structure and colour are all important, as are the presence of markers and their
structure and colour. Microscopic detail including diameter and inclusions, mechanical
fits and the amount of rope manufactured all have some bearing on the commonness. If
the manufacturer is traced, conclusions can be more readily drawn, but, as stated earlier,
this is often impossible because, unlike garments, ropes only rarely carry labels (in the
form of tape markers) which may allow a manufacturer to be identified. The number and
Figure 3.3 Check lists for natural fibre rope/twine and man-made fibre rope/twine. (Reproduced
from Wiggins, K. G., Sci Justice, 35, 53–58, 1995.)
type of comparative points will indicate the strength of evidence available. An unusual
number of strands, a rare fibre type or a distinctive colour and marker yarn may make the
rope unique.
This chapter is based on a paper written by the author in 1995.
References
Adolf, F.-P., 1995. Physical fits between textiles. In Proceedings of the 3rd Meeting of the European
Fibres Group (pp. 36–41), Linkoping, Sweden.
Anon., 1975. Identification of Textile Materials (7th ed.). Manchester: The Textile Institute.
Anon., 1995. Textile terms & definitions (10th ed.). In J. E. McIntyre and P. N. Daniels (Eds.),
Manchester: The Textile Institute.
Borwick, G. R., 1973. Mountaineering ropes, Alp. J., 78, 62–69.
Ford, J. E., 1974. Rope, Build. Build. Mater. Text, 3, 36–40.
Grieve, M. C., 1997. From Vietnam to Berlin. Polypropylene clothes line. In Proceedings of the 5th
Meeting of the European Fibres Group (pp. 112–117), Berlin.
Luniak, B., 1948. Die Unterscheidung der Textilfasern (2nd ed.). Zurich: Verlag Leeman.
Wiggins, K. G., 1995. Recognition, identification and comparison of rope and twine, Sci. Justice, 35,
53–58.
From Crime Scene
to Laboratory
JAMES ROBERTSON
CLAUDE ROUX
4
Contents
4.1 Introduction 99
4.2 Crime Scene Considerations 101
4.2.1 General Guidance for Searching the Crime Scene 103
4.2.2 Item Collection at the Scene 103
4.2.2.1 Handpicking 104
4.2.2.2 Lifting 104
4.2.2.3 Sweeping 106
4.2.2.4 Combing or Brushing 106
4.2.2.5 Vacuuming 106
4.2.2.6 Cutting 107
4.2.3 Should Trace Materials be Recovered at the Scene? 108
4.3 Transfer, Persistence and Recovery of Fibres 109
4.3.1 Transfer 109
4.3.2 Persistence 113
4.3.3 Conclusions: Transfer and Persistence 116
4.3.4 Recovery of Fibres 116
4.3.4.1 Methods of Retrieval 117
4.3.5 Case Management Considerations 118
4.4 Protocols for Fibre Examination and Initial Examinations 119
4.4.2 Trace Fibre Evidence 119
4.4.2.1 Case Scenario 119
4.4.2.2 Protocol 119
4.5 Microscopic Comparison 126
4.6 Natural Fibres 129
4.6.1 Vegetable Fibres 129
4.6.2 Animal Fibres 131
4.6.2.1 Non-Human Hair 135
4.7 Summary and Conclusions 140
References 140
4.1 Introduction
Wherever he steps, whatever he touches, whatever he leaves, even unconsciously, will serve as
silent witness against him. Not even his fingerprints or his footprints, but his hair, the fibres
from his clothes, the glass he breaks, the tool marks he leaves, the paint he scratches, the blood
or semen he deposits or collects – all of these bear mute witness against him. This is evidence
99
that does not forget. It is not confused by the excitement of the moment. It is not absent because
human witnesses are. It is factual evidence. Physical evidence cannot be wrong; it cannot per-
jure itself; it cannot be wholly absent – only its interpretation can err. Only human failure to
find it, study and understand it, can diminish its value. [our emphasis]
–Harris v. United States, 331 U.S. 145, 1947
The purpose of this chapter is to provide a broad forensic context for forensic fibre exami-
nations, starting at the crime scene and working through to protocols and procedures for
the laboratory examination of fibres. Other chapters in this book cover in greater depth
individual aspects of the forensic examination of fibres and of fabrics and other materials
constructed from fibres such as ropes and cordage.
Fibres have long been recognised as a category of physical materials which have poten-
tial utility to provide useful information in investigations, and many forensic laborato-
ries around the world continue to invest effort and time in recovering and examining
fibres. However, as we have previously discussed (Robertson and Roux, 2010) trace evi-
dence including fibres, can only have a strong future provided we can demonstrate value
in answering relevant questions in the investigative and intelligence phases of the criminal
justice process. It is axiomatic that meaningful examinations can only be conducted if
there is recognition that trace materials may be present and a specific and conscious effort
to ‘find it’ and recover trace materials. We would suggest that a significant limitation of
forensic science in today’s world is that there is a poor understanding of the potential value
of trace examinations and that its value has been diminished by the failure to ‘find it’.
Peterson et al. (2013) have demonstrated that the collection of crime scene evidence was
predictive of arrest and subsequent examination predictive of charging, conviction and
sentence length. Of course evidence can only be examined if recognised and collected and
this same paper drew attention to the historically low rate of collection of physical materi-
als from even serious crime scenes.
In the 17 years which have passed since the 2nd Edition of this book, there have been
major advances in the DNA analysis of biological materials. This has unquestionably
been a major advance for forensic science and the criminal justice system (CJS) but,
arguably at the cost of less capability on the global level in the forensic examination of
trace material and, certainly, less emphasis in the general collection of trace materials.
Putting aside debate and discussion as to the ability of DNA analysis to ‘identify’ there
is no doubt that the vast numbers associated with DNA ‘results’ have placed the spot-
light firmly on trace examiners to better answer the weight and substance question with
better answers than ‘could have come from’ statements. These questions are considered
elsewhere in this book.
The real value in examining trace materials, including fibres, has to be in contributing
to a more holistic consideration of forensic investigations and, in particular, in attempting
to answer questions as to ‘what happened’ through event reconstruction. By definition this
has to start at the locus or scene of the alleged (criminal) event. Hence, this chapter starts
by considering the crime scene from the perspective of a trace examiner.
Interestingly, it was recently shown that a brighter future for trace evidence, including
fibres, requires recapturing the holistic view of trace evidence and considering the crime
scene as a scientific endeavour (Roux et al., 2015).
From Crime Scene to Laboratory 101
4.2 Crime Scene Considerations
Crime scene examination is a specialist role requiring specific training and competencies.
Kelty et al. (2011) has identified seven critical skills found in high performing crime scene
examiners (CSE’s). These can be summarised as follows:
Cognitive abilities
• Lateral thinker, open, curious to new ideas and alternative methods
• Objective and considered decision making
• High-level multi-tasking abilities
• High-level short and long term planning abilities
• High-level consequential thinking
• Analytical deductive reasoning
Knowledge base
• University degree (not necessarily science/forensic science)
• Legal, policing and investigation (holistic) knowledge
• Sound knowledge of scientific principles
Work orientation
• Good time-management
• Genuine interest/dedication to role (goes beyond job description)
• Self-motivated learners persistent and results driven
Communication skills
• Active listeners with good negotiation/assertive interpersonal skills
• Inclusive team orientated (teacher/trainers and mentors)
• High level written and verbal skills
Professional demeanour
• Unassuming and modest, respected, high credibility
• Does not whinge, ignores pettiness and minor internal politics
• Willing to defend decisions. Self-confident in own abilities but not arrogant,
admits when wrong, learns from mistakes
• Not judgemental, not easily influenced by external factors or people
Approach to life
• Fitness and health orientation.
• Positive world view/positive about life, realistic about life events.
• Consistent, grounded and stress resilient.
• Black humour.
• Creative and innovative (at work and at home).
• Clear life/work separation. Strong social/family support.
Of course it is highly probable (although there is no direct research) that many labora-
tory forensic scientists would share these qualities. What most laboratory forensic scientists
would not share is specific training in the practical aspects of crime scene examination.
Unless a laboratory forensic scientist has received such detailed training they should not
purport to be a CSE. Should laboratory forensic scientists involved in trace examinations
be trained as CSE’s? In an ideal world we would suggest that this would have many benefits
but, practically, it is unlikely to be the general situation due to the traditional divisions in
the field forensic and laboratory forensic organisations and employment spaces with field
forensic still predominately to be found in law enforcement organisations. We would argue
strongly that trace examiners must have a good grounding in scene examination proce-
dures and protocols to the extent which would enable them to integrate into a multidisci-
plinary crime scene team such as might be established in a major incident. This is not to say
that the trace examiner needs to be a specialist across the broader skills and competencies
of a specialist CSE. For example, a trace examiner would not usually be responsible for
recording the scene except for anything specific to their examination and, even here, the
trace examiner may well have the CSE take the actual images.
Hence, in an ideal world the trace examiner would be part of a team in which the vari-
ous specialists would understand their roles and usually work under the direction of the
CSE with overall responsibility for the forensic management of the scene.
Unfortunately, in many parts of the world this ideal situation is a rare event and it
would be even rarer for a trace examiner to have any field involvement, except in the most
significant incidents and where there is an obvious need for trace expertise. Thus, the real-
ity is that in most incidents whether or not trace materials are considered and/or recovered
will rely on the CSE. In so-called volume crime, trace materials are only likely to be recov-
ered where they are not really trace but visible to the naked eye! It is certainly our view
that there is a need to re-engage and re-connect trace examiners and CSE’s to improve the
understanding of CSE’s as to the potential for trace materials and their value as evidence.
This needs to start with a broad understanding by the CSE that trace by definition will
most often not be readily seen by the naked eye and that it will usually only be recovered
effectively if there is a recognition of:
• When trace materials may be present

• How to preserve items against the loss of trace materials
• How to avoid contamination
• When it is appropriate to recover trace materials at the scene
• Appropriate packaging to preserve trace materials to maximise the potential for
meaningful examinations in the laboratory
Many police forensic groups have assumed responsibility for evidence recovery from
items collected at the scene and for triage and selection of what is then sent to a specialist
laboratory for more specific analysis. This is a model which has some potential benefits for
the treatment of trace materials but also has potential drawbacks – we shall consider this
subject in more detail elsewhere in this chapter.
Given the reality of the above model (if trace examinations are to deliver useful
outcomes) it is even more critical that the CSE, the police forensic examiner (by whatever
name) and the laboratory scientists work in a holistic manner and that all have an appropri-
ate detailed understanding of their roles, responsibilities and, importantly, limitations. This
model demands an improved understanding of fibres as trace material on the part of the CSE
and whoever conducts the subsequent recovery and preliminary selection of trace materials.
4.2.1 General Guidance for Searching the Crime Scene

There is a plethora of introductory forensic textbooks all of which include guidance on
search methodologies at crime scenes. Mostly this guidance is similar but there can be
jurisdictional variation which can reflect specific legal considerations and requirements as
well as historical and traditional views of best practice.
Australia has recently produced a standard (AS 5388.1–2012), Forensic Analysis, Part 1:
Recognition, recording, recovery, transport and storage of material, which includes the
following underpinning principles that whatever specific techniques are used they will all
ensure that:
• Recovery is relevant and optimal.

• The integrity of the physical material is not compromised.
• The potential for contamination is minimised.
• Evidence continuity and security is maintained.
• The potential for analysis is optimized.
The standard also includes the following guidance with respect to preservation of
forensic material.
From the point at which physical material is considered to be of potential evidentiary

value, the preservation of that material becomes of paramount importance. Material,
or samples from that material, must be recovered and stored in such a way as to
minimize the risk of loss, deterioration, contamination or alteration, to ensure that
the conclusions based on any scientific examination of the material are reliable and
verifiable.
Forensic material shall be collected and stored in accordance with this standard,
prior to examination, with consideration of the following variables:
1. OHS considerations for the examiner at the scene

2. The dignity and wellbeing of persons involved
3. The type of material
4. The size or quantity of material required for analysis
5. The environment or matrix in which the material is situated
6. The persistence of the material
7. The stability and volatility of the material
8. If possible, the preservation of material or samples from that material for
re-examination.
4.2.2 Item Collection at the Scene

Priorities must be established regarding the order in which evidence will be collected, with
the most fragile or transient materials being collected first, progressing to the more stable
(less ephemeral).
Collection must be undertaken in a manner that will minimise loss or contamination.
In addition, collection sequences should be structured so that initial techniques do not
compromise subsequent processing or collection methods. This will generally mean start-
ing with the least intrusive collection methods and moving to more intrusive techniques.
Information from the attending police may be a good starting point for the formation
of a hypothesis on what to collect from where. Such information would never, however,
be taken as certain and alternative hypotheses should always be considered. Information
should be limited to that which is relevant to the examiner’s making informed decisions
on the management of the scene and collection of relevant materials. Care always must
be taken to recognise the potential for undue cognitive bias through domain irrelevant
information sharing.
The CSE should collect material that may or may not support the proposition posed
by the investigating police. The CSE should also be looking for anything that appears out
of context or inconsistent with the proposition. At a practical level the examiner should:
• Collect a representative sample of the material that is sufficient for the examina-
tions required with provision for sufficient material, where possible, for re-analysis
at a later time and testing on behalf of the defence.
• Avoid examining, collecting or recording material that is not relevant to the
investigation, or is unlikely to provide information relevant to the investigation.
• Collect material in appropriate packaging.
• Label material collected following the required guidelines.
• Record the collection of material in contemporaneous notes.
• Transport the material to the site of analysis or hand the material to the investigator
if required to do so.
Once this has been completed, a final scene walkthrough is recommended to ensure
that no material has been left behind and no potential evidence overlooked.
Various collection techniques are available to the scene examiner, including removal
of the whole item (e.g. bedspread, furniture, weapon, etc.), handpicking, tape-lifting,
vacuuming, swabbing, scraping, and sweeping. The choice of technique will depend on the
circumstances and the type of physical material.
Fibres can be collected with one or more of the following methods.
4.2.2.1 Handpicking
Material may be collected by handpicking using disposable gloves and disposable forceps.
If the material is visible serious consideration should be given to handpicking at the scene
noting the location and recording any special factors which may be relevant to its retention.
The value and importance of not only collecting fibres at the scene but noting and recording
how they are attached to the item is clearly demonstrated in the Tizzone case (see
Roux et al., 1999 and Bennett et al., 2010). This case is discussed in detail elsewhere in
this chapter.
4.2.2.2 Lifting
Clear adhesive tape (tape-lifts) may be used to collect trace material such as fibres. The tape
used needs to be kept in a closed container and checked for extraneous materials before
use. New commercial tape-lifts are becoming available.
The size of tape-lifts varies depending on local preference and includes large (and
sometimes broad) pieces of tape placed on clear plastic sheets, or smaller pieces of tape
placed on glass slides (Figure 4.1a and b).
The choice of tape recovery largely depends on the preferred approach to subsequent
searching of the tape-lifts. As a general observation CSE’s will often take too few tape-lifts
which may then be overloaded. Overloading is where the density of recovered material on
the tape-lift is such that it would have significantly lost its adhesive ability to pick up fibres.
Overloading makes subsequent searching less efficient and less effective.
Some jurisdictions use, so called, one-to-one tape-lifting where the entire surface
being examined has single pieces of tape placed on the surface until it is entirely cov-
ered (Biermann, 1998; Coyle et al., 2004; Nehse, 2004) (Figure 4.2). In this way a highly
detailed map of recovered fibres can be developed. An alternative approach is to use a zonal
method where defined areas of the surface are tape lifted. The size of the zones can only be
(a)
(b)
Figure 4.1 Fibre tape-lifts using glass microscope slides (a) and acetate sheet (b).
Figure 4.2 Use of 1:1 tape-lift of body at scene. (Image courtesy of Kornelia Nehse).
determined on a case-by-case basis taking into consideration the item/surfaces being

examined. Regardless of the actual number of tape-lifts, tape-lifting and other recovery
methods need to be thorough and fit for their intended purpose.
4.2.2.3 Sweeping
Sweeping may be used to collect material from inaccessible areas or where there is a mass
of dispersed material. A clean and contaminant free brush should be used.
4.2.2.4 Combing or Brushing

It is not uncommon for offenders to wear items such as hats, beanies or balaclavas on their
heads which, if recovered, may provide a link between a suspect and such an item (often
discarded at a scene or nearby). Several reports (McKenna and Sherwin, 1975; Salter and
Cook, 1995 and Griffin and Crawford, 1997) have shown the value of combing or brushing
hair to recover fibres which may have been retained by the wearer of a head item. Combs
used for such recovery need to be contaminate free and will often have cotton wool placed
between the teeth (see Figure 4.3).
4.2.2.5 Vacuuming
Vacuuming may be used to remove trace material from inaccessible areas or where there
is a mass of dispersed material. Vacuuming is not the method of first choice for fibres as
Figure 4.3 Use of comb with cotton wool for recovery of fibres from hair.
searching recovered material for millimetre-sized fibres is difficult and time consuming.
Where vacuuming is used, it needs to be conducted using specialised equipment with a
purpose-built collection chamber. Once again equipment needs to be clean and contami-
nant free prior to use.
4.2.2.6 Cutting
As well as collecting potential evidentiary material, the scene examiner must also collect
relevant known and control samples from the scene. For materials such as carpet and fab-
ric, a piece may be cut out and placed in a suitable container. Clearly cutting a sample is
destructive and official permission may be required prior to sampling. Such samples will
be required by the laboratory analyst to assist with the interpretation of results or for com-
parison purposes with trace material recovered from other sources.
Care needs to be taken to avoid cross contamination between involved persons, victims,
physical material, the collector and/or scenes. The following precautions should be followed.
• Gloves should be worn. Some practitioners choose to double glove. They do this
because it is very difficult to put on a pair of disposable gloves without touching
their exterior surfaces – this is aimed primarily at minimising transfer of biologi-
cal trace materials between items.
• Collection equipment (disposable forceps, swabs etc.) should be discarded after
each scene – if this is not possible, then the collection equipment should be cleaned
by washing with soap and water and disinfected prior to the next scene. Victim
and suspect samples must be collected separately (i.e. preferably separated by time,
packaging and space).
• Protective clothing must be changed between scenes.
The appropriate level of personal protective equipment (PPE) will be determined by

the circumstances of the case. However, the work of Palmer and Burch (2009) on the popu-
lation, transfer and persistence of fibres on the skin of living subjects is informative as it
demonstrates that skin can retain numerous fibres for intervals of up to 24 hours. PPE
should cover arms and gloves should be firmly attached to sleeves to avoid fibres being
‘shed’ and causing potential contamination of items being examined.
These procedures are primarily aimed at minimising the transfer of trace biological
material but they have the dual role of also minimising the unintentional transfer of fibres.
4.2.3 Should Trace Materials Be Recovered at the Scene?

Consideration should always be given to recovering trace materials in situ at the scene
particularly to preserve the integrity of the location of trace materials such as fibres. It is
inevitable that in the collection of items from a scene, and their subsequent movement to a
morgue or laboratory, there will be both loss and potential redistribution of fibres.
Hence the CSE needs to be aware of what trace evidence may be present and the poten-
tial for this to be lost.
When dealing with a deceased body, the deceased’s head, jaw, hands and feet should be
protected to minimise potential loss of trace material. If fibres are visible then it is highly
recommended that they are recorded at the scene, in situ, paying particular attention to
how they may be attached or retained. They should then be removed using forceps or twee-
zers as appropriate and placed in an appropriate container such as a druggist’s fold, as show
in Figure 4.4a through f.
Where there are visible signs that the body has been disturbed, such as indications
of sexual assault, then serious consideration should be given to tape lifting at the scene
using a 1:1 or zonal approach. Specific recovery of any biological trace should be consid-
ered before general tape lifting for fibres.
Another example where recovery of fibres at the scene should be considered
is the examination of a typical bedroom scene involving an allegation of sexual
(a) (b)
(c) (d)
(e) (f )
Figure 4.4 (a through f) Use of druggist fold for collection of physical materials.
assault. If bedding is packaged, without prior recovery of fibres, there will be a movement
of these during packaging which could result in the loss of locational evidence.
Murder victims are often buried. The process of recovering a body from a grave is
known as exhumation. Invariably, the grave is shallow and subsides as the body decomposes
and soil settles. CES’s play an important role at these scenes examining the body, search-
ing the soil and the surrounding area for potential evidence. Deceased remains need to be
examined in a systematic manner paying attention to the environment, the position and
the state of the body, as well as physical damage and injuries. If the body is in an open-
air location then it will be necessary to consider protection from the environment. CSE’s
need to be aware of the potential for the recovery of fibres and examination of recovered
clothing.
CSE’s and trace examiners also need to be cognisant of the potential for fibres to be
present in body bags used to transport deceased persons.
In summary, whilst there are no fixed rules with respect to recovery of trace materi-
als at the scene, serious consideration should be given to recovery at the scene if there are
concerns about loss of trace materials or where the location of recovered fibres may assist
in determining what has taken place.
4.3 Transfer, Persistence and Recovery of Fibres
4.3.1 Transfer
The forensic examination of trace evidence, including fibres, is largely based on a theory
postulated by Edmond Locard, Founder and Director of the Institute of Criminalistics at the
University of Lyon (Locard, 1920). This theory simply states that every contact leaves a trace.
This is often quoted as the Locard exchange principle, although Locard himself referred to a
German chemist (Liebig) for his initial ideas. It follows that, in the extreme, if all traces from
a crime were available one could reconstruct the steps of the event and trace back its author.
However, in reality when a transfer has taken place, it may not be detected. The amount
of material transferred may be so small that it is not detected or be identifiable by current
technology. Also, the rate of loss of some materials after transfer may be so great that the
transfer cannot be detected at a very short time after transfer.
For fibres, proof of the theory of transfer first became available in 1975 when stud-
ies were reported from a group working at the then Home Office Central Research
Establishment (HOCRE) at Aldermaston, England.
Pounds and Smalldon (1975a), using wool and acrylic donor and recipient garments,
showed that the number of fibres transferred depended on a number of factors, including:
• The area of the surfaces in contact.

• The number of contacts repeated contacts over the same area were found to cause
transfer of some fibres back to the garment of origin.
• The force or pressure of contact, more fibres being transferred with increasing
pressure.
• The nature of the recipient garment – a cotton laboratory coat with a smooth
surface proved to be a poor recipient (the coarseness of the recipient seems to be
important).
• As high-pressure and coarse-recipient garments produced a greater proportion of

short fibres than low-pressure and smooth recipient garments, it was suggested
by the authors that fragmentation of fibres during contact may be an important
mechanism in fibre transfer.
Pounds and Smalldon (1975c) also studied the mechanisms of fibre transfer and pro-
posed that three mechanisms may be involved:
• Transfer of loose fragments already on the surface of the fabric

• Loose fibres being pulled out of the fabric by friction
• Transfer of fibre fragments produced by the contact itself
Kidd and Robertson (1982) carried out similar experiments but used a wider variety of
fibre types as both donor and recipient garments. Their work showed:
• The importance of the nature of both the donor and recipient garments with
respect to both fibre composition and texture of the fabric.
• That whilst force or pressure of contact is important, the number of fibres trans-
ferred increased with pressure only up to a point beyond which increasing pres-
sure led to no further increase – this was supported by Kriston (1984) who also
showed that fibre transfer can happen even when the pressure of contact is weak.
• That far fewer polyester and viscose fibres were transferred than cotton, acrylic, or
wool fibres, using donor fabrics of these fibre types.
• That the proportion of polyester and viscose fibres transferred from a mixed fabric
was close to the ratio of these fibres in the donor fabric.
• That most (over 80%) of the polyester or viscose fibres transferred were under 5 mm
in length.
Further work by Salter et al. (1984) and Parybyk and Lokan (1986) did not support one
of the above findings. Both these groups highlighted the differential shedding phenom-
enon, wherein for blended fabrics the number of fibres transferred of the different types is
not proportional to the stated fibre composition of the garment. For example, Salter et al.
(1984) showed that with 65/35 polyester: cotton, 55/45 polyester: wool and 70/30 polyester-
viscose donor fabrics, most fibres were shed from the minor component. Indeed, in some
experiments with the polyester/wool mix only wool fibres were transferred. Parybyk and
Lokan (1986) found that the shedding ratio of synthetic polymer blends approximated to
their composition only when expressed by fibre number. It is important to realize that on
garment labels the proportions refer to weight composition.
This issue of differential shedding is important, as situations can arise where the num-
ber of fibres of different types recovered from a case item are not in proportion to the stated
composition of an alleged donor. For example, Mitchell and Holland (1979) reported a
homicide case where only wool fibres were transferred even though the victim’s trousers
also contained synthetic fibres. It is critical to understand the dynamics of both fibre trans-
fer and subsequent persistence in attempting to interpret real-life findings. Sometimes the
explanation can be simple. For example, cord trousers are often a polyester/cotton mixture.
Polyester may not be transferred because the surface of the garment presents only cotton
fibres, with the polyester threads in the underlying construction of the fabric. Fibre exam-
iners also need an understanding of fabric construction.
Fibre transfer is a dynamic and complex process, and other factors can also come into
play. For example, Cordiner et al. (1985) showed that the diameter of wool fibres influ-
enced their fragmentation under pressure. In donor fabric containing wool fibres of vary-
ing diameter, more fine fibres may be recovered after a transfer.
The morphology and thickness of the fibres are definitely significant factors, as shown
by the work of Kolar (1994) and Quattrini (1997) with microfibres. Their findings showed
that some microfibre fabrics are excellent transfer fabrics and can generate up to 17 times
more fibres than cotton in the same conditions. Others do not transfer so well. This will
depend on the surface characteristics of the microfibre garment.
Hellwig (1997) showed that knitting construction is a very important factor in deter-
mining shedding capacity. Other factors that have a definite influence on the shedding
capacity of knitwear garments include the construction of the yarn or twist, the number of
stitches per area and the staple length of the fibres.
Furthermore fibre transfer may be modified as the garment ages or becomes more
worn (Kriston, 1984; Grieve and Biermann, 1997). As there may be a considerable time gap
between the commission of a crime and a suspect being nominated, it is important to real-
ize that transfer properties of clothing collected in these circumstances may have altered
because of wear, washing, other treatment or environmental impact (Kriston, 1984; Bresee
and Annis, 1991; Hellwig, 1997).
Another important issue of fibre transfer is related to the phenomenon of secondary
transfer. Fibres may be transferred from, say, a jumper onto a car seat – the primary trans-
fer. At some time later a second person sits in that same seat, and fibres from the jumper
are transferred to the clothing being worn by the second person – the secondary transfer.
Similarly one could define tertiary, quaternary, etc. transfers. These phenomena have been
extensively studied by Lowrie and Jackson (1994) using garments and car seats as inter-
mediate recipients. They showed that secondary and tertiary transfers between items of
clothing would normally involve only very small number of fibres. However, the second-
ary transfer via car seats appeared to be more significant. There was secondary transfer
in 88 per cent of the cases, and as many as 74 fibres were transferred in one experiment.
This phenomenon is obviously related to the transfer and persistence characteristics of the
fibres and fabrics involved. As pointed out by Grieve and Biermann (1997), it is not possible
to say that a large number of fibres (e.g. 50) must be due to a primary transfer, any more
than to say that a small number of fibres must be due to a secondary transfer.
Many of the above studies were carried out by using artificial or simulated contact. It
is necessary to control the many variables which would otherwise make the interpretation
of results more difficult. However, one can question whether or not it is valid to extrapolate
these laboratory-based findings into real-life situations. While it is possible to carry out
quite sophisticated experiments to test a hypothesis in a specific case, it is naïve to expect
that the results will always give an unequivocal answer. It is more likely that the number of
possibilities will be narrowed. There are real-life problems in real life, and the use of gen-
eral assumptions made from published data can be dangerous. The most accurate picture
of transfer behaviour is obviously provided by the use of live tests simulating the actual
case conditions. Unfortunately, pressure of casework in most laboratories makes this type
of research impossible in all but the most critical cases.
This re-enactment approach has been presented in several instances in the literature.
For example, Grieve et al. (1989) carried out a series of experiments in an attempt to help
interpret real case findings where a large number of fibres were found on a bed sheet,
nightdress and wrist of a deceased. The authors wished to investigate whether or not these
fibres could have originated by casual contact or secondary transfer according to one of
four hypothetical situations:
1. By the wearer of the cardigan walking by the bed, sitting on the bed or removing
the cardigan while close to the bed – all of these actions might have caused loose
fibres to land on the sheet at some time.
2. By the wearer taking off the cardigan and laying it on the bed during a previous
visit to the apartment.
3. After picking up the red acrylic fibres on her outer clothing at the party, on return-
ing home the victim had taken off these items, thrown them on the bed and some
of the fibres had been transferred to her nightdress and the bed sheet.
4. The nightdress had picked up the red acrylic fibres by secondary transfer from
another item in the apartment, for example a chair.
Grieve and Biermann (1997) also performed tests to assess the transfer of wool fibres to
vinyl and to leather vehicle seat following a terrorist case involving fibre transfer onto vinyl
vehicle seats. Roux et al. (1999) carried out a series of tests involving vehicle carpets and
shoe soles to address several issues relevant to the interpretation of a case in which a large
number of fibres had been recovered from the shoe soles of a murder victim.
It is beyond the scope of this review to examine the results of these studies in detail:
it is sufficient to point out that in all cases the replication of the circumstances provided
crucial information to support a given hypothesis, or alternate hypotheses, to support or
not the case findings.
Most of the work described above relates to garment-to-garment transfers. However,
other works have been published dealing with:
• Transfer from automobile carpet fibres to clothing fabrics (see Scott, 1985)
• Transfer from carpet fibres to shoe soles (see Robertson and de Gamboa, 1984;
Roux et al., 1999)
• Transfer from clothing to seat and seat belts (see Robertson and Lim, 1987; Grieve
et al., 1989; Roux et al., 1996; Grieve and Biermann, 1997)
The results of these studies show that the underlying principles of fibre transfer,
emerging from the studies of Pounds and Smalldon, and Robertson and Kidd, hold true.
The quantum of fibres transferred will be determined by the precise circumstances, with
the nature of the recipient and donor materials being critical. Specifically, it is often the
case that car seat fabrics and loose car seat covers will shed very few fibres. Fibres trans-
ferred to car seats are redistributed and, generally, some time after transfer will be found at
the meeting point of the back of the seat and the seat itself (Roux et al., 1996; Roux, 1997).
In the case of transfer of fibres from carpets to shoes, the number transferred will
depend on the nature of the shoe sole, including roughness and wear, and of the carpet,
including texture, mode of manufacture and fibre type. In general, the number of fibres
transferred will be small in comparison with fabric-to-fabric contacts.
In conclusion, fibre transfer is a complex subject. It is difficult to isolate a single mech-

anism that predominates in a given situation. Whilst subtle differences should not be
ignored, overemphasis on detail is equally bad, obscuring the overall trends which emerge
from studies such as those cited above. Highly sophisticated analysis is possible, but in real
life the variables contributing to the number of fibres transferred are so numerous and
unknown that simulated trials can give only general guidance. The value in the work pub-
lished to date lies in its ability to assist the scientist to interpret case findings, being aware
of the factors involved and the limitations in reaching a conclusion.
4.3.2 Persistence
Persistence is the other half of the equation which will determine whether or not fibres will
be found after a transfer (Siegel, 1997). Depending on the nature of the donor and recipient
garments and many other factors, the number of fibres transferred at the time of contact
may range from only a few to many hundreds. Whatever the nature of the transfer involv-
ing garment recipients, it has been clearly established that there is a rapid initial loss of
fibres. The early work of Pounds and Smalldon (1975b) showed an initial loss of about 80%
after only four hours, and only 5%–10% remaining after 24 hours – see Figure 4.5.
The number of fibres lost and the rate of loss depend on numerous factors, some of
which were spelled out by Robertson et al. (1982), who found that fibre persistence was
lessened by four factors:
• The continued wearing of the recipient garment – loss of fibres depends on the
item being moved or worn
• Other garments being worn over or on top of the recipient
100
80
Fibre persistence (%)
60
40
20
8 16 24 32 40
Time of wear (hours)
Figure 4.5 Persistence of wool fibres on three articles of clothing: circles = wool jacket;
triangles = wool/nylon sweater; x = wool jacket. (From Pounds, C. A. and Smalldon, K. W., J
Forensic Sci Soc, 15, 29–37, 1975.).
• The transferred fibres being situated on an area of the recipient more prone to
contact with other surfaces
• The pressure of the original contact being low
These authors also found that longer fibres were lost more quickly than shorter frag-
ments (below 2.5 mm).
Pounds and Smalldon (1975c) proposed that there are three states of transferred fibres,
loosely bound, bound and strongly bound. As time elapses after the initial transfer, the
loosely bound and bound fibres are lost first, with the strongly bound being those that
become physically trapped in the weave of the recipient fabric. This is important in deter-
mining the way in which items are searched. If only strongly bound fibres are likely to
remain, the most efficient recovery technique would be required.
Once fibres have been transferred to a particular area of a garment they can also be
redistributed over the garment and, indeed, onto other garments (Robertson and Lloyd,
1984). In real-life situations it is often not known whether or not all the clothing worn has
been submitted for examination. If small numbers of fibres are found on items it may, for
example, be because:
• There has been a long time gap between contact/transfer and examination.
• The fibres have arrived on these garments by redistribution of a secondary, or sub-
sequent, transfer.
• The recovery method was not efficient.
• The donor is a poor shedder.
• This is not a real transfer but coincidence.
Caution is necessary when interpreting the finding of a small number of fibres, espe-
cially to items such as underclothing. Such potentially damaging evidence may have less
significance than at first glance! Note that the failure to package items in separate bags will
also lead to redistribution between garments. A recent study by Chewning et al. (2008)
showed that when ski masks were seeded with test fibres, and subsequently packaged and
transported, there was a redistribution of fibres between the inner and outer surfaces. The
authors concluded that an examiner cannot conclude with certainty on which surface of
the ski mask fibres were originally deposited.
Another factor which should be considered is the potential for differential loss of fibres
in a fabric blend. For example, smooth polyester fibres have been shown to be lost more
rapidly than rougher viscose fibres (Kidd and Robertson, 1982). Salter et al. (1984) have
pointed out that if one type of fibre from a blended fabric is lost much more rapidly than
the other, this could result in no fibres of one type being recovered.
It should be noted that the normal Pounds and Smalldon theory of fibre loss does not
apply in all circumstances. For example, Grieve and Biermann (1997) showed that fibres
on an unoccupied vehicle seat may be sufficiently bound to the seat surface to remain
undisturbed during normal vehicle usage, due to the absence of direct mechanical forces.
Similarly, a high percentage of fibres persist on garments which have been exposed to the
open air, even after several weeks. Krauss and Hildebrand (1995) showed that under these
circumstances the persistence depends on the fibre type and fibre length, the texture of
the recipient garment and the weather. Light rainfall and wind velocities of up to 17 m s–1
affected fibre persistence only to a small degree, whereas higher amounts of precipitation
resulted in higher fibre loss. This is significant for daily casework activities, because the
probability of finding fibres originating from the offender’s clothing on a homicide victim
can be high, even when the corpse has been exposed to the elements for weeks (Spencer,
1994; Krauss and Hildebrand, 1995; Palmer and Polwarth, 2011). Krauss and Hildebrand
(1996) have also studied fibre persistence in skin under open-Air conditions.
Quattrini (1997) has studied the persistence of microfibres as both donor and recipi-
ent fabrics. While the initial loss is comparable to the findings of Pounds and Smalldon,
the rate of loss is much lower, especially when the microfibres are used as recipients. In the
study above, microfibres transferred onto microfibre garments persisted for up to 14 days
of normal activity.
The effect of garment cleaning on fibre persistence has also been studied (Grieve et al.,
1989; Robertson and Olaniyan, 1986; Talalienë and Vasiliauskienë, 1996; Palmer, 1997;
Szewcow et al., 2011). From these studies it is clear that while cleaning does result in both
loss and redistribution of fibres, it is still worthwhile to examine garments which have been
cleaned. Washing a garment may not be efficient in removing transferred fibres. It is not
unusual to recover dead persons from an aqueous environment, such as a river, or other
inland watercourse (pond, lake) or from the sea. Palmer (2008) describes one such case
involving the murder of five females in and around Ipswich in the UK in late 2006. The
naked bodies of two of the deceased were recovered from a river. Fibres recovered from the
hair of these victims established a common link with the other victims and the offender.
Although DNA analysis did lead to the suspect, DNA could not link all five deceased to the
suspect and his environment.
Palmer (2010) also reviewed several other papers reporting the successful recovery
of fibres in cases in which aquatic environments were involved. Hence, whilst an aquatic
environment may offer unique challenges it is still worth considering the potential for fibre
recovery. Palmer and Polwarth (2011) have also shown that fibres may persist on bodies
exposed to outdoors environmental conditions for up to 12 days.
Fibres can persist for periods of many days and even weeks when transferred to car
seats and seat belts, as shown by the work of Robertson and Lim (1987). The rate of loss was
shown to depend on the nature of the seat covering, with greater retention for fabric seat
covers than for vinyl coverings. The relatively high persistence of fibres on fabric seats and
low persistence on vinyl seats have been confirmed by Grieve et al. (1989) and Grieve and
Biermann (1997) respectively. Roux (1997) showed that the persistence also depends on the
nature of the donor garment. It seems that garments with high shedding capacity remove
more fibres from the seat surface than garments with poor shedding capacity. Coxon et al.
(1992) report on a method to assess the fibre shedding potential of fibres. De Wael et al.
(2010) have proposed a practical method to assess shedding potential based on the use of
a visual scale.
Ashcroft et al. (1989) have shown that fibres transferred from ski masks onto head hair
can persist for up to six days where hair was not washed, and for up to three days when
hair was washed. Fibre transfer, persistence and removal in connection with human head
hair have also been reported on by Salter and Cook (1995).
The persistence of fibres on shoes is extremely poor in normal circumstances, with few
fibres remaining after minutes. Robertson and de Gamboa (1984) have shown that the pat-
tern of the shoe sole and its composition are important. However, for fibres to stay on soles
for any length of time it would seem they have to be trapped or adhering to a sticky deposit.
Roux et al. (1999) have reported on the transfer and persistence of fibres from automotive
carpets onto the soles of a variety of shoes. This study showed that the number of fibres
transferred was always small, often single fibres and in no case more than 100. The shoe
which attracted most fibres had a high profile (i.e. deep-patterned) sole and the composi-
tion of the sole resulted in a very soft and easily worn surface. Even in this example, normal
wear usually resulted in the transfer of only 30 or 40 fibres. In this study, persistence was
measured after the wearers of shoes had walked on a variety of surfaces including vinyl,
concrete, asphalt, and grass. In no cases did fibres persist for more than five minutes unless
they were physically caught in the rubber at the edge of the sole or in a recessed area of the
sole. Even in these cases, all fibres were eventually lost. Trials in which the soles of shoes
were dipped in coke or petrol and then allowed to dry partially failed to retain fibres for
more than minutes. It would appear that finding any fibres on the shoes of an individual
would indicate very recent contact unless there were obvious reasons why fibres had not
been lost.
4.3.3 Conclusions: Transfer and Persistence

In summary, to maximize the recovery of fibres and interpret their potential to answer
case relevant questions:
• It is important to collect clothing and other fabrics from complainants and sus-
pects as soon as possible after an alleged offence.
• Because fibres are so readily lost and retransferred, undue significance should not
be placed on the exact distribution of a small number of fibres.
• Unless a suspect is apprehended fairly quickly subsequent to an incident, failure
to find fibres matching the complainant’s clothing does not necessarily imply lack
of contact.
• Evidence of contact, and hence association, found through comparison of trans-
ferred fibres will generally involve recent transfers.
• It is vital to the integrity of fibre evidence that minimization and contamination
prevention procedures are considered at scenes and in the laboratory.
• As the time of wear increases, those fibres which do remain will be very
persistent and difficult to remove, hence efficient methods of recovery need to be
used.
4.3.4 Recovery of Fibres

It should be clear from the preceding sections that the opportunities for fibres to be trans-
ferred are great and that the fibres will often be too small to be readily detected with the
naked eye. The potential for the use of this type of evidence should be apparent at nearly
all major crime scenes and in sexual and assault cases involving contact. In handling items
and in making decisions about what should or should not be collected, one should always
keep in mind the GIFT (get it first time) principle.
Scenes and items should be treated as though the case will eventually depend on physi-
cal evidence. If, in the fullness of time, it becomes clear that the facts do not warrant the
examination of the physical evidence, then nothing has been lost. If, however, the scenes
or items have been incorrectly dealt with and altered or contaminated, then meaningful
subsequent examinations will not be possible.
A final word of introduction regarding the nature of physical materials: while much of
it is too small to be readily detected, there will be cases where there are more or less obvi-
ous visible tufts of fibre, for example on a broken window, attached to a wire fence, or on a
knife or other weapon.
Less obvious, but well worth looking for, are fused fibres on the plastic trim inside
a vehicle. Fibres can become fused as a result of high temperatures caused by a vio-
lent impact in a vehicle accident (Masakowski et al., 1986; Pabst, 1992). Fabric impres-
sions may also be created. Colour changes may have occurred in the recovered fibres
(Schiller, 1995).
4.3.4.1 Methods of Retrieval

Where fibres are visible, it needs to be decided whether or not to recover them where they
are found. The use of a specialist light source may also be useful in locating the presence
of fibres. Where possible the evidence should be photographed in situ, removed, and pro-
tected before being sent to the laboratory. If there is any chance that the evidence may be
lost during transport it should be removed at the earliest opportunity. Fibres can be col-
lected by using fine forceps aided by the use of low magnification with a hand lens. The
fibres can then be placed into a folded sheet of clean paper and put into a paper or plastic
envelope. With a deceased person, fibres should be collected at the scene before removal to
the mortuary.
Apart from picking off visible materials, the main methods which can be used to
recover fibres are discussed below. As discussed earlier in this chapter the guiding prin-
ciple in deciding whether or not to collect fibres at the scene is, ‘Will the potential for evi-
dence be lost in transit?’ If in doubt, collect!
4.3.4.1.1 Tape-lifts This method was first proposed by Frei-Sulzer (1951). Pieces of clear
adhesive tape are pressed onto the surface of the item being examined, and the whole item
is systematically searched in a grid fashion. Tapes with different adhesive qualities can be
obtained. The stickier the tape, the more effective it is in recovering fibres. However, the
cost is that the fibres comprising the items being searched are also recovered more effi-
ciently, creating an often dense background of fibres which can hinder subsequent search-
ing of tape-lifts for target fibres. 1:1 and zonal tape lifting were considered at 4.2.2.
4.3.4.1.2 Vacuuming By using special-purpose vacuum equipment, large areas can be
quickly searched. The technique is useful for the recovery of particulate materials (glass,
paint, soils) from difficult areas such as car boots and interiors. For fibres the technique
can cover large areas quickly. Its efficiency depends on the strength and effectiveness of the
vacuum motor. A potentially serious drawback is that it can recover a lot of ancient history.
The material collected may be extremely difficult and time-consuming to search.
The advantages and drawbacks of the above techniques have been discussed in detail
by Pounds (1975) and by Lowrie and Jackson (1991). It is quite clear that the method of
choice for fibre recovery is the use of the tape lifting technique. Other techniques such as
brushing or shaking have, in our opinion, such serious drawbacks that they should not be
used to recover extraneous fibres.
Finally, recovery techniques should aim to cause the minimum of disruption or dis-
placement of a fibrous surface. For this reason we do not recommend scraping techniques,
as they will inevitably create loose fibres in the search room (Moore et al., 1986). In this
one aspect, we do not support the SWGMAT (Scientific Working Group Material Sciences
Advisory Group) guidelines released in 1998. In other aspects, the guidelines are extremely
useful and provide the first formal guidance for trace evidence examiners.
4.3.5 Case Management Considerations

All serious cases, and many less serious cases, should be considered for their potential to
yield fibre evidence, and scenes/items must be treated accordingly. First, secure the evi-
dence. Thereafter the items worth searching can only be decided upon when information
becomes available. It is essential that the scientist has the fullest possible relevant knowl-
edge. This will result in resources being channeled where they are needed, with the effec-
tive use of available expertise.
Thus, it is very important to evaluate the case history before starting laboratory exami-
nations. A case conference involving the scientist, crime scene, and investigating officers
can be invaluable. There is no point in conducting a lengthy search to show that fibres are
present in a location where they may reasonably be expected to be as a result of a legiti-
mate transfer. Equally futile is a lengthy search which fails to reveal the presence of fibres,
only for it to turn out that there is good reason to believe that the suspect’s clothing was
not involved or not worn during an incident under investigation. The type of information
which should be sought includes:
• What is alleged to have taken place, who is involved and how?

• Where is the incident said to have taken place? If it was in a house or in a car, who
was the occupier or owner?
• With a sexual assault, did it occur on a bed, on the floor? Is it possible to recon-
struct the sequence of events? Were bed covers present and were they moved?
• When did the incident take place, and was there any delay before the scene was
examined?
• Did any person involved have legitimate access to the scene or legitimate contact
with the other person or persons before the incident?
• Are reliable descriptions available of what was being worn by the offender?
• Were items of clothing removed during the incident?
This type of information will enable the scientist to concentrate on what is likely to be
productive.
Once a decision has been made to proceed on the basis of the case information avail-
able, it then becomes necessary to evaluate possible transfers that are worth following up.
This will be based on a mix of case information and technical considerations relating to the
fibre composition and colour of target fibres.
In summary, the recovery phase is the critical phase for the fibre examiner. It is not
a case of ‘rubbish in, rubbish out,’ but rather nothing found, no potential evidence. The
fibres examiner can never be certain that not finding a fibre transfer is because no con-
tact occurred, because all the fibres may have been lost or because the recovery technique
may have failed to collect the transferred fibres. Recovery procedures must balance the
efficiency of recovery of relevant transferred fibres with the efficiency in recovering back-
ground fibres, either from previous non-relevant (to the case in question) contacts or from
the item itself. Tape lifting may be technically undemanding, but effective tape lifting
requires considerable thought and judgment.
4.4 Protocols for Fibre Examination and Initial Examinations
4.4.1 Introduction
The protocol used to examine fibres will depend on a number of factors, the most impor-
tant being the nature of the fibres to be examined. For example the protocol for the exami-
nation of man-made fibres will differ from that followed for animal fibres or vegetable
fibres. A second major factor which will influence the examination sequence will be the
starting material. For example, protocols will differ for a piece of fabric, a tuft or yarn of
fibres, vacuumings or tape-lifts.
Other authors in this book cover the detailed procedures used in the examination of
these various starting materials.
4.4.2 Trace Fibre Evidence

4.4.2.1 Case Scenario
To illustrate the approach and techniques used, the following case scenario will be used.
Joan Smith, a single mother, lives in a ground floor unit. She is in the habit of sleeping
in a rear bedroom and does not shut the window. In the early hours of the morning entry
has been gained to her bedroom through the window. It is alleged that the offender has then
threatened the complainant and had vaginal intercourse before leaving the scene. At the time
of the alleged assault she was wearing a white cotton nightdress which the offender roughly
pulled up around her breasts. The bedsheets are a very pale yellow colour, and the label
indicates the fibre composition to be polyester/cotton. A blue blanket was also present on the
bed. Although the complainant cannot be certain, she thinks the offender was wearing blue
denim jeans and a red windcheater. A suspect has been apprehended within twelve hours of
the alleged incident.
4.4.2.2 Protocol
Figure 4.6 shows a flow sheet for fibre examination. The search for extraneous fibres is
based on the hypothesis that contact may have occurred between the offender and the
complainant. The purpose of the examination is to see if there are any fibres that would
associate the suspect with the complainant and/or her environment and vice versa.
The first step is to examine the clothing, bedding, carpets, etc. relating to the com-
plainant and the suspect or scene from which fibres may have been transferred. These are
donor items from known sources.
All items should firstly be visually examined. An illuminated low-power magnifier
is useful and/or a swing arm stereomicroscope mounted over a large search table (see
Figure 4.7a and b). Use of an alternative light source may also reveal the presence of fibres.
Visible-sized extraneous material should first be collected. Items should be thoroughly
and systematically examined for the presence of hairs, stains, damage, etc. as previously
Fibres from known items Extraneous fibres

(tape lift and yarn) (questioned)
Loose Tape lift
Preliminary examination stereo Search extraneous fibres for

and compound microscopy ‘targets’
Nominate ‘target’ fibres Similar fibres marked for

recovery
Mount known fibres from Remove and mount fibres

tape lift and from fabric or individually in mountant
yarns in mountant
Comparison microscopy of known

and recovered fibres (visible/UV)
VIS – microspectrophotometry (MSP) of fibres (OR) UV–VIS MSP

accepted after comparison microscopy
Dye extraction and chromatography on

known and recovered fibres
Polarizing microscopy of known and all

recovered man-made fibres
FTIR of sample of known, recovered, and

reference man-made fibres
Optional other techniques,

Py-GC, Micro Raman Spectroscopy
Figure 4.6 Flow diagram for fibre examination.
stated. It is essential that items from the complainant be searched in a separate location
from those of the suspect, and that all possible precautions against contamination are
taken. These precautions should include the use of designated search rooms and specialist
equipment designed for the recovery of trace materials. There should be specific quality
assurance protocols and procedures to minimize the potential for contamination and to
reduce or eliminate background fibres. Procedures should be in place to actively monitor
background fibre populations.
As fibre examinations will rarely be conducted in isolation and independently of other
forensic examinations, careful consideration needs to be given to the sequence of forensic
examinations with the guiding principle being that examinations to recover fibres need to
take place before items are extensively handled or otherwise tested.
(a) (b)
Figure 4.7 (a and b) Use of Magilamp (a) and swing over low magnification microscopy in
evidence examination and recovery (b).
As previously covered any visible fibres would firstly be removed noting both their
locations and how the fibres may be retained. A useful way to retain these fibres is to place
them on a ‘druggist fold’ under a ‘post-it note’ as shown in Figure 4.4a through f.
Once visible fibres have been removed items should then be systematically tape lifted.
A series of tape-lifts is thus created for each item. The location and number of fibres may
become an issue in attempting to reconstruct the events surrounding a case scenario. Case
notes should be sufficiently detailed to enable the examiner to answer questions which
require information about location and numbers. The condition and wear of a fabric and
the presence and degree of damage can be significant factors in determining the potential
for fibre transfer. It is important to collect known fabric/fibre samples from these known
items. A small sample of the fabric (to include both warp and weft yarns for woven fab-
rics) and a tape-lift of the surface should be taken. The tape-lift will help to familiarize the
searcher with the appearance of the target fibres on the tape, and may give more accurate
information about the fibres available on the surface for transfer. It will not necessarily give
an accurate estimate of the sheddability of the fabric (Coxon et al., 1992). With carpets and
items which cannot be readily or easily moved, yarns can be carefully cut from the item.
As for any known item, the samples must fulfill two criteria. They must be sufficient in
number or size and they must be representative, that is, include all variation in the item.
The next stage in the examination is to carry out a preliminary study of the known
fibre samples using both a low-magnification stereo microscope, and, where necessary, a
compound light microscope with transmitted light and higher magnifications. The authors
prefer, at this point, to use tape-lifts for the stereo microscopic examination in the same
way that the recovered tape-lifts are later examined.
In our case scenario, the items submitted are as given in Table 4.1. For the sake of this
discussion, this preliminary examination reveals the information given in Table 4.2.
Table 4.1 Items Submitted

Complainant Suspect
White cotton nightdress Blue denim jeans
Pale yellow sheets Red windcheater
Blue blanket White underpants
Table 4.2 Results of Preliminary Examination

Item Colour Fibre Description
Cotton nightdress Microscopically colourless Twisted or convoluted
Sheets Almost colourless Mixture of convoluted cotton fibres
and delustred man-made fibres
Blanket Mid-blue Lightly delustred man-made fibres
Jeans Pale to dark blue Short convoluted cotton fibres
Windcheater Pale and mid-red Pale convoluted cotton fibres and
darker delustred man-made fibres
Underpants Microscopically colourless Convoluted cotton fibres
With this information, target fibres are selected. There are some rules which assist in
making this selection. Some types of fibre are not well suited to be target fibres. These can
be divided into four groups:
• Fibres which are extremely common. Examples are colourless cotton and, in most
cases, blue cotton fibres from denim fabric. Sometimes these may be of value if
they have been transferred to a location where they would not normally be found
(Grieve and Biermann, 1997). In some situations, work-related clothes (e.g. uni-
forms) will yield fibres which in normal circumstances would be of value, but are
ubiquitous in the specific case.
• Garments which are constructed from very smooth, shiny fabrics are unlikely to
shed or to retain fibres in any quantities. Sheddability can be tested by using con-
trolled fabric-to-fabric experiments (Parybyk and Lokan, 1986) or by the method
described by Coxon et al. (1992).
• Fibres which are undyed or those where the dye concentration is so low that the
fibres appear to be microscopically colourless.
• Fibres from items constructed of shoddy (the terms refers to fabric made from
reprocessed fibres which are so variable that it is not possible to establish what is a
proper known sample).
It cannot be overemphasized that each case must be assessed on its merits.

As previously stated, there will be circumstances where colourless or almost colour-
less fibres can provide useful information. In one murder case, large numbers of blue
denim cotton fibres were found on the pubic hair of the female deceased. She had not
been wearing jeans. Normally, the presence of blue denim cotton fibres would not be
powerful information, but common sense would dictate that any suspect would have to
have blue jeans if they were indeed the offender. This may seem like a statement of the
obvious. The point is that it is important to make sure the obvious is considered, as the
absence of the obvious may be the trigger for the alert examiner when something is not
quite right in a case.
Gaudette (1988) suggests a series of criteria which should be considered at this point in
the case management.
• Highly sheddable garments are difficult and time-consuming to tape, but are likely
sources of questioned fibres.
• Garments of low sheddability are good for taping but are not likely sources of
questioned fibres.
• The more colour contrast there is between the questioned fibres and the garment
taped the better.
• Fibres that fluoresce under ultraviolet radiation make good questioned fibres, par-
ticularly if the garment to be taped does not fluoresce, or fluoresces to a lesser
extent.
• The less common the fibre type, the better it is as a questioned fibre.
• The darker the fibre, the better it is as a questioned fibre, since the chances of a suc-
cessful microspectrophotometric or dye analysis will be greater.
• The coarser the fibre, the more identification and comparison tests can be per-
formed on it; hence, the better it is as a questioned fibre.
• If a garment is damaged, it is a likely source of questioned fibres.
• Special attention should be paid to important areas of the garment, as indicated
by the circumstances of the case (for example the seat of a pair of slacks where it is
believed that the person was sitting on a seat cover).
• If possible, it is desirable to establish a two-way contact or association.
When all of these factors are taken into consideration, it may be that even if the overall
circumstances of the case warrant fibre examination, this may not be possible owing to a
lack of meaningful target fibres.
In our case scenario, taking these criteria into account, the likely target fibres would
be as follows:
• Blue man-made fibres (blanket relating to complainant)

• Red cotton and man-made fibres (windcheater relating to suspect)
The target fibres will have to be identified and their characteristics recorded. An exam-
ple of an appropriate data sheet is shown in Figure 4.8.
The next step would be to search the tape-lifts taken from the items relating to the sus-
pect for blue fibres similar to those comprising the blue blanket; and search the tape-lifts
taken from the items relating to the complainant for red fibres similar to those comprising
the red windcheater.
Tape-lifts from clothing or other items will have background fibres from the items
themselves along with a mixture of other fibres picked up over time from the many con-
tacts the wearer has made. Searching for target fibres is like looking for the proverbial
needle in a haystack. Where target fibres are present, a small number will usually be found
in a background of up to hundreds of extraneous fibres on a single tape lift. The analyst
should note the presence of any large number of fibres of particular types and/or colours.
Other than in exceptional circumstances, it would be impractical and serve no useful pur-
pose to attempt to determine the origin of all the extraneous fibres found to be present, as
is sometimes suggested as being a practical and sensible option.
An exercise which fibre examiners may be asked to conduct is to search items relating
to a complainant or deceased/victim for the presence of fibres which might assist the inves-
tigator by indicating the type of garment or surface with which contact might have been
made. If there is a suspect, the request will be to look for fibres on the alleged victim which
may relate back to the suspect or the suspect’s environment, or the request may be to look
for a common shared contact if no primary source can be determined. Clearly there are
case circumstances that warrant this type of exercise. However, extreme caution should be
exercised, especially if the ‘contact’ established is based on a very small number of fibres or
when the source of fibres cannot be established. In the latter scenario, it is usually proposed
that the item from which such fibres originated has been disposed of by the suspect. This
Case Number: Exhibit number:
Examination Room: Article:
Area sampled: Warp/Weft Filament/Staple

Colour (macroscopic): Colour (microscopic):
Shedding potential: high/medium/ low-nil
Target suitability: too common/ micro.colourless/non-shed
Fibre elongation:
Fibre thickness (μm): Path difference (nm): Birefringence:
Refractive Index: northo npara

* *
COTTON WOOL MAN-MADE( ) OTHER VEG. OTHER ANIMAL
Melting point (if applicable)
Solubility (if applicable)
Delustrant: none coarse
very slight fine
semi-dull regular distribution
dull irregular distribution
aggregates
Markings: cross porosity marks
pits draw marks (fish-eyes)
hollow channels anti-stat
fine striations other
FTIR spectroscopy: yes/ no
Confirmed polymer type: Sub-type:
Fluorescence UV: UV + Blue Green:
Microspectrophotometry
Disc no. File no.
*
use appropriate sheet
Case Reporting Officer Signature Date Examined:

Co-worker Signature Date Check ed:
Figure 4.8 Example record form for known fibre samples.

is only one possible explanation. Generally, the authors consider this type of exercise to
have many dangers of which the fibre examiner should be fully aware. The examiner must
ensure the investigator and, if appropriate (and certainly if the findings are to be produced
as evidence), the prosecutor is made fully aware of the potential limitations of the findings.
Returning to the hypothetical case scenario, fibres similar to the targets are then
marked on the tape-lift with a permanent marker (see Figure 4.9a through f).
(a) (b)
(c) (d)
(e) (f)
Figure 4.9 Sequence for searching tape-lifts for fibres of potential interest, recovery and
mounting of individual fibres, (a – top left) search under stereo microscope (SM), (b – top right)
marking fibres of interest, (c – middle left) cutting a ‘recovery’ window, (d – middle right) plac-
ing xylene onto fibre, (e – bottom left) placing fibre into mountant, (f – bottom right) placing
coverslip onto mountant.
These are the questioned fibres. They may be removed by cutting a small window on
top of the tape-lift, placing a drop of xylene on the fibre, and picking the fibre of interest
off with fine forceps. This is usually done with the assistance of low-magnification micros-
copy. Xylene-based mountants are potentially hazardous and due care needs to be exer-
cised when using them. The recovered fibre is placed in a suitable mountant and a cover
slip added. There are two advantages to mounting recovered fibres individually under
small (12 mm) circular cover glasses (Figure 4.9). First, it allows them to be easily removed
for any subsequent examination which requires an alternative form of preparation, for
example microspectrophotometry including the UV range, or FTIR-microspectroscopy.
Secondly, every fibre can be allotted a unique identifying number, making it a simple task
to record the various examinations that have been carried out on any particular recovered
fibre. In addition, it facilitates examination with the comparison microscope, and makes
quality control checks which may have to be performed by a second examiner much easier.
Choice of mountant is important, as the mountant should meet some essential criteria.
These have been discussed by Cook and Norton (1982) and more recently by Wiggins and
Drummond (2007). There is no single correct mountant, but mountants such as Entellan
and Permmount are widely used. Examiners using microspectrophotometry in the UV
range, which necessitates making temporary mounts on fluorescence-free quartz slides
and cover slips, may find it helpful to use a water-based mountant such as phytohistol
(Grieve and Deck, 1995). Fibres from the known sources will also be prepared in perma-
nent mountant. These should include fibres recovered from tape-lifts of the known item
and also fibres teased out from yarns or threads from the fabric. Temporary mounts, using
glycerine for example, are not recommended. They are accident-prone, become sticky and
collect dust, and are unsuitable for long-term storage. Some problems with dye bleeding
have also been reported.
Subsequent examination consists of three aspects: microscopic comparison, colour
analysis and fibre identification. Only the first aspect is considered here. The other two
aspects are considered in other chapters of this book.
4.5 Microscopic Comparison
The side-by-side comparison of recovered/questioned fibres with fibres from known

sources is conducted using a comparison microscope. This consists of two compound light
microscopes connected by an optical bridge (Figure 4.10a and b). A number of manufac-
turers market comparison microscopes; Leica instruments are most commonly found in
forensic laboratories.
Some examiners believe that it is not possible to achieve balanced illumination with
a comparison microscope and prefer to mount both recovered and control fibres under
the same cover slip. However, the latter procedure has the following flaws and is not
recommended.
• Demounting a recovered fibre for further tests involves re-mounting the known
fibre; the more the fibres are handled, the greater is the danger of confusing the
samples.
• It allows only the one-to-one comparison of a recovered fibre with a particular
known fibre. The range of microscopical features existing within a known sample
(a)
(b)
Figure 4.10 Comparison microscopy of hairs (a) comparison microscope (b) comparison of
recovered and known red fibres.
(easily seen in a preparation consisting of many fibres under one cover slip) cannot
be quickly taken into account for a particular recovered fibre.
• In order to mount a known fibre next to every recovered fibre, a frequently
opened source of these fibres must be in close proximity, creating the potential for
contamination.
Using the comparison microscope, fibres can be compared side by side. It is critical
to look for points of identity (the same features present) and, equally, points of difference.
Much is made of the fact that the examiner may concentrate too much on points of identity
and ignore or give insufficient weight to differences. There is no simple answer to this criti-
cism, but the following discussion may help.
With natural fibres, it is probably fair to say that no two fibres are identical. It is gener-
ally accepted that man-made synthetic fibres are highly uniform; at most any variation is
within narrow limits. However, the fibres and items examined by the forensic worker are
not straight off the production line. The donor item may be old, it may have been exposed
to weathering, or it will have been washed or dry-cleaned. All of these factors, and pos-
sibly others, mean that there is a much greater potential for variation in the population of
fibres which are available on any item for transfer. After transfer, changes may take place in
transferred fibres to further complicate the comparison process. Thus it is facile to believe
that even with man-made fibres, it will always be possible to find identical fibres in the
known sample. Clearly, the way in which the known sample is taken is important. Every
effort needs to be made to take a sample that reflects what would be available to transfer
during contact.
Thus, examiners talk of no significant or meaningful differences. There is an ele-
ment of interpretation as to what is a significant or meaningful difference, and this may
be expected to vary depending on many factors, especially experience and the inherent
discriminating ability of examiners. The latter needs to be established and tested during
training and through ongoing proficiency testing.
Returning to the case scenario, at this comparison stage, the examiner will decide
whether or not each of the recovered/questioned fibres shows no significant differences
from one or more of the known fibres. Use is made of bright-field microscopy (normal
compound light microscopic conditions), and, if the microscope is fitted with the appro-
priate accessories, incident fluorescence microscopy. The types of features which will be
seen include:
• Fibre diameter and variation along fibre length.

• Fibre shape – to study this accurately, cross-sections may have to be made.
• Surface features may be seen, depending on the refractive index (RI) of the fibre
surface relative to that of the mountant.
• Internal detail including the presence, amount, size, shape and distribution of
delustrant (titanium dioxide is used as a delustrant, the effect being to reduce the
brightness of man-made fibres).
• Colour.
• Fluorescence.
Assuming that all other characteristics are common, colour is the critical feature. In
many garments a range of shade will be present in the known sample for colour (hue and
shade). Where a number of fibres have been recovered, the examiner would also wish to be
satisfied that the range of shade found in the known sample and in the recovered fibres is
comparable. The authors believe that this cannot and should not be viewed from a math-
ematical or statistical viewpoint. However, where a large number of recovered fibres are
found, the latter may be possible.
No mention has been made of older techniques such as solubility, thermal properties
and other microchemical testing procedures which were once common. These procedures
have all but been replaced by more sophisticated (and expensive) instrumental approaches.
Many of these tests are described in The Textile Institute’s Identification of Textile Materials
(1975). The focus of this publication is not at the micro level, but many of the test proce-
dures can be conducted on a cavity microscope slide with the aid of a low-magnification
microscope. In parts of the world where expensive equipment, with micro facilities, may
not available, much can still be achieved with a combination of these older techniques and
a polarizing microscope. Interest in the use of simple solubility tests has not entirely disap-
peared with Suga et al. (2014) reporting on the use of solubility tests to discriminate acrylic
and modacrylic fibres.
The protocol outlined above is suitable for most fibre examinations involving trace
evidence where the majority of fibres recovered will be perhaps 1–3 mm in length, and will
rarely exceed 0.5 cm. Such textile fibres may be of man-made, animal or vegetable or other
natural origin. Where a tuft of fibres has been recovered, the same protocol can be followed
except that it will not always be necessary to mount all the fibres. A representative sample
may be sufficient. Where the ‘tuft’ is in fact a piece of fabric, preliminary examination of
the construction of the fabric will be necessary. In instances where it is suspected that the
fibres are not man-made, such as with a fur-type fabric, or where coarse vegetable fibres
have been used, it is necessary to employ different approaches.
4.6 Natural Fibres
For natural fibres such as cotton, wool, other animal fibres, mineral fibres, and vegeta-
ble fibres, microscopic examination is the most important method. These fibres would be
examined as follows.
• Fibres are mounted in a suitable mountant and viewed with bright-field illumina-
tion. For vegetable fibres pretreatment of the fibres is almost invariably necessary.
• Scale casts can be made by a number of methods to reveal the surface pattern of
animal fibres.
• Cross-sections may be made by using a microtome or by improvised means.
Cross-sections have some value in the identification of animal hairs and vegetable
fibres. For man-made fibres, they are a useful comparison feature and can give
some indication of the method of manufacture of the fibre. Appropriate references
can be found in Grieve (1990), and the articles by Palenik and Fitzsimons (1990a,b)
are helpful.
• Determination of optical properties using polarizing microscopy. The optical
properties of a fibre arise from the chemical or molecular composition of the fibre,
and changes in orientation and spacing of the constituent molecules or other
structural units that occur during spinning, stretching, and other treatments. This
aspect is covered in Chapter 5 in this book dealing with microscopy.
4.6.1 Vegetable Fibres

A scheme of analysis for vegetable fibres is shown in Figure 4.11 and a description of
features is given in Table 4.3. The reader is referred to Catling and Grayson (1982) for a
detailed treatment of this topic. Gordon Cook (1984) and The Textile Institute (1975) are
also invaluable primary references. With the exception of cotton, which is almost ubiqui-
tous, vegetable fibres are now rarely seen in forensic casework. Their most common end-
product use is in cordage, ropes and mats. The forensic examination of these is considered
in Chapter 3.
It is important to understand the potential for confusion in the use of the term fibre for
vegetable fibres. For example, one of the more common fibres is jute. The commercial fibre
varies greatly in the coarseness or thickness of the ‘fibre’. In fact what is called a fibre is a
bundle of true botanical fibres. The term ‘fibre’ in strict botanical use refers to an elongate
cell with pointed ends. With vegetable fibres, these individual cells are referred to as ulti-
mates. In the identification of vegetable fibres it is important to study features associated
with the groups of ultimates and the individual ultimates themselves. The fibre bundles
MACROSCOPIC EXAMINATION
Visual examination and use of low-magnification stereo microscope
SAMPLE
MICROSCOPIC EXAMINATION
Fibres present in bundles Single fibres
Fibre Appearance and confirmatory tests
COTTON Convoluted fibres of uneven diameter: variable lumen
KAPOK Very thin, smooth-walled fibre with tapered blunt end
FLAX (cottonized) Similar to cotton but with X nodes along fibre: twist test
Vascular tissue absent
FLAX As for cottonized flax but in groups
JUTE Phloroglucinol test – heavily lignified: cross-section: ashing
HEMP Low lignin content: cross-section: ashing: twist test
OTHER BAST FIBRES
COIR Very coarse fibre with short (<1 mm) ultimates
Vascular tissue present
ABACA Cross-section: ashing: Billinghame’s test
SISAL Cross-section: ashing: Billinghame’s test
Figure 4.11 Scheme for analysis of plant fibres (for methods of conducting the twist test,
examination of fibre ash and Billinghame’s test, see The Textile Institute. Identification of
Textile Materials, The Textile Institute, Manchester, 1975).
require pretreatment to separate the individual cells. Polarizing microscopy is then a very
useful tool in observing the features of these separated cells. The identification of vegetable
fibres is quite specialized, and few, if any, forensic scientists see these fibres often enough
to be confident of identifying the precise fibre. It may only be possible to identify a group
such as leaf or bast fibre.
Cross-sectioning of vegetable fibres other than cotton is invariably helpful. Methods
vary from the plate method (The Textile Institute, 1975) through a wide range of tech-
niques (Grieve, 1990).
Groups of ultimates can be separated by treating a small sample with 1% sodium
hydroxide with gentle heating. A more extreme measure is to place the sample in chromic
acid. In both methods, the progress of the separation needs to be carefully monitored. If
fibres are left for too long in chromic acid, they will simply dissolve. Both methods can
be carried out on a cavity slide, but extreme care should be taken with handling chro-
mic acid. Treated samples may be carefully washed in water and a 5% solution of acetic
acid prior to mounting. Methylene blue is a useful stain to further aid visualization of the
ultimates. Staining with a 2% alcoholic solution of phloroglucinol will show the degree of
lignification.
Ashing is used to reveal the presence of crystals of calcium oxalate which can then be
visualised easily under polarized light. As this technique is destructive, it can only be used
when there is sufficient material. Since only a few groups of ultimates are required and it
can be carried out on a microscope slide or very small dish, this is not usually a limiting
factor. These approaches are suitable where, as is often the case, the vegetable fibre has not
been dyed. If the fibre has been dyed, then useful additional information will be gained by
examination of the colour.
4.6.2 Animal Fibres

Animal fibres are composed of either keratin or fibroin protein. Only silk is made of fibroin,
and in the true sense it is not an animal fibre. Of the keratin-based fibres or hairs, the
examination of those from humans is by far the most important in forensic applications.
The examination of human hairs is nevertheless only a specialized example of the general
examination of hairs. For most hairs, the main purpose is to identify (as far as possible) the
species of animal from which the hairs originated. Limited individualization is possible
from microscopic examination. The approach to the examination of all animal hairs has
common elements, and requires knowledge of the structure of hairs.
The basis of all hair work is microscopic examination, and there is no substitute for
a detailed and systematic examination of individual hair shafts. With respect to meth-
ods, the procedure to be taken is visual examination, followed by examination with a ste-
reo microscope and then with a compound light microscope. Hairs may be mounted on
microscope slides in temporary mountant, usually glycerine, water (9:1 v/v), or in a suit-
able permanent mountant. The selection of mountant is important, and there are points in
favour of and against temporary and permanent mountants. Roe et al. (1991) have evalu-
ated mountants suitable for use in forensic hair examination. De Forest et al. (1987) recom-
mended Meltmount with a refractive index of 1.539 as a suitable mountant.
If permanent mountant is to be used, then impressions of the hair surface (scale casts)
should be made before mounting the hair. Scale casts may be prepared in a number of
ways. A simple method is to layer carefully a thin coat of clear nail varnish onto a micro-
scope slide. A hair is, or hairs are, then placed gently into the nail varnish and left until the
Table 4.3 Summary of Features of Common Bast and Leaf Fibres
132
Length of
Ultimates or
Botanical Common Fibre Cells Extraneous Lumen and Crystals,
Origin Species Name (mm) Material Cross-Markings Cell Wall Pits Silica Ash
Bast fibres Linum Flax 1.6–24.0 Epidermis with Rare. When seen, Cell wall Very fine. Not No crystals
usitatissimum paracytic stomata. often very regular occasionally obvious. Can be reported or
Parenchyma. along whole length of striated. Cell seen by using seen.
Xylem elements. fibre cell. wall thick. polarized light.
Lumen narrow,
regular.
Corchorus Jute 0.6–5.3 Few. Mostly Few, faint. Occasional Lumen of Bordered, Cubic crystals
capsularis and parenchyma, marks from varying width, funnel-shaped in chains
olitorius sometimes with chambered cells. often varying in side view. sometimes
cubic or cluster Scalloped edges to regularly along mixed with
crystals. Very fibres cells. whole length of occasional
occasional vessels. cell. cluster crystals.
Single cluster
crystals.
Cannabis sativa Hemp 1.0–34.0 Laticiferous Variable. Some cells Cell wall Slit-like, Cluster crystal
elements in with fine regularly striated. Lumen parallel to the in short chains,
unmacerated spaced marks in every most long axis of cell, often three or
fibre. Parenchyma specimen. Marks from commonly sometimes four together.
of various types. chambered cells. three to five coalescing. Single cluster
Cluster crystals Several series on one times the width crystals. Very
free or in cells. fibre cell. More of cell wall. occasional
Hairs. Epidermis. frequent than in flax. cubic or
More rarely, Occasional remains of rhombic
blocks of tissue cells attached to crystal in some
and xylem marks. specimens.
elements.
Table 4.3 (Continued) Summary of Features of Common Bast and Leaf Fibres
Length of
Ultimates or
Boehmeria Ramie 13.0–82.7 Parenchyma. Common, fine, nearly Lumen difficult Elongated, Cluster crystals
nivea Cluster crystals always with attached to see because slit-like, parallel in chairs.
free or in remains of (a) it varies, (b) to long axis of Single cluster
parenchyma. parenchyma cells. cell wall is cell, sometimes crystals.
Cluster crystals in Several series on one striated, (c) coalescing.
From Crime Scene to Laboratory
chambered cells. fibre cell. Occasional fibre cells

Few hairs and marks of chambered tangle. Lumen
vessel elements. cells. commonly two
to three times
the width of
cell wall.
Leaf fibres Agave sisalana Sisal 3.0–6.7 No epidermal cells No cross-marks seen. Lumen has Obvious, Large acicular
or stomata seen. Scalloped edges to maximum slit-like, at crystals,
Mesophyll. Blocks some cells. width five to six approximately sometimes
of tissue. times the 40° to long axis with organic
Tracheary thickness of cell of cell. matter
elements with wall, most annealed to the
annular commonly four surface.
thickening or to five times.
opposite to
scalariform
pitting. Spirals,
‘butt’ fibre cells.
(Continued)
133
134
Table 4.3 (Continued) Summary of Features of Common Bast and Leaf Fibres
Length of
Ultimates or
Musa textilis Abaca 2.0–9.3 Parenchyma. Variable, never very Lumen has Slit-like, at Silica bodies,
Stegmata. common, occasionally width two to slight angles to very occasional
Epidermis with regularly spaced. several times long axis of cell. small acicular
stomata and the thickness of crystals in
sub-epidermal the cell wall, some
sclerotic layer. most specimens.
Occasional blocks commonly
of tissue. three to five
Tracheary times. Wide
elements with variation
spiral or annular within and
thickening. between
specimens.
Source:Adapted from Catling, D. L. and Grayson, J., Identification of Vegetable Fibres, Chapman & Hall, London, 1982.
Table 4.4 Human v. Non-Human Animal Hair

Feature Human Other
Colour Relatively consistent along Often showing profound colour
shaft changes and banding
Cortex Occupying most of width of Usually less than width of medulla
shaft – greater than medulla
Distribution of pigment Even, slightly more towards Central or denser towards medulla
cuticle
Medulla Less than ⅓ width of shaft Greater than ⅓ width of shaft
Scales Imbricate, similar along shaft Often showing variation in pattern
from root to tip along shaft from root to tip
varnish has dried. This process can be accelerated by placing the slide on a hot plate for a
short time. Once dry, the hair is peeled off—it helps to leave the root end just free of the
varnish—leaving an impression of the scale pattern. Other methods for obtaining scale
casts have included the use of Polaroid black-and-white land film coater (Ogle et al., 1973)
and Meltmount (Petraco, 1986).
Scale patterns can also be revealed using scanning electron microscopy. This approach
suffers from the disadvantages that it is effectively destructive and it is not usually possible to
study easily the sequence of scale patterns over the length of a complete hair. Furthermore,
after preparation of a hair for SEM, unlike with the simpler scale cast method, the hair is
not suitable for microscopic examination.
The first question which arises in any hair examination is whether the hair is of human
or non-human origin. The differentiation of human hair from other hairs is usually quite
straightforward. Table 4.4 lists a comparison of features found in the two groups.
Once it has been established that the hair is human or non-human, different proto-
cols are followed. Only non-human hair will be considered here. The reader is referred to
Robertson (1999) for a more complete consideration of human hairs.
4.6.2.1 Non-Human Hair

Hairs are composed of three anatomical regions, the cuticle, the cortex and, when present,
the medulla. The attempted identification of non-human hairs requires that the fine detail
associated with these regions be studied in a systematic way.
With non-human hairs, different types of hair can be present in the fur or peltage.
These are usually visually clearly different, based on their degree of coarseness. While dif-
ferent classification systems exist, the most commonly found types are guard hairs and
underhairs. Guard hairs are longer and coarser than the underhairs, which are usually fine.
Guard hairs also display the widest range of microscopic features, which makes them the
most useful for identification. Guard hairs and underhairs should be examined separately
if present. The features shown in Figure 4.12 should be assessed.
Non-human animal hairs should be examined from their root end to their tip end,
and variation in the features recorded. In particular, changes in medulla and scale pattern
appear to be consistent enough to have value in identification. Where possible, questioned
animal fibres should always be compared with authenticated standards.
Many scientific papers describe the microscopy of animal hairs, but few are of real
assistance. Usually they deal with only a very small group of species and consider only
those features which have most value in discriminating the members of that group. Often
descriptions also assume that the scientist has a bulk sample, whereas in forensic work each
hair has to be treated separately. Thus, the identification of all but the more common hairs
is often not possible, and analysis is usually comparative.
It is helpful to use a checklist to record features. It also encourages systematic observa-
tion. An example of such a checklist is presented in Figure 4.13.
This is a very general guide to the approach used in the examination of animal hairs.
Three general references with useful descriptions of a range of hairs are Appleyard (1978),
1 PROFILE (general shape)
Shield
Straight
Symmetrically thickened
Wavy
2 CUTICLE OR SCALE FEATURES
2.1 Scale margin
Smooth
Crenate Sharp pointed teeth
Rippled Indentations deeper than above and rounded
Scalloped Margins with broad rounded teeth
2.2 Distance between scales
Close
Near
Distant
2.3 Scale pattern
Mosaic Regular mosaic
Irregular mosaic
Wave Simple regular wave
Interrupted regular wave
Streaked wave
Irregular waved, mosaic
Regular waved, mosaic
Chevron Single
Double
Figure 4.12 Features for non-human animal hairs.

(Continued)
Pectinate Coarse
Lanceolate
Petal Irregular
Diamond
3 MEDULLA
Note whether present or absent. Where present, it may be continuous,

interrupted, or fragmented.
In non-human hairs, it is often continuous with a defined structure. The structure

can be of two main classes, ladder or lattice.
A ladder medulla is so called because it looks like the rungs of a ladder. Where
there is a single row of ‘rungs’ this is a uniseriate ladder, and with several rows, a
multiseriate ladder.
A lattice medulla is so called because it has the appearance of a lattice made up of

‘struts’ of keratin which outline polyhedral shaped spaces, each of which is
continuous with its neighbours. A special type of lattice medulla, called an
aeriform lattice, differs in that the shapes giving the appearance of the lattice
have arisen because of cell collapse leaving air-filled gaps roughly polyhedral in
shape.
4 COLOUR
The colour of hairs results from pigment particles deposited in the cortex. Overall
visual and macroscopic colour can be important in non-human hairs, but the
pigmentation in the cortex is less important than in human hair. Pigment should
be assessed with respect to:
a. amount sparse or dense
b. distribution along the shaft even
across the shaft denser near centre
denser near cuticle
5 CROSS-SECTION
This is not always carried out, because it is distructive. information from cross
-sections is threefold:
a. a good appreciation of pigment distribution across the staff is gained
b. the position of the medulla, which can be in the middle (centric) or off to
one side (eccentric) is determined (both of these features can be assessed
by optical sectioning when fibres are viewed in longitudinal plane)
c. the shape of the hair is determind.
Figure 4.12 (Continued) Features for non-human animal hairs.
Brunner and Coman (1974) and Debrot et al. (1982). As stated previously, it is often pos-
sible to carry out only a comparative examination.
A comprehensive reference collection of hairs is invaluable, but there is no substi-
tute for experience. Animal hairs are ubiquitous, but this is not reflected in their use as
evidence of contact. Paradoxically, this may be because they are ubiquitous, and forensic
workers lack the ability to discriminate sufficiently to give them evidential significance.
It is our belief that there are many more cases in which animal hairs could provide useful
information than is currently realized. However, few forensic laboratories have people with
the necessary expertise.
The application of DNA techniques to the individualization of animal hairs opens up
exciting possibilities for the future (Menotti-Raymond et al., 2008; Wictum et al., 2013).
Because of the time and effort necessary for DNA analysis, microscopic examination will
remain the core to the examination of animal fibres. Differences between broad groups
of animals are usually obvious at a microscopic level based on a systematic examination
ANIMAL HAIR EXAMINATION 1 CASE NUMBER:
ITEM NUMBER:
Date: Time: Page of
MACROSCOPIC HAIR NUMBER
FEATURE 1 2 3 4 5
LENGTH CM
SHAFT PROFILE SHIELD
STRAIGHT
SYMM THICK
WAVY
COLOUR
GENERAL DESCRIPTION:
EXAMINER: NOTES CHECKED BY:
ANIMAL HAIR EXAMINATION 2 CASE NUMBER:
ITEM NUMBER:
Date: Time: Page of
MICROSCOPIC HAIR NUMBER
FEATURE 1 2 3 4 5
PIGMENT DENSITY NONE
LIGHT
MEDIUM
HEAVY
Figure 4.13 Check list of features for examination of non-human animal hairs.
(Continued)
PIGMENT DISTRIBUTION TOWARDS CENTRE
EVEN
TOWARDS CUTICLE
MEDULLA DISTRIBUTION NONE
FRAGMENTED
INTERRUPTED
CONTINUOUS
MEDULLA TYPE AMORPHOUS
LADDER
LATTICE
AERIFORM LATTICE
SCALE EDGE SMOOTH
CRENATE
RIPPLED
SCALLOPED
DISTANCE BETWEEN SCALES CLOSE
NEAR
DISTANT
SCALE PATTERN MOSAIC
SIMPLE WAVE
INTER. WAVE
WAVED MOSAIC
SINGLE CHEVRON
DOUBLE CHEVRON
COARSE PECTINATE
LANCEOLATE
PECTINATE
OVOID BODIES
EXAMINER: NOTES CHECKED BY:
Figure 4.13 (Continued) Check list of features for examination of non-human animal hairs.
of scale features and the medulla, paying particular attention to how these vary along the
length of hairs. Many eliminations will be possible with microscopic examination. DNA
promises a degree of individualization that has not previously been possible using micro-
scopic examination.
4.7 Summary and Conclusions
The ubiquitous nature of fabrics and their constituent component fibres ensures their pres-
ence as a physical material that forensic scientists should consider in forensic investiga-
tions and subsequent examinations. Active consideration needs to be given at the crime
scene and ‘evidence’ recognition stage if this type of physical material is to be properly
recorded and recovered. As stated in the quotation opening this chapter ‘only human fail-
ure to find it, study and understand it, can diminish its value’.
An objective of this chapter is to increase awareness of the potential for fibres to be
recovered and then to describe protocols and approaches for their subsequent examination
in the context of active and effective case management. Subsequent chapters deal with the
more detailed technical examinations possible for different fibres and then how to interpret
and place weight on fibres as potential evidence in case context.
Nonetheless all of the more detailed work can only be possible when the front end of the
forensic examination places appropriate balance on the recovery of all physical evidence.
References
Appleyard, H. M., 1978. Guide to the Identification of Animal Fibres (2nd ed.). Leeds: WIRA.
Ashcroft, C. M., Evans, S., and Tebbett, I. R., 1989. The persistence of fibres in head hair, J. Forens.
Sci. Soc., 28, 289–293.
Bennett, S., Roux, C. P., and Robertson, J., 2010. The significance of fibre transfer and persistence—A
case study, Aust. J. Forensic Sci., 42, 221–228.
Biermann, J., 1998. The advantages and disadvantages of 1:1 taping. In Proceedings of the 6th European
Fibres Group Meeting, Dundee.
Bresee, R. R. and Annis, P. A., 1991. Fibre transfer and the influence of fabric softener, J. Forensic Sci.,
36, 1699–1713.
Brunner, H. and Coman, B. J., 1974. The Identification of Mammalian Hair. Melbourne: Inkata Press.
Catling, D. L. and Grayson, J., 1982. Identification of Vegetable Fibres. London: Chapman & Hall.
Chewning, D. D., Deaver, K. L., and Christensen, A. M., July 2008. Persistence of fibers on ski masks
during transit and processing, Forensic Sci. Comm., 10, https://archives.fbi.gov/archives/about-
us/lab/forensic-science-communications/fsc/july2008/research/2008_06_research01.htm.
Cook, R. and Norton, D., 1982. An evaluation of mounting media for use in forensic textile fibre
examination, J. Forensic Sci. Soc., 22, 57–63.
Cordiner, S. J., Stringer, P., and Wilson, P. D., 1985. Fibre diameter and the transfer of wool fibres, J.
Forensic Sci. Soc., 25, 425–426.
Coxon, A., Grieve, M., and Dunlop, J., 1992. A method for assessing the fibre shedding potential of
fabrics, J. Forensic Sci. Soc., 32, 151–158.
Coyle, J., Larkin, A., Smith, K., Mayo, S., Chan. A., and Hunt, W., 2004. Fibre mapping—A case study,
Sci. Justice, 50, 192–194.
De Forest, P. R., Shankles, B., Sacher, R. L., and Petraco, N., 1987. Meltmount 1.539 as a mounting
medium for hair, Microscope, 34, 249–259.
De Wael, K., Lepot, L., Lunstroot, K., and Gason, F., 2010. Evaluation of the shedding potential of
textile materials, Sci. Justice, 44, 179–186.
Debrot, S., Fivaz, G., Mermod, C., and Weber, J. M., 1982. Atlas des Poils de Mammifères d’Europe.
Neuchatel: Institut de Zoologie, Université Neuchatel.
Frei-Sulzer, M., 1951. Die Sicherung van Mikrospuren mit Klebeband, Kriminalistik, 10/51, 190–194.
Gaudette, B. C., 1988. The forensic aspects of textile fibre examination. In R. Saferstein (Ed.), Forensic
Science Handbook (Vol. 2). Englewood Cliffs, NJ: Prentice Hall.
Gordon Cook, J., 1984. Handbook of Textile Fibres, Natural Fibres (5th ed.). Durham: Merrow.
Grieve, M. and Biermann, T., 1997. The population of coloured textile fibres on outdoor surfaces, Sci.
Justice, 37, 231–240.
Grieve, M. and Deck, S., 1995. A new mounting medium for the forensic microscopy of textile fibres,
Sci. Justice, 35, 109–112.
Grieve, M. C. and Biermann, T. W., 1997. Wool fibres—Transfer to vinyl and leather vehicle seats and
some observations on their secondary transfer, Sci. Justice, 37, 31–38.
Grieve, M. C., Dunlop, J., and Haddock, P. S., 1989. Transfer experiments with acrylic fibres, Forensic
Sci. Int., 40, 267–277.
Grieve, M., 1990. Fibres and their examination in forensic science. In Maehly A. and Williams R.
(Eds.), Forensic Science Progress (Vol. 4). Berlin: Springer-Verlag.
Griffin, R. M. E. and Crawford, C., 1997. An improved method for the preparation of combs for use
in hair combing kits, Sci. Justice, 37, 109–113.
Hellwig, J., 1997. The effect of textile construction on the shedding capacity of knitwear. In Proceedings
of the 5th Meeting of the European Fibres Group (pp. 102–106), Berlin.
Kelty, S. F., Julian, R., and Robertson, J., 2011. Professionalism in crime scene examination: The seven
key attributes of top crime scene examiners, Forensic Sci. Pol. Manag., 2, 175–186.
Kidd, C. B. M. and Robertson, J., 1982. The transfer of textile fibres during simulated contacts, J.
Kolar, P., 1994. Transfer experiment with microfibres. In Proceedings of the 2nd Meeting of the
European Fibres Group (10–11), Wiesbaden.
Krauss, W. and Hildebrand, U., 1995. Fibre persistence on garments under open-air conditions In
Proceedings of the 3rd Meeting of the European Fibres Group (32–35), Linköping.
Krauss, W. and Hildebrand, U., 1996. Fibre persistence on skin under open-air conditions. In FBI
Symposium on Trace Evidence in Transition, San Antonio, TX.
Kriston, L., 1984. Über den Beweiswert der Textilmikrospüren, Arch. Krim., 173, 109–115.
Locard, E., 1920. L’enquète criminelle et les méthodes scientifiques (p. 139). Paris: Flammarion.
Lowrie, C. M. and Jackson, G., 1991. Recovery of transferred fibres, Forensic Sci. Int., 50, 111–119.
Lowrie, C. N. and Jackson, G., 1994. Secondary transfer of fibres, Forensic Sci. Int., 64, 73–82.
Masakowski, S., Enz, B., Cothern, J. E., and Rowe, W. F., 1986. Fiber–plastic fusions in traffic accident
reconstruction, J. Forensic Sci., 31, 903–912.
McKenna, F. J. and Sherwin, J. C., 1975. A simple and effective method for collecting contact evi-
dence, J. Forensic Sci. Soc., 15, 277–280.
Menotti-Raymond, M., David, V. A., Pflueger, S. M., Lindblad-Toh, K., Wade, C. M., O’Brien, S. J.,
and Johnson, W. E., 2008. Patterns of molecular genetic variation among cat breeds, Genomics,
91, 1–11.
Mitchell, E. J. and Holland, D., 1979. An unusual case of identification of transferred fibres, J. Forensic
Sci., 19, 23–26.
Moore, J. E., Jackson, G., and Firth, M., 1986. Movement of fibres between working areas as a result
of routine examination of garments, J. Forensic Sci. Soc., 26, 433–440.
Nehse, K., 2004. Using 1:1 taping to reconstruct a source. In M. Houck (Ed.), Trace Evidence Analysis
(pp. 191–210), Elsevier, London.
Ogle, R. R., Mitosinka, B. A., and Mitosinka, G. J., 1973. A rapid technique for preparing hair cuticu-
lar scale casts, J. Forensic Sci., 18, 82–83.
Pabst, H., 1992. Anschmelzspuren, Kriminalistik, 8–9/92, 527–549.
Palenik, S. and Fitzsimons, C., 1990a. Fiber cross sections, Part 1, Microscope, 38, 187–195.
Palenik, S. and Fitzsimons, C., 1990b. Fiber cross sections, Part 2, Microscope, 38, 313–320.
Palmer, R., 1997. The retention and recovery of transferred fibres following the washing of
recipient garments. In Proceedings of the 5th Meeting of the European Fibres Group (60–62),
Berlin.
Palmer, R., 2008. The Ipswich serial killings. In Proceedings of the 16th Meeting of the European Fibres
Group, Budapest.
Palmer, R., 2010. The forensic examination of fibres—A review 2007–2010. In Proceedings of the 16th
International Forensic Science Symposium, ICPO–Lyon: Interpol, Taylor and Francis.
Palmer, R. and Burch, A. J., 2009. The population, transfer and persistence of fibres on the skin of
living subjects, Sci Justice, 49, 259–264.
Palmer, R. and Polwarth, G., 2011. The persistence of fibres on skin in an outdoor deposition crime
scene scenario, Sci Justice, 51, 187–189.
Parybyk, A. E. and Lokan, R. J., 1986. A study of the numerical distribution of fibres transferred from
blended fabrics, J. Forensic Sci. Soc., 26, 61–68.
Peterson, J. L., Hickman, M. J., Strom, K. J., and Johnson, D. J., 2013. Effect of forensic evidence on
criminal justice case processing, J. Forensic Sci., 58, 78–89.
Petraco, N., 1986. The replication of hair cuticle scale patterns in Meltmount, Microscope, 34,
341–345.
Pounds, C. A., 1975. The recovery of fibres from the surface of clothing for forensic examinations, J.
Pounds, C. A. and Smalldon, K. W., 1975a. The transfer of fibres between clothing materials during
simulated contacts and their persistence during wear—Part 1: Fibre transference, J. Forensic
Sci. Soc., 15, 17–27.
Pounds, C. A. and Smalldon, K. W., 1975b. The transfer of fibres between clothing materials during
simulated contacts and their persistence during wear—Part 2: Fibre persistence, J. Forensic Sci.
Soc., 15, 29–37.
Pounds, C. A. and Smalldon, K. W., 1975c. The transfer of fibres between clothing materials during
simulated contacts and their persistence during wear—Part 3: A preliminary investigation of
the mechanisms involved, J. Forensic Sci. Soc., 15, 197–207.
Quattrini, A., 1997. Transfer of microfibre onto conventional garments and microfibre garments
during simulated contacts and their persistence during wear, In Proceedings of the 1st European
Academy of Forensic Sciences Meeting, Lausanne.
Robertson, J. (Ed.), 1999. Forensic and microscopic examination of human hair. In Forensic
Examination of Hair. London: Taylor & Francis.
Robertson, J. and de Gamboa, X. M., 1984. The transfer of carpet fibres to footwear. In Proceedings of
the 10th International Association of Forensic Sciences Meeting, Oxford.
Robertson, J., Kidd, C. B. M., and Parkinson, H. M. P., 1982. The persistence of textile fibres trans-
ferred during simulated contacts, J. Forensic Sci. Soc., 22, 353–360.
Robertson, J. and Lim, M., 1987. Fibre transfer and persistence onto car seats and seatbelts. In
Proceedings of the 11th International Association of Forensic Sciences Meeting, Vancouver.
Robertson, J. and Lloyd, A., 1984. Redistribution of textile fibres following transfer during simulated
contacts, J. Forensic Sci. Soc., 24, 3–7.
Robertson, J. and Olaniyan, D., 1986. Effect of garment cleaning on the recovery and redistribution
of transferred fibres, J. Forensic Sci., 31, 73–78.
Robertson, J. and Roux, C. P., 2010. Trace evidence: Here today, gone tomorrow? Sci Justice, 50,
18–22.
Roe, G. M., Cook, R., and North, C., 1991. An evaluation of mountants for use in forensic hair
examination, J. Forensic Sci. Soc., 31, 59–66.
Roux, C., 1997. La valeur indiciale des fibres textiles découvertes sur un siège de voiture: Roblèmes
et solutions. PhD thesis, University of Lausanne, Sydney, Australia: UTS Printing Services, pp.
110–116.
Roux, C., Chable, J., and Margot, P., 1996. Fibre transfer experiments onto car seats, Sci. Justice, 36,
143–152.
Roux, C., Langdon, S., Waight, D., and Robertson, J., 1999. The transfer and persistence of automo-
tive carpet fibres on shoe soles, Sci. Justice, 39, 239–251.
Roux, C., Talbot-Wright, B., Robertson, J., Crispino, F., Ribaux, O., 2015. The end of the (forensic sci-
ence) world as we know it? The example of trace evidence. Phil. Trans. B, 370(1674), 20140260,
doi:10.1098/rstb.2014.0260.
Salter, M. and Cook, R., 1995. Transfer of fibres to head hair, their persistence and retrieval, Forensic
Sci. Int., 81, 211–221.
Salter, M. T., Cook, R., and Jackson, A. R., 1984. Differential shedding from blended fabrics, Forensic
Sci. Int., 33, 155–164.
Schiller, W. R., 1995. Textilfasern in Anschmelzspuren, Kriminalistik, 11/95, 728–730.
Scientific Working Group for Materials Analysis, 1998. Forensic Fibre Examination Guidelines.
Washington, DC: SWGMAT–Fibre Subgroup FBI Lab. Div.
Scott, H., 1985. The persistence of fibres transferred during contact of automobile carpets and cloth-
ing fabrics, J. Can. Soc. Forensic Sci., 18, 185–199.
Siegel, J. A., 1997. Evidential value of textile fibre—Transfer and persistence of fibres, Forensic Sci.
Rev., 9, 81–96.
Spencer, R., 1994. Significant fibre evidence recovered from clothing of a homicide victim after expo-
sure to the elements for twenty-nine days, J. Forensic Sci., 39, 854–859.
Suga, K., Narita, Y., and Suzuki, S., 2014. A simple solubility tests for the discrimination of acrylic
and modacrylic fibers, J. Forensic Sci., 59, 750–753. doi:10.1111/1556–4029.12370.
Szewcow, R., Robertson, J., and Roux, C. P., 2011. The influence of front-loading and top-loading
washing machines on the persistence, redistribution and secondary transfer of textile fibres
during laundering, Aust. J. Forensic Sci., 43, 263–274.
Talalienë, D. and Vasiliauskienë, D., 1996. Evidential value of textile fibre examinations. In Proceedings
of the 4th Meeting of the European Fibres Group (27–31), London.
The Textile Institute, 1975. Identification of Textile Materials (7th ed.). Manchester: The Textile
Institute.
Wictum, E., Kun, T., Lindquist, C., Malvick, J., Vankan, D., and Sacks, B., 2013. Developmental vali-
dation of Dogfiler, a novel multiplex for canine DNA profiling in forensic casework, Forensic
Sci. Int. Genet, 7, 82–91.
Wiggins, K. and Drummond, P., 2007. Identifying a suitable mounting medium for use in forensic
fibre examination, Sci. Justice, 47, 2–8.
Microscopic Examination
of Fibres
SAMUEL J PALENIK
5
Contents
5.1 Introduction 146
5.2 The Tasks of Forensic Fibre Microscopy 146
5.2.1 Comparison 146
5.2.2 Fibre End-Use or Manufacturer 147
5.3 Instruments for Forensic Fibre Microscopy 148
5.3.1 Stereomicroscope 148
5.3.2 Polarizing Microscope 148
5.3.3 Comparison Microscope 148
5.3.4 Fluorescence Microscope 149
5.3.5 Hot Stage Microscope 149
5.3.6 Instrumental Analysis Hyphenated with a Microscope 149
5.4 Bright-Field Microscopy and Morphological Features 149
5.4.1 Vegetable Fibres 149
5.4.1.1 Seed Fibres 150
5.4.1.2 Bast and Leaf Fibres 150
5.4.2 Wool and Speciality Fur Fibres 152
5.4.3 Silk 153
5.4.4 Synthetic and Regenerated Fibres 153
5.4.4.1 Longitudinal Appearance 154
5.4.4.2 Cross-sections 155
5.5 Polarized Light Microscopy and the Optical Properties of Man-Made Fibres 157
5.5.1 Refractive Indices 159
5.5.2 Birefringence 161
5.5.3 Sign of Elongation 162
5.5.4 Isotropic Refractive Index 163
5.5.5 The Standort Diagram 164
5.5.6 Fibre Identification 164
5.6 Thermal Microscopy 167
5.7 Observing Colour on Fibres through the Microscope 168
5.7.1 Dichroism 169
5.7.2 Fluorescence 172
5.7.3 Microchemical Testing 175
5.8 Conclusion 175
Acknowledgements 176
References 176
145
5.1 Introduction
Microscopy is the essential element in almost all aspects of the forensic examination and
comparison of fibres. On the most fundamental level, microscopes are necessary to see
fibrous evidence no matter what observations or tests are being performed. Thus, micro-
scopical observation (whether by humans or by instruments) is used to detect fibres, to
characterize them, and in most cases to compare them to a suspect source. Of the three
essential components in a forensic fibre investigation (sample, analytical instrument,
and specimen preparation), only the instrument and techniques of sample preparation
are normally under the control of the analyst, since a typical sample will consist of only a
single fibre or fibre fragment.
To counteract the disadvantage of working with small quantities, i.e. minute pieces
of evidence, the analyst is forced to get the most out of tools he/she employs. In the
case of microscopes, this implies that the scientist must become a good microscopist.
That means that he or she must understand the theoretical and practical aspects of
each of the microscopes which they use, so that not only can observations and tests
be intelligently performed, but results can be rationally interpreted. Once the micro-
scope has been optimized for the particular type of observation (e.g. bright-field, com-
parison microscopy, refractive index determination, fluorescence), it is then necessary
to be certain that sample preparation is optimal. This may be as simple as ensuring
that a fibre which is to be examined by fluorescence microscopy is mounted in a non-
fluorescing mounting medium, or it may be a somewhat more difficult task, such as
preparing a cross-section to determine the actual shape and modification ratio of a
colourless trilobal carpet fibre. In any event, sample preparation, though frequently
taking a little extra time, often makes the difference between a thorough examination
and a merely cursory one. The only way to know the best method to prepare a particular
fibre for a particular type of examination is to understand both the microscope and the
type of fibre with which one is dealing. The purpose of this chapter is to help make these
relationships more clear.
5.2 The Tasks of Forensic Fibre Microscopy
The types of problem encountered by the forensic fibre microscopist fall naturally into two
categories (although occasionally ‘one of a kind’ problems may be submitted). These are as
follows.
5.2.1 Comparison
This is by far the most common type of examination request made of forensic fibre
examiners, and the only type of analysis ever made by many of them. A target fibre is
compared with fibres from a suspect source. In this case, the target fibre will normally
(but not always) be searched for on the basis of its colour or some outstanding morpho-
logical property (in the case of colourless fibres or animal textile or vegetable fibres).
This traditionally time-consuming task may soon be relieved by the newly developed
automated fibre finders which show great promise in carrying out this chore while the
analyst is employed in duties requiring human participation.
Microscopic Examination of Fibres 147
Once fibres worth comparing have been located, they are identified by genus (e.g. nylon,
acrylic, polyester). This can typically be accomplished for the majority of regenerated and
synthetic fibres rather quickly using polarized light microscopy, while the specialty fur
and vegetable fibres can be categorized by class (animal or vegetable) on the basis of their
morphological features. Once the genus of a fibre is known, the analyst can then devise a
series of tests designed to distinguish the questioned and known fibres from one another.
The nature of these tests will vary based on the type of fibre being examined. The most
characteristic features of one genus of fibre may be of little value when dealing with another,
especially when colourless or light-coloured fibres are involved. For example, in the case of
nylon fibres, infrared microspectroscopy is normally of little value compared with melting
range, whereas with acrylic fibres the opposite is true.
When the fibres in question are distinctly coloured, this property is normally the most
important comparison characteristic, with other properties generally having a secondary
significance. Comparison microscopy by bright-field, polarization and fluorescence are
the most important first observations followed by visible microspectrophotometry and,
if conditions permit and it is deemed necessary, thin-layer or high-performance liquid
chromatography. With lightly coloured and colourless fibres, properties such as polymer
composition determined from infrared microspectroscopy, refractive indices, birefrin-
gence, thermal behaviour and other tests and observations may assume added significance.
Fluorescence microscopy and microspectrophotometry may also be useful in these cases if
fibres have picked up fluorescent contaminants or contain optical brighteners or dyes that,
although faint in visible light, fluoresce under short wavelength excitation.
5.2.2 Fibre End-Use or Manufacturer

Occasionally, the fibre microscopist receives a request to identify the source or possible
end-use of a fibre which has been collected as evidence in an investigation in progress. For
example, in the Atlanta child murders in Georgia in the late 1970s, it was helpful to estab-
lish that a green trilobal carpet fibre was produced by a particular manufacturer (Deadman,
1984a and b). This led to records as to how much was made and where it was sold. In a similar
way, it may be helpful under certain circumstances to know that colourless polyester fibres
were intended for use as fibrefill (e.g. pillow stuffing) rather than in a cotton/polyester shirt or
blouse fabric. These questions can sometimes be answered through an understanding of the
properties of different fibres within a genus and the methods for extracting this information
from single fibres. They also require familiarity with the fibres which are being made and the
companies which produce them, a skill gained only through experience.
Such questions can occasionally be answered. In the Atlanta case, the unique shape
of the cross-section, created to circumvent a patent, led directly to the manufacturer of
the carpet fibre. In the second example, a likely end-use might be determined by an exact
measurement of the birefringence of the fibre and examination of its cross-section. Colour
usually plays a minor role in such studies, but it can occasionally be of great significance.
For example, one could cite the importance of finding that a fibre was solution dyed, since
the pigments used to colour it must have been selected by the fibre manufacturer and intro-
duced by it into the polymer before it was extruded. The chances of success in such research
are directly proportional to the experience and resources of the analyst called upon to
conduct the investigation. The fact that these investigations are even sometimes successful
illustrates the point that they could probably be undertaken more often than they are.
5.3 Instruments for Forensic Fibre Microscopy
Several types of microscope are necessary in any laboratory in which forensic fibre analyses
are to be conducted. A stereo microscope and a polarizing microscope should be regarded
as a minimum, and other microscopes are highly desirable for some types of observations
and essential for others. A desirable assortment in the well equipped laboratory would
include the following:
5.3.1 Stereomicroscope
This is used for low-magnficication examinations, where its long working distance and
upright image assist in searching small items of evidence or tapes and in recovery and
manipulation of individual fibres. The microscope should be equipped for observa-
tion in both transmitted and reflected light. Accessories can be obtained which permit
observations between crossed polars and with fluorescence. A boom stand is useful since
it permits the microscope to be used over large objects which could never fit the stage of
a conventional instrument. Hand rests attached to the transmitted light base are also a
convenience when manipulations are being performed.
5.3.2 Polarizing Microscope

The polarizing microscope is similar to a conventional biological microscope with addi-
tional features which facilitate observations and measurements with plane polarized
light. These refinements include a rotating stage graduated in degrees, a polarizer located
beneath the condenser, another polarizer (called the analyzer) located above the objec-
tives, a slot for oriented crystal plates called compensators and strain-free objectives. One
of the eyepieces should be equipped with a cross-hair graticule which accurately reflects
the vibration directions of the polarizer and analyzer and incorporates a micrometre
scale for making measurements. A rotating analyzer and a swing-out polarizer are also
advantageous.
The polarizing microscope permits the microscopist to observe and measure the
principal optical properties of a fibre regardless of size. A rapid estimation of these
optical properties is normally sufficient to allow the generic type of most synthetic fibres
to be identified within moments. One of the first people to recognize the value of this
for forensic examinations was Culliford (1963). Exact measurements can be made when
they will help to strengthen a comparison or give information about a potential end-
use or manufacturing source. Observations of dichroism on coloured fibres can also be
made. Some laboratories have equipped their comparison microscopes with equipment
for observations in polarized light, which has obvious advantages for those who can
afford the expense.
5.3.3 Comparison Microscope

This instrument permits direct comparison of such fibre properties as morphology, diam-
eter and colour. The utility of a comparison microscope is improved if it is arranged with
equipment for polarized light and fluorescence. Before any comparison microscope is put
into service, it should be tested for light balance and magnification.
5.3.4 Fluorescence Microscope

This microscope should be of modern design, equipped for incident illumination. A selec-
tion of broad band excitation filters should be available covering the range from ultra-
violet through violet, blue and green. Narrow band interference filters are unnecessary,
unless there is a special reason for using them, since the purpose of this method is to search
for and observe fluorescence originating from dyes, optical brighteners or contaminants
and not to excite particular fluorochromes such as those used in immunohistochemistry.
5.3.5 Hot Stage Microscope

While not actually a microscope, but an accessory, this is important enough to list as
an essential piece of equipment. A hot stage for fibre work must be capable of reaching at
least 300°C and should fit the stage of the polarizing microscope. It should be calibrated
with melting point standards supplied by the manufacturer of the instrument, since
melting points determined microscopically will differ from those determined by conven-
tional techniques such as the capillary tube method used by organic chemists.
5.3.6 Instrumental Analysis Hyphenated with a Microscope

UV-vis microspectrophotometers (UV-vis MSPs), infrared microspectrophotometers
(micro-FTIRs) and Raman microprobes have established themselves as important ana-
lytical tools for the forensic examination of fibres. This situation warrants their own
chapters (Chapters 6–8).
5.4 Bright-Field Microscopy and Morphological Features
Natural fibres such as vegetable, wool and speciality fur fibres rely on a careful study of
their morphological features for their identification and comparison. For the vegetable
fibres, unless they are coloured, accurate identification is as far as comparison can be
carried. Thus, to be able to say with certainty that two bast fibres are, for example, ramie
(Boehmeria nivea) would be as far as that particular identification and comparison could
be carried by microscopical means. This is also true of wool and the speciality fur fibres,
although these are frequently coloured, giving an additional property for comparison.
The three most important aspects of natural fibre identification are the accumula-
tion of authentic reference specimens, sample preparation and careful observation of the
most characteristic features by the appropriate means. It is essential in this regard that
the analyst become used to using the true scientific names of the plants from which the
fibres are derived. In this way, problems inherent in the use of common names are avoided.
Methods of sample preparation and the types of observations which can be made are
discussed here.
5.4.1 Vegetable Fibres

Background on these fibres is given in Chapter 1. This discussion is limited to the non-
woody fibres which are used in textile, cordage and packaging applications and includes
representatives of the following groups.
5.4.1.1 Seed Fibres

The fibres grow from a single epidermal cell from the outer surface of a seed. Thus they will
always, if not broken, show the open end where they broke away from the epidermal cell.
An entire seed fibre consists of a single cell.
The most commonly encountered seed fibre for the forensic microscopist is cotton
(Gossypium spp.). Other seed fibres, with the exception of kapok, are rarely used com-
mercially, but it is not impossible that the analyst might encounter seed fibres from cattail,
thistle, and milkweed and may want to learn to recognize them.
Cotton can be recognized by its flattened and twisted appearance and its failure to
go to extinction between crossed polars (a characteristic which it shares with many other
vegetable fibres). Since it is ubiquitous, colourless cotton has virtually no evidential value,
although at times certain questions may arise which can be answered. For example, the
presence or absence of an optical brightener, as shown by fluorescence microscopy, may
answer the question of whether or not a particular fibre is likely to have originated from
a certain textile. Mercerized cotton can usually be recognized since the process swells the
fibres and deconvolutes them. Since most cotton is coloured, this characteristic becomes
the most important attribute in the comparison of these fibres. The first step is to deter-
mine how the colourant is applied. Most fibres are dyed, but the analyst should be aware of
naturally coloured cottons which have entered the marketplace, albeit in small quantity, in
the past few years (Anon, 1996). When the dye is observed to be irregularly smeared on the
surface of the fibre, it is usually a sign that the fibre originates from a printed fabric. Colour
comparisons are initiated on the comparison microscope. If a microspectrophotometer is
available, the absorption spectra of the known and questioned fibres should be acquired
and compared. For deeply coloured fibres, when a spectrophotometer is not available or
to further the comparison, thin layer chromatography may be attempted. Fluorescence
microscopy will show the presence of optical brighteners and cause certain dyes to flu-
oresce under particular excitation conditions. A range of excitation wavelengths should
always be tried to determine the fluorescence characteristics of even weakly dyed cotton.
If at all possible, cotton and other vegetable fibres should be mounted in water, glycer-
ine or glycerine jelly. These media swell the fibres to their characteristic maximum size and
fully display the fibres’ morphological characteristics, and are not themselves fluorescent.
Neither water nor glycerine mounts are permanent, however, and many analysts prefer
resinous mountants for all their fibre preparations. No matter which mounting medium is
utilized, both known and questioned fibres should be mounted in the same material.
5.4.1.2 Bast and Leaf Fibres

These fibres, with the exception of linen (Linum usitatissimum), ramie (Boehmeria nivea)
and recently hemp (Cannabis sativa), are usually encountered as the technical fibre, i.e. as
a fibre bundle consisting of individual cells (fibre ultimates) along with other cells and
tissues which are part of the stem or leaf from which the fibre of commerce originated.
Thus for all practical purposes the technical fibre is considered the individual fibre in terms
of both transfer and analysis.
The identification (and thus comparison) of these fibres must be based on the ana-
lyst having previously acquainted himself or herself with the characteristic properties of
authenticated specimens either from plants backed up by vouchered herbarium specimens
or secondary standards verified by a competent microscopist with access to the primary
standards. The samples are taken through a series of tests and observations designed to be
carried out on a single technical fibre, and to exploit all the distinctive properties of that
fibre. Such a series, devised by the author for his own education in this area, is given in
Table 5.1. By using a single technical fibre all the way through the process, the microscopist
ensures that an important observation made on any part of the fibre applies to all of it. It
is possible that technical fibres in a particular product can be mixed. If this were to occur,
the results could be confusing or lead to an erroneous conclusion if single technical fibres
were not used throughout the analysis as the sample unit.
Table 5.1 Procedure for the Identification of Bast and Leaf Fibres
I. Examination of the Technical Fibre
1. Preliminary examination. Examine the fibre under the stereomicroscope and observe its colour, texture
and stiffness.
2. Fibre scrapings. Scrape the length of the fibre gently with a razor, collecting the scrapings over a glass
slide. Use the tip of the razor to collect the fine particles together in the centre of the slide, being certain
to scrape off the residue on the tip of the razor and add this to the little pile of debris. Mount the residue
in a drop of glycerine–alcohol (1:1) and examine with the polarizing microscope for:
A. epidermal tissue
B. crystals of calcium oxalate
C. other cells and tissues such as spiral thickenings and parenchyma cells.
3. Cross-section. Prepare a simple cross-section of the entire technical fibre and observe the overall shape
and the shapes of the individual cells.
4. Test for degree of lignification. Take a small piece from the cross-section and add a drop of phloroglucinol
reagent.a Observe the colour, which will range from colourless to pink to deep red.
5. Clear the fibre. The technical fibre is cleared by boiling it on a slide under a coverlslip in a 1:1 mixture of
alcohol and glycerine. The alcohol evaporates and the glycerine (with a refractive index of 1.48)
penetrates the fibre and displaces the air. Examine the interior of the cleared fibre for crystals or other
characteristics.
II. Examination of the Fibre Ultimatesb

1. Length. The average length of the ultimates should be determined by either actual measurement or
estimation.
2. Dislocations and nodes. The presence, shape and frequency of the dislocations along the length of the cells
should be noted. In general, the number and complexity of dislocations is greater in the bast than in the
leaf fibres.
3. Lumen and cell diameter. The thickness of the central canal (lumen) in the individual cells is noted with
respect to regularity and its thickness with respect to the fibre diameter.
4. Herzog test. An individual cell, isolated from a single technical fibre, is aligned parallel to the vertical
cross-hair of the polarizing microscope in what would be an extinction position for a synthetic fibre. The
first-order red compensator is then inserted and the colour of the cell wall is noted. If it is yellow the
cellulose helix has an S twist, and if it is blue, a Z twist.
5. Other tissues or cells. These should be observed, noted and compared to similar cells isolated from known
fibres.
a Phloroglucinol reagent: a bright red colour appears if the material being tested contains lignin. A nearly satu-
rated solution of phloroglucinol in alcohol mixed with an equal quantity of concentrated hydrochloric acid.
b The fibre ultimates, as well as other cells and crystals, are gently released by treating the technical fibres in a
small test tube with enough of a freshly prepared mixture of 30% hydrogren peroxide and glacial acetic
acid to cover them. The tubes are placed in a small beaker of water and heated until the fibres turn white. This
will take about 4–8 hours. Under no circumstances should the tubes be left unattended and allowed to go to
dryness, since explosive peroxides may detonate in the dry tubes. The fibres are removed from the tubes, washed
and teased out on a microscope slide and then mounted in water, glycerine or glycerine jelly. A stain may be
used (e.g. methylene blue), but some of the fibre ultimates should always be left unstained for the Herzog test.
A few comments regarding some of the suggestions made in the table may help to
clarify some points. The fibre scrapings and fibre clearing are an alternative to the com-
monly recommended process of ashing the fibre. Crystals can be seen clearly in either
of these preparations, especially between crossed polars. The microscopist can observe
the distinctive crystals directly and not merely as pseudomorphs left behind after the
ashing process. In the author’s opinion, ashing merely destroys evidence. A cross-section
is probably the simplest way to distinguish bast and leaf technical fibres. In bast fibres
the outer boundaries of the cells are polygonal, while in the leaf fibres they tend to be
rounded. Cross-sections are also particularly useful for distinguishing among many
of the jute substitutes. Here the external shape of the technical fibre seems to be most
useful. The test for lignin with phloroglucinol will be negative for fibres consisting of
practically pure cellulose such as linen and hemp, while true jute (Corchorus spp.) will
assume a bright red colouration. Partially lignified fibres such as manila (Musa textilis)
stain pink (The Textile Institute, 1975).
While the length of the fibre ultimates can be measured, a more practical approach is
to determine the length relative to the field of view. Using a low magnification (e.g. a 4×
objective and 10× ocular), observe the cells spread out in the field of view. For example,
fibres of coir (Cocos nucifera) are very short and it would take several of them to bridge
the field of view. Fibres of Corchorus spp. fit easily inside a single field, while linen fibres
extend well outside the field of view even when curled back on themselves. Fibre ultimates
of ramie (Boehmeria nivea) can reach an inch in length. The Herzog test requires practice
on known fibres before it can be relied upon for the distinction between flax and hemp
(for which it was originally designed). The theory of this test is explained by Valaskovic
(1991). For details on the interpretation of these observations the reader is referred to the
excellent works by Catling and Grayson (1982) and Luniak (1953).
5.4.2 Wool and Speciality Fur Fibres

Morphology is also the key to the identification of these fibres, while colour is once again
the principal means of comparison. The wool and fur fibres are identified as such on
the basis of surface scales comprising the cuticle. The distinction between sheep’s wool
(Ovis spp.) and the speciality fur fibres such as cashmere (Capri falconeri), mohair
(Capri spp.), camel (Camelus spp.) and alpaca (Llama glama) is based on careful obser-
vation of a limited number of features. The situation is complicated by the fact that
different breeds of sheep provide wool of different colours and diameters. Experimental
programmes to cross-breed many of the animals which produce the speciality fur fibres
may result in fibres with intermediate properties. An extensive collection of known wool
from sheep of different breeds and the speciality fur bearing animals should be available
in the laboratory for training and reference purposes. The United States Department of
Agriculture sells sets of wool and mohair grade standards consisting of authenticated
specimens of these fibres, classified by diameter, which is a useful addition to the labora-
tory’s reference collections.
The principal morphological characteristics which can be observed and compared
on wool include diameter, scale thickness, scale prominence and scale count. The diam-
eter can be measured directly using a calibrated ocular micrometer, or fibres can be
compared directly with a comparison microscope. Fibres for use in clothing will origi-
nate from sheep breeds such as the merinos and average 17–24 μm in diameter. Other
breeds may produce coarser fibres for use in blankets and carpets. These may contain
medullary cells that may at first be confused with the hair of other animals. However,
they are frequently dyed, which gives a clue as to their origin. The finest wools have the
same diameter as cashmere, while mohair is somewhat thicker. The classical method of
distinguishing wool, cashmere and mohair has been to compare the diameter, deter-
mine the number of scales per 100 μm and observe the prominence of the scales. This
procedure can give reasonable results, but the accuracy is less than 100%, as Wortmann
and Wortmann (1991) have shown. This problem is particularly troublesome when the
distinction to be made is between the finest sheep’s wool and cashmere. Blind trials
have shown that scale thickness, as measured in the SEM, is a reliable guide in this
regard, with blind trials giving accurate diagnoses 100% of the time. Wortmann’s origi-
nal papers should be consulted for details.
In light of these difficulties, the analyst should be cautious when identifying a wool-
len fibre by name. It is often sufficient simply to refer to these fibres as wool. Two wool
fibres that originate from the same source (regardless of their animal origin) would then
be expected to show the same approximate diameter, scale count and scale prominence as
well as colour. Colour is compared in the same manner as described for cotton above.
5.4.3 Silk
Cultivated silk from the silk worm (Bombyx mori) is the most commonly encountered
type, although so-called wild silks are also articles of commerce. Tussah (Antherea
paphia) is the most common of the wild varieties. These two types can be readily dis-
tinguished by their cross-sectional shapes and diameter. Cultivated silk is finer and has
a somewhat triangular cross-section, while tussah silk is broader with striations along
its length and has a ribbon-like cross-sectional appearance. In the not too distant past,
cultivated silk was easily distinguished from all other fibres by its narrow diameter, but
the advent of microfibres has changed this. Now the microscopist must be more care-
ful, since nylon microfibres and silk can be confused because of the similarities in their
diameter and their infrared spectra. Silk, however, is normally less regular in appearance
along its length than a microfibre. An easy way to observe this irregularity is between
crossed polars using the interference colours in the same way that one views a topo-
graphic map. The most definitive difference, if difficulties are encountered, is to place a
short segment from the fibre (a cross-section will do nicely) in the hot stage. Nylon will
melt while silk will not.
Colour is the principal point of comparison once it has been established with certainty
that the fibre is silk and what type of silk it is. Fluorescence microscopy may provide addi-
tional features based on any fluorescence of the dyes.
5.4.4 Synthetic and Regenerated Fibres

Since the man-made fibres cannot be identified on the basis of their morphology as can
the natural ones, the bright-field microscopy of these fibres is limited to the features used
for comparison. Identification is achieved through a determination of the fibre’s optical
properties or from an infrared spectrum. A variety of features may be observed on fibres
which are useful for further classification and to aid in trying to reduce the possibilities for
comparison. Some of the most common are as follows.
5.4.4.1 Longitudinal Appearance

The analyst should note the appearance of the fibre as it lies on the slide. Does it appear
to be circular in cross-section or does it appear to be modified? Examination between
crossed polars, although not a bright-field technique, can allow the surface topography
of the fibre to be estimated to a first approximation. The surface of the fibre should be
examined and one should note if it is smooth or striated and, if striated, the relative
lengths of the striae noted. This examination should be conducted with the fibre either
unmounted (i.e. in air) or in an index liquid far from the average refractive indices of the
fibre. Otherwise a lack of contrast makes it difficult to distinguish surface features from
internal structure.
Internal features and inclusions can best be characterized and identified if the sur-
rounding polymer is made invisible by mounting in a liquid near or equal to one of the
refractive indices of the fibre. For example, n⊥ for nylon fibres is close to 1.520. If the fibre
is mounted in a liquid of RI 1.520 and is oriented on the stage of a polarizing microscope
so that this vibration direction is parallel to the polarizer of the microscope, the fibre will
seem to disappear. Any inclusions, such as delustrant, pigment or drawn-out spherulites
will become visible since the polymer is now invisible. In the example of a nylon fibre, the
delustrant would probably be sub-micrometre crystals of titanium dioxide with refrac-
tive indices much higher than those of the fibre. However, other delustrants may be used.
In the past, Monsanto made nylon carpet fibres containing a paraffin wax delustrant.
Fibre companies are experimenting with new products all the time, so it should not be
assumed that delustrant particles are all titanium dioxide. The density of delustrant par-
ticles is a useful point of comparison. It should not be assumed that all particles in a
fibre are delustrant. Flame retardants such as antimony oxide are added to some fibres,
particularly modacrylics intended for sleepwear. Although these crystals are white, they
tend to be larger than pigment-grade titanium dioxide. Certain fibres, particularly poly-
propylene and some others such as nylon for use in automotive carpet, are ‘solution dyed’.
This unfortunate choice of words denotes a fibre which has been coloured by inclusion
of a pigment as the fibre was manufactured. Polypropylene fibres, having no functional
groups to react chemically with dye molecules, are almost always solution dyed, although
manufacturers continue to experiment with modifying these fibres so that they can be
dyed with conventional dyestuffs. To observe these pigments in a polypropylene fibre, it is
mounted in 1.520 index of refraction oil and rotated so that the length of fibre is parallel
to the polarizer. This is approximately equal to n parallel for the fibre, causing it to disap-
pear so that the pigment crystals stand out in contrast. Examination of a fibre mounted in
this manner provides an extremely sensitive method of comparison, for if it is examined
at high magnification (with a high dry or oil immersion objective) and searched care-
fully along its length, one can observe trace amounts of pigments which may be present
as either minor components or contaminants. Thus, for example, the examination of a
questioned polypropylene fibre revealed the presence of single crystals of a blue pigment
at the level of approximately one per field of view, interspersed among the carbon black
particles. The identical pigment was found at the same level in the known fibres submitted
for comparison.
Before leaving the subject of longitudinal appearance, it should also be pointed out
that the effects of texturing on the fibre should be noted. A variety of methods are used
to add bulk to fibres (Backer, 1972). These manifest themselves on the external shape,
usually in the form of a regular or irregular crimp. In some instances, it is possible to pos-
tulate the crimping process (e.g. gear, stuffer box, false twist) from an examination of the
fibre along its length. These examinations are usually best performed on the unmounted
fibre under a stereo microscope, although occasionally a mounted fibre or cross-section
will be necessary to confirm the method used.
5.4.4.2 Cross-sections
The cross-section of a fibre can usually be inferred to a first approximation by the methods
described above. However, there is no substitute for an actual fibre cross-section in estab-
lishing the true nature of this aspect of a fibre. The cross-section of a fibre may result from
one of three factors (Palenik and Fitzsimons, 1990a), as follows.
• As a result of the process used to manufacture the fibre. For example, the crenulated
cross-section of a viscose fibre is due to shrinkage as it is regenerated in the acid
bath. Similarly, the dogbone shape of a dry spun acrylic fibre (such as Orlon®) is a
result of the collapse of a round fibre in the dry air stream as the organic solvent
is evaporated.
• It has been engineered for specific reasons. Engineered cross-sections are normally
observed on melt spun fibres such as nylon, polyester and polypropylene, with
nylon getting the most attention. The most commonly encountered cross-sectional
design in nylon fibres is a trilobal shape. However, within this broad classification
there are many possible variations and many of them have been the subject of
patents.
One way of numerically designating a trilobal fibre is by means of its modification

ratio (MR). This is the ratio of the diameters of two circles inscribing and circumscribing
the cross-section. Figure 5.1 illustrates how this number is derived and, as can be seen, is a
way of expressing a deviation from circularity, since a round fibre will have a modification
ratio of 1. Examples of several trilobal fibres with their modification ratios are shown in
Figure 5.2.
Although they can be difficult to produce, almost any idea for a novel cross-section
can be produced today. Only some of these ever make their way to market, but even the
MR = R
r
Figure 5.1 Method of determining the modification ratio. (From Palenik, S. and Fitzsimons, C.,
Microscope, 38, 313–320, 1990b. With permission.)
Figure 5.2 Cross-sections of fibres and their modification ratios. (From Palenik, S. and
Fitzsimons, C., Microscope, 38, 313–320, 1990b. With permission.)
ones that do can be interesting and, if difficult (and thus expensive) to make or subject to a
patent, can be characteristic and thus traceable to a particular manufacturer.
• It has been modified by processing or accident. Various texturing processes may

characteristically modify the shape of the cross-section. For example, false twist
texturing produces a polygonal shape which is quite characteristic of that process.
Other processes can produce modifications, and the forensic microscopist must
be alert for any subtle features which can help to aid in a comparison. Accidental
causes such as crushing are too obvious to require further comment.
In addition to the shape of a fibre, a good deal more can be learned by the examination of
a cross-section – just how much will often depend on the quality of the section produced.
Although the best sections are produced by a microtome, classical microtomy is not always
the best solution for forensic purposes. High-quality microtomy requires that the sample be
embedded, located and sectioned, usually on an expensive instrument operated by an experi-
enced microtomist. The requirements for forensic fibre microscopy are that a thin section be
produced from a small fibre or fibre fragment, which is perpendicular to the fibre length, and
that the remaining piece (or pieces) be recoverable after the section has been produced. These
requirements are all met by the polyethylene film method devised by Palenik and Fitzsimons
(1990b). Although this method is ideal for single fibres, it is not efficient when greater numbers
of fibres must be sectioned (e.g. when a carpet tuft must be cut to determine the full range of
cross-sections present). In these cases, the Joliff method using perforated plastic slides (Heyn,
1954) or fibre microtomes such as those of Hardy (1935) or Schwartz can be utilized.
It must be recognized, however, that some of the observations described below can-
not be satisfactorily performed on sections produced by the Joliff method because of the
thickness of the sections or the presence of packing fibres. Additional observations which
can be made on thin sections of man-made fibres, such as those made by the polyethylene
film method include the following.
• Dye penetration. Exhaustive dyeing can be distinguished from ring dyeing, and the
depth of dye penetration of ring dyed fibres can be compared. In addition to obser-
vations by bright-field transmitted light, an examination with various wavelengths
of fluorescent light can be advantageous, especially when one dye component
Figure 5.3 Cross-section of a ring dyed trilobal nylon fibre; magnification 960x.
fluoresces and the other does not. In these cases, a comparison between the bright-
field and fluorescence images may be rewarding. In an analogous manner, the dis-
tribution of pigments in solution dyed fibres, as well as delustrant particles, can be
compared. The cross-section of a ring dyed fibre is shown in Figure 5.3.
• Examination of spherulites. Many fibres crystallize as spherulites. Nylon and
polyolefins exhibit classic examples of this behaviour. In nylon fibres, particularly
some large denier carpet fibres, conditions may be right for the formation of spheru-
lites which are large enough to observe by light microscopy. Although these crystals
originally grew with spherical symmetry, drawing causes them to elongate along
the fibre axis as illustrated in Figure 5.4. If they are large, they can be seen under
the microscope as elongated streaks inside the fibre, when the fibre is mounted in
a medium matching one of its refractive indices. When observed in cross-section
these spherulites have a round outline and are a good deal larger than delustrant
crystals. If the polars are crossed, a ‘Maltese cross’, which remains in the same posi-
tion as the stage is rotated, can be observed inside each of them (Figure 5.5).
• Air spaces and gas voids. A number of fibres are now produced with air channels
arranged in various patterns. Cross-sections can clearly show the arrangement
and number of these channels, which it is often impossible to establish in lon-
gitudinal view. Some fibres, especially certain acrylics and modacrylics, contain
gas voids which have been purposely introduced during manufacturing. The true
nature and arrangement of these features across the diameter of the fibre can be
clearly seen in thin sections.
5.5 Polarized Light Microscopy and the Optical

Properties of Man-Made Fibres
The polarizing microscope is perhaps the single most important instrument for identi-
fying fibres and performing comparative fibre analyses. To utilize it fully and success-
fully, the analyst must understand its use and the attendant optical theory of polarized
light. These topics are treated in numerous texts, some of which are listed in the references
(Hartshorne and Stuart, 1970; Bloss, 1961; McCrone et al., 1978). It is possible here only to
Figure 5.4 Drawn spherulites seen parallel to the axis of a nylon fibre, mounted in a liquid of
refractive index 1.520. When the fibre axis is parallel to the polarizer, the spherulites become
invisible; magnification 593x.
Figure 5.5 Cross-section of a nylon fibre, showing ‘Maltese crosses’ in the spherulites when
viewed between crossed polars; magnification 960x.
describe briefly the origins of the principal optical properties of fibres, to illustrate how
they relate to fibre identification and comparison, and to summarize the methods by which
they may be determined.
Four basic optical properties (excluding colour) can be determined for any fibre. These
properties serve to characterize the man-made fibres to the extent that, if they are esti-
mated for an unknown fibre, the genus of the specimen can be quickly and accurately
determined. If the properties are carefully measured, they can at times provide valuable
comparative data. These properties are: the refractive indices, isotropic refractive index,
birefringence and sign of elongation. They are the same properties by which optical crystal-
lographers have been identifying minerals with the petrographic (i.e. polarizing) micro-
scope for over 150 years.
5.5.1 Refractive Indices

To a first approximation, all fibres have two refractive indices, one parallel to the length of
the fibre (n||) and the other perpendicular to the length n⊥. These two values reflect the fact
that fibres are anisotropic, with certain properties along the length of the fibre which are
different from those perpendicular to the length. An average refractive index, the so-called
isotropic refractive index (niso), can be calculated from the formula
[n + 2(n⊥ )]
niso =
3
This value approximates the average refractive index of the fibre were it not anisotropic,
but simply an isotropic solid. If the isotropic refractive indexes of the various man-made
fibres are compared, it will be found that they differ from one another, but are similar to
the values given in the literature for the bulk polymers from which they are prepared. This
is because the refractive index of a (polymer) molecule is a result of its molar refractions,
which are in turn derived from the atomic refractions of the atoms of which the molecule
is constructed (Hartshorne and Stuart, 1970). Since most synthetic and regenerated fibres
are made from the same elements (carbon, hydrogen and oxygen), the isotropic refractive
indexes and anisotropic refractive indices of most synthetic fibres all fall within a limited
range of about 1.5. As a dramatic example of the effect of atomic refraction on refractive
index of a polymer, it is necessary only to consider Teflon® fibres which have an isotropic
refractive index of about 1.35. This extremely low value is due to the unusually low atomic
refraction of fluorine, which in turn results from its unique electronic structure.
Light waves, which are oscillating electric vectors, affect the electronic fields of the
binding electrons through which they pass as they propagate through matter. Their effect
is to form dipoles, whose strength is determined by the polarizability of the molecules along
which the ray passes. It is from these bond polarizabilities that molecular refraction origi-
nates (i.e. they are proportional to the ease with which a light ray can pass through an elec-
tric field). The majority of molecules in an oriented polymer fibre are normally arranged so
that the length of the molecules (or polymer chains) run parallel to the fibre’s axis. When
the fibre axis is oriented so that plane polarized light passes along it, the individual rays of
light interact with individual molecules oriented in the same general direction. In particu-
lar, the electric vector (the light ray) and the electronic configuration of the binding elec-
trons affect one another. This interaction manifests itself in the refractive index which we
can measure along the fibre axis. For most fibre forming polymers, the bond polarizibility
is greatest along the length of the molecule, resulting in a higher refractive index paral-
lel to the length of the fibre than perpendicular to it. However, in some cases the bond
polarizability is stronger perpendicular to the length of the fibre than parallel to it. Fibres
of polyacrylonitrile (PAN) are an excellent example of this. The nitrile groups (–C≡N)
are pendant to the carbon–carbon chains in these fibres and thus are oriented roughly
perpendicular to the length of the molecules. The result is that n⊥ in PAN fibres is greater
than n||. This is so unusual that these fibres can usually be identified on the basis of this
characteristic alone. Since most synthetic fibre molecules are based on a carbon–carbon
chain, n values tend to be in about the same region of 1.5. However, polyester fibres, in
which phenyl groups form part of the molecular chain, have n|| values of 1.7 and greater.
This is due to the highly polarizable π electrons associated with the benzene rings and the
effects of induction along the polymer chain.
In practice, the refractive indices of a fibre can be determined by the immersion method,
in which a fibre with one of its principal vibration directions (n|| or n⊥) aligned parallel to the
polarizer of the microscope is successively mounted in index of refraction oils until the liq-
uid–solid interface disappears. The relief (contrast) is used to estimate the difference in refrac-
tive index between the fibre and index of refraction oil. As an index match is approached, the
contrast decreases. The Becke line is used to determine whether the fibre or liquid has the
higher refractive index. Details of the method can be found in any textbook of optical crys-
tallography. With temperature correction and monochromatic light, the refractive indices
of a fibre can be measured accurately to the third decimal place (The Textile Institute, 1975).
The refractive indices can also be measured with much greater accuracy using interference
microscopy (Figure 5.6). Heuse and Adolf (1982) have described the method in detail and its
application to forensic science.
Figure 5.6 Zeiss interference microscope showing (right) refractive index oils, differing in RI
steps of 0.002, and portable fume extraction hood.
5.5.2 Birefringence
The birefringence (Γ) is the numerical difference between n|| and n⊥. It is an aid in fibre
identification, and within a particular fibre genus provides a relative measure of molecular
orientation. It can be obtained by subtraction after determining the refractive indices as
outlined above (i.e. n|| – n⊥), or calculated from the thickness at a point where the retar-
dation (amount by which one ray in an anisotropic medium has been slowed down with
respect to the other) is known by the relationship
R
Γ=
1000T
where R is the retardation in nanometres and T is the thickness in micrometres. Strictly

speaking, it is the path difference and not the retardation which is determined, but the tra-
ditional term retardation is used here (see The Textile Institute (1975) for a clear discussion
of this subject).
The retardation colours can be observed and the retardation estimated from the
Michel–Lévy chart or they can be estimated with a quartz wedge. Such an estimate is
normally adequate when the purpose is merely to identify the fibre genus. The birefrin-
gence of a fibre can be estimated from the retardation colours which are observed when
the polars are crossed and the fibre rotated into its position of maximum brightness at 45°
to the cross-hairs (i.e. 45° to the polarizer orientation). For an estimation this approach
can be justified because most textile fibres have diameters which lie in a similar range.
Thus, fibres showing first-order grey colours are probably acrylic, modacrylic or acetate
fibres. They could not be nylon, polypropylene or polyester. Fibres with bright second-
order colours are probably nylon, polypropylene or perhaps rayon. High-order white prob-
ably indicates polyester or a specialized high-performance fibre. In this way, many fibres
can be eliminated from consideration and the possibilities reduced almost at a glance.
The newly introduced microfibres are changing these assumptions. Since they are consid-
erably thinner than most normally encountered man-made fibres, they cannot be drawn
as much as ordinary fibres or they would break. Thus they have birefringence values that
are lower than might be expected.
On occasion, it may be useful to determine the birefringence of a fibre exactly. The
birefringence may help to infer the end-use of a fibre. For example, high-strength nylon
for use in ropes and shoelaces has a birefringence of approximately 0.060. Nylon used in
carpets has values ranging from about 0.052 to 0.057. Nylon for use in luggage is lower
still. For example, the exact birefringence of a colourless polyester fibre could be of value
in distinguishing fibres from a cotton/polyester fabric and from a tyre cord. The exact
measurement of birefringence may be of comparative value in some cases, but in others
it would be a waste of time. Consider, for example, a colourless polyester fibre fragment
found on the blade of a knife. If the fibre originated from a cotton/polyester fabric used
to make a shirt or blouse. It may have been spun and drawn under carefully controlled
conditions so that all the fibres have the same molecular orientation and thus the same
birefringence. While the chances might be good in this case that an exact measurement
of the birefringence would be a useful comparison characteristic, the only way to decide
this for certain would be to measure the birefringence of a number of fibres from the shirt
in question. On the other hand, if the suspected source of the fibre was a pillow, filled with
polyester fiberfill, the chance that the fibres were produced under carefully controlled
conditions would be small; still, the only way to know for certain whether birefringence
would be of value in this comparison would be to measure some of the known fibres.
In measuring thickness, the task is simple for a fibre with a circular cross-section:
the thickness is the same as the diameter and the problem is trivial. For modified cross-
sections, obtaining an accurate thickness becomes more of a problem, but the method
described by Gaudette (1988) can be used successfully with a little experience. Gorski and
McCrone (1998) have proposed a general solution based on tested assumptions. Still greater
accuracy can be obtained if a cross-section is cut, using the polyethylene film method, at
the same location at which the retardation is measured.
With a good value for thickness, the problem is reduced to obtaining an accurate value
for the retardation at the point where the thickness is known. If the retardation is less than
or equal to about four orders, this can be determined with good accuracy using a tilting
compensator of the Ehringhaus or Berek varieties. The fibre is rotated into a subtractive
position (see section 5.5.3), in which the slow vibration directions (high refractive indices)
of the compensator and fibre are perpendicular to one another. The compensator is then
rotated until the fibre becomes black, indicating complete compensation. The retardation
is then read from a table (Ehringhaus) or calculated (Berek). If the retardation is greater
than four orders, it is likely that false compensation bands will be obtained. This problem
is a result of the fact that the inorganic crystals of which compensators are constructed and
organic fibres have very different dispersion curves (i.e. change in refractive index with
wavelength). The result is that wildly inaccurate values of retardation are obtained and, if
not recognized, may be used to calculate improbable values of birefringence. An elegant
solution to this problem was described by Seminski (1975), who made a wedge cut on the
edge of the fibre, which in monochromatic light permits the almost effortless counting of
full orders of retardation. The fraction of an order of retardation at the thickest point on
the fibre is then determined with a compensator and the two values are added. In this way
very accurate values can be obtained for fibres with very high birefringence.
5.5.3 Sign of Elongation

The convention for this property is that if n|| is greater than n⊥, the fibre is said to be posi-
tive (+). If the opposite obtains, the fibre is said to be negative (–). If both refractive indices
of the fibre have been measured, the sign is determined by subtraction. In practice it is
usually determined by means of a compensator. The fibre should be centred and placed
in the 45° position. If the retardation colours are one order (530 nm) or less, the sign is
most often determined with a first-order red compensator. The fibre should be centred
and placed in the 45° position with crossed polars. This compensator is sometimes called a
sensitive tint plate. When retardation of 100–200 nm is added, the colour becomes second-
order blue or blue-green; when the same amount of retardation is subtracted, the object
becomes yellow or orange. For retardations greater than first order, a quartz wedge or
tilting compensator can be employed.
The sign of elongation of most fibres, as stated previously, is usually positive. However,
there are certain exceptions. The most notable are the acrylic and modacrylic fibres, most
of which have negative signs of elongation. However, as the amount of polyacrylonitrile
in a modacrylic fibre falls beneath a threshold limit, the fibre becomes optically positive.
This can be observed with Verel® and Dynel® modacrylic fibres. Verel has a negative and
Dynel a positive sign of elongation, reflecting the relative amounts of polyacryonitrile
in these two fibres. Recall that modacrylic fibres are required to contain only 35% poly-
acrylonitrile by United States Federal Trade Commission rules. Verel and Dynel are no
longer manufactured. Most modacrylics in current production (Grieve and Griffin, 1999)
have positive elongation distinguishing them from acrylic fibres, but within the range of
Kanekalon fibres variations can be found.
Secondary acetate fibres have a low birefringence and typically exhibit first-order
grey interference colours between crossed polars. In a triacetate fibre, with approximately
90% of its hydroxyl groups esterified, the birefringence decreases as n⊥ increases. This is
due to the high molar refraction of the acetate groups and their position on the cellulose
molecule, roughly perpendicular to the polysaccharide chain. Thus the sign of elongation
of triacetate fibres can range from negative to positive, with most of them appearing to be
practically isotropic until the first-order red compensator is inserted.
5.5.4 Isotropic Refractive Index

This number is an approximation of the refractive index of the unoriented polymer from
which the fibre is spun. It provides the analyst with an approximation of the relative crys-
tallinities of two fibres of the same type, and thus gives insight into an aspect of fibre
structure which can only be fully understood by X-ray and thermal methods of analysis.
To appreciate this, it is first necessary to recognize the difference between orientation and
crystallinity in a fibre. Orientation is imparted to a fibre during a specific drawing opera-
tion in which the fibre is spun between two rollers which operate at different speeds. As the
fibre is stretched, the diameter is reduced, and the molecular orientation is increased paral-
lel to the length of the fibre, resulting in an increase in birefringence. Fibre manufacturers
control this orientation to achieve properties suitable to the end use for which a particular
fibre is intended. Thus, a given genus of fibre, for example nylon, will be produced in a
variety of draw ratios (although it would not be advertised and sold this way), depending
on the use to which it is to be put.
In spite of this longitudinal orientation, the polymer may still be largely amorphous
if there is little lateral arrangement between the long randomly arranged molecules. As
this alignment increases, so does the order. It is these regions of high three-dimensional
order that are responsible for crystallinity in a polymer. Because these crystalline regions
are usually too small in fibres to be observed with the light microscope (although they can
sometimes be seen as spherulites in large denier carpet fibres and nylon monofilaments),
they are normally studied by X-ray diffraction, transmission electron microscopy with
electron diffraction and thermal methods such as differential scanning calorimetry and
density. Density determinations are routinely used in the laboratories of some fibre manu-
facturers as a quality control test for crystallinity, since the greater the three-dimensional
order in a given volume, the greater will be the mass. Thus, for two polymers of the same
type, the one in which the molecules are most efficiently packed (i.e. the one with the
higher crystallinity) will have the higher density of the two. The Gladstone and Dale rule
shows that there is a direct relationship between density and refractive index. It is stated as
n − 1 = Kρ
Table 5.2 The Effect of Crystallinity on Refractive Index

Dupont Nylon from Heat-set Antron Carpet Dupont Type 746 Nylon Fibre
n|| 1.579 1.575
n⊥ 1.527 1.523
Γ 0.052 0.052
niso 1.544 1.541
where n is the isotropic refractive index, K is the atomic refractivity of the material and ρ is
the density. Thus the isotropic refractive index can be used to compare otherwise identical
fibres which may differ only in their crystallinity. This observation can be of more than
theoretical value in fibre cases involving lightly coloured or colourless nylon carpet fibres
with non-distinctive equilateral trilobal cross-sections. In these cases neither colour nor
cross-section is of much help, and the microscopist has little to draw on beyond birefrin-
gence and melting range.
Many carpet fibres undergo a process called heat setting. In this process, the fibres,
twisted into a tuft, are heated above their glass transition temperature (which is still well
below their melting point). At this temperature they are in a ‘rubbery’ state. They are then
cooled back to room temperature and their ‘glassy’ state. The purpose of this is to ‘set’ the
twist in the carpet, much like a permanent wave is set into hair in a hair salon. While
above the glass transition temperature, the molecules of a fibre are free to undergo some
limited movement and do so by arranging themselves into more ‘comfortable’ low-energy
orientations which result in an increase in the three-dimensional order in these crystal-
line regions. The result of this increase in crystallinity can be observed by the increase in
the isotropic index of a nylon carpet fibre after heat setting as shown in Table 5.2. Thus it
may be possible to distinguish two otherwise similar, nondescript fibres on the basis of
their relative crystallinities from this readily determined property. Note in Table 5.2 that
the birefringence of the fibres remains constant because there was no further longitudinal
orientation of the molecules by stretching during the process.
5.5.5 The Standort Diagram

Heuse and Adolf (1982) have developed a most useful way of plotting and using optical
data from fibres called the Standort diagram. An example of this is shown in Figure 5.7.
The refractive indices are plotted on the ordinate (n||) and abscissa (n⊥), and the diagonal
lines indicate the birefringence. The sign of elongation is determined from the zero bire-
fringence line. Fibres above it have a positive sign of elongation and those beneath it a nega-
tive one. When the optical data of the various fibre genera are plotted on it, one can quickly
see the relationships between fibres and their optical constants, as well as the relationships
between different fibre types. Using this chart, it becomes a simple matter to select mount-
ing media to identify an unknown fibre, distinguish between two closely related fibres,
select a mountant to look at internal features or particles by causing the fibre to disappear
optically, and observe the range of any optical property within a particular group of fibres.
5.5.6 Fibre Identification

With a little practice, it is usually possible to identify the common synthetic and regener-
ated fibres on the basis of their optical properties by means of relative refractive index,
5 0 5 0
0.2 0.2 0.1 0.1
n
1.70
5
0.0
PET
PBT
Polyester
1.65
0
Δn
0.0
PCT
Polyamide PA 6.6
PA 6 PPT
1.60 Regerate
zellulosen .05
–0
PUR
1.55
PVC
.10
–0
PP Proteinbasis
1.5
AL
PUE Polyacrylnitril
1.50
.15
–0
Zellulose
0
derivate 0.5
1.45
1.5
1.45 1.50 1.55 1.60

1.5
Figure 5.7 Optical ‘Standort’ diagram (loci) of man-made fibre types. (From Heuse, O. and
Adolf, F.-P., J Forens Sci Soc, 22, 103–122, 1982. With permission.)
an estimation of birefringence and a determination of the sign of elongation. Although

a number of schemes for the identification of man-made fibres on the basis of their
optical properties have been devised and a few published (Gaudette, 1988; Grieve, 1990;
Stoeffler, 1996), the author recommends that the analyst devise his or her own scheme
of identification as he or she learns to identify fibres by means of their optical proper-
ties. Only in this way will the microscopist become familiar with the properties and
peculiarities of fibres which are so essential not only in identifying them but also in
comparing them.
This is not to say that established laboratory procedures should be disregarded. Once
the analyst has learned to identify fibres confidently, he or she can be introduced to the
procedures used in the laboratory in question. As a result of their training, however, they
should know when and why deviations from the standard procedures should be adopted,
and may even be in a position to improve upon procedures which need updating or change.
Some fibres do not lend themselves to identification by optical methods. These fibres
are usually dyed or pigmented so deeply that light cannot penetrate them. In this situation,
the interference colours cannot be reliably interpreted due to absorption, and the Becke
lines cannot be observed to determine the relative refractive indices. In these cases, a small
piece of the fibre should be cut off and prepared for infrared microspectrophotometry or,
if a microspectrophotometer is not available, microchemical or solvent tests. The most
important rule here is that before a fibre comparison can be called complete, it is essential
that the identity of the fibres being compared be unequivocally established. Table 5.3 gives
typical optical properties and melting points for man-made fibres.
Table 5.3 Typical Optical Properties and Melting Points for Man-Made Fibresa
Fibre Type n|| n⊥ Birefringence Sign Melting Range (°C)
Acetate
Diacetate 1.474–1.479 1.473–1.477 0.002–0.005 + 250–260
Triacetate 1.469–1.472 1.468–1.472 0.000–0.001 +, – 288–300
Acrylonitrile
Acrylic 1.510–1.520 1.512–1.525 0.001–0.005 – does not melt (dnm)
Modacrylic
Verel® 1.538–1.539 1.538–1.539 0.000–0.001 – dnmb
Dynel® 1.528–1.535 1.523–1.533 0.002–0.005 + 190b
Aramid
Kevlar® 2.050–2.350 1.641–1.646 0.200–0.710 + 425
Nomex® 1.800–1.900 1.664–1.680 0.120–0.230 + 371
Cellulosic
Rayon (viscose) 1.541–1.549 1.520–1.521 0.020–0.028 + dnm
HT viscose 1.544–1.551 1.505–1.516 0.035–0.039 + dnm
Cupro 1.548–1.562 1.519–1.528 0.021–0.037 + dnm
Lyocell 1.562–1.564 1.520–1.522 0.044 + dnm
Fluorocarbon 1.389 1.350 0.039 + 280–300
Nylon
Nylon 6 1.568–1.583 1.525–1.526 0.049–0.061 + 215–228
Nylon 6.6 1.577–1.582 1.515–1.526 0.056–0.063 + 254–265
Polyolefin
Polyethylenec 1.568–1.574 1.518–1.522 0.050–0.052 + 119–135
Polypropylene 1.520–1.530 1.491–1.496 0.028–0.034 + 167–179d
Polyester
PET 1.699–1.710 1.535–1.546 0.147–0.175 + 256–268
PBT 1.688 1.538–1.540 0.148–0.150 + 221–222
PCDT 1.632–1.642 1.543–1.542 0.098–0.102 + 278
PEN e.g Pentex® 1.862 1.589 0.273 + 274
Saran 1.599–1.610 1.607–1.618 0.008 – 170
Spandex e e 0.010 + 230
Sulfar 1.849 1.738 0.111 + 283
Vinal 1.540–1.547 1.510–1.522 0.025–0.030 + dnm
Vinyon (chlorofibre) 1.527–1.541 1.524–1.536 0.002–0.005 + dnm
a These are typical examples. Values for each genus are represented by a locus of points on the ‘Standort’ diagram
(see Figure 5.3).
b The sign of elongation is variable in modacrylic fibres. Dynel®, a vinyl chloride co-polymer, is the only one
where a melting point can be found in the literature. Dynel fibres are no longer manufactured.
c Example: Spectra®.
d See Hartshorne, A. W. and Laing, D. K., Forens. Sci. Int., 26, 45–52, 1984.
e These fibres absorb ordinary index of refraction oils and their values change. Refractive indices must be
measured by interference microscopy using silicone oils.
5.6 Thermal Microscopy
Observation of the behaviour of thermoplastic fibres as they are heated while being observed
under the microscope often provides valuable information for both fibre identification and
comparison. The type of hot stage used is not particularly important. Temperatures which
are read from a thermometer are as good as those taken from a digital display panel. The
most important considerations in selecting a hot stage for thermal microscopy of fibres are
as follows:
• It should be capable of being mounted on the stage of a polarizing microscope.

• A clear image of the fibre should be visible at a magnification of at least 100× by
transmitted light between crossed polars.
• The stage should be capable of reaching a temperature of at least 300°C.
• It should be possible to raise the temperature by as little as 4°C min–1 and prefer-
ably less.
• The three most common reasons for observing the thermal behavior of a fibre
include are as follows:
• Fibre identification. A determination as to whether or not a fibre melts, and the
melting point if it does, can aid in identifying fibres which are too heavily dyed
or pigmented to be recognized by polarized light microscopy.
• Sub-classification. The classic example of this is for the distinction between
nylons 6 and 6.6. Although this can also be accomplished by infrared spectros-
copy, it is still of use when the fibre contains both nylon 6 and 6.6, which can
confuse the interpretation of the infrared spectrum.
• As a comparison characteristic of thermoplastic fibres. The melting point
or range of a polymer is determined by an often complex variety of factors
such as molecular weight, degree of crystallinity and the presence of organic
additives. Some of these factors can be studied directly by thermal methods
such as differential scanning calorimetry, but these methods cannot be
applied to a single fibre fragment. Although the individual contributions
from the various elements which affect it cannot be directly related, the
melting point, when carefully and reproducibly determined, is a useful
comparison characteristic.
Melting point determinations may be carried out in air or in silicone oil. Only a small
length of fibre is necessary: a few hundred micrometres is sufficient in almost all cases.
The author commonly uses rejected cross-sections (those which are too thick and
lie on their side) for this purpose. The fibre is situated on the proper area of the slide,
covered with a coverslip and placed in the hot stage. The melting point can be measured
for a single specimen, or a short segment of a questioned fibre can be mounted next to
a slightly longer piece of the known and they can both be observed as they are heated.
This perhaps somewhat controversial suggestion – that the known and questioned sam-
ple be mounted together – is justifiable since only small pieces of the fibres and not the
entire samples are being placed together, and a scientist should be able to tell the dif-
ference between a longer piece of fibre and a shorter piece of fibre by using a consistent
convention. The fibres should be roughly parallel to one another so their interference
colours can be watched as they begin to melt.
The actual melting points of synthetic fibres are not often absolute (Grieve, 1983). In
practice melting behaviour over a range of temperature is commonly observed. Rather
than record a single temperature, the author notes several temperatures as melting
progresses. In some cases certain fibres will shrink dramatically as the temperature
begins to rise. This is an important characteristic when it occurs, although it must be
admitted here that this characteristic is far more subtle when only a short segment is
utilized. Another important temperature is that at which the first evidence of melting
occurs, followed by the temperature at which the fall in interference colours begins.
The temperature at which the fibre becomes first-order grey, and finally the tempera-
ture at which the interference colour disappears, complete the determination. The tem-
perature at which the first-order grey disappears, is often called the ‘crystalline melting
point.’ This term originates from photometric melting point determinations where
light intensity is monitored over a field of view covered by the polarizer. As the solid
melts, the interference colours fall. The isotropic melt is black under these conditions.
As the melt fills the field of view, the light intensity drops. When the film is completely
melted, the intensity drops to zero because the crystallinity has disappeared. When
one is melting single fibres or fibre segments, it is sometimes noticed that a small patch
of first-order grey material remains long after the vast bulk of the fibre has melted.
Although this should be noted, the temperature at which most of the fibre becomes
black under crossed polars should be recorded, not that at which the anomalous region
finally melts.
5.7 Observing Colour on Fibres through the Microscope
The colour of fibres under the microscope is one of their most striking and characteris-
tic features. As a result, target fibres are most often selected on the basis of their colour.
In fact, this is the basis on which automated fibre finders work. Colour comparisons of
fibres are generally completed by means of UV-visible microspectrophotometry and/or
thin-layer chromatography. However, they are started at the microscope. The purpose
of this section is to discuss some of the microscopical aspects of colour on fibres which
will help to ensure that only the best candidates are chosen for spectrophotometry and
chromatography.
First of all, the colour must be visible under the microscope. If one observes an off-
white, beige, tan, pale yellow or similar carpet laid out in a room, there is no difficulty
in stating the colour within the limits of one’s colour vocabulary. However, the situa-
tion changes dramatically when we observe only a single fibre from one of these carpets
by transmitted light, in a mounting medium on a microscope slide. In such cases, the
fibre often appears to be colourless and no colour comparison is possible, although
some colour may be discerned on the unmounted fibre under a stereomicroscope by
darkfield reflected light. In these situations, the examiner will exploit the other charac-
teristics described throughout this chapter to the fullest extent possible in making the
comparison. It is no use forcing the issue of colour in these cases. The author has been
shown spectrophotometric curves from pale yellow fibres and told that the examiner
who produced them had testified that the slightly inclined, almost flat lines on the charts
not only showed that the fibres were dyed with the same dye, but that they were charac-
teristic of the dye as well. The fibres were trilobal carpet nylon, but the analyst refused
to make a cross-section or a melting point determination to find out whether they were
nylon 6 or 6.6.
If the colour can be observed under the microscope, the microscopist can compare
the colours of known and questioned fibres directly under the comparison microscope.
In this regard it is important that the examiner compare the colours critically. Although
always important, this is particularly so when comparing natural fibres. By this, we mean
that the colours be actually looked at and studied, and not just noted. In a recent case,
we were asked to perform some follow-up work for a state forensic laboratory involving
a number of fibres from a knife blade which had been compared to the clothing of the
suspect. We were astonished to find that blue cotton fibres which the laboratory claimed
matched fibres from the suspect’s sweater could be distinguished from those comprising
the sweater by visual comparison. The ‘blue’ fibres from one source were greyish and from
the other purplish. It was not necessary to use spectrophotometry to see the difference,
in spite of some initial similarity.
Careful observation will show the source of colour in a fibre. Thus it is possible,
as stated earlier in this chapter, to determine whether a fibre is coloured with a dye
or pigment by simply looking at it under high magnification, preferably after match-
ing the refractive index of one of the vibration directions. In a similar manner, colours
applied to the surface of a fibre by printing can be seen as well. If these observations are
neglected before performing a spectrophotometric analysis, valuable information may
be missed entirely.
5.7.1 Dichroism
Pleochroism is the change in absorption colour with vibration direction; when it occurs in
fibres it is an important comparison characteristic. Pleochroism is a general term for this
phenomenon. In most fibres, which exhibit only two vibration directions in longitudinal
view, the correct term for this effect is dichroism. Not all dyes produce dichroism on fibres.
The first essential characteristic of a candidate dye is that its molecule be elongated; specifi-
cally that its chromophoric groups be in linear alignment. Molecules of the dye Congo Red
(C.I. 22120) have such a shape, as shown in Figure 5.8.
Crystals of this dye are pleochroic when illuminated with plane polarized light, due
to the fact that absorption is greater for light vibrating along the length of the dye mol-
ecule than it is for light vibrating perpendicular to it, as shown in Figure 5.9. A viscose
fibre which has been dyed with Congo Red and aligned with its length parallel to the
NH2 H 2N
N=N N=N
SO3Na NaO3S
Figure 5.8 Molecular structure of Congo Red.

Figure 5.9 Pleochroism of a Congo Red dye crystal. Upper, light vibrating across the dye mol-
ecules; lower, light vibrating along the dye molecules; magnification 593x.
polarizer appears bright red. When the fibre is rotated so that n⊥ is aligned with the polar-
izer, it appears practically colourless (Figure 5.10). This effect is the result of an interaction
between the molecules of the fibre and the dye, and is the reason why dichroism is such a
valuable characteristic when it occurs.
The second necessary characteristic is that the bound dye molecules are aligned
with the polymer molecules. Modern theory holds that dyeing takes place in the amor-
phous regions of a polymer, and not in the crystalline ones. Recall, however, that the
polymer molecules in the amorphous regions of most man-made fibres are highly ori-
ented along the fibre axis. Linear dyes, such as Congo Red, bind with these axially
oriented fibre molecules and are thus arranged in very much the same orientation as
they are in the crystalline dye itself. This results in a dichroic formula (i.e. the colour
associated with each vibration direction) for the fibres which is practically identical to
that observed on the dye crystals themselves. Thus, for two fibres to exhibit both the
same colour and same dichroic formula, it is necessary that the molecules of at least
one of the dyes in both fibres be linear, that they have similar structures (to produce
the same colours), and that they are both bonded to polymer molecules with approxi-
mately the same degree of linear orientation. This is a lot to expect by coincidence in
two fibres which are also similar in all other characteristics. Microspectrophotometry
can be used to record the dichroic absorption spectra in the visible region for docu-
mentation, although photomicrography makes a strong visual impression for a jury. A
dramatic instance of dichroism on a viscose fibre is shown in Figure 5.11, where n|| is
green and n ⊥ is pink.
Figure 5.10 Dichroism in a viscose fibre dyed with Congo Red. Upper, n ⊥ aligned with the
polarizer, the fibre appears practically colourless; lower, n∥ aligned with the polarizer, the fibre
appears red; magnification 593x.
Figure 5.11 Dichroism in a viscose fibre. Upper, n ⊥ aligned with the polarizer, the fibre appears
pink; lower, n|| aligned with the polarizer, the fibre appears dark green; magnification 593x.
It is frequently noted that solution dyed fibres such as polypropylene exhibit dichro-
ism. Figure 5.12 shows an example of this. In light of the discussion above, the reason for
this should now be obvious. Pigment crystals with an acicular or even an oval habit are
oriented by the shear forces which develop in the viscous polymer matrix during drawing.
The result is that most of these pleochroic (they are usually biaxial) crystals are oriented
with their long axes more or less parallel to the fibre axis, giving rise to dichroism which
has the same formula as the pigment crystals themselves.
5.7.2 Fluorescence
This discussion of colour has been concerned so far with the effects of visible light. Many
substances exhibit a surprising effect when illuminated with light of short wavelengths
(i.e. ultraviolet light). They emit light of a longer wavelength than that with which they
were excited. If this emission stops when the exciting radiation is cut off, the phenom-
enon is known as fluorescence. Modern fluorescence microscopes are constructed so that
the specimen is illuminated from above, which is more efficient than older microscopes
which operated by transmission. The illuminator is a high-pressure mercury burner in an
adjustable vertical mount. This high-intensity light is directed through a series of excita-
tion filters which absorb unwanted wavelengths, and a heat-absorbing filter. This excitation
energy is then directed at a right angle down through the objective onto the specimen by
means of a dichromatic mirror matched to the excitation filters which are in use. These
dichromatic mirrors are constructed so that they reflect light of certain wavelengths and
transmit higher wavelengths. A selection of excitation filters and matching dichromatic
mirrors should cover the range of ultraviolet through green. Figure 5.13 shows a Leica
DMR microscope equipped with a reflected light fluorescence accessory. If the specimen is
Figure 5.12 Dichroism in solution dyed polypropylene fibre. The pigment crystals are orien-
tated with their long axis parallel to the fibre; magnification 960x.
Figure 5.13 Leica DMR microscope illustrating the reflected light fluorescence accessory.
Different filter block combinations can be selected by rotating the large knurled disc in the
centre.
fluorescent, the longer wavelengths of the fluorescent light emitted from it will pass back
through the objective and dichromatic mirror and form an image. A final barrier filter is
included to block harmful ultraviolet radiation from reaching the eyes of the microscopist
(Herman, 1998; Holz, 1979).
Both longitudinal mounts and cross-sections of fibres may be examined with advan-
tage. The primary requirement is that they be mounted in a non-fluorescing mounting
medium. When examining fibres by fluorescence microscopy, the analyst will normally
be searching for autofluorescence which is due to substances in the fibre that fluoresce
when illuminated with a certain range of excitation energy. In organic fibres, this will
usually originate from optical brighteners, conventional dyes and pigments or contami-
nants. Optical brighteners are dyes, usually stilbene derivatives, which exhibit a blue-
white fluorescence with ultraviolet excitation and make fibres ‘whiter than white’. They
may be added to the fibre during spinning, as in the case of polyester which is intended
to be mixed with cotton for the manufacture of fabric for shirts and blouses. The natu-
ral fluorescence of cotton fibres with ultraviolet excitation is a faint dull green. Optical
brighteners may be applied to cotton during the manufacture of the fabric or it may be
acquired during laundering, since almost all commercial detergents contain these com-
pounds. It may be useful here to point out that cross-linked and resin-treated cotton,
especially when new, may fail to absorb optical brighteners, so that it is not impossible
to encounter cotton fibres which have been washed in detergent and still show no fluo-
rescence with ultraviolet excitation. This became an important issue in a murder case in
Seattle several years ago when the defendant’s girlfriend testified that she had washed
his shirt in a commercial detergent, yet the questioned cotton fibres (from a tuft of green
cotton and polyester) showed no evidence of fluorescence. Microchemical testing proved

that the cotton fibres were cross-linked.
Dyes and pigments frequently fluoresce and since the colour in a fibre is usually
achieved with more than one colourant, we can expect that some of these components
will fluoresce. In contrast to optical brighteners, which always fluoresce blue-white with
ultraviolet excitation, normal dyes fluoresce in a variety of colours and with different
excitation wavelengths. When dye fluorescence is observed, the full range of excitation
wavelengths should be employed and the emission colours noted (or if this is being done
on a microspectrophotometer the emission spectra should be recorded) and compared.
A questioned fibre should behave in the same way as fibres from the known source. The
same trilobal fibre is shown under ultraviolet excitation (Figure 5.14) and blue light excita-
tion (Figure 5.15). These photomicrographs suggest the wide range of variation which one
may encounter from fluorescence examinations.
Occasionally, irregular fluorescence that appears to originate from outside the fibre
will be observed. This will usually be some form of contamination which, if it occurs on
both the questioned and known fibres, may be a strong additional point of comparison.
In these cases, an examination of the longitudinal view of the fibre may not permit a
definitive answer to the question of whether the fluorescent substance is on the sur-
face. An examination of the cross-section will usually answer this question, although
the section must be thin, since thick sections, such as those produced with the Joliff
method, suffer from interference due to light scattering from the packing fibres. It may
be possible to extract the fluorescing material with solvents and identify it by infrared
microspectrophotometry. The fluorescence can be used to trace the small amount of
substance, since it will show if the solvent has extracted it from the fibre and then if it
is in the solvent.
Figure 5.14 Fibre fluorescence. A trilobal nylon fibre under ultraviolet excitation; magnifica-
tion 400x.
Figure 5.15 The same fibre as in Figure 7.14 under blue light excitation; magnification 400x.
5.7.3 Microchemical Testing

If the laboratory does not have a microspectrophotometer, it is still possible to conduct
further testing of colourants on fibres by means of comparative microchemical tests.
Short segments of the known and questioned fibres, such as those used for thermal
microscopy, are placed adjacent to each other on a slide and examined at moderate
magnification with the compound microscope. A droplet of reagent is placed at the
edge of the coverslip and the fibres are observed as it flows around them and reacts
with the dyes. Four or five such slides can be set up and observed as the various
reagents react. A useful series might include hydrochloric acid, alcoholic potassium
hydroxide, sodium hypochlorite (bleach), sodium dithionite (reducing agent) and con-
centrated sulphuric acid. The reagents should be chosen to produce a variety of chemi-
cal reactions on the dyestuffs. The solutions should all be strong and it may be necessary
to crush the fibres in some cases by pressing the coverslip with a steel needle to make
certain that they can reach the dyestuff or pigments. If differences are noted in the
colour reactions which occur, the two fibres could not have been coloured with the same
dyestuffs.
5.8 Conclusion
The greater part of any forensic fibre examination is performed under microscopical
observation. It is therefore critical that the fibre examiner fully understand the micro-
scopes that are being used, not only in terms of the optical theory which governs such
parameters as image formation, resolution and the determination of optical properties,
but also in terms of their practical operation and performance. The manner in which
a sample is presented to a particular microscope may have a profound bearing on the
information which can be obtained from it. The knowledge and ability to prepare sam-
ples optimally and reduce artefacts is yet another goal of the forensic fibre microscopist.
However, the best theoretical background and skill in sample preparation are virtually
meaningless if the analyst does not possess a sound knowledge of the characteristics of
single fibres, their similarities and how they can be differentiated from and compared
to one another. This can only be achieved by studying a wide variety of known fibres,
beginning with samples of known provenance and treatment and extending the study to
fibres from every source obtainable as the microscopist’s level of skill and understanding
improve. The best scientists will approach their cases with the viewpoint of the analyst,
in an attempt to extract the most important information from each specimen with the
least expenditure of sample, as opposed to the technician who merely follows some-
one else’s procedures and is thus reduced to the role of mere comparator. The results
of the former will bear the hallmark of real science and will withstand the scrutiny to
which so many cases today are subject.
Acknowledgements
The author wishes to acknowledge the assistance of Christopher Palenik, who reviewed the
manuscript and made numerous helpful suggestions and comments. Don Felty and Donna
Knoop, formerly microscopists at Monsanto and Allied Signal Fibers respectively, freely
shared information over the years, which helped to fill in gaps which invariably appear as
an outsider attempts to make sense of such a complex field. Although after 25 years they are
now too numerous to mention and in some cases even to recall, credit is also extended to
colleagues in forensic science laboratories around the world, former students, clients from
fibre manufacturing companies and both district attorneys and defence lawyers who have
submitted the wide variety of cases and projects involving fibres of all types which have
made my life interesting and given me the opportunity to learn something about these
fascinating microscopic filaments.
References
Anon, 1996. Calgene gets patent on cotton engineered with color pigments, Chem. & Eng. News,
July, 29.
Backer, S., 1972. Yarn, Sci. American, 227, 47–56.
Bloss, D., 1961. An Introduction to the Methods of Optical Crystallography. New York, NY: Holt,
Rinehart and Winston.
Catling, D. L. and Grayson, J., 1982. Identification of Vegetable Fibres. London, UK: Chapman & Hall.
Culliford, B. J., 1963. The multiple entry card index for the identification of synthetic fibres, J. Forens.
Sci. Soc., 4, 91–97.
Deadman, H., 1984a. Fiber evidence and the Wayne Williams Trial (Part 1), FBI Law Enf. Bull.,
March, 13–20.
Deadman, H., 1984b. Fiber evidence and the Wayne Williams Trial (conclusion), FBI Law Enf. Bull.,
May, 53 (5), 10–19.
Gaudette, B. D., 1988. The forensic aspects of textile fibre examination. In Saferstein, R. (Ed.), Forensic
Science Handbook (Vol. 2). Englewood Cliffs, NJ: Prentice Hall.
Gorski, A. and McCrone, W. C., 1998. Birefringence of fibers, Microscope, 46, 3–16.
Grieve, M. C., 1983. The use of melting point and refractive index determination to distinguish
between colourless polyester fibres, Forens. Sci. Int., 22, 31–48.
Grieve, M. C., 1990. Fibres and their examination in forensic science. In Maehly, A. and Williams, R. L.
(Eds.), Forensic Science Progress (Vol. 4). Berlin, Germany: Springer-Verlag.
Grieve, M. C. and Griffin, R. M. E., 1999. Is it a mod-acrylic fibre? Sci. & Just., 39, 151–162.
Hardy, J., 1935. A practical laboratory method for the making of thin cross sections of fibres, Circular
No. 378, Washington, DC: United States Department of Agriculture.
Hartshorne, A. W. and Laing, D. K., 1984. The identification of polyolefin fibres by infrared spectros-
copy and melting point determination, Forens. Sci. Int., 26, 45–52.
Hartshorne, N. H. and Stuart, A., 1970. Crystals and the Polarizing Microscope (4th ed., pp. 556–588).
Bath, UK: Pitman Press.
Herman, B., 1998. Fluorescence Microscopy, Oxford, UK: Bios Scientific.

Heuse, O. and Adolf, F.-P., 1982. Non-destructive identification of textile fibres by interference
microscopy, J. Forens. Sci. Soc., 22, 103–122.
Heyn, A. N. J., 1954. Fiber Microscopy (pp. 297–302). New York, NY: Interscience Publishers.
Holz, H., 1979. Worthwhile Facts about Fluorescence Microscopy, Oberkochen, Germany: Carl Zeiss.
Luniak, B., 1953. The Identification of Textile Fibres, London, UK: Sir Isaac Pitman & Sons.
McCrone, W., McCrone, L. and Delly, J. G., 1978. Polarized Light Microscopy, Ann Arbor, MI:
Ann Arbor Science Publishers.
Palenik, S. and Fitzsimons, C., 1990a. Fiber cross sections: Part 1, The Microscope, 38, 187–195.
Palenik, S. and Fitzsimons, C., 1990b. Fiber cross sections: Part 2, A simple method for sectioning
single fibers, Microscope, 38, 313–320.
Seminski, M. A., 1975. A note on the measurement of birefringence, Microscope, 23, 35–36.
Stoeffler, S., 1996. A flowchart system for the identification of common synthetic fibres by polarised
light microscopy, J. Forens. Sci., 41, 297–299.
The Textile Institute, 1975. Identification of Textile Materials (7th ed., p. 262). Manchester, UK:
The Textile Institute.
Valaskovic, G. A., 1991. Polarized light in multiple birefringent domains: A study of the Herzog
effect, Microscope, 39 (3–4), 269.
Wortmann, F.-J. and Wortmann G., 1991. Scanning Electron Microscopy as a Tool for the Analysis of
Wool/Speciality Fibre Blends, Aachen, Germany: Deutsches Wollforschungsinstitut.
Colour Analysis of Fibres
THOMAS W BIERMANN
KENNETH G WIGGINS
6
Contents
6.1 Microspectrophotometry 180
6.1.1 Colour 180
6.1.2 History of Microspectrophotometry 182
6.1.3 Theory of Microspectrophotometry 184
6.1.4 Instrumentation, Validation and Calibration 189
6.1.4.1 Microscope 190
6.1.4.2 Spectrophotometer Devices 191
6.1.4.3 Data Control, Processing, Recording 196
6.1.4.4 Validation 196
6.1.4.5 Instrument Calibration 197
6.1.5 Spectral Measurement 199
6.1.5.1 Absorption Measurement 200
6.1.5.2 Fluorescence Emission Measurement 203
6.1.5.3 Sampling and Sample Preparation 205
6.1.6 Spectral Comparison 206
6.1.6.1 Quantitative and Qualitative Limitations 206
6.1.6.2 Discriminating Power 207
6.1.6.3 Spectral Comparison 208
6.1.7 Numerical Colour Coding 213
6.1.7.1 Visual Comparison Using Standards 214
6.1.7.2 Instrumental Measurements/the CIE System 216
6.1.7.3 Complementary Chromaticity Coordinates (CCC) 219
6.1.8 Conclusions 220
References 221
6.2 Fibre Dyes Analysis 224
6.2.2 Classification of Fibre Dyes 224
6.2.2.1 Colour Index Number: CI 14780 224
6.2.2.2 Generic Name: CI Direct Red 45 225
6.2.2.3 Commercial Name: Thiazine 225
6.2.3 Chemical Classification 225
6.2.3.1 Acid Dyes 225
6.2.3.2 Azoic Dyes 225
6.2.3.3 Basic Dyes 226
6.2.3.4 Direct Dyes 226
6.2.3.5 Disperse Dyes 226
6.2.3.6 Metallized Dyes 226
6.2.3.7 Reactive Dyes 227
179
6.2.3.8 Sulphur Dyes 227

6.2.3.9 Vat Dyes 227
6.2.3.10 Pigments 227
6.2.3.11 Ingrain 228
6.2.4 Fibre/Dye Combinations 228
6.2.5 Dye Classification and Extraction 229
6.2.6 Dye Analysis by Thin Layer Chromatography (TLC) 234
6.2.6.1 Thin Layer Chromatography of Non-reactive Fibre Dyes 234
6.2.6.2 Thin Layer Chromatography of Reactive Fibre Dyes 239
6.2.6.3 Non-Extractable Dyes 241
References 242
6.1 Microspectrophotometry
THOMAS W BIERMANN
6.1.1 Colour
The individual experience called colour does not exist in reality. Colour is ‘only’ an interaction
of light with a substance, perceived by a human being. The human eye is sensitive to radiation
of the visible (vis) region from the blue to the red part of the electromagnetic radiation spec-
trum with a maximum sensitivity for green colour. The physical stimulus of radiation causes
reactions in the eye brain system which we associate with the sensations of light, brightness
and colour. The eye-brain system is reported to be extremely sensitive for the detection of
colour and particularly for the detection of small colour differences. It is estimated that more
than six million colour shades can be distinguished (Welsch and Liebmann, 2003).
If the eye receives the unattenuated radiation of the vis region, for example daylight,
the psychophysiological result is recorded as white, in contrast to a total absorption of
radiation, which is recognized as black. If the same portions of radiation are absorbed at
each wavelength of the light, then we get the impression of a grey shade. There are many
different grey shades, known as the achromatic range with white and black at the opposite
ends. In contrast, the sensation of colour is only stimulated when the amount of the radia-
tion absorbed varies at the different wavelengths across the vis region. Colours are known
as the chromatic range. The spectral colours – the bands of radiation which make up natu-
ral light—are the purest and most intense ones. They start with violet, occurring at around
400 nm, followed by blue, green, yellow, orange, and red, ending at around 700 nm.
It has already been mentioned that colour recognition by human beings is not an
objective event. There are different reasons. One is the influence of the environment. This
can easily be demonstrated by an effect according to Friedrich Wilhelm von Bezold, a well
known German physicist and metereologist (1837–1907). This effect demonstrates that the
actual impression of a colour shade is influenced by the brightness and colour of the fore-
ground. In Figure 6.1 the impression of the grey background in the left and right half of the
picture is different because of the white and black foreground. As a consequence, not only
must visual colour comparison be carried out under a defined illumination, but also each
Colour Analysis of Fibres 181
Figure 6.1 The Bezold effect.
environment must be illuminated under identical physical conditions. The colour match-
ing boxes used in many parts of the textile industry with their neutral grey insides are a
common example of the realization of these demands.
Colour impression produced by the eye-brain system also depends on psychophysi-
ological factors such as eye adaption and fatigue. Thus, the colour impression perceived
from the same stimulus varies, not only from one observer to another, but for the same
observer at different times. It also is known that the two eyes of the same individual
may differ in colour sensitivity. It is equally important to know that even in the case of a
convincing visual match between the colour of two samples, one cannot be sure if both
samples are actually dyed with identical dyes. It is daily practice in textile dye houses,
where textile materials are usually dyed in batches of varying size that the contents of
the dye baths used to produce the same colour shade are not constant but are varied
by using different dyes. This process is known as ‘topping-up’ (Wiggins and Holness,
2005). Objects such as fabrics or fibres which show the same hue, but are dyed with dif-
ferent dyes or mixtures of dyes, are called metameric. Metamerism is difficult to detect
by the human eye. Illumination with different light sources (e.g. D65, C, A or UV) is
necessary.
All these inadequacies of the human colour vision system lead to certain restrictions
in use which a forensic fibre examiner must bear in mind. For example, use of the com-
parison microscope for visual colour comparison is limited by the fact that the light beams
cannot be made identical in the physical sense. Therefore, the samples are not really being
observed under the same optical conditions. A second relevant point is that of metameric
colours, which the eye-brain system cannot detect without instrumental aid. Finally, it
must be taken into account that the colour of a single fibre, as seen microscopically, is
determined by other properties such as dye uptake, thickness, shape, delustrant content.
These characteristics can be extremely variable, particularly in natural fibres. As a result,
a variety of colour shades can be found in single fibres which all originate from the same
textile. Again, an observer needs instrumental aid to decide whether those fibres are really
identically dyed, i.e. dyed with the same colourants.
From these limitations, an inexperienced fibre examiner may get the impression that
his eyes are virtually useless for colour assessment of fibre evidence. This opinion must of
course be corrected. The sensitivity of the eye, and its speed of colour detection, are unsur-
passed. These characteristics are both required in the searching and screening procedures
which dominate the beginning of each fibre case. However, the visual findings must be
controlled and verified by an objective method in one of the subsequent steps of the chain
of examination. Besides chromatographical methods such as thin-layer chromatography
(TLC) or ultra-high performance liquid chromatography (UHPLC) the most widely spread
method is microspectrophotometry (MSP).
6.1.2 History of Microspectrophotometry

Microspectrophotometry is not a new approach in forensic science. Indeed, the method
has already been used for about 55 years in forensic applications, and for nearly 90 years
in biological and non-biological disciplines. In 1927, Weigel and Habich, and Weigel and
Ufer, described the identification of minerals from their absorption spectra. The colour in
the microscopic image of the specimen was measured using a polarized light microscope
equipped with a photocell. Casperson (1936, 1940) developed MSP for use in biological
science in order to locate and identify chemical constituents within cells. MSP was fre-
quently used for biochemical studies during this early period of development. Since then,
MSP has been used to study almost all materials (Piller, 1979) such as organic and inor-
ganic crystals, amorphous solid particles, biological tissues, textile fibres, glass fragments,
paper and patterns on paper, plastic particles, etc. The application of MSP to the forensic
examination of fibres was first described by Amsler in 1959. Important contributions to the
theoretical background and principles were published by Suchenwirth and Brück (1968),
Halonbrenner and Meier (1973), Halonbrenner (1976), Brück and Röhm (1975), Macrae
et al. (1979), and Pabst (1980).
The only standard textbook about MSP, Microscope Photometry, was published by Piller
(1977). In the early sixties Piller constructed the UMSP 1 – a double-beam photometer – and
in 1977 he started the development of a new generation of instruments in cooperation with
the Bundeskriminalamt in Wiesbaden. The MPM 03TM was a single-beam version using
modern grating monochromators which increased the value of resolution (Figure 6.2).
A second impetus was provided by the availability of stable and sensitive photoelectric
devices and stable light sources. Real progress was made when it became possible to couple
computers to the instruments. The type MPM 800TM which was manufactured in the early
nineties was the third and together the last version of microspectrophotometers produced
by Zeiss.
Since the seventies, other microspectrophotometers have been on the market: the Leitz
MPVTM, the Nanospec IOSTM produced by Nanometrics and the Rofin Micro-ColoriteTM
system in Australia. Due to their attachment to normal transmitted light microscopes,
these systems could only measure the visible region of the electromagnetic spectrum. The
descendants of the Leitz and Nanometrics systems were on the market for a longer period
of time, with products named MPV SPTM from Leica and SEE 1000TM and SEE 2000TM from
S.E.E. Inc. respectively. Both the hardware and software had of course been modernized.
Because scientific and technical development never stops, forensic microspectropho-
tometry has again undergone modernization. In the middle of the nineties photodiode-
array detectors (PDA) were used as spectral measuring devices (Adolf, 1997). With such
Figure 6.2 The Zeiss MPM 03TM.
photodiodearray detectors or even with charged coupled device detectors (CCD) it was
possible to record visible spectra from dyed fibres of the same quality as those measured
using a grating monochromator as a scanning device, but in a very short measuring time of
less than a second. Additionally, a PDA or CCD detector can be easily and quickly mounted
at the photo port of a microscope. This new development enabled forensic fibre examiners
to implement spectral measurements of fibre dyes as a screening tool into the very initial
phase of fibre identification/comparison.
Today PDA and CCD detectors cover both the ultraviolet and the visible range of the
electromagnetic radiation. There are several systems on the market. For example Craic Tech-
nologies (USA) offers the Microspectra 121TM UV-Visible-NIR Micro-spectrophotometer,
the Microspectra 10TM Microscope Spectrophotometer or the Microspectra 1TM Microscope
Photometer. Depending upon the microscope optics and sources, the spectral range is
from the deep UV to the near infrared region.
Another company which offers microspectrophotometers which are suitable for foren-
sic use is J&M Analytik AG in Germany. There are several systems available. The so called
TIDAS MSP 800TM which is equipped with a PDA covers the range from 240–900 nm of
the electromagnetic radiation. Another system like the TIDAS MSP 400TM is designed for
measurements in the visible range of the electromagnetic radiation and organic petro-
graphic analysis respectively. On request the company configures microspectrophotom-
eters with CCD or even with photomultipliers as detectors.
Angstrom Sun Technologies Inc. (USA) produces a system called MSP500
Microspectrophotometer and Film Thickness Measurement SystemTM with a CCD detec-
tor for measurement from 250 to 1700 nm. Another German company offering micro-
spectrophotometers is A.S.&Co. GmbH. The forensic scientist has the choise between
systems equipped with PDA or CCD detectors or systems running grating monochroma-
tors together with photomultiplier detectors covering the full UV-vis range of the electro-
magnetic radiation.
Today in laboratories of forensic fibres sections all over the world, different systems
are in use ranging from old scanning MSP systems like Zeiss MPM 03TM, UMSP 80TM and
MPM 400/800TM or Leica MPVTM to multichannel PDA or CCD MSP systems like SEE
2000/2100TM, Craic QDI 1000TM and J&M Tidas MSP/MPM 800TM (Wiggins et al., 2008).
There is also a great variety of software packages in use, namely Lambda Scan, Spectralys,
GRAMS, Panorama and several others. Because of this variety the European Fibres Group
(EFG now called European Textile and Hair Group, ETHG) and the Scientific Working
Group for Materials (SWGMAT) Fibres section in the USA initiated a research project
that should answer the question if these different MSP systems around the world produce
comparable dye spectra. In total 42 laboratories from 21 different countries took part in
the project. They all are members or associated with the EFG, SWGMAT or Crim-SAG
(Criminalistics Scientific Advisory Group of the Senior Managers of Australien and New
Zealand Forensic Science Laboratories). The results of this study show that the tested MSP
systems produce reliable and comparable dye spectra.
6.1.3 Theory of Microspectrophotometry

Microspectrophotometry belongs to the wide range of spectroscopic methods. These are
physical methods which play an important role in experimental inorganic chemistry
(Williams and Fleming, 1966). In principle, electromagnetic radiation of different wave-
lengths is used to examine different aspects of the molecular structure. Depending on the
range of the wavelength used and the kind of information gained, the methods are called
X-ray diffraction, X-ray structural analysis, X-ray fluorescence, UV-vis spectroscopy,
IR-spectroscopy, electron spin resonance spectroscopy (ESR), and nuclear magnetic reso-
nance spectroscopy (NMR). MSP is a special method within UV-vis spectroscopy. It allows
measurement of the absorption of electromagnetic radiation by microscopical amounts of
many kinds of materials in the visible region (vis) and—depending on the instrumental
equipment—the ultraviolet region (UV) of the electromagnetic radiation spectrum.
Figure 6.3 illustrates the location of the UV-vis region in the complete electromagnetic
radiation spectrum, the wavelengths and frequencies of the different radiations and which
interactions can be observed between radiation and matter. It shows that the UV-vis region
is the region where excitation of electrons occurs. Because dyes are substances with conju-
gated systems of excitable electrons, UV-vis spectroscopy in general but MSP in particular
are suitable methods for dye examination and control of dye production. In forensic sci-
ence, MSP has therefore been established as a quick and nondestructive standard method
for the examination of trace evidence which consists of coloured microscopic particles
such as fibres, paints and inks. The UV-vis region continues into the infrared region (IR)
which is also important for forensic examination of fibres, but where excitation results in
molecular vibration and rotation of the functional groups. These phenomena provide a
different form of spectral information about the object being measured which can be used
to determine its chemical composition.
Light is characterized by attributes of waves and of particles as well. On the one hand
we can measure single photons on the other hand light is able to form interferences and is
scattered at gratings. According to Welsch and Liebmann (2003) this particle-wave dual-
ism resembles the question if an object can possess corners even though it is round simul-
taneously. Transferred into a two-dimensional flat world one would recognize only a circle
from a cylinder crossing the flat world perpendicular to its length axis. If the cylinder
crosses this world parallel to its length axis a rectangle would be recognized. Only the view
from a higher dimension solves this contradiction. The contradictory aspects of such phe-
nomena are called complementary principle.
Y-radiation
Ionisation
< 0.1 Å
X-rays
0.1–10 Å
Vacumm UV
Electron excitation
100–200 nm
Optical radiation
Near UV
200–400 nm
VIS
400–800 nm
IR
oscillation/ rotation
0.8–50 μm
Molecule
Far IR
50–500 μm
Middle waves
500 μm–30 cm
Radio waves
0.1–100 km
Figure 6.3 The electromagnetic radiation spectrum.
As mentioned above, the radiation in the UV-vis region is absorbed by the molecules
by the excitation of the electrons. The spectrum therefore provides information about the
presence of conjugated electron systems as one part of the structural elements of the mol-
ecule. Furthermore, the visible part of the spectrum provides fundamentally important
information about the colour of the object being measured.
Spectroscopy in the UV-vis region involves observation of the excitation of electrons.
Such excitable electrons are the valency electrons which occur in atomic orbitals as well
as in molecular orbitals. This subject has a complex theoretical background and needs
detailed knowledge about the structure of atoms and molecules which would be beyond
the scope of this book. For those who wish to go deeper into the theoretical principles,
Gauglitz (1994) may offer a useful starting point.
Most of the phenomena involved in the interaction between electromagnetic radiation
and matter can be understood by looking at the wavelike character of radiation. This espe-
cially concerns the UV-vis wavelength region. An electron becomes excited if the frequency
of the incident electromagnetic radiation matches or closely corresponds to the difference
in energy between two electronic states. This leads to a resonance excitation, a change in
the electron density distribution and, finally, to an electronic transition from the high-
est occupied molecular orbital to the lowest unoccupied molecular orbital. This process is
named absorption (Figure 6.4). The energy difference depends on the electronic configura-
tion of the molecule and is also influenced by its environment. Because the excited state
(S1, S2) of a molecule is not the normal state it is followed by a deactivation process. Three
types of deactivation are known: non-radiative, which can be a thermal equilibration, an
internal conversion or an intersystem crossing; radiation transfer, which is combined with
spontaneous emissions known as fluorescence and phosphorescence; and photochemical
Excited singlet states

S2
Internal
conversion
S1
Intersystem
crossing
Excited
triplet
Non-radiative relaxation
state
or quenching
Fluorescence
Absorption
Phosphorescence
S0
Ground state
Figure 6.4 Energy diagram according to Jablonski.
reactions, which often result in ionization, cleavage, etc. We all know that it is more com-
fortable to wear white shirts instead of deeply dyed or black shirts in the summer sunlight.
What is the reason for? The white garment completely reflects the light. In contrast, the
black dye completely absorbs the light and produces heat when the electrons fall down
again to the lower energy level – the ground state.
The absorption of radiation is always calculated based on the measurement of only
two values. These are the total radiation-flux in the system without the object (I0) and the
radiation-flux with the object (I). Any calculations of spectral parameters are based on
these two values. In forensic examination of fibres, there are two spectral parameters in
use - the transmission (T) and the absorbance (A) - calculated as
I
T= ×100(%)
I0
I0
A = log
I
The physical result of absorption measurements is a spectrum. Spectroscopy in the

UV-vis region does not result in line spectra. This is caused by the fact that in addition
to the excitation of the electrons, vibrational and rotational transitions can also occur. In
consequence, the absorption bands of spectra in the UV-vis region are generally broad.
Compared with those from the IR region, they provide less information about the molecu-
lar structure and functional groups present. Therefore, the main use of UV-vis spectros-
copy is not to identify substances, although it can often be used as a quick and inexpensive
method for identifying classes of materials, especially in the pharmaceutical industry. In

spite of certain limits, UV-vis spectroscopy is also used for quantitative analysis not only to
determine concentrations but also in classical multicomponent analysis, chromatography,
fluorimetry and reflectometry. For forensic fibre examination it is important to be aware
that the absorption spectra recorded give information about a certain part of the chemical
constitution of the fibre as a whole. The visible part of the spectrum is also the basis for
colorimetry. Forensic fibre examination uses UV-vis spectra or vis spectra respectively for
qualitative comparison purposes, not for the identification of a substance.
In the UV-vis region the absorption of electromagnetic radiation is caused by the exci-
tation of valency electrons. There are three types of valency electron: the σ-electrons of
the molecular frame, the π-electrons of the double bonds and triple bonds, and the pairs
of non-binding electrons (known as ‘ion pair’ or n-electrons) found, for example, in oxy-
gen, sulphur and nitrogen atoms. In order to excite σ-electrons, an energy of more than
150 kcal mol−1 is necessary. Because MSP is concerned only with radiation of wavelengths
above 240 nm, excitation of the σ-electrons does not occur in this method. In contrast, the
absorption of electromagnetic radiation in the UV-vis region in question is caused by the
excitation of the π-electrons and n-electrons which require less energy for excitation. If
double bonds and triple bonds are conjugated (e.g. in alternating positions), the π-electrons
become more and more easily excitable as the number of conjugated bonds increases. Such
extended conjugated π-systems also absorb radiation at longer wavelengths (vis) and there-
fore the substance appears coloured.
Basically, the chemical theory of colour says that absorption of radiation in the visible
region requires some groups of conjugated double bonds to be present. The absorption by
such a system is particularly supported if the π-electrons are easily displaceable, which is
the case if some resonating structures are present in the compound in its basic state. If in
addition electron donors such as hydroxyl and amino groups are present in the compound,
the capacity for mesomerism will be greatly enhanced. From these basic facts it becomes
obvious that structural factors determine whether or not a molecule will absorb in the
visible range, and also decide where such an absorption will occur.
These explanations on the chemical theory of colour can be illustrated by two
practical examples. Figures 6.5 and 6.6 illustrate the absorption spectra of benzene (in 95%
methanol) and synthetic indigo (dyed on cotton) as examples. Benzene is a simple aromatic
ring system consisting only of single bonds and a few double bonds. Benzene in methanol
is a clear and colourless solution. Four sharp absorption bands between about 240 nm and
270 nm are characteristic for this solution. Above 280 nm, there is no more absorption.
In contrast, synthetic indigo (C.I. 73000, vat blue 1), the dyestuff mostly used for dyeing
Jeanswear, is a larger molecule, linked not only by number of single bonds but also by dif-
ferent double bonds. This molecule absorbs radiation in the whole of the UV-vis region,
showing an absorption maximum near 660 nm in the yellow-orange-red region. The sub-
stance therefore appears coloured and shows the complementary colour shade blue.
The part of a coloured molecule which is responsible for the absorption of radiation
in the UV-vis region is called a chromophore. Chromophores are simple unsaturated
groups attached to benzene, or fused benzene rings (Waring and Hallas, 1990). There are
two groups of chromophores (Figure 6.7). One group of chromophores contains π-bonds
beside σ-bonds, but where only the π-electrons are excited. The second group also contains
non-binding n-electrons. The azo group, cyan group, aldehyde group, keto group, and car-
boxylic acid belong to this group.
2.0 255, 5
248, 5
1.8
1.6
243 262, 5
1.4
1.2 Benzene
ABS
1.0
0.8
0.6
0.4
0.2
0.0
240.0 245.0 250.0 255.0 260.0 265.0 270.0 275.0 280.0
nm
Figure 6.5 Absorption spectrum of benzene in 95% methanol.
100.0
90.0
80.0
70.0
60.0
%T
50.0
40.0 CI 73000 Indigo
30.0
20.0
10.0
0.0
250.0 300.0 350.0 400.0 450.0 500.0 550.0 600.0 650.0 700.0 750.0
nm
Figure 6.6 Synthetic indigo dyed on cotton showing absorption in the whole UV-vis region.
Double bonds –(C C)n
Triple bonds –(C C)n
Azo group –N N
Cyan group >C N
Aldehyde group R >C O

H
Keto group R >C O

R
O
Carboxylic acid R–C
OH
Figure 6.7 Chromophoric groups.
Further, the chemical theory of colour is connected with the term auxochrome.
Auxochromic groups are basic, salt-forming groups such as the hydroxyl group and the amino
group, both have only weakly bonded electrons which are easily movable. The introduction
of auxochromic groups into a coloured molecule leads to an increasing depth of the colour
and simultaneously allows the molecule to bond onto another substance (e.g. fibres). To put
it simply, dyes originate from the combination of a chromophore with an auxochrome. The
absorption of the chromophore of a dyestuff may be additionally influenced by substituents
which not only change its chemical constitution but also influence the electronic configura-
tion within the molecule. In Figure 6.8 the spectrum of synthetic indigo (C.I. 73000, vat
blue 1) is compared with the spectrum of one of its derivatives (C.I. 73065, vat blue 5) which
has four bromine atoms as substituents. It can clearly be seen that the substituents cause a
shift of the absorption maximum from about 660 nm to about 600 nm to the blue (hypso-
chromic shift) and in the UV region - close to 290 nm - the spectrum is shifted by almost 10
nm to the red (bathochromic shift).
6.1.4 Instrumentation, Validation and Calibration

Microspectrophotometers are composite instruments. The main components are the
microscope, the photometer devices, the computer with the software, and the printer to
obtain hard copies of the spectral data. The application of computers and software pack-
ages, which include extensive administration, evaluation and special user programs to
record the spectra and to manipulate and compare them using mathematic-statistical
methods, is a prerequisite for comfortable handling of the entire system, and is absolutely
necessary for routine laboratory work. The software packages usually also offer a colorim-
etry programme (see Section 6.7) which can be used not only for objective colour measure-
ment but also for compiling data collections.
Experience gained over more than four decades shows that in forensic practice, under
the burden of a heavy caseload, the fact that microspectrophotometers are highly sophisti-
cated instruments is often overlooked. Specialized knowledge of each component is needed
100.0
CI 73065 Tertrabromoindigo
CI 73000 Indigo
90.0
80.0
70.0
60.0
50.0
%T
40.0
30.0
20.0
10.0
0.0
250.0 300.0 350.0 400.0 450.0 500.0 550.0 600.0 650.0 700.0 750.0
nm
Figure 6.8 UV-vis spectrum of vat blue 1 (synthetic indigo) and of vat blue 5, demonstrating
the influence of substituents on the absorption characteristics.
so that they can be optimally adjusted and calibrated in order to contribute to a valid result
according to quality assurance.
Two different types are to be differentiated, the multichannel microspectrophoto-
meter which is equipped with a photodiodearray detector (PDA) or a charge coupled device
detector (CCD) and the (single channel) scanning microspectrophotometer with a grating
monochromator, a photomultiplier tube and a photodetector.
6.1.4.1 Microscope
The microscope is the centre of a microspectrophotometer system. Its function is not only
to place the object in a defined position - the stage plane - but also to allow reproducible
focusing of the radiation onto the sample. The task of the microscope is to transport light
energy. Therefore, the concepts of geometrical optics of the microscope must be known
and must be strictly followed. This subject is covered in depth and in a comprehensible
manner by Piller (1977). This book is a useful source of information about the different
aspects of microspectrophotometry. It is not very commonly known that the microscope
plays its own role in obtaining high-quality spectra. The user must know how to set up
the microscope for Koehler illumination. This is indispensable for correct centring of the
microscope. Incorrect centering may cause variation in intensity and decrease the repro-
ducibility of measurements.
From the theory of geometrical optics, the operator must understand that there are
different ray paths and certain interrelations of lenses and diaphragms. Consequently, the
same image is formed in several optical planes in a microscope. These are called conju-
gated focal planes. They play an important role in multichannel systems and in partic-
ular in scanning microspectrophotometers because the images received by the essential
parts of the photometric system as light source, monochromator exit slit, illumination
diaphragm, measurement diaphragm, photomultiplier are conjugated. For example, in

normal bright-field illumination, which is the standard situation for fibres measurement,
the surface of the lamp, the entrance and exit slits of the monochromator, the condenser-
aperture diaphragm, and the back image of the objective are conjugated. The back image
of the objective can be easily seen with the so-called auxiliary telescope which should have
been delivered with the system. The centering and the focus of the lamp as well as of the
exit slit of the monochromator can be observed in this plane and corrected if necessary.
Experience shows that an inaccurately adjusted lamp is one of the most frequent reasons
for noisy spectra. An optimal lamp adjustment is critical if a xenon lamp is used.
The Zeiss MPM 800TM is an excellent example of the extraordinary importance of the
geometrical optics in a microspectrophotometer. In practice, MPM 800 spectra - particu-
larly from round, non-delustred man-made fibres with pale colour shades, measured with
the conditions which have been set according to theory, often show transmittance of more
than 100% in the visible region. This problem can be overcome when the original illumina-
tion diaphragm is used as measurement diaphragm and vice versa.
The main function of the microscope is to transmit light energy, therefore the optical
flux and the transmittance of the system are additional criteria which must be observed
(Piller, 1977). The size of the field and aperture diaphragms and the diameters of the
exit pupil of the microscope and of the photocathode will all influence the optical flux.
Transmittance of the optical system is the amount of the light remaining after loss by
absorption, reflection, scattering and diffraction, i.e. after interaction between the light
beam and all optical elements in the system. These parameters cannot be influenced to any
extent or varied by the user of a microspectrophotometer. Their optimization is the respon-
sibility of the manufacturer. The user does have some possibilities for selecting the appro-
priate objectives, fibre optics and/or condensers. The condenser should have a relatively
low aperture value because light rays transmitted under higher angles are more strongly
absorbed than those transmitted vertically. The literature recommends a condenser with
an aperture not exceeding 0.6.
For measurements in the vis region only, the objectives should be apochromatic fluo-
rit objectives. They should have a high image contrast like planapochromatic objectives.
Measurements in the UV region require special optical equipment with so-called mirror
objectives or lens objectives which are made from quartz glass. Of course, all other opti-
cal parts of the apparatus - the condenser, etc. - must also transmit UV radiation. Mirror
objectives and lens objectives are suitable for the vis region too. Because of their special
construction, mirror objectives have the disadvantage of a shadow in the centre, resulting
in a loss of contrast and an increase in reflections (Gerlach, 1976). This may be why the
highest quality spectra from single fibres are those measured with quartz lens optics. Using
a PDA connected to a workstation microscope, it is important to remember that in most
microscopes there are optical reflectors made from glass plates, mirrors or prisms. These
devices often cause interference fringes which can be seen in a spectrum as a regular wave
structure with short frequencies (Figure 6.9).
6.1.4.2 Spectrophotometer Devices

6.1.4.2.1 Scanning Microspectrophotometer The scanning spectrophotometer con-
sists of three components: a light source, a monochromator and a photomultiplier. For
fibre measurements using the UV-vis range, the monochromator is usually placed in front
of the object (Figure 6.10).
100.0
90.0
80.0
70.0
60.0
%T
50.0
40.0
30.0
20.0
10.0
0.0
250.0 300.0 350.0 400.0 450.0 500.0 550.0 600.0 650.0 700.0 750.0
nm
Figure 6.9 Spectrum with interference fringes originating from optical reflectors in the
microscope.
Photomultiplier tube
Dispersion device
detector
Entrance slit Exit slit
Source
Monochromator
Figure 6.10 Scheme of a scanning microspectrophotometer consisting of a light source,

a monochromator and a photomultiplier detector.
6.1.4.2.1.1 Light Source The type of light source used depends on the spectral region to
be measured. Tungsten lamps (halogen 12 V/100 W) provide the vis region, whereas xenon
lamps (XBO 75 W type) are only necessary if the UV region is to be included. In both
cases the light source must be stabilized. Xenon lamps produced by several companies
have considerable differences in life expectancy and stability, not only between the prod-
ucts of different producers but also within products from a particular company and from
batch to batch. Experience has shown that it is more economical to invest a little more
money in the selection of xenon lamps of higher quality than to change a lamp after every
100 or 200 hours and to readjust the system. Experiments at the BKA have shown that
the dimensions and construction of the lamp house also exert considerable influence on
the lamp stability. Experience shows that in practice the lamp and the lamp house are
among the most critical points of the system, as all the energy originates from here. Also,
the alteration and the adjustment of the lamp—especially of the xenon lamp—the most
common reasons for poor-quality spectra.
6.1.4.2.1.2 Monochromator A monochromator produces monochromatic light of
a certain spectral bandwidth. The object is scanned with these spectral bands step by
step and the spectrum is created point by point. The measurement time is usually about
a minute. This scanning principle is known as scanning spectroscopy. Modern mono-
chromatizing devices are exclusively of the grating variety. Grating monochromators
allow linear calibration of the wavelength scale. Continuous-interference filters or prism
monochromators are not suitable in modern microspectrophotometers. The gratings in
UV-vis monochromators are usually reflection gratings of the ‘Echelle’ type and nor-
mally have 1200 grooves. They are holographically blazed. Monochromators with this
type of grating have a pronounced reflectance maximum at a particular wavelength,
called the ‘blaze wavelength’, with a maximum of light due to specular reflection. For
wavelengths longer or shorter than the blaze wavelength, the reflectance (or the trans-
mittance of the monochromator) decreases rapidly. At this specific point a polarization
of diffracted energy occurs, causing enhancement of the signal which often is reproduced
in a spectrum as a distinct transmission maximum. This effect is called Wood’s anom-
aly and should be suppressed with the help of a polarizer placed in the optical path in
front of the object. Further, the grating produces a series of spectra with different orders
of diffraction, causing a series of harmonics with wavelengths superimposed on the set
wavelength. Harmonics must be suppressed with the help of blocking filters. In the MPM
800TM these filters change at about 380 nm and 630 nm, depending on the specific values
of the individual filters.
6.1.4.2.1.3 Photomultiplier At the end of the optical path a photosensitive device is
necessary to multiply the low-light signal and to transform the photons of the light into
electrical energy. In the scheme of a photomultiplier detector presented in Figure 6.11 the
1 3 5
9
Photon e–
a
2 4 6 8
b 0
Photomultiplier tube (PMT)
Figure 6.11 Scheme of a photomultiplier detector. 0: photocathode, 1–8: dynodes, 9: anode, a:

incident light, b: focusing device.
incident light beam (a) is concentrated by a focusing device (b) and hits the photocathode (0).
The evolving secondary electrons are multiplied through a cascade of dynodes (1–8) and
finally detected by the anode (9).
6.1.4.2.2 Multichannel Microspectrophotometer The multichannel spectrophotom-
eter consists of three components as well: a light source, a polychromator, and a photodio-
dearray (PDA) or charged coupled device (CCD) detector (Figure 6.12).
6.1.4.2.2.1 Light Source Modern PDA UV-vis microspectrophotometers use a mixture
of light emitted by a deuterium lamp (UV) and a tungsten lamp (vis). This light is directed
onto the fibre sample either via glass fibre optics or the normal light pass of the microscope.
6.1.4.2.2.2 Polychromator / Photodiodearray Detector PDA detectors have been used in
MSP since the middle of the 1990s (Adolf, 1998). At first covering only the visible range of
electromagnetic radiation the last generation of multichannel microspectrophotometers
uses the whole UV-vis region from about 240 nm to 760 nm. Photodiode detectors have
a wider dynamic range and, as solid-state devices, are more robust than photomultiplier
tube detectors. Multichannel spectroscopy allows spectra to be created from simultaneous
recording across the entire spectral range being scanned. The measuring time is there-
fore reduced to less than a second. In a photodiode, light falling on the semiconductor
material allows electrons to flow through it, thereby depleting the charge in a capacitor
connected across the material. The amount of charge needed to recharge the capacitor at
regular intervals is proportional to the intensity of the light. A diode array consists of a
series of photodiode detectors positioned side by side on a silicon crystal. Each diode has
a dedicated capacitor and is connected by a solid-state switch to a common output line. A
shift register controls the switches. Initially, the capacitors are charged to a specific level.
PDA/CCD detector
Dispersion device
Source
Entrance slit
Polychromator
Figure 6.12 Scheme of a multichannel microspectrophotometer consisting of a light source,

a polychromator and a PDA or CCD detector.
When photons penetrate the silicon, free electrical charge carriers are generated that dis-
charge the capacitors. The capacitors are recharged at regular intervals that represent the
measurement period for each scanning cycle.
PDA detectors are usually preferred if high signal intensities must be measured,
because of their better signal/noise ratio and their high saturation charge. CCD detectors
are approximately 100 times more sensitive than PDAs, and are said to be advantageous
in situations such as fluorescence emissions, where only low intensities are present. The
main difference between PDAs and CCDs is that a photodiode cannot accumulate weak
signals of a special wavelength. In contrast, within a CCD several sensitive devices are
charged coupled and are therefore able to accumulate very weak signals with a very good
signal/noise ratio. To lower the noise, CCDs can be cooled down below room temperature.
Different detectors such as CCDs, intensified CCDs and electron multiplying CCDs are
available.
A PDA is composed of a grating fixed together in one unit with a photodiodearray as
the photosensitive device onto which the spectrum is reproduced. In contrast to scanning
MSP systems in PDA MSP the full intensity of the light source is directed onto the fibre.
This can cause bleaching of sensitive fibre dyes with a low colour fastness (Figure 6.13). In
Table 6.1 different features of scanning and multichannel MSP are set against each other.
100.0
600 sec
90.0
80.0 300 sec
70.0
180 sec
60.0 120 sec

%T
50.0 60 sec
40.0
30.0
20.0
10.0
0.0
250.0 300.0 350.0 400.0 450.0 500.0 550.0 600.0 650.0 700.0 750.0
nm
Figure 6.13 Bleaching of fibre dyes with low colour fastness. The spectra presented show the
bleaching of irradiation intervals of 1, 3, 5, 10, 20, 30, 60, 120, 180, 300, and 600 sec.
Table 6.1 Characteristics of Photomultiplier Tubes (PMT) and Photodiodearrays (PDA)

Characteristics PMT PDA Characteristics PMT PDA
Wavelengths 200–900 200–2000 Voltage Very high Low
nm nm
Performance/Costs Fair Good Light intensity Low High
Sensitivity Excellent Very good Fibre bleaching No May occur
Linearity Good Excellent Measuring time Slow Very rapid
Noise performance Fair Very good Construction Fragile glass Solid
Vacuum tube Semiconductor
Dynamic range Very good Excellent Size Large Small
Stability Very good Very good Ruggedness Poor Excellent
Drift Yes No Ease of Poor Easy
customization
Reproducibility Very good Excellent
6.1.4.3 Data Control, Processing, Recording

Microspectrophotometry requires the input of different settings for operation, the con-
trol of these data, the processing of many measurement values and, finally, the record-
ing of all the data as spectra and figures. This multi-task is managed with the support of
a microscope system processor unit containing the amplifier, the feed-forward control for
the shutters, etc., and is assisted by a microcomputer and a Printer/Plotter. Commercially
available microspectrophotometers have fully integrated these components.
6.1.4.4 Validation
To meet the demands of quality assurance a new MSP must be validated before it is intro-
duced into the examination of textile fibres. Reasons to substitute a microspectrophotom-
eter are for example the reduced reliability of an old system, a limited efficiency, the limited
availability of spare parts or problems with service. It is recommended to produce a valida-
tion plan including an appropriate set of parameters. Each parameter will have a standard
and a target. Standard means a criterion which must be met by the tested system; target
means a performance criterion which is desirable but not essential. Before starting the vali-
dation the measurement parameters must be specified. The validation includes parameters
concerning the instrument and the robustness of the method:
• Diaphragm setting with optimized size and boundary accuracy

• Accuracy of wavelength registration
• Accuracy of photometric measurement
• Minimum noise level
• Second order effects like light from multiple orders
• Spectral distortion due to system overload
• The impact of focus level on shape and wavelength positions of peaks
• Appropriate detection limits
• Fluorescence measurement
• Time needed for the measurement of a spectrum
• Tests for operator effects
• Tests for bleaching effects

• Impact of external interferences like light, heat, vibration, etc.
• Does polarization affect the spectra
Other parameters concern the software of the system:
• Accuracy of implemented algorithms (colorimetry)

• Accuracy of the output spectrum of the data in both intensity and absorbance
plots
• Is the input of case data (e.g. case number, etc.) possible to an acceptable level
Most of the above mentioned parameters are tested using secondary standards which are
described in the next section.
6.1.4.5 Instrument Calibration

To ensure accuracy in spectral measurement, the microspectrophotometer must first be
calibrated. It is a prerequisite for accurate routine work as well as for intra- and inter-
laboratory comparison of the results (Hartshorne and Laing, 1991a) and should therefore
be part of any quality assurance procedure. In MSP, control by calibrants is required in
three areas: wavelength accuracy and spectral resolution, absorbance and spectral linear-
ity, and colorimetry (Section 6.7). Calibration should be done twice a year using certified
secondary standards which are calibrated to NIST primary standards (National Institute
of Standards and Technology, USA). According to an established quality assurance system
detailed measuring conditions should be fixed.
Wavelength accuracy and spectral resolution over the UV-vis region can be quickly
checked using secondary standards composed of holmium filter glasses. Such holmium
filter glasses show—depending on the batch—certified narrow peaks at 279.3 nm, 360.9 nm,
453.5 nm, 536.3 nm, and 637.9 nm (Figure 6.14). Control of absorbance is required par-
ticularly if colorimetry (see Section 6.7) is to be used for recording and comparison of
colour in a fibre data collection. Hartshorne and Laing (1991a) specifically reported the
problems which must be considered as well as the production of a standard that could be
used by many laboratories. The absolute absorbance accuracy can easily be controlled with
a set of 3 secondary standards composed of neutral density filters calibrated to the NIST
primary standards as well. Simultaneously, these filters also roughly indicate the spec-
tral linearity of the system. If the system will be used to measure samples which obey the
Lambert–Beer law, the spectral linearity should be more exactly controlled, for example by
a self-made step-filter as described by Pabst (1980). Such a step-filter is a useful accessory
and may be assembled from ten layers of CB16 filter glass which are stuck together using
a neutral adhesive.
In daily routine work the 100% transmittance line offers a suitable control for the con-
sistency of a microspectrophotometer. This line is acquired by measuring an area without
the object. Because no absorbing matter is measured, the transmittance at all wavelength
steps must be 100. A deviation of ±1% is acceptable. The 100% transmittance line should
be constant about 30 minutes after the lamp and the system have been switched on for
thermal and electrical equilibration. It can also be used as an indicator for alterations in
the behaviour of the lamp, especially the xenon lamp.
1.6
446.0
1.5
242.0
1.4
1.3
1.2
1.1
1.0
0.9
ABS
0.8
460.0
279.0 361.0
0.7 454.0
288.0
0.6
0.5
0.4
334.0 419.0 537.0
0.3
484.0
638.0
0.2
0.1
250.0 300.0 350.0 400.0 450.0 500.0 550.0 600.0 650.0 700.0 750.0
Filter 666–015 nm
Figure 6.14 Secondary standard consisting of a holmium filter glass showing certified narrow
peaks at 279.3 nm, 360.9 nm, 453.5 nm, 536.3 nm, and 637.9 nm.
6.1.4.5.1 Colorimetry In colorimetry, the accuracy of the tristimulus values X, Y, Z and

the chromaticity coordinates x, y, z are the critical criteria. The calibrant must be transpar-
ent. Again, filter glasses must be used, which are calibrated by standardization agencies.
Experience shows that a set of three filters should be available, for the blue, green and red
part of the spectrum respectively. Again these certain filter glasses with specific charac-
teristics are exclusively used for calibration purposes in MSP. In general, these secondary
standards must be kept in conditions which prevent any alterations occurring, for exam-
ple, as a result of exposure to light and they must periodically be recertified.
2.0
Dye solution
1.8
Dyed fibre
1.6
1.4
1.2
ABS
1.0
0.8
0.6
0.4
0.2
0.0
250.0 300.0 350.0 400.0 450.0 500.0 550.0 600.0 650.0 700.0 750.0
nm
Figure 6.15 Spectrum of a purple dyestuff applied on PA 6.6 compared with the spectrum of
pure dye solution.
The absorption characteristics of a molecule are also influenced by its chemical envi-
ronment. With dyes, the chemical environment is the substrate onto which the dyestuff is
bonded. Figure 6.15 illustrates this effect, which results in spectra with different shapes.
It is obvious that the dyestuff fixed in the fibre (in this case a polyamide 6.6) produces
a spectrum which is definitely different from that of the dye in solution.
The influence of the environment on the shape of the spectrum has some practical
consequences. Man-made fibres such as polyamide, polyester or polyacryl usually have
a homogeneous chemical constitution. In comparison to natural fibres such as cotton and
wool, the dye in man-made fibres is bonded to a relatively constant chemical environment.
In consequence, the spectra of dyed man-made fibres usually show less intra-sample varia-
tion with respect to the wavelength position of the absorption bands. On the other hand,
cotton and particularly wool are composed of many different chemical components which
in addition are inhomogeneously distributed throughout the fibre matrix. Due to this,
intra-sample variation in natural fibres is the rule rather than the exception. From these
facts it is easy to conclude that the application of MSP to forensic examination of fibres is
subject to some special conditions concerning sample selection and preparation and the
comparison and evaluation of the resulting spectra.
6.1.5 Spectral Measurement

The measurement procedure as well as the actual measurement parameters must conform
to the demands arising from the object being measured. Fibres usually are transparent
objects. They can therefore be examined in the transmission mode. This is an advantage
compared with opaque materials such as paints, because absorption spectroscopy provides
more spectral information than reflectance spectroscopy.
Fibres both undyed and dyed often fluoresce due to optical brighteners and/or dyes
applied to them. Fluorescence emission is a coloured spectral message about a substance
that offers further information about the chemical structure of a substance different from
its absorption characteristics. Currently, in forensic examination of fibres, fluorescence
emission is examined visually using incident fluorescence excitation with the workstation
microscope. Fluorescence spectroscopy may offer a more objective procedure to evaluate
fluorescence emission of fibres. Especially modern PDA or CCD microspectrophotometers
offer several advantages in the measurement of fluorescence emission.
6.1.5.1 Absorption Measurement

The spectrum of a single fibre is usually measured in the transmission mode. Experience
shows that after a short period of learning how to handle an MPM system, different users
develop a standard measurement procedure with distinct measurement conditions which
rarely need to be changed because they fit more than 90% of cases. The measurement proce-
dure must take account of the fact that MSP is a special use of photometry and a special use of
the microscope, and must also consider the particular characters of fibres. Fibres are long and
thin, they are not plane objects, they are birefringent and they show diffraction phenomena.
If the measurement is extended to the UV region, it must be remembered that the
samples are exposed to a high level of radiation and that dyed fibres with a low colour fast-
ness may be bleached within few seconds if measurement conditions do not prevent this
behaviour. In consequence concerning scanning MSP, the fibre must be measured using
monochromatic light of narrow bandwidth, and the monochromator must be placed in
front of the sample. Using multichannel MSP a special shutter should cut off the full inten-
sity of the light source except during spectrum measurement.
6.1.5.1.1 Microscope and Geometrical Optics Once the object to be measured has been
placed on the microscope stage and focused, the system must at first be set up according to
Koehler illumination using white light. Next, the measurement and illumination slits both
must be focused and centred in the viewing field of the microscope. Because Koehler illumi-
nation does not display the illumination source in the object plane, and in scanning MSP the
monochromator exit slit is used as illumination, the slit position and homogeneity of the illu-
mination of the exit slit of the monochromator should be checked. The exit slit is displayed
in the back image of the objective and can be easily observed with the auxiliary telescope.
6.1.5.1.2 Object Characteristics Fibres are linear objects. This means that a long, nar-
row rectangular shape is the most suitable one for the measurement slit in comparison
to a circular diaphragm, where the area of measurement is limited by the fibre diameter.
A slit aperture increases the light level and averages any local variations along the fibre.
Experience shows that a proportion of 10:1 length to width of the measurement slit is a
standard size which fits most fibre types. Of course, the slit must be centred within the
fibre. Further, the more or less round shape of the fibres requires that the actual width of
the slit does not cover more than 1/3 of the fibre diameter. The shape of the illumination
slit must also be rectangular (unless a pinhole measurement diaphragm is used). It should
generally be larger than the measurement slit, but there is no conformity to a particular
size. Therefore the size which should be used in a particular laboratory must be found by
tests with standard objects.
In a multichannel microspectrophotometer the incident light is transported via glass
fibre optics. These fibre optics are connected to the normal light pass of the microscope—in
this case all the optics, shutters and diaphragms of the microscope can be used—or the
fibre optics are placed just beneath the object where normally the condenser is placed. As in
scanning microspectrophotometers in a multichannel system the width of the illumination
diaphragm and the measuring diaphragm directly influences the quality of dye spectra. An
illumination diaphragm which is opened too much results in false light and values above
100% transmission. In contrast, an illumination diaphragm which is too narrow results in
absorption values which are too low. As a consequence, when setting up a system calibration
using secondary standards has to be done to find the optimal diaphragm setting.
Fibres are birefringent. Some of them, such as cotton, even show extreme polarizing
effects which may cause serious artefacts in the spectra. This, and the already mentioned
polarizing effects in the microspectrophotometer itself, make it necessary with scanning
MSP to use a polarizer which must be placed in the front of the object (Nasse, 1998, pers.
comm.). Because the polarizer produces linear polarized light, i.e. light with only one direc-
tion of vibration, the fibre must be oriented parallel to this direction. In many laboratories
the orientation of the polarizer, the fibre, and the diaphragms is therefore standardized to
the north-south direction.
Another object characteristic of fibres is known as pleochroism. Pleochroism is the
variation in colour of a material based on its orientation under polarized light. Because
fibres act microscopically as anisotropic uniaxial crystals, they can exhibit only two
such colours and are said to be dichroic. The difference in colour between the fibre as
oriented in n┴ and n║ is based on the fibre’s and dye’s molecular orientation. The clari-
fication of this molecular orientation provides additional determination of the struc-
tural properties and relationship between the two materials. Houck (1997) developed an
instrumental procedure to characterize fibre dichroism by MSP in order to help in the
specificity of reporting fibre properties in forensic comparisons. De Wael and Vanden
Driessche (2011a, b), De Wael and Lepot (2011, 2012) and De Wael (2012) extended the
research on measuring dichroism of fibres by MSP in detail. In these research projects
they focused on dyed polyester, polyamide, wool, silk, cotton, viscose, acrylic, acetate
fibres and pigmented fibres. The authors showed that the strong dichroic effects of
polyester fibres are related to hypochromic effects in intensity and hypsochromic and
bathochromic shifts in the absorption bands (De Wael and Vanden Driessche, 2011a).
In the three peptidic fibre classes polyamide, wool and silk, the dichroic effects are
much smaller compared to that of polyester fibres and therefore may not be observed
(De Wael and Vanden Driessche, 2011b). In cellulosic fibres, namely cotton and viscose,
dichroic effects depend on the chemical structure of the colourants. Dyes of the azo,
stilbene, thiazole and oxazin type show always dichroism whereas dyes of the sulphur,
diphenylmethane, triarylmethane, and phtalocyanine type never show dichroic effects
(De Wael and Lepot, 2011). In contrast almost 75% of the tested pigmented fibres of
all man-made fibre types and all colour shades showed dichroism mainly depending
on the crystal structure and draw ratio of the fibres (De Wael and Lepot, 2012). The
low birefringent fibres of the acrylic and acetate type only show very small dichroic
effects which are hard to observe with light microscopy. However, using MSP these
small effects could be measured (De Wael, 2012).
6.1.5.1.3 Measurement Conditions of Scanning Microspectrophotometers In routine
work the electromagnetic radiation is measured from 240 to 760 nm. The wavelength reso-
lution of microspectrophotometers in use is at least 1 nm. Therefore the step width (dis-
tance between two points of measurement in a spectrum) as well as the bandwidth of the
monochromator (width of the exit slit) should be less than 5 nm. The experience gained
over almost 40 years shows that best results are obtained by setting these parameters to
2.5 nm. Modern software additionally offers the possibility of varying the number of scans
per spectrum and selecting the number of measurement values which are summarized for
the calculation of the intensity at each wavelength step (the average). In fibre examination,
spectra of high quality are usually obtained by measuring with a higher number of scans
rather than with a high average. The reference, i.e. the I0 spectrum, is measured from a
blank spot in the slide beside the fibre.
6.1.5.1.4 Measurement Conditions of Multichannel Microspectrophotometers As
in scanning systems the range of radiation to be measured is from 240 to 760 nm, though
the PDA or the CCD in use covers a spectral range from c. 200 to 980 nm (1044 pixels).
The integration time can be selected between 7 ms and 8 s and should be set to a value to
exceed 30,000 counts. Due to the use of a combination of deuterium and halogen light
source artefacts can be detected in dye spectra at about 660 nm. These artefacts arise
from a very intense and sharp peak at 660 nm of the deuterium lamp. To prevent such
artefacts the use of a blue quartz glass filter is recommended, which cuts off the radiation
above 400 nm of the deuterium lamp. In Figure 6.16 the radiation spectrum of the deu-
terium/halogen lamp is presented with and without a cut-off filter. A comparison of dye
spectra measured with well adjusted scanning and multichannel MSP reveals that the
peaks, troughs, shoulders and points of inflection are at identical wavelength positions.
The only difference that is found concerns the so-called dynamic range which is accord-
ing to Jochem (2007) a little bit lesser with multichannel systems (Figure 6.17)
5.25E4
4.5E4
3.75E4
3.0E4
E
Without cut off

2.25E4 quartz filter
1.5E4
7500
With cut off quartz filter
0
250 300 350 400 450 500 550 600 650 700
nm
Figure 6.16 Radiation spectrum of a combined deuterium-halogene light source without and
with a cut off quartz glass filter.
100.0
90.0
CCD
80.0
70.0
60.0
T%
50.0 MPM 800
40.0
30.0
20.0
10.0
0.0
250.0 300.0 350.0 400.0 450.0 500.0 550.0 600.0 650.0 700.0 750.0
Figure 6.17 Comparison of a dyed fibre measured with a MPM 800 and a CCD multichannel
MSP.
6.1.5.2 Fluorescence Emission Measurement

Many substances in our world have the property of fluorescence. Therefore, fluorescence
is used for discrimination in many areas in forensic science (Gibson, 1977; Siegel, 1996).
Fluorescent substances absorb light energy and then emit the energy, usually at a longer
wavelength. Because the absorption process is generally linked to multiple bond conju-
gation and aromaticity (see Section 6.3), many textile dyes and especially optical bright-
eners are fluorescent, making fluorescence emission an additional characteristic of high
interest in forensic fibre examination (Kubic et al., 1983; Bresee, 1987). In most laborato-
ries, fluorescence is visually evaluated by incident fluorescence microscopy. The applica-
tion of microspectrofluorimetry for spectral measurement of the fluorescence emission
from single textile fibres is not a standard method in forensic fibre examination.
Until now the practice of the application of microspectrofluorimetry has been reported
only by Adolf (1987) and Hartshorne and Laing (1991b, c, d). Adolf concentrated on using
the additional spectral information to achieve further differentiation, whereas Hartshorne
and Laing concentrated on the application of colorimetry to fluorescence emission spec-
tra. The microscope of the microspectrophotometer must be equipped with an incident
fluorescence illuminator. The illuminator should be fitted with filter cubes for excitation
by the mercury lines at 365 nm and 436 nm respectively. The best results were achieved
using plan-neofluar immersion objectives. Because of the lack of an absolute fluorescence
standard and the usually low level of fluorescence intensity of fibres, it was found useful to
display only relative fluorescence spectra. The formula for computing the values of fluores-
cence at each wavelength step must take account not only of parasitic light, but also of the
theoretical values of a tungsten lamp with a colour temperature of 3200 K as the reference,
and the real intensity as standard. In case of very weak fluorescence a sensitive scanning
MSP is the appropriate system to record the signals but since the introduction of PDA/
CCD MSP and their adaptation to the photoport of normal microscopes’ strong fluores-
cence of colourants and in particular of optical brighteners are preferentially and simply
measured with such systems.
Figures 6.18 and 6.19 illustrate two important points. First, optical brighteners have
two different types of spectra (Figure 6.18). Second, the fluorescence emission spectra of
metameric dyed fibres may have different shapes which can be used to distinguish between
metamers (Figure 6.19).
2.0
Blancophor Blancophor
REU ANR
1.8
1.6
1.4
1.2
Cts
1.0
0.8
0.6
0.4
0.2
0.0
400.0 420.0 440.0 460.0 480.0 500.0 520.0 540.0 560.0 580.0 600.0
nm
Figure 6.18 Fluorescence spectrum of Blankophor ANR and REU.
2.0
1.8
1.6
1.4
1.2
Cts
1.0
0.8
0.6
0.4
0.2
0.0
450 475 500 525 550 575 600 625 650 675 700 725 750
nm
Figure 6.19 Fluorescence spectra of two metameric dyed fibres.

6.1.5.3 Sampling and Sample Preparation

In microspectrophotometry, the general handling and tracking of samples must guarantee
the continuity of the recovered fibres. Known fibre sample selection must represent the
complete range of fibre colours and dyeing depths in the known textile or fibre source
which means sampling must be representative for a garment. Figure 6.20 shows different
coloured linings in the two pockets of an anorak.
Additionally, it must be taken into account that the extent of wear, bleaching and laun-
dering (Was-Gubala and Grzesiak, 2010) and biological, thermal and/or mechanical influ-
ences (Was, 1997) may cause serious artefacts which increase the colour variation within
a textile or fibre source. Therefore, it is not really possible to standardize the numbers of
known fibre samples selected for measurement. General experience has shown that natural
fibres have a higher degree of intra-sample variation than man-made fibres, meaning that
it is normally necessary to measure more natural fibres to get an overview of the spec-
tral variation within the sample. For example, ten natural fibres and five man-made fibres
would be the minimum number normally measured.
The known fibre sample should be selected and well separated under a stereo micro-
scope. It is not advisable to prepare the fibres at random on the slide. The fibres should not
only be mounted in a single layer but should also be oriented in a line as parallel as possible
to each other and to the short edges of the slide. This systematical preparation procedure
saves searching time. If these basic requirements have been considered, 50–70 individual
fibres can be measured per day with the scanning MSP (full UV-vis range), and more than
300 with a multichannel MSP. Since the introduction of quality assurance systems it has
proved necessary to mount only up to 6 questioned fibres on one slide and each fibre under
a small round cover slip to prevent a possible loss during sample preparation.
Microspectrophotometry requires the specimen mounting medium to have low or
no fluorescence. Questioned fibres and known fibres should be mounted using the same
medium. Many products are available, of which Phytohistol (Grieve and Deck, 1995) and
Entellan New (Wiggins and Drummond, 2007) are examples. If measurement in the UV
region is necessary, the use of quartz slides and cover slips is necessary, as well as non-
fluorescent glycerol as mounting medium.
Figure 6.20 Anorak showing different coloured linings in the two pockets.
6.1.6 Spectral Comparison

All analytical techniques special limitations are inherent. Therefore, a prerequisite for
a reliable comparison of dye spectra measured by microspectrophotometry is a profound
knowledge about these limitations and additionally about the discrimination power of the
applied technique.
6.1.6.1 Quantitative and Qualitative Limitations

Quantitative limitations may be encountered when dealing with very pale coloured fibres
or very deeply coloured fibres. That means that the fibres may absorb either too little or too
much radiation, so that the spectra do not show recognizable patterns. Although MSP still
produces a clear absorption feature a very pale dyed single fibre is perceived as colourless
by the human eye. In practice this is more a theoretical problem because when searching
fibres in tapes the main search criteria is the colour of a fibre because, as a rule, colourless
fibres, especially colourless cotton fibres, are not suited to indicate a fibre transfer. With
very deeply dyed fibres, which therefore have low transmittance values, the full spectral
information will be obtained only if the spectrum is calculated in absorbance units. If this
is unsatisfactory, cross-sections can be made or synthetic fibres can be squashed.
Qualitative limitations may arise from fibres which are dyed using colourants with
very similar chemical structure. According to Adolf (1986), synthetic indigo (CI 73000)
can be clearly distinguished from its derivatives and homologues. However, within
the derivatives there are some bromine-substituted dyes which apparently give identi-
cal spectra (Figure 6.21). From this one might conclude that due to similar chemical
100.0
Tetrabromoindigo
90.0
80.0
70.0
Hexabromoindigo
60.0
50.0
%T
40.0
30.0
20.0
10.0
0.0
450 475 500 525 550 575 600 625 650 675 700 725 750
nm
CI 73065: 5,7,5‘,7‘–Tetrabromoindigo
5,7,5‘,7‘ and CI 73075: 4,5,6,4‘,5‘,6‘-Hexabromoindigo
Figure 6.21 Spectra of the bromine-substituted derivatives of synthethic indigo, showing very
similar shapes.
structures of dyes there may be serious limitations to the application of MSP in forensic
fibre examination. Again, in reality, textile dye houses would regard this problem as the
exception rather than the rule, because dyes are usually applied in mixtures of two or
three components. The application of only one dye component, as in the case of denim,
is an exception.
Besides indigo dyed blue cotton fibres the only real limitation arising from the struc-
ture of spectra which has been found to play a role in practice is presented in Figure 6.22.
The spectra originate from the cotton fibres of a black T-shirt dyed with sulphur black 1,
the most widely used black sulphur dye. Such spectra are easy to recognize. The fibres
appear microscopically grey and the spectra are normally very simple featureless curves
with wide and weak absorption bands and flat transmission maxima. It is very difficult
to draw conclusions about the possible origin of recovered fibres dyed with sulphur dyes,
because of the wide range of variation (due to concentration effects) and shifts in peak
maxima up to 40 nm which occur in the spectra from known samples (Grieve et al., 2001).
In such cases it is very difficult or even impossible to come to any conclusion, let alone to
decide whether spectra match. Even after many measurements it may be impossible to
make a serious decision as to whether the target fibres match the known sample or not. But
as we will see in the next section, fortunately, this is only true for cotton fibres dyed with
indigo or sulphur black 6.
6.1.6.2 Discriminating Power

The discriminating power (DP) of microspectrophotometry was first evaluated by Macrae
et al. (1979). Various samples of red, blue and black wool fibres were examined by MSP and
by TLC. The DP was found to be practically the same for both methods. For example, the
DP of MSP was found for the red wool fibres to be 0.94 and for the blue wool fibres to be
0.99. From these results it was concluded that MSP and TLC are complementary methods.
Another complementary trials carried out by Wiggins et al. (1995) and De Wael et al. (2015)
100.0
90.0
80.0
70.0
60.0
%T
50.0
40.0
30.0
20.0
10.0
0.0
250.0 300.0 350.0 400.0 450.0 500.0 550.0 600.0 650.0 700.0 750.0
Sulphur Black 1 nm
Figure 6.22 Spectra from three different cotton fibres from a T-Shirt dyed with sulphur black 1,
illustrating extreme wavelength shifting.
concerned red cotton fibres and black, blue and red cotton fibres respectively. As a result
of this researches it can be stated, that MSP—especially if the UV range is included—is a
highly discriminating technique and generally complementary to TLC.
The initial evaluation of the technique by Macrae et al. was followed by several other
studies which focused on dyed polyester, wool and cotton fibres (Beattie et al., 1979, pers.
comm.; Jenne, 1981; Grieve et al., 1988, 2001, 2003, 2005; Cassista and Peters, 1997; Robson,
1997; Biermann, 2007; Palmer et al., 2009). The reported DP values depend on the investi-
gated fibre/colour combination with highest DP-values of about 0.999 for blue cotton fibres
(Biermann, 2007).
6.1.6.3 Spectral Comparison

In this context there are four questions of particular interest. These often provoke serious
discussions and controversial opinions, as they affect the logistical and economic problems
to be faced when dealing with a heavy case load.
• How many individual fibres must be microspectroscopically measured from

a group of questioned fibres, in order to conclude with a high degree of certainty
that all individual fibres of the group have a similar spectral behaviour?
• Should the full UV-vis region be measured and, if so, should this be done regularly?
• Should the spectra be recorded in absorbance units or in transmittance units?
• What is the best way to compare spectra?
The question of how many individual fibres must be microspectroscopically measured from
a group of fibres in order to conclude with a high degree of certainty that all individual
fibres of the group have similar spectral behaviour applies not only to the questioned fibre
material but also to the known fibre material. This represents a basic problem in material
science, i.e. what is the representative size of a random sample? On the one hand it is likely
to become an important issue now that forensic science is being forced to examine criti-
cally its economic side and to present precise answers. Only a few laboratories were more
deeply involved in the problem of optimal sampling size in forensic fibre examinations
(Leijenhorst, 1998, pers. comm.). It is obvious that statistics and probability calculations
will be involved. But it is still an unsolved problem although there has been an approach by
a sub-working group of the European Fibres Group (EFG). To summarize, it is clear that
the question of optimal sample size with special regard to fibres and MSP is far from being
resolved. On the other hand experience gained over the last years shows that the introduc-
tion of fast scanning (vis) microspectrophotometers into the first step of fibre examination
dramatically reduced the amount of fibres which have to be measured in the UV region.
From the purely scientific point of view, spectra from the full UV-vis region generally
provide more spectral information and consequently enhance the discriminating power
(Grieve et al., 2001, 2003, 2005; Biermann, 2007). However, it does not follow that the full
UV-vis region must always be measured. It is a question of the economics of the examina-
tion procedure that the UV region is not measured if the spectra of the questioned fibres
already differ in the vis region but if the spectra cannot be clearly distinguished in the vis
region, the UV region has to be measured or another complementary method such as TLC
should be carried out as a control and verification step.
Figures 6.23 and 6.24 illustrate that in some cases spectral differences between fibres
from different origins can only be detected in the UV region. It is estimated that about 10%
100.0
90.0
80.0
70.0
60.0
%T
50.0
40.0
30.0
20.0
10.0
0.0
250 300 350 400 450 500 550 600 650 700 750
Blue CO from different sources nm
Figure 6.23 Spectra from two blue cotton fibres from different garments, which are clearly
different only in the UV region.
100.0
90.0
80.0
70.0
60.0
%T
50.0
40.0
30.0
20.0
10.0
0.0
325.0 375.0 425.0 475.0 525.0 575.0 625.0 675.0 725.0
Red PES from different sources nm
Figure 6.24 Spectra from two red polyester fibres from different garments, which are clearly
different only in the UV region.
of the questioned fibres examined show similar spectral behaviour to the control fibres in
the vis region but are different in the UV part of the spectrum. It is also estimated that
for more than 50% of questioned fibres the visible spectrum alone is not very informative,
necessitating additional examination. The measurement of the full UV-vis region offers the
quickest way of doing this and is non-destructive.
The financial aspects of recording spectra in the full UV-vis region must also be men-
tioned. The necessity for optics transmitting UV radiation makes MSP systems nearly
twice as expensive as systems for the use in the vis region only. The necessary quartz slides
and cover slips are expensive as well. On the other hand, if a fibre laboratory has only a vis
range instrument available it is essential to use an additional method for dye analysis. TLC
is the only alternative at present which offers similar discriminating power to a measure-
ment of the full UV-vis region.
The software of microspectrophotometer systems normally offers a choice of recording
absorbance or transmittance curves. In photometry of solutions in cuvettes, absorbance
is commonly used to record the spectrum because there normally is a linear relationship
between concentration and absorbance (Beer’s Law). This custom may be the reason that
absorbance spectra are usually the preferred choice over transmittance spectra when ana-
lyzing dyed fibres. With dyed fibres (in contrast to dye solutions) the relationship between
the dye concentration and the absorbance is normally nonlinear (e.g. Was-Gubala and
Starczak, 2015), as different factors influence the absorbency of the dyed fibre. Therefore
absorbance spectra cannot provide the usual support. Nevertheless, absorbance spectra
may facilitate comparison of spectra in casework. As mentioned before, if transmittance
values are lower than 10%, which may happen with deeply dyed fibres, an absorbance spec-
trum is preferable (Suzuki et al., 1994). Absorbance spectra are necessary if colorimetry is
to be carried out using complementary chromaticity values (see Section 6.7).
Most spectral analyses require data to be taken from only one point on a curve i.e. the
analytical wavelength. In contrast, the spectral information of an absorption spectrum of a
dyed fibre is not located in one point. The information is assembled not only by the intensi-
ties and wavelength positions of absorption minima and maxima but also by their shapes
and the shapes of shoulders and other features such as points of inflection or the incline of
the different sections of the curve. All these features cannot be simply described by math-
ematical procedures but are quickly registered by the eye-brain system.
In conclusion, correlation between all these features, including the general shape of the
curve, must be established before a spectral match is concluded. Therefore it is extremely
important when comparing spectra of dyed fibres to consider the curve in its entirety.
The easiest and most common way to do this is by use of a light box or on the computer
screen. The spectrum of a questioned fibre can be overlaid on spectra of the known fibres
which function as standards. If many spectra are to be compared, it is advisable to do this
systematically. One way is to control separately the intra-sample variation of the known
fibres and the questioned fibres first (Figure 6.25). This provides an overview about the
homogeneity/inhomogeneity of the fibres group to be compared, and allows the establish-
ment of spectral subgroups. This can either be done by comparing and sorting the spectra
on the screen or with printouts or more comfortably by creating so called libraries of the
spectra under investigation and computer-aided spectral comparison which is offered by
the software packages.
An ideal spectral match is of course concluded if the questioned and the known spectra
are absolutely congruent. This may not be the case in practice, especially when comparing
natural fibres. The basic requirements are that the spectrum from a suspect fibre must lie
within the range exhibited by the replicate standard spectra, and that all structural details
must be similar in both spectra. In connection with this it must always be remembered that
the spectral resolution in MSP is 1 nm, i.e. differences of 2 nm in the absorption maxima
between two spectra may already mean two different colourants.
100.0
90.0
80.0
70.0
60.0
%T
50.0
40.0
30.0
20.0
10.0
0.0
250.0 300.0 350.0 400.0 450.0 500.0 550.0 600.0 650.0 700.0 750.0
CO, black, jogging trousers nm
Figure 6.25 Intra-sample variation of black cotton fibres of a jogging trousers.
This approach to spectral shape comparison is considered by some to be relatively

crude and unsophisticated. Macrae et al. (1979) explored more refined ways of comparing
spectra. The approach taken was to convert spectra to unit areas. Normalized spectra were
obtained by dividing the absorbance values by total band areas. Five so-called difference
parameters were established:
• The sum of the absolute differences between corresponding data points at 10 nm

intervals
• The sum of the squares of the differences between corresponding data points
• The sum of the absolute differences in gradients at corresponding data points
• The maximum difference between the normalized spectra, i.e. the longest vertical
distance between two curves
• The maximum difference between the normalized cumulative distributions
The wavelength position of the absorption maximum (λ max) was also used as a discrimi-
nating parameter. The results showed that in this trial the most powerful discrimination
was achieved by using a combination of sums of squares and λ max. At present, the use of
difference parameters for spectral comparison is mainly focused on search procedures in
spectral data libraries.
Another way to compare spectral data is to calculate derivative spectra (Coyle, 2002;
Grieve et al., 2005; Biermann, 2007). Since new software has been available the comparison
of derivative spectra offers several benefits. Due to the given signal to noise ratio (s/n) of the
UV-vis spectra it is possible to compare 1st derivative spectra. These are calculated using the
algorithm according to Savitzky and Golay (1964). The Savitzky–Golay method essentially
performs a local polynomial regression on a distribution to determine the smoothed value
of each point. In their epochal publication they also provided methods for calculating the
1st up to the 5th derivatives. The higher the number of points selected the smoother is the
result. Selectable values of points are 5, 7, 9, …, 21. To smooth a dye spectrum a factor of 5 is
appropriate. This smoothing factor is not high enough to calculate good 2nd derivatives, but
with higher factors characteristic details of a spectrum will be lost and shifts in wavelength
positions of peaks of up to 10 nm occur (Figure 6.26). Due to derivative calculation peaks
0.02
0.015
0.01
0.005 5
21
0.0
ABS/nm
–0.005
–0.01
21
–0.015
–0.02 5
–0.025
–0.03
250.0 300.0 350.0 400.0 450.0 500.0 550.0 600.0 650.0 700.0 750.0
Derivative spectra reactive black 5 nm
Figure 6.26 Impact of different smoothing factors from 5 up to 21 on the shape of derivative
spectra showing loss of shoulders and peak shifts up to 10 nm (smoothing factors 5 and 21 are
indicated).
1.0 Absorbance
0.5
0.0
1st Deriative
0.01
0.00
–0.01
Figure 6.27 Principle of 1st derivative spectra.
and troughs of the original dye spectra become zero crossings of the x-axes in 1st deriva-
tive spectra. Furthermore, the peaks and troughs of the 1st derivative spectrum represent
the exact wavelength positions of the points of inflection of the dye spectrum (Figure 6.27).
In Figure 6.28 an example of two blue cotton fibres from different garments is presented,
showing the UV-vis spectra and their 1st derivative spectra. In the corresponding deriv-
ative spectra differences in wavelength positions of the troughs at 310 nm and 320 nm
respectively and points of inflection at about 290 nm and 550 nm/570 nm are shown.
Figure 6.29 shows the spectra and their 1st derivatives of another pair of blue cotton
fibres from different sources. These two spectra cannot be differentiated even when look-
ing at the derivative spectra. All the wavelength positions of peaks, troughs, shoulders and
points of inflection are exactly at the same wavelengths positions.
Generally, it is of great importance to analyse the full range of dye variability of a
representative sample. Otherwise false exclusions can occur which is proved by a study
of Wiggins et al. (2007). They found calculation of 1st derivative spectra as a useful aid in
comparison of spectra but they emphasized that a high degree of caution is required to
avoid false exclusions where a large intra-sample variation in colour occur.
6.1.7 Numerical Colour Coding

Chamberlin and Chamberlin (1980) have ascertained that two broad but different
approaches are used in defining and measuring colour:
100.0
90.0
80.0
70.0
60.0
%T
50.0
40.0
30.0
20.0
10.0
0.0
250.0 300.0 350.0 400.0 450.0 500.0 550.0 600.0 650.0 700.0 750.0
nm
0.0125
0.01
0.0075
0.005
0.0025
0.0
ABS/nm
–0.0025
–0.005
–0.0075
–0.01
–0.0125
–0.015
–0.0175
–0.02
250.0 300.0 350.0 400.0 450.0 500.0 550.0 600.0 650.0 700.0 750.0
nm
Figure 6.28 Example of two blue cotton fibres from different garments showing the UV-vis
spectra and their 1st derivative spectra.
100.0
90.0
80.0
70.0
60.0
50.0
%T
40.0
30.0
20.0
10.0
0.0
250 300 350 400 450 500 550 600 650 700 750
0.005
0.0025
0.0
–0.0025
–0.005
ABS/nm
–0.0075
–0.01
–0.0125
–0.015
–0.0175
250 300 350 400 450 500 550 600 650 700 750
Figure 6.29 Example of two blue cotton fibres from different garments showing the UV-vis
spectra and their 1st derivative spectra.
• Visual comparison with known physical standards of colours which are accepted
as references, e.g. the application of systematized and accurately reproducible
colour atlases.
• Instrumental measurement of the constituent parts of the colour in terms of the
relative amounts of each wavelength present. First, this gives an unequivocal fin-
gerprint of the colour. Second, to turn this into a visualizable description of what
the colour looks like, one has to find a way of relating these basic stimuli to the
colour image that would be produced by this stimulus in the brain of a hypotheti-
cal standard observer.
6.1.7.1 Visual Comparison Using Standards

Humankind has dealt with colour analysis for thousands of years. Aristotle defined a sys-
tem to describe colours just before the christian calender. Many others followed Aristotle
and defined more complex colour systems. Some of the most common are the Ostwald
system, the DIN system and the system according to Albert Henry Munsell (1858–1918)
who developed the basic principles of his colour model in 1905. In 1915, he published The
Munsell Atlas of Colour, displaying coloured specimens of a range of values and chromas
for ten hues. Each colour is described by hue, value and chroma. In his system Munsell
strived for visual equal distances for each colour. Although these systems have certain
differences, each of them formulates a colour space based on three variables: hue, chroma
or saturation and value or lightness. Thus, each colour in these systems is defined by three
values.
The Munsell concept of colour space illustrates that colour space is three-dimensional.
The structure is shown in Figure 6.30. The hues can be thought of as being arranged in a
circle. It is divided into ten principal hues: red (R), yellow-red (YR), yellow (Y), green-yellow
(GY), green (G), blue-green (BG), blue (B) purple-blue (PB), purple (P), and red-purple (RP)
The centre of the circle is occupied by the achromatic axis (vertical oriented) with black and
white at its ends. Chroma is the distance measured from the achromatic axis towards the
periphery. Value (lightness) is measured as a distance along the vertical scale from black
to white. One value step is visually equal to two chroma steps. The Munsell Book of Colour
now contains well over 1500 systematically ordered actual colour samples which are pre-
cisely defined. Munsell specifications, as well as the DIN system, which contains about 2000
colour samples, can also be converted into the basic CIE system of colour designation.
Value Hue
Chroma
YR
RP
Y
P
GY
PB
B G
BG
Figure 6.30 The Munsell concept of colour space.

While the samples in the Munsell Book of Colour, the DIN system, etc. are solid chips
dyed with pigments, another system, called the Scotdic® system, offers an atlas with speci-
fied colours where the samples are made from dyed pieces of a polyester or cotton fabric
(see Figure 6.32). The specification follows the rules of the Munsell system.
Colour atlases are mainly used to communicate more precise information about
colours in industry: about automotive paints, colours in the textile industry, etc. They are
a proven tool needed to overcome the problem of verbal descriptions of colour using com-
mon names. Common colour descriptions such as ‘raspberry red’, ‘sky blue’, ‘grass green’
and many others, used without a second thought every day in colloquial speech, only sim-
ulate exactness. In reality, if one were to ask four individuals to describe one of these com-
mon colour phrases, it is unlikely that their descriptions would agree closely.
The use of colour atlases has not yet played a role in colour description for forensic
examination of fibres. This way of describing the colour of an object is mainly applicable
to larger samples. Nevertheless, the colour of a single fibre can be more clearly defined by
using colour atlases as a guide. In addition, it is often necessary to record the macroscopic
colour of clothing and other textiles. The help offered by a colour book such as Scotdic®
makes this task easier and more objective. It requires only a quick visual comparison which
leads to a numerical description for the colour concerned. A forensic fibres examiner
should have some general knowledge about colour atlases and how they work.
6.1.7.2 Instrumental Measurements/the CIE System

The science of numerical colour coding is called colorimetry. Colorimetry is a part of
spectrophotometry. Besides their chromaticity, the numerical specification of colours
also defines their brightness or lightness (Kowaliski, 1973). The aim of colorimetry is to
establish colour order systems. A colour order system is a prerequisite for determining
and specifying colours. Colorimetry is based on a number of principles put forward by
Grassmann (1853).
• The human eye can distinguish only three dimensions of colour - hue, saturation,
intensity (according to CIE), or hue, chroma, value (according to Munsell).
• When one part of a two-part mixture is changed, the colour of the mixture changes.
• Separate lights that appear to be the same colour to the human eye will produce
identical effects in mixtures, whatever their spectral composition.
The colour of an object is known to depend on two factors, besides the characteristics of
the observed object itself: the kind of light used for illumination, the viewing conditions
and colour response of the observer. Therefore, before a mathematical description of colour
can be arrived at, the light sources as well as the viewing geometries and the eye response
factors have to be standardized. The first international agreement on the mathematical
treatment of colour data, so that there could be a common basis for calculations, came at
the 1931 Commission Internationale de L’Eclairage Conference held in Cambridge, UK.
Since then a number of illuminants have been defined from which Illuminant C (rep-
resentative of average daylight with an overcast sky and designated as having a colour tem-
perature of 6774 K) is the most commonly used in transmission work. Other illuminants
of interest in forensic fibres examination may be Illuminant A (incandescent light, 4874 K),
Illuminant B (noon sunlight, 2856 K) and Illuminant D65 representing daylight and des-
ignated as having a colour temperature of 6500 K. Illuminant D65 is replacing Illuminant
C more and more.
The matter of viewing geometry includes an agreement about the area of the field of
view, the angle of illumination and the angle of viewing. The area of the field viewed is
approximately equal to the size of one’s thumbnail as viewed at arm’s length. The light
should be falling at 45° and the sample viewed perpendicularly, or vice versa if preferred.
From these three standards, the angle of viewing is only variable in transmission work with
MSP. There are definitions for 2° and 10° observers, but to date the latter is mainly used.
The definition of a ‘standard observer’ to represent the human colour response has
been the most difficult and voluminous task in establishing colorimetry. It was accom-
plished by getting a number of people to review numerous shades of colour. They were
asked to reproduce or match a shade by means of mixing red, green and blue lights. As
reported above, on stimulation by light the human eye reacts as if it has three distinct
colour receptors for red, blue and green which have been given the symbols x, y and z
respectively. Enough information was collected in this way to define the values X, Y and Z,
known as the tristimulus values, and x, y and z, which are the chromaticity coordinates
(Graham, 1983). The chromaticity coordinates are calculated by normalizing the tristimu-
lus values, that is, dividing each of them by their sums, as follows:
X Z Y
x= z= y=
X +Y + Z X +Y + Z X +Y + Z
The sum of x, y and z is always 1. Subsequently only two coordinates are necessary for
colour specification. The x and y values were chosen for this purpose. They define hue and
chromaticity. The third dimension for specifying colour is the Y tristimulus value. It has
special significance, for it was established as a direct measure of luminance or lightness.
A plot of the chromaticity coordinates x and y for the spectral colours between 380 nm
and 700 nm forms a horseshoe-shaped spectrum locus known as a chromaticity diagram
(Figure 6.33). All real colours are located within the roughly triangular area of the diagram.
In this diagram the point N, called the neutral point, indicates the position of the
illuminant, which is source D65 in most transmittance work. The third-dimension Y axis
is perpendicular to the x-y plane at the neutral point (N) and has a scale of 0%–100%. It
could be thought of as the location of the achromatic range. Locations on a line between
N and a point on the periphery of the diagram represent shades of the same hue. These
shades differ only in their saturation. Saturation of a hue increases from the neutral point
to its spectral locus. The spectral colour has maximum saturation. The values from 2000
up to 25000 along the curved line represent the colour temperature of the black body radia-
tion. The measurement of colour data and the calculation of the CIE coordinates plays a
dominant role in forensic paint analysis and in the establishment of paint data collections
(Fourweather et al., 1976; Hudson et al., 1977). Measurement is carried out in the reflec-
tance mode. In forensic fibre examination there are other priorities.
CIE L*a*b* (CIELAB) is the most complete colour space specified by the International
Commission on Illumination. It describes all the colours visible to the human eye and was
created to serve as a device independent model to be used as a reference (Figure 6.31).
The three coordinates of CIELAB represent the lightness of the color: L = 0 indicates
black and L=100 yields white. +a indicates red while -a yields green and +b indicates yel-
low while -b yields blue. The asterisk (*) after L, a and b are part of the full name, since they
represent L*, a* and b*, to distinguish them from Hunter’s L, a and b. Because the red/green
and yellow/blue opponent channels are computed as differences of lightness transforma-
tions of (putative) cone responses, CIELAB is a chromatic value colour space.
White
Yellow
+b*
–a* L* +a*
Green Red
–b*
Blue
Black
Figure 6.31 CIE L*a*b* (CIELAB) colour space specified by the International Commission on
Illumination. Black: L* = 0, white: L* = 100, red: +a*, green: -a*, yellow: +b*, blue: -b*.
97
01
03
04
05
95
96
93
06
92
07
89
08
87
11
5RP 6.25RP 10RP 2.5R 3.75R 5R

P 2.5RP 7.5RP 6.25R 7.5R
1.25R
85
8.7
12
10P 5R
10
.5P R
83 7
5P
1.2
5Y
R
13
75 14
2.5
P
2.5
YR
Index
3.7
PB
15
5Y
74
10
R
PB
5YR
8.75
73 17
7.5YR
5PB
Achromatic colors
6.25PB 7.
8.75YR 10YR
72 N10 N15 N20 N25 N30 N35 N40 N45 N50 N55 N60 N65 N70 N75 N80 N85 N90 N95 18
5PB
69
N1 N1.5 N2 N2.5 N3 N3.5 N4 N4.5 N5 N5.5 N6 N6.5 N7 N7.5 N8 N8.5 N9 N9.5 19
2.5Y
3.75f
68
23
B
2.5P
5Y
67
25
PB
Cotton
6.2
5
5
1.2
66 26
7.5
B
10
B 10
Y
7.5
65
5B GY
2.5 27
2.5B Chromatic colors 54HUES 5GY
63
7.5GY
31
0BG
7.5BG 1 5BG 2.5BG 10G 7.5G 5G 2.5G 10GY 8.75GY
61
33
59
35
57
37
55
38
53
51
49
47
45
43
41
Figure 6.32 Index of the Scotdic® colour atlas. The different hues are dyed on pieces of cotton
garments.
y
520 525
515 530
0.8 535
510 540
545
0.7 550
505 555
560
0.6 565
570
500 Yellowgreen 575
0.5 580
Green
Yellow 585
495 3,000 2,000 590
0.4 4,000 595
Orange 600
Bluegreen + 6,000 610
N Red 620
0.3 490 10,000
15,000 650
25,000 700
0.2 485
Purple
480 ine
0.1
Blue
rp le l
Pu
470
460
440 380
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 x
Figure 6.33 Plot of the chromaticity coordinates x and y for the spectral colours between
380 nm and 760 nm forms a horseshoe-shaped spectrum locus known as a chromaticity diagram.
6.1.7.3 Complementary Chromaticity Coordinates (CCC)

In forensic examination of fibres, data collections always have been of interest in order to
try to improve the assessment of the meaning of examination results. Without doubt, it is
desirable to be able to store information on colour in such a data bank in such a way that
it is easily researchable. CIE tristimulus values and chromaticity coordinates vary with
changes in colourant concentration (Venkataraman, 1977). Tristimulus values decrease
nonlinearly with increasing concentration, and if the chromaticity values are plotted on
a chromaticity diagram they are found to move as a function of dilution in an arc that
reaches the neutral point at infinite dilution.
The nonlinear relationship between tristimulus values and colourant concentration
decreases the utility of conventional colorimetry in forensic fibre examination, where sin-
gle fibres may be measured in the transmittance mode. The calculation of the chromaticity
coordinates is based on % transmittance; they therefore depend on dye concentration. This
is a drawback when a number of measurements are required from a standard fibre sample,
because - as we know - intra-sample variation of the colour shade occurs in most textiles.
Operating in absorbance instead of transmittance, the complementary chromaticity
coordinates (CCC), x’ and y’, can be produced. By substituting absorbance for transmittance
the desired linearity is obtained, at least for solutions (Rounds, 1969) and assuming that the
colourant obeys Beer’s Law. However, measuring the colour of dyed fibres, as opposed to
solutions, gives rise to complications. Such factors as selective uptake from multicomponent
dyes and the inherent fibre colour can produce variations in the values of the coordinates
(Laing et al., 1986). Further investigation by Hartshorne and Laing (1987) into factors which
may cause variations in CCC values showed that presence or absence of delustrant and vary-
ing levels of delustrants can cause colour shifts between fibres dyed with the same colourant.
Hartshorne and Laing (1988) showed that variations in CCC values form a roughly
elliptical shape when plotted in the CIE diagram. This ellipse was called an error ellipse.
The size of the ellipse reflects the degree of variation in the sample; the larger the ellipse,
the greater is the variation. An extensive fibre database was established in England (Laing
et al., 1987) and, when studying colour-matching within the database, Hartshorne and
Laing found out that these ellipses required too much computer space; in 1988 they pro-
posed a simpler alternative method using squares.
6.1.8 Conclusions
Beside the morphology of fibres and the fibre substance, colour is the third characteristic
which dominates forensic fibre examination. The human eye is very sensitive to colour
differences and is therefore used as a screening tool in the initial phase of fibre examina-
tion, where similar fibre groups must be recognized, but it cannot detect whether differ-
ences in the colour shade of individual fibres are due to different dye uptake or to different
colourants. This deficiency is covered by microspectrophotometers. Therefore, MSP is the
pivotal method in colour and dye examination of individual fibres.
Microspectrophotometry belongs to the spectroscopic methods and covers the UV-vis
region of the electromagnetic radiation spectrum. The absorption spectra gained primarily
provide information about conjugated π-electron systems in a molecule. Additionally, the visi-
ble part of a spectrum is linked with the phenomenon of colour and is the basis for colorimetry.
Concerning fibres, microspectrophotometry fulfils economic demands. The method
does not need difficult preparation, it is quick and non-destructive and its discriminating
power is very high. It has been established for more than 50 years in forensic fibres exami-
nation and is accepted as being complementary to TLC.
Modern PDA or CCD multichannel MSP covering the vis region are currently a very
valuable screening tool in the very first step of single fibre examination and dramatically
reduce the amount of fibres which have to be analysed in the UV region. But it must be
clearly stated, if a microspectrophotometer will operate only in the vis region, the assess-
ment of the chemical identity of the colourants of two individiual fibres is limited. In this
case an additional method such as TLC is required or the statement of the strength of the
evidence concerning the likelihood of a common origin of two fibres must be restricted.
Except indigo, sulphur black 1 and a few other dyes microspectrophotometry cannot
identify colourants. One aim of the research work during the last decade was to establish
a method for identifying dyes mainly by liquid chromatography (LC) and liquid chroma-
tography mass spectrometry (LC-MS). Although there are promising results (Roux et al.,
2007) LC-MS is not established as a routine method in forensic fibre analysis until today. In
the opinion of the author besides expensive costs a main reason are difficulties to establish
reference data collection for that kind of LC-MS spectra.
Acknowledgements
In memory of my friends Michael Grieve and Kenneth Wiggins. I wish to cordially

thank James Robertson and Claude Roux for the invitation to completely revise the
chapter of microspectrophotometry which was basically written by Franz-Peter Adolf

and James Dunlop. Thank you very much Stefan for measuring spectra with the new-
est generation of CCD multichannel MSP. Last but not least, thanks to Georg Jochem
and Steffen Dillinger my former colleagues and successors at the Textile Section of the
Bundeskriminalamt.
References
Adolf, F. P., 1986. Microscope photometry and its application in forensic science—The use for the
examination of transparent objects. In Proceedings of the 6th Meeting of the Scandinavian
Forensic Science Laboratories, Helsinki, Finland.
Adolf, F. P., 1987. Experiences in the application of microspectrofluorimetry in forensic fibre exami-
nation. In Proceedings of the 11th IAFS Meeting, Vancouver, Canada.
Adolf, F. P., 1997. Photodiode array detectors—Their possible use as standard equipment for fibre
examination. In Proceedings of the 5th European Fibres Group Meeting (pp. 82–89), Berlin,
Germany.
Adolf, F. P., 1998. Fibre examination—A critical view of the present state and new developments.
In Proceedings of the International Workshop on the Forensic Examination of Trace Evidence,
Tokyo, Japan.
Amsler, H., 1959. Die Mikro-Spektralphotometrie, ein wichtiges Hilfsmittel fur den Farbvergleich
kleinster corpora delicti, Arch. Krim., 124, 85–94.
Biermann, T. W., 2007. Blocks of colour IV: The evidential value of blue and red cotton fibres, Sci.
Justice, 47(1), 68–87.
Bresee, R. R., 1987. Evaluation of textile fiber evidence: A review. J. Forensics Sci., 32, 510–521.
Bruck, H. J. and Rohm, E., 1975. Mikrospektralphotometrische Absorptionsmessungen an
Textilfaserspuren. In H. Freund (Ed.), Handbuch der Mikroskopie in der Technik, Bd VII,
Mikroskopie in der chemischen Technik (pp. 177–207), Frankfurt, Germany: Umschau Verlag.
Casperson, T., 1936. Uber den chemischen Aufbau der Strukturen des Zellkerns, Skand, Arch.
Physiol., 73, 1–149.
Casperson, T., 1940. Methods for the determination of the absorption spectra of cell structures,
Trans. R. Microsc. Soc., 60, 8–25.
Cassista, A. and Peters, A. D., 1997. Survey of red, green and blue cotton fibres, Can. Soc. Forensic
Sci. J., 30, 225–231.
Chamberlin, G. J. and Chamberlin, D. G., 1980. Colour—Its Measurement, Computation and
Application. London: Heyden & Son.
Coyle, T., 2002. The use of first derivate data as an aid to spectral comparison. In Proceedings of the
10th European Fibres Group Meeting (pp. 93–97), Paris, France.
De Wael, K., 2012. Dichroism measurements in forensic fibre examination Part 4—Dyed acrylic and
acetate fibres, Sci. Justice, 52, 81–89.
De Wael, K. and Lepot, L., 2011. Dichroism measurements in forensic fibre examination Part 3—
Dyed cotton and viscose fibres, Sci. Justice, 51, 173–186.
De Wael, K. and Lepot, L., 2012. Dichroism measurements in forensic fibre examination Part 5—
Pigmented fibres, Sci. Justice, 52, 161–167.
De Wael, K. and Vanden Driessche, T., 2011a. Dichroism measurements in forensic fibre examina-
tion Part 1—Dyed polyester fibres, Sci. Justice, 51, 57–67.
De Wael, K. and Vanden Driessche, T., 2011b. Dichroism measurements in forensic fibre examina-
tion Part 2—Dyed polyamide, wool and silk fibres, Sci. Justice, 51, 163–172.
De Wael, K., VanDijck, K. and Gason, F., 2015. Discrimination of reactively-dyed cotton fibres with
thin layer chromtography an UV microspectrophotometry, Sci. Justice, 55, 422–430.
Fourweather, C., May, R. W., and Porter, J., 1976. The application of a standard colour coding system
to paint in forensic science, J. Forensic Sci., 21, 629–635.
Gauglitz, G., 1994. Ultraviolet and visible spectroscopy. In Ullmann’s Encyclopedia of Industrial
Chemistry, Tool, Analytical Methods I (Vol. B 5, pp. 383–428). Weinheim, Germany: VCH
Verlagsgesellschaft.
Gerlach, D. 1976. Das Lichtmikroskop—Eine Einführung in die Funktion, Handhabung und
Spezialverfahren für Mediziner und Biologen. Stuttgart: Georg Thieme Verlag.
Gibson, E. P., 1977. Review: Application of luminescence in forensic science. J. Forensic Sci., 22,
680–696.
Graham, L. A., 1983. Color order systems, color specification and universal colour language. In
G. Celikiz and R. G. Kuehni (Eds.), Colour Technology in the Textile Industry (pp. 35–152).
Research Triangle Park, NC: AATCC.
Grassmann, H., 1853. Zur Theorie der Farbmischung, Ann. Phys. Chem., 89, 69–84.
Grieve, M. C., Biermann, T. W. and Davignon, M., 2003. The occurrence and individuality of orange
and green cotton fibres, Sci. Justice, 43(1), 5–22.
Grieve, M. C., Biermann, T. W., and Davignon, M., 2001. The evidential value of black cotton fibres,
Sci. Justice, 41(4), 245–260.
Grieve, M. C., Biermann, T. W., and Schaub, K., 2005. The individulaity of fibres used to providse
forensic evidence—Not all polyesters are the same, Sci. Justice, 45(1), 13–28.
Grieve, M. C. and Deck, S., 1995. A new mounting medium for the forensic microscopy of textile
fibres, Sci. Justice, 35(2), 109–112.
Grieve, M. C., Dunlop, J., and Haddock, P. S., 1988. An assessment of the value of blue, red and black
cotton fibres as target fibres in forensic science investigations. J. Forensic Sci., 33, 1332–1334.
Halonbrenner, R., 1976. Mikrospektralphotometrische Untersuchungen an Textilfasern, Arch. Krim.,
157, 93–106.
Halonbrenner, R. and Meier, J., 1973. Mikrospektralphotometrische Untersuchungen an Textilfasern,
Kriminalistik, 27, 344–350.
Hartshorne, A. W. and Laing, D. K., 1987. The definition of colour for single textile fibres by micro-
spectrophotometry, Forensic Sci. Int., 34, 107–129.
Hartshorne, A. W. and Laing, D. K., 1988. Colour matching within a fibre data collection, Forensic
Sci. Int., 33, 1345–1354.
Hartshorne, A. W. and Laing, D. K., 1991a. An absorption standard for microspectrophotometry:
Results of a collaborative exercise, Forensic Sci. Int., 51, 263–272.
Hartshorne, W. and Laing, D. K., 1991b. Microspectrofluorimetry of fluorescent dyes and brighteners
on single textile fibres: Part 1—Fluorescence emission spectra, Forensic Sci. Int., 51, 203–220.
Hartshorne, W. and Laing, D. K., 1991c. Microspectrofluorimetry of fluorescent dyes and brighteners
on single textile fibres: Part 2—Colour measurement, Forensic Sci. Int., 51, 221–237.
Hartshorne, W. and Laing, D. K., 1991d. Microspectrofluorimetry of fluorescent dyes and brighteners
on single textile fibres: Part 3—Fluorescence decay phenomena, Forensic Sci. Int., 51, 239–250.
Houck, M., 1997. Measuring dichroism in fibres by use of the microspectrophotometer. In Proceedings
of the 5th Meeting of the European Fibres Group, Berlin, Germany.
Hudson, G. D., Andahl, R. O., and Butcher, S. J., 1977. The paint index—The colour classification and
use of a collection of paint samples taken from scenes of crime, J. Forensic Sci. Soc., 17, 27–32.
Jenne, K., 1981. Zur Differenzierung schwarzer Wollfasern—Mikroskopische, mikrospektralpho-
tometrische und dunnschichtchromatographische Untersuchungen, Arch. Krim., 168, 17–22.
Jochem, G., 2007. Addition of a CCD Spectrometer to a Zeiss MPM800 Microspctrophotometer. In
Proceedings of the 15th European Fibres Group Meeting, Madrid, Spain.
Kowaliski, P., 1973. The role of luminance in colorimetry, Colour 73—Survey lectures. In Abstracts
of the Papers Presented at the 2nd Congress of the ICA (pp. 298–299, B116), University of York,
London: Adam Hilger.
Kubic, T. A., King, J. E., and Dubey, I. S., 1983. Forensic analysis of colorless textile fibres by fluores-
cence microscopy, Microscope, 31, 213–222.
Laing, D. K., Hartshorne, A. W., Cook, R., and Robinson, G., 1987. A fiber data collection for forensic
scientists—Collection and examination methods, J. Forensic Sci., 32, 364–369.
Laing, D. K., Hartshorne, A. W., and Harwood, R. J., 1986. Colour measurements on single textile
fibres, Forensic Sci. Int., 30, 65–77.
Macrae, R., Dudley, R. J., and Smalldon, K. W., 1979. The characterization of dyestuffs on wool fibres
with special reference to microspectrophotometry, J. Forensic Sci., 24, 117–129.
Pabst, H., 1980. Anwendung der Mikrospektralphotometrie in der Kriminaltechnik, Microsc. Acta.,
83 (Suppl. 4), 189–193.
Palmer, R., Hutchinson, W., and Fryer, V., 2009. The discrimination of (non-denim) blue cotton, Sci.
Justice, 49(1), 12–18.
Piller, H., 1977. Microscope Photometry. New York, NY: Springer-Verlag.
Piller, H., 1979. Domains of microscope photometry in materials science, J. Microsc., 116, 295–310.
Robson, R., 1997. Spectral variation within red cotton dyes. In Proceedings of the 5th European Fibres
Group Meeting (pp. 66–74), Berlin, Germany.
Rounds, R. L., 1969. A colour system for absorption spectroscopy, Text. Chem. Colour., 1, 297–300.
Roux, C., Austin, C., Benett, S., Beavis, A., Dawson, M., Doble, P., Lennard, C., and Robertson,
J., 2007. Ultra Performance Liquid Chromatography an Emerging Technique for the Forensic
Analysis of Trace Evidence, Global Forensic Sci. Today, 1, 29–37.
Savitzky, A. and Golay, M. J. E., 1964. Smoothing and differentiation of data by simplified least square
procedures, Anal. Chem., 36, 1627–1639.
Siegel, J. A., 1996. Application of fluorescence spectroscopy to forensic science, Forensic Sci. Rev., 8,
2–11.
Suchenwirth, H. and Bruck, H. J., 1968. Uber den Aussagewert von mikrospektralphotometrischen
Messungen an Textilfaserspuren, Arch. Krim., 142, 16–25, 111–120.
Suzuki, S., Marumo, Y., and Adolf, F. P., 1994. Microspectrophotometric discrimination of black
single fibres using absorbance profiles of the ultra violet and visible region, Rep. Nat. Inst. Police
Sci., 47(2), 39–45, Tokyo, Japan.
Venkataraman, K. (Ed.), 1977. The Analytical Chemistry of Synthetic Dyes. London and New York,
NY: Wiley.
Waring, D. R. and Hallas, G. (Eds.), 1990. The Chemistry and Application of Dyes. New York, NY and
London: Plenum Press.
Was, J., 1997. Identification of thermally changed fibres, Forensic Sci. Int., 85, 51–63.
Was-Gubala, J. and Grzesiak, E., 2010. The kinetics of colour change in textiles and fibres treated
with detergent solutions: Part II—Spectrophotometric measurements, Sci. Justice, 50, 55–58.
Was-Gubala, J. and Starczak, R. 2015. UV-Vis microspectrophotometry as a method of differentia-
tion between cotton fibre evidence coloured with reactive dyes, Spectrochim. Acta Mol. Biomol.
Spectrosc., 142, 118–125.
Weigel, O. and Habich, G., 1927. Uber Mineralfarbungen. 1. Die Absorption rot gefarbter Mineralien
im sichtbaren Teil des Spektrums, N. Jahrb. Mineral., Beilagenband 57 A, 1.
Weigel, O. and Ufer, H., 1927. Uber Mineralfarbungen. 11. Die Absorption einiger rot gefarbter
Mineralien und kunstlicher Praparate im sichtbaren und ultravioletten Teil des Spektrums, N.
Jahrb. Mineral., Beilagenband 57, A, 397–500.
Welsch, N. and Liebmann, C. Chr., 2003. Farben—Natur, Technik, Kunst. Berlin: Spektrum
Akademischer Verlag GmbH Heidelber.
Wiggins, K. G., Crabtree, S. R., Adolf, F. P., and Grieve, M. C., 1995. The importance of analysis of
reactive dyes on cotton fibres, Crime Lab. Dig., 22, 89.
Wiggins, K., Davis, E., Schennum, C., and Drummond, P., 2008. An investigation into fibre variation
across garments, Global Forensic Sci. Today, 6, 2–16.
Wiggins, K. and Drummond, P., 2007. Identifying a suitable mounting medium for use in forensic
fibre examination, Sci. Justice, 47(1), 2–8.
Wiggins, K. and Holness, J.-A., 2005. A further study of dye batch variation in textile and carpet
Wiggins, K., Palmer, R., Hutchinson, W., and Drummond, P., 2007. An investigation into the use
of calculating the first derivative of absorbance spectra as a tool for forensic fibre analysis, Sci.
Justice, 47(1), 9–18.
Williams, D. H. and Fleming, I., 1966. Spectroscopic Methods in Organic Chemistry. Maidenhead:
McGraw-Hill.
6.2 Fibre Dyes Analysis
KENNETH G WIGGINS
6.2.1 Introduction
Fibres encountered are both synthetic and natural and although they are originally
opaque, colourants are added to make them commercially useful. The colourants may
be a single component or they may be mixtures of dyes. Thin-layer chromatography is
one analytical technique that allows the constituent components of dyes to be separated
and hence allows the forensic scientist to compare dyes extracted from different textile
fibres.
Synthetic colourants used for the dyeing of fibres are organic, and the colour of the
dye is related to its chemical structure. Ethanol, β-napthol and Acid red 88 are organic
compounds, but only Acid red 88 is capable of behaving as a dye. Acid red 88 absorbs
radiation not only in the ultraviolet region, as in the case of ethanol and β-napthol, but
also in the visible region of the electromagnetic spectrum. This gives both a UV and a
visible spectrum with maximum absorbance in the visible region, and therefore appears
coloured. In order for visible radiation to be absorbed, a compound must contain at least
one chromophore. Characterized by areas of high electron density, a chromophore is the
part of a molecule that absorbs light. Most dyes also contain an auxochrome, which is a
functional group that, on addition to a molecule, brings about a change in colour intensity.
Auxochromic groups include –COOH, –SO3H, –N(CH3)2 and –NH2, which can influence
dye solubility and hence its ability to bond to a fibre.
6.2.2 Classification of Fibre Dyes

Dyes are classified generally using their method of application or chemical class, but
occasionally by the type of fibre to which dyes are applied. Other classifications are avail-
able but rarely used. The method of application of the dye and the fibre type to which it
is applied are influenced by the relative solubility of the dye in water. The Colour Index
(Society of Dyers and Colourists, 1985) is an extensive reference work listing the chemical
class of all dyes. Updates are available periodically. An example of how a dye is classified
is given below. The structure of a typical dye with colour index number CI 14780 is shown
in Figure 6.34.
6.2.2.1 Colour Index Number: CI 14780

A colour index number (CI) is given to all dyes and consists of five digits. The dye above is
an azothiazole dye. Acidic dyes which are produced as different salts are given identical CI
numbers but each is given a different suffix, e.g. 1, 2, etc.
NaO3S
S HO
H3C
C N N
N
NaO3S
Figure 6.34 Structure of a typical dye (CI 14780).
6.2.2.2 Generic Name: CI Direct Red 45

The dye class is described by the generic name, which also gives an indication of its use.
When dyes have more than one generic name it is an indication that there are additional
applications. The other generic name in this example is CI Food Red 13.
6.2.2.3 Commercial Name: Thiazine

The commercial name generally originates from the manufacturer. If a particular dye is
produced by more than one dye house, different commercial names may exist.
6.2.3 Chemical Classification

As already stated, there are many different types of dye. These dyes can be classified
either by their chemical structure or according to their method of application. This
section will deal with the dyes specifically encountered in the forensic examination of
fibres, and the classification will be according to their method of application. General
information on dyes can be found in Venkataraman (1977) and Waring and Hallas
(1990).
6.2.3.1 Acid Dyes

Acid dyes are generally applied under acidic conditions to fibres such as polyamide, wool,
silk and occasionally polypropylene. When acid dyes are applied to any of the above fibre
types, ionic bonds or salt linkages result between the dye molecules and the polymer. The
dye site is the point on the fibre polymer where the dye becomes attached. In wool there are
many amino groups which act as dye sites, whereas in nylon it is a terminal amino group.
During dyeing the amino group becomes positively charged and attracts the dye anion
which has a negative charge. Because wool fibres have many amino groups and the fibre is
very amorphous, dye penetration is high, hence dark shades are achievable. Silk has fewer
amino groups, so dark shades can be difficult to achieve. The more crystalline structure of
polyamide compared to wool and silk and the low level of amino groups in them means
that acid dyes cannot be used to produce dark shades in these fibres. Hydrogen bonds and
van der Waals’ forces are also formed between the acid dyes and the fibre structure, but the
retention of the dye is governed by the ionic bonds.
6.2.3.2 Azoic Dyes

Azoic dyes can be applied to viscose and cotton, i.e. cellulosics, but on the rare occasions
when they are seen in forensic fibre examinations it is usually on cotton fibres. Azoic dyes
consist of a coupling component, in the form of a napthol, and the base or diazo compo-
nent. The first stage of colouration is to impregnate the fibre/fabric with a naphthol solu-
tion. The diazo component is usually a stabilized diazonium salt or ‘Fast Salt’ which can be
dissolved in water. The naphthol-treated fibre/fabric can then be passed through the ‘Fast
Salt’ solution to effect a reaction between the salt and the naphthol which results in one
larger insoluble molecule. This is the coupling reaction.
6.2.3.3 Basic Dyes

Basic dyes are applied under acidic conditions to polyacrylonitrile, modified acrylic and
occasionally to polyester and polypropylene fibres. These dyes, which are ammonium, sul-
phonium or oxonium salts, have glacial acetic acid added to improve solubility. They ionize
in solution with the coloured component of the dye being a cation (hence their alternative
name, ‘cationic dyes’). The negatively charged fibre surface attracts the dye cation, which
results in the fibre being neutralized. Raising the temperature of the dye bath enables the
dye to enter the fibre. Cationic retarders are added to the dye bath to prevent uneven dye-
ing. This is achieved by the retarders initially taking up some of the dye sites on the fibre.
As the dye has a greater substantivity for the fibre compared with the retarder, it will slowly
take over the sites and ensure an even dyeing.
6.2.3.4 Direct Dyes

Direct dyes are applied directly to cellulosic fibres from an aqueous medium containing an
electrolye, e.g. sodium chloride. The positively charged sodium ion is attracted to the nega-
tively charged surface of the fibre, neutralizing the surface, enabling the dye anion to enter
the fibre. Heating swells the fibre, increases the energy of the dye solution components and
ultimately increases the dyeing rate.
6.2.3.5 Disperse Dyes

Disperse dyes are applied from an aqueous dispersion to polyester and acetate fibres.
Occasionally they are encountered in polyacrylonitrile, polyamide and polypropylene
fibres. The aqueous dispersion consists of dye, water and a surface active agent. Heating
allows the fibre to swell and assists in the dye’s penetration of the fibre. Hydrogen bonds
and van der Waals’ forces hold the dye molecules in the fibre. Carriers and high-temper-
ature dyeing (100°C to 130°C) enable moderately dark shades to be achieved. These are
normally restricted due to the fibres being hydrophobic and highly crystalline.
6.2.3.6 Metallized Dyes

Metallized dyes are encountered on wool fibres and very occasionally on polypropylene.
Three methods can be used to form metal complexes in fibres.
• Chrome mordant method: The mordant, i.e. chrome, is applied to the fibre from an
aqueous acidic medium (containing a dichromate) which is brought to the boil.
The fibre is transferred to a bath containing a chrome dye in an acid medium.
With heat, the dye molecule is fixed to the chrome in the fibre and a complex is
formed.
• Metachrome method: The dye and the mordant are applied to the fibre simultane-
ously. If certain dyes are used and the time interval between addition of dye and
mordant is not correctly controlled, the dye complex may be formed in the dye
bath rather than in the fibre.
• After chrome method: Acid dyes are applied to the fibre and the temperature is
slowly raised to the boil. After keeping it at this temperature for approximately
one hour, the mordant is added. After another hour the dye complex is formed in
the fibre.
Another type of dye that falls under the heading of metal-complex dyes is premetal-
lized. The metal is incorporated in the dye molecules during the dye’s manufacture.
2:1 premetallized dyes are easier to apply than 1:1 premetallized dyes and have become
more popular. The chromophore is anionic and is attracted to the amino groups of wool
fibres. The ionic link and van der Waals’ forces ensure that the dye is retained in the
fibre.
These metal complex dyes are encountered only infrequently in forensic fibre exami-
nation. It is thought that this is probably due to the health and safety problems associated
with the disposal of effluent. Although methods are available for recovery of the mordants,
the cost is high.
6.2.3.7 Reactive Dyes

Reactive dyes react chemically with the fibre, forming covalent bonds with the functional
groups during the dyeing process. They are regularly used to dye wool and cellulosics and
occasionally to dye polyamide fibres. Reactive dyes are very similar in structure to acid
dyes but with the addition of a nucleophilic group. The method of application is similar to
that used for direct dyes, but application is carried out in an alkaline medium for cellulos-
ics and an acidic medium for wool.
6.2.3.8 Sulphur Dyes

Sulphur dyes are used to dye cellulosics and are so called because they contain sulphur
atoms in their molecules. Sulphur dyes must be reduced with sodium sulphide or sodium
hydrosulphite, normally in an alkaline medium, to produce the leuco form of the dye. Heat
is applied to the dye liquor to aid dye penetration and to speed up the process. Once the
dye has penetrated the fibre, the leuco form is oxidized to its original insoluble form. The
resulting colours are generally dark blue, black, mustard yellow and olive green.
6.2.3.9 Vat Dyes

Vat dyes are rarely encountered in forensic fibre examinations other than in denim gar-
ments and overalls. Their only use is in the dyeing of cellulosics. They have good colour
fastness and the name originates from the large wooden vats used to apply the dyes. The
process of dyeing is detailed; it involves dispersion, production of a leuco form, application
of the dye to the fibre, oxidation of the leuco form and removal of insoluble dye. As the dye
from denim is rarely examined, these dyes are unusual in a forensic context and the appli-
cation will not be discussed in detail.
6.2.3.10 Pigments
Pigments are not dyes but are seen as colourants in a variety of fibre types including poly-
propylene, viscose, acrylic, polyamide and polyester. These pigments have no affinity for
fibres and are generally added in the melt, and are therefore not subsequently extractable.
Pigments can also be bonded onto fibres or fabrics with a bonding agent.
6.2.3.11 Ingrain
Ingrain dyes are also very rarely encountered. They generally bear the Alcian or Phthalogen
trade mark and are particularly noted for their turquoise colour. The colour is unique and
results from a copper phthalocyanine chromophore.
6.2.4 Fibre/Dye Combinations

Variation in the chemical structure of both natural and man-made fibres means that
some fibre/dye combinations are more usual than others. Table 6.2 shows the relationship
between fibre type and the dye classes that tend to bind to them. Table 6.3 shows the fibre/
dye combinations encountered in the Metropolitan Laboratory of the Forensic Science
Service, March 1993 to April 1994.
It can be seen from Tables 6.2 and 6.3 that expected and actual fibre/dye combina-
tions are very similar. Acid and reactive dyes are most commonly encountered with wool.
Reactive dyes have increased at the expense of metallized dyes in the past few years.
Polyamide fibres have a terminal amino group which makes them amenable to acid dyes.
The ease of this process and their consequent dominance is supported by the data in
Table 6.2 Colour Absorbed versus Colour Observed by the Eye

Colour Absorbed Colour Observed, by Eye Wavelength (nm)
Violet Yellow-green 380–430
Blue Yellow 430–480
Green-blue Orange 480–490
Blue-green Red 490–500
Green Purple 500–560
Yellow-green Violet 560–580
Yellow Blue 580–590
Orange Green-blue 590–610
Red Blue-green 610–750
Table 6.3 Textile Fibre Type and Associated Class of Dye

Dye Class Fibre Type
Acid Wool, silk, polyamide, protein, polyacrylonitrile,
polypropylene
Basic Polyacrylonitrile, modified acrylic, polyester,
polyamide
Direct Cotton, viscose
Disperse Polyester, polyacrylonitrile, polyamide,
polypropylene, acetate/triacetate
Reactive Cotton, wool, polyamide
Sulphur Cotton
Vat Cotton
Metallized Wool, polypropylene
Azoic Cotton, viscose
Ingrain Cotton
Table 6.3. As polyester is difficult to colour, disperse dyes are the most frequently encoun-
tered dye with this fibre type. Basic dyes are dominant with acrylic fibres. Cotton fibres
show more variety. Reactive, sulphur and direct dyes are all popular in cotton, but azoics
also occur occasionally. It can be seen that many dyes are available for the various fibre
types commonly encountered in forensic textile fibre examination. As the dyes are pro-
duced and applied in various ways, both the extraction procedure and the separation tech-
nique that will ultimately be employed must be carefully considered.
6.2.5 Dye Classification and Extraction

Feeman (1970) compared the so-called classical methods of dye identification with the
then more modern methods. At that time, the best known schemes for dye identification
were based on determining the application class (acid, direct, basic, etc.) and generic struc-
ture (azo, anthraquinone, etc.). They involved fibre identification followed by chemical
tests. However, although the schemes were easy to follow and gave good results when the
dye was homogeneous, they failed when mixtures of dyes were encountered and they did
not identify the dye chemically, even when it was homogeneous.
Feeman cited the following as the newer methods of dye identification: chromatog-
raphy and electrophoresis for dye separation and classification, and infrared and ultra-
violet spectroscopy for dye identification. He emphasized that it was possible to identify
many commonly found dyes without knowing their chemical structures, because IR
and UV spectra acted as fingerprints unique to each dye. This did imply, of course,
the need for analysts to have reference spectra of standard dyes to compare with the
‘unknowns.’
Figure 6.35 shows Feeman’s scheme for dye identification. He also listed combinations
of solvents for extracting particular dye classes from certain fibre types, and elution sys-
tems for thin layer chromatography separation of extracted dyes. He himself was working
Dye Dye
(solution, paste or solid) (on fibre or fabric)
Solution or Solvent
extraction extraction
Dye solution
Separation
(chromatography)
Pure dye
Infrared
spectrophotometry
Dye identification
Figure 6.35 New scheme for dye identification. (From Feeman, J. F., Can Text J, 87, 83–89, 1970.)
with large non-forensic samples, but his work has formed the basis of many of the systems
developed for use in textile fibre dye analysis in forensic science laboratories.
Dye extraction/classification schemes have been developed which generally allow sin-
gle fibres to be sequentially extracted with a range of solvents. These schemes not only
determine the most efficient method for dye extraction but also allow the dye present to be
classified with a degree of caution. Extraction and classification studies have been carried
out on the dyes for wool, polyamide, polyacrylonitrile, polyester, cellulosics, polypropylene
and acetate fibres (Figure 6.36).
Macrae and Smalldon (1979) developed a three-part procedure for the major dye
classes found on wool, i.e. acid, metallized and reactive. Single fibres were extracted at
room temperature or at 90°C with pyridine/water 4:3 v/v. Good extraction indicated an
acid dye being present. No extraction meant removing the pyridine/solution and then a pre-
treatment with 2% oxalic acid for 20 minutes. After this time the oxalic acid was removed
and the fibre dried before an attempt was made to extract the dye with pyridine/water solu-
tion. If the extraction improved then the dye was classified as chrome or metal complex.
A reactive dye was indicated if no extraction was observed. Schemes for the extraction and
classification of dyes on single nylon, acrylic and polyester fibres (Beattie et al., 1979), cel-
lulose acetate fibres (Beattie et al., 1981a) and polypropylene fibres (Hartshorne and Laing,
1984) followed. Much later, a scheme for cellulosic fibres was developed (Laing et al., 1991).
Resua (1980) introduced thin-layer chromatography into the classification schemes.
He used chloroform–methanol–acetic acid (70:20:10) as a screening solvent system which
indicated the dye class and the best solvent system to be used for TLC. Fong (1989) criti-
cized the work for the large samples used in the tests, but Resua’s work helped establish the
systematic approach to dye classification/extraction. Resua et al. (1981) also investigated
the use of uncorrelated paired solvent systems.
Without doubt, the most comprehensive and widely used schemes for the classification/
extraction of fibre dyes encountered in forensic science are those used by the Forensic
Science Service in England, Figures 6.37 through 6.41.
The fibre type can be identified using bright-field and polarized microscopy and/or
infrared spectroscopy. Once the fibre type is known, the sequential classification schemes
can be followed. The schemes involve both solvent extraction and chromatography; gener-
ally only a single fibre is necessary. However, a second fibre is necessary to complete the
cotton and viscose classification. Should it be necessary to classify very pale fibres, a small
tuft of fibre would be needed. In casework, dye classification procedures will generally be
Stage
1 Pyridine/water (4:3) 100°C 10 min
Good extraction ACID DYE
Little or no extraction Go to stage 2

2 2% aqueous oxalic acid 100°C 20 min
then pyridine/water (4:3) 100°C 10 min
Improved extraction METALLIZED DYE

Little or no extraction REACTIVE DYE
Figure 6.36 Classification of dyes from wool fibres.

Stage
1 Glacial acetic acid 100°C 20 min
Good extraction AZOIC DYE

Good extraction DIRECT DYE

3 Dithionite/polyvinylpyrrolidone* 100°C 20 min Check fibre colour.
Extract applied to TLC plate. Check colour of spot.
Fibre colour changed REACTIVE DYE
No coloured spot or spot not original fibre colour
Fibre colour unchanged INGRAIN DYE
No coloured spot or spot not original fibre colour
Fibre colour changed Go to stage 4

Original coloured spot
4 With new fibre,
10%–14% sodium hypochlorite 100°C 10 min
Fibre colour changed SULPHUR DYE
Fibre colour unchanged VAT DYE

*80 mg sodium dithionite, 30 mg polyvinylpyrrolidone, 450 μL 10% sodium
hydroxide, 9 mL water; use immediately and discard any excess.
Figure 6.37 Classification of dyes from cotton and viscose fibres.
Stage
1 Formic acid/water (1:1) 100°C 20 min
Good extraction Go to stage 2
2 TLC procedure–methyl acetate eluent
movement DISPERSE DYE
no movement Go to stage 3
3 TLC procedure –methanol eluent
Sharp line at solvent front ACID DYE
Little or no movement or smeared BASIC DYE
Figure 6.38 Classification of dyes from polyacrylonitrile fibres.
carried out on the control items. Once the dye class and best extraction procedure have
been established for the control fibres, thin-layer chromatography can be used to com-
pare other control fibres with recovered fibres. If sufficient control material is available,
as is normal, different fibres can be used for each part of the classification. This prevents
Stage
1 Chlorobenzene 150°C 15 min
Good extraction DISPERSE DYE

Good extraction Go to stage 3
Little or no extraction REACTIVE OR DIAZO DYE

3 TLC procedure – methanol eluent
Sharp line at solvent front ACID DYE
Little or no movement or smeared BASIC DYE
Figure 6.39 Classification of dyes from polyamide fibres.
Stage
1 Chlorobenzene 130°C 10 min

2 Dimethyl formamide/formic acid (1:1) 100°C 20 min
Good extraction BASIC DYE
Figure 6.40 Classification of dyes from polyester fibres.
Stage
1 Methyl acetate/water/acetic acid (5:5:1) 100°C 20 min

Some extraction Go to stage 3
No extraction PIGMENT
3 2% aqueous oxalic acid 100°C 20 min
then pyridine/water (4:3) 100°C 20 min
Improved extraction METALLIZED DYE
No improvement ACID DYE
Figure 6.41 Classification of dyes from polypropylene fibres.
washing and drying of the fibre and hence potential loss. It also speeds up the procedure.
It is best practice to use one long control fibre cut into a number of smaller pieces to carry
out classification, always remembering to use one piece as a blank. This piece is placed in a
glass tube with water and heated in the same way as each test piece, allowing an easy com-
parison to see if extraction has occurred. The glass tubes used are about 2.5 cm in length,
with an internal diameter of 1.5 mm, sealed at one end. A fine wire is used to push the fibre
down the tube and approximately 10 μl of water or solvent (sufficient to cover the fibre) is
added using a glass pipette or syringe. The tube is then heat-sealed to avoid evaporation,
and incubated for the relevant time and temperature in an oven.
Care is taken to classify a dye only as being equivalent to, or acting as, a particular
dye class. For example, if a wool fibre extracts in pyridine/water 4:3 v/v after 10 minutes
at 100°C, this is consistent with it being dyed with an acid dye. This covers the eventuality
of any new dyes being produced that may not conform to the present scheme. For many
years acid, direct, metallized, disperse, basic and azoic dyes were considered as the only
dye classes which are extractable and are therefore amenable to TLC. The other major
classes – reactive, vat, ingrain and sulphur – were considered not to extract. Home and
Dudley (1981) reported that a 1.5% aqueous solution of sodium hydroxide would extract
the majority of reactive dyes encountered on cellulosics. In the author’s experience, suc-
cess is unusual, and those that do extract are often changed in colour. In their dye extrac-
tion and classification schemes, Laing et al. (1991) identified reactive dyes on cotton by an
exclusionary procedure. Organic solvents could not extract reactive dyes, but when the
fibres were treated with a reducing agent (sodium dithionite in sodium hydroxide), the
original dye, if azo in nature, was irreversibly decolorized. This distinguishes azo reactive
dyes from other dye classes, but prevents dye analysis by TLC. Cheng et al. (1991) described
a similar extraction and classification procedure. They reported the best solvent for reac-
tive dyes as 400 g/l hydrazine in water, but still many could not be extracted. Another
approach is to use enzymatic hydrolysis. Rendle et al. (1994) reported that pre-swelling
with sodium hydroxide, followed by digestion of the cotton fibres dyed with reactive dyes
using cellulase over a 24-hour period, produces homogenous coloured solutions. Typically
the fibre disappears. They are not true dye extracts but probably contain dye chromophores
bound to glucose molecules. These products are amenable to separation by TLC. The pro-
cess is successful for bulk and single fibres.
Extraction problems also occur with reactive dyes on wool. The high fastness of the
reactive dyes to the normal extraction processes is again likely to be due to covalent bonds
between the dyestuff and fibre (Macrae and Smalldon, 1979). These processes are designed
on the basis that dye will react with nucleophilic groups, such as amino, thiol and hydroxyl
functions in the amino acid residue side-chains of the keratin fibres. Crabtree et al. (1995)
reported a method for the alkaline hydrolysis of wool fibres dyed with reactive dyes and
subsequent analysis of the coloured products by TLC. Fibres were digested using 0.75 M
sodium hydroxide for 24 hours at 40°C. The resulting alkaline solutions were buffered
with 0.3 M citric acid in methanol, the methanol preventing precipitation of the coloured
components. The method has also been successfully applied to single wool fibres. Indeed,
Wiggins et al. (1996) demonstrated that TLC analysis of reactive dyes yields important
information over and above that obtained using techniques such as comparison micros-
copy and visible light microspectrophotometry.
Both enzymatic digestion of dyed cotton fibres and alkaline hydrolysis of wool fibres
are destructive methods of analysis. Thus, analysts have to decide whether, in terms of evi-
dential value in a particular case, the benefits of chromatographic analysis will outweigh the
loss of the fibre and whether the court will agree with the judgement made. In most cases,
only a proportion of the recovered fibres need to be examined in this way and others will be
available for re-examination if required. The question of ‘destroying’ fibres is far from being
a new one. The choice between including or omitting TLC methods in forensic fibre dye
comparisons is often a matter of local policy. Some laboratories have always regarded TLC,
with its implied dye extraction (for acid, basic, disperse, metallized, direct and azoic dyes)
to be a destructive technique, and as such they avoid it or use it only as a last resort. It can
be argued that microspectrophotometry over the full spectral range is sufficient, being both
discriminating and non-destructive. In other establishments, any additional discrimina-
tion offered by dye extraction and TLC is seen as too important to ignore.
6.2.6 Dye Analysis by Thin Layer Chromatography (TLC)

TLC has been used to compare control and recovered fibres in criminal cases for over
25 years, and even with the advent of instrumental methods it still has an important place
in forensic textile examination. TLC eluent sytems have been described for separating dyes
used on polyester, nylon and acrylic fibres (Beattie et al., 1981b), cellulosic fibres (Home
and Dudley, 1981) and polypropylene fibres (Hartshorne and Laing, 1984). Eluents for wool
fibre dyes had already been described (Macrae and Smalldon, 1979) and after a further ten
years a scheme for azoic dyes on cotton fibres was developed by Laing et al. (1990b). In
1989, Golding and Kokot looked at the possibility of combining normal and reversed phase
chromatography for the comparison of disperse, acid and reactive dyes.
If a mixture is applied in solution to a thin layer of support medium and solvents are
allowed to pass across it, the different components travel with the solvents at different rates
depending on their physical and chemical properties. Visually similar colours can be made
up of different component dyes and can therefore be distinguished quickly and easily using
TLC. Owing to the spectral properties of dye molecules and their high absorptivities, the
detection levels required can generally be achieved by the human eye. However, TLC does
have some limitations. As stated earlier, dye extraction may be difficult, or even impossible,
for particular dye classes, e.g. sulphur. Colours such as pale yellow will not be easily visible
on a TLC plate and fading of other colours does occur with time. Some dye classes, e.g.
azoic are more susceptible to this problem than others. Green (1982) showed that the fad-
ing can be slowed by spraying the plate with silicon. Assessing the colour and intensity of
the TLC bands is rather subjective. Golding and Kokot (1990) attempted to overcome this
by using a scanning densitometer to obtain the position, colour and relative proportions
of the various dye components in a semi-quantitative form. The instrument can also be
used in a fluorescence mode, providing extra information to help analysts form an opinion
on the similarity or non-similarity of the fibres in question. Kokot, in fact, reported the
detection limit for yellow dyes using scanning densitometry to be 10–20 times lower than
that obtained by visual means using reflected light. Having mentioned densitometry, it is
true to say that in general the coloured components are visible to the eye without the use
of sophisticated instrumentation.
Further comparison of TLC plates can be made by examining them under long-wave
UV light. Caution must be exercised in interpretation, however, as ‘extra’ bands may result
not from dye components, but from fibre finishing agents or from contaminants on the
control material (e.g. if the fibres have been taken from an area stained with grease or oil).
6.2.6.1 Thin Layer Chromatography of Non-reactive Fibre Dyes

This section will deal with the TLC of the following dye classes: acid, azoic, basic, direct,
disperse and metallized.
6.2.6.1.1 Principles of the Method Dye from the control garment is classified and
eluent systems are evaluated to achieve optimum separation of the dye extract. Dye is then
extracted from single, recovered and control fibres and the TLC chromatograms compared.
Extraction and eluent solutions are prepared as shown in Tables 6.4 and 6.5. Extraction
solutions should be tested before using to make sure that they have not become contami-
nated. Eluent solutions should also be checked to ensure they are performing as expected.
A standard dye, prepared as in Table 6.6, relevant to the eluent being tested is spotted
onto a TLC plate, e.g. Merck DC Alufolien Kieselgel 60F254 (7.5 × 5.0 cm), alongside the
relevant extraction solution. The plate should rest on a hot plate (at approximately 70°C).
The extracts (standard dye and extraction solution) are applied 1 cm from the lower
edge of the TLC plate using a finely drawn capillary to produce a spot of approximately
2 mm diameter. After the spots are fully dry the process is repeated to ensure that the
standard dye spot is strongly coloured. The TLC plate is placed in an oven at approximately
100°C for 5 minutes to ensure that it is completely dry.
The prepared plate is placed in approximately 10 mL of the appropriate eluent in a 250
mL beaker which is then covered with a glass Petri dish. Suitable brand-name TLC cham-
bers may be used as described by Laing et al. (1990a). The plate is eluted to 2 cm above the
origin, removed and dried in a hot air stream. Used eluent is discarded. If the extractant
solution track is clear, it is obvious that the extractant is free from contaminants and can be
Table 6.4 Fibre/Dye Combinations Encountered in

the Metropolitan Laboratory, Forensic Science Service,
March 1993 to April 1994
Textile Dye No. of Items
Cotton Sulphur 28
Cotton Direct 12
Cotton Reactive 33
Cotton Azoic 2
Polyester Disperse 36
Polyester Pigmented 2
Polyacrylonitrile Basic 97
Wool Reactive 6
Wool Acid 17
Wool Metallized 5
Polypropylene Pigmented 6
Cellulose (viscose) Direct 4
Cellulose (viscose) Pigmented 2
Acetate Disperse 3
Polyacrylonitrile + acrylamide Basic 5
Polyamide Acid 12
Table 6.5 Composition of Extraction Solutions

Pyridine/water 4:3 v/v, prepare 100 mL and use until exhausted
Formic acid/water 1:1 v/v, prepare 100 mL and use until exhausted
2% aqueous oxalic acid (0.2 g in 10 mL water), use immediately and discard excess
Table 6.6 Composition of Eluents

Eluent No. Solvents Proportions (v/v)
1 n-Butanol, acetone, water, 5:5:1:2
ammonia
2 Pyridine, amyl alcohol, 10% 4:3:3
ammonia
3 n-Butanol, ethanol, ammonia, 8:3:4:4:3
pyridine, water
4 Methanol, amyl alcohol, water 5:5:2
5 Toluene, pyridine 4:1
6 Chloroform, ethyl acetate, ethanol 7:2:1
7 n-Hexane, ethyl acetate, acetone 5:4:1
8 Toluene, methanol, acetone 20:2:1
9a n-Butanol, acetic acid, water 2:1:5
10 n-Butanol, ethanol, ammonia, 4:1:3:2
pyridine
11 Chloroform, butanone, acetic acid, 8:6:1:1
formic acid
12a n-Butanol, acetic acid, water 4:1:5
Note: The ethanol used is 99% and the ammonia is 0.880 SG
unless otherwise stated. Eluents 6 and 11 should be discarded
daily, all others on a weekly basis.
aThese eluents form an upper and lower phase. Use the upper
phase as the eluent.
used. The standard dye chromatogram is compared to previously stored chromatograms.

If it is separating as normal, the eluent may be used for casework chromatography.
6.2.6.1.2 Choice of Extraction Solution Dye classification indicates the best extraction
solution for a particular dye class and fibre type. This is summarized in Table 6.7.
6.2.6.1.3 Choice of Elution Solution A tuft of fibres is placed into a Durham tube (mea-
suring 2.5 cm × 0.3 cm internal diameter) and approximately 100 μL of extraction solution is
added. The open tube is placed in a sand bath and heated to approximately 100°C. The extrac-
tion progress is checked at 15 minute intervals for up to a maximum of 1 hour. If the sample to
be eluted is a mixture of fibre or dye types these are first separated. A few single fibres of each
type are placed in glass tubes and extracted as described for dye classification. The resulting
extracts are spotted on TLC plates as described for standard dye testing. A minimum of two
eluents should be evaluated for each dye extract. Common combinations of fibre type and dye
class, together with the eluents which give the best separation, are shown in Table 6.8.
If other fibre type/dye classes are encountered, then the appropriate eluents for that dye
class should be used. Once the plates have been eluted in the appropriate eluents, usually
to a distance of 2 cm from the origin, five parameters are considered when selecting the
optimum eluent:
• Separation of component bands
• Sharpness of bands
• Movement from the origin
• Components travelling at or close to the solvent front
• Strength of dye extract from recovered (questioned) fibres
Table 6.7 Preparation of Standard Dye Mixtures

Solution A for eluents 1, 2, 3, 4, 9, 10, 12
Solway green G (C1 acid green 25)
Solway blue RNS (C1 acid blue 47)
Naphthalene fast orange 2GS (C1 acid orange 10)
Solution B for eluents 5, 7, 8
Superacet fast orange G (C1 disperse orange 3)
Superacet fast violet B (C1 disperse violet 8)
Superacet scarlet 2G (C1 disperse orange 1)
Solution C for eluent 6
Solution D for eluent 11
Note: Approximately 5 mg of each dye component is
made up to a final volume of 25 mL with pyridine/
water 4:3 v/v. Use until the supply is exhausted.
Table 6.8 Choice of Extraction Solution

Dye Class Fibre Type Extraction Solution
Acid Wool Pyridine/water 4:3 v/v
Silk Pyridine/water 4:3 v/v
Polyamide Pyridine/water 4:3 v/v
Protein Pyridine/water 4:3 v/v
Polyacrylonitrile Pyridine/water 4:3 v/v
Polypropylene Pyridine/water 4:3 v/v
Azoic Cotton Pyridine/water 4:3 v/v
Viscose Pyridine/water 4:3 v/v
Basic Polyacrylonitrile Formic acid/water 1:1 v/v
Modified acrylic Formic acid/water 1:1 v/v
Polyester Pyridine/water 4:3 v/v
Direct Cotton Pyridine/water 4:3 v/v
Viscose Pyridine/water 4:3 v/v
Disperse Polyester Pyridine/water 4:3 v/v
Polyacrylonitrile Pyridine/water 4:3 v/v
Polypropylene Pyridine/water 4:3 v/v
Acetate Pyridine/water 4:3 v/v
Triacetate Pyridine/water 4:3 v/v
Metallized Wool 2% aqueous oxalic acid then
pyridine/water 4:3 v/v
Polypropylene 2% aqueous oxalic acid then
pyridine/water 4:3 v/v
If the eluents suggested produce poor separation, others appropriate to the dye class are eval-
uated. In exceptional circumstances eluents appropriate to other dye classes may be used.
6.2.6.1.4 Equivalent Fibre Testing If a recovered fibre to be tested is short in length or pale
in colour, there may be insufficient dye to obtain a result from TLC. In these circumstances
a control fibre, equivalent in all respects, should first be analyzed. It is impossible to say with
any degree of certainty that a a black cotton fibre 5 mm in length will be insufficient to obtain
a TLC whereas a black wool of the same length will be sufficient. Gaudette (1988) constructed
a table showing the minimum suggested fibre length needed for successful TLC, but this
should be treated with caution. The depth of colour, the length, the ease of extraction and
the experience of the analyst are all factors which affect whether a result will be obtained. If a
result is obtained from an equivalent control fibre, then the recovered fibre should also give a
result if it originates from that or from another textile made of identical fibres.
6.2.6.1.5 Single-Fibre Procedure Both recovered and control fibres normally need to
be removed from microscope slides following microscopy and microspectrophotometry.
Any traces of marker pen ink should be cleaned from the coverslip using an appropriate
solvent, e.g. acetone. The coverslip should be cracked around the fibre and an appropri-
ate solvent should then be used to soften the mountant. The fibre is removed and washed
in the solvent before placing it in a glass tube as previously described. After labelling the
tube it is stored in a covered Petri dish. Additional recovered fibres that require testing are
prepared in the same way as control fibres. The control fibres should be chosen so as to be
of an equivalent length and depth of dyeing to the recovered fibres, and should always be
treated in an identical manner.
The appropriate extractant is added to the tubes and, after sealing, they are placed in
an oven to extract. A TLC plate is labelled ensuring that a standard dye is included and
recovered fibre(s) extracts are spotted between control fibre extracts. The extracts are spot-
ted on the plate and eluted in a beaker or TLC chamber as described previously.
6.2.6.1.6 Comparison of Large Fibre Samples If large samples are available they can be
eluted in Durham tubes. Where possible, ‘co-chromatography’ is used where a mixture of
known and recovered extracts is also included. This is particularly useful if it is thought
that one of the samples is contaminated with a solvent or similar substance that may alter
the separation, e.g. petrol from a piece of material in an arson case.
6.2.6.1.7 Interpretation of Results The band position(s) and colour(s) produced by the
control and recovered fibres can be compared using visible and long-wave ultraviolet light.
A positive association is noted when band colours and positions are identical in control
and recovered fibres. A negative association is noted when either the control or the recov-
ered patterns show no similarities or where there are a number of coincident bands but
one or more bands are missing from the recovered or control sample. An inconclusive
association is noted when there are no bands on the TLC plate because of insufficient dye
present in the extract. If equivalent fibre testing is carried out, inconclusive results should
only rarely occur. Where single-fibre chromatograms from several control fibres show a
large amount of variation and the recovered fibre chromatograms show features that are
similar, but an exact match cannot be found, a ‘positive within range’ association is noted.
Subtle differences between control and recovered fibres should be carefully considered
before a negative association is reported. Testing of additional areas of the control garment
may result in a positive association being found. This happens when dye batch variation is
encountered (Wiggins et al., 1988). R f values, the distance from the origin to the centre of
the band divided by the distance travelled by the eluent, are difficult to measure, especially
on chromatograms obtained from pale colours or small amounts of extract. These values
vary with temperature and differences in eluent composition. Although they should be the
same for one plate, it would be impossible to use R f values to compare between plates.
6.2.6.2 Thin Layer Chromatography of Reactive Fibre Dyes

This section will deal with reactive dyes encountered on wool and cotton fibres.
6.2.6.2.1 Background and Principles of the Method Reactive dyes are covalently
bound to the fibre and cannot be removed by conventional extraction methods but can be
released from wool and cotton by disrupting the fibre. The coloured solutions are not true
dye extracts but are none the less amenable to separation and analysis by TLC. The sensi-
tivity of this technique enables the amount of dye present in single fibres to be analyzed.
Coloured solutions are released from a sample of the control garment by either chemical
or enzymatic digestion for wool and cotton fibres respectively. The solution is tested with
appropriate reagents to achieve optimum separation. Dye is then obtained from single
fibres and the component colours separated out by TLC. Recovered and control chromato-
grams are compared. Fibre disruption solutions and eluents are prepared as in Tables 6.9
and 6.10.
Table 6.9 Fibre Type/Dye Class and Appropriate Eluent

Fibre Type Dye Class Eluent Nos.
Wool or silk Acid or metallized 1, 2
Cotton or viscose Direct 1, 4, 3
Cotton or viscose Azoic 5
Polyester Disperse 6, 7, 8, 5
Polyacrylonitrile Basic 11, 12, 1
Polyamide Acid 9, 10
Table 6.10 Composition of Fibre Disruption Solutions

Acetic acid 0.5M glacial acid (prepare 100 mL and use until stock
is exhausted)
Cellulase 1.6 mg/mL sodium acetate buffer (prepare 50 mL and
discard at the end of each week)
Citric acid 0.3M in methanol (prepare 10 mL and discard at the
end of each week)
Sodium acetate buffer 0.1M in water adjusted to pH5±0.2 with glacial acetic
acid (prepare 50 mL and discard at the end of each
week)
Sodium hydroxide (for wool) 0.75M in water (prepare 100 mL and use until stock
is exhausted or discard at the end of one month)
Sodium hydroxide (for cotton) 3M in water (prepare 100 mL and use until stock is
exhausted or discard at the end of one month)
Note: Cellulase (Penicillium funiculosum) should be stored at –18°C and used until the
stock is exhausted.
Checks should be carried out just prior to use on the eluent performance, and the fibre
disruption solution should also be tested to ensure that it has not been contaminated. The
dissolution process is checked to ensure that it is working correctly by analyzing a standard
dyed fibre sample. Examples of standard dyed fibres are shown in Table 6.11.
Dye is released from the standard fibres using the following methods.
• Wool: Large fibre tufts. Place a tuft of fibres in a 500 μl centrifuge tube, add 100 μl
of 0.75M sodium hydroxide and incubate at 45°C for 24 hours with regular inver-
sion. Add 66 μl of 0.3M citric acid, mix and centrifuge at 7000 rpm for 5 minutes.
• Wool: Single fibres or small tufts. Push the fibres to the bottom of a glass tube as
previously described. Add 3 μl of 0.75M sodium hydroxide and seal the tube before
incubating at 45°C for 24 hours with continuous agitation. Open the tube and add
2 μl of 0.3M citric acid, mix and centrifuge at 7000 rpm for 5 minutes.
• Cotton: Large fibre tufts. Place a tuft of fibres in a 500 μl centrifuge tube, add 50 μl of
3M sodium hydroxide and keep at 0°C for 4 hours. The solution is then discarded.
Resuspend the sample in 50 μl of 0.5M acetic acid for 20 seconds and discard the
solution. Wash the sample twice with 150 μl of cellulase solution and decant the
solution. Resuspend the sample in 150 μl of cellulase solution and incubate at 45°C
for 20 hours with regular inversion. Centrifuge at 7000 rpm for 5 minutes. Transfer
an aliquot of the dye solution to a new 500 μl centrifuge tube, add an equal volume
of methanol and mix.
• Cotton: Single fibres or small tufts. Push the fibres to the bottom of a glass tube as
previously described. Add 5 μl of 3M sodium hydroxide and seal the tube. It is then
kept at 0°C for 4 hours before discarding the solution. Resuspend the sample in 5
μl of 0.5M acetic acid for 20 seconds and discard the solution. Resuspend the sam-
ple in 3 μl of cellulase solution, reseal the tube and incubate at 45°C for 20 hours
with continuous agitation. 3 μl of methanol is added, mixed and centrifuged at
7000 rpm for 5 minutes.
In order to check the eluents, a standard dye (see Table 6.12) is spotted onto a TLC plate,
e.g. Merck DC Alufolien Kieselgel 60F254 (7.5 × 5.0 cm), alongside the standard fibre dye
and the disruption solution. The spotting is performed while the plate is resting on a hot-
plate (70°C approximately). The plate is dried as previously stated and then eluted in the
appropriate eluent. The standard dye chromatogram and that obtained from the standard
Table 6.11 Composition of Eluents

Eluent Proportions
No. Solvents (v/v)
1 n-Butanol, acetone, water, ammonia 5:5:1:2
2 Pyridine, amyl alcohol, 10% ammonia 4:3:3
3 n-Butanol, ethanol, ammonia, pyridine, water 8:3:4:4:3
4 Methanol, amyl alcohol, water 5:5:2
13 Propan-1-ol, methanol, water, ammonia 6:3:1:4
Note: The ammonia used is 0.880 SG unless otherwise stated. All eluents should
be discarded at the end of the week in which they are prepared.
Table 6.12 Preparation of Standard Dye Mixtures

Solution A for eluents 1, 2, 3, 4, 13, 14, 15
Solway blue RNS (C1 acid blue 47)
Naphthalene fast orange 2GS (C1 acid orange 10)
Note: Approximately 5 mg of each dye component is made up to a final vol-
ume of 25 mL with pyridine/water 4:3 v/v. Use until this supply is exhausted.
fibre dye solution are checked against stored chromatograms to ensure that the separation
is adequate and that they match in all aspects. The disruption solution is also checked to
ensure that there are no visible bands present.
Once control fibre extracts have been prepared as described, they should be evaluated
in a minimum of two eluents.
Reactively dyed wool extracts separate well in eluents 13, 1 and 2, whereas cotton
extracts separate well in eluents 14, 15, 3 and 4. When the best eluent has been decided on
using the same parameters as described for non-reactive fibre dyes, equivalent fibres can be
prepared and tested as previously described.
6.2.6.2.2 Single-Fibre Procedure The sample preparation is as previously described

and once the single fibres are in tubes, the method for cotton or wool single fibre dye
release is followed. Extracts are spotted onto TLC plates and eluted in the appropriate elu-
ent. Large samples can be compared using co-chromatography. The results are interpreted
as discussed for non-reactive fibre dyes.
6.2.6.3 Non-Extractable Dyes

This section will deal with ingrain, sulphur and vat dyes. If classification indicates that a
non-extractable dye or pigment is present, then one recovered and one control fibre should
be placed in glass capillary tubes. Approximately 10 μL of pyridine/water 4:3 v/v should be
added and an attempt made to extract them at 100°C for 1 hour. If neither fibre extracts,
then it can be recorded as a positive result. If the recovered extracts but the control does
not, then it is recorded as a negative result. Care should be taken to observe that on rare
occasions fibres coloured with these dye classes can bleed dye into the extraction solution.
In this case there may be sufficient dye in solution for analysis.
6.2.7 Conclusions
Comparison microscopy and visible microspectrophotometry are the first choice tech-
niques for the comparison of dyes that are encountered in textile fibres in the field of
forensic science. Many laboratories have microspectrophotometers that allow the colou-
rants to be analyzed in the visible region, but few allow for analysis to be carried out in
the ultraviolet range. Thus it is still accepted that a third comparative technique should
be used in many cases. Now that methods are available for analysis of reactive dyes,
most dyes encountered in forensic fibre examination can be compared using TLC. This
technique is used extensively because it is cheap and, with practice, relatively easy to
perform.
Acknowledgements
Much of the groundwork for this chapter was done while preparing a paper entitled
‘Forensic analysis of textile fibre dyes’ (Rendle and Wiggins, 1995). Ken Wiggins would
therefore like to acknowledge the work of Dr David Rendle of the former Metropolitan
Laboratory of the Forensic Science Service.
References
Beattie, I. B., Dudley, R. J., and Smalldon, K. W., 1979. The extraction and classification of dyes on
single nylon, polyacrylonitrile and polyester fibres, J. Soc. Dyers Colour., 95, 295–301.
Beattie, I. B., Roberts, H. L., and Dudley, R. J., 1981a. The extraction and classification of dyes from
cellulose acetate fibres, J. Forensic Sci. Soc., 21, 233–237.
Beattie, I. B., Roberts, J. L., and Dudley, R. J., 1981b. Thin layer chromatography of dyes extracted
from polyester, nylon and polyacrylonitrile fibres. Forensic Sci. Int., 17, 57.
Cheng, J., Wanogho, S. O., Watson, N. D., and Caddy, B., 1991. The extraction and classification of
dyes from cotton fibres using different solvent systems, J. Forensic Sci. Soc., 31, 31–40.
Crabtree, S. R., Rendle, D. F., Wiggins, K. G., and Salter, M. T., 1995. The release of reactive dyes from
wool fibres by alkaline hydrolysis and their analysis by thin layer chromatography, J. Soc. Dyers
Colour., 111, 100–102.
Feeman, J. F., 1970. An introduction to modern methods of dye identification—Chromatography
and spectrophotometry, Can. Text. J., 87, 83–89.
Fong, W., 1989. Analytical methods for developing fibers as forensic science proof: A review with
comments, J. Forensic Sci., 34(2), 295–311.
Gaudette, B. D., 1988. The forensic aspects of textile fiber examination. In R. Saferstein (Ed.) Forensic
Science Handbook (Vol. 2, pp. 209–272). Englewood Cliffs, NJ: Prentice Hall.
Golding, G. M. and Kokot, S., 1989. The selection of non-correlated thin layer chromatographic
solvent systems for the comparison of dyes extracted from transferred fibres. J. Forensic Sci.,
34, 1156–1165.
Golding, G. M. and Kokot, S., 1990. Comparison of dyes from transferred fibres by scanning densi-
tometry, J. Forensic Sci., 35(6), 1310–1322.
Hartshorne, A. and Laing, D. K., 1984. The dye classification and discrimination of coloured poly-
propylene fibres, Forensic Sci. Int., 25, 133–141.
Home, J. M. and Dudley, R. J., 1981. Thin layer chromatography of dyes extracted from cellulosic
fibres, Forensic Sci. Int., 17, 71–78.
Laing, D. K., Dudley, R. J., Hartshorne, A. W., Home, J. M., Rickard, R. A., and Bennett, D. C.,
1991a. The extraction and classification of dyes from cotton and viscose fibres, Forensic Sci.
Int., 50, 23–35.
Laing, D. K., Hartshorne, A. W., and Bennett, D. C., 1990b. Thin layer chromatography of azoic dyes
extracted from cotton fibres, J. Forensic Sci. Soc., 30, 309–315.
Macrae, R., Dudley, R. J., and Smalldon, K. W., 1979. The characterization of dyestuffs on wool fibres
with special reference to microspectrophotometry, J. Forensic Sci., 24, 117–129.
Macrae, R. and Smalldon, K. W., 1979. The extraction of dyestuffs from single wool fibres, J. Forens.
Sci., 24, 109–117.
Rendle, D. F., Crabtree, S. R., Wiggins, K. G., and Salter, M. T., 1994. Cellulase digestion of cotton
dyed with reactive dyes and analysis of the products by thin layer chromatography, J. Soc. Dyers
Colour., 110, 338–341.
Rendle, D. F. and Wiggins, K. G., 1995. Forensic analysis of textile fibre dyes, Rev. Prog. Color,
25, 29–34.
Resua, R., 1980. A semi-micro technique for the extraction and comparison of dyes in textile fibers,
J. Forensic Sci., 25, 168–173.
Resua, R., De Forest, P. R., and Harris, H., 1981. The evaluation and selection of uncorrelated
paired solvent systems for use in the comparison of textile dyes by thin-layer chromatogra-
phy, J. Forensic Sci., 26, 515–534.
Society of Dyers and Colourists, 1985. Colour Index (Vols. 1–6, 4th ed.). Bradford: Society of Dyers
and Colourists.
Venkataraman, K. (Ed.), 1977. The Analytical Chemistry of Synthetic Dyes. London and New York,
NY: Wiley.
Waring, D. R. and Hallas, G. (Eds.), 1990. The Chemistry and Application of Dyes. New York, NY and
Wiggins, K. G., Cook, R., and Turner, Y. J., 1988. Dye batch variation in textile fibers, J. Forensic Sci.,
33(4), 998–1007.
Wiggins, K. G., Crabtree, S. R., and March, B. M., 1996. The importance of thin layer chroma-
tography in the analysis of reactive dyes released from wool fibers, J. Forensic Sci., 41(6),
1042–1045.
Wiggins, K. and Holness, J.-A., 2005. A further study of dye batch variation in textile and carpet
Wiggins, K., Palmer, R., Hutchinson, W., and Drummond, P., 2007. An investigation into the use
of calculating the first derivative of absorbance spectra as a tool for forensic fibre analysis, Sci.
Justice, 47(1), 9–18.
Williams, D. H. and Fleming, I., 1966. Spectroscopic Methods in Organic Chemistry. Maidenhead:
McGraw-Hill.
Infrared Microspectroscopy
of Fibres
K PAUL KIRKBRIDE
7
Contents
7.2 Infrared Microspectroscopy 246
7.2.2 Spectral Photometric Accuracy 250
7.2.3 Specimen Contributions to Photometric Inaccuracy 250
7.2.3.1 Specimen Thickness 252
7.2.3.2 Specimen Flatness 252
7.2.3.3 Scattering 254
7.2.3.4 Refraction and Cross-Sectional Shape 255
7.2.4 Instrumental Contributions to Photometic Inaccuracy 256
7.2.4.1 Stray Light 256
7.2.4.2 Diffraction 257
7.2.4.3 Beam Polarization 265
7.2.5 Recommended Techniques 266
7.2.5.1 Diamond Cell 267
7.2.5.2 Rolled Fibres 269
7.2.5.3 Setting-up Apertures 269
7.2.5.4 Transmission-Reflection 270
7.2.6 Micro-ATR Spectroscopy 271
7.2.7 Infrared Imaging 274
7.2.7.1 Polarized Infrared Microspectroscopy 275
7.2.8 IR Spectrum Interpretation 278
7.2.8.1 Computer-Assisted Processes 279
7.2.8.2 Characterization Using ‘First Principles’ 279
7.3 Conclusions 286
References 286
7.1 Introduction
Infrared spectroscopy and Raman spectroscopy both probe the same fundamental char-
acteristic of a molecule, namely its vibrational modes. Accordingly, infrared and Raman
spectroscopy are regarded as particular types of the more general field of vibrational spec-
troscopy. In the interval between the first edition of this book (i.e. the early 1990s) and the
present one, infrared microspectroscopy has become a mature, cornerstone technique for
forensic fibre analysis and Raman spectroscopy has gained strong acceptance.
Each technique is now a field unto itself and as the operational details of infrared
and Raman are quite different, especially with regards to the wavelength of radiation used
to collect a spectrum, the instrumentation, development and strengths and weaknesses
245
this book will now devote separate Chapters to infrared and Raman microspectroscopy,
respectively chapters 7 and 8.
This chapter is a guide to the practices associated with infrared microspectroscopy as
applied to single textile fibres rather than an overview of the theory of infrared spectros-
copy or Fourier transform techniques. For those looking for an introduction to these topics
the works by Stuart (2004) and Griffiths and De Haseth (2007) are recommended.
The contributions of Mary Widmark Tungol to the previous version of this Chapter are
acknowledged and a significant quantity of the present text is based on her contribution.
7.2 Infrared Microspectroscopy
7.2.1 Introduction
Textile fibres of interest to forensic science are usually only a few millimetres in length and
between 10 and 100 microns in diameter. In principle it should be possible to place such a
fibre into the beam of a conventional infrared spectrometer and acquire an infrared spec-
trum. However, as the beam usually measures about 10–15 millimetres in diameter only
an extremely small percentage of the beam will pass through the sample and carry spectral
information to the detector. Most of what the detector receives is called ‘stray radiation’
or ‘stray light’, that is, radiation that has reached the detector but not passed through the
specimen; this situation is depicted in Figure 7.1. As is indicated later, stray radiation leads
to perturbation of spectral data and steps should be taken to eliminate it.
In principle it is possible to mount the fibre in a mask of some sort, so that stray radia-
tion is blocked from reaching the detector (Figure 7.2). In this instance the detector is
‘under-filled’ and a huge percentage of its active surface is susceptible to random noise.
Therefore the weak signal arising from the specimen is swamped by background noise.
In order to avoid these problems, forensic fibre analysis is usually performed with the
aid of a microspectrometer, in which as much as possible of the radiation from the source
is directed through the specimen, and as much as possible of the signal from the specimen
is collected and presented efficiently to the detector (Figure 7.3).
Microspectroscopy can be achieved using either a beam-condensing accessory, or an
infrared microscope. Both devices function, in principle, as outlined in Figure 7.3, but the
IR beam
Fibre
Detector
Figure 7.1 Fibre placed into standard beam of an infrared spectrometer. Under these circum-
stances only a very small portion of the beam can be modulated by the fibre, and the detector
receives mostly stray light (i.e. radiation that has not passed through the specimen).
Infrared Microspectroscopy of Fibres 247
IR beam Masks
Fibre
Detector
Figure 7.2 Stray light prevented from reaching the detector by use of a mask. Detector is
under-filled, therefore data acquired will have a poor signal to noise ratio.
IR beam
Fibre
‘Lens’ Detector
‘Lens’
Figure 7.3 The ideal features for microspectroscopy. All of the beam passes though the sample
(thereby eliminating stray radiation), and all transmitted radiation is collected and presented
to all of the detector element.
infrared microscope offers such ease of use and performance superiority that it will be
the only apparatus discussed in this chapter. One advantage of the infrared microscope is
that the specimen can be seen at high magnification with visible light, and the stage can
be moved along three orthogonal axes enabling the specimen to be brought into the field
of view and into sharp focus prior to analysis. For infrared analysis the microscope offers
good sensitivity due to its dedicated detector and it offers great convenience as the circular
beam can be trimmed to a shape closely matching the shape of the specimen, thus mini-
mizing stray light.
A typical infrared microscope is depicted in Figure 7.4 and a general functional
description follows.
The beam from the optical bench passes through the interferometer before it is diverted
into the transfer optics of the microscope. The beam is then converged to a small diameter
at the first focal plane of the microscope. Located at this focal plane is a collection of four
opaque (to IR radiation) knife-edges that can be moved independently into the infrared
beam (see Figure 7.4 inset), these are used to trim the circular beam into shapes such a
rectangular slit or a square, which can be arranged to reduce stray radiation further and/
or delineate a particular region of interest in the specimen. Strictly speaking, these knife-
edges form a diaphragm at the focal plane, but they are more commonly referred to as a
focal plane aperture. In deference to common usage, in this chapter the term aperture will
also be used and the aperture at the focal plane between the source and specimen will be
referred to as the first aperture.
From IR
source
Eyepieces
‘Lenses’
Apertures
Detector
Figure 7.4 Schematic diagram of an infrared microscope. Inset shows the construction of a
focal plane aperture.
The beam then proceeds to an optical element that functions as a lens, which focusses
the beam down to a small spot a few hundreds of microns across at the specimen plane
(i.e. at the stage of the microscope). With respect to the infrared beam this lens functions
as a condenser element and it will be referred to as such for the remainder of this chap-
ter. As infrared radiation is absorbed by glass, the objective uses a combination of convex
and concave mirrors in what is referred to as a Schwarzchild configuration to achieve
magnification.
After transmission through the sample, the radiation is collected by another reflecting
‘lens’, which is an objective element and will be referred to as such for the rest of this chap-
ter. The objective gathers radiation and in some instruments passes it through a second
set of knife edges located at the rear focal plane (which will be referred to as the second
aperture) before it is focussed onto an IR detector.
In microspectrometers the detector is a mercury cadmium telluride (MCT) type rather
than the deuterated triglycine sulphate (DTGS) type that is encountered in macro sample
spectrometers. The MCT detector element is a small chip (ranging from 250 μm square to
50 μm square depending upon the magnification of the microscope optics) that is main-
tained at low temperature in a bath of liquid nitrogen so that stray thermal noise is reduced
to a minimum. MCT detectors can be obtained in narrow-, medium-, and wide-band form
that detect down to about 700, 600, and 450 cm−1, respectively. Narrow-band detectors are
about 2.5 more sensitive than wide-band detectors, so they are a typical choice for fibre
analysis.
Although the field of view of the microscope might be about 1300 μm when a total
magnification of 150x is used, the actual dimension of the functional IR beam at the speci-
men is much less. Furthermore, the intensity of the infrared beam is not constant across
the specimen plane. Figure 7.5 shows the intensity of the beam as a function of location
within the specimen plane (measured as align signal passed through an 8 μm × 100 μm
aperture at various locations in the sample plane). As can be seen, the intensity of the beam
drops off towards the edges of the specimen plane, and is at a maximum in the centre. With
the example shown only a region of approximately 100–150 μm of the full 1300 μm field
of view is usable for spectroscopy. Although it should be the case that this active spot is in
the centre of the field of view, poor alignment of the optics (which might arise from jolts to
the instrument or injudicious human interference) results in movement of the maximum
away from the centre of the specimen plane. Clearly it is critical to ensure that microspec-
trometers are aligned, and it is good practice to record foreground and background spectra
in the same region of the beam and close to its centre. Some modern microscopes are
arranged such that the aperture is always centred in the field of view and therefore there is
no choice but to record the foreground and background in the same relative location in the
beam which, if the beam is aligned, will be the centre of it.
Assuming that the fibre is correctly positioned in the beam it would appear that an
infrared microspectrometer and a conventional spectrometer are functionally equiva-
lent, that is, the spectrum of a nylon fibre acquired using a microspectrometer would be
identical to the spectrum of a nylon sheet acquired using a conventional spectrometer.
For a number of reasons the two types of instruments are not equivalent however, and it
has even been stated (Davis, 2011) that infrared microspectrometry is effectively useless
for identification of the molecular composition of intact fibres of small diameter. First, the
infrared beam in a microspectrometer is brought to a sharp focus on or in the specimen,
therefore if the specimen is planar then the extremities of the beam traverse a longer path
250
Align signal (mV)
200
150
100
50
0
100 80 60 40 20 0 20 40 60 80 100
Distance from centre of specimen plane
Figure 7.5 Energy transmitted by focal plane aperture (8 × 100 μm) between source and speci-
men plane moved horizontally (top curve) and vertically (bottom curve) across the field of view.
through the specimen than the centre of the beam. This is not the case in a conventional
spectrometer where the beam is only slightly convergent, if not parallel. Therefore, even in
the case of simple specimens spectra acquired using the two types of instrument will be
slightly different due to the apparent thickness of the specimen. Fibres are usually not flat
and their shape further complicates their apparent thickness and in addition causes then to
act as a complex refractor (e.g. as a cylindrical lens in the case of round fibres) that causes
some wavelength-dependant refractive losses of the beam compared to a bulk, planar speci-
men. Finally, microscopic specimens (including fibres) sometimes have width and thick-
ness that approach the dimensions of the wavelengths in the mid-infrared range. This,
together with unavoidable consequences of forcing mid-infrared radiation through the
optical train of the microspectrometer (apertures and ‘lenses’) and past edges and bound-
aries of and within the specimen results in a number of distortions of spectral data due
to diffraction. Compared to spectroscopy using conventional spectrometers therefore the
use of an infrared microspectrometer to acquire spectral data from microscopic speci-
mens results in some distortions of spectral data. Some are unavoidable and of minor con-
sequence for forensic fibre comparisons, such as the perturbation of apparent specimen
thickness due to the convergence of the beam in a microspectrometer. Some carry more
significant consequences, which although unavoidable, can be minimized by the user.
Some distortions can be avoided altogether. A number of the following sections deal with
the origins of these distortions, how to recognize them, and how to minimize their impact.
For additional discussion the reader is directed to the works of Davis et al. (2010a and b,
and 2011) and Roush (1987).
7.2.2 Spectral Photometric Accuracy

An infrared spectrum is a plot of two data values; absorbance and frequency. Many spec-
troscopists would be (rightly) appalled if a spectrometer was to produce significant errors
in frequency data, yet many ignore or do not recognize errors in absorbance values. Such
errors in absorbance are termed photometric inaccuracy. For a spectrum to be photomet-
rically accurate every absorption throughout the spectral range should reach its correct
absorbance value and every point on the baseline should be at an absorbance of zero.
For forensic purposes it is desirable that spectral data acquired are photometrically
accurate. If the task at hand is to identify a fibre’s polymeric composition by comparison
of its infrared spectral properties against an infrared spectral database then an inaccurate
spectrum might not yield a satisfactory search result. Even if the task is comparative,
such as comparison of recovered fibres with their putative source, accuracy is important.
Distortions of spectral data might be interpreted as genuine differences, with the result
that a falsely negative conclusion as to association might be drawn (i.e. a Type 1 error). The
following sections describe various sources of photometric inaccuracy arising from the
specimen itself and from the instrument used with particular reference to infrared micro-
spectroscopical analysis of textile fibres.
7.2.3 Specimen Contributions to Photometric Inaccuracy

Radiation interacts with matter through a variety of mechanisms. If the specimen trans-
mits radiation then it will also refract it and probably absorb some of it. The specimen can
also reflect and scatter incident radiation. Just as absorption of radiation is not constant
.9
.8
Amide A and B
.7
.6
.5
.4
.3
.2
.1
0
3500 3000 2500 2000 1500 1000
1.2
.8
.6
.4
.2
3500 3000 2500 2000 1500 1000

1.6
1.4
1.2
1
.8
.6
.4
.2
0
3500 3000 2500 2000 1500 1000
Figure 7.6 Horizontal axis in wavenumbers, vertical axis in absorbance. TOP: Zeftron 500
fibre (nylon, 46.5 μm) lightly pressed in a diamond cell. Note the relative absorbance values
of the major peaks in the spectrum, the lack of resolution between peaks, and the spikes on
the tops of the amide A and B peaks. MIDDLE: Same specimen pressed further. Although
resolution is improved, the amide A and B peaks are still noisy at their maxima. BOTTOM:
Same specimen pressed even further. Compared to the top spectrum note the following: the
narrowness of the major peaks and better resolution apparent; the amide A and B peaks are
smooth and exhibit single maxima; the higher absorbance values of major peaks. In order to
produce this spectrum high pressure and two separate pressings were required to produce a
specimen of required thickness. This is a characteristic of thick nylon fibres.
across the spectral range (this is, after all, the basis of all spectroscopic techniques), nei-
ther is scattering, reflection, nor refraction. Distortions due to these phenomena will be
greater at one end of the spectral range or the other. Fortunately, these effects tend to vary
smoothly and in a predictable fashion across the spectral range (unlike absorption, which
exhibits peaks).
7.2.3.1 Specimen Thickness

Ideally, organic polymeric specimens should be no thicker than about 10–20 μm for infra-
red analysis. Recovered fibres are usually extremely small, but ironically most are too thick
for analysis without some kind of modification. Figure 7.6 illustrates what can happen to
photometric accuracy as specimen thickness is increased. Up to a certain thickness Beer’s
law is obeyed, that is, as thickness is increased absorbance levels for every peak increase
by a proportional amount. For example, if specimen thickness is doubled, then absorbance
values for every peak in the spectrum are doubled. Above a certain thickness however, the
ideal linear relationship described by Beer’s law breaks down. As thickness increases, peaks
will eventually reach complete absorbance. Obviously as thickness increases still further,
peaks that have reached complete absorbance cannot increase their absorbance further,
only those peaks that are not completely absorbing can absorb further. The outcome is
that when a sample is too thick there will be a high level of photometric inaccuracy. In
the extreme case, resolution of peaks will be lost. Furthermore, in the regions of very high
beam absorbance, there is very little light reaching the detector. As a consequence the spec-
trum will exhibit low signal-to-noise ratio in these regions; the tops of the peaks will be
very noisy, and exhibit splitting. Specimen thickness is usually the biggest contributor to
photometric inaccuracy, but it is easy to control through rolling or squashing (as described
later). Further examples of the affects of sample thickness can be found in articles by
Tungol et al. (1991) and Carter et al. (1989).
7.2.3.2 Specimen Flatness

Another important spectral distortion, one that arises not from the substance within the
specimen, but from any flat, smooth surfaces that it might possess (or that the sampling
accessory confers to it) is interference fringing. If a fibre has been flattened into a smooth
ribbon-shape then the beam from the infrared microscope can suffer internal reflection on
the flat surfaces (see Figure 7.7). Compared to the signal that passes straight through the
specimen the doubly reflected signal reaches the detector having traversed a longer path.
The detector is not able to recognize that the signal from the longer path is actually an echo
of the other and treat it accordingly. Instead the detector produces a summed response as
shown in Figure 7.7.
The position of the echo is entirely dependent upon the thickness and refractive index
of the specimen; a thin specimen of low refractive index will have the subsidiary located
close to the main centreburst, while a thick specimen of high refractive index gives a sub-
sidiary further away from the main centreburst. Upon Fourier transformation the summed
interferogram shown in Figure 7.7 yields the infrared spectrum of the sample, but the nar-
row spike of the subsidiary interferogram is transformed into a broad-band artefact in the
spectrum, which is manifested as a sinusoidally modified or rippled baseline (Figure 7.8).
The ripples are referred to as interference fringes. Just as the position of the echo in the
interferogram depends upon the thickness of the sample, the periodicity of the baseline
sinusoid is a function of sample thickness as described by Equation 7.1.
Specimen
Figure 7.7 Depiction of transmission and multiple internal reflection (MIR) of an infrared
beam in a planar sample. The interferogram (bottom) recorded is the sum of the principle
interferogram (top) and the subsidiary (bottom) produced by MIR.
1 ⎡ 10N ⎤
Thickness (mm) = η ⎢ (7.1)
2 ⎣ ( ν1 − ν2 ) ⎥⎦
where η is the refractive index of the sample, ν1 is the frequency of one fringe in the sinusoid,
ν2 is the frequency of another fringe in the sinusoid, and N is the number of fringes present
between ν1 and ν2.
Unlike other contributors to photometric inaccuracy, interference fringes can be con-
fused with actual absorptions, making it difficult to interpret and compare spectra. With
the exception of acrylic, the region between 1900 and 2800 cm−1 for fibres is usually free of
broad absorptions and careful examination of this region for ripples as shown in Figure 7.8
will indicate whether interference fringes might be present. They can be either removed
using data treatments or prevented from arising in the first place. After data acquisition the
spike can be removed (with some difficulty) from the interferogram. Spikes are not easy to
recognise, particularly when they are situated close to the centreburst. Furthermore, if the
spike is close to the centreburst, replacing it with a string of zeros in the interferogram might
degrade the spectral information (a spike some distance from the centreburst can easily
.5
–.5
3500 3000 2500 2000 1500 1000
.5
–.5
3500 3000 2500 2000 1500 1000
Figure 7.8 Horizontal axis in wavenumbers, vertical axis in absorbance. TOP: Spectrum of
nylon fibre (Zeftron 500). BOTTOM: Spectrum of Zeftron fibre pressed flat. Interference fringes
are obvious as a sinusoidal modification of the baseline. The most obvious spurious peaks are
present below 1000 cm–1. If the effects due to interference are not recognized, then the obvious
(but erroneous) conclusion is that the two spectra in this figure represent different fibre types.
be replaced by a string of zeros, however, as this has only an effect upon narrow features
in the transformed spectrum). One procedure, described by Krishnan and Ferraro (1982)
and Hirschfeld (1978), is to record two spectra of the specimen; one where the specimen
is normal to the beam, and another where the specimen is inclined. The inclination alters
the path length through the specimen, and therefore alters the position of the subsidiary
centreburst in the interferogram. Subtraction of the two interferograms yields a difference
interferogram that contains only features due to the subsidiary centreburst. This enables
the subsidiary to be easily found (for example in the interferogram recorded with the spec-
imen normal to the beam) and removed, data from the other interferogram can then be
used to ‘patch the hole’ where the subsidiary used to be.
However, prevention of interference fringing is better, and easier to achieve, than a
cure for it. Section 1.2.5 describes how sample preparation techniques can be used to mini-
mize fringing.
7.2.3.3 Scattering
As indicated above, it is desirable that the specimen does not have plane, parallel,
smooth surfaces, otherwise interference fringing is likely to cause problems. If, on the
other hand, the surface of the specimen is rough, then the situation might still not
be perfect because the incident beam might be scattered by the surface. The presence
of solid particles in the specimen, such as titanium dioxide delusterant in fibres, will
also cause scattering of the beam. Effectively, scattering is deviation of the incident
beam in random directions so that a fraction of it does not reach the detector, which
effectively becomes a broad absorption feature in the spectrum. It is not a constant
phenomenon across the range, scattering very quickly becomes more pronounced at
shorter wavelengths. Scattering can be recognized as a shift of the baseline to higher
absorbance values at the left hand side of the spectrum (i.e. at short wavelengths).
Simple subtraction of arbitrary levels of absorbance to return the baseline to zero (as
performed by standard baseline correction routines supplied by instrument software
manufacturers) achieves a cosmetic correction of the baseline, but leaves minor under-
lying photometric inaccuracies. If the task at hand is to compare two fibres on the
basis of their infrared spectral data, and the level of scattering in each fibre is about
the same, then inaccuracies generated by this phenomenon can safely be ignored.
However, if a baseline-corrected spectrum is compared with library spectra recorded
in the absence of scattering, it is important to recognise that absorbance values of
peaks at high frequency (C–H, O–H, and N–H stretches, for example) might be a little
low in the unknown spectrum.
7.2.3.4 Refraction and Cross-Sectional Shape

Refraction of the infrared beam within the specimen is a phenomenon that sometimes
has a surprisingly large impact upon spectral data. The refractive index of a compound is
a function of wavelength, but it is not a smooth function. In the wavelength region where
absorption take place, the refractive index of a compound abruptly rises, falls below its
unperturbed value and then rises again to its unperturbed value. This causes a concomi-
tant rise, fall, and rise in reflectivity of the specimen over the wavelength range, which
means that the light transmitted by the specimen to the detector falls, rises, and then falls
again. This perturbation is superimposed upon the profile of the infrared absorbance (i.e.
its peak shape) causing it to rise, fall below the baseline then rise up to it. This phenomenon
is referred to as the Christiansen Effect, and it is often observed in the spectra of acrylic
fibres. Although the perturbation of the baseline is annoying, a more serious concern is
that the Christiansen Effect causes the top of an absorption to move to a slightly different
frequency. Before concluding that two fibres do not match on the basis of slight frequency
mismatches the fibre analyst should rule out the possibility that different extents of the
Christiansen Effect are in play.
The combination of refraction, scattering, rapid fluctuations in refractive index across
the spectral range and cross-sectional shape of fibres can also have an impact upon spec-
tral data. Fibers with a circular cross-section function like a cylindrical lens of short focal
length. As described by Davis et al. (2011), the cylindrical shape gives rise to significant
distortion of the optical fields in and around them, which in turn results in significant
spectral distortion, especially when the thickness of a fibre approaches the wavelength of
the spectrometer beam.
The distortions can be manifested in spectra as small, high frequency ripples or large
and broad “humps”, sometimes both. Fibers with irregular cross-sections will refract
radiation, but in a more complicated manner. The cross-sectional shape of fibers also has
another affect. In the case of an untreated fiber, no matter which cross-sectional shape
is involved, some parts of the IR beam travel through a longer path length than others.
The outcomes of this phenomenon have been described by Hirschfeld (1979a, b). In a
forensic context, where the usual task is to compare questioned and known fibres that
have the same shape and size, the affects should be minimal, assuming that the known
and unknown fibres are analyzed using identical instrument settings. As indicated above,
squashing or rolling fibres will reduce spectral distortions arising from excessive thick-
ness, but there is the additional benefit that these treatments will destroy the cylindrical
shape of fibres and therefore spectral distortions arising from the cross-sectional shape
will also be minimized. Anomalies will be apparent, however, if spectral data acquired
from fibres are compared to data acquired from bulk samples.
7.2.4 Instrumental Contributions to Photometic Inaccuracy

7.2.4.1 Stray Light
Stray light is radiation that reaches the detector but has not passed through the sample. An
example of a spectrum acquired with a large amount of stray light is given in Figure 7.9a.
In this example a polyester fibre (Enkron 7151, 19 μm in diameter) was pressed to reduce
its thickness and widen it to about 50 μm, but the spectrum was acquired without the
assistance of any apertures. The effective beam diameter at the specimen was about
180 μm, therefore the proportion of the effective beam area (π × 902 = 25,400 μm2) that
illuminated the fibre (50 × 180 = 9000 μm2) is about 35%. Expressed another way, 65%
of the incident beam does not interact with the fibre, and reaches the detector as stray
.1
.08
.06
.04
.02
3000 2000 1000
1.5
.5
3000 2000 1000
Figure 7.9 (a and b) Horizontal axis in wavenumbers, vertical axis in absorbance. (a) Top spec-
trum; flattened Enkron 7151 fibre spectrum recorded using an unapertured beam (high stray
light). (b) Bottom spectrum; same flattened Enkron fibre, spectrum recorded with effectively
no stray light.
light. These figures are slightly underestimated because the beam is much brighter in the
centre of the field of view compared to the wings (as illustrated in Figure 7.5). The affects
of stray light can be seen by comparing the spectrum in Figure 7.9a (which features a large
contribution from stray light) with that in Figure 7.9b, which was recorded using the same
specimen and identical instrument parameters but less stray light. First, the signal to noise
in Figure 7.9a appears low, as though a thinner fibre was used or fewer scans. Second, small
peaks appear to be too large (look, for example, at the bands near 3000 and 700 cm−1) and
third, the absorbance values of all peaks are very low (for example the absorbance of the
C = O stretch is about 0.09 vs 1.8).
In the microspectrometer apertures are used to block stray light and thereby improve
photometric accuracy, but before a discussion of apertures takes place it is important to
consider the affects of diffraction in the microspectrometer.
7.2.4.2 Diffraction
The phenomena arising when radiation encounters a specimen (e.g. refraction, scattering
and absorption) have been discussed earlier. Diffraction is a phenomenon that arises as a
result of passing radiation through the ‘lens’ of the microspectrometer and past the edges
of objects such as apertures and the specimen, and it causes radiation to deviate from its
usually straight line causing blurring of what should be sharp images.
A fundamentally important aspect of diffraction for infrared microspectroscopists is
the point-spread function for the ‘lenses’ in the instrument. If one was to magnify an infi-
nitely small but bright spot of infrared radiation originating at the specimen plane through
the ‘lens’ of a microspectrophotometer, the image at its focal plane is not an infinitely small
spot. Instead it is diffracted into a broad spot of light (called an Airy disc) surrounded by
rings of light; the ‘lens’ cannot produce an image of the point source as a point of light, no
matter how well it is constructed or focussed.
Figure 7.10 shows the intensity of radiation across the image of the diffraction pattern
at the focal plane and the point-spread function (i.e. the intensity distribution across the
plane) is approximately described by Equation 7.2
1.22 × λ
D= (7.2)
NA × n
where λ is the wavelength of light used in the experiment, NA is the numerical aperture of
the ‘lens’, n is the refractive index of the medium through which the specimen is viewed
and D is the diameter of the Airy disc. For mid-infrared radiation of 1000 cm−1 λ = 10 μm,
and if an objective with NA of 0.71 is used in air, then the diameter of the Airy disc is about
17 μm. The maximum intensity of the spot, compared to the point source, is approximately
84%. This means that 16% of the source radiation is diffracted away from the centre of
the pattern by at least 8.5 μm (i.e. the radius of the Airy disc). Nishikida (2004) presents a
more advanced discussion of the point spread function applicable to the ‘lenses’ in infrared
microspectrometers when apertures are in place.
It will be appreciated from Equation 7.2 that the degree of spread of the diffraction pat-
tern is inversely proportional to the numerical aperture of the optical element. This means
that elements with a higher numerical aperture offer the potential for a sharper image.
Furthermore, elements with higher numerical aperture transmit more energy (exponen-
tially so). Although numerical aperture is usually not an instrumental parameter that can
8.5 μm
100%
I
n
t
e
n
s
i
t
y
0 11.5 18.5 μm
Schwarzchild
optical element
Point source
Figure 7.10 Image of point source of radiation of λ = 10 μm transmitted through the objective
of an infrared microscope of numerical aperture 0.71. The curve (top) indicates the intensity
distribution across the rear focal plane. Note the image is not a point but a diffuse spot of
radiation surrounded by bright and dark rings. The central diffuse spot, called the Airy disc
has a radius of about 8.5 μm, rings of light (two are shown) surround the Airy disc with radii of
about 11.5 and 18.5 μm.
be varied from experiment to experiment (unless the microspectrometer is equipped with

a number of objectives), a choice of optics might present itself when the time comes to
purchase equipment. Unlike visible light microscopy, infrared microspectrometers cannot
make use of oil immersion optics in order to increase the refractive index of the viewing
medium (although Chan and Kazarian [2013] report the use of a hemispherical calcium flu-
oride lens on top of transmission cells, which in addition to increasing numerical aperture
has the collateral benefit of eliminating chromatic aberration of the focused infrared beam).
An important feature of diffraction is that it is directly proportional to the wavelength

of the light used for the experiment. As a consequence, the affects of diffraction are much
less pronounced with visible light compared to infrared light, and just considering the
latter the effects are more pronounced at the long-wavelength end of the mid-infrared
range compared to the short-wavelength end. That is to say, diffraction affects are mani-
fested more severely at the right hand side of infrared spectra than at the left hand side.
This can be demonstrated by putting 0.71 for NA into Equation 7.2 and then λ = 14 μm
(approximately 700 cm−1) and λ = 3 μm (approximately 3,300 cm−1); it can be seen that
the point-spread function across the entire mid-infrared wavelength range varies almost
5-fold between approximately 24 μm and 5 μm.
When radiation passes an edge (such as the specimen or an aperture) diffraction also
occurs, causing blurred shadows of the edge rather than a sharp shadow. This is illustrated in
Figure 7.11 by the graph which shows the intensity (`qualitatively) of radiation that falls into the
shadow region compared to the ideal intensity distribution that would be realized in the absence
of diffraction. In the ideal case there should be no radiation reaching the shadow region.
The apertures in a microspectrometer are formed by 4 knife-edges and these will
diffract radiation that passes through the aperture. When the aperture is between the
100%
Radiation
intensity
Figure 7.11 Diffraction pattern produced at a high contrast edge such as an aperture or speci-
men edge. Dotted line (top) shows ideal radiation intensity distribution across edge (ie, edge
image). Solid line (top) indicates actual edge image intensity distribution. Note that there is
significant radiation diffracted into the ‘shadow’ region which should exhibit intensity of zero.
100%
Radiation
intensity
Figure 7.12 Diffraction pattern produced across a slit-like aperture. Dotted line (top) repre-
sents the ideal radiation intensity distribution across the slit (ie, slit image). Solid line indicates
actual image intensity distribution. Note significant radiation is diffracted into the shadow
region which should exhibit intensity of zero.
Beam from IR source
Aperture
Specimen
Figure 7.13 Infrared microscope probe beam resulting from aperture established between
source and specimen.
source and the specimen, the specimen will be illuminated by a somewhat diffuse pattern.
Figure 7.12 illustrates this for a slit-like aperture set up to illuminate a fibre; an estimate of
the spatial intensity of the beam at the specimen is plotted in the graph, together with the
ideal slit profile. Clearly a fraction of the radiation passing through the aperture does not
fall onto the fibre and it will be stray radiation if it reaches the detector.
The edge of the specimen also diffracts radiation. If apertures are placed between the
specimen and the detector and the blades of the aperture are set to coincide with the edges
of the fibre then some radiation from just beyond the edge of the fibre will be bent, pass
through the aperture and also be recorded as stray radiation.
The situation for infrared microspectroscopy of fibres therefore is that even though
the specimen image appears sharp (as far as the observer is concerned, who is of course
receiving visible light) and apertures might be put in place on the edge of the specimen
in order to reject as much stray light as possible, the infrared radiation bends around the
aperture and around the specimen and the image of the fibre through the ‘lens’ is blurred,
so the detector still receives more stray light than is apparent to the operator.
Diffraction originating at the ‘lens’ is out of the control of the spectroscopist, but the
apertures are not; therefore there is some scope for the analyst to minimize diffraction
Beam from IR source
Specimen
Aperture
To detector
Figure 7.14 Infrared microscope probe beam resulting from aperture established between
specimen and detector.
originating there. In an ideal diffraction-free world the situation where apertures are
located at the focal plane between the source and the specimen beam (Figure 7.13) would
be equivalent to the situation where the apertures are located between the specimen and
the detector (Figure 7.14), and furthermore there would appear to be no benefit in using
apertures both before and after the specimen.
As Figure 7.15 shows, this is not the case. This Figure features a stacked plot of an
expanded region of 4 spectra recorded from a single specimen (the flattened Enkron fibre
discussed above) using varying aperture conditions and a table giving the absolute and
relative absorbance values for the peaks at 1020 and 1105 cm−1. The bottom spectrum was
acquired with no apertures at all; this spectrum exhibits the worst-case example of excess
stray light in microspectrometry. The next two spectra (progressing from the bottom of the
figure) were acquired using single apertures between the specimen and the detector and
between the source and the detector, respectively. The top spectrum was acquired with aper-
tures both between the specimen and the source and between the specimen and the detec-
tor. This Figure highlights the inverse relationship between photometric accuracy and stray
light, and the relative efficiency of various aperture configurations. Clearly the performance
obtained using a single aperture between the source and the specimen is better than that
obtained using apertures between the specimen and the detector, and using both apertures
offers some additional improvement. Messerschmidt (1995) has indicated that compared to
single apertures the use of dual apertures allows a reduction in specimen size of about 50%
for equivalent photometric accuracy. Another way to demonstrate the contribution that stray
.5
1150 1100 1050 1000

Absorbance Absorbance Absorbance
1105 cm–1 1020 cm–1 ratio 1105/1020
Top curve 1.01 0.41 2.45
Second top curve 0.86 0.35 2.45
Second bottom curve 0.45 0.21 2.13
Bottom curve 0.06 0.04 1.48
Figure 7.15 (a) Spectra acquired from a single flattened Enkron fibre as in Figure 9 with various
aperture configurations. Bottom curve, no apertures. Second bottom curve, aperture between
specimen and detector. Second top curve, aperture between source and specimen. Top curve,
apertures both between source and specimen and between detector and specimen. (b) Table of
absorbance values for peaks at 1020 and 1105 cm–1 from the four spectra given above. Table also
indicates the ratios of these absorbance values for each spectrum. Notice that in the case of
the bottom two spectra stray light has the apparent effect of pushing the larger peaks into the
baseline, as evidenced by a reduction in the 1105:1020 ratio.
light makes to a measurement is to compare the signal-to-noise ratio for various aperture
arrangements. Nishikida (2004) did this for both single and dual apertures at sizes of 100 ×
100 μm and 7 × 7 μm. For the larger aperture the signal-to-noise ratio increased by a modest
15% when single apertures were used instead of double apertures. In the case of the smaller
aperture, the signal-to-noise ratio increased by 400% when single apertures were used instead
of double apertures. The increased signal achieved is purely a result of radiation from outside
the apertures reaching the detector, that is, an undesirable signal increase due to stray light.
100%
Radiation
intensity
Figure 7.16 Approximate diffraction patterns produced by slit-like apertures of different

width. Top: Aperture spacing (slit width) of at least 5λ (i.e. about 50 μm in the mid infrared
domain) and spatial intensity distribution of radiation across the specimen plane (dotted line
indicates ideal intensity distribution, solid line indicates actual image). Good beam transmis-
sion is evident with only a small amount of diffracted radiation reaching the shadow region.
Middle: Aperture spacing of about 3λ. At this spacing the diffraction patterns produced by each
edge begin to interfere with each other. Transmission by the slit begins to drop off and radia-
tion diffracted into the shadows increases. Bottom: Spacing of about 1λ and below. Radiation
transmission is much less than expected and the percentage of radiation diffracted into the
‘shadow’ region is great.
40
30
20
10
0
3000 2000 1000
Figure 7.17 Spectra recorded through slit-like apertures of spacing 80 × 100mm (top), 40 × 100
(middle), and 20 × 100mm (bottom) ratioed against background spectrum acquired without an
aperture. Notice that in the bottom spectrum absorbance at the long wavelength end is about
50% of that at the high wavelength end. Horizontal axis in wavenumbers, vertical axis in
absorbance.
Another aspect of diffraction is that it rapidly worsens as apertures are brought closer
and closer together; the amount of energy transmitted by the aperture is reduced dispro-
portionately and the percentage of radiation cast into the shadow becomes greater and
greater. This situation is depicted qualitatively in Figure 7.16 and Messerschmidt (1995),
Sommer and Katon (1991), and Nishikida (2004) describe it more comprehensively.
Figure 7.17 demonstrates the phenomenon using a microspectrometer. The spectra
were acquired through apertures of various widths with no specimen present and with a
wide-open aperture for the background. In the absence of diffraction each of the spectra
should be a horizontal line with transmittance offset from 100% by a value proportional
to the amount of light attenuated by the apertures (i.e. if half of the beam is attenuated
the line should be at 50% transmittance). It can be seen that as the apertures get narrower
they transmit much less light than a simple calculation based on their area would predict.
The reason is that as the apertures get narrower a disproportionate amount of radiation is
being diffracted and not reaching the detector. Figure 7.17 also illustrates the phenomenon
that radiation of longer wavelength is affected more by diffraction than radiation of short
wavelength. The slope to the right for each spectrum (which gets progressively worse as the
apertures narrow) is a direct result of the longer wavelengths of radiation suffering greater
diffraction and not reaching the detector.
The phenomenon shown in Figure 7.17 illustrates a point as to spectroscopic technique.
Foreground and background spectra should be recorded with identical aperture configu-
rations. Obviously, diffraction is not eliminated by this practice, but occurs to the same
extent in the foreground and background, therefore baseline slopes and massive transmit-
tance (or absorbance) offsets are avoided. What is not avoided is that a significant amount
of radiation is lost at the long wavelength end of the spectrum in both the foreground and
the background, which causes a reduction in the signal-to-noise ratio.
The discussion so far has dealt with the contribution that diffraction makes to stray
light and its impacts on photometric accuracy; however, diffraction can also lead to ‘spec-
tral leakage’. Spectral leakage is only a problem when the specimen has compositional
boundaries and when apertures are used in order to attempt to isolate a particular region
for analysis. This is of relevance in a number of areas of forensic analysis (examination of
Figure 7.18 Depicting an aperture established to view the white section (region of interest)
of a 3 section, 2 component (blue and white) specimen (components of microscope omitted
for simplicity). Point spread functions (blue curves) associated with 2 points within the blue
components just outside the apertured region of interest are depicted. The point spread func-
tions indicate that some radiation that has passed through the blue sections pass through the
aperture and on to the detector.
paint or plastic laminate cross-sections for example) but in the field of fibre examination
it is really only of relevance to the analysis of bicomponent fibres or fibre cross-sections
that are mounted in a resin support. Figure 7.18 illustrates the origin of spectral leakage.
It was indicated earlier that a point at the specimen plane spreads into an Airy disc and
rings at the aperture. The Figure depicts a side-by-side 3 layered two component specimen
with an aperture established over the central region of interest. Shown are the point spread
functions for two points on the specimen just outside the apertured region of interest.
Due to the point spread of the ‘lens’ some spectral information originating from outside
the region of interest ‘leaks’ through the aperture and reaches the detector. Sommer and
Katon (1991) demonstrate the extent of spectral leakage by successively recording spectra
of air through an 8 μm single slit-like aperture before the specimen and close to the edge
of a sheet of cellulose acetate. It was found that a small signal from the acetate was detected
even when the aperture was established 40 μm from the edge of the acetate. Nishikida
(2004) also illustrates spectral leakage by recording the spectrum of a 25 × 25 μm void
in a sheet of polymer using apertures set at 25 × 25 μm. Spectral leakage was effectively
stopped if apertures were established both before and after the specimen, but not if single
apertures were used. Davis et al. (2010a and b) have written an excellent description of the
interaction between homogeneous and heterogeneous specimens and the infrared beam
emerging from the microspectrometer.
7.2.4.3 Beam Polarization

As described towards the end of Section 1.2.1, the intensity of the infrared beam is not con-
stant across the field of view. With respect to another important property, polarization, the
beam is also not homogeneous. The radiation emanating from the infrared source is free
from polarization, but on traversing the beamsplitter in the interferometer and the various
mirrors in the optical train, the infrared beam acquires a small degree of polarization. This
is of no consequence if we wish to record the spectrum of a specimen that has no crystal-
linity, or no selective orientation of functional groups. However, if there is some crystal-
linity or net orientation of the polymer chain within the specimen, then the orientation of
the specimen within the beam will have some effect upon the spectral data obtained. For a
more complete discussion of the interaction of polarized light with polymers, see the later
section in this chapter. At this point, it is sufficient to understand that absorbance values
of some peaks will change as the orientation of the fibre is changed from being parallel to
the residual polarization of the beam to being perpendicular to it (that is to say, the pho-
tometric accuracy of the spectrum might depend upon the orientation of the fibre within
the beam). Figure 7.19 shows the align signal obtained through an infrared microscope as
an infrared polarizer located near the first focal plane is rotated through 360°. If the beam
had no polarization, the align signal should not change during rotation of the polarizer;
obviously, the signal does vary. The magnitude of the variation will differ from instrument
to instrument. As the exact state of orientation of polymer within an unknown fibre is
also unknown, it is good practice to ensure that one particular orientation of known and
unknown fibres relative to the beam is maintained during analysis, for example, all ori-
ented horizontally across the field of view or all oriented vertically.
Beam polarization will have no affect if the specimen has no net polymer orientation,
or if its orientation is destroyed during specimen preparation. Crushing a specimen, melt-
ing it, or dissolving it in a solvent is likely to destroy any polymer orientation.
7.2.5 Recommended Techniques

If the task before us is to compare two fibres using infrared microspectroscopy then the
message from the above discussion is that steps should be taken to minimize variations in
the level of stray light intruding into our analysis from specimen to specimen. Failure to do
so will result in differences in relative peak heights that might be misinterpreted as genuine
differences between the fibres (Type 1 error) or degrade chemometric processes.
As discussed previously, the shape and thickness of a fibre can have affects upon
the spectral data acquired from it. Furthermore, in order to minimize stray light effects,
7.5
7
Align signal
6.5
5.5
0 40 80 120 160 200 240 280 320 360
Degrees rotation
Figure 7.19 Align signal recorded as a polarizing filter located at the focal plane between the
source and the specimen was rotated through 360°. At 0, 180, and 360° the polarizer axis ran
north south across the specimen plane while at 90 and 270° the polarization axis went east-west
across the specimen plane. This indicates that upon traversing the optical components of the
microscope, the beam develops a small degree of polarization in the north-south direction.
apertures should be used to trim the size of the beam that illuminates the specimen. For
very small specimens a narrow aperture must be used, with the result that aperture trans-
mittance drops off and stray light increases due to diffraction.
Flattening the fibre has many beneficial affects; the irregular or cylindrical cross-
section is destroyed, the specimen thickness will be reduced, and the sample will become
wider, therefore apertures can be set further apart thereby minimizing diffraction-induced
stray light while maximizing the amount of energy transmitted by the apertures. There are
some risks with this sampling strategy, however. If the fibre is flattened into a very uni-
form ribbon, interference fringes will intrude in the spectral data. If the fibre is treated too
harshly it might be turned into an intractable smear that is too thin to be of use. Finally,
as indicated above, crystallinity in the fibre might be altered. Not only will this make it
impossible to conduct infrared dichroism studies (see later) it increases the likelihood for
Type 1 and Type 2 errors under certain circumstances. Nylon 6/6 can be differentiated
from nylon 6 by the presence of a weak band at 935 cm−1, but as shown by Tungol et al.
(1991, 1995a), this band vanishes as pressure is applied to the specimen. It is conceivable
that two identical Nylon 6/6 fibres subjected to different flattening forces could yield dif-
ferent spectral data, with the possible outcome being a Type 1 error. Conversely, under
high flattening forces, Nylon 6/6 and nylon 6 might be indistinguishable; the outcome in
this event is a Type 2 error. Similar logic applies to analysis of poly(ethylene terephthal-
ate) (PET) fibres. The absorbance values of peaks due to methylene deformation modes in
the spectrum of PET (1371 and 1340 cm−1) vary with increasing pressure (see for example
Tungol et al., 1991, 1995a). As a consequence, different flattening forces applied to two
identical PET fibres could result in quite different spectral data in the 1300–1400 cm−1
region, with the possible outcome also being a Type 1 error. In short, the affects of pressure
upon the specimen should be borne in mind whenever a sample is flattened for analysis.
Three recommended techniques for flattening fibres prior to analysis are given below.
7.2.5.1 Diamond Cell

Micro-diamond anvil cells make use of two tiny finely polished diamonds that are used to
flatten specimens such as fibres and other forms of trace evidence prior to infrared micro-
spectroscopy. Even hard fibres, such as nylon that can be difficult to flatten with any other
technique, readily deform in the cell. Diamonds transmit enough IR radiation to permit
them to be used as a transmission window, therefore after the specimen has been flattened
its IR spectrum can be recorded while it is still in situ. The first step in mounting a fibre for
analysis is to place a short section of it centrally across the face of one of the diamonds. This
task requires a steady hand under the stereomicroscope, and finely crafted instruments to
handle the specimen. The second diamond is then brought into contact with the fibre and
then gently pressed in order to flatten it. Experienced operators are capable of sampling
and manipulating 0.5 mm of fibre or less.
On occasions the diamond can acquire a static charge, and it will vigorously repel
fibres. As a precaution it can be wise to discharge the cell faces before attempting to posi-
tion the specimen. A convenient way to do this is to stroke the diamonds with a carbon
fibre brush, such as the type sold to discharge vinyl disc music recordings, or discharge
them with an inexpensive ionizing unit as described by Suzuki and Pettit (1994).
During the flattening process it is beneficial to view the fibre through the diamonds
with the aid of a stereomicroscope using either a transmitted light source or an incident
source directly above the cell. This will allow the analyst to proceed with enough force to
flatten the fibre but not smear it. A recommended approach is to apply only gentle force
to the specimen, then record its spectrum. If the specimen is still too thick, it is a simple
matter to subject it to additional pressings and record additional spectra. In this way a
specimen will not be inadvertently destroyed before any data can be collected, and it is
likely that some information due to crystallinity will be recorded before it is lost forever.
Once the fibre has been flattened adequately its spectrum and background can
be collected, either by recording directly through the cell, or after the two halves have
been separated. When separated it will be found that the fibre sticks to only one diamond,
and it is that half that should be presented to the infrared microscope with the fibre on top.
As a result of its finely polished diamond surfaces the micro-diamond anvil can induce
interference fringing. First, if the background is collected directly through an empty region
of the assembled cell then reflection of the beam from the two internal surfaces of the cell will
cause a subsidiary centreburst spike, as described earlier. This will result in interference fringing
in the background, which will be carried through to the final spectrum. Obviously if the fore-
ground and background spectra are recorded using the disassembled cell this is not an issue.
However, if the fibre has been flattened into a smooth ribbon, interference fringing can still
occur because multiple reflections can originate internally from the surfaces of the fibre itself. If
the cell permits it, a slight rocking of the diamonds can be employed during the flattening oper-
ation. This technique seems to produce irregular films, but care must be taken not to crack the
diamonds (the manufacturer’s instructions should be consulted before trying this technique).
If the spectrum is acquired using an assembled cell, then steps can be taken to minimize
interference fringing by minimizing reflection at the internal diamond surfaces. One way is
to include a small crystal of potassium bromide in the cell, which sinters into a transparent
film while the sample is being pressed. The background should be acquired through the
film of potassium bromide; its refractive index is much higher than air, therefore internal
reflection at the diamond faces (and therefore interference fringing) is reduced. Another
way is to include a micro-drop of Nujol between the diamonds and record the background
through it (see Flynn et al., 2006, for example). Obviously the final spectrum will contain
contributions due to Nujol, which must be removed by spectral subtraction.
Diamond windows have a profound affect on the infrared beam they transmit due
to their high refractive index. If the microscope permits it, before any attempt is made to
record foreground or background spectra any lens that is imaging the specimen through
a diamond window should be adjusted to compensate for the spherical aberration intro-
duced by the diamond. This is accomplished by off-setting the separation between the two
mirrors in the lenses. The exact offset required for a fibre sample on a substrate depends
upon the optics, the refractive index of the substrate (in this case diamond) and the thickness
of the substrate. Instrument manufacturers generally give a guide as to offsets applicable to
various window thicknesses and materials and how to achieve the offset.
Background spectra should be acquired fairly close to the fibre as the thickness of the
diamond window can vary slightly across its area and the frame of cell might vignette the
beam (this is not obvious if the specimen is viewed with reflected light). At the same time,
one must take into consideration the diffraction effects discussed previously. The back-
ground should be acquired sufficiently far enough away from the specimen to ensure that
no radiation can pass through it and reach the detector. With a dual aperture microscope
a distance of about 50 μm will suffice.
It should be noted here that spectra acquired using this single-diamond technique will
often exhibit baselines below zero absorbance units (above 100% transmittance). This is
because the diamond surface has a reflectivity much higher than the specimen. Therefore
the amount of radiation reflected from the diamond during acquisition of the background
exceeds reflection from the specimen during acquisition of the foreground. The result
of this is that less energy reaches the detector in the background spectrum than in the
foreground, therefore the baseline exhibits absorbance below zero.
It is quite important to ensure that both the fibre and the top and bottom surfaces of each
diamond are free of foreign particles as they could contribute signals to the spectrum of the
fibre. The application of a carbon fibre bush can be sufficient to dislodge loose foreign par-
ticles but will not shift debris that is tightly attached to the diamonds or that is trapped on the
back faces of the diamonds, which are usually located in a tiny aperture in the metal backing
plate. Dental micro-applicators, which are miniature nylon flock buds attached to thin and
flexible plastic sticks, can be useful to assist cleaning. Stubborn deposits can be rubbed-off
with a micro-applicator moistened with methanol and then polished with a dry one.
7.2.5.2 Rolled Fibres

There is no doubt that the diamond cell is a very convenient and effective tool. It is, how-
ever, a fairly costly accessory. It is possible to obtain spectra of fibres simply by rolling them
flat and presenting them to the infrared microscope. With microscopic samples this can
sometimes be an awkward process.
Thermo Fisher provides a tool that is convenient for rolling fibres. It is the size and
shape of a pencil, with a small, polished, hardened steel roller attached to one end. Once
a length of fibre has been collected for analysis it is placed onto a hard surface (such as a
microscope slide or a hard halide crystal window) and flattened by running the roller along
the fibre. It is necessary to conduct this operation under a stereomicroscope. Effort must
be made to make sure the roller is kept flat on the hard surface, if just one edge of the roller
is applied the fibre will not be flattened because the roller will not make contact with it. It
is possible to produce a very uniform ribbon if the fibre is rolled on a very smooth surface.
In this event it is very likely that the resulting spectrum will show evidence of interference
fringing. A remedy for this state of affairs is to roll the fibre on a matte surface, such as the
frosted portion of a microscope slide, or to roughen the surface of the roller with a very fine
grade of abrasive paper.
If the fibre has been rolled on a halide crystal, all that remains is to present the sample
to the microscope and acquire the spectrum. If the fibre has been rolled on a microscope
slide, it must be removed from the glass and supported on a substrate such as a halide
crystal. The advantages to this method are that it is inexpensive and a single halide crystal
can hold many specimens, which opens up the possibility for efficient batch processing if
access to an automated motorized microscope stage is available.
7.2.5.3 Setting-up Apertures

From the work of Sommer and Katon (1991) a few broad generalizations can be drawn as
to effective ways to establish apertures. For homogeneous specimens of greater than 30 μm
in width one can adopt the tactic of ‘over-aperturing’ in order to reduce stray light. This
causes most of the beam, even the diffracted components, to pass through the sample and
there is not a major affect on the beam energy. As a consequence the signal-to-noise ratio
is high and stray light is minimal.
For specimens about 30 μm in width it is better to operate with the first aperture set to
the edges of the specimen image, and also put apertures in place between the specimen and
the detector, rather than ‘over-aperture’ a single set of apertures (for some microscopes, it
is not possible to control each set of apertures independently, in which case the apertures
should be set to the edge of the specimen).
If the apertures are brought together to define an object about 10 μm in width, approx-
imately 30% of the beam energy does not impinge upon the specimen. The situation is far
worse for objects less than 10 μm in width. For objects much smaller than 30 μm in width,
two options present themselves, each with their attendant compromises. One is to set all
apertures on the specimen edges, and put up with low energy; the result will be a spectrum
of low signal-to-noise ratio. Alternatively, the apertures can be set wider than the speci-
men, which gives a spectrum with a higher signal-to-noise ratio but the consequences of
higher stray light must be put up with. In any event it is not wise to over-aperture small
specimens.
7.2.5.4 Transmission-Reflection
In this technique, also known as reflection-absorption or double-pass transmission, the
fibre is flattened onto a highly reflective metal substrate or flattened first and then placed
on the substrate. With the infrared microscope operated in the reflection mode the beam
passes through the specimen, reflects off the metal surface, then passes back through the
specimen a second time. Reflective metal substrates range from highly polished strips of
metal to aluminium foil. Gold- or aluminium-coated glass microscope slides specially
produced for infrared spectroscopy can be purchased, but they are expensive. Aluminium
foil applied to a common microscope slide is an inexpensive, functional alternative or one
can easily coat slides with a film of gold using a sputter coater as used for treatment of
specimens prior to scanning electron microscopy.
With reflection-absorption care must be taken to ensure that the specimen is not made
too smooth otherwise interference fringing will result.
This technique is quite convenient, but as with other techniques reliant upon flattening
a single fibre by hand, a certain level of skill is required to flatten the specimen. The tech-
nique is also more successfully applied to certain fibres than to others. With double-pass
transmission the effective specimen path length is doubled, therefore this technique should
not be applied to the analysis of fibres which are difficult to flatten or those with a large
diameter or high absorptivity. A common example would be thick nylon carpet fibres,
which exhibit all three of these characteristics. On the other hand, thin fibres with low
absorptivity, such as acrylics, lend themselves to this technique.
In some microspectrometers the optical train restricts the energy available to reflec-
tance measurements to 50% of that available for transmission. For minute fibres this energy
penalty can inhibit performance.
The main problem with transmission-reflection spectrometry is a host of severe spec-
tral distortions that can arise. These originate from the complex variation of refractive
index of the fibre across the spectral range and as a result of reflection/interference of
the infrared beam from the multiple interfaces that a fibre pressed on a surface present
to the beam. This is made more complicated by the fact that beams strike each interface
at an angle and not perpendicularly. For a number of years a phenomenon called an elec-
tronic standing wave was also thought to be responsible for large and broad absorption-
like features in the spectral data (see Bassan et al., 2013, for example). However, a recent
study by Lee, (2017) has shown that the features are due purely to multiple reflections and
interference.
7.2.6 Micro-ATR Spectroscopy

Internal reflection spectroscopy (IRS), often termed attenuated total reflectance (ATR)
spectroscopy, is a near-surface analysis technique which has been incorporated into spe-
cialized objectives for several infrared microscopes. The term ATR will be used here,
although IRS is the preferred terminology (ASTM, 1981).
Internal reflection occurs when a beam of radiation strikes a high-refractive-index
material at an angle that exceeds the critical angle. This infrared-transparent, high-refrac-
tive-index material is termed the internal reflection element (IRE) or, more simply, the
crystal. All of the beam energy is reflected upon internal reflection. The beam does, how-
ever, penetrate slightly beyond the IRE’s surface in the form of a standing or evanescent
wave. The amplitude of the electric field of this wave decreases exponentially with distance
from the surface of the IRE. When a sample of lower refractive index is placed in contact
with the IRE the evanescent wave penetrates the sample’s surface and can be absorbed or
attenuated. This attenuated radiation, when plotted as a function of wavelength, yields an
absorption spectrum similar to that obtained in normal transmission mode.
The depth of penetration (dp) of the evanescent wave is defined as the point where
the beam intensity drops to 1/e of its original magnitude and can be calculated using
Equation 7.3 (Harrick, 1967):
λ
dp = 1
⎡ ⎛ η ⎞ ⎤2
2
2Π η1 ⎢ sin 2 θ − ⎜ 2 ⎟ ⎥ (7.3)
⎢⎣ ⎝ η1 ⎠ ⎥⎦
where η1 and η2 are the refractive indices of the sample and crystal, respectively, λ is the
wavelength, and θ is the angle of incidence. It is apparent from this equation that the depth
of penetration of the infrared radiation into the sample is dependent upon the refrac-
tive index of the two materials, the wavelength of radiation, and the angle of incidence.
Traditional macro ATR instrument designs typically consist of a trapezoidal IRE through
which the infrared beam is internally reflected many times as it traverses the length of the
element. The sample to be analyzed is placed on one or both sides of the crystal as shown
in Figure 7.20. The effective path length (EPL) is equal to the number of reflections striking
the sample times the depth of penetration.
The ATR microscope objective, on the other hand, is designed around a single-bounce,
hemispherical crystal also shown in Figure 7.20. Although the surface of the element
appears flat in the drawing, it is actually slightly convex so that the center of the crystal
face makes intimate contact with the sample. Because this is a single-bounce design, the
EPL is simply equal to the penetration depth.
The ATR crystal acts like an immersion medium for the ‘lens’. Therefore it substan-
tially reduces the size of the effective beam diameter at the specimen and the functional
size of apertures placed before the specimen by a factor proportional to the refractive index
of the crystal. For example, a 100 μm aperture produces a 42 μm spot at the specimen with
a ZnSe crystal.
Several types of IREs are available for use with the ATR objective. Currently, zinc
selenide (ZnSe), diamond, silicon (Si) and germanium (Ge) are offered. Properties of these
materials are shown in Table 7.1. The choice of IRE material depends on several factors
including cost, durability, penetration depth and ease of use. Diamond and ZnSe IREs are
Sample
IRE
From To
source Sample detector
IRE
Sample
Figure 7.20 Top: trapezoidal IRE typically used for macro samples. Bottom: hemispherical,
single-bounce IRE used in the ATR objective.
Table 7.1 Properties of Various Internal Reflection Elements
Refractive Index Hardness Transparent Effective path length (μm)

IRE (@ 1000 cm−1) (Knoop #) in visible for sample η2 = 1.5
4000 2000 1000 500
cm−1 cm−1 cm−1 cm−1
ZnSe 2.4 150 yes 0.50 1.0 2.0 4.0
Diamond 2.4 – yes 0.50 1.0 2.0 4.0
Silicon 3.4 1150 no 0.21 0.42 0.85 1.7
Germanium 4.0 550 no 0.17 0.33 0.66 1.3
transparent to visible light, thus, they are much easier to align than are opaque materials
such as Si and Ge. Because ATR is a contact technique, it is easy to damage the surface of
the IRE and thereby significantly degrade the performance of the optic. The technique
is ideal for flat, flexible samples. Hard, irregular-shaped samples are apt to damage the
crystal surface. The harder the IRE material, the less likely it is to be damaged during use.
Obviously, diamond is ideal due to its hardness, but it is also the most expensive of the four
choices. The cost, however, should be a one-time expense as the crystal will never have
to be replaced if it is treated properly. Finally, penetration depth may be a consideration.
The higher the refractive index of the material, the shallower the penetration depth. For
example, at 2000 cm−1, a Ge IRE yields an effective penetration depth into the sample of
0.33 μm, whilst ZnSe or diamond IREs yield an analysis depth of 1.0 μm. Garside and
Wyeth (2007) have applied Equation 7.3 to the examination of a typical polymeric fibre
(η2 = 1.6) using a diamond IRE (η1 = 2.4) with a 45° angle of internal reflection and gen-
erated a plot of depth of penetration versus wavelength. Under these circumstances the
depth of penetration is about 1 μm at 4000 cm−1, 1.5 μm at 2000 cm−1, 2 μm at 1500 cm−1,
3.5 μm at 1000 cm−1, and about 4 μm at 700 cm−1. Clearly, the higher the refractive index of
the IRE, the greater the contribution of surface chemistry to the spectrum of a fibre.
As with all techniques, there are advantages and disadvantages to the use of ATR spec-
troscopy for single fibre analysis. The most readily apparent advantage is greatly reduced
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
4000 3000 2000 1500 1000
Figure 7.21 Horizontal axis in wavenumbers, vertical axis in absorbance. Nylon spectra
acquired by ATR from two different locations on a sofa. Top: Fibre from top surface of seat
cushion. Bottom: Fibre from same fabric tacked underneath the sofa frame. Note the addition
bands in the upper spectrum due to surface materials.
sample preparation. With sufficiently hard IREs sample preparation is eliminated. The
fibre sample need only be placed on a suitable substrate for analysis. Glass microscope
slides work well. The back of the slide should be covered with adhesive tape so that if too
much pressure is applied and the glass breaks, fragments will not fall into the condenser
element. Acrylic or polycarbonate sheet cut to the size of a standard microscope slide also
works well and will not break. When using softer crystals such as ZnSe, the fibre should be
flattened slightly first to avoid denting the crystal surface. The lack of sample preparation
offers a distinct advantage for hard fibres such as nylon which can be extremely difficult to
thin sufficiently for transmission analysis (Bartick et al., 1994).
The major analytical distinction between ATR and transmission spectroscopy is the
enhanced contribution of surface chemistry to the ATR spectrum. This can be a major advan-
tage or disadvantage depending upon the circumstances. For surface-coated fibres, such as
those found in many stain-resistant carpets, ATR can yield spectral information on the coat-
ing material. This is a distinct advantage when conducting fibre examinations as it adds yet
another point of comparison to the analysis. Figure 7.21 demonstrates this point with fibres
removed from a sofa. Environmental surface contamination, unfortunately, also is enhanced
in ATR spectra. Thus, two identical fibres can yield different spectra if they have differing
contaminant materials on their surfaces. Whilst this is also the case with transmission spec-
tra, because the fibre material comprises the bulk of the analyzed sample the affect is greatly
reduced often to the point of not being observable. Flynn et al. (2006) also observed that
ATR microspectroscopy was not effective for the analysis of bicomponent fibres. Whilst this
is not surprising in the case of sheath-core bicomponent fibres (i.e. being a surface-sensitive
technique, the beam is not likely to penetrate to the core of the fibre, as indicated by Cho et al.
0.020
0.015
0.010
0.005
0.012
0.010
0.008
0.006
0.004
0.002
0.000
3500 3000 2500 2000 1800 1600 1400 1200 1000 800
Figure 7.22 Acrylonitrile:styrene:butadiene (ABS) rubber spectra acquired by ATR spectros-

copy. Note the higher peak intensities at lower frequency in the original ATR spectrum (top)
compared to the same spectrum corrected to resemble an absorbance spectrum (bottom).
[2001]) it is somewhat surprising in the case of side-by-side construction. Evidently the IRE
was only making contact with one component of the fibre and, surprisingly, this phenom-
enon was repeated over all the fibres studied and at several locations along a given fibre.
As shown in Equation 7.3, penetration depth is directly proportional to wave-
length. Thus at longer wavelengths (lower frequencies) peak intensities are increased
relative to shorter wavelengths in ATR spectra when compared to transmission spectra
(see Figure 7.22). When comparing two spectra, this can be an advantage. Peaks in the
‘fingerprint’ region of the spectrum are enhanced in intensity. This provides a better com-
parison between two fibre spectra particularly in the finer spectral features.
The Christiansen effect induces slight shifts in peak frequencies in ATR spectra, par-
ticularly for strongly absorbing bands. The refractive index of a material changes rapidly
in the region of an absorption with a derivative-shaped appearance centered around the
center of the absorption wavelength. The sharp rise in refractive index may cause η1 to
exceed η2 in this region and the criterion for internal reflection is lost. The net result is a
slight distortion of the peak shape and a shift to lower frequency.
The difference in peak intensities and possibly peak frequencies necessitate careful
interpretation when comparing ATR spectra with transmission spectra. While transmis-
sion reference libraries can be used to identify ATR spectra, ideally an ATR library of fibre
standards should be created by the forensic laboratory.
7.2.7 Infrared Imaging

The conventional way to tackle IR microspectroscopy has involved the analysis of a sin-
gle, small region of interest within a specimen using either transmission optics or an ATR
accessory. Over the last few years an alternative to this ‘single spot’ analytical approach has
arrived that allows simultaneous collection of many spectra from many contiguous locations
in the specimen. A special detector, called a focal plane array (FPA), is responsible for this
advancement. As the name implies, the FPA involves an array of detector elements and these
are arranged in either a strip or a square pattern. In effect, a microspectrometer equipped with
an FPA achieves the same outcome as if a conventional microspectrometer were to record an
array of spectra at contiguous locations in a specimen. However, the FPA achieves the result
in one ‘scan’ where a conventional spectrometer would require many scans with movement
of the specimen between each scan and achieves higher spatial resolution than a conventional
microspectrometer. The sequential approach using a conventional microspectrometer is called
‘spectral mapping’ whereas the use of an FPA produces a ‘spectral image’. Both mapping and
imaging have been used to characterize the spatial composition of specimens, including poly-
mers (see Bhargava et al., 2000, for example), but the ease, speed, and spectral resolution of
spectral imaging ensure that it will make significant contributions to materials analysis.
Spectral imaging has been conducted in either transmission or ATR modes. In the case
of transmission imaging, spatial resolution down to about 10 μ is achievable with appropri-
ate aperturing while with ATR the high refractive index of the crystal allows even better
resolution; in the latter case the performance of the instrument has been reported to rival
the spatial resolution achieved using a synchrotron source and conventional microspec-
trometer (Chan et al., 2005).
The strength of spectral imaging lies in the capability for the analysis of inhomoge-
neous specimens. As fibres are usually homogeneous there have been few applications of
spectral imaging to forensic fibre examination reported so far. Flynn et al. (2006) examined
the application of spectral imaging to bicomponent fibres. While there have been success-
ful attempts to identify bicomponent fibres using conventional transmission (Tungol et al.,
1991) and a combination of transmission and ATR IR microspectroscopy (Cho et al., 2001),
both of these methods rely upon the analyst’s being aware that the fibre has a bicomponent
structure in the first place, which is a serious drawback because the structure is often not
obvious. Flynn et al. (2006) reported that transmission spectral imaging allowed simple,
rapid, and clear definition of the spatial arrangement of the two components (i.e. side-by-
side or sheath and core) in 6 out of the 11 types of fibres examined. In order to be success-
ful, the two components must exhibit sufficient spectral differences and the analyst must
inspect carefully all the spectra acquired across the fibre for evidence of compositional
changes. In several instances fibres seemed to vary between single component and bicom-
ponent along the length of the fibre. Flynn et al. also reported that spectral imaging in
ATR mode can be used to identify bicomponent if they are examined as transverse cross-
sections, although the method was described as being tedious. In a similar vein a Perkin-
Elmer application note (2006) indicates that the very high spatial resolution achievable
using ATR IR microspectroscopical imaging allows differentiation of the spectral proper-
ties of the medulla, cortex and cuticle in human hair cross-sections.
7.2.7.1 Polarized Infrared Microspectroscopy

The most effective interaction between a functional group and infrared radiation (i.e. that
which leads to the most effective absorption of energy) takes place when the electrical vector
of the radiation aligns with a dipole moment present in the functional group. In most cases
of infrared microspectroscopical analysis of fibres this phenomenon has no impact because
electrical vectors point in all directions relative to the fibre and the functional groups of the
fibre also are oriented randomly relative to the fibre axis. However, when both are not ran-
domly oriented there is a noticeable affect. This takes place when polarized radiation is used
and the specimen has some defined molecular orientation, which usually arises when fibres
are stretched (or drawn) after extrusion. The greater the draw ratio the greater the amount
of orientation of polymer chains along the axis of the fibre and therefore the greater the dif-
ference between absorbance of an infrared beam that has its polarization aligned with the
axis of the fibre compared to a beam that is aligned perpendicular to it. In the case of poly-
acrylonitrile, for example, stretching of fibres results in the carbon backbone of the polymer
becoming aligned with the axis of the fibre and the nitrile groups becoming aligned across
the axis of the fibre. The result is that the peak at 2244 cm−1 in the IR spectrum of the fibre
is much more intense when the spectrum is recorded with beam polarization perpendicular
to the axis compared to the situation when the polarization is aligned with the axis of the
fibre (Cho et al., 1999a). PET also becomes aligned under tension, with the ethylene gly-
col residues rotating into a trans conformation that results in the C-O stretch at 973 cm−1
more strongly absorbing parallel polarized radiation compared to perpendicularly polar-
ized radiation (Cho et al., 1999a). On the other hand, this results in alignment of certain
aromatic C-H bonds across the axis of the fibre, with the result that the aromatic C-H out-
of-plane bending absorption peak at 876 cm−1 is stronger when the beam is perpendicular
to the fibre. Cho et al. (1999a) also indicate that certain bands in the spectrum of Nylon 6
exhibit infrared dichroism (3082 cm−1 and 997 cm−1). Some plant fibres have natural struc-
tural orientation of micro-crystalline cellulose wound in a helical pattern either clockwise
or anticlockwise around the axis of the fibre. Under these circumstances the preferred ori-
entation is at some positive or negative angle to the axis of the fibre rather than exactly par-
allel or perpendicular; the magnitude of this angle relates to how tightly or loosely the helix
is wound. Proteinaceous fibres such as silk also exhibit some level of natural orientation.
Polarized beam experiments therefore allow infrared spectroscopy to probe polymer
chain conformation as well as composition. The relevance of this phenomenon is that
the use of polarized beam experiments might allow discrimination beyond that which is
achievable using standard beam techniques.
Examples of parallel and perpendicular polarized spectra for two PET fibres with dif-
ferent manufacturing processes are shown in Figure 7.23. The two fibres obviously differ
in the degree of orientation. The differences in peak intensities between parallel and per-
pendicular polarized spectra for the first fibre indicate a high degree of orientation of the
polymer chains within the fibre, while the similarities between parallel and perpendicular
polarized spectra for the second fibre indicate a fairly random orientation with little order.
This is the result of the first fibre’ having been drawn during the manufacturing process,
while the second fibre, produced for melt bonding, was not.
It is possible to make use of infrared dichroism in a purely qualitative sense as described
above, but it is more useful to calculate a dichroic ratio with respect to particular spectral
bands. The simplest approach, which is applicable to manufactured fibres that exhibit ori-
entation along the axis of the fibres, is that given by Equation 7.4:
A
R= (7.4)
A⊥
where A// and A┴ are the absorbances for a particular absorption band obtained with par-
allel- and perpendicular-polarized radiation. The experimentally obtained dichroic ratio
Parallel
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
2.2 Perpendicular
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
2000 1800 1600 1400 1200 1000 800
2.2 Parallel
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
2.2 Perpendicular
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
2000 1800 1600 1400 1200 1000 800
Figure 7.23 Horizontal axis in wavenumbers, vertical axis in absorbance. Parallel and per-
pendicular polarized spectra for two poly(ethylene terephthalate) (PET) fibres. Over-absorbing
bands have been truncated for clarity. TOP: Fibre with a high degree of orientation. BOTTOM:
Fibre with a low degree of orientation.
will differ from the actual physico-structural ratio due to several factors including overlap-
ping absorption bands, scattering and reflection losses, polarization scrambling, and stray
light. Beam convergence in high-numerical-aperture objectives will also affect the experi-
mentally determined dichroic ratio for a single fibre (Fraser, 1953; Quynn, 1954). Even
taking all these factors into account, it is possible to obtain reasonable polarization data
from single fibres using infrared microspectroscopy. Stray light losses tend to cancel out
in the calculation of dichroic ratio (Chase, 1988) and scattering and reflection losses have
also been demonstrated to be minimal, presumably for the same reason (Tungol, 1995b).
Polarization scrambling in infrared microscopes has been shown to be minimal (Chase,
1987, 1988; Tungol et al., 1995b), and finally, the correction for beam convergence has been
shown to be less than 10% (Tungol et al., 1995b).
Although infrared microspectroscopy has been demonstrated to yield valid polariza-
tion data for single fibres (Chase, 1987, 1988; Krishnan, 1984; Young, 1988; Church, 1991;
Cho et al., 1999a), few forensic studies have been reported in this area. Tungol et al. (1995b)
examined dichroic ratios for 11 types of PET fibres. Repetitive analyses yielded relative
standard deviations (RSDs) on the order of 5% and no significant variation was noted
between different single fibres of the same type. Certain fibres types were differentiated
and grouped by manufacturer using dichroism of four absorption bands. More recently,
Cho et al. (1999b) has classified 32 polyester fibre samples into 13 groups based on dis-
criminant analysis of infrared dichroic spectra.
Although Garside and Wyeth (2007) reported the analysis of cellulosic textile fibres
(ramie, sisal, high tenacity rayon and low tenacity rayon) for conservation purposes, their
work is of relevance to forensic fibre examiners. They applied both transmission and ATR IR
microspectroscopy and found the latter to be more beneficial with regards to discrimination
between ramie (which exhibits a maximal dichroic ratio when the polarization angle of the
IR beam is –7.5° to the axis of the fibre as a result of an anticlockwise helix of cellulose fibrils)
and sisal (which exhibits a maximal dichroic ratio at 22.5° as a result of a clockwise helix). It is
obviously important for cellulosic fibres therefore to record absorbance of peaks at a number
of angles between parallel to the fibres and the two perpendicular directions. A particular
angle to the fibre axis will be found where the ratio is at a maximum; not only can different
fibres exhibit different ratios, they can exhibit different angles at which the ratio is maximized.
Garside and Wyeth concluded that because the majority of the cellulose orientation
lies just under the surface of the fibre in the secondary cell wall, and because the structure
winds completely around the fibre, ATR allows just one part of the secondary cell wall to
be measured without contribution from the opposite side (which of course has cellulose
aligned in an opposite sense). As transmission spectroscopy records absorbance from both
‘front’ and ‘back’ of cell walls the polarization signals from both are convoluted to some
extent. Garside and Wyeth also found that ATR microspectroscopy also allowed much
better discrimination between high tenacity rayon and low tenacity rayon than trans-
mission microspectroscopy. They postulated that this observation is evidence for a high
level of surface orientation in high-tenacity rayon compared to low-tenacity rayon (i.e. in
ATR microspectroscopy that part of the specimen probed is almost completely oriented,
therefore the dichroic ratio is high, while in transmission microspectroscopy a significant
fraction of the absorbance arises from unoriented polymer in the centre of the fibre, there-
fore the dichroic ratio is ‘diluted’).
7.2.8 IR Spectrum Interpretation

Fibre characterization can be achieved in two ways: by computer-assisted processes; or by
skilled interpretation of infrared spectral features (‘first principles’). The proof of any iden-
tification conjecture, however, is when spectra of a standard fibre and those that might be
confused with it are compared with the spectrum of the unknown.
7.2.8.1 Computer-Assisted Processes

In addition to database searching, which is a trivial exercise nowadays assuming compre-
hensive databases are available, computers attached to infrared spectrometers are powerful
enough to perform a variety of mathematical manipulations on spectral data. One tech-
nique that is finding increasing application to forensic fibre comparison is chemometric
analysis of spectra. Chemometrics is a mathematical process that allows correlations
between patterns within spectral data and membership of classes to be identified. As an
example, Gilbert et al. (1993), using diffuse reflectance infrared spectroscopy and chemo-
metric methods, were able to differentiate cotton fabrics (bleached and raw voile, raw and
bleached muslin, and raw poplin), and the level of processing to which they had been sub-
jected (raw, singed, bleached, mercerized, and optically brightened). In order to achieve
this level of sophistication, the authors had to ‘train’ the software with a calibration data-
base of 96 infrared spectra from which consistent principal spectral components for each
of the fabric types and processes were extracted.
It would appear that chemometric techniques do have application in forensic fibre
analysis as a special tool to extend conventional generic and sub-generic classification.
However, chemometrics does require operators to minimize spectral artefacts that would
confuse the algorithm, it requires a large number of samples to train the database and
calibration databases are instrument specific (i.e. sharing databases is not possible). The
technique is also relatively complicated to explain to a jury. As a consequence, chemomet-
rics has not become a common practice in forensic science.
7.2.8.2 Characterization Using ‘First Principles’

Usage of the following first principles guidelines and associated flow diagrams should
enable characterization of most fibres. The flow diagrams have been updated from the pre-
vious edition of this chapter; the earlier versions have been adopted and re-published by
the Scientific Working Group for Materials Analysis (SWGMAT).
Included in the flow diagrams are references to articles that discuss techniques other
than infrared microspectroscopy that can yield further discrimination within a class
determined by infrared microspectroscopy.
Two key diagnostic features in the infrared spectra of fibres relate to the presence (or
absence) of peaks arising from carbonyl and nitrile moieties, which can be found in the
range 1500–1800 cm−1 and 2245 cm−1, respectively. The following sections are broken down
into the four possible combinations of nitrile present/absent and carbonyl present/absent.
7.2.8.2.1 Absence of a Nitrile Absorption and Absence of a Carbonyl Absorption The

simplest interpretative situation is when the infrared spectrum of a fibre does not con-
tain either a nitrile or carbonyl absorbance. This indicates that the fibre could be cellulose
(natural plant-based, or manufactured via the lyocel or the xanthate-processes), polyolefin
(polyethylene, polypropylene), saran, fluorocarbon (‘Teflon’) or polybenzimadazole (PBI).
Assuming microscopy has eliminated the possibility of natural fibres, confirmation of
manufactured cellulose fibres includes the presence of strong C-O and O-H stretches at
1067/1027 and 3300 cm−1, respectively. Carrillo et al. (2004) and Brinsko (2010) describe
the IR discrimination of viscose fibres (derived from pulped wood, bamboo, etc., dissolved
in alkaline carbon disulphide and extruded into a coagulating bath— the so-called
xanthate process and lyocel fibres (also derived from cellulose products, but alginate
and chitin can also be included—the lyocell process involves dissolution of the polymer in
N-methylmorpholine oxide before extrusion). Lyocel fibres have a high degree of crystallin-
ity and orientation whereas the viscose fibres are amorphous, which results in a better defi-
nition of bands at 1335, 1315, and 1277 cm−1 in lyocel compared to rayon and modal and a
lower 1420/893 absorbance ratio for lyocel. SeaCell (a cellulose-alginate lyocel-process fibre)
has been found by Brinsko (2010) to be indistinguishable from other lyocels, but chitin-cel-
lulose lyocel fibres were found to exhibit a broad shoulder at 1500–1600 cm−1. As discussed
above, Garside and Wyeth (2007) have indicated that dichroic ratios calculated using ATR
microspectroscopy can be used to discriminate high tenacity rayon from other forms.
Spectra of polyethylene and polypropylene fibres are dominated by strong peaks due to
C–H stretches at 2800–3000 cm−1. The main structural difference between polypropylene
and polyethylene is that the former has an abundance of pendant methyl (CH3) groups
attached to a methylene (–CH2-) backbone, whereas the latter is almost entirely composed
of methylene units. Methyl groups undergo a symmetrical deformation, which is mani-
fested in the infrared spectrum as a peak at about 1375 cm−1. The corresponding deforma-
tion for methylene groups results in a peak at about 1470 cm−1. In polyethylene, which has
methyl groups only at the ends of polymer chains, the 1470 cm−1 peak is of greater intensity
than the peak at 1375 cm−1. In polypropylene however, the peak at 1375 cm−1 is of greater
intensity that that at 1470 cm−1. Polyethylene also exhibits a small peak at 720 cm−1 which
arises from a rocking vibration of a chain of methylene groups. As polypropylene does not
contain adjacent methylene groups, its spectrum does not show this peak.
Spectra of polybenzimadazole fibres are dominated by strong N–H stretches at
3500 cm−1 (these peaks are much stronger than the corresponding bands in nylons or pro-
tein fibres) and have additional distinguishing bands at 1630, 1460, and 1050 cm−1.
Halogenated polymers (fluorocarbons and sarans) are characterized by the presence
of dominant peaks in the 1000–1200 cm−1 region; peaks due to CH vibrations are small,
if present at all. Sarans show a strong doublet at 1070/1048; Teflon™ shows a doublet at
1212/1155 cm−1.
The above information can be navigated by use of the flow chart in Figure 7.24.
7.2.8.2.2 Presence of a Nitrile, Absence of Carbonyl The presence of a peak at 2245 cm−1
indicates that the fibre is based upon polyacrylonitrile (PAN) and it belongs to the acrylic or
modacrylic class. By definition, acrylic fibres contain more than 85% PAN, whereas mod-
acrylic fibres contain less than 85% but more than 35% PAN. Three types of comonomers
are blended with PAN to make acrylic and modacrylic fibres: these are halogen-contain-
ing comonomers (added to make the fibre flame retardant); neutral comonomers (which
are added to control shrinkage, allow penetration of disperse dyes, and modify solubility
for spinning); and ionic comonomers (added to introduce attraction for ionic dyes) (Kim,
2009). The presence of vinyl chloride, vinylidene chloride, or vinyl bromide as comonomer
indicates that the fibre is a modacrylic. Grieve (1995) and Grieve and Griffin (1999) have
shown how it is possible to differentiate further within these classes of fibres using char-
acteristic IR absorptions of various comonomers and other additives and manufacturing
residues; the ensuing discussion is based on the information in these two articles by Grieve.
The IR spectra of both vinyl chloride and vinylidene chloride feature strong bands
C-Cl stretching bands at lower than 700 cm−1, which will not be visible if a narrow band
MCT detector is employed. The only C-Cl band visible from these copolymers above
700 cm−1 is the peak at about 710–717 cm−1 for vinylidene chloride, which can be con-
firmed by the presence of a prominent C-C stretching absorption at about 1070–1080 cm−1
and a C-H stretch between 1216–1225 cm−1. The peak at 1070–1080 cm−1 is absent in vinyl
chloride-containing modacrylics; instead, two C-H stretching peaks are obvious at about
1240–1260 cm−1 and about 1330–1380 cm−1 are present. See Figure 7.24 for a summary of
the above in a flow chart form.
7.2.8.2.3 Presence of a Nitrile and Carbonyl As indicated above, if PAN is present (i.e.
a peak is present at 2245 cm−1) but vinylidene chloride or vinyl chloride can be ruled out
then the fibre is an acrylic. A very important feature is the presence or absence of a domi-
nant carbonyl (C = O) stretch found in the region 1500–1750 cm−1, which indicates the
presence of an ester-, amide- or pyrrolidone-containing copolymer.
In regards to amide-containing acrylics, some include either N-methylolacrylamide or
acrylamide copolymers. The principle carbonyl stretches for these amides are found at quite
low frequency (about 1680–1684 cm−1). In acrylamide a second band is found at about 1613
cm−1, and as the amide moiety includes an N–H bond it gives rise to a strong peak at about
3365 cm−1. The presence of N-methylolacrylamide can be confirmed by the presence of a C–N
stretch at 1530 cm−1. Methacrylamide comonomer has been found in modacrylic fibres, and
it gives rise to a C = O stretch at 1650 cm−1 and 1546 cm−1. A related comonomer is polyvi-
nylpyrrolidone that is unusual in that it features a tertiary cyclic amide (lactam); its presence
is indicated by a strong C = O single stretch at a frequency lower than the amide peak of
the acrylamides (i.e. at about 1670 cm−1). In addition to carbonyl-containing comonomers,
Carbonyl absorption absent
Both C O and O H stretches Cellulose fibre (lyocel or xanthate process)

present? (1067/1027 and 3300) YES discriminate by another technique, see Garside and
NO Wyeth (2007), Brinsko (2010) and Carillo (2004)
Strong 1260 present confirmed by presence of 1330–
CN stretch present? YES
1380? (YES PAN/VC modacrylic)
(2245) Additional discrimination provided by
NO presence/absence of 1030 (alumina hydrate), 737
(antimony trioxide), or 1038 and 1003–1011 (styrene
sulphonate)
NO
1070–1080 confirmed by presence of 1216–1225 and
710–717? (YES PAN/VDC modacrylic)
Additional discrimination provided presence or
absence of 1670 (DMF), or antimony trioxide, or
styrene sulphonate, as above
NO
PAN homopolymer acrylic

C H stretch present? YES 1470 intensity > 1375? (YES Polyethylene)
(2800–3000)
NO
NO 1470 intensity < 1375? (YES Polypropylene)
N H stretch present? YES 1630, 1460, and 1050 present? (YES PBI)
(3500)
NO
Strong peaks in the 1000–1200 YES Strong doublet at 1070/1048? (YES Sarans)
region? NO
Strong doublet at 1212/1155? (YES Teflon)
Figure 7.24 Flow chart for polymer identification from first principles. PAN = polyacryloni-
trile, VC = vinyl chloride, VDC = vinylidene chloride.
some acrylic fibres include methylvinylpyridine in order to accept acid dyes more effectively.
The presence of methylvinylpyridine is indicated by a small but sharp peaks at about 1492,
738, and 1602 cm−1. Another point of differentiation between fibres of both the acrylic and
modacrylic classes is the presence of non-polymeric additives. N, N-dimethylformamide
Strong carbonyl absorption present
Carbonyl below YES Sharp peak at 2244 YES Carbonyl at 1680–1684 and 1530?
1700? present? (YES NMAA modified acrylic, check
NO for presence of MA)
NO NO
Carbonyl at 1680–1684, 1613, and 3365
present? (YES AA modified acrylic)
NO
Carbonyl at 1650 and 1546?
(YES MAA modified modacrylic, check
for presence of MA)
NO
Carbonyl at 1670 present? (YES PVP
modified acrylic)
Weak C H stretch YES Carbonyl at 1656, 1608, 1536, and 1487
present? present? (YES Nomex)
(2800-3000) NO
NO Carbonyl at 1648, 1540, 1515, and 1407
present? (YES Kevlar)
NO
Protein fibres (silk, azlon, differentiate
using another technique, see Brinsko
(2010))
Weak 935 present ? YES 720 and single, dominant peak at 1240
(care!) present? (YES Nylon 6/10)
NO
NO 720 and two maxima at 1275 and 1235
present? (YES Nylon 6/12)
NO
720 and a single, dominant peak at 1280
present? (YES Nylon 11)
NO
720 and a single, dominant peak at 1270
NO
720 absent and dominant peak at 1210
NO
720 absent and dominant peak at 1274
present? (YES Nylon 6/6)
Prominent peaks at YES Qiana

900 and 960
present? NO Nylon 6
Figure 7.25 Flow chart for polymer identification from first principles. PVP = polyvinylpyrro-
lidinone, PET = polyethyleneterephthalate, TA = terephthalic acid, E.G. = ethylene glycol, PHBA
= p-hydroxybenzoic acid, PBT = polybutyleneterephthalate, NMAA = N-methylolacrylamide
(also known as N-(hydroxymethyl)acrylamide), MAA = methacrylamide, AA = acrylamide,
DMF = dimethylformamide, MVP = methylvinyl pyridine, VA = vinyl acetate, MA = methac-
rylate, MMA = methylmethacrylate.
(Continued)
Carbonyl between YES Sharp peak at 2244 YES 1220 (major) and 1130 peaks present?
1700 and 1750 present? (YES PAN/MMA) additional
NO discrimination provided by
NO presence/absence of 1670 (DMF). Check
for presence of MA or VA (see below).
NO
1170 (major) and 1204, 1229, and 1250
present? (YES PAN/MA, check for
presence of MAA as above) additional
discrimination provided by presence
of DMF, 1492, 738, and 1602 (MVP), 1040
with either 1000–1011 or 1667 (various
sulphonates) or 1590 (itaconic acid).
NO
1235 (major) and 940 present?
(YES PAN/VA) additional
discrimination provided by
presence/absence of MVP
Weak 903, principle YES Cellulose acetate, differentiate further
C O stretch at 1235, using another technique
1370 and > 1430
present?
NO
principle C O YES strong peak at 1605 and 767? (YES PHEB)
absorption at 1240– NO
1270, aromatic prominent peaks at 1408 and 1339 present?
absorption at 730– (YESPET or TA/EG/PHBA copolymer).
770 present? Refer to Cho (1999b) and Causin (2004) for
further discrimination with the PET class
NO
1408 only? (PBT)
Carbonyl above YES Polylactic acid. Refer to Brinsko (2010) for further discrimination
1750?
Figure 7.25 (Continued) Flow chart for polymer identification from first principles.
PVP = polyvinylpyrrolidinone, PET = polyethyleneterephthalate, TA = terephthalic acid, E.G. =
ethylene glycol, PHBA = p-hydroxybenzoic acid, PBT = polybutyleneterephthalate, NMAA =
N-methylolacrylamide (also known as N-(hydroxymethyl)acrylamide), MAA = meth-
acrylamide, AA = acrylamide, DMF = dimethylformamide, MVP = methylvinyl pyridine,
VA = vinyl acetate, MA = methacrylate, MMA = methylmethacrylate.
(DMF) is an important solvent that is associated with the production of fibres. It results
in a sharp peak at 1670 cm−1 accompanied by other, smaller peaks at 1400, 1380, and 1090
cm−1. The relative intensity of the peak due to DMF is highly variable as it is a volatile solvent
that gradually evaporates from the fibre over a long period of time after spinning. Acetone
and ethylene carbonate and are used as solvents and residues of them can be detected by
characteristic peaks at 1708–1710 cm−1 and at 1785 and 1805 cm–1, respectively. Aromatic or
aliphatic sulphonates are blended into fibres if uptake of basic dyes is required. A peak com-
mon to sulphonates can be found at about 1040 cm−1, with an additional peak at about 1000–
1011 cm−1 due to styrene sulphonate, and in an undisclosed sulphonate, 1667 cm–1. Itaconic
acid is another additive used to facilitate the uptake of basic dyes; its presence is indicated by
a peak at 1590 cm−1. Antimony trioxide and alumina trihydrate, both flame retardants, can
be identified by the presence of strong peaks at about 737 cm−1 and 1030 cm−1, respectively.
For ester-containing acrylics differentiation can be achieved by an examination of
absorptions in the C–O stretch region (1100–1300 cm−1). The presence of methylmethacrylate
is indicated by a characteristic sharp C–O stretch at 1127–1130 cm−1 and another peak at
about 1170 cm−1 that is seen as a shoulder on a broader and stronger absorption at about 1220
cm−1, the presence of vinyl acetate can be confirmed by the presence of a strong C–O absorp-
tion at 1235 cm−1 accompanied by a small peak at 940 cm−1, whereas the presence of methyl-
acrylate is indicated by a sharp peak at 1170 cm−1 together with small but characteristic triplet
at 1204, 1229, and 1250 cm−1. See Figure 7.25 for a summary of the above in flow chart form.
Clearly the acrylic and modacrylic fibre classes can be discriminated to a substantial
extent on the basis of the presence and absence of various additives and copolymers. Causin
et al. (2005) and Tungol et al. (1993) have enhanced the discrimination process further by
using the quantitative information present in spectral data to reach some level of discrimina-
tion within a given group. Tungol showed that the relative proportion of polyacrylonitrile to
carbonyl-containing comonomers (as indicated by the ratio of absorbances at 2240 cm−1 and
1730 cm−1) to be useful. Causin extended the method of Tungol by plotting the 2240/1730
absorbance ratio of a range of fibres against their 1730/1370 ratio (which measures the relative
proportions of carbonyl-containing monomer to total methylene content) and was able to
demonstrate that several fibres of the same class were separated in the two-dimensional space.
7.2.8.2.4 Absence of a Nitrile and Presence of a Carbonyl If the carbonyl functionality
is found below 1700 cm−1, the fibre is likely to be a polyamide of some type or one of the
amide-modified acrylics discussed above. This section discusses those fibres that do not
contain PAN and therefore do not exhibit a peak at 2244 cm−1. As indicated above, amides
yield distinctive infrared spectra dominated by the so-called amide I and amide II bands,
due to C = O stretch and NH deformations, respectively. The amide I and II bands form a
strong doublet at about 1640 and 1530 cm−1.
The most important non-PAN polyamide-containing fibres are the nylons, protein-
aceous fibres (silk, wool and azlons), and aramids (Nomex or Kevlar). From its microscopi-
cal appearance, it should be quite obvious if the specimen is wool or other hair fibre, and
these will not be discussed further.
Spectra of azlons, silk, and aramids are characterized by very low intensity CH stretch
absorption bands (2800–3000 cm−1). Further differentiation of silk from azlons can be
made by very careful comparison against spectral data from a silk reference fibre, on the
basis of its microscopical appearance and transverse cross-section, and its high refractive
index and birefringence. Two new azlons Soy Silk (made from soy protein) and Silk Latte
(made from milk protein) have been described by Brinsko (2010). Using infrared spectros-
copy they can be differentiated from earlier azlons and silk.
The aramids are easy to recognise and differentiate; Nomex shows four strong peaks
at 1656, 1608, 1536, and 1487 cm−1, while Kevlar shows strong peaks at 1648, 1540, 1515,
and 1407 cm−1.
In order to differentiate nylons, it is important to carefully examine the data to establish
whether a peak at about 935 cm−1 is present; if so, the fibre is of the nylon 4, 6/6, 6/10, 6/12,
11 or 12 type. The (genuine) absence of this peak indicates Qiana or nylon 6 (as described
above, it is important to ascertain whether the absence is genuine, or as a result of too much
pressure upon the sample). Two prominent peaks at 900 and 960 cm−1 indicates Qiana.
In order to resolve the six nylons that show a 935 peak, the region below 1300 cm−1
must be carefully looked at. Nylon 4 and 6/6 do not show prominent peaks near 720 cm−1,
there is a dominant peak in the spectrum of nylon 6/6 at 1274, while the dominant peak is at
1210 cm−1 for nylon 4. Nylon 6/10, 6/12, 11 and 12 all show a peak near 720 cm−1. However,
nylon 6/10 shows only a single dominant peak in the 1300–1200 range at 1240 cm−1, nylon
12 shows a single peak at 1270, nylon 11 a single peak at 1280, while nylon 6/12 shows two
maxima at 1275 and 1235 cm−1.
The presence of a carbonyl absorption band above 1700 cm−1 indicates polyester or cel-
lulose acetate fibres. Further differentiation of these fibres can be achieved by examination
of the most intense band in the C-O stretch region (1100–1300 cm−1) in particular, and
other specific features.
Cellulose acetate is very easy to recognize due to the characteristics of the acetate moe-
ity, which gives rise to a small peak at 903 cm−1, and a principle C-O stretch moved to quite
high frequency (1235 cm−1) by hyperconjugation. Furthermore, due to the abundance of
methyl groups in the structure, acetate fibres show a methyl deformation band (1370 cm−1)
that is stronger than the methylene band (1430 cm−1) as described above for polypropylene;
in other fibre-forming polymers the methylene band is stronger than the methyl band.
The C-O bond in aromatic acid-based polyesters (PET, etc) is strengthened due to con-
jugation with an aromatic ring, therefore the strongest absorbance in the C-O range is
found at quite high frequency (1240–1270 cm−1). Furthermore, the para-disubstituted aro-
matic residues in these polyesters give rise to characteristic peaks at about 730 cm−1. Some
discrimination within the PET class has been reported. Causin et al. (2004) based discrim-
ination upon the ratio of trans- to gauche-PET, which was measured using absorbances at
846 cm−1 and 1340 cm−1 (due to trans-PET) and absorbances at 896 cm−1 and 1370 cm−1
(gauche-PET). As indicated above, the examination of infrared dichroic ratios has been
shown to be of use in the further discrimination of fibres found to be in the PET class. Cho
et al. (1999b) classified 32 PET fibre samples into 13 groups based on discriminant anal-
ysis of infrared dichroic spectra. When fibre diameters and cross-sectional shapes were
included, 22 unique groups and five paired groups were identified. Other polyester fibres
are readily discriminated from PET.
Polylactic acid fibres are readily distinguished by the presence of a carbonyl absorption
at very high frequency in their infrared spectra; see Brinsko (2010) for a discussion regard-
ing further discrimination within this group. See Figure 7.25 for a summary of the above
information in flow chart form.
7.2.8.2.5 Dyes and Pigments Peaks arising from pigments or dyes are not likely to be
obvious in infrared spectra of highly coloured fibres, and are likely to be below the limit of
detection for pale coloured fibres. Due to the weakness of the signal arising from pigmenta-
tion, usually only one or two peaks from the pigment are distinguishable from the larger
spectral contribution of the fibre. Usually these peaks are distinguished only after detailed
comparison of the spectral data of the fibre with data from an unpigmented reference. In gen-
eral terms, pigments likely to be encountered have aromatic or azo (-N = N-) moieties present
in their molecular structure. Both of these functionalities are only moderately active infrared
chromophores, but fortunately they do give rise to sharp peaks, not broad absorptions, there-
fore the full limits of detection of IR spectroscopy can be exploited, such as they are.
Grieve et al. (1998) has published spectral data relating common dyes that could be
encountered during examination of acrylic fibres. Characteristic peaks of various pigments
were observed at 1600–1602, 1598, 1586, 1563, 1561, 1557, 1530, 1520, and 1152 cm−1. It is
possible to confuse peaks arising from atmospheric water vapour with peaks due to pigment.
Furthermore, peaks due to amides, lactams, and aromatic compounds, which are often pres-
ent at relatively high levels in fibre, absorb strongly in this region and can mask the pigment.
The selection rules for IR absorption and Raman emission are complementary, with
the outcome being that Raman spectroscopy is more sensitive to non-polar compounds
such as pigments and less sensitive to polar compounds such as polymers compared to IR
spectroscopy. The reader is referred to the sections dealing with Raman spectroscopy for
additional information regarding fibre discrimination based upon their dyes and pigments.
7.3 Conclusions
As indicated at the beginning of this chapter, infrared microspectroscopy is a cornerstone

technique in forensic fibre comparison. It is a very versatile technique that allows the com-
position of an entire fibre to be analysed by using transmission measurements or its surface
composition to be analysed by using ATR measurements. It is also a very reliable technique;
in this author’s experience infrared microspectroscopy has rarely failed to yield useful spec-
tral data for textile fibres submitted for forensic examination. As the sophistication of Raman
microspectrometers has progressed in leaps and bounds over the last decade the question as
to which variant of vibrational microspectroscopy – Raman or infrared – is best for forensic
fibre examination is a relevant one. Miller and Bartick (2001) have shown that there is little to
recommend one technique over the other as far as class and sub-class discrimination power
is concerned, with the possible exception of the forensically very important acrylic fibre class
where infrared offers better sub-class discrimination. Raman offers the possibility of surface-
enhanced techniques that in one step offer better signal to noise performance and fluorescence
suppression. Through confocal operation it is possible to collect Raman spectral data from any
position within a fibre. On the other hand, infrared microspectroscopy offers simple infrared
polarization experiments that yield important information as to polymer chain orientation
and with ATR, infrared microspectroscopy offers the ability to conduct surface analysis. One
major drawback of Raman microspectroscopy, which is sometimes insurmountable, is the
interference of strong background fluorescence. This is not an issue with infrared spectroscopy
which, as indicated above, is a technique that can be reliably and universally applied without
the intrusion of too many unpredictable interferences. Although the question as to which vari-
ant of vibrational spectroscopy is best is relevant, its answer, unfortunately, is not clear.
Even more important than realizing the strengths of a given analytical technique is the
recognition of its limitations in order to not draw incorrect or overreaching conclusions or
to not apply it when the technique has nothing to offer.
It is important to keep in mind that polymer composition might be the least discriminating
fibre characteristic examined in forensic fibre examinations. For example, it is entirely plausible
that light red and light blue fibres of different sizes and shapes may share a common polymer
composition and, therefore, yield indistinguishable infrared spectra. Relatively few polymers
are currently used to produce fibres. On the other hand, there are many dyes and pigments
available and their combinations make colour analysis a very discriminating technique. Thus,
infrared spectroscopy is not typically recommended for fibre comparisons until after micro-
spectrophotometry has been conducted. In regards to natural fibres, skilled usage of optical
microscopy will quickly and simply yield a wealth of information as to the animal or plant
origin of fibres that infrared microspectroscopy can only confirm rather than resolve further.
References
ASTM, 1981. Standard Practices for Internal Reflection Spectroscopy, E573-81. Philadelphia, PA:
American Society for Testing and Materials.
Bartick, E. G., Tungol, M. W., and Reffner, J. A., 1994. A new approach to forensic analysis with infrared
microscopy: Internal reflection spectroscopy, Anal. Chim. Acta, 288, 35–42.
Bassan, P., Lee, J., Sachdeva, A., Pissardini, J., Dorling, K.M., Fletcher, J.S., Henderson, A., Gardner,
P., 2013. The inherent problem of transflection-mode infrared spectroscopic microscopy and
the ramifications for biomedical single point and imaging applications, Analyst, 138, 144–157.
Bhargava, B., Wall, B. G., and Koenig, J. L., 2000. Comparison of the FT-IR mapping and imaging
techniques applied to polymeric systems, Appl. Spectrosc., 54(4), 470–479.
Brinsko, K., 2010. Optical characterization of some modern ‘eco friendly’ fibers, J. Forensic Sci.,
55(4), 915–923.
Carrillo, F., Colom, X., Sunol, J. J., and Saurina, J., 2004. Structural FTIR analysis and thermal char-
acterisation of lyocell and viscose-type fibres, Eur. Polym. J., 40, 2229–2234.
Carter, R. O., Carduner, K. R., Paputa Peck, M. C., and Motry, D. H., 1989. The infrared analysis of
polyethylene terephthalate fibres and of their strength as related to sample preparation and to
particle size, Appl. Spectrosc., 43, 791–794.
Causin, V., Marega, V., Guzzini, G., and Marigo, A., 2004. Forensic analysis of polyethylenetere-
phthalate fibers by infrared spectroscopy, Appl. Spectrosc., 58(11), 1272–1276.
Causin, V., Marega, V., Schiavone, S., and Marigo, A., 2005. A quantitative differentiation method for
acrylic fibers by infrared spectroscopy, Forensic Sci. Int., 151, 125–131.
Chan, K. L. A., Kazarian, S. G., Mavraki, A., and Williams, D. R., 2005. Fourier transform infrared
imaging of human hair with a high spatial resolution without the use of a synchrotron, Appl.
Spectrosc., 59(2), 149–155.
Chan, K. L. A., and Kazarian, S. G., 2013. Correcting the effect of refraction and dispersion of light
in FT-IR spectroscopic imaging in transmission through thick infrared windows, Anal. Chem.,
85, 1029–1036.
Chase, D. B., 1987. Infrared microscopy: A single-fiber technique. In P. B. Roush (Ed.), The Design,
Sample Handling, and Applications of Infrared Microscopes (pp. 4–11). ASTM STP 949,
Philadelphia, PA: American Society for Testing and Materials.
Chase, D. B., 1988. Dichroic infrared spectroscopy with a microscope. In R. G. Messerschmidt and
M. A. Harthcock (Eds.), Infrared Microspectroscopy: Theory and Applications (pp. 93–102). New
York, NY: Marcel Dekker.
Cho, L., Reffner, J. A., Gatewood, B. M., and Wetzel, D. L., 1999a. A new method for fiber comparison
using polarized infrared microspectroscopy, J. Forensic Sci., 44(2), 275–282.
Cho, L., Reffner, J. A., Gatewood, B. M., and Wetzel, D. L., 2001. Single fiber analysis by internal reflc-
tion infrared microspectroscopy, J. Forensic Sci., 46, 1309–1314.
Cho, L., Reffner, J. A., and Wetzel, D. L., 1999b. Forensic classification of polyester fibers by infrared
dichroic ratio pattern recognition, J. Forensic Sci., 44(2), 282–291.
Church, S., 1991. Polarized infrared microspectroscopy of fibers and crystals, Spectra-Tech Scan
Time, 22, 1–4.
Davis, B. J., Carney, P. S., and Bhargava, R., 2010a. Theory of midinfrared absorption microspectros-
copy: I. Homogeneous samples, Anal. Chem., 82, 3474–3486.
Davis, B. J., Carney, P. S., and Bhargava, R., 2010b. Theory of midinfrared absorption microspectros-
copy: II. Heterogeneous samples, Anal. Chem., 82, 3487–3499.
Davis, B. J., Carney, P. S., and Bhargava, R., 2011. Theory of infrared microspectroscopy for intact
fibers, 83, 525–532.
Flynn, K., O’Leary, R., Roux, C., and Reedy, B. J., 2006. Forensic analysis of bicomponent fibers using
infrared chemical imaging, J. Forensic Sci., 51(3), 586–596.
Fraser, R. D. B., 1953. The interpretation of infrared dichroism in fibrous protein structures, J. Chem.
Phys., 21, 1511–1515.
Garside, P. and Wyeth, P., 2007. Use of polarization spectroscopy as a tool for examining the micro-
structure of cellulosing textile fibers, Appl. Spectrosc., 61(5), 523–529.
Gilbert, C., Kokot, S., and Meyer, U., 1993. Application of DRIFT spectroscopy and chemometrics
for the comparison of cotton fabrics, Appl. Spectrosc., 47, 741–748.
Grieve, M. C., 1995. Another look at the classification of acrylic fibres, using FTIR microscopy, Sci.
Justice, 35, 179–190.
Grieve, M. C., and Griffin, R. M. E., 1999. Is it a modacrylic fibre? Sci. Justice, 39, 151–162.
Grieve, M. C., Griffin, R. M. E., and Malone, R., 1998. Characteristic dye absorption peaks found in
the FTIR spectra of coloured acrylic fibres, Sci. Justice, 38, 27–37.
Griffiths, P. and De Haseth, J. A., 2007. Fourier Transform Infrared Spectrometry (2nd ed.). Hoboken,
NJ: John Wiley & Sons.
Harrick, N. J., 1967. Internal Reflection Spectroscopy. New York, NY: Interscience.
Hirschfeld, T., 1978. New trends in the application of Fourier transform infrared spectroscopy to
analytical chemistry, Appl. Opt., 17, 1400–1412.
Hirschfeld, T., 1979a. Quantitative FT-IR: A detailed look at the problems involved. In J. R. Ferraro
and L. J. Basile (Eds.), Fourier Transform Infrared Spectroscopy Applications to Chemical Systems
(Vol. 3, pp. 215–218). New York, NY: Academic Press.
Hirschfeld, T., 1979b. Diagnosis and correction of wedging errors in absorbance subtract Fourier
transform infrared spectrometry, Anal. Chem., 51, 495–499.
Kim, H. C., 2009. The structure of acrylic, polyvinylalcohol, and polyvinylchloride fibers. In S. J. Eichorn,
J. W. S. Hearle, M. Jaffe, and T. Kikutani (Eds.), Handbook of Textile Fibre Structure. Fundamentals
and Manufactured Polymer Fibres (Vol. 1, pp. 305–306). New Delhi: Woodhead Publishing.
Krishnan, K., 1984. Applications of FT-IR microsampling techniques to some polymer systems, Poly
Prepr., 25, 182–184.
Krishnan, K. and Ferraro, J. R., 1982. Techniques used in Fourier transform infrared spectroscopy.
In J. R. Ferraro and L. J. Basile (Eds.), Fourier Transform Infrared Spectroscopy Techniques Using
Fourier Transform Interferometry (Vol. 3, pp. 193–198). New York, NY: Academic Press.
Lee, J., 2017. On the non-existence of the so-called ‘electric field standing wave effect’ in transflection
FTIR spectra, Vibrational Spectroscopy, 90, 104–111.
Messerschmidt, R. G., 1995. Minimizing optical nonlinearities in infrared micsrospectroscopy. In H.
J. Humecki (Ed.), Practical Spectroscopy, Practical Guide to Infrared Microspectroscopy (Vol. 19,
pp. 1–39). New York, NY: Marcel Dekker.
Miller, J. V. and Bartick, E. G., 2001. Forensic analysis of single fibers by Raman spectroscopy, Appl.
Spectrosc., 55(12), 1729–1732.
Nishikida, K., 2004. Spatial Resolution in Infrared Microscopy and Imaging, Application Note: 50717.
Madison, WI: Thermo Electron Scientific Instruments Corp.
Perkin Elmer, 2006. ATR FT-IR Imaging of Human Hair Cross-Section. Shelton, CT: Perkin Elmer.
Quynn, R. G., 1954. An infrared microscopic study of orientation in fibers. Ph. D. Dissertation,
Princeton University.
Roush, P. B. (Ed.), 1987. The Design, Sample Handling, and Applications of Infrared Microscopes,
ASTM STP 949. Philadelphia, PA: American Society for Testing and Materials.
Sommer, A. J. and Katon, J. E., 1991. Diffraction-induced stray light in infrared microspectroscopy
and its effects on spatial resolution, Appl. Spectrosc., 45, 1633–1640.
Stuart, B. H., 2004. Infrared Spectroscopy: Fundamentals and Applications. Chichester: Wiley & Sons.
Suzuki, E. M. and Pettit, W. E., 1994. An antistatic device for use when sampling with a diamond
anvil cell, J. Forensic Sci., 39, 904–905.
Tungol, M. W., Bartick, E. G., and Montaser, A., 1991. Analysis of single polymer fibres by Fourier
transform infrared microscopy: The results of case studies, J. Forensic Sci., 36, 1027–1043.
Tungol, M. W., Bartick, E. G., and Montaser, A., 1993. Forensic analysis of acrylic copolymer fibers
by infrared microscopy, Appl. Spectrosc., 47, 1655–1658.
Tungol, M. W., Bartick, E. G., and Montaser, A., 1995a. Forensic examination of synthetic textile
fibres by microscopic infrared spectrometry. In H. J. Humecki (Ed.), Practical Spectroscopy,
Practical Guide to Infrared Microspectroscopy (Vol. 19. pp. 245–286). New York, NY: Marcel
Dekker.
Tungol, M. W., Bartick, E. G., Montaser, A., 1995b. Polarized infrared study of poly(ethylene tere-
phthalate) fibers, Crime lb. Digest, 22, 86–87.
Young, P. H., 1988. The characterization of high-performance fibres using infrared microscopy,
Spectroscopy, 3, 25–30.
Raman Spectroscopy of Fibres
JANE HEMMINGS
8
Contents
8.2 Raman Spectroscopy Theory 290
8.3 Potential Limiting Factors 292
8.3.1 Fluorescence 292
8.4 Instrumentation 293
8.4.1 Instrument Set-up 293
8.4.2 Lasers 294
8.4.3 Microscope 296
8.5 Analysis Parameters 297
8.5.1 Commonly Used Techniques, Sample Handling and Preparation 297
8.5.2 Raman Analysis Conditions. 300
8.6 Techniques for Special Applications, SERS/SERRS 301
8.7 Data Handling, Treatment and Results 302
8.7.1 Spectral Presentation 302
8.7.2 Assigning Bands—First Principles 302
8.8 Interpretation 302
8.9 Conclusions 305
References 305
Further Reading 306
8.1 Introduction
There are two main spectroscopic techniques used to detect molecular vibrations; infrared
(IR) spectroscopy, as discussed in chapter 7; and Raman spectroscopy. Both techniques
provide structural detail and are able to analyse samples ranging from gas to solid and
from microscopic to bulk analysis. Raman spectroscopy historically has been used less
frequently than infrared spectroscopy (particularly in the forensic sciences) due to once
typical issues of sample degradation, fluorescence, and high equipment cost and complex-
ity. Major advances in technology in the last 10–20 years have enabled Raman spectros-
copy to move from predominantly academic and industrial laboratories to applications in
operational forensic laboratories.
One field where Raman spectroscopy has increased in popularity is in the analysis of
forensic fibre evidence. Raman spectroscopy, unlike infrared spectroscopy, allows for the
in-situ analysis of a single fibre. This advantage of no (or very minimal) sample preparation
coupled with the fact that the technique is almost totally non-destructive to the sample
has seen its application in fibre examinations grow significantly since the early 2000’s and
especially since the introduction of the European Fibre Group’s (now the European Textile
and Hair Group) Raman spectroscopy sub-group.
289
In the analysis of forensic fibre samples Raman spectroscopy is predominantly applied

to the analysis of the dyes and pigments used to colour the fibre because of the relatively
strong Raman signals from these substances. In some instances it can also be used to char-
acterise the fibre substrate, however, infrared spectroscopy is still the method of choice
for fibre identification as fibre substrates tend to have weak Raman signals that can be
swamped by the Raman signal of dyes or pigments.
Modern Raman spectroscopy is relatively simple. Complex experimental set-ups are
no longer a hindrance as there are few instrument variables, and data interpretation may
be as simple as a direct comparison between spectra. This chapter introduces the basic
principles of Raman spectroscopy, its application in the field of forensic fibre examinations
and what sample, instrument and data interpretation factors the forensic fibre examiner
will need to take into consideration. This chapter is not meant to be a comprehensive dis-
cussion on advanced Raman theory, but instead will focus on the application of Raman
spectroscopy to the analysis of textile fibres, from fibre identification to dye analysis. More
detailed descriptions of Raman theory, technique and instrumentation can be found in
texts such as those listed at the end of this chapter.
8.2 Raman Spectroscopy Theory
In order to understand the benefits of Raman spectroscopy in the analysis of fibre

dyes and pigments it is necessary to understand the technique. When photons of
light interact with matter they may be scattered, absorbed or travel through. If the
energy of the photon is equal to the difference between the ground state and the excited
state of the molecule then the photon may be absorbed and the energy used to promote
the molecule to the excited state. The measurement of this change in photon energy is
absorption spectroscopy (and is the principle on which IR spectroscopy is based). The
same photon could interact with a molecule and be scattered by it. In this instance,
the photon does not need to have the same energy as the difference between the ground
state and the excited state in order for it to be scattered. The scattering may be elastic
(i.e. at the same wavelength as the photon, this is known as Rayleigh scattering) or
inelastic (i.e. at a different wavelength from the photon; these are known as Stokes and
anti-Stokes scattering).
Sir CV Raman described the inelastic scattering of light experimentally in 1928.
Raman discovered that when radiation was passed through a transparent medium, a small
fraction of the scattered radiation differed in wavelength from that of the incident beam.
Furthermore, the shifts in wavelength were dependent on the chemical structure of the
sample. This became known as Raman scattering and was a result of the photons of light
changing the distance between the nuclei of a molecule during interaction. Raman dif-
fers from infrared spectroscopy in that the measurement of the vibrational energy of the
molecule/s is based on different selection rules.
Raman scattering occurs when a photon of light is able to induce a change in the polar-
isability of the electron cloud surrounding the nuclei. Symmetric changes generally induce
the largest scattering. Conversely, with infrared absorption the greatest change occurs
with asymmetric vibrations that change the dipole moment. This means that not all mol-
ecules will be both Raman scatterers and infrared absorbers and the two spectroscopic
techniques tend to give different spectral patterns.
Raman Spectroscopy of Fibres 291
Raman spectroscopy and infrared spectroscopy differ in the way in which they utilise
radiation. In infrared spectroscopy, radiation with a wide range of frequency is impinged
on the sample. Absorption of a frequency occurs when that frequency matches the energy
difference between the ground state and the excited vibrational state of the molecule. This
‘absorption’ of the frequency is measured using a detector. Conversely, Raman spectros-
copy utilises radiation of a single frequency directed onto the sample. Raman spectra are
obtained using a powerful laser source (UV, visible or infrared monochromatic radiation)
to irradiate a sample (refer to Figure 8.1). Once the monochromatic light (νlaser) impinges
on the sample, it interacts with the sample and the resulting light is scattered elastically
(νscattered) or inelastically (Raman scattered light). The spectrum of the scattered radiation
is measured using a suitable spectrometer. The scattered radiation is of three types, Stokes,
anti-Stokes, and Rayleigh.
Figure 8.2 is a representation of a Raman spectrum and it shows that the Rayleigh scat-
tering is at the same wavelength of the excitation source (e.g. argon ion laser, 480.0 nm).
Rayleigh scattering is the most intense since most photons will scatter elastically. Generally,
anti-Stokes lines are less intense than the corresponding Stokes lines and the anti-stokes
Raman scattered light;

stokes and anti-stokes
νlaser– νscattered = νRaman (weak; 1 in 106 photons)
νlaser νscattered – Rayleigh scattered light
Laser Incident light Sample
Figure 8.1 Schematic of the Raman effect showing Raman and Rayleigh scattering.
Rayleigh scattering
(wavelength 480.0 nm)
Intensity
Stokes line Anti-stokes line
Δν, cm–1
Figure 8.2 Raman spectrum showing relative differences in intensity of Rayleigh scattering,
Anti-Stokes and Stokes radiation.
scattering will weaken as the frequency of the vibration increases and will become stronger
as the temperature of analysis rises. For this reason the Stokes region of the spectrum is
used (the negative sign is reversed and the Raman shift is expressed in cm–1).
In a Raman spectrum, wavenumbers (cm–1) are displayed on the x-axis. Strictly speak-
ing Wavenumbers (cm–1) are not an SI Unit and Raman scattering should be shown as the
change (or shift) in energy from the wavelength of the radiation (Δcm–1), however, as it is
standard practice for spectroscopists to use cm–1, this common nomenclature will be used
in this chapter.
8.3 Potential Limiting Factors
Raman has significant advantages over other analytical techniques that are used for the
analysis of dyes and pigments, such as thin-layer chromatography (TLC) and microspec-
trophotometry (MSP). This is predominantly due to easy sample preparation and analysis
and the potential to identify a sample. It does however have some inherent technical issues,
the two greatest being
• Fluorescence
• Degradation of the sample (such as pyrolysis of the sample through thermal inter-
actions and photochemical reactions from light exposure)
Of these two problems, fluorescence proves to be the most common.
8.3.1 Fluorescence
Impinging a sample with photons from a laser (hvex) will cause excitation of some molecules
from ground state (S0) to an excited state (S’1). The Raman effect results in the molecule’s
either gaining energy or losing energy to the vibration and rotation of the. Fluorescence,
however, is a radiative decay process (from S’1 to S1) that occurs after excitation of the elec-
tronic molecular states and produces an energy change (ΔEFluorescence) thereby a change in
emitted wavelength (hvem). Figure 8.3 shows this decay process.
S’1 ΔEFluorescence
S1
hvex
Energy
hvem
S0
Figure 8.3 Fluorescence decay process.

In Raman spectroscopy, the problem arises because the spectrometer measures changes
in photon energy (i.e. wavelength) after molecular excitation. If fluorescence occurs, the
resultant energy change (ΔEFluorescence) will be recorded along with the energy change as a
result from the Raman effect. The energy change from the Raman effect occurs in a small
number of instances (1 in 106 photons). If energy changes due to fluorescence outnum-
ber those produced by the Raman effect in intensity fluorescence will swamp any Raman
signal.
Fluorescence is an unpredictable phenomenon (colour is no guide to what may
fluoresce when analysed by Raman) that does not occur in every instance and appears as a
large featureless background that swamps Raman bands from the sample being analysed.
8.4 Instrumentation
The forensic fibre examiner will need to make several practical choices in relation to
instrumentation, sample presentation and data handling prior to commencing analysis.
The choices made will be partially determined by instrumentation available to the exam-
iner as well as what questions the examiner wants answered. Some common questions/
considerations are
• Which instrument set-up is best?

• Do I use a dispersive or non-dispersive spectrometer?
• What laser wavelength will provide the optimal results for my sample?
• Do I need a microscope attachment with my instrument?
• How should my sample be presented to the instrument?
• What analysis parameters are most suitable?
Answering these questions is essential for a Raman analysis that will produce results
in which the examiner can be confident.
8.4.1 Instrument Set-up

One decision is whether to utilise a dispersive or a non-dispersive spectrometer. A disper-
sive spectrometer utilises a grating based dispersive unit and CCD detector to separate the
different wavelengths (refer to Figure 8.4 for an example of a dispersive Raman spectrom-
eter). A non-dispersive spectrometer modulates the different wavelengths into frequencies
(also known as Fourier Transform Raman, FT-Raman) similar to the Fourier Transform
interferometers used to modulate the infrared beam in Fourier Transform Infra-Red
(FTIR) spectroscopy.
A comparison of the visible (dispersive) and the Near Infra-Red (NIR)/Fourier
Transform (FT) instruments (non-dispersive) is made more complex by the fact that the
visible systems can be extended into the NIR region by using lasers with excitation lines
in the 785–850 nm region. The CCD chips used in dispersive instruments lose sensitivity
at wavelengths above 1000 nm and this means that a 785 nm laser used on a dispersive
instrument is close to the end of the detector range and a drop in sensitivity for the higher
frequency peaks will result. This disadvantage is offset by the increase in range of samples
Figure 8.4 An example of a dispersive Raman Spectrometer.
that can be examined without fluorescence interference. Both types of instrumentation

(dispersive and non-dispersive) have their advantages and disadvantages. For the forensic
fibre examiner the most important factor is the difference in sensitivity once each type of
instrument is coupled with a microscope. Whilst the spectral resolution is dependent on
the grating used for a dispersive spectrometer it is generally more sensitive when coupled
to a microscope than a non-dispersive spectrometer.
8.4.2 Lasers
The necessity of quality monochromatic radiation for effective Raman scattering coupled with
their limited lifetime means that laser/s can be quite expensive. While this is an obvious con-
sideration for laboratories, it is by no means the only consideration when it comes to selecting
the laser source/s for use in fibre examination. Lasers are available in a wide range of wave-
lengths from the deep UV (<200 nm) to the NIR (1064 nm). The wavelength of the excitation
source is directly related to the intensity of Raman scattering (known as the ν4 dependency).
This means that the short wavelengths will have higher intensities (more efficient and less
noise) but will increase the possibility of exciting fluorescence in your sample (though this is
reduced when moving from the visible to the UV). Longer wavelengths will reduce the effect
of fluorescence but will result in a decrease in signal strength and an increase in noise.
Some common lasers are:
• Ultra Violet: 244 nm, 257 nm, 325 nm and 364 nm

• Visible: 457 nm, 473 nm, 488 nm, 514 nm and 532 nm
• Near Infrared: 785 nm, 830 nm, 980 nm and 1064 nm (usually installed on
FT-Raman instruments)
Surveys carried out by the European Textile and Hair Group (ETHG), formerly known
as the European Fibres Group (EFG), led by Professor Massonnet of the University of
Lausanne, Switzerland; indicate that having at least two lasers enabled the analysis of most
fibre samples. A visible laser (514 or 532 nm) and a NIR (785 or 830 nm) were considered
the most complimentary.
Because of resonance enhancement effects, fluorescence and the ν4 dependence of
scattering intensities on laser frequency, Raman spectra of the same substance, at differ-
ent laser wavelengths, can exhibit different relative peak intensities (refer to Figure 8.5).
Therefore, caution has to be exercised when measuring and interpreting Raman spectra, as
data from different laser wavelengths cannot always be compared directly.
Several studies have looked at a range of laser wavelengths for specific fibre/dye samples.
Table 8.1 is a summary of the fibre types, colour and the wavelengths used in each study.
The collective results of the studies listed in Table 8.1 are that the optimal laser is
dependent on the sample set being analysed and having a system with multiple lasers was
the most beneficial.
A study by Massonnet et al. (2012) researched a series of dye concentration detection
limits. This study included the Raman analysis of reactive dyed cotton fibres where the
concentration ranges were 5%–0.005% (w/w). Raman spectroscopy was identified as being
the most sensitive technique, detecting concentrations as low as 0.005%.
514.5 nm laser
632.8 nm laser
Intensity
785 nm laser
830 nm laser
1500 1000 500

Counts/Raman shift (cm–1)
Figure 8.5 Analysis of Sumifix Navy dye with (from top) 514.5nm, 632.8nm, 785nm and
830nm laser wavelenths sources. (From Thomas, J. M., PhD thesis, University of Technology,
Sydney, Australia, 2005.)
Table 8.1 Fibre Type, Colour and Wavelengths That Have Been Investigated
Reference Fibre Type Colour Wavelength/s (nm)

Lang et al. (1986) Polyester Colourless 514.5 and 476.5
polyacrylonitrile
Bouffard et al. (1994) Polypropylene Various pigments 632.8
Bourgeois and Church Acrylic Red and blue 1064
(1990)
Keen et al.(1998) Rayon Colourless 632.8
Wool Colourless 632.8
Polypropylene Colourless 632.8
Polyester Colourless, red 632.8 and 780
Nylon Colourless, red 632.8 and 780
Kokot et al. (1997) Cotton Red 1064
Miller et al. (2001) Various man-made Colourless 785
Jochem et al. (2002) Polyacrylonitrile Various 514 and 633
Viscose Various 514 and 633
Massonnet et al. (2005) Acrylic Red 458, 488, 514, 532, 633,
685, 785, 830,1064 for all
Wool Red
Massonnet et al. (2004) Cotton Green 458, 488, 514, 532, 633,
685, 785, 830,1064
Thomas (2005) Cotton Grey, blue, black 514.5, 632.8, 785, 830
Buzzini and Massonnet Cotton, Wool, acrylic Black, blue and red 514.5, 632.8, 785 and
(2013) 830 nm
Lepot et al. (2008) reviewed five fibre cases that had utilised both Raman spectroscopy
and microspectrophotometry (MSP). The study highlighted the value of including Raman
spectroscopy in the fibre examination sequence and concluded by recommending that
Raman spectroscopy be applied in routine fibre dye analysis following optical microscopy
and MSP measurements. Additionally, Jochem and Lehnert (2002) observed that Raman
spectroscopy is capable of detecting and identifying at least the main (most concentrated)
dye component on a fibre.
8.4.3 Microscope
It is critical that the Raman spectrometer has a microscope attachment to allow the analy-
sis of a single fibre. Without a microscope only gross measurements of textiles and refer-
ences can be undertaken. The major advantage of the microscope is that any part of the
sample that can be magnified and visualised can also have the Raman spectrum measured.
The laser spot size will depend on the laser wavelength being used as well as the mag-
nification of the microscope objective. The theoretical spatial resolution of a microscope-
coupled Raman spectrometer is calculated as
λ
1.22
NA
Where: λ = wavelength of laser being used and NA = the numerical aperture of the micro-
scope objective.
For example a 532 nm laser wavelength with a 0.9/100× objective would have a theo-
retical spot size of 0.72 μm, and the same objective with a 785 nm laser will have a theoreti-
cal spot size of 1.1 μm. In practice the optical processes for Raman are more complex than
for optical microscopy, and as a result of this complexity the actual spot sizes for analysis
are larger than the theoretical calculations.
The fact that the microscope effectively focuses the beam also means that it is possible
to use relatively low powered lasers for analysis. The focussing of the beam by the micro-
scope will also create a large collection angle and the resultant small excitation volume
requires only a small spectrometer.
The microscope can be set up as normal or for confocal analysis. In the confocal
arrangement, the Raman spectrometer utilises a pinhole in the focal plane, this enables
only the light focussed on the plane to be collected efficiently (as light not focussed on the
plane is stopped). This enables examination of the sample in the z-plane resulting in depth
profiling of the sample. Confocal analysis is not routinely used in fibre examinations as the
fibre is either uniformly dyed or the dye is concentrated at the fibre surface. The confocal
set-up is advantageous when there are very fine fibres and the depth of focus needs to be
reduced (e.g. 514 nm laser with 100x objective will reduce depth of focus from 3.3 μm to
2 μm): this will however come at a loss of Raman signal.
8.5 Analysis Parameters
8.5.1 Commonly Used Techniques, Sample Handling and Preparation

Raman spectroscopy is well known for the minimum handling and preparation required
for analysis. Some texts (see Smith et al., 2005) recommend that fibres be loosely packed
into a small diameter glass tube (or alternately wrapped around the tube) to increase the
signal to noise. Practically, as forensic fibre examiners know, this is not possible except
with a sample of known or reference fibres.
Several approaches to mounting fibres (of sizes commonly encountered in casework)
have been undertaken. Table 8.2 summarises the different set-ups tested.
Bourgeois and Church (1990) designed a cell specifically to optimize the Raman signal
in their fibre experiments. The fibres were compressed between a glass window and an
adjustable concentric metal rod which was located with a screw fitting. This arrangement
Table 8.2 Mounting Methods Used in Various Studies

Reference Mounting (covering) Mounting (base or mountant)
Bourgeois and Church (1990) Glass Specially constructed cell
Keen et al. (1998) No cover (fibres taped Glass slide metal plate
at each end)
Miller et al. (2001) No cover Glass slide
Glass coverslip Aluminium foil permount (mountant)
Jochem et al. (2002) No covering Specially constructed aluminium cell
Thomas (2005) No covering Glass slide
Glass coverslip XAM (mountant)
Koens (2013) No covering Aluminium foil
Glass coverslip Entellan New (mountant)
ensured that unwanted voids were not present in the sampling area and that a high density
of sample was impinged by the laser beam.
Jochem et al. (2002) also constructed a cell for the analysis of their fibre samples. This
was simply an aluminium slide with a small hole that the fibre was placed over. In this way
the authors avoided the influence of a specimen holder.
Keen et al. (1998) analysed their samples on either glass slides or metal plates and
encountered no problems with this form of sample mounting.
Miller et al. (2001) also examined samples with no covering on glass slides and metal
plates. Initially the authors taped the end of their fibre samples on to glass slides and then
analysed the exposed fibres. They found that this method produced an underlying fluores-
cence spectrum from the glass. The authors found that changing the glass to aluminium
foil removed the underlying spectrum and produced clear spectra for the samples.
Typically the best results are produced by direct laser impingement on the samples.
However, as highlighted by the Miller et al. study, sample information can be obtained
when fibres are mounted in less than ideal situations (i.e. Permount). If samples can be
analysed directly after mounting then this could potentially reduce time spent in changing
the mounting conditions after microscopy or MSP. With samples that are in a matrix (such
as a fibre mounted in XAM) the relative scattering intensities of the matrix need to be con-
sidered before the sample is analysed and the resultant spectra is interpreted.
An example of the effect of the mounting condition is shown in Figure 8.5. Thomas
(2005) investigated two mounting techniques: one where there coloured cotton fibre was
mounted on double sided tape secured to a glass slide (bottom spectra in Figure 8.5), and
the second where the same fibre was mounted in the permanent mountant XAM under a
glass cover slip (top spectra in Figure 8.6).
The fibre analysed in the XAM mounted conditions produced a significantly different
spectrum to that of the uncovered fibre. This spectrum was dominated by a broad, intense
peak at ~1450 cm–1 which can be attributed to glass. Some of the smaller peaks observed in
Intensity (arbitrary)
1800 1600 1400 1200 1000 800 600 400 200

Raman shift (cm–1)
Figure 8.6 Raman spectra of: (top) Black cotton through glass coverslip (mounted in XAM),
and (bottom) black cotton sample with no covering. (From Thomas, J. M., PhD thesis, University
of Technology, Sydney, Australia, 2005.)
the mounted spectrum may be attributed to the fibre sample as they have a similar Raman
shift. The other smaller peaks can be attributed to the XAM contribution (refer to Figure 8.7).
While analysing a fibre in an unmounted condition produces unimpeded results it
does introduce an additional step of demounting a fibre before Raman analysis. This pro-
cess can be difficult and lead to a breakage of the fibre or even the loss of a fibre.
Recent work by Koens (2013) has shown that the analysis of the fibre in the mounted
condition can be optimised to minimise the spectral interferences of both the glass cover-
slip and the mountant. The optimisation primarily focussed on the focal point for analy-
sis and the results shown in Figure 8.8 and 8.9 show the limited contribution of either
Intensity (arbitrary)
1800 1600 1400 1200 1000 800 600 400 200

Counts/Raman shift (cm–1)
Figure 8.7 Raman spectra of: (top) XAM through glass coverslip, (middle) XAM, and (bottom) glass
coverslip. (From Thomas, J. M., PhD thesis, University of Technology, Sydney, Australia, 2005.)
250,000
Aluminium
200,000 Glass
150,000
Counts
100,000
50,000
0
100 600 1100 1600 2100 2600 3100
Figure 8.8 Raman spectra of sample a Reactive Black 5 dyed cotton (top) under glass coverslip
with mountant, and (bottom) ends of fibre taped onto aluminium. (From Koens, F. S., Thesis
[Hons.], University of Canberra, 2013.)
350,000
300,000 Aluminium
Glass
250,000
200,000
Counts
150,000
100,000
50,000
0
100 600 1100 1600 2100 2600 3100
Figure 8.9 Raman spectra of sample a Sulphur Black dyed cotton (top) under glass coverslip
with mountant, and (bottom) ends of fibre taped onto aluminium. (From Koens, F. S., Thesis
[Hons.], University of Canberra, 2013.)
coverslip or mountant (top spectrum) to the spectra of the dyed fibres. The benefit of being
able to examine in situ is not only the elimination of the risk of losing the sample but
also the reduction in sample preparation time. The study by Koens highlighted the strong
relationship between the intensity of the colour of the fibre and the success of acquiring
a spectra of the mounted fibres. This result correlates with the findings from the study of
sensitivity limits conducted by Massonnet et al. (2005).
8.5.2 Raman Analysis Conditions

Polymers are generally very weak scatterers. This can be quite an advantage for Raman
analysis of coloured polymers as the resultant spectrum is predominantly attributed to the
dye or pigment. The most influential conditions are:
Wavelengths range: The range that provides the most information for Raman analy-
sis is the 3600–200 cm–1 range (though different wavelength ranges are possible).
This range includes most characteristic vibrational modes of a molecule. For the
analysis of dyed fibres most information is obtained from the 2000–200/300 cm–1
range. More polymer characteristics can be observed for some man-made fibres
if the range is extended up to 4000 cm–1.
Microscopic objective: Generally, a 50× or a 100× objective is best; however, different
objectives should be tested on the fibre samples to be examined.
Laser power (or intensity): The primary consideration for the laser power inten-
sity is that it should not degrade the sample, secondly is that it should not
saturate the detector. It is suggested that the examiner starts with a low laser
power and increase it as determined by the results obtained. Photobleaching
of the sample may occur at high laser powers, which may not physically alter
the sample but will change the molecular structure (and therefore the resultant
spectra).
Time of analysis, number of accumulation, number of measures: These parameters

will determine how long the sample is irradiated (time of analysis) and how
many times the spectrum is acquired (number of accumulations/number of
measures). The optimal settings for these parameters are very sample depen-
dent. If a good quality spectrum is obtained with short analysis time and a small
number of accumulations then variation in these parameters is not required.
If, however, the first spectrum is weak then these parameters will need to be
adjusted by increasing the time of analysis and the number of accumulations/
measures.
Grating: Some instrument set-ups allow the grating to be varied (thereby varying
the resolution). The grating can vary from 300 to 1800 lines per mm. The greater
the number of lines per millimetre the higher the resolution (and the longer the
analysis time). It is generally agreed that a grating of 1200 L/mm will produce a
well-resolved spectrum with a reasonable analysis time.
Sample focus: For the analysis of fibres (as with other solids) collection of the scat-
tered radiation should be from as large an area as possible. The sensitivity can be
increased by using a reflective surface (such as an aluminium or mirrored slide)
behind the sample being measured. To obtain the maximum signal from a solid
the surface should be at (or near) the focal point. This is particularly important in
those fibres that are dyed at the fibre or yarn stage as these fibres have the greatest
concentration of the dye at the fibre surface. Acceptable spectra can be obtained
from samples not at the focal point, however, the examiner needs to be aware that
relative band strength may vary depending on the distance the sample surface is
from the focal point.
8.6 Techniques for Special Applications, SERS/SERRS
Several sub-types of Raman spectroscopy have been developed to enhance the weak Raman
signal. These include resonance Raman scattering (RR), surface enhanced Raman scatter-
ing (SERS), and surface enhanced resonance Raman scattering (SERRS).
Resonance Raman scattering (RR) is a phenomena that happens when the energy of
the photon from the excitation laser beam is the same as that of an electronic transition a
chromophore within the dye/pigment being analysed. This can induce resonance that can
selectively enhance the bands that belong to the chromophore by factors of 103–105.
Surface enhanced Raman scattering (SERS) is a phenomenon where the applica-
tion of a colloidal metal to the surface of the sample being analysed (generally colloidal
silver is used) can give a Raman signal enhancement of about 106. Early work on the
application of colloids to fibres conducted by White (1996, 2000) indicate that there is
potential for this technique but as yet this has not been significantly realised in opera-
tional laboratories.
The combination of resonance enhancement (RRS) and surface enhancement (SERS)
is referred to as surface enhanced resonance Raman scattering (SERRS). The matching of
excitation energy to the absorbance maxima of a chromophore that is also in contact with
a metal colloid can produce signals up to ten orders of magnitude greater than that of stan-
dard Raman spectroscopy.
8.7 Data Handling, Treatment and Results
The resultant spectra of a sample being analysed by Raman spectroscopy is influenced by

not only the chemical composition of the sample itself but also by the instrument set-up and
any treatment of the data post collection. It is therefore important to understand the differ-
ent influences of instrument set-up (as discussed in sections 8.4 and 8.5) and data treatment.
When looking at untreated data the examiner should be aware that the scales tend
to automatically adjust. As a result, a weak spectrum may not be immediately obvious
and examiners should consult the y-scale to determine the strength of the spectrum. If a
spectrum is weak, the examiner should consider that the sample could be from too small
a sample (i.e. low amount of dye), or that the sample is a weak Raman scatterer. The exam-
iner should also be aware that any peaks in a weak spectrum could be from a contaminate
that is a strong scatterer.
Spectra can be scaled for comparison purposes; this is generally done by choosing a
band common to both spectra. Noise and the relative intensity of the band selected should
be considered. Spectra can be treated mathematically through the software by smoothing,
baseline correction, spectral subtraction and numerous other functions. Once the spectrum
has been generated care should be taken with any data treatment. Whilst many of these
features make the spectra look pretty they will result in the loss of information. If manipu-
lation is required, such as baseline correcting, this function should be undertaken with an
understanding of what is occurring to the original data and what may have been introduced/
lost with the data processing. It is recommended that the only function utilised is that which
removes peaks caused by cosmic rays, and baseline correction for comparing spectra.
8.7.1 Spectral Presentation

Spectral presentation is generally straightforward with the Stokes spectra presented and
the anti-Stokes omitted. The major variation is the way in which the wavelength scale is
presented. It is either high to low wavenumber or low to high wavenumber. Arguments
for both forms exist, what the fibre examiner needs to be aware of is that for comparison
purposes the scale needs to be the same for the questioned and known.
8.7.2 Assigning Bands—First Principles

As for FTIR interpretation, it is possible to assign frequency ranges for the characteristic
Raman scattering of the common functional groups. To interpret Raman spectra this way
requires knowledge of the approximate energy and intensity of bond vibrations. Texts such
as Lin-Vien et al. (1991) can be of great help when determining what a peak in a Raman
spectrum may be.
In order to assign specific peaks to specific vibrations, analysts generally utilise published
data and libraries of spectra with the aid of computer software for comparing similarities
(and differences) with standards, reference samples or casework based comparison samples.
8.8 Interpretation
In practice the use of Raman to determine an unknown compound comes a very poor sec-
ond to infrared spectroscopy. Examiners should be aware that whilst Raman spectra are
often clearer than IR spectra, they can be a lot less easy to fingerprint. This is because some
groups do not give strong bands and the number of recorded, published reference spectra
is significantly less.
If a fibre examiner is to determine what they are examining (in terms of dye/pigment
and substrate) then it is essential that they utilise all available information about the fibre
sample when interpreting the spectra.
Some of the factors to be considered are related to the sample itself whilst others are
related to the instrument used for collecting the data.
1. Sample: An understanding of how the sample came to be presented for exami-

nation will provide a lot of information. Whilst in Raman there is usually zero
preparation required, the nature of fibre examinations generally means that a fibre
may have been, at the least, demounted from a permanent mountant in order for
its presentation for examination. Understanding the sample will help sense check
resultant spectra and that the bands present and overall shape is not affected by the
sample and its preparation. Some questions that the examiner should ask is
• Has the sample been washed with a solvent?

• Has the work-up conditions introduced an impurity?
• Is the sample mounted on double sided tape?
• Is there likely to be a polarisation or orientation effect?
• Has the sample been rolled/flattened for IR spectroscopy analysis?
• Has the dye been extracted for TLC purposes?
Whilst the history of a fibre sample is generally available, the points listed above
should be considered if the resultant spectrum of the sample does not appear as
expected. Any sample with no known history should be interpreted with great
care.
2. Instrument: Anomalous result may occur due to instrumentation and software

effects. The examiner must consider the following instrument/software factors
when interpreting any sample spectrum:
• What laser wavelength was used?
• Could resonance of a chromophore have occurred?
• Could self-absorption have affected band strength?
• Has automatic software correction occurred removing background
fluorescence?
• Is the spectrum as strong as it appears or has the scale auto corrected?
• Has any smoothing function been applied?
• Are any broad bands due to sample fluorescence or burning?
• Are the sharp bands due to cosmic rays? Or fluorescent room lights (if you are
not examining in an enclosed chamber)?
When undertaking Raman analyses the examiner must be aware that the Raman spec-
trum is a combination of information from the dye or pigment, delustrants (if present) and
the fibre substrate itself. The ratio of contribution of each of these factors to the spectrum is
dependent on their relative concentration in the sample and their Raman scattering. A pale
or colourless fibre will generally result in a spectrum dominated by the fibre substrate
whereas a deeply coloured fibre may result in a spectrum of the dye or pigment only with
no contribution from the fibre substrate.
Due to the weak scattering of the OH bond, Raman spectroscopy can be used to charac-
terise some natural and regenerated cellulosic fibres. Raman is, however, not generally used for
the identification of the fibre substrate and it is recommended that infrared spectroscopy and/
or optical microscopy is used for this. It is, however, very important to have an understanding
(and preferably a reference set of Raman spectra) of the possible bands/peaks that the fibre
substrate can contribute when assessing the resultant Raman spectra for coloured fibres. This
is particularly important where there is a variation in the dye intensity of the target fibre which
can lead to variations in the contribution of the fibre substrate to the samples spectrum.
When assessing the Raman spectrum of a dyed fibre it is very difficult to determine
the structure of the dye and thereby identify the dye present. This is for a variety of reasons
including:
• There is a huge range of dyes (numbering several thousand) available in today’s market.
• The spectrum of chemically different dyes can be quite similar.
• Some dyes (such as reactive dyes) bind or chemically interact in some manner with
the fibre substrate.
• Dyes can interact with other dyes or pigments present.
• A comprehensive database of dyes (and their various interactions) is not available.
(The European Fibres Group (EFG) has produced a database of dyes with a focus
on red/blue and black reactive dyes utilising 514, 532 and 785 nm lasers).
Conversely, the number of pigments available is significantly smaller, numbering sev-

eral hundred, and almost all have been databased by the European Paint Group (ENFSI-
EPG) using 458, 514, 633 and 785 nm lasers. Each pigment has a unique Raman spectrum
and pigment identification by this means is relatively straight forward. A more complex
issue is when there is combinations of pigments present producing a mixed spectrum.
It is for the reasons listed above that the utilisation of Raman spectroscopy for forensic
fibre examinations is predominantly for the direct comparison of one sample (target fibre)
with another (recovered or questioned fibre). When comparing two samples it is important
that:
• The within sample variation is understood before any between sample variation is
investigated.
• The same laser wavelength is utilised for both samples.
• The major and minor peaks, their relative position and intensities are compared
(generally achieved by overlaying the spectrum of both).
• Variations in the noise are taken into account as this may make it difficult to visu-
alise some spectral details.
If there are no discernible differences then the two samples contain the same detectable
components and could share a common source. If differences are observed in peak posi-
tions and intensities (or peaks are missing) then the fibres do not have the same chemical
composition.
8.9 Conclusions
Raman spectroscopy is a technique that is growing in popularity for the analysis of mate-
rials. It is quick, discriminatory, non-destructive and requires minimal sample prepa-
ration. The research conducted to date highlights the value of this technique for fibre
examinations.
The fact that fibre polymers are weak Raman scatterers means that this technique is
excellent for examining dyes and pigments without the need for extraction. Many limita-
tions by way of expense and instrumentation size have been overcome however; the fibre
examiner may still encounter significant issues associated with fluorescence.
While there is still a large amount of research that can be conducted into fibre exami-
nations, the use of Raman spectroscopy in fibre examination has now moved from the
experimental arena to being part of the standard fibre examination sequence.
References
Bouffard, S. P., Sommor, A. J., Katton, J. E., and Godber, S., 1994. Use of molecular microspectroscopy
to characterize pigment-loaded polypropylene single fibres, Appl. Spectrosc., 48, 1387–1393.
Bourgeois, D. and Church, S. P., 1990. Studies of dyestuffs in fibres by fourier transform raman spec-
troscopy, Spectrochim. Acta Part A: Mol. Spectrosc., 46, 295–301.
Buzzini, P., Massonnnet, G., 2013. The discrimination of colored acrylic, cotton, and wool textile
fibres using micro-Raman spectroscopy. Part 1: in-situ detection and characterisation of dyes,
J. Forensic Sci., 58(6), 1593–1600
Fleishman, M., Hendra, P. J., and McQuillan, A. J., 1974. Raman spectra of pyridine absorbed at a
silver cathode, Chem. Phys. Lett., 26, 163–167.
Jochem, G. and Lehnert, R. J., 2002. On the potential of Raman microscopy for the forensic analysis
of coloured textile fibres, Sci. & Justice, 42, 215–221.
Keen, I. P., White, G. W., and Fredericks, P. M., 1998. Characterization of fibers by Raman micro-
probe spectroscopy, J. Forensic Sci., 43(1), 82–89.
Koens, F. S., 2013. The routine integration of Raman spectroscopy into the examination of black cot-
ton, Thesis (Hons.), University of Canberra
Kokot, S., Tuan, N. A., and Rintoul, L., 1997. Discrimination of reactive dyes on cotton fabric by
Raman spectroscopy and chemometrics, Appl. Spectrosc., 51(3), 387–395.
Lang, P. L., Katon, J. E., O’Keefe, J. F., and Schiering, D. W., The identification of fibers by Infrared and
Raman microspectroscopy, Microchem. J., 1986. 34(3), 319–331.
Lepot L., De Wael K., Gason F., and Gilbert B. (2008), Application of Raman spectroscopy to forensic
fibre cases, Sci. & Justice, 48(3), 109–117.
Massonnet, G., Buzzini P., Jochem, G., Fido, L., Beattie, R., Bell, S., Stauber, M., Coyle, T., Roux, C.,
Thomas, J., Leijenhorst, H., Van Zanten, Z., Wiggins, K. G., Russell, C., Chabli, S., Rosengarten,
A., Marshall, L., Bartick, E., and Murphy, M., 2004. Raman Group 2004—Green Cotton, 12th
EFG Meeting. Prague, Czech Republic.
Massonnet, G., Buzzini, P., Jochem, G., Stauber, M., Coyle, T., Roux, C., Thomas, J., Leijenhorst, H.,
Van Zanten, Z., Wiggins, K., Russell, C., Chabli, S., and Rosengarten, A., 2005. Evaluation of
Raman spectroscopy for the analysis of coloured fibres: A collaborative study, J. Forensic Sci.,
50, 1028–1038.
Massonnet, G., Buzzini, P., Monard Sermier, F., Jochem, G., Fido, L., Bell, S., Marshall, L., Stauber,
M., Coyle, T., Roux, C., Thomas, J., Leijenhorst, H., Van Zanten, Z., Wiggins, K., Russell, C.,
Drummond, P., Chabli, S., Rosengarten, A., Meile, C., Ketterer, S., Blumer, A., and Zingg, C.
2005. Raman trial 2005: Reactive dyes on cotton fibres. Proceeding of the ENFSI European Fibre
Group Meeting, Bled, Slovenia.
Massonnet, G., Buzzini, P. Lepot, L., Gason, F., DeWael, K., Hemmings, J., Jochem, G., Van
Zanten, Z., Van der Weerd, J. Arnold, J., Blumer, A. Palmer, R. Ketterer, S., amd Rosengarten,
A. 2007. EFG Raman trial 2007, Proceeding of the ENFSI European Fibre Group Meeting,
Madrid, Spain.
Massonnet, G., Buzzini, P., Monard,, F., Jochem, G., Fido, L., Bell, S. and Stauber, M. et al. 2012.
Raman spectroscopy and microspectrophotometry of reactive dyes on cotton fibres: Analysis
and detection limits, Forensic Sci. Int., 222, 200–207.
Miller, J. V. and Bartick, E. G., 2001. Forensic analysis of single fibers by Raman spectroscopy, Appl.
Spectrosc., 55, 1729–1732.
Raman, C. V. and Krishnan, K. S., 1928. A new type of secondary radiation, Nature., 121, 501–502.
Thomas, J. M., 2005. Raman spectroscopy for the analysis of reactively dyed cotton. PhD thesis,
University of Technology, Sydney, Australia.
Further Reading
General
Lin-Vien, D., Norman B. Colthup, N. B., Fateley, W. G., and Grasselli, J. G, 1991. The Handbook
of Infrared and Raman Characteristic Frequencies of Organic Molecules. California: Academic
Press.
McCreery, R. L., 2000. In J D Winefordner (Ed.), Raman Spectroscopy for Chemical Analysis (Vol.
157). New York, Chichester, Weinheim, Brisbane, Singapore, Toronto: Wiley-Interscience
Publication.
Smith, S. and Dent, G., 2005. Modern Raman Spectroscopy—A Practical Approach. England: John
Wiley and Sons Ltd.
Raman for Identifying Fibre Class and Subclass

Coyle, T. and Fairchild, J., 2003. Raman spectroscopy of cellulosic fibres, Forensic Sci. Int., 136(1),
125.
Coyle, T., Fairchild, J., Feilden, C., and Revell, D., 2007. Raman microspectroscopy and its place in
forensic fibre examination—The identification of man-made cellulosic fibres, Global Forensic
Science Today, Issue 3, 26–31.
Edwards, H. G. M., Farwell, D. W., and Webster, D., (1997). FT Raman microscopy of untreated
natural plant fibres, Spectrochimica Acta Part A, 53, 2383–2392.
Raman for Analysing Dyed/Pigmented Fibres/Textiles

Buzzini, P., 2007. L’application de la spectrométrie Raman en criminalistique pour l’analyse du col-
orant des fibres textiles en acrylique, coton et laine. PhD thesis, School of Forensic Science,
University of Lausanne, Switzerland.
Coupry, C., Sagon, G., and Gourget-Ballesteros, P., 1997. Raman spectroscopic investigation of bBlue
Contemporary Textiles, J. Raman Spectrosc., 28, 85–89.
Lepot, L., 2006. In situ study of Cotton Dyed Fibers by Raman Spectroscopy. ENFSI European Fibres
Group (EFG) Annual Meeting, Rome, Italy.
Thomas, J., Buzzini, P., Massonnet, G., Reedy, B., and Roux, C., 2005. Raman spectroscopy and the
forensic analysis of black/grey and blue cotton fibres: Part 1. investigation of the effects of vary-
ing laser wavelength, Forensic Sci. Int., 152, 189–197.
SERRS/SERS/RR
Munro, C. H., Smith, W. E., White, P. C., 1995. Qualitative and semi-quantitative trace analysis
of acidic monoazo dyes by surface enhanced resonance Raman scattering, Analyst, 120,
993–1003.
Umemura, J. and Park, S. R., 2005. Resonance Raman spectra and tautomeric structures of a surface-
active azo dye, CI Acid Red 138, adsorbed at the aqueous solution-air surface, Vib. Spectrosc.,
38, 29–32.
White P.C., 2000. SERRS spectroscopy—A new technique for forensic science? Sci. & Justice, 40,
113–119
White P. C., Munro, C. H., and Smith, W. E., 1996. In situ surface enhanced resonance Raman scat-
tering analysis of a reactive dye covalently bound to cotton, Analyst, 121(6), 835–838.
Other Instrumental Approaches
to Fibre Examination
CLAUDE ROUX
RACHEL MORISON
9
PHILIP MAYNARD
Contents
9.2 Fibre Identification by Pyrolysis Techniques 311
9.2.1 Pyrolysis Mechanisms 311
9.2.2 Types of Pyrolyzers 312
9.2.3 Types of Detectors 315
9.2.4 Pyrolysis-Mass Spectrometry (Py-MS) 315
9.2.5 Non-Discriminating Pyrolysis–GC 316
9.2.6 Concluding Comments and Future Directions about Pyrolysis–GC 316
9.3 Isotope Ratio Mass Spectrometry (IRMS) 317
9.3.1 Isotope Ratios 317
9.3.2 Instrumentation 318
9.3.3 Sample Introduction 318
9.3.4 Application to Fibre Analysis 318
9.4 Scanning Electron Microscopy and Elemental Analysis 319
9.4.1 Scanning Electron Microscopy as an Imaging Tool 319
9.4.2 Elemental Analysis of Fibres 322
9.4.2.1 Source of Inorganic Constituents 322
9.4.2.2 Methods for Elemental Analysis of Fibres 323
9.4.2.3 Concluding Comments and Future Directions about Elemental
Analysis 326
9.5 High Performance Liquid Chromatography (HPLC) 328
9.5.1 Basic Instrumentation 328
9.5.2 Types of HPLC 329
9.5.2.1 Normal Phase Chromatography 329
9.5.2.2 Reversed Phase Chromatography 329
9.5.3 Detection Systems 330
9.5.3.1 Monochromatic Detectors 330
9.5.3.2 Photodiode Array (PDA) Detectors 330
9.5.4 Application of HPLC for Dye Analysis 330
9.6 Capillary Electrophoresis 331
9.6.1 Capillary Electrophoresis Principles 332
9.6.2 Capillary Zone Electrophoresis (CZE) 333
309
9.6.3 Micellar Electrokinetic Capillary Chromatography (MEKC) 334

9.6.4 Mass Spectrometry 335
9.6.4.1 Triple Quadrupole Mass Spectrometer 336
9.6.4.2 Quadrupole Time-of-Flight (QTOF) Mass Spectrometer 336
9.6.5 Application of CE for Dye Analysis 337
9.6.6 CE-ESI-MS 338
9.6.7 Advantages and Disadvantages of Capillary Electrophoresis 338
9.7 Conclusions 339
References 340
9.1 Introduction
The objectives of a typical forensic analysis of fibres are (1) to identify the fibre type and
(2) to compare one or several fibres of an unknown origin to fibres from a known source
to determine if they could share a common origin. Collectively, the pool of techniques
employed should fulfil the needs for accurate identification and a high degree of discrimi-
nation between similar fibres. In a particular case, choosing appropriate techniques from
the wide range available will depend on several factors including the quantity of material
available, the exact circumstances of the case under investigation and the instrumental
techniques available to the laboratory. As described earlier, a number of techniques are
well established and commonly used in forensic science laboratories worldwide. Examples
include microscopic (see Chapter 5) and microspectrometric (see Chapter 6) examinations,
infrared (IR) spectroscopy (see Chapter 7) and Raman spectroscopy (see Chapter 8). While
these techniques meet the requirements expressed above in general casework, the context
of the case, the nature of the fibre specimens and the questions being asked sometimes
justify the application of other techniques. In parallel, new potential techniques are con-
stantly being developed by the broad analytical scientific field and ought to be presented to
the forensic fibre examiner.
This chapter discusses instrumental approaches that are not presented elsewhere. It is
not meant to be comprehensive and does not include all ‘exotic’ techniques, but the discus-
sion will focus on the techniques that have been shown to have practical value or the ones
with the best potential to add value to a typical forensic examination of fibres. For the ease
of presentation, they will be classified as follows:
Techniques primarily focusing on the fibre:
• Pyrolysis techniques
• Elemental and micro-structural analysis
• Isotope ratio mass spectrometry
Techniques primarily focusing on the fibre dyes:
• High-performance liquid chromatography

• Capillary electrophoresis methods, including hyphenated with advanced mass-
spetrometry techniques
Other Instrumental Approaches to Fibre Examination 311
9.2 Fibre Identification by Pyrolysis Techniques
As described in Chapters 6 and 7, the comparison of fibres, in particular synthetic fibres, is

usually performed by comparing the colour using microspectrophotometry and by quali-
tatively characterising the polymer composition using Fourier transform infrared (FT-IR)
spectroscopy. However, when colourless fibres are encountered and/or a higher level of
discrimination is required, an additional technique is required. One technique that has
been explored is pyrolysis. This technique is based on the high-temperature fragmentation
of a substance in an inert atmosphere. The molecular fragments produced by this ther-
mal decomposition are usually characteristic of the composition of the macromolecular
material being examined. The pyrolysis products are generally detected and identified by
coupling the pyrolysis unit to a gas chromatograph (GC), a mass spectrometer (MS) or a
combination of both (Py-GC-MS). Analytical pyrolysis is sensitive and gives good dis-
crimination, involves minimal sample manipulation and generates reliable and reproduc-
ible results.
In particular, pyrolysis–GC is a well-established technique in forensic analysis often
used by forensic chemists to analyse a vast array of polymeric materials (e.g. analysis of ton-
ers, inks, tapes, adhesives, paint and drugs, in addition to synthetic fibres). The technique has
been studied for various fibre types (Armitage et al., 2001; Blackledge, 1992; Challinor, 1999;
Górecki and Poerschmann, 2001; Hacaloglu, 2012; Lattimer, 1997; Stauffer, 2003). The main
advantage of the technique is that it can quickly differentiate most fibre types.
Examples of results obtained by pyrolysis–GC are provided below, illustrating the
clear differentiation at the sub-class level (Figure 9.1).
9.2.1 Pyrolysis Mechanisms

The degradation of a molecule that occurs during pyrolysis is caused by the dissociation
of a chemical bond and the production of free radicals. The Pyrolysis products reflect the
molecular structure, free radical stability, substitution and internal rearrangements of the
polymers constituting the sample material (Wampler, 2006)
There are three main pathways for thermal degradation of polymers. In some cases,
more than one mechanism occurs during pyrolysis of the polymer (Challinor, 1999;
Özgen, 2012; Sobeih et al. 2008).
Random chain scission involves the random breaking of the polymer’s C-C bonded
backbone as all the bonds are of equal strength. This results in the formation of products
including alkanes, alkenes and alkadienes of smaller sizes. An example of random chain
scission is the thermal degradation of polyethylene which forms a homologous series of
alkanes, mono-olefines and di-olefines (Figure 9.2)
Side-group scission occurs when the side groups attached to the backbone are broken
away resulting in the backbone becoming polyunsaturated. The backbone chain then frag-
ments and cyclization takes place to yield aromatic compounds such as benzene, toluene,
ethylbenzene, styrene and naphthalene. Polymers such as styrene, vinyl and some rub-
bers will undergo side-group scission. For example in the side group scission of polyvinyl
chloride (PVC) the side groups are eliminated to form hydrochloric acid (HCl) leaving the
remaining polyene macromolecule to form aromatic molecules (Figure 9.3).
Nylon 6
Nylon 6.6
Qiana
4 8 12 16 20 24 28 32 36
Minutes
Figure 9.1 Pyrolysis profiles of nylon 6, nylon 6.6 and Qiana.
During monomer reversion (depolymerisation) the polymer unzips and reverts to

its original monomeric version. Usually, only one predictable compound is produced.
Polymers known to undergo this mechanism include polymethylmethacrylate, polytetra-
fluoroethylene, poly-methylstyrene and polyoxymethylene (Figure 9.4).
9.2.2 Types of Pyrolyzers

The three most commonly used and recognized pyrolyzers for GC are the micro furnace,
Curie-point and resistively heated filament (Sobeih et al., 2008).
The micro furnace rapidly raises the temperature of the sample until it reaches the
pyrolysis temperature and then it remains at this temperature. The samples are introduced
into the pyrolysis zone using a liquid syringe, solid plunger syringe or by using a small cup.
The disadvantages of this method are that there are problems with reproducibility and
a rise time of several seconds is needed (Fifield and Kealey, 2000). Care also has to be taken
that the pyrolysis zone is free from any contaminants from previous experiments.
The Curie-point pyrolyzers accurately reproduce pyrolysis conditions using ferromag-
netic metals. This technique employs the inductive heating of a pure metal or an alloy
H H H H H H H H
... C C C C C C C C ...
H H H H H H H H
... ...
H H H H H H H H
... C C C C C C C C ...
H H H H H H H
H ... C ...
C
H
H H H H H H H H
... C C C C C C C C ...
H H H H H H H
H H H
H C C C ...
H H H
H H H H H H H Saturated hydrocarbon
H
... C C C C C C C C ...
H
H H H H H H
Radical Unsaturated hydrocarbon
Figure 9.2 Random chain scission of polyethylene. (From Stauffer, E., Sci. Justice, 43, 29–40,
2003.)
wire using a high-frequency oscillating current in a coil surrounding the wire which holds
the sample. The wire will heat until it reaches its Curie point temperature (the exact tem-
perature where the ferromagnetic material loses its magnetism) and then remains at that
temperature until the system turns off. The rise time of this technique is much faster than
the micro furnace method, however, the pyrolysis temperatures are limited to the Curie-
points of available materials (Parsi et al., 2007).
Filament pyrolyzers can acquire a controlled pyrolysis temperature extremely quickly
by using a piece of resistive metal. The metal heats rapidly at a high voltage until it reaches
the programme pyrolysis temperature. This temperature can then be maintained by reduc-
ing the voltage. Samples that are soluble in a volatile solvent are pyrolysed using a ribbon
probe. In this system, the sample is added directly to the ribbon probe. Samples that are
not soluble in volatile solvents are placed in quartz tubes and heated using a coil probe
(Blackledge, 1992). The advantages of this method are that it has very good reproducibility
(if the sample size and positioning are kept constant), it has a very short rise time, and the
pyrolysis temperature can be varied continuously from about 200°C–1000°C. The heated
filament and Curie-point pyrolysis also produce fewer secondary pyrolysis products, and
thus the pyrograms produced are easier to interpret.
H Cl H Cl H Cl H Cl H Cl
C C C C C C C C C C
H H H H H H H H H H
–HCL
C C C C C C C C C C
H H H H H H H H H H
Aromatics
Benzene Toluene Ethylbenzene Styrene Phenylethyne Naphthalene
Figure 9.3 Side group scission mechanism of polyvinyl chloride. (From Stauffer, E., Sci. Justice,
43, 29–40, 2003.)
A newer method of pyrolysis is laser micropyrolysis GC-MS, an analytical technique

that dates back to the early 1970s, however most studies since then have focused on instru-
mental development rather than its applications. Laser micropyrolysis GC-MS offers
several advantages over traditional methods of analytical pyrolysis. First, it only needs
a very small sample size as the laser output can be focused on specific areas. The technique
also has relatively high sensitivity as there is an optimum transfer of pyrolyzates to the
GC. In terms of its application to the forensic examination of fibres the technique shows
promise as it can detect over 100 pyrolyzates that can be used to easily distinguish samples
(Armitage et al., 2001).
However, there are still several disadvantages to the technique. First, the analysis is
limited by the laser used. Armitage’s research showed that synthetic fibres are transparent
to IR radiation and so additional laser sources (e.g. UV or visible lasers) may be needed to
enhance the analytical scope of the method. However, this study also found that the fibres
could be indirectly pyrolysed when embedded in a graphite matrix. The reproducibility
limitations of the technique also need to be rectified before the technique can be reliably
used in case work.
CH3 CH3 CH3

... CH2 C CH2 C CH2 C ... Polymer
COOR COOR COOR
Beta
scission
CH3 CH3 CH3

... CH2 C + CH2 C + CH2 C ..
COOR COOR COOR
Monomer
Figure 9.4 Schematic diagram of monomer reversion mechanism of polymethylmethacrylate.

(From Stauffer, E., Sci. Justice, 43, 29–40, 2003.)
9.2.3 Types of Detectors

The purpose of a detector used in conjunction with pyrolysis–GC is to monitor the car-
rier gas as it leaves the column and respond to changes in its composition as solutes are
eluted (Sobeih et al., 2008). An ideal detector has a rapid response time, high sensitivity
and a wide range of linear response.
The most commonly used detection system in pyrolysis–GC is the flame ionization
detector (FID) in which the pyrolysis products are identified by comparing their retention
times with those of known standards. While the FID is used for detecting all compounds,
other detectors have been designed to measure specific compounds. For example, the
alkali flame ionization detector (AFID) specifically measures nitrogen- and phosphorus-
containing compounds. The flame photometric detector (FPD) is used for monitoring sul-
phur- and phosphorus containing compounds and the electron capture detector (ECD) is
specific for halogenated hydrocarbons and other electron-accepting organic compounds.
An infrared spectrometer can also be coupled with pyrolysis–GC to detect and iden-
tify compounds. The pyrolysis products are collected using both the light pipe and matrix
isolation techniques. The advantage of the FTIR detector is that it has a sensitivity of low
nanogram levels and it can identify structural isomers (Kuckuk et al., 1994).
The most favoured and modern detector is the mass spectrometer (MS) as it is sensi-
tive, reproducible and can be used for both quantitative and qualitative analysis. Electron-
impact (EI) mass spectral libraries make it possible to rapidly identify many of the pyrolysis
products allowing for the characterisation of polymers, the identification of additives
and contaminants in the polymers and the determination of degradation mechanisms.
Chemical ionization can be used to give molecular weight/molecular formula information
of pyrolysis products as it produces molecular ions in greater abundance.
9.2.4 Pyrolysis-Mass Spectrometry (Py-MS)

Another technique for the analysis of pyrolysis products is direct Pyrolysis–Mass
Spectrometry (Py–MS). In this method, the fragments generated by pyrolysis are directly
introduced into the ionisation source of the MS. Total Py-MS experiments can be
performed in a few minutes. A large number of Py-MS configurations can be designed by
varying pyrolyser-MS coupling (off-line; on-line), pyrolyser type (resistively and induc-
tively heated devices, micro-furnace, laser pyrolysers), on-line mode (in-source, near-
source, in front of the source), mass spectrometer type (QMS, QQQ, ToFMS, FT-ICR-MS),
ionisation mode (EI, LVEI, CI, APCI, ENCI, FI, FD, FAB, PI), and heating mode (flash
vs. temperature resolved pyrolysis) (Armitage et al., 2001). This technique is useful for
the determination of composition, microstructure and additives of industrial polymers
(Hacaloglu, 2012).
One main advantage of direct pyrolysis mass spectrometry, DP-MS, is that it is the
only technique in which secondary and condensation reactions are at least partly avoided
(as thermal degradation occurs inside the mass spectrometer and the pyrolyzates are
rapidly transported from the heating zone to the source region and ionized before fur-
ther reactions can occur). The detection of high-mass pyrolyzates and unstable thermal
degradation products is also possible as the high vacuum inside the mass spectrometer
favours vaporization and the mass spectrometer is rapid enough to detect products before
further degradation (Hacaloglu, 2012). These features allow a better understanding of a
better understanding of the thermal characteristics, polymerization, crosslinking and char
formation processes.
The choice of ionisation method for Py-MS is important. Although electron ionization
(EI) is useful as it allows comparison of results to commercial mass spectral libraries,
it is recommended to use weaker ionisation methods. It is also not a very reproducible
technique as the fragmentation changes depending on the instrumental conditions. As the
most characteristic information is obtained from the molecular ions of high mass thermal
degradation products, a technique is required that will have high yields and high sensitivity
for high mass ions. Soft ionisation modes, such as chemical ionization (CI), FI, FD and PI,
are therefore preferred as they prevent fragmentation and generate mainly molecular ions.
9.2.5 Non-Discriminating Pyrolysis–GC

In the pyrolyzers mentioned previously the resulting pyrolyzates may not get flushed from
the pyrolysis zone fast enough, which might potentially result in secondary reactions (e.g.
defunctionalisation, release of low molecular weight analytes including CO and CO2, for-
mation of carbonaceous residue, etc.). Some pyrolysis products (especially ones with high
molecular weights) may not be transferred from the pyrolysis zone to the analytical sys-
tem, creating difficulties in the analysis of the pyrolysis products. To combat this problem,
a new technique was introduced in 2001 to minimize the discrimination of high-molecular
weight pyrolysis products (Górecki and Poerschmann, 2001). This technique involves inert
gas being pushed through the sample to flush all the pyrolyzates from the hot zone and
potentially reducing the occurrence of secondary reactions. Studies have shown that this
technique has improved recovery of higher molecular weight pyrolyzates; however, studies
into its applications are still limited.
9.2.6 Concluding Comments and Future Directions about Pyrolysis–GC

Despite the significant advantages described above, pyrolysis–GC is not universally used in
routine analysis by forensic science laboratories. There are still a number of disadvantages
that limit the technique’s application for fibre examination. Obviously, the technique is
destructive and so should only be performed if there is enough of the specimen to allow
for it. While only a small amount is needed (equivalent to a single fibre of around 10 mm),
this may already constitute a problem in some cases. However, if there is enough specimen
the technique should not be ruled out. This is particularly true when trying to address the
challenging case of the analysis of colourless synthetic fibres. Structural variation of the
polymer can significantly affect the composition of the mixture of the pyrolysis products,
allowing such fibres to be differentiated at the sub-class level more easily compared to
other methods. For example, polyamides are readily distinguished. The method also has
the potential for the identification of additives.
Another disadvantage is that, due to the large number of parameters in Pyrolysis–GC
(such as the temperature and duration of the pyrolysis, the rate of temperature rise and
the pyrolysis chamber atmosphere), it is hard to obtain reproducible results (Badri, 2008).
As a result, challenges exist on inter-laboratory comparisons and databasing. It is how-
ever believed that combined developments in instrumental design, sample automation,
advanced mass spectrometry techniques and advanced clustering methods for the data
treatment will mitigate most of these drawbacks. There may be some resurgence for this
technique in the future.
9.3 Isotope Ratio Mass Spectrometry (IRMS)
One important aim of the forensic analysis of fibres is to distinguish one specimen of the
same substance from another very similar but coming from a different source. In the broader
forensic science field, one technique probably epitomised the potential and limitations to
ever reach this aim more in the last ten years: isotope ratio mass spectrometry (IRMS).
The value of this technique relies on the assumption that two substances which are chemi-
cally the same may have different stable isotope compositions if either their origin or his-
tory differ. Recent research demonstrated applications for the analysis of drugs, explosives,
glass, paints, papers, inks, plastics, adhesives and materials in general, including fibres
(Benson et al., 2006; Daéid et al., 2011; Gentile et al., 2015)
9.3.1 Isotope Ratios

Isotopes are different forms of a single element. They are atoms with the same number of
protons, but differing numbers of neutrons in their nuclei resulting in different mass num-
bers. All but 12 elements exist as mixtures of isotopes. Each element has a dominant light
isotope (e.g. 12C [carbon], 14N [nitrogen], 16O [oxygen], 32S [sulphur], and 1H [hydrogen]),
and one or two heavy isotopes (e.g. 13C, 15N, 17O, 18O, 33S, 34S, and 2H) with a natural
abundance of a few percent or less (Boutton, 1996). Variations in abundances of these iso-
topes in samples can be used to discriminate to the extreme (some may say ‘individualise’)
a large variety of forensic traces.
Early research on this technique focused on its usefulness for forensic analysis in
terms of its specificity and discrimination power. Jasper (2002) used an example where
the carbon isotope ratio values of a number of compounds within an ignitable liquid mix-
ture were analysed. It was found that by combining the specificities of a number of the
compounds, the overall specificity of this technique may be estimated. In this example, he
found that if 5 compounds each had 67 different isotopic variations and the carbon isotopic
composition of these compounds were measured, the combined specificity would be 1 in
1.4 billion (i.e. there would be approximately 1 in 1.4 billion chance that a random pet-
rol sample would be identical to the questioned sample. Meier-Augenstein and Liu (2004)
explored the discrimination power of the technique and found that the more isotope ratios
that can be measured in a sample, the more individualised the sample can be (i.e. more of
a ‘fingerprint’ can be made of the sample).
9.3.2 Instrumentation
A good practice guide detailing the instrumentation and set up of IRMS has been writ-
ten by the Forensic Isotope Ratio Mass Spectrometry (FIRMS) Network (Carter and
Barwick, 2011).
IRMS instruments are designed to detect small differences in the abundances of iso-
topes such as 2H/1H, 15N/14N, 13C/12C, 18O/16O, and 34S/32S. Before they enter the ion source,
these ions are converted into simple gases (hydrogen (H2), nitrogen (N2), carbon diox-
ide (CO2), carbon monoxide (CO), and sulphur dioxide (SO2)) which represent the isotope
composition of the original sample. The IRMS then measures the ratio of ions that corre-
spond to these gases.
9.3.3 Sample Introduction

There are two common sample introduction techniques for IRMS analysis. Both techniques
require that solid, liquid and gaseous samples are converted to pure gases (Table 9.1).
9.3.4 Application to Fibre Analysis

Currently, although there has been a lot of research into various forensic applications of
IRMS, there has been very little research into its application in the forensic analysis of
Table 9.1 IRMS Analysis–Comparison between Dual Inlet and Continuous Flow
Techniques
Dual Inlet IRMS (DI-IRMS) Continuous Flow IRMS (CF-IRMS)
Type of gas entering the Pure gases such as hydrogen Mixed gas. Helium carrier gas carries pure
mass spectrometer (H2), nitrogen (N2), carbon analyte gas into the ion source. (e.g. CO2 +
dioxide (CO2), or sulfur dioxide He)
(SO2)
Amount of sample Enough sample is needed to As Helium carrier gas creates viscous
required create viscous flow in the flow, a smaller amount of sample is
capillaries. needed. ≈100 nmol.
≈10 μmol of the sample is
needed.
How the analyte and Pure analyte gas and reference The pure analyte gas is introduced into the
reference gas are entered gas are alternated rapidly with system just once. The reference gas is
into the IRMS variable volume bellows that introduced before and/or after the analyte.
create reference signal
responses of equal intensity.
fibres. A study by Nic Daéid et al. (2011) looked at the 2H and 18O isotopic data of un-dyed
spun cotton fibres grown in a range of geographical regions and found there was potential
for isotopic analysis to be used as an aid to discriminate natural fibres from each other or
elucidate specimen history based on geographical origin. This potential was confirmed
with the successful use of the isotopic signature to discriminate between unprocessed U.S.
and non-U.S. cotton samples (Meier-Augenstein et al., 2014). These promising findings are
significant because the analysis of colourless cotton fibres is challenging as conventional
microscopic and chemical analysis of these fibres is generally unsuccessful because of their
similar morphology. This justifies more research in this area in the future.
9.4 Scanning Electron Microscopy and Elemental Analysis
For several reasons, fibre analysis protocols rarely include scanning electron microscopy
(SEM) and elemental analysis. This situation is probably justified for SEM as an imaging
tool. The often stated advantages of SEM, high magnification and large depth of field, bring
limited extra information in routine cases compared to the data obtained by standard ana-
lytical protocols. Except for some applications described below, the main interest of SEM
lies in its combination with suitable spectrometers to obtain an elemental X-ray micro-
analysis (electron probe microanalysis, EPMA). However, elemental analysis of fibres is no
longer restricted to SEM, as new techniques are emerging. Several studies have shown that
elemental analysis can be usefully employed to supplement the pool of existing methods
and to help achieve the highest possible degree of discrimination.
9.4.1 Scanning Electron Microscopy as an Imaging Tool

In SEM, high-energy electrons are focused into a fine beam, which is scanned across the
surface of the specimen. Complex interactions of the beam electrons with the atoms of
the specimen produce a wide variety of radiation products. The signals of greatest interest
are the secondary and backscattered electrons since these vary according to differences
in surface topography as the electron beam sweeps across the specimen. This radiation is
collected by a detector (scintillator–photomultiplier detector), and the resulting signal is
amplified and displayed on a cathode ray tube, on a television screen or by digital imaging.
In most cases, interpretation of the result is quite straightforward.
It is beyond the scope of this chapter to discuss SEM technology in detail; suffice it
to say that SEM has obvious advantages as an imaging tool thanks to the high resolu-
tion that can be obtained (1.5 nm for commercial below lens instruments) and to the
three-dimensional appearance of the specimen with a large depth of field. The ability to
examine the specimen at high and continuous magnifications (up to 300,000 times for
below lens SEM) is another useful feature. The theoretical and practical aspects of SEM,
along with X-ray microanalysis, have been extensively covered by Goldstein et al. (1992).
For fibre examination, the preparation of the specimen is relatively simple. Single fibres
can be stretched between double-sided adhesive strips placed a few millimetres apart on a
polished carbon disc. The disc is then attached to a specimen stub using non-conducting
adhesive. The samples are generally dried and carbon coated under vacuum before exami-
nation to eliminate or reduce the electric charge that builds up rapidly in a non-conducting
Acrylic
Figure 9.5 Use of SEM to examine surface features: unusual scale-like structure on the
outer surface of Cashmilon bicomponent acrylic fibres. (With kind permission from Springer
Science+Business Media: Forensic Science Progress, Vol. 4, A. Maehley and R. Williams [Eds.],
Fibres and their examination in forensic science, Grieve, M.C., 1990.)
specimen when it is scanned by a beam of high-energy electrons. However, this step is no

longer necessary with the new generation of so-called environmental SEMs.
When applied to fibre examination, SEM is useful in revealing morphological fea-
tures of the surface, the cross-section or the fibre ends. For example, SEM was used by
Smalldon (1973), Grieve and Cabiness (1985) and Grieve et al. (1988) to reveal details of
the surface structure of acrylic and modacrylic fibres. Examples of distinct morphological
features revealed by SEM are shown in Figures 9.5 through 9.8.
Surface imaging using SEM as an aid in the identification of animal hair structure
has been reported (Robson, 1994). It can also help in the detection of trace debris on
fibres. The revelation of modifications of surface features such as washer/dryer abrasion
(Goynes, 1971) and acid-wash treatment of denim garments (Hecking and Erikson, 1993)
has also been reported. Some authors have examined fibre bonding in nonwoven fabrics
and shrink-proofing treatment of wool (Greaves and Saville, 1995).
Owing to the large depth of field of SEM, the examination of fibre extremities permits
an easy and rapid characterization of the cross-sectional shape in comparison to other
methods. This application is related to another area which has attracted a great deal of
interest: examining and characterizing fibre damage. Hearle et al. (1989) produced an atlas
of over 100 SEM micrographs depicting different sources and types of fibre damage. Some
authors suggested that SEM could assist in differentiating the causes of fibre damage: cuts
from tears (Stowell and Card, 1990), a knife from scissor cuts (Choudhry, 1987) and sharp
instruments from canine bites (Chaikin, 1986; Robinson, 1987). Some of these results were
challenged by Pelton (1995), who carried out an extensive study involving over 600 fibre
ends and observed overlapping characteristics for a scissor cut, knife cut and torn fabrics.
Pelton’s study also showed that fibre-end features could be attributed to or influenced by
various manufacturing processes. In his conclusion, Pelton suggested that SEM validity
and reliability should be verified using a blind review procedure and that this exercise
should include both textile and forensic scientists at different sites. It is beyond the scope of
this chapter to discuss this controversial area. Damage to textiles is covered in chapter 2 of
the book. However, it is worth mentioning that:
Modacrylic
Figure 9.6 Use of SEM to examine surface features: highly serrated surface of themodacrylic
fibre Kanekalon 7. (With kind permission from Springer Science+Business Media: Forensic
Science Progress, Vol. 4, A. Maehley and R. Williams [Eds.], Fibres and their examination in
forensic science, Grieve, M.C., 1990.)
Figure 9.7 Use of SEM to examine fibre damage: scissor-cut ends of polyester fibres from
woven fabric in a coat. (Courtesy of F.-P. Adolf, Bundeskriminalamt, Germany.)
• Caution should be exercised when interpreting fibre-end fracture morphology to

determine the cause of textile damage.
• Fibre-end fracture morphology as determined by SEM is only one aspect of
damage analysis, the examination being applied at the fibre level only. In most
cases, a proper analysis should also involve the examination of the damaged
area at the fabric and yarn levels using the naked eye and a stereo microscope
(Taupin, 1996).
Figure 9.8 Use of SEM to examine fibre cross-sections: cross-sections of trilobal nylon 6 carpet
fibres from Allied Signal. (Courtesy of D. Knoop, microscopist, Allied Signal, Petersburg, VA.)
9.4.2 Elemental Analysis of Fibres

About the elemental analysis of fibres in forensic science, the main questions are as follows.
• Can various fibres made of the same material be distinguished by their trace
element content?
• What is the method of choice for such an analysis?
• What is the optimum and minimum fibre length sufficient for such an analysis?
• Is extra discrimination gained compared to conventional techniques (microscopy,
MSP, FTIR, etc.)?
• Are the differences in the trace element content significant enough to determine a
manufacturing source?
• What is the influence of external or environmental contamination, and how can
this be minimized?
9.4.2.1 Source of Inorganic Constituents

Most textile fibres are organic polymers but contain inorganic constituents which come
from three main sources:
• Residues from the manufacturing process

• Additives/finishing agents
• Environmental contaminations
Manufacturing processes involve the use of water-soluble redox systems such as

polymerization initiators, solvents and co-monomers. They can be present in the form of
residues and contain elements that may be detected in the final product.
Many manufacturers also add elements to their final products or incorporate them
into the polymer matrix during the manufacturing process. These additives modify and/
or improve certain mechanical, physical or chemical properties of the fibre. They pro-
tect the polymer from the effects of light, heat and bacteria; change the density, lubricat-
ing and thermal properties; improve fire resistance and reduce the formation of smoke;
Table 9.2 Inorganic Additives Used by the Textile Industry

Additive Type Elementsa
Pigments Fe, Cr, Pb, Cd, Mn, Co
Delustring agents Ti, Mn, Al
Antioxidants, light stabilizers p, s
Flame retardants, flame suppressor Cl, Br, P, Sb, Al, Mg, Mo, B
Antistaining agents F
Antistatic agents K, Sn, Ti, Zr
Antibacteriological agents As, Cu, Zn, S
Lubricants Various metals
Metallic stabilizers Na, K, Mg, Ca, Sr, Ba, Zn, Cd, Sn, Pb, Sb
Fillers Ca, Ba, Al, Si, Mg, K
Markers (tracers) In, Sm, Er
a These elements may be present in different forms: oxides, hydroxides, salts, etc.
facilitate the dyeing process and reinforce colours; improve biodegradability and electrical
conductivity; etc. The additives employed depend on the intrinsic qualities of the polymer
base and the qualities desired in the final product. The principal elements employed by
the textile industry are summarized in Table 9.2. For further information see Katz (1990),
Kroschwitz (1990) or Edenbaum (1992).
As shown by Prior (1991) and Jallard (1995), the wearing or washing of garments may
modify the elemental profile of fibres. This can add uniqueness to these fibres. However,
it also means that wearing or washing a garment can constitute an important source of
environmental contaminations. This must be kept in mind when interpreting localized
variations in elemental profiles.
9.4.2.2 Methods for Elemental Analysis of Fibres

Inorganic constituents are only present in minute quantities in single fibres. Therefore
extremely sensitive methods are required for their detection. These methods include:
• Neutron activation analysis (NAA)

• X-ray diffraction (XRD)
• Inductively coupled plasma techniques
• Inductively coupled plasma atomic emission spectrometry (ICP-AES)
• Inductively coupled plasma mass spectrometry (ICP-MS)
• Laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS)
• X-ray fluorescence techniques
• Electron probe microanalysis (EPMA, aka EDX)
• X-ray fluorescence spectrometry (XRF)
• X-ray fluorescence microanalysis (micro XRF, aka XRFM)
• Total reflection X-ray fluorescence spectrometry (TXRF)
9.4.2.2.1 Neutron Activation Analysis Neutron activation analysis is based on the
measurement of radioactivity that has been induced in samples by irradiation with neu-
trons or charged particles. The method offers several advantages including high sensitivity,
minimal and often nondestructive sample preparation and ease of calibration. As a result,
NAA attracted a great deal of interest for forensic applications in the 1970s, including fibre
examination. For example, Bush et al. (1973) indicated the presence of markers or discrim-
inating elements in samples of wool and acrylic fibres using NAA. However, the technique
never really entered into routine use because of the need for special facilities for handling
and disposing of radioactive materials.
9.4.2.2.2 X-ray Diffraction X-ray diffraction is the result of interference (both con-
structive and destructive) which takes place when X-rays are scattered by the ordered envi-
ronnment in a crystal. The analytical interest of this phenomenon is based on the fact that
an X-ray diffraction pattern is unique for each crystalline substance. If a ‘match’ can be
found between the pattern of an unknown and an authentic sample, chemical identity can
be achieved. By measuring the intensity of the diffraction lines and comparing them with
standards, it is also possible to achieve a quantitative analysis.
Lynch (1981) employed X-ray diffraction techniques for the analysis of different sam-
ples of animal, vegetable and synthetic fibres. Characteristic diffractions using CoKα
radiations indicated that it was possible to differentiate between different generic classes.
The fine bands produced by the polycrystalline delustring agents were employed as a point
of comparison. However, the technique proved to have a lower discriminating power than
more conventional techniques (FTIR, Py-GC). For this reason, the application of XRD to
fibre examinations is rather limited.
9.4.2.2.3 Inductively Coupled Plasma Techniques Inductively coupled plasma (ICP)
techniques are specialized emission spectroscopic techniques using an argon plasma as a
high-energy atomization source. They can achieve rapid and highly sensitive multielemental
analyses. Spectra for dozens of elements at the ppb level can be recorded simultaneously. In
ICP-AES instruments, the lines are detected and their intensities are measured by optical
detection using gratings and photomultiplier tubes. In ICP-MS, the ICP torch is linked
to a mass spectrometer. The spectra produced in this way consist of a series of isotope
peaks, the main advantage being a significant (approximately 10 × to 100 ×) increase in
detection limits for most elements. In a forensic context, these advantages are somewhat
negated by the sample introduction procedures which generally involve the digestion of
the sample. This drawback is overcome when using a laser ablation ICP-MS (LA-ICP-MS).
In LA-ICP-MS, a laser is used to ablate surface particles of a sample into a carrier gas that
are then transported into the plasma for analysis. In this case, the technique is relatively
non-destructive as it uses minimal amounts of sample (in the order of 300 ng).
Roux (1995) showed that ICP-AES was highly discriminating for the analysis of colour-
less polyester fibres. The results obtained permitted the classification of 77 different sam-
ples into 13 different groups. It was also observed that samples with the same trade names
were often classified in the same group. However, it was pointed out that the technique suf-
fered from two major disadvantages that make it impractical for routine forensic work: it is
totally destructive, and an unrealistic quantity (at the mg level) of fibres is required per
analysis.
Gallo and Almirall (2009) applied ICP-MS and LA-ICP-MS to the forensic elemen-
tal analysis of cotton fibres. LA-ICP-MS was found to be more precise and slightly faster
than the solution-based approach. The LA-ICP-MS method also had the advantages of very
low sample consumption compared with the solution-based method and the possibility of
re-analysing the samples. All of the raw cotton samples from different sources could be
distinguished from one another, as could all the cotton T-shirts.
Hiroma et al. (2010) used LA-ICP-MS to discriminate PET fibres obtained from
a variety of car trunk mats. It was possible to discriminate all 31 samples based on the ana-
lytical results of a single fibre by LA-ICP-MS combined with those of FT-IR and SEM-EDS.
LA-ICP-MS was found to have a good analytical sensitivity and required a much shorter
preparation time and a smaller sample size than any other conventional element analysis
method.
9.4.2.2.4 X-ray Fluorescence Techniques X-ray fluorescence techniques are based on
the emission of characteristic X-rays from the elements in the sample due to their excita-
tion by absorption of a primary beam of X-rays or electrons. XRF techniques can perform
rapid and nondestructive qualitative multielemental analyses of barely visible samples,
which constitutes a major advantage in a forensic context. Modern XRF instruments can
also produce quantitative analyses of complex materials. However, most XRF techniques
are not as sensitive as other methods described previously.
Andrasko et al. (1984) were able to differentiate polymer samples by EPMA by varia-
tions caused by the use of different additives. Grieve and Cabiness (1985) were also able to
show differences between acrylic and modacrylic fibres and within the class of modacrylic
fibres by chlorine and bromine contents. Unfortunately, the detection of markers such as
indium or samarium required an unrealistic quantity of fibres. It clearly appears that the
concentrations of most elements required for fibre comparison are too low to be readily
detectable by EPMA, although the technique can accommodate small sample sizes and is
relatively nondestructive.
Koons (1996) investigated the feasibility of measuring the elemental composition
of carpet fibres by XRF for purposes of classification and comparison. This study was
designed to address the question of whether elemental compositions can be used to classify
carpet fibres into two intended usage classes, automotive and residential. The usefulness of
elemental composition as a point of comparison for discriminating among similar fibres
was also evaluated. The method was found to be useful to characterize carpet fibres as
small as 5 mm in length. Attempts to classify fibres into automotive and residential use
were only partly successful. Several transition metals (Co, Cr and Zn) were generally pres-
ent in the automotive fibres and absent in the residential carpet fibres. Nylon fibres did
not contain metallic elements when intended for residential use. Koons concluded that
elemental analysis may afford some degree of additional discrimination and that micro
XRF could improve significantly on these results.
These two hypotheses were then confirmed by Cartier et al. (1997), who applied
micro XRF to the analysis of samples of colourless acrylic and polyester fibres which
were indistinguishable using bright-field, fluorescence and FTIR-microscopy. They used a
Kevex Omicron energy dispersive micro XRF system able to accommodate collimators of
100–20 μm in diameter. The aim was to determine whether this technique could be suc-
cessfully applied to single fibres of relatively fine titre and whether it would be beneficial to
add it to the existing sequence of techniques used to compare colourless fibres. The extent
of intra-garment variation and the possible effects of tape and mounting media residues
on the elemental analyses were also investigated. The results confirmed the high value
of fluorescence microscopy within the existing examination sequence and showed that
single-fibre analysis using micro XRF not only is feasible but improved the discriminating
power between colourless samples by about 50%.
Hyphenation of Raman with micro-XRF showed some promising results with a range
of trace evidence materials including fibres (Zieba-Palus et al., 2008).
Another type of XRF technique has attracted interest in forensic science: total reflec-
tion X-ray fluorescence (TXRF). TXRF differs from conventional dispersive energy XRF in
that the incident X-rays arrive near the critical angle of reflection without penetration into
the sample substrate. The sensitivity is greatly improved (in the picogram range) thanks to
a better signal-to-noise ratio. Another advantage of the technique is its ability to perform
quantitation using a single-element internal standard. TXRF appears to combine the much-
needed requirements of high sensitivity and the ability to analyze minute solid samples.
Only a few studies have been reported (see below); however, TXRF is seen as an excellent
method for the elemental analysis of individual fibres in forensic science. An example of
elemental profile comparison carried out using TXRF is shown in Figure 9.9.
Prange et al. (1995) analyzed a total of 35 colourless samples (single fibres and weighable
amounts) of wool, viscose and polyester using TXRF. Fibres as small as 3 mm in length
could be successfully analyzed. Quantitative comparison and identification were possible
using ratios of specific elements such as P, Ti, Mn or Sb, 23 out of 35 single fibres being
definitely assigned to one source. Contamination problems with omnipresent elements
such as Ca, Fe and Zn were also reported, especially when dealing with single fibres.
Buscaglia and Koons (1998a) applied TXRF to the determination of elemental
concentrations in various polymeric materials, including synthetic fibres. Elements such
as Cr, Co, Cu and Zn were found suitable for source discrimination purposes. In another
study, Buscaglia and Koons (1998b) investigated the application of TXRF to the analysis of
individual automotive carpet fibres. Samples selected for this study were composed of the
same polymeric material and were not readily distinguishable by macroscopic colour or
polarized light microscopy. Elements such as Cr, Co, Cu and Zn, when present, were good
elements for discrimination of fibres. Fe, Ca and Ti could also be suitable for discrimination
purposes; however, these elements are major components of environmental contamination
and are affected by sample handling and preparation.
9.4.2.3 Concluding Comments and Future Directions

about Elemental Analysis
As already stated, the elemental analysis of fibres is rarely, and then only briefly, mentioned
in forensic protocols. One reason for this is that no in-depth study has fully investigated
the actual value of fibre elemental profile analysis; for example, the level of variation within
and between garments of one type is yet unknown. The information to date has become
available through focused and necessarily limited research projects such as those described
above. However, these studies have brought valuable information, as a result of which the
following trends can be identified.
• Various fibres made of the same material (i.e. same polymer or same generic class)
can be distinguished by their trace element content.
• Specialized XRF techniques such as micro XRF and especially TXRF as well as
LA-ICP-MS appear to be promising methods for the elemental analysis of single
fibres. These methods fulfil the requirements of sensitivity and ability to deal with
minute solid samples, including single fibres as small as 5 and 2 mm in length
respectively.
16,000
Y
(1 ng) Mo tube lines
12,000
Si
Counts
8000 Fe
Ca
Zn
Ar
Co
4000 K
S Cu
Ti Cr Ni Pb Sr
Pb
0
0 2 4 6 8 10 12 14 16 18 20
keV
6000
Si Y
(1 ng)
Mo tube lines
Ar Fe
4000
Counts
Ca
Cu
2000 Zn
K Co
S
Ti Sr
Cr Ni
Pb Pb
0
0 2 4 6 8 10 12 14 16 18 20
keV
Figure 9.9 Similar but differentiable TXRF spectra of grey nylon carpet fibres from two differ-
ent automobiles of the same year, make and model. These spectra were acquired after plasma
ashing for 2000 s with Mo excitation (50 kV, 38 mA) using a thin filter and 1 ng Y as inter-
nal standard. (Courtesy of J. A. Buscaglia, Forensic Science Research and Training Center, FBI
Academy, Quantico, VA.)
• TXRF and micro XRF permit extra discrimination when applied to colourless
fibres in comparison with conventional techniques (microscopy, MSP, FTIR, etc.).
The increase in discrimination power has not yet been investigated when applied
to coloured samples.
• The data currently available are not extensive enough to confirm that the manu-
facturing source can be identified by the elemental profile to the exclusion of all
other sources in the world; however, matching elemental profiles are good indica-
tors of a common manufacturing source. When added to the other analytical data
available, and in conjunction with industrial enquiries, this information may lead
to a positive identification.
• When dealing with highly sensitive techniques, such as those needed for the elemen-
tal analysis of fibres, it is very important to use appropriate plastic micro-tools to
minimize external contamination. Environmental contamination due to wearing
or washing are generally unavoidable. As a result, extra care must be taken when
interpreting localized variations and/or when the known sample (e.g. suspect’s gar-
ment) has been sequestred a long time after the transfer of the unknown specimen
(e.g. fibres found on the victim).
9.5 High Performance Liquid Chromatography (HPLC)
High-perfomance liquid chromatography (HPLC) is one of the most widely used analyti-
cal techniques for the forensic analysis of fibre dyes. Dyes play an important role in the
identification and comparison of fibres in forensic casework as it provides information
of great value in the comparison of single fibres, more than that allowed by comparison
of fibre type and characterization. Two other methods used in the analysis of fibre dyes
are thin-layer chromatography (TLC) and microspectrophotometry; however, both these
methods have limitations that are not present with HPLC. Microspectrophotometry has
difficulties analyzing dark color fibres and can also give complementary chromaticity
coordinates for different dyes which can be disadvantageous when scanning stored spec-
tral data where the original sample is not present. TLC has difficulties when it comes to
analysis as it requires relatively large quantities of dye (which may not be possible with
light colored fibres or the limited amount of fibre samples that can be present in a forensic
context), requires different eluent systems for various dye classes, and provides only a
semi-permanent record. HPLC has a number of potential advantages over TLC such as the
better resolution of the dyes, quantitation and reproducible retention times of separated
components. HPLC is applicable to a wide range of analytes such as amino acids, proteins,
nucleic acids, carbohydrates, drugs, pesticides and steroids. In forensic science, HPLC has
been used extensively as an analytical technique for the qualitative and quantitative analysis
of drugs and metabolites, organic and inorganic explosives, marker dyes and inks as well
as fibres.
9.5.1 Basic Instrumentation

The basic instrumentation that makes up the HPLC system is:
• The solvent reservoir contains the mobile phase. The mobile phase is typically
made up of mixtures containing water and an organic modifier (such as acetoni-
trile or methanol and/or buffer solutions) however, more complex mobile phases
are needed for more complex separations. If the composition of the mobile phase is
constant, the method is termed isocratic. The composition of the mobile phase can
also be programmed to change during the separation in a technique called gradi-
ent elution. This technique is useful if the same contains a compound with a range
of polarities or chemical properties, something often seen in fibre dyes.
• The pump forces the mobile phase through the column at high pressure and is mon-
itored and controlled by the pressure regulator. Different pump types are available;
binary and quaternary pumps are the most frequently employed. A binary pump
can drive two mobile phases at the same time, whereas a quaternary pump can
drive four different mobile phases at the same time if required. The pump must
be able to generate high pressures, deliver flow rates from 0.1 to 10 mL min–1,
have a reproducibility of 0.5% (or better) and be resistant to corrosion by a variety
of solvents.
• The injector port allows controlled the introduction of the liquid sample into the
HPLC instrument via a rotating valve system.
• The column contains the packed stationary material. The components of a mixture
are resolved in the column depending on their selectivity towards either the mobile
phase or the stationary phase. Columns can vary in length, packing particle size,
and internal diameter.
The separated compounds are sequentially eluted from the column and identified by
means of a detection system. The detector type will vary depending on the application with
the most common detector being diode array UV/Vis though other detectors can be used
such as electrochemical and mass spectrometry detectors. The detector chosen will depend
on the type of chromatography and the analyte being detected. For example, an electro-
chemical detector is needed when performing ion exchange chromatography as a UV/V
is detector cannot detect.
9.5.2 Types of HPLC

9.5.2.1 Normal Phase Chromatography
Normal phase HPLC (NP-HPLC) separates analytes based on their polarity. NP-HPLC
uses a polar stationary phase and a non-polar mobile phase. The packing materials within
the normal phase columns are composed of unmodified silica spherical beads (cyano,
amine, or diol packing materials can also be used) with the mobile phase consisting of
nonpolar organic solvents such as ethanol, chloroform, propanol, or hexane. The polar
analyte interacts with and is retained by the polar stationary phase with the adsorption
strength increasing with the increased analyte polarity (thus increasing the elution time).
In the early days of HPLC, most separations were performed on columns packed with
silica particles and a non-polar mobile phase such as hexane. However, this method has
problems in coping with the water content of the samples and the mobile phase, which can
lead to separation problems.
9.5.2.2 Reversed Phase Chromatography

Reversed phase HPLC (RP-HPLC or RPC) has a slightly polar aqueous mobile phase and
a non-polar stationary phase. Generally, the packing material is composed of silica beads
coated with hydrophobic alkyl chains of varying length. Typical coating materials are
C4, C8, and C18; the longer the chain is, the more nonpolar the stationary phase will be.
RPC operates on the principle of hydrophobic interactions, which result from repulsive
forces between a polar eluent, the relatively non-polar analyte and the non-polar station-
ary phase. The binding of the analyte to the stationary phase is proportional to the contact
surface area around the non-polar segment of the analyte molecule upon association with
the ligand in the aqueous eluent.
9.5.3 Detection Systems

The choice of the detection system is crucial to the effective trace analysis of fibre dyes
which have a distinctive ultraviolet/visible (UV/vis) spectrum.
9.5.3.1 Monochromatic Detectors

Monochromatic (single-wavelength) detectors are sensitive enough to detect dyes
extracted from casework sized fibres. However most fibres are coloured with a mixture
of different dyes. In these cases, the dye extracts will need to be identified with a number
of different wavelengths to detect the different components. The wavelengths usually
chosen are 400, 500 and 600 nm as this increases the chance of detecting any yellow,
red or blue dyes at or close to their respective wavelength maxima, and can potentially
give information about the relative quantities of the dyes present in the fibres analysed.
However, this is a lengthy process, and there is no guarantee that the wavelength used
will be the wavelength maxima of the dye which may result in some dyes not being
detected (Griffin et al. 1994).
9.5.3.2 Photodiode Array (PDA) Detectors

Photodiode array (PDA) detectors can be used to resolve this problem. These detectors are
capable of obtaining a complete UV and visible spectrum (200–800 nm) of each eluting
peak within a single injection, cutting down on run times and sample dilution.
There are several advantages to PDA as, in combination with HPLC, PDA detectors
can produce a full spectral analysis. As a result of this, the visible spectra of each dye
component in the mixture can be compared to spectra of standard dyes, etc. to assist iden-
tification. PDA also allows for immediate qualitative assessment of peak homogeneity and
for the rapid differentiation of spectra which allows for subtle differences in chromato-
grams of mixtures to be seen that could have been found using a fixed wavelength detector.
(Tebbett, 1991). It allows the differentiation of compounds based on their absorbance ratios
(White and Catterick, 1987)
9.5.4 Application of HPLC for Dye Analysis

A number of different dye classes are used to colour fibres, and within each class of dye,
there can be many widely differing chemical structures which can be divided into three
main groups: acidic, basic and neutral. In order for an HPLC system to be developed, a
number of factors had to be taken into account such as the structure of the dyes, variation
within the dye classes, the extraction and potential degradation of the dyes, the compo-
nents of the system (column, detector, sensitivity), unresolved dyes and the time required
for the analysis (Griffin and Speers, 1999).
It is difficult to separate the three dye groups on a single chromatographic system, so a
lot of research has been undertaken to develop optimum HPLC techniques for the separa-
tion of various dye classes. The main problems encountered during dye extraction are that
the solvents need to be able to remove all dyes from a fibre efficiently, reproducibly and with
the minimum of degradation of the dye or the fibre. However, in some cases, the only solvent
that meets these needs is not compatible with the HPLC column. Chen et al. (1998) used
a methanol and aqueous ammonium acetate mobile phase to separate azo sulfonated acid
dyes and azo, and anthraquinone acid dyes were separated by Holčapek et al. (1999) with a
similar gradient. Griffin et al. (1994) developed an isocratic solvent system for the separation
of an unknown mixture of basic dyes. Laing et al. (1988) developed a complex mobile phase
consisting of methanol, water, tetramethylammonium chloride, and ammonium dihydro-
genphosphate to separate acid dyes while Speers et al. (1994) used acidified acetonitrile and
water for the separation of acid, basic, and disperse dyes. Other research has shown that a
methanol and water mobile phase could be used to separate disperse, acid, basic, vat, and
direct dyes (Huang et al. 2004; West, 1981; Yinon and Saar, 1991).
Due to the small sample sizes of fibres collected in forensic casework, it is important
that an analytical technique is sensitive enough to provide information. Laing (1988)
determined that a fibre 2–10 m min length contains around 2–200 ng with lighter shades
containing far less. It was also found that if the dye is made up of mainly one component it
can be detected readily at 5 ng level, but any minor constituents are not detected. However,
if it contains at least two major components, then detection is only possible at around
50 ng. The sensitivity of HPLC was also found to increase when the column volume was
decreased (Griffin et al., 1988; Wheals et al., 1985). Speers et al. (1994) analysed standard
dye mixtures ranging from 0.5 to 25 ng using 5 μL injections to determine the LOD of
HPLC. An LOD range was found of 40–800 pg for basic dyes, 230–1250 pg for acid dyes,
and from 80 to 1280 pg for disperse dyes.
Using mass spectrometry detection greatly improves the sensitivity and hence the
wider application and usefulness of such analyses. For example, Huang et al. (2005) suc-
cessfully used liquid chromatography-mass spectrometry (LC-MS) for the discrimination
of textile dyes that could not easily be distinguished based on their UV/visible absorp-
tion profiles. LC-MS was also capable of discriminating between two pairs of cotton fibres
which were indistinguishable by microspectrophotometry.
Petrick et al. (2006) could separate and identify a mixture consisting of 15 basic and
13 disperse dye standards. LC-MS was also applied to the detection and analysis of dyes
extracted from individual 0.5 cm acrylic and polyester fibres. Interestingly, the advantage
of LC-MS compared with other traditional methods was demonstrated by the differen-
tiation of three black acrylic fibres that could not be fully differentiated by their UV-Vis
spectra and/or chromatograms.
9.6 Capillary Electrophoresis
Capillary electrophoresis and its related techniques offer greater separation efficiency
than traditional chromatographic techniques and hold some promise for analytes ranging
from small ionisable compounds to large nonpolar species. The technique has been devel-
oped since the 1980s. Jorgenson and Lukacs (1981) refined the previous slab method of
electrophoresis and were the first to demonstrate high separation efficiency with high field
strength in narrow capillaries. Since that time there has been a large amount of research
with developments in technique and instrumentation (Landers, 1996; Li, 1992).
As described in other chapters, current routine methods for the examination of fibre
colour and dyes either examine the colour resulting from the dye or combination of dyes
in-situ (microspectrophotometry) or require the extraction of dyes for subsequent ana-
lytical separation by thin-layer chromatography (TLC), for example. The tedious nature
of the extraction step and the minute quantity of dyes being extracted from a single fibre
remain a significant challenge. During the first decade of the years 2000, a significant
attempt to overcome part of this challenge was made through some significant research
into the use of Capillary Electrophoresis for the analysis of fibre dyes (Fakhari et al., 2006;
Stefan et al., 2009; Trojanowicz et al., 2003; Xu et al., 2001). Due to the low abundance and
complexity of this kind of analyte, a highly efficient technique such as CE presents some
potential attractiveness Tagliaro et al. (2010).
9.6.1 Capillary Electrophoresis Principles

Capillary Electrophoresis has an exceptional separation capacity with simple instru-
mentation consisting of a high-voltage power supply, two buffer reservoirs, a capil-
lary, a detector of some sort and data collection (Figure 9.10). Separations in CE are
based upon the differential migration of charged species through a fused silica capil-
lary filled with the background electrolyte. These capillary columns are made of fused
silica which is externally coated with a polymer (polyamide) to improve its mechanical
strength. The internal diameter of these capillaries ranges from 25 to 200 μm, with
50–75 μm being common while lengths vary from a few centimetres to one metre. These
narrow-bored capillaries allow higher voltages to be applied than in conventional gel
electrophoresis. High voltage in gel electrophoresis results in convection which causes
the gel to be heated and therefore limits the separation efficiency. Due to the small size
of CE capillaries, convection is hindered in CE. Thus field strengths above 100 kV/m
can be used for separation purposes Wätzig and Günter (2003). Detection is typically by
ultraviolet-visible absorbance spectrophotometry, resulting in the graph, which looks
similar to a chromatographic separation and is often referred to as an electropherogram
Lewis (2009). Commercial Capillary Electrophoresis systems allow for temperature
control, programmable power supply and data analysis which runs through a computer
interface. There are variations in injection devices and detectors. Method development
is rapid because of the ability to perform a number of different types of separation
quickly on the same instrument, due to the ease and speed with which buffers, and
other parameters, of the basic system, can be varied.
A variety of different capillary electroseparation methods exist, including:
• Capillary zone electrophoresis (CZE) (Baker, 1995)
• Capillary gel electrophoresis (CGE) (Dolník, 1994)
High voltage
power supply
Detector:
UV, MS, etc.
Sample
+ –
Figure 9.10 Schematic diagram of capillary electrophoresis instrumentation.

• Micellar electrokinetic capillary chromatography (MEKC or MECC) (Terabe

et al., 1984)
• Capillary electrochromatography (CEC) (Robson et al., 1997)
• Capillary isoelectric focusing (CIEF) (Rodriguez-Diaz et al., 1997)
• Capillary isotachophoresis (CITP) (Gebauer and Boček, 2002)
• Microchip electrophoresis (ME) (Manz et al., 1990)
The two techniques that will be examined further are the commonly used capillary
zone electrophoresis (CZE) and micellar electrokinetic capillary chromatography (MEKC
or MECC) which has the most forensic significance due to its ability to separate neutral
species.
9.6.2 Capillary Zone Electrophoresis (CZE)

The basic mechanism of separation occurs along the length of a fused silica capillary,
resembling typical chromatography, but uses an applied voltage of 5–30 kV along the
capillary to separate the analytes based on their size-to-charge ratio. The contribution of
electroosmosis to separations can be manipulated and influenced through variances in
Blumelhuber (1997):
• pH: Affects the charge of the analytes and electro-osmotic flow (EOF). The higher
the pH, the faster is the EOF. pH is the single most important factor influencing
the quality of the separation.
• Ionic strength: Affects peak shape and resolution, contributes to Joule heat (the
higher the ionic strength, the more current flows through the capillary, and the
higher the amount of Joule heat generated in the capillary).
• Voltage: If the applied voltage is increased, the speed of migration of the EOF is
increased, thus giving shorter analysis time. However, this can also lead to broader
peaks, sample decomposition and buffer boiling due to heat production from high
currents.
• Temperature: Affects buffer viscosity, shifts equilibria, pH and the critical micellar
concentration during micellar electrokinetic capillary chromatography (MEKC or
MECC). The efficient control of temperature is essential for analytical performance.
• Organic modifiers: Alter viscosity, analyte solubility, hydration radius and
separation selectivity.
• Internal diameter (ID) of capillary: Affects sensitivity - the larger the ID, the larger
the optical path length of UV detection.
The capillary tube is made of fused silica containing internal surface silanol groups
which may become ionized in the presence of the buffer. The interface of the silica tube
wall and the buffer consists of three layers:
• Negatively charged silica surfaces at pH>2

• An immobile layer of cations which are tightly held by the negatively charged silica
surface and
• A mobile layer of cations which are further away from the silica surfaces and there-
fore not held as tightly and tend to migrate toward the cathode under an electric field
When a potential is applied across the capillary under normal conditions, the
cations in the mobile layer are dragged towards the cathode (which has a negative poten-
tial), generating a bulk flow of solution towards the cathode as more cations are attracted
to the capillary wall, forming something called the electroosmotic flow (EOF). Smaller,
positively charged ions (cations) are the most mobile towards the cathode and travel ahead
of the EOF, followed by larger cations, as their mobility is greater than that of the buffer.
Large negatively charged ions (anions) are attracted towards the anode, but they are still
dragged along with the bulk EOF, which has a much higher flow velocity than the anions
mobility towards the anode. These are followed last by small anions. Neutral molecules
move at the same velocity as the EOF and do not separate, and while they will all comi-
grate, it still allows all components, cations, anions and neutrals to be analysed in a single
CE run (Figure 9.11).
The EOF can also be reversed so anionic species are eluted first. This is achieved by
adding a quaternary amine the electrolyte as a flow modifier. The positively charged qua-
ternary amine is attracted to the negatively charged Si-O- groups on the capillary wall.
The analytes migrate through the capillary at different speeds under the applied elec-
tric field, and this results in the separation of the analytes. The velocity at which each
analyte travels through the capillary is directly proportional to the electrophoretic mobil-
ity of that species which is in turn related to the size and charge of the particle. Mobility is
also affected by the solution in which the species exists, i.e. the buffer. The apparent mobility
of the analyte is calculated using the migration times of the analyte and electroosmotic
flow, the applied voltage and the length of the capillary:
IL
μa =
tV
where: μa = apparent mobility of the analyte

l = effective capillary length to the detector
L = total capillary length
t = migration time
The apparent mobility of the analyte can then be calculated by calculating the mobility
of the EOF (μEOF) using the migration time of the electroosmotic flow and subtracting it
from the apparent mobility of the analyte (μa):
μa = μ e + μ EOF
9.6.3 Micellar Electrokinetic Capillary Chromatography (MEKC)

Neutral molecules cannot be separated using CZE, and so this separation technique was
modified in the 1980s to form a new separation technique, micellar electrokinetic capil-
lary chromatography (MEKC) (Terabe et al. 1984). The instrument and detector are the
same as CZE. However a surfactant is added to the buffer. The surfactant forms spheri-
cal micelles, with a charged, hydrophilic exterior and a neutral, hydrophobic interior.
This unique shape allows the neutral components to partition in and out of the micelle
at a rate determined by other factors such as its size and hydrophobicity. The more
hydrophobic solutes interact with micellar phase and thus migrate more slowly than
Sample region
– +
– +
Figure 9.11 Separation of analytes in a capillary electrophoresis.
hydrophilic compounds. These charged micelles will migrate faster or slower than the
EOF depending on their charge, and the neutral components ideally will separate and
migrate between the EOF and these micelles. The properties of micelles can be altered
and hence, selectivity in MEKC separations can be attained by varying the surfactant,
by using mixtures of surfactants and, increasingly, by adding a variety of modifiers.
The most common surfactant used in an anionic surfactant is sodium dodecyl sul-
phate (SDS) [CH3(CH2)11OSO3–Na+]. This is of particular interest to forensic casework,
where complex samples consisting of large, often neutral, molecules such as dyes, inks,
drugs or explosives can now be separated without further sample preparation such as
derivatisation.
9.6.4 Mass Spectrometry

Traditional UV detection has low sensitivity and is of limited use for fibre dye analysis.
Research has taken place in recent years into coupling mass spectrometry with capillary
electrophoresis to overcome this issue. This led to interesting results. However, interfacing
these two instruments is not without challenges.
Electrospray ionisation (ESI) is the most common interface used. In this case, a poten-
tial is required between the sprayer tip and the inlet of the mass spectrometer. Typical
sprayer voltages do not exceed 6000V, so this will not severely affect or reverse the electroos-
motic flow in the capillary Schmitt-Kopplin and Frommberger (2003). Mass spectrometry
also requires that volatile solvents are used (which increases sensitivity and selectivity),
and so volatile buffers need to be used in the capillary electrophoresis. As the buffer solu-
tion, concentration and pH are important factors in determining separation efficiency, and
since the buffers used in CE are ionic solutions, they generally lack the necessary volatility.
The buffer selection process, therefore, requires careful consideration of not only the buf-
fer itself but also additives in the buffer as surfactants and other additives such as chiral
selectors have been shown to greatly increase the noise, can stain the MS inlet and do
not evaporate easily (Schmitt-Kopplin and Frommberger, 2003; Scriba, 2008; Smith et al.,
1993). Along with this, any analytes trapped in non volatile components will not enter the
MS, lowering the sensitivity of the technique. This means that MEKC cannot be used in
conjunction with MS.
The problem of interfacing the high voltage capillary with a spray ionisation system
has been fixed by manufacturers. For example, Agilent™ have developed an ESI interface
using a special nebuliser needle that runs sheath liquid and drying gas through coaxial
stainless steel capillaries outside of the CE capillary. This sheath liquid is made up of the
buffer solution and serves to make an electrical connection between the CE anode and
the electrospray cathode (allowing the sample to be drawn out of the CE capillary) and
to provide a high flow rate in addition to the EOF to facilitate the electrospray. The buf-
fer is attracted to the electrospray chamber, and this attraction in combination with the
sheath flow draws the liquid out of the capillary which is turned into a fine spray using
the nebulizer gas. These fine droplets are able to easily evaporate into a gaseous form with
the heated drying gas entering the chamber, leaving charged analyte droplets. The mass
spectrometer then pulls in these analyte droplets by a charge applied to the inlet. As this is
a soft ionization technique, there is very little fragmentation.
Mass spectrometers accurately measure ions using their mass-to-charge ratio (or m/z).
Simple mass spectrometers achieve this using a single quadrupole by filtering out ions that
do not resonate at the same amplitude as the ion that is being scanned for. However, there
are other forms of mass spectrometer that can be used with capillary electrophoresis such
as the triple quadrupole mass spectrometer and the quadrupole time-of-flight (QTOF)
mass spectrometer.
9.6.4.1 Triple Quadrupole Mass Spectrometer

The triple quadrupole mass spectrometer utilises three quadrupole units in sequence to
focus on specific ions and fragments. The advantage of this is that there is less interference
so that analytes can be detected with a high signal-to-noise ratio. The first quadrupole acts
filters the analyte by mass and then sends the selected masses to the second quadropole
which contains a hexapole collision cell which fragments the selected ion (by accelerating
it through a collision gas (at a specified energy). The fragments are then passed through the
third quadrupole that is set to acquire specific ions to determine all of the fragment ions.
This process is known as tandem MS (MS-MS) and is very useful for structural elucida-
tion. This is more useful than the single quadropole MS which regular quadrupole mass
spectrometer which can either only acquire one specified ion at a time (SIM mode) or a
range of ions (scan mode).
9.6.4.2 Quadrupole Time-of-Flight (QTOF) Mass Spectrometer

The quadrupole time-of-flight mass spectrometer incorporates a scanning quadrupole to
act as a mass filter in the first stage, selecting ions based on their mass/charge to pass on to
the flight tube for time-of-flight analysis. The ToF flight tube measures the mass-to-charge
ratio of an ion by how long it takes to travel a path in an electric field. The ion is acceler-
ated upwards along a cylindrical tube towards a reflectron (a charged repellor plate that
‘reflects’ the ions in a parabolic path towards the detector). The degree to which the ion is
reflected, and hence the total time taken in the flight tube, is indicative of the ions precise
mass measured to four decimal places.
9.6.5 Application of CE for Dye Analysis

Capillary electrophoresis (CE) is an emerging technique having advantages over TLC and
HPLC particularly regarding separation efficiency and the separation range of dyes with a
single condition. Initially, early studies on the use of CE for fibre dye analysis focused on
water-soluble dyes and conventional buffer systems (Resua et al., 1981). Early applications
of this technique suffered from limited reproducibility of migration times, poor sensitivity,
and the inability to separate non-ionizable dyes and so it was not commonly used. However, it
was discovered that the use of micelles in the electrophoretic buffer (micellar electrokinetic
chromatography or MEKC) allowed for the separation of water-insoluble dyes (such as
natural dyes like flavonoids and anthraquinones) (Beattie et al., 1979).
The main problem faced with the forensic analysis of fibre dyes with capillary elec-
trophoresis is that the analysis is of very small amounts of dye extracted from single short
pieces of fibre. The amount of dyes extracted from such a small single fibre typically having
a diameter of 15–30 μm and a length of 2–20 mm is normally far from sufficient for a CE
analysis. As a result research into this technique has focused on optimising a method to
analyse only small quantities of various dyes and textiles.
Initial research on the analysis of acid dyes extracted from wool fabrics (Robertson
et al., 1993; Sirén and Sulkava, 1995) identified a number of problems, including the fact
that dyes appeared to stick to the capillary walls. To overcome this problem, it was neces-
sary to use stringent purges to clear the capillary between runs, which greatly increased
analysis time. Another weakness of capillary electrophoresis is the small injection volume
required, and so high analytical concentrations are necessary. When materials are coloured
with dyes of low concentration, they are extremely difficult to extract and screen with CE
due to their high solubilities in water. This in turn has caused problems with repeatability
in past research (Sirén and Sulkava, 1995) whereby, in an attempt to help extract dyes with
a low concentration and high solubility in water, high ionic strengths were used. However,
this had a negative effect by changing migration times from run to run. Marker techniques
were introduced to improve screening of the electropherograms and give better repeat-
ability; however, the migrations of the dye components were only repeatable within a day.
Fresh samples had to be made fresh within one day to obtain an accurate result which is
not suitable for casework. The major problem also faced in this research is that the sample
size has been limited to squares of textiles, rather than a single fibre as would be expected
in case work.
Further research has involved both basic and acidic dyes being micro-extracted and
analyzed by using both CE-UV and CE-MS (Stefan et al. 2009). Basic dyes were extracted
from acrylic fibres with 88% formic acid/12% water and then electrophoretically separated
in a 45 mM ammonium acetate in ACN/water solution at pH 4.7; MS was used as the detec-
tion system. Acidic dyes, namely anthraquinone, azo and a metal complex, were instead
extracted from nylon fibres by using pyridine/ammonia/ water solution and electrophoret-
ically separated in a 15 mM ammonium acetate in ACN/water (40:60, v/v) buffer at pH 9.3.
Diode array spectra were obtained and used to discriminate even dyes of similar colour.
When capillary electrophoresis and MEKC were combined with a related technique,
called sample-induced isotachophoresis, the sensitivity was greatly increased, the dye
content of both synthetic and natural fibres could be analysed, and the sample size could
be reduced to a single fibre. This greatly increased the usefulness of CE in forensic case-
work Xu et al. (2001). Lastly, although the technique has not yet been applied to fibres,
nonaqueous capillary electrophoresis has shown promise for hydrophobic basic dyes
Momenbeik et al. (2006).
9.6.6 CE-ESI-MS
One of the main limitations faced by the use of CE-ESI-MS analysis for fibre dyes is the
small number of buffers able to be used due to the interface of the CE with the MS. These
limitations have meant that MEKC and aqueous buffers cannot be used, leaving only non-
aqueous CE (NACE) as the only option Given that not all dyes are water soluble, a mixed
buffer system comprising of organic and aqueous components could also improve the
compatibility with several classes of dyes.
CE-ESI-MS alone is unable to facilitate efficient fibre dye analysis as the co-elution of
neutral dyes due to limited buffer suitability will remain a problem for regular mass spec-
trometers, which are unable to resolve molecular ions and their corresponding fragment
ions Schmitt-Kopplin and Frommberger (2003). These instruments rely on buffer systems
tailored to each class of dye to enable full resolution of compounds. For this reason, mass
spectrometers capable of MS/MS analysis are required to resolve these dyes.
Research carried out over the years has focused several fibre types and dye types
using single quadrupole mass spectrometers and Q-ToF mass spectrometers. The methods
employed in these studies used a different method for each of the types of dyes (based on
whether the dye was positively or negatively charged, its polarity and solubility in water).
The extraction methods for single fibres in these studies were tailored for each specific
fibre/dye combination, with solvents and temperatures chosen for the easiest compatibility
with the buffers or with automated sample extraction instruments (Herrero et al., 2005;
Morgan et al., 2009; Poiger et al., 2000; Rittgen et al., 2012; Stefan et al., 2009; Varesio and
Serwe, 2009). The main results of these studies were that non-aqueous buffers were the
most effective method for CE-ESI-MS analysis. The importance of the choice of organic
additive was also found to be important as an increase in the boiling point of the com-
pound lowered the signal-to-noise significantly. Optimisation of the spray chamber and
mass spectrometer acquisition parameters has also been shown to greatly effect sensitivity
Herrero et al. (2005). Tandem mass spectrometry has also been found to be useful to sepa-
rate co-migrating and overlapping compounds (Scriba, 2007).
9.6.7 Advantages and Disadvantages of Capillary Electrophoresis

The main disadvantages of capillary electrophoresis is the lack of sensitivity and selectiv-
ity as UV-visible detection is not able to uniquely identify analytes, nor is it able to detect
all analytes in every sample (though this can be improved on by using CE-MS), lack of
knowledge or development which has limited CE to research purposes only and not indus-
try acceptance and the issue of reproducibility which has been a problem with UV-visible
detection.
However, there are far more advantages associated with capillary electrophoresis. The
first, which is very important for forensic casework, is that it requires very little material
(only 1–50 nL of sample is injected into the capillary), the technique is also highly efficient
(with a high number of theoretical plates for a small column) and is fast with most com-
pounds being able to be resolved in less than 20 minutes. The CE method is also highly
adaptable with variables such as the buffer, solution pH and modifiers affecting how the
separation occurs, and what compounds can be analysed. This process can also be auto-
mated. The instrumentation is also fairly simple with detection able to be performed on the
capillary itself (with UV-Vis detection) without requiring external apparatus. The cost of
using the method is also low with only small amounts of buffer (1–2 mL) needed.
Therefore while the limitations of CE and CE-MS limit its use in casework at the
present time, the advantages of the method show that with further research, it could be
a very useful addition to the forensic analysis of fibres. Of particular interest is the recent
development of microchip electrophoresis (ME) that can be seen as a miniaturized version
of CE where the capillary is replaced by a chip capillary device. Advantages of ME over
traditional CE include shorter analysis times and lower sample volumes, as well as the
potential for the creation of disposable devices and automation. While there has been little
research investigating applications of ME (and more generally microfluidics) to fibre anal-
ysis, ME has reached maturity in the past decade, and the instrumentation is constantly
being developed and improved. It is, therefore, foreseeable that more research activity and
possibly casework applications will emerge in the next few years.
9.7 Conclusions
Almost all fibres are complex (natural or man-made) polymeric material. This means that
an optimized set of data should include morphological, physical and chemical features.
This is often achieved by traditional techniques such as those described in the remainder
of the book. As a result, in most routine cases, these will suffice. However, some problems
remain challenging, for example, the discrimination of colourless or denim cotton fibres,
of some colourless synthetic fibres (e.g. polyester); elucidation of the specimen history and/
or its environment; textile damage; just to name a few. Attempting to address these chal-
lenges often justifies the application of other instrumental techniques that are not part
of the standard forensic science laboratory procedure. The field of analytical sciences is
very dynamic and regularly presents new instrumental techniques that may be applied
to the forensic examination of fibres. It is, therefore, important for the fibre examiner to
stay abreast of these developments and recognize the potential when a new specialized
technique is being proposed. More often than not, it will require further research within
a forensic science context and proper validation before that specialized technique can be
used in casework. Further, given the current economic climate experienced by most coun-
tries, such research will have to include a robust cost-benefit analysis.
From all of the above, it appears that the following areas can supplement the pool of
existing methods and warrant our continued attention:
• Pyrolysis–GC techniques, especially when automation and advanced MS detec-

tion are considered
• Isotope ratio mass spectrometry
• Scanning electron microscopy, in particular, more recent environmental system
• Elemental analysis in general and micro-XRF or LA-ICP-MS in particular
• Dye analysis using CE, especially when coupled with advanced MS detection, for
example, Triple Quadrupole or QTOF MS
• Microfuidics in general offering growing opportunities for automation and field
deployment
It is anticipated that, in the future, with the advances in research in these fields, these
techniques, along with hyperspectral imaging techniques (see Chapter 7) will be more
widely used and accepted in the forensic examination of fibres.
References
Andrasko, J., Haeger, L., Maehly, A. C., and Svensson, L., 1984. Comparative analysis of synthetic
polymers using a combination of three analytical methods, Forens. Sci. Int., 25, 57-70.
Armitage, S., Saywell, S., Roux, C., Lennard, C., and Greenwood, P., 2001. The analysis of forensic
samples using laser micro-pyrolysis gas chromatography mass spectrometry, J. Forensic Sci.,
46(5), 1043.
Badri, B., 2008. The influence of reactive dyes on the pyrolysis of cotton, J. Anal. Appl. Pyrolysis,
81(2), 162–166.
Baker, D. R., 1995. Capillary Electrophoresis. New York, NY: Wiley.
Beattie, I., Dudley, R., and Smalldon, K., 1979. The extraction and classification of dyes on single
nylon, polyacrylonitrile and polyester fibres, J. Soc. Dyers Colour., 95(8), 295–302.
Benson, S., Lennard, C., Maynard, P., and Roux, C., 2006. Forensic applications of isotope ratio mass
spectrometry—A review, Forensic Sci. Int., 157(1), 1–22.
Blackledge, R. D., 1992. Application of pyrolysis gas chromatography in forensic science, Forensic
Sci. Rev., 4(1), 1–16.
Blumelhuber, F., 1997. Capillary electrophoresis–A new tool for dye identification. In Proceedings of
the 5th Meeting of the European Fibre Group, Berlin.
Boutton, T., 1996. Stable carbon isotope ratios of soil organic matter and their use as indicators of
vegetation and climate change. In T. Boutton and S. I. Yamasaki SI (Eds.), Mass Spectrometry of
Soils. New York, NY: Taylor & Francis.
Buscaglia, J. A. and Koons, R. D., 1998a. The application of total reflection X-ray fluorescence spec-
trometry to the elemental analysis of trace evidentiary materials, Proc. 50th Meeting of the
American Academy of Forensic Sciences, San Francisco, CA.
Buscaglia, J. A. and Koons, R. D., 1998b. The characterization of automotive carpet fibers by total
reflection X-ray fluorescence spectrometry, Proc. 50th Meeting of the American Academy of
Forensic Sciences, San Francisco, CA.
Bush, H. D., Butterworth, A., Pearson, E. F., and Pounds, C. A., 1973. An evaluation of trace element
analysis in the forensic investigation of fibre samples, J. Radioanal. Chem., 15, 245-264.
Carter, J. F. and Barwick, V. J., 2011. Good Practice Guide for Isotope Ratio Mass Spectrometry. FIRMS.
Self-published.
Cartier, J., Roux, C., and Grieve, M.C., 1997. A study to investigate the feasibility of using X- ray
fluorescence microanalysis to improve discrimination between colourless synthetic fibers, J.
Forens. Sci., 42, 1019-1026.
Chaikin, M., 1986. Unpublished statement, Royal Commission of Inquiry into Chamberlain
Convictions, Northern Territory Government, Darwin, Australia.
Challinor, J. M., 1999. Instrumental methods used in fibre examination: Fibre identification by pyrol-
ysis techniques. In J. Robertson (Ed.), Forensic Examination of Fibres (2nd ed.). Boca Raton, FL:
CRC Press.
Chen, M., Moir, D., Benoit, F.M., and Kubwabo, C., 1998. Purification and identification of several
sulphonated azo dyes using reversed-phase preparative high-performance liquid chromatogra-
phy, J. Chromatogr. A, 825(1), 37–44.
Choudhry, M., 1987. The use of scanning electron microscopy for identification of cuts and tears in
fabrics: Observations based upon criminal cases, Scan. Microsc., 1, 119-125.
Daéid, N. N., Meier-Augenstein, W., and Kemp, H. F., 2011. Investigating the provenance of un-dyed
spun cotton fibre using multi-isotope profiles and chemometric analysis, Rapid Commun. Mass
Spectrom., 25(13), 1812–1816.
Dolník, V., 1994. Capillary gel electrophoresis, J. Microcolumn Sep., 6(4), 315–330.
Edenbaum, J., 1992. Plastics Additives and Modifiers Handbook. New York, NY: Van Nostrand
Reinhold.
Fakhari, A. R., Breadmore, M. C., Macka, M., and Haddad, P. R., 2006. Non-aqueous capillary elec-
trophoresis with red light emitting diode absorbance detection for the analysis of basic dyes,
Anal. Chim. Acta, 580(2), 188–193.
Fifield, F. W. and Kealey, D., 2000. Principles and Practice of Analytical Chemistry. Oxford: Blackwell
Science.
Gallo, J. M. and Almirall, J. R., 2009. Elemental analysis of white cotton fiber evidence using solution
ICP-MS and laser ablation ICP-MS (LA-ICP-MS), Forensic Sci. Int., 190(1), 52–57.
Gebauer, P. and Boček, P., 2002. Recent progress in capillary isotachophoresis, Electrophoresis,
23(22–23), 3858–3864.
Gentile, N., Siegwolf, R. T., Esseiva, P., Doyle, S., Zollinger, K., and Delémont, O., 2015. Isotope ratio
mass spectrometry as a tool for source inference in forensic science: A critical review, Forensic
Sci. Int., 251, 139–158.
Goldstein, J. I., Newbury, D. A., Echlin, P., Joy, D. C., Romig, A. D., Lyman, C. E., Fiori, C., and
Lifshin, E., 1992. Scanning Electron Microscopy and X-Ray Microanalysis. New York and
Górecki, T. and Poerschmann, J., 2001. In-column pyrolysis: A new approach to an old problem,
Anal. Chem., 73(9), 2012–2017.
Goynes, W. R., 1971. A scanning electron microscope study of washer-dryer abrasion in textile fibres,
Proc. 28th Annual EMSA Meeting, 346-347.
Greaves, P. H. and Saville, B. P., 1995. Scanning electron microscopy. In Microscopy of Textile Fibres,
pp. 51-67, Oxford, UK: Bios Scientific Publishers.
Grieve, M. C. and Cabiness, L. R., 1985. The recognition and identification of modified acrylic fibres,
Forens. Sci. Int., 29, 129-146.
Grieve, M. C., Dunlop, J., and Kotowski, T. M., 1988. Bicomponent acrylic fibres – Their character-
ization in the forensic science laboratory, J. Forens. Sci., 28, 25-34.
Griffin, R., Kee, T., and Adams, R., 1988. High-performance liquid chromatographic system for the
separation of basic dyes, J. Chromatogr. A, 445, 441–448.
Griffin, R. and Speers, J., 1999. Other methods of colour analysis, In J. Robertson (Ed.), Forensic
Examination of Fibres(2nd ed.) (p. 311).Boca Raton, FL: CRC Press.
Griffin, R., Speers, S., Elliott, L., Todd, N., Sogomo, W., and Kee, T., 1994. An improved
high-performance liquid chromatography system for the analysis of basic dyes in forensic case-
work, J. Chromatogr. A, 674(1), 271–280.
Hacaloglu, J., 2012. Direct insertion probe mass spectrometry of polymers. In M. Hakkarainen (Ed.),
Mass Spectrometry of Polymers—New Techniques (Vol. 248, pp. 69–103). Berlin Heidelberg:
Springer.
Hearle, J., Lomas, B., Cooke, W., and Duerdon I., 1989. Fibre Fracture and Wear of Materials: An Atlas
of Fracture, Fatigue and Durability. Manchester, UK: The Textile Institute.
Hecking, L. T. and Erikson, C., 1993. Characterisation of acid-washed denim garments and acces-
sories using the stereomicroscope and the scanning electron microscope, Cust. Lab. Bull., 5, 3.
Herrero, M., Arráez-Román, D., Segura, A., Kenndler, E., Gius, B., Raggi, M.A., Ibáñez, E., and
Cifuentes, A., 2005. Pressurized liquid extraction–capillary electrophoresis–mass spectrometry
for the analysis of polar antioxidants in rosemary extracts, J. Chromatogr. A, 1084(1), 54–62.
Hiroma, Y., Nakai, I., and Hokura, A., 2010. Forensic identification of Trunk Mat by trace element
analysis of single fiber with laser ablation ICP-MS, Bunseki Kagaku, 59(9), 759–769.
Holčapek, M., Jandera, P., and Přikryl, J., 1999. Analysis of sulphonated dyes and intermediates by
electrospray mass spectrometry, Dyes Pigm., 43(2), 127–137.
Huang, M., Russo, R., Fookes, B.G., and Sigman, M.E., 2005. Analysis of fiber dyes by liquid chroma-
tography mass spectrometry (LC-MS) with electrospray ionization: Discriminating between
dyes with indistinguishable UV-visible absorption spectra, J. Forensic Sci., 50(3), 526e34.
Huang, M., Yinon, J., and Sigman, M. E., 2004. Forensic identification of dyes extracted from textile
fibers by liquid chromatography mass spectrometry (LC-MS), J. Forensic Sci., 49(2), 238–249.
Jallard, R., 1995. Application de la microfluorescence X lors de la discrimination de fibres de blue
jeans, unpublished seminar thesis, University of Lausanne, Lausanne, Switzerland.
Jasper, J., Edwards, J., Ford, L., and Corry, R., 2002. Putting the arsonist at the scene: ‘DNA’ for the
fire investigator, Fire Arson Invest., 51(2), 30–34.
Jorgenson, J. W. and Lukacs, K. D., 1981. High-resolution separations based on electrophoresis and
electroosmosis, J. Chromatogr. A, 218, 209–216.
Katz, H. S., 1990. Additives. In Rubin, I. I. (ed.), Handbook of Plastic Materials and Technology, pp.
675-681, New York, NY: John Wiley & Sons.
Koons, R. D., 1996. Comparison of individual carpet fibers using energy X-ray fluorescence, J.
Forens. Sci., 41, 199-205.
Kroschwitz, J. I., 1990. Polymers – Fibers and Textiles, a Compendium. Encyclopaedia Reprint Series,
New York: John Wiley & Sons.
Kuckuk, R., Hill, W., Burba, P., and Davies, A. N., 1994. Pyrolysis–GC-FTIR for structural elucidation
of aquatic humic substances, Fresen. J. Anal. Chem., 350(7–9), 528–532.
Laing, D., Gill, R., Blacklaws, C., and M Bickley, H., 1988. Characterisation of acid dyes in forensic
fibre analysis by high-performance liquid chromatography using narrow-bore columns and
diode array detection, J. Chromatogr. A, 442, 187–208.
Landers, J. P., 1996. Handbook of Capillary Electrophoresis (2nd ed.). Boca Raton, FL: CRC Press.
Lattimer, R. P., 1997. Pyrolysis field ionization mass spectrometry of hydrocarbon polymers, J. Anal.
Appl. Pyrolysis, 39(2), 115–127.
Lewis, S., 2009. Analysis of dyes using chromatography. In M. M. Houck (Ed.), Identification of
Textile Fibres (pp. 1–396). Cambridge, UK: Woodhead.
Li, S. F. Y., 1992. Capillary Electrophoresis: Principles, Practice and Applications. Amsterdam: Elsevier.
Lynch, B., 1981. Investigation of single fibres by X-ray diffraction in forensic analysis, X-Ray
Spectrosc., 10, 196-197.
Manz, A., Graber, N., and Widmer, H. M., 1990. Miniaturized total chemical analysis systems: A
novel concept for chemical sensing, Sens. Actuators B Chem., 1(1), 244–248.
Meier-Augenstein, W., Kemp, H. F., Schenk, E. R., and Almirall, J. R., 2014. Discrimination of unpro-
cessed cotton on the basis of geographic origin using multi-element stable isotope signatures,
Rapid Commun. Mass Spectrom., 28(5), 545–552.
Meier-Augenstein, W. and Liu, R., 2004. Forensic applications of isotope ratio mass spectrom-
etry. In J. Yinon (Ed.), Advances in Forensic Applications of Mass Spectrometry (pp. 149–80).
Boca Raton, FL: CRC Press.
Momenbeik, F., Johns, C., Breadmore, M. C., Hilder, E. F., Macka, M., and Haddad, P. R., 2006. Sensitive
determination of carbohydrates labelled with p-nitroaniline by capillary electrophoresis
with photometric detection using a 406 nm light-emitting diode, Electrophoresis, 27(20),
4039–4046.
Morgan, S. L., Vann, B. C., Baguley, B. M., and Stefan, A. R., 2009. Advances in Discrimination Of
Dyed Textile Fibers Using Capillary Electrophoresis/Mass Spectrometry, viewed 17 November
2013, http://projects.nfstc.org/trace/docs/final/morgan_dyed_textiles_revised.pdf.
Nic Daéid, N., Meier-Augustine, W., and Kemp, H., 2011. Investigating the provenance of un-dyed
spun cotton fibre using multi-isotope profiles and chemometric analysis, Rapid Comm. in
Mass. Spec., 25(13), 1812–1816.
Özgen, C., 2012. The Characterization of Some Methacrylate and Acrylate Homopolymers,
Copolymers and Fibers Via Direct Pyrolysis Mass Spectroscopy. Turkey: Middle East Technical
University.
Parsi, Z., Hartog, N., Gorecki, T., and Poerschmann, J., 2007. Analytical pyrolysis as a tool for the
characterization of natural organic matter—A comparison of different approaches, J. Anal.
Appl. Pyrolysis, 79(1), 9–15.
Pelton, W. R., 1995. Distinguishing the cause of textile fibre damage using the scanning electron
microscope (SEM), J. Forens. Sci., 40, 874-882.
Petrick, L., Wilson, T., and Fawcet, W., 2006. HPLC UV–Visible spectroscopy electrospray ionisation
mass spectrometry method for acrylic and polyester fibre dye analysis, J. Forensic Sci., 51(4),
771–779.
Poiger, T., Richardson, S. D., and Baughman, G. L., 2000. Analysis of anionic metallized azo and
formazan dyes by capillary electrophoresis–mass spectrometry, J. Chromatogr. A, 886(1),
259–270.
Prange, A., Reus, U., Böddeker, H., Fischer, R., and Adolf, F.-P., 1995. Microanalysis in forensic sci-
ence: characterization of single textile fibers by total reflection X-ray fluorescence, Anal. Sci.,
11, 483–487.
Resua, R., DeForest, P., and Harris, H., 1981. The evaluation and selection of uncorrelated paired sol-
vent systems for use in the comparison of textile dyes by thin-layer chromatography, J. Forensic
Sci., 26(3), 515–534.
Rittgen, J., Pütz, M., and Zimmermann, R., 2012. Identification of fentanyl derivatives at trace lev-
els with nonaqueous capillary electrophoresis–electrospray–tandem mass spectrometry (MSn,
n = 2, 3): Analytical method and forensic applications, Electrophoresis, 33(11), 1595–1605.
Robertson, J., Wells, R. J., Pailthorpe, M. T., David, S., Aumatell, A., Clark, R., 1993. An assessment
of the use of capillary electrophoresis for the analysis of acid dyes in wool fibres. In B. Jacob and
W. Bonte (Eds.), Advances in Forensic Sciences (Vol. 4, pp. 247–249). Forensic Criminalistics 2.
Berlin, Germany: Verlag.
Robinson, V., 1987. Unpublished statement, Royal Commission of Inquiry into Chamberlain
Convictions, Northern Territory Government, Darwin, Australia.
Robson, D., 1994. Fibre surface imaging, J. Forens. Sci., 34, 187-191.
Robson, M. M., Cikalo, M. G., Myers, P., Euerby, M. R., and Bartle, K. D., 1997. Capillary electro-
chromatography: A review, J. Microcolumn Sep., 9(5), 357–372.
Rodriguez-Diaz, R., Wehr, T., and Zhu, M., 1997. Capillary isoelectric focusing, Electrophoresis,
18(12–13), 2134–2144.
Roux, C., 1995. The analysis of inorganic additives in polyester fibres by inductively coupled plasma
atomic emission spectroscopy (ICP-AES), Adv. Forens. Seci., 4, 2 16-224.
Schmitt-Kopplin, P. and Frommberger, M., 2003. Capillary electrophoresis–mass spectrometry: 15
years of developments and applications, Electrophoresis, 24(22–23), 3837–3867.
Scriba, G. K., 2007. Nonaqueous capillary electrophoresis–mass spectrometry, J. Chromatogr. A,
1159(1), 28–41.
Sirén, H. and Sulkava, R., 1995. Determination of black dyes from cotton and wool fibres by capillary
zone electrophoresis with UV detection: Application of marker technique, J. Chromatogr. A,
717(1–2), 149–155.
Smalldon, K. W., 1973. The identification of acrylic fibres by polymer composition as determined by
infrared spectroscopy and physical characteristics, J. Forens. Sci., 18, 69-81.
Smith, R. D., Wahl, J. H., Goodlett, D. R., and Hofstadler, S. A., 1993. Capillary electrophoresis/mass
spectrometry, Anal. Chem., 65(13), 574A–84A.
Sobeih, K. L., Baron, M., and Gonzalez-Rodriguez, J., 2008. Recent trends and developments in
pyrolysis–Gas chromatography, J. Chromatogr. A, 1186(1), 51–66.
Speers, S., Little, B., and Roy, M., 1994. Separation of acid, basic and dispersed dyes by a single-gra-
dient elution reversed-phase high-performance liquid chromatography system, J. Chromatogr.
A., 674(1), 263–270.
Stauffer, E., 2003. Concept of pyrolysis for fire debris analysts, Sci. Justice, 43(1), 29–40.
Stefan, A., Dockery, C., Nieuwland, A., Roberson, S., Baguley, B., Hendrix, J., and Morgan, S.,
2009. Forensic analysis of anthraquinone, azo, and metal complex acid dyes from nylon
fibers by micro-extraction and capillary electrophoresis, Anal. Bioanal. Chem., 394(8),
2077–2085.
Stefan, A. R., Dockery, C. R., Baguley, B. M., Vann, B. C., Nieuwland, A. A., Hendrix, J. E., and
Morgan, S. L., 2009. Microextraction, capillary electrophoresis, and mass spectrometry for
forensic analysis of azo and methine basic dyes from acrylic fibers, Anal. Bioanal. Chem.,
394(8), 2087–2094.
Stowell, L. and Card, K., 1990. Use of scanning electron microscopy (SEM) to identify cuts and tears
in a nylon fabric, J. Forens. Sci., 35, 947–950.
Tagliaro, F., Pascali, J., Fanigliulo, A., and Bortolotti, F., 2010. Recent advances in the application of
CE to forensic sciences: A update over years 2007–2009, Electrophoresis, 31(1), 251–259.
Taupin, J. M., 1996. Damage analysis – A method for its analysis and application in case of violent
crime, Proc. 14th International Association of Forensic Sciences Meeting, Tokyo, Japan.
Tebbett, I., 1991. Chromatographic analysis of inks for forensic science applications, Forensic Sci.
Rev., 3, 71–82.
Terabe, S., Otsuka, K., Ichikawa, K., Tsuchiya, A., and Ando, T., 1984. Electrokinetic separations with
micellar solutions and open-tubular capillaries, Anal. Chem., 56(1), 111–113.
Trojanowicz, M., Wójcik, L., and Urbaniak-Walczak, K., 2003. Identification of natural dyes in his-
torical Coptic textiles by capillary electrophoresis with diode array detection, Chem. Analit.,
48(3), 607–620.
Varesio, E. and Serwe, M., 2009. Analysis of Amphetamines in Urine by CE-ESI-MS, viewed 17
November 2013, http://www.chem.agilent.com/en-US/Search/Library/_layouts/Agilent/
PublicationSummary.aspx?whid=58902&liid=2306.
Wampler, T. P., 2006. Applied Pyrolysis Handbook. Boca Raton, FL: CRC Press.
Wätzig, H. and Günter, S., 2003. Capillary electrophoresis-a high performance analytical separation
technique, Clin. Chem. Lab. Med., 41(6), 724–738.
West, J. C., 1981. Extraction and analysis of disperse dyes on polyester textiles, J. Chromatogr. A,
208(1), 47–54.
Wheals, B., White, P., and Paterson, M., 1985. High-performance liquid chromatographic method
utilising single or multi-wavelength detection for the comparison of disperse dyes extracted
from polyester fibres, J. Chromatogr. A, 350, 205–215.
White, P. and Catterick, T., 1987. Evaluation of absorbance ratioing for solute identification in
high-performance liquid chromatography using a diode array detector, J. Chromatogr. A, 402,
135–147.
Xu, X., Leijenhorst, H. A. L., Van den Hoven, P., De Koeijer, J. A., and Logtenberg, H., 2001. Analysis
of single textile fibres by sample-induced isotachophoresis–micellar electrokinetic capillary
chromatography, Sci. Justice, 41(2), 93–105.
Yinon, J. and Saar, J., 1991. Analysis of dyes extracted from textile fibers by thermospray
high-performance liquid chromatography-mass spectrometry, J. Chromatogr. A, 586(1), 73–84.
Zieba-Palus, J., Borusiewicz, R., and Kunicki, M., 2008. PRAXIS—combined μ-Raman and μ-XRF
spectrometers in the examination of forensic samples, Forensic Sci. Int., 175(1), 1–10.
Interpretation of Fibre Evidence
10
MICHAEL GRIEVE
CLAUDE ROUX
KENNETH G WIGGINS
CHRISTOPHE CHAMPOD
FRANCO TARONI
Contents
10.1 Factors Influencing Interpretation 346
10.1.2 The Influence of Case Circumstances 348
10.1.3 Fibre Frequency 351
10.1.4 Target Fibres and Case Strategy 353
10.1.5 The Influence of Discrimination within a Generic Type 354
10.1.5.1 The Extreme Variability of Synthetic Fibres and Target
Fibre Studies 354
10.1.5.2 Comparison of Fibre Characteristics 355
10.1.6 The Number of Matching Fibres Recovered 358
10.1.7 Quality Assurance 361
10.1.8 Report Writing 363
10.1.9 Examples of Casework Findings 366
10.1.9.1 Example 1 366
10.1.9.2 Example 2 366
10.1.9.3 Examples 3 and 4 (Recovery of a Single Fibre Only) 367
10.1.9.4 Example 5 368
10.1.9.5 Example 6 368
10.1.9.6 Example 7 370
References 371
10.2 Aids to Interpretation 375
10.2.2 Fibre Reference Collections 375
10.2.3 Data Collections on Fibre Frequency 377
10.2.4 Target Fibre Studies 379
10.2.5 Population Studies 381
10.2.6 Industrial Enquiries 384
10.2.6.1 Value of Industrial Enquiries 385
10.2.6.2 Methods of Tracing Manufacturers 386
10.2.6.3 Type of Information Required 386
10.2.6.4 Examples of Industrial Enquiries 387
345

Acknowledgement 392
References 392
10.3 A Probabilistic Approach to the Evaluation of Fibre Evidence 395
10.3.2 Interpretation Process 396
10.3.3 Likelihood Ratios for the Evaluation of Trace Evidence 398
10.3.4 Likelihood Ratios in Various Scenarios 403
10.3.4.1 Evidence Left by the Offender(s) 403
10.3.4.2 Evidence Possibly Left by the Offender(s) 406
10.3.5 Bayesian networks 417
10.3.6 The Need for Data 421
10.3.7 Conclusion 422
References 423
10.1 Factors Influencing Interpretation
MICHAEL GRIEVE
CLAUDE ROUX
10.1.1 Introduction
Interpreting the value and significance of the analytical findings in a fibres case and
expressing them concisely in a written report so that it can be understood, without the risk
of ambiguity, by scientists, lawyers and lay persons alike is one of the most difficult tasks
facing the forensic scientist. In recent years, this important step of the forensic science
process has become increasingly important, to the point that it is argued that a large part
of the future of trace evidence, including fibres, depends on how this challenge can be
resourced and addressed. As explained by Roux et al. (2015), failing to do so means that
‘Overall, more often than not, the value of trace evidence will remain unclear to the non-
specialist and of relatively poor cost–benefit to the manager’.
This chapter deals with the complex and interacting factors influencing the
interpretation of fibre transfer evidence. Reference is made to research results which can
be used to help forensic scientists to support their evaluations and conclusions. Some case-
work examples are discussed.
General articles covering problems of fibre interpretation have been written by
Deadman (1984), Brüschweiler (1993), Grieve (1994), Brüschweiler and Rey (1997) and
Roux and Robertson (2013a). More information can be found in the works of Gaudette
(1988) and Grieve (1990). References relating to the use of a probabilistic framework based
on likelihood ratios (often simplistically called a Bayesian approach) can be found in
Section 10.3 of this chapter.
There are two schools of thought about assessing the value of fibre evidence. The clas-
sical approach favours the use of statistical probability based on fibre frequencies (i.e. how
common the fibres are) and expressing the result in terms of statistical odds (i.e. what is
the chance to find these fibres). The alternative, Bayesian, approach considers a ratio of the
likelihood of the observed results being caused by two competing probabilities – those
of the evidence being present if the suspect did, or did not, commit the crime. Prior
Interpretation of Fibre Evidence 347
odds (background information about the case) can be modified in the light of additional
information (the evidence) to give posterior odds for particular circumstances. Frequencies
are only one aspect considered using a Bayesian approach. The latter also embeds other
important factors such as transfer and persistence phenomena. For these reasons, as
explained in Section 10.3, probabilistic theory has been offered as a coherent model for
interpreting and evaluating forensic evidence (Aitken and Taroni, 2004), notwithstand-
ing the fact that attempts to use statistics without a proper foundation can lead to highly
speculative and dangerously misleading conclusions (Gianelli and Imwinkelried, 1986).
Sceptics dislike the Bayesian approach on the grounds that it uses too many assump-
tions and estimations (Grieve, 2000). The authors believe that it combines some subjectiv-
ity with the increasing amount of objective data compiled through research undertaken
over the last twenty years or so, and that consideration of these data relating to fibre popu-
lations forces forensic scientists to give much more careful thought to different hypotheses
and previously unconsidered aspects within casework. At this point, it is worth highlight-
ing that the Bayesian approach is at the core of the framework for formulating evaluative
reports that has been recently recommended by the European Network of Forensic Science
Institutes (ENFSI, 2015).
Gaudette (1988) dealt with the problem of error probabilities with respect to fibres
evidence, and concluded that the main problem was to avoid type 2 errors (incorrect asso-
ciations). The value of fibre evidence is inversely proportional to type II errors. Subsequent
research using target fibre studies and population studies (see Section 10.2), and the large
amount of information on the discriminating power of the methods used for fibre identifi-
cation and comparison (Adolf, 1993; Grieve et al., 1988, 1990; Hartshorne and Laing, 1991;
Wiggins, 1994; Cassista and Peters, 1997; Grieve et al., 2001; Grieve and Biermann, 2003;
Grieve et al., 2003; Grieve et al., 2005; Grieve et al., 2006; Coyle et al., 2007; Massonnet and
Buzzini, 2008; Biermann, 2007; Palmer et al., 2009), have shown that the probability of
associative fibre evidence being presented as the result of coincidence is very small indeed.
Much of the original work has been carried out by Stoney on the evaluation of transfer
evidence. Stoney (1991) described the basic concepts in transfer evidence interpreta-
tion, defined relevant statistical terms, emphasized the need for both bulk textile source
frequency data (Biermann and Grieve, 1996, 1998) and environment-specific population
data (Grieve and Biermann, 1997a; Roux and Margot, 1997; Cantrell et al., 2001; Dignan
and Murphy, 2002; Cresswell et al., 2003; Palmer and Oliver, 2004; Was-Gubala, 2004a;
Watt et al., 2005; Marname et al., 2006; Lazic et al., 2012; Almazrooei et al., 2012), and
discussed the use of Bayes’ theorem.
Samples from known sources are required to support an inference of contact, to verify
that the crime scene is not a possible source of fibres believed to have been left by the
offender, and, conversely, that the suspect’s environment is not a source of fibres suppos-
edly coming from the crime scene. Comparing collectives of fibres from different sources,
without having a control sample of known origin, is to be discouraged, except for the pur-
pose of providing investigative leads. The evidential implications of this are discussed by
Stoney (1994). Both he and Aitken (1995) first drew attention to the necessity to consider
not only matching fibre types, but also the whole range of potentially incriminating ‘for-
eign’ fibres.
Many aspects must be considered before formulating conclusions about the results of
fibres analysis. They are summarized below, and can be divided into two categories: known
and unknown facts.
Facts That Are Known from the Outset (or Can Subsequently Be Determined):
• The circumstances of the case

• The time that elapsed before cosllection of the evidence
• The suitability of the fibre types involved for transfer, recovery and comparison
• The extent of the information which can be derived from the evidence submitted
• What evidence might be expected in the light of conflicting hypotheses
• The number of different types of matching fibres
• The number of different types of non-matching fibres
• Whether there has been an apparent cross-transfer of fibres
• Whether there is additional evidence of secondary transfer(s) of environmental
background fibres
• The quantity of the matching fibres recovered
• The location of the recovered fibres
• The analytical techniques used to conduct the examinations
Unknown facts:
• The degree of certainty that specific items were in contact, and to what extent
• The area, duration and pressure involved in contact(s)
• The shedding and retaining potential of the textiles involved
• The frequency of occurrence of the matching fibre types
Much of the information available to fibres analysts to help them with interpretation
has been obtained from research projects. In spite of the self-limiting, labour-intensive
nature of these research projects, considerable advances have been made during the past
twenty years, partly due to increased international cooperation. Aspects of these interpre-
tational aids (see Section 10.2) include:
• Fibre transfer and persistence studies

• Studies on primary and secondary transfer
• Studies on differential shedding
• Live trials (transfer experiments replicating specific case circumstances)
• Experiments on the alteration of fibre characteristics in accordance with localized
conditions specific to a case
• ‘Target fibre’ studies
• Population studies
• Use of data collections
• Industrial enquiries
• Use of a Bayesian approach to evidence evaluation
10.1.2 The Influence of Case Circumstances

It has been argued for some time that in the interest of complete impartiality it is preferable
that analysts be in total ignorance of the circumstances of the case that they are dealing
with. Starrs (1990) was one of first authors to make this claim. Over the last ten years the
topic of contextual bias gained some momentum across the whole forensic science disci-
pline through a number of very active researchers, in particular Dror (2011, 2012). While
we recognize that forensic scientists should not have access to irrelevant and potentially
biasing information, on balance, in a field like fibres, it is more desirable to have too much
than not enough contextual information. The same analytical result may be highly valu-
able or essentially useless (if not completely misleading) depending on the context. In
times of economic restraints and the need for effective case management, an analytical
strategy designed to give the most useful results in the shortest time can only be possible
if the full story is known. The critical aspect of any fibres case is how the crime scene was
treated and the circumstances under which the evidence was collected. Some knowledge
of the background fibre environment of the persons involved in the incident may provide
information crucial to proper interpretation of the findings.
Examiners should have all information relating to the crime scene (see Section 4.2) and,
if they have not visited the scene themselves, they will need comprehensive details about the
chain of events, and the people involved, including answers to the following questions:
• Where did the offence occur; who had (legitimate) access to premises/vehicles
involved?
• If a rape or assault took place in a living area, did it happen on a bed, couch, or
floor?
• At what time did the incident happen, and when was the evidence collected?
• What (exactly) was being worn by the participants?
• Have the appropriate fibre standards been obtained and submitted?
• If the crime scene is a residence or furnished area, are there textiles in that area that
would be an obvious source of target fibres (bright colours/good shedding poten-
tial) which might be expected to contribute to the background fibre population
of that area and therefore be of potential value in providing secondary evidence?
A fibre case involving many items usually means a labour-intensive and time-con-
suming examination. The more information available, the better the assessment of the
likelihood of possible transfer and persistence of fibres between items will be, and the
examination can be planned accordingly. It may be possible to limit a fibre examination
to the most significant items only, for example, the undergarments in a rape case. By con-
sidering the number and distribution of transferred fibres, information may be gained on
whether a transfer is likely to have been primary or secondary. Background knowledge
may help to increase the accuracy of estimations of the chances of transfer and persistence
of fibres from particular garments, which are necessary when using a probabilistic frame-
work for the interpretation.
It is vital that the integrity and security of evidence be maintained at all times. The
analyst must be confident that pollution* has not occurred. The necessary precautions have
been dealt with in Sections 4.2 and 4.3. Strict compliance with quality assurance mea-
sures (see Section 10.1.7) should prevent mishaps. If the defence can successfully allege that
transfer of seemingly incriminating fibres took place after the crime, even the best fibre
evidence will be rendered inconsequential.
* Most readers would be familiar with the term contamination. However, according to the ENFSI Forensic
Glossary (http://esc-dev.unil.ch/moodle/mod/glossary/view.php?id=84) the term pollution is more accu-
rate in this case: ‘Contamination is the natural background noise on investigation scenes. It is normal
forensic science conditions and we have to live with it! We need to understand potential affects on inter-
pretation and conclusions. Pollution, on the other side, is another topic. Pollution is what has been added
due to careless handling. It should be avoided by adopting protective measures or can be documented in
certain circumstances (e.g. medical help).’
The following factors tend to lower the evidential value of findings in fibre transfer
cases:
• Any possibility, no matter how slight, that pollution may have occurred.
• Indications that the transfer may have taken place during a legitimate contact between
subject and victim, e.g. that they were drinking, dining, dancing together, or were
together socially in a particular environment. Time may play an important role here.
Sound knowledge about fibre transfer and persistence is required (see Chapter 4).
• When both suspect and victim share/occupy the environment where the alleged
offence took place, for example assault in a family environment. The number and
distribution of transferred fibres may be of vital importance in such cases. A strong
argument can be presented for not carrying out a fibre transfer examination at all
under these circumstances, as the results can never be interpreted with complete
reliability. An extremely cautious approach, and the use of live tests to replicate the
circumstances as accurately as possible, would certainly be advisable.
• If the fibres involved are of a type that is common and have a widespread distribu-
tion, for example, grey or black cotton fibres that have been dyed with sulphur black.
Conversely, the following factors may tend to enhance evidential value in fibre
transfer cases:
• When an apparent transfer can be demonstrated between complete strangers

involving several colours and types of fibres, and including transfer of other types
of physical evidence besides fibres.
• If there is an apparent cross-transfer, especially when it involves a large number
of fibres.
• If the fibres involved are man-made. Population studies (Grieve and Dunlop, 1992,
Grieve and Biermann, 1997a, Roux and Margot, 1997, Cantrell et al., 2001, Dignan
and Murphy, 2002, Cresswell et al., 2003, Palmer and Oliver, 2004, Was-Gubala,
2004b, Watt et al., 2005, Marname et al., 2006, Lazic et al., 2012; Almazrooei et al.,
2012) have shown that synthetic fibres are highly polymorphic. All these studies
have shown similarities enabling general conclusions to be drawn (see section
10.2.5), indicating that the evidential value of synthetic fibres can be very high.
The fibre colour may also play an important role here.
• When the delay between the offence and the collection of the evidence has been
minimal. This reduces the likelihood of matching fibres, present in quantity, having
originated from other putative sources, unless a contact had occurred within the
same time frame. An interpretation favouring a direct transfer will be more likely
under these circumstances.
• When transfer of incriminating fibres has occurred onto, or better still between,
items having special significance, e.g. underwear, in cases of a sexual nature.
• When the results of a transfer analysis follow in detail a pattern that is totally
consistent with the reported case circumstances.
• If the fibres are unusual in some way (see Section 10.1.5.2).
• If the fibres are of a type which can be objectively said to be uncommon.
• In the event of a large number of matching target fibres being recovered from a
surface which has a low fibre retention potential.
Even very common fibre types, such as blue denim cotton, may assume increased
importance if they are present (in large numbers) in circumstances where it would not be
normal to expect them (Grieve and Biermann, 1997a).
10.1.3 Fibre Frequency

Deadman (1984) categorized fibres into groups depending on their frequency of occur-
rence. On the one hand, there are those fibres that are indisputably common and are
used in the production of a massive number of articles. The best example is white cotton.
Because the sources of these fibres are so numerous and fibres from different sources are
indistinguishable, they can have little value as evidence.
On the other hand, by virtue of an unusual morphological characteristic, extremely
specific end-use, limited production or obsolescence, fibres can be deemed uncommon.
Deadman believed that all other fibre types lie between these two extremes. In the past
twenty years, considerable advances have been made in obtaining information relating to
fibre frequency through:
• Population studies: Which detail the components of a fibre population on a chosen

surface
• Target fibre studies: Which determine the chance of encountering a particular cho-
sen fibre type in a random population of fibres
• Databases: The saving of data relating to frequency of occurrence of fibres of
different types within a given population
These are all dealt with in more detail in Section 10.2, as is the use of industrial
enquiries to obtain information on production and distribution of textiles.
It is apparent to fibres examiners that some types of fibre are much more frequently
encountered than others. This observation needs to be quantified objectively. Figures can
be obtained from the Food and Agriculture Organisation of the United Nations and from
textile journals, e.g. Chemical Fibers International, which reflect production and use of the
main generic types. Useful information can also be obtained through the monitoring of
specialized websites. Examples include:
• http://www.fabriclink.com
• http://fibre2fashion.com
• http://fs2012.empa.ch/
• http://www.fashion-links.de/
• http://www.fashionseek.net/
• http://www.ita.rwth-aachen.de/
• http://www.techexchange.com/
• http://www.texdata.com/
• http://www.texi.org/
• http://www.textile.fr/
• http://www.textileweb.com/
• http://www.textileworld.com
• http://www.textilexpert.com/
• http://www.textilserver.de/
• http://www.trevira.de/
• http://www.textileworld.com
• http://textilesupdate.com
These figures may vary depending on whether they are based solely on clothing or also
on fibres from home textiles, furnishings and technical textiles. Comparisons can be made
with figures from forensic databases and the results of population studies. Scientists should
take care that data they use are representative of the country in which they are operating.
In any case, there are six dominant fibre types: cotton, wool, polyester, polyamide, acrylic
and viscose.
Various factors may influence the frequency of a particular fibre type within the
general fibre population of a specific geographical region, for example:
• Country-specific peculiarities, e.g. cultural and climatic influences

• Production numbers of garments
• The market share held by certain retailers, especially chain stores
• The relationship between the price and the quality of garments – ‘value for money’
Consumer taste (the effects of age, income, social status, fashion trends and consumer
advertising) is not likely to affect the frequency of the commonest fibre types. Garments
are constructed from the fibres which can most suitably fulfil the required specifications.
On the other hand, surveys of personal clothing (Biermann and Grieve, 1998) showed that
the contents of people’s wardrobes are likely to be extremely individual.
The large number of studies undertaken so far allows us to reach a consensus: the vast
majority of fibres in the general fibre population occur as only a very small percentage of the
total (usually less than 1.0%), which increases their evidential value. The following fibre types
fall into this category: acetate, cupro, linen, modal, modacryl, polypropylene, silk and triac-
etate. These fibre types can now be defined as infrequent among the general fibre population.
Even within a generic type, fibres of a particular polymer composition, or manufactured
using a particular solvent, may be relatively rare. Acrylic fibres are a good example. Those
copolymerized with methylacrylate (~60%) and vinyl acetate (~30%) are by far the com-
monest varieties. The value of other types (Grieve, 1995) is correspondingly elevated.
However, we must be careful with simplistic generalization. For example, in the
previous edition of this book, a cautionary tale was presented about a type of acrylic fibre
considered rare because it was manufactured using ethylene carbonate as the solvent. These
fibres were produced in relatively small quantity by a company in Romania and another in
North Korea. It was cautioned that a forensic scientist examining fibres in Romania might
not find them so unusual. Changes which occur in fibre production affect frequency; thus,
bicomponent acrylics are not seen as often as they were two decades ago. Manufacturing
plants change hands and some fibre types are discontinued. Nevertheless, forensic scien-
tists must be prepared to see discontinued fibres occasionally. ‘Verel’ modacrylic fibres from
a car seat cover played an important role in a murder case examined in Finland in 1996,
although production ceased in the early 1980s (Rovas, personal communication, 1998).
Within a generic type, colour plays a major part in the assessment of frequency.
This was illustrated by the work of Grieve and Biermann (1997a) in their survey of out-
door fibre populations, confirming observations made in the earlier study of Grieve and
Dunlop (1992). This work has been further supported by the work of Roux and Margot
(1997) and subsequently by all population studies undertaken. It should be pointed out
that in such surveys colour is typically only assessed subjectively, without the aid of micro-
spectrophotometry (MSP). The percentages of colours accurately measured by MSP would,
of course, be far smaller and if the synthetic fibres were further subdivided into generic
type, the percentage of colour type combinations in this group would become very small
indeed. Collectives of fibres in these types and colours will be immediately recognizable,
and will also be highly significant evidentially.
Conclusions which can be drawn from frequency studies carried out so far have
shown that
• The majority of color/morphology/generic type combinations (approximately 65%)

are represented by a single fibre type only.
• Synthetic fibres form only a low percentage (13%–20%) of any population yet
studied.
• Synthetic fibres exhibit a very high degree of polymorphism.
• The chance of one type representing >1% of the total population is remote.
Through these studies, it is possible to produce realistic numbers relating to fibre fre-
quencies and to show without doubt that certain fibre types occur very infrequently. While
it is necessary to use these figures with caution, they can be taken into account when con-
sidering a probabilistic approach to interpretation and can be used to give an objective
foundation for placing the evidential value into the appropriate step on a verbal scale of
probability (Evett, 1990; Rudram, 1996).
10.1.4 Target Fibres and Case Strategy

The first step in fibre transfer examinations is to decide which items contain suitable ‘target
fibres’. These are the fibres that the analyst will look for to see if he or she can establish
that a transfer has taken place. This choice is related to, and may ultimately affect, the final
assessment of the value of the findings in the case.
The analytical strategy applied in different laboratory systems will depend on the
following:
• Whether the laboratory adopts a partial or full commercial position with the police
paying for the examinations; laboratory policy may thus be related to financial
considerations.
• The case load system within the laboratory at any particular time.
• Whether there are legal requirements to provide examination results within
a certain time limit, e.g. 90 days.
• The degree of priority given to any particular case.
• The degree of priority given to fibre examination in any particular case.
Some of these factors may favour situations where not all exhibits in each case will be
examined, or where searches will not be made for all possible fibre transfers. This situa-
tion demands that the examiner be able to assess the evidential value of the different fibre
types involved in each case before the search commences. (Practical considerations on the
choice of target fibres are discussed in Section 4.3.) The examiner must assess different
aspects of the case to consider what examinations might be performed and what results
may be expected from them in relation to different hypotheses (Cook et al., 1998a and
b). It should not be forgotten that the absence of evidence of a transfer or the apparent
transfer of only a very small number of fibres may also be significant. Whatever fibres are
chosen as targets, it must be remembered that the inability to find any transferred fibres
does not mean that contact did not take place, because complex factors influence transfer
and retention of fibres.
Estimation of the shedding potential of textiles at the outset of a case, or before
planning a series of ‘live tests’ to replicate particular case circumstances, is important.
These test results may play a crucial part in influencing the examiners’ interpretation of
the case findings. However, all shedding tests (e.g. using a length of adhesive tape or more
sophisticated systems) suffer from limitations and only provide crude estimates. When
carrying out live tests, analysts certainly need to be aware of these limitations, and to do
their best to replicate actual case circumstances under very difficult conditions. Forensic
scientists are often pressed in court to try to give better explanations for their findings,
but the limitations have to be understood. For example, to dismiss the use of a textile-clad
stuffed sack, appropriately weighted to represent a child’s body, as unrealistic because it
does not have arms or legs (an actual situation) shows little appreciation for the efforts that
are being made to overcome interpretational problems.
10.1.5 The Influence of Discrimination within a Generic Type

10.1.5.1 The Extreme Variability of Synthetic
Fibres and Target Fibre Studies
Synthetic fibres can be characterized by a large number of morphological parameters which
give them a high degree of variability even before the colourant is considered. Colourless
fibres of one type, for example polyesters produced by one manufacturer, may differ in sev-
eral respects: cross-sectional shape, diameter, concentration and distribution of delustrant
particles. They can have different refractive indices or birefringence, and different melting
points or infrared spectra due to varying polymer composition or crystalline structure. All
these differences can be determined without difficulty by the forensic scientist.
By the time fibres from this production series have been dyed, the result will be an
extensive range of differentiable fibres within one generic type-colour combination.
A small range of dyes can be used to produce a very large number of colours by combina-
tion using the CIELAB colour system. For example, 12 serilene dyes, used for polyester
staple, can be mixed to give 220 possible colour combinations. The total number of acid
dyes used throughout the industry to dye polyamide fibres is only about 30, but again
a large number of combinations is possible.
Consider a textile (ladies’ pullover) made from an average synthetic fibre, for example,
a yellow, non-delustred, bean-shaped acrylic. How many additional textiles must one
examine before finding another one containing fibres that are not differentiable in every
respect? Despite the perception of the public in general, and defence lawyers in particular,
that of course all yellow acrylic fibres are the same, nothing could be further from the
truth. Research (Biermann and Grieve, 1997a) shows that the chances of finding a garment
containing identical yellow, non-delustred, bean-shaped acrylic fibres among a random
population of garments (in Germany) are about 1 in 20,000. If the fibres are of a less com-
mon type and colour, e.g. turquoise acetate, the chances may decrease to around 1 in
250,000 or more.
Natural fibres do not display the same wide variety of morphological characteristics:
their main comparative feature is colour. This, and the high usage of cotton and wool,
tends to reduce somewhat their evidential value compared with synthetics.
The high degree of polymorphism in synthetic fibres has been verified by ‘target fibre
studies’. These studies give some idea of what proportion of the population has fibres of
a particular type on their clothing (or car seat, cinema seat, pub seat, etc.). These data
incidentally will automatically take into account not only the rarity of a fibre type, but
also the tendency for this fibre to be transferred or to persist on the receiving fabric.
Considerable advances in target fibre studies have been made in the past twenty years,
thanks to increased international cooperation that has contributed to reducing the enor-
mous amount of labour necessary for these projects.
Details about target fibre studies can be found in Section 10.2.4. Sufficient studies have
now been carried out to allow some important general conclusions to be drawn from them:
• The chance of finding a particular colour/type combination among a random

population of foreign fibres is very low.
• The chance of a ‘collective’ of these fibres being present is even lower, especially if
synthetic fibres are involved.
• The occurrence of a group of fibres on a surface that match those from a particular
textile source constitutes strong evidence of contact with that source.
• The chance of finding differences between potentially matching target fibres
recovered from a particular surface will increase proportionately to the number of
comparative tests used to examine them.
• The environment where the fibres are recovered may be significant, as the chance
that ‘matching fibres’ will be found on an item that has been subjected to ‘relentless
contacts’ (e.g. seating in public places) is theoretically higher.
The extreme polymorphism and high evidential value of synthetic fibres can be
effectively demonstrated by showing that matching fibres providing incriminating
evidence could not have originated from other putative sources (alibi specimens) which
might be put forward by the defence. In a murder case where black acrylic fibres recov-
ered from the victim’s body matched those of the suspect’s pullover, 31 additional samples
of known black acrylic fibres from other pullovers were submitted for comparison. Some
differed microscopically, and in no case did the dye composition match that of the fibres
from the suspect’s pullover. This provides an effective rebuttal to the allegation ‘of course,
black pullovers containing these fibres are very common.’
10.1.5.2 Comparison of Fibre Characteristics

The more characteristics that two fibres have in common, the greater is the chance that
they originated from the same source. The comparative information that can be obtained
depends on the equipment available to the analyst and on the discriminating power of the
techniques used. Thanks to the use of microspectrophotometers which operate in the ultra-
violet and visible spectral ranges and the development of FTIR microscopes and Raman
microprobes, problems associated with small sample size have largely disappeared. FTIR
microscopes have eliminated the time-consuming preparation problems which were often
detrimental to obtaining good quality infrared spectra, meaning that polymer composi-
tion of virtually any fibre fragment can be analyzed using this technique. Extending the
use of microspectrophotometry into the ultraviolet range has largely eliminated the need
to use thin-layer chromatography as a complementary technique for colour comparisons.
Similarly, Raman microspectroscopy is excellent for examining fibre dyes and pigments
without the need for extraction.
Failure to use state of the art techniques for fibre identification and comparison can
lead to a reduction in evidential value, as the number of potential alternative sources will
rise considerably if all comparative possibilities are not exhausted. Not all the techniques
available to the forensic fibre analyst must be used in every case. As indicated in previous
chapters, a large pool of analytical methods can be used appropriately depending on the
type of fibre specimen and the case circumstances. At this stage, it must be emphasized
that a technique is only as good as it allows the forensic scientist to answer the question that
is being asked. Some guiding principles are described below.
Fibre colour comparisons at present rely on comparison microscopy (bright-field and
fluorescence) plus microspectrophotometry (240–760 nm) or in the visible range only with
additional examination of the extracted dye using a chromatographic method.
The following features may enhance the evidential value of individual fibres
(see Figure 10.1):
• Unusual polymer composition

• If they are bicomponent
• Adherence of characteristic trace debris, e.g. adhesive residue
• Alteration due to the effects of heat
• Overdyeing
• An unusual colourant for that fibre type, e.g. pigmented acrylic or polyester fibres
• If they are dyed with a dye belonging to a class not normally used on that type of
fibre, e.g. acid-dyed acrylic
• Manufacturing faults, damage, voids, inclusions, cross-markings or porosity
marks
• Channels or anti-static inclusions
• Presence of flame retardant
• Unusual cross-sectional shape (or unusual for that fibre type)
• If they have undergone unusual physical processing
• Modified appearance due to subjection to localized conditions, e.g. immersion in
water, exposure to strong sunlight
• If that fibre type is a very unusual component in that particular type of textile
Polyester fibres have been produced in the traditional round or trilobal shapes for
many years. Pentalobal or octalobal forms, however, are less common. Variations of cross-
sectional shape are usually associated with an attempt to impart special properties to the
fibre. Du Pont and Allied Signal have recently produced a number of polyamide fibres with
interesting and characteristic cross-sectional shapes. Forensic scientists should use textile
journals, e.g. Chemical Fibers International, to keep informed about these new develop-
ments and have some idea of the production quantity and uses of these new fibres.
Cross-sections can provide information on the fibre manufacturer, the spinning process
used, the end-use, physical processing, fibre quality and the dyeing method (Palenik and
Fitzsimmons, 1990a, b). The modification ratio of synthetic fibres can be used to identify
the manufacturer, thus opening the way to an industrial enquiry.
(a) (b)
(c) (d)
(e) (f )
(g) (h)
Figure 10.1 Unusual features found in fibres which may enhance their evidential value: (a) car-
bon anti-static inclusion in ICI ‘Timbrelle’ nylon fibres; (b) damaged areas caused by faulty
production in a Hoechst ‘Trevira’ polyester fibre; (c) vacuoles in an acrylic fibre—one of several
recovered in casework—cause unknown; (d) microvoids, believed to have an anti-static function,
in ICI nylon 66 bulked continuous filament trilobal carpet fibres; (e) ribbon-form acrylonitrile/
vinyl acetate co-polymer fibres (this cross-sectional shape is often seen in modacrylic fibres, but
very seldom in acrylics); (f) ‘fish eyes’ in ‘Kanekalon’ modacrylic wig fibres—these are draw-
marks caused by particles of undissolved material interrupting the polymer flow; (g) an example
of an acrylic fibre that has been dyed with the Courtauld’s ‘Neochrome’ process, producing the
‘tiger-tail’ effect; (h) an Orlon 21 bicomponent acrylic fibre—these mushroom-shaped section
fibres have a dyed delustred component and a colourless non-delustred component; the dye may
‘bleed’ into the colourless part. ([a, c and h] With kind permission from Springer Science+Business
Media: Forensic Science Progress, Vol. 4, 1990, A. Maehly and R. Williams [Eds.].)
Sometimes, a particular dyeing process may give fibres a very characteristic appear-
ance, such as the ‘tiger tails’ which result from the Neochrome process developed by
Courtauld’s to dye Courtelle acrylic fibre (Todd, 1991). The overdyeing of shoddy fibres
will also produce fibres where traces of the original colour are still visible. Features such as
these enhance evidential value.
Care must be taken to avoid false exclusions after fluorescence examinations where
unexpected differences occur between recovered fibres and the reference fibres although
they match in all other respects. Taking additional controls will normally reveal that these
differences arise owing to local conditions, for example:
• Surface fibres may fluoresce differently from those deeper within the fabric.
• Alterations from excessive wear, for example in the central area of a well-worn car seat.
• The fabric has been washed or dry-cleaned recently or has some localized staining.
• Alterations caused by subjection to environmental factors – exposure to strong
sunlight, or immersion in water or another fluid.
Once an explanation for these inconsistencies is forthcoming, additional tests may result
in a positive comparison, after which the evidential value will be enhanced.
It is often asked whether or not it is possible to assign or negate the possibility of ques-
tioned and known fibres sharing a common origin based on characteristics resulting from
exposure to particular environmental conditions using sensitive instrumental techniques.
The problem is that the frequency of occurrence of particular characteristics or groups of
characteristics and the degree of variation both in and between garments are at present
largely unknown. Also, changes as a result of environmental influences could take place
after a fibre transfer has occurred, which might lead to a false exclusion.
Finally, it is worth mentioning an approach recently proposed by Stoney et al. (2015a,
2015b) focusing on the automatic detection and analysis of very small particles (VSPs)
present on fibres using SEM-EDX. These VSPs come from the environment and essentially
impart an acquired dimension to mass-produced items like fibres. The evidential value
of these fibres is therefore significantly enhanced if the VSP profiles found on unknown
specimens and a putative source are statistically non-differentiable. At present, the full
applicability of this approach has not been demonstrated in routine casework but it is
highly promising.
10.1.6 The Number of Matching Fibres Recovered

The interpretation of a fibre transfer examination is strongly influenced by the number of
matching fibres recovered. The problem is that the number of fibres originally transferred
will always be unknown. Does the number recovered represent a high or low percentage
of this number? For example, recovery of 20 matching fibres from a high-shedding donor,
could, depending on other circumstances, be attributable to a secondary transfer. If, on
the other hand, the donor is a very poor shedder, the recovery of 20 fibres may be a strong
indication of a primary contact. In addition, the rate of loss over the intervening (and
sometimes unknown) time of wear after the transfer is also unknown.
Since the 1970s a number of studies have provided a sound basis for an understanding
of the factors to be considered in the interpretation of the transfer and persistence of fibers
in forensic investigations. The most important concepts and seminal studies are presented
in Section 4.3. They are summarized in Roux and Robertson (2013b) and Roux et al. (2013).
It has been suggested that it is helpful in assessing the evidential value to recover
all matching fibres from each item of evidence, unless the numbers run into hundreds.
Present constraints in forensic science laboratories mean that this approach may no longer
be considered practical or justifiable. In some cases, recovery of perhaps 20 matching fibres
of good evidential value may be enough to elicit a pre-trial confession. This way of think-
ing has been spawned by economic considerations, and its acceptance may depend on the
requirements of the legal system under which the laboratory is operating. Can anything be
gained by the recovery and exhaustive testing of all matching fibres? The question is how
many fibres constitutes a representative sample of what is originally a specimen of fibres
that had been transferred in an uncontrolled manner in the first place?
Information on fibre distribution (which may fit the case circumstances) can be obtained
by recovery of all target fibres (or fibres in a collective), but because of redistribution due to
packaging (Robertson and Lloyd, 1984; Robertson and Olaniyan, 1986), the information is
only reliable in circumstances where 1:1 taping has been used, for example from a body at
a crime scene (naked or clothed) before it has been moved; or in an instance where the fibre
distribution turns out to be overwhelmingly disproportionate.
Recovery of all target fibres may help to provide information on whether a transfer
has been primary or secondary (Grieve et al., 1989; Lowrie and Jackson, 1991, 1994),
particularly if only low numbers of fibres are involved.
Several reasons can be used to explain recovery of only a few fibres, all of which must
be considered when assessing examination results. They are as follows:
• It could be the result of inefficient fibre collection. Jackson and Lowrie (1987)
studied the efficiency of using adhesive tape, which was found to vary from
32% to 100% depending on the degree of adhesiveness of the tape used. Further
information on the efficiency of different methods of fibre recovery was provided
by Pounds (1975) and Roux et al. (2001).
• There may have been a considerable time lapse between the offence and recovery of
items for examination. A number of fibre persistence studies have been undertaken
over the years, including by by Pounds and Smalldon (1975), Robertson et al. (1982),
Lowrie and Jackson (1991) and Akulova et al. (2002). They all confirmed that the
loss of transferred fibers starts immediately after the contact which resulted in the
transfer and that this loss is often rapid in the first few hours. The effect of weather
conditions on fibre persistence has been studied by Krauß and Hildebrand (1995)
and Krauß and Doderer (1997). Palmer and Burch (2009) performed a combined
investigation into the transfer and retention properties and fibre population study
of the skin of living subjects. Palmer and Polwarth (2011) carried out a study to
investigate the persistence of fibres on skin in an outdoor deposition crime scene
scenario. Hong et al. (2014) gathered data on the transfer of fibres on hands and
their persistence during hand washing. Lepot and Vanden Driessche (2015) studied
the persistence of fibres on immersed garments.
• If there has been a delay, fibres will be lost by dropping off or by further transfer
to other surfaces, both textile and non-textile. Redistribution of fibres after their
original deposition has been originally discussed by Robertson and Lloyd (1984).
Chewning et al. (2008) examined the persistence of fibres on the interior and exte-
rior surfaces of ski masks during transportation to the FBI Laboratory and during
evidence processing in order to assess if and how these fibres are redistributed.
A study investigating into the redistribution of fibres on garments being washed

has been undertaken by Szewcow et al. (2011).
• It could be due to the low shedding capacity of the textile donor. Hellwig
(2011) investigated the effect of textile construction on the shedding rate of
knitwear.
• The duration, pressure and/or area of contact involved may have been minimal.
• The garments involved may have been subsequently washed or dry-cleaned. The
effects of these treatments have been studied by Robertson and Olaniyan (1986),
Grieve et al. (1989), Palmer (1997) and Szewcow et al. (2011).
• When a garment is made from a fibre mixture, the various fibre types will not
necessarily be transferred in the same ratio as the percentages of the individual
components. Transfer will depend on the textile construction and how easily the
fibres fragment. Some types may transfer substantially less than others. The pro-
cess is called ‘differential shedding’, and information can be found in Mitchell and
Holland (1979a), Parbyk and Lokan (1986), Salter et al. (1987) and Hellwig (1997).
A case example is included in this chapter.
• A few matching fibres may be the result of a secondary or tertiary transfer, as
opposed to a primary one. In experiments (Jackson and Lowrie, 1987) involving
items of clothing, secondary transfer of at least one fibre took place at least 50%
of the time. However, in only 3% of the 120 tests were more than five fibres
transferred. Secondary transfer is likely to play a more significant role with seating
than with clothing, as seats are immobile, less frequently cleaned, and may not be
subjected to continuous contacts. A pool of fibres may be created during a primary
transfer, from which fibres may persist after several secondary contacts or remain
undisturbed for a long period of time (Grieve and Biermann, 1997b). The ques-
tion of secondary transfer between the garments of one individual was also briefly
discussed by these authors. Palmer and Banks (2005) considered the effects of fibre
type, hair style, time and fibre persistence on the secondary transfer of mask fibres
to pillowcases, via head hair.
Mathieson and Elliott (1994) made significant observations on secondary transfer.

These can be summarized as follows:
• There is an area of overlap between the number of matching fibres found after
a direct (primary) or an indirect (secondary) transfer.
• It is not possible to say that a low number must be attributable to a secondary
transfer any more than it is possible to state categorically that a larger number, e.g.
50 fibres, must be due to a primary transfer.
• The number of fibres remaining will be a function of the original number
transferred; estimation of this number is critical to considerations of primary
versus secondary transfer.
• If a large number, e.g. more than 50 fibres, are recovered from undergarments,
it may be possible to say that this is the result of a primary transfer.
In all interpretations involving primary versus secondary transfer, it is necessary to

consider all factors that can have an effect on fibre transfer and persistence. Under certain
conditions (not uncommon in forensic science casework), the recovery of 50 (or fewer) fibres
may be a strong indication of a primary transfer. There are also circumstances which could
result in a secondary transfer of over 50 fibres. Every case must be assessed individually.
10.1.7 Quality Assurance

It has already been mentioned that the choice and extent of the analyses used during a fibre
examination may affect the amount of information gained and the conclusions which can
be drawn. If the results are incorrect, the conclusions drawn will also be flawed. Quality
assurance (QA) is inextricably linked to ensuring that conclusions and opinions presented
at the end of an examination can be relied upon.
QA is defined as ‘a program for the systematic monitoring and evaluation of the vari-
ous aspects of a project, service, or facility to ensure that standards of quality are being met’
(Merriam-Webster’s Online Dictionary, 2016). It is an activity designed to provide evidence
to all stakeholders that activities related to the delivery of a service or the manufacture of
a product are being performed effectively and to an appropriate standard (Lennard, 2013).
Requirements for a properly established QA program generally include:
• An organizational structure appropriate for the range of activities undertaken

• Adequate staffing levels and resources
• Required education and training of staff
• Appropriate facilities, including laboratory, office, and storage areas (with ade-
quate security and access control)
• ‘Fit-for-purpose’ equipment that is regularly maintained and calibrated
• Case file and exhibit management systems
• Documented methods and procedures (to cover both administrative and technical
activities)
• Regular audits, quality control (QC) tests, and staff proficiency tests for the ongo-
ing monitoring and assessment of performance
• Routine peer review of case files, results, and reports
A QA program needs to be based on relevant international standards, with ISO 17025

being the most widely used standard in forensic science laboratories and ISO 17020 by
some crime scene units. Further, all policies, procedures, and operating instructions that
may impact on the quality of the services and the analytical results provided need to be
fully documented in standard operating procedures (SOPs). Further, each procedure needs
to be validated to demonstrate that it is capable of providing reproducible, reliable and
robust results that are fit for purpose. Validation should be appropriate and proportionate.
Individual laboratories may also require independent verification even when there has been
extensive broad validation. This is important to demonstrate that a procedure or technique
is working according to specifications.
In the last 20 years, a considerable effort has been directed towards standardization
in forensic science, including in the forensic examination of fibres. Various organisations
internationally have developed and recommended relevant guidelines and standards,
including:
• Manual of Best Practice of the European Network of Forensic Science Institutes’

Fibres Group (now European Hairs and Textiles Group – HTG);
• Forensic Fiber Examination Guidelines of the Scientific Working Group for

Materials (SWGMAT), USA*
• SWGMAT Trace Evidence Recovery Guidelines, USA
• ASTM International, through Committee E30 on Forensic Sciences
• AS 5388 Australian Standards on forensic analysis, respectively on recognition,
recording, recovery, transport and storage of material; analysis; interpretation;
and reporting
During that time, a large number of laboratories have become accredited by an

accreditation body such as the Laboratory Accreditation Board of the American Society
of Crime Laboratory Directors (ASCLD/LAB), the Forensic Specialties Accreditation
Board (FSAB), the National Association of Testing Authorities (NATA) in Australia or
the European cooperation for Accreditation (EA). Formal accreditation provides an
independent verification that an organization’s quality system is fit-for-purpose and
addresses all of the relevant requirements of the standard (Lennard, 2013). This can assist
in ensuring quality and integrity whenever scientific evidence is presented in court.
At the individual level, schemes have also been introduced for the individual
certification of forensic scientists, as practised by the American Board of Criminalists
and the Chartered Society of Forensic Sciences. It remains the responsibility of employer
organisations to authorise caseworkers.
Guidelines for fibres examination should cover the following subjects: general quality
assurance/quality control guidelines (also applicable to all trace evidence areas) with
specific sections on evidence recovery, microscopy, microspectrophotometry, thin- layer
chromatography, infrared spectroscopy, other instrumental techniques, textile stucture
and alterations to textiles, as well as ropes and cordage.
The function of having guidelines is not to dictate rigid adherence to certain methods,
because there may be alternative ways of achieving the same objective and factors such
as the availability and cost of instrumentation, the training and experience of staff make
standardization within different laboratory systems difficult. Guidelines demonstrate the
need to follow internationally accepted procedures in order to guarantee a result that meets
acceptable standards at an international level. There is also a drive to move guidelines
towards a more robust ‘standards’ level of authority.
Some aspects of fibre examination lend themselves much more readily to standard-
ization than others. Within the classical microscopical methods used for measuring the
optical properties of fibres, there is little variability in results, although there are alternative
methods of measurement. Spectrophotometers need to be properly set up and calibrated.
Spectra need to be produced under defined, standardized operating conditions and can
only be compared with spectra run under the same conditions using the same instrument.
Reference collections of infrared spectra should be produced ‘in-house’, using authenti-
cated standards obtained from manufacturers. Hard and soft copies of all results should
be kept and should be sufficient to allow an independent review if that should ever become
necessary. Verification of results and cross-checking of reports, peer and supervisory
* SWGMAT documents are being updated and gradually replaced as documents of the Materials (Trace)
Subcommittee of the Organization of Scientific Area Committees (OSAC) of the National Institute of
Standards and Technology (NIST).
review are now considered essential. All these will be captured in the quality system or
program of an accredited facility.
With fibre transfer examinations it is particularly difficult to monitor a crucial part
of the examination – the securing of the evidence (often done outside the laboratory) and
the recovery of fibres from adhesive tape-lifts. Much rests on the degree of skill and con-
scientiousness applied during these stages. Quality assurance addresses these problems by
developing training programmes for new examiners, considering requirements for con-
tinuing professional development (including the need to take part in proficiency testing,
to attend training courses and to keep abreast of current literature) and advising on mini-
mum qualifications for examiners and supervisory staff. The use of a checklist for persons
submitting exhibits for examination may assist in ensuring the integrity of the process.
Such a checklist might cover the history of the collection process and whether packaging
meets the required standards. A checklist can also be useful during the case review process
after completion of the examination.
The final aspect for any fibre examination is interpretation and reporting of findings.
AS 5388 (parts 3 and 4) provides a broad standard that can be applied to fibre examination.
This standard allows for the use of a likelihood ratio approach but does not mandate its
use. It provides guidance for formulating an opinion and estimating probabilities. AS 5388
recognises the need to use professional judgement but this needs to be properly defined
in individual quality systems to avoid inappropriate applications. The recent ENFSI
guidelines for formulating evaluative reports (ENFSI, 2015) provide stronger guidance
towards a more general use of the likelihood ratio approach.
The overall goal of a quality system remains that of reducing the chance of an incor-
rect result or unsupportable opinion leaving the laboratory and minimising the possibil-
ity of human error. For this reason, over the last 20 years, quality management in forensic
science materialized into the quality triangle of standardisation, accreditation and certi-
fication. However, despite the benefits described above, quality management should not
be seen as a magic bullet replacing sound holistic case management relying on the initial
recognition and appropriate treatment of fibres at the scene and in the laboratory and on
appropriately educated and trained forensic scientists. Fibres, as a field of forensic science
endeavour, epitomize this quote by Willis (2014): ‘[…] it is necessary to refocus on people.
Forensic science is not a bundle of tasks or routine processes although it includes these.
It needs productive thinking, problem solving, cognitive ability and decision-making
about which, when and why. None of this is at variance with accreditation. In fact if used
in a continuous improvement way, accreditation acknowledges these skills. However the
more fragmented the process becomes, the less accreditation helps address the need for
an overall approach to the questions raised to help solve crime and evaluate findings in
the judicial process.’
10.1.8 Report Writing

In the authors’ opinion, the forensic scientist is obliged to give some indication of the
significance and value of his/her evidence in the report, provided it is legally permissible.
The expert witness is under a moral obligation to be as helpful as possible when explain-
ing the complexities of scientific evidence, while maintaining complete impartiality. The
forensic scientist is in the best position to give an opinion on the value of the fibres evidence.
There are no simple and universal rules about what should be included in a report. This
is also variable and depends on the jurisdiction, legal system and local common practices.
Having said that, needlessly complicated and badly structured reports full of complex
scientific terminology are not appropriate. The product should be succinct and easy to
follow. Over-caution and reluctance to express an opinion will certainly not assist a judge
and jury.
According to ENFSI (2015), reports should include (not-exhaustive list):
• Conditioning information used

• Mandate and questions asked, if required
• The propositions of interest
• Relevant items collected/received
• Items examined
• Significant findings
• Discussion and evaluation
• Conclusion(s)
• A caveat that any change in conditioning information may require assessments,
conclusions and/or propositions to be reviewed
ENFSI (2015) also provides excellent guidance about general reporting require-
ments and clarifies four essential requirements for reports: balance, logic, robustness and
transparency. Further, the same document identifies these concepts as follows:
• Balance: The findings should be evaluated given at least one pair of propositions:
usually one based upon one party’s account of the events and one based upon
an alternative (opposing party’s) account of the events. If no alternative can be
formulated, the value of the findings cannot be assessed. In that case, forensic prac-
titioners should state clearly that they are not reporting upon the value of the find-
ings. As explained in Section 10.3, the formulation of appropriate propositions at
the appropriate level – source or activity – is crucial to the interpretation process.
• Logic: Evaluative reports should address the probability of the findings given the
propositions and relevant background information and not the probability of the
propositions given the findings and background information. The report should
not contain statements that are transposing the conditional (see Section 10.3).
• Robustness: The reporting should be capable of sustaining scrutiny and cross-
examination. It should be based upon sound knowledge and experience of
the trace type(s) and the use of data. The forensic practitioner will be satisfied
that the results of the observations and analyses upon which inferences and
conclusions are drawn are robust. When there are insufficient data, the likeli-
hood ratio approach provides the practitioner with a framework for structured
and logical reasoning based on his experience, as long as he can explain the
grounds for his opinion together with his degree of understanding of the par-
ticular trace type.
• Transparency: The reported conclusions should be derived from a demonstrable
process in both the case file and the report. The report should be written in such a
way that it is suitable for a wide audience of readers (i.e. participants in the justice
system). It may include supplements explaining the technical background.
In courts, a great deal of emphasis is often placed on fibre frequency. The ability of
the expert to assess this will depend on experience, familiarity with research on popu-
lation and target fibre studies and whether or not he or she has access to any form of
representative database. It must be made clear that recovered matching fibres could not
only have originated from a particular garment, but also from a textile containing identical
fibres (provided there was an opportunity for contact with such a second item). Thanks to
the large body of research available today, analysts now have documented material point-
ing to the very high degree of polymorphism of synthetic fibres and to the low chance of
fibres non-differentiable to those in a given textile being present, either in another garment
chosen from a random population, or as loose foreign fibres in a random population.
It should be possible to decide whether findings are consistent with a recent direct
contact, a casual or non-recent direct contact or a secondary contact. Often, various
alternative explanations can, and should, be offered.
Beyond some level of generality, there will be differences of opinion as to what information
should be included in a report. The opinion of over 100 fibre examiners in Europe and in
North America was sought by Grieve (2000a, 2000b) The percentages of respondents who
were in favour of always including the following information were as follows:
• The techniques that were used in carrying out the examination – Europe: 88%;
North America: 63%
• The objective(s) of the examination (e.g. to look for any evidence of fibre transfer
between items X and Y) – Europe 84%; North America 29%
• A summary offering possible interpretations of the findings – Europe 78%; North
America 33%
• The number of matching fibres found – Europe 70%; North America 65%
• A statement of which target fibres were being sought (not all fibre types involved
are suitable as targets) – Europe 65%; North America 49%
• Explaining if, and why, certain items were not examined – Europe 59%; North
America 23%.
When asked if they felt further explanatory material should always be included, the
response was as follows:
• That under the known circumstances, legitimate contact could have occurred – 65%.
• There could have been a possibility of pollution because of faulty packaging –
63% (in many replies it was stated that under the above two circumstances no
examination would be carried out).
• Whether certain fibre types could be considered as being very common, with a
reason – 49%.
• The finding of obvious ‘collectives’ of fibres (other than the target fibres) on tapes – 33%.
• The possibility of secondary transfer – 29%.
• The ‘sheddability’ (= donor capacity) of various garments – 23%.
The great majority of the remaining answers were in favour of including some or all of
this information sometimes, on an ‘as needed’ basis.
In formulating conclusions, it can be useful for the examiner to consider opposing
hypotheses (in favour of prosecution and defence arguments respectively) in relation to
findings to answer questions about different aspects of the case. The survey showed that the
exact formulation of these questions is a matter of critical importance.
10.1.9 Examples of Casework Findings

Unless stated otherwise, it is assumed that the techniques used for identification included
bright-field and polarized light microscopy and FTIR microscopy, and that comparisons
were made using bright-field and fluorescence comparison microscopy and microspectro-
photometry either between 240 and 760 nm or in the visible range only, with additional use
of thin layer chromatography.
10.1.9.1 Example 1
The subject attempted to murder his girlfriend by attaching an explosive device to the underside
of her car. It failed to detonate, and was submitted to the laboratory for examination. The
timing device and batteries were held together by adhesive tape.
Numerous fibres were recovered from the sticky surface of the adhesive tape, including
animal hairs, blue denim cottons, a pale green polyester and two yellow-green polypropylene
fibres. As a result of this examination, the suspect’s apartment was searched for possible
sources of these fibres.
The search revealed a folding camp seat with polypropylene fabric, a green polyester/
cotton sports shirt, a pair of blue jeans and a black cat! All the fibre types matched those from
their putative sources.
Factors which increase the value of the findings are as follows:
• The protected location of the fibres under the tape surface indicates that they were
deposited during assembly of the bomb.
• Four types of matching fibre are involved, even though two of them are in
themselves common.
• Polypropylene fibres are not very common among the general fibre population,
and infrared spectroscopy and melting point examination showed them to be of
the same type as those in the camp chair.
• Information revealed that only 12 chairs of this type had been sold from the local
Post Exchange store, one of them to the suspect.
10.1.9.2 Example 2
The subject approached the victim from the rear, placed a hand over her mouth, dragged her
into a nearby field and raped her. The man was arrested shortly afterwards and his jeans and
T-shirt seized as evidence within four hours.
The victim was wearing a dress made of very characteristic turquoise viscose rayon
fibres. Over 1200 of these fibres were recovered from the suspect’s T-shirt, mainly from the
chest area. In addition, numerous examples of the same type were recovered from his jeans.
Over 80 blue cotton fibres matching those in the suspect’s T-shirt were recovered from the
victim’s dress.
The exceptionally high number of recovered fibres supports the interpretation of a
recent, considerable, primary contact. The presence of so many matching viscose fibres
on the chest area of the T-shirt is consistent with her back resting and rubbing against the
suspect’s chest as he dragged her into the field. Turquoise viscose fibres are not particularly
common (see below) and this, together with a cross-transfer involving all items submitted
(which is also consistent with the circumstances), helps to strengthen the evidential value
of the findings. The subject had worn a condom; no serological evidence was available,
emphasizing the value of the fibres evidence.
If the case had taken place after the establishment of the Catalogue Data Base
(Biermann and Grieve, 1998), this would have shown that of the 2106 dresses included
among 81,898 garments, only three were made of 100% turquoise viscose rayon (0.142%);
and that the chance of finding such a dress from a random population of garments is 3.6 in
100,000. If morphological data indicating that only 50% of viscose fibres encountered are
non-delustred are included, the chances diminish to 1.8 in 100 000 garments.
The important question to answer in order to evaluate the evidence properly is ‘what
are the chances of these fibres not only originating from such a dress but also from any
other textile source containing these fibres?’ As an example, among the 81,898 garments
are 263 others made of turquoise viscose. If only 40% of these contained viscose fibres that
are irregular in cross-section and non-delustred (morphological data), this would produce
a figure of 1.2 in 1000. If a figure relating to the individual dye spectral frequency could be
included (based on discrimination between green reactive dyes used on cellulosic fibres),
a realistic end figure would be of the order of 2 in 100,000 textiles.
10.1.9.3 Examples 3 and 4 (Recovery of a Single Fibre Only)

The suspect invited a girlfriend to his apartment and subsequently had sex with her, against
her will, on top of his bed. He denied using force and said that intercourse was consensual,
but had taken place on the living-room floor – she had never been in the bedroom. Since the
victim was naked at the time of the offence, and a demonstration of contact between their
clothing or a search for carpet fibres on her clothing would be of no value, a fibres examina-
tion was not requested.
In the combings from the victim’s pubic hair was a long, dark golden yellow polyester
fibre. The fibre was non-delustred and was pentalobal in cross-section, making it a very
unusual and characteristic fibre. It matched the fibres in the suspect’s bedspread.
While not helping with the question of consent, the unusual fibre type, coupled with
its recovery from a seemingly incriminating location, seemed to lend support to the vic-
tim’s statement.
However, another plausible hypothesis might be that as the polyester bedspread was on
the suspect’s own bed, a fibre fragment had at some time been transferred to his pubic hair
where it had remained and had subsequently undergone secondary transfer to the victim
during intercourse on the living-room floor. This example illustrates the need to consider
all possible alternative explanations when interpreting fibre transfers!
After a breaking and entering involving the smashing of a wooden store-room door, a single
long curly acrylic fibre was recovered, caught on splintered wood. The fibre had an unusual poly-
mer composition (methylmethacrylate co-polymer) and was dyed yellow-brown – the dye having
many different components. The fibre matched those in a brown pullover taken from a suspect.
As in the previous example, the value of the fibre lies very much in its unusual char-
acteristics, together with its point of recovery, indicating that the wearer of a garment
containing these fibres had contact with the door as, or shortly after, it was splintered. In
the majority of cases, strong emphasis should not be placed on the transfer of just one fibre;
but these two examples show that highly characteristic fibres should not be dismissed on
the grounds of quantity alone.
10.1.9.4 Example 5
Two men were involved in a fight in a bar. One stabbed the other in the chest and fled. On
their way to the scene, the police saw a man running away and arrested him. His clothing did
not appear bloodstained, but he could not give a satisfactory explanation for his behaviour.
A knife was recovered from the pavement outside the bar.
The knife and the victim’s shirt and leather jacket were submitted to the laboratory.
The suspect’s fingerprints were found on the knife. Human blood was found on the blade,
but no DNA result was obtainable.
The upper left-hand side of the victim’s jacket and shirt had cuts in them. The jacket
lining consisted of dark green nylon 6 fibres. The shirt was made from 50:50 polyester–cot-
ton, and had narrow vertical stripes in the design which were grey, brown, green and
dark green.
13 Dark green Nylon 6 Matching the lining of the victim’s jacket

5 Dark blue/grey Polyester Matching a seam thread in the victim’s jacket
4 Grey Cotton Matching fibres in the victim’s shirt
1 Dark green Polyester Matching fibres in the victim’s shirt
4 Brown Cotton Matching fibres in the victim’s shirt
2 Brown Polyester Matching fibres in the victim’s shirt
2 Pale green Polyester Matching fibres in the victim’s shirt
Forty-seven fibre fragments were removed from the blade of the knife:
The remaining fibres were 10 colourless cotton, three colourless polyester, two blue
denim cotton and one dark grey cotton.
Although none of the fibres are particularly unusual types, the evidence linking the
weapon to the offence is strengthened by the large number of different fibre types and
colours (seven) recovered from the knife blade matching three different sources in the vic-
tim’s clothing. The percentage of potentially incriminating fibres on the knife blade in
relation to the general background fibre population on it was very high, suggesting recent
deposition. Linking the knife to the suspect is dependent on the fingerprint evidence. He
could claim that he had picked the knife up in the street and then discarded it. The chain
of evidence would have been complete if the blood of the victim had been found on the
suspect’s clothing or hands.
10.1.9.5 Example 6
An 18-year-old woman was hitchhiking home on a summer evening after visiting a girlfriend
in the country. She was picked up by the suspect, who drove her into a wooded area and raped
her inside the car. He wore a condom which was not recovered. He threw her out of the car
and drove off. She went to the police and described the suspect and his car, as a result of which
the suspect was apprehended and his car impounded.
The following items were submitted for examination.
Victim’s clothes:
Shirt Blue acetate/blue polyamide
Halter top Grey acrylic with black sequins
Hot pants Blue acetate/blue polyamide
Panties White cotton
Suspect’s clothes:
T-shirt White cotton
Jeans Blue denim cotton
Underpants Purple cotton
From the vehicle:

• Piece of fabric
• Tapings from passenger seat
The findings were as follows.

Suspect’s T-shirt 0
Suspect’s underpants 10
Suspect’s jeans 12
Tapings from passenger seat 33
A small piece of material appeared to have been torn out of the victim’s shirt. The
fabric construction, composition and dimensions of the piece of material recovered from
the suspect’s car were such that it was completely consistent with originating from the
victim’s shirt.
Blue acetate fibres, matching those in the victim’s shirt and hot pants were found
as follows.
Shirt 3 Delustred 2 Non-delustred
Hot pants 1 Delustred 1 Non-delustred
No fibres matching the blue polyamide fibres in these items, or grey acrylics from the
halter top, were found on the suspect’s clothes.
The vehicle seats were partly leather, with cloth inserts containing two types of poly-
ester fibres (50:50). Fibres found matching these were found on the victim’s clothes, as
follows.
A pill of purple cotton fibres, matching those in the suspect’s underpants, was found
on the inside of the victim’s panties.
The evidence very strongly supports the probability that there was contact between the
victim and the suspect and between the victim and the suspect’s vehicle. It supports the
probability that violence was used. Its strength lies in the following factors.
There is cross-transfer between the suspect and victim and between the victim and the
vehicle seat. Four fibre types are involved, two of which can be shown by existing data to
occur infrequently among the general fibre population. The probability of these fibre types
being present by coincidence is very low. The blue acetate fibres are a very rarely encoun-
tered type of synthetic fibre, and purple is not a common colour for cotton. The location of
the pill of purple cotton fibres is incriminating. The piece of material found in the suspect’s
car fits almost exactly (apart from some missing threads) into damaged area of the victim’s
shirt. The lack of polyamide fibres transferred from the shirt/hot pants could be attributed
to differential shedding, which was verified by shedding tests.
10.1.9.6 Example 7
During the course of a party, the suspect was alleged to have removed the clothing of a six-
year-old child in the bathroom and indecently assaulted her. He was wearing casual jeans
made from black cotton and a wool pullover, mainly dark blue, but also containing small
areas of red, green and pale blue wool fibres. The child was wearing pyjamas and a pair of
white cotton underpants. The pyjamas did not contain suitable target fibres, being made of
very pale green (microscopically colourless) cotton.
Recovered from the pyjamas were 11 dark blue and two red fibres matching those in the
pullover, and 14 grey cottons matching those in the suspect’s jeans. Two matching dark blue
wools and seven grey cottons were recovered from the underpants.
The evidential value of the findings in this example can be considered to be low, as in
his defence the suspect said that the child had sat on his knee earlier in the evening, thus
explaining the presence of the matching fibres on the pyjamas. The two dark blue and two
red wools on the underpants could be the result of a secondary transfer (after the previous
legitimate contact). The grey cotton fibres were dyed with sulphur black – a type which is
widely found in T-shirts and black jeans, which are very difficult to discriminate and which
are present on many articles of underclothes (Grieve et al., 1988; Grieve and Dunlop, 1992).
Some kinds of natural fibres, like cotton and wool, may have a lower evidential value than
synthetic fibres, due to their relative lack of comparative characteristics.
It is easy to offer alternative explanations for the findings in this case.
10.1.10 Conclusions
In the prevailing climate in forensic science laboratories, fibre examinations, in common
with other trace evidence, are going through a testing time. The key to survival lies as
much in being able to speed up the examinations without sacrificing quality as in being
able to produce useful analytical results which, when interpreted, have a high level of evi-
dential significance that contributes to the effectiveness of the criminal justice system. The
interpretation must be meaningful enough to capture the imagination of the police, to
make them appreciate the usefulness of fibres evidence: to convince them that the effort of
collecting and submitting fibres evidence to the laboratory has been worthwhile.
The body of information relating to fibre populations and fibre frequencies is quite solid
and ever expanding. The same can be said about fibre transfer and persistence characteristics.
This body of knowledge clearly shows that the chances of a particular man-made fibre type
occurring in a random fibre population can run into one in thousands, or even hundreds
of thousands. The possibilities for increasing these numbers have not yet been exhausted,
and even then, the numbers will err in favour of the accused. For the first time, the distinct
characteristics of synthetic fibre types can be emphasized and illustrated by numerical data.
This basic knowledge can be applied in assessments of evidential value. Interest and
work in the use of a probabilistic framework based on likelihood ratios have grown sig-
nificantly in the last 20 years, up to the point that they are now part of the recommended
framework for formulating evaluative reports by the European Network of Forensic
Science Institutes (ENFSI, 2015). Over that time, it has been demonstrated that the abili-
ties of mathematicians and statisticians can be effectively combined with the practical
casework experience of the fibres analyst to improve the reliability and accuracy of fibre
evidence. In the authors’ opinion, being required to consider a wide range of circum-
stances and possibilities within the framework of a particular case, in order to apply a
likelihood ratio approach, can only help to emphasize impartiality on the part of the
forensic scientist. This gives great cause for optimism in the future, and should encour-
age fibre examiners that their subjective opinions about the value of fibres as evidence are
actually exceeded by reality.
References
Adolf, F.-P., 1993. Discrimination of Polyester Fibres by Thickness, Delusterant and Fluorescence,
Internal BKA Research Report, Wiesbaden.
Aitken, C. G. G. and Taroni, F., 2004. Statistics and the Evaluation of Evidence for Forensic Scientists
(2nd ed.). Chichester, UK: John Wiley & Sons Ltd.
Aitken, C. G. G., 1995. Statistics and the Evaluation of Evidence for Forensic Scientists (pp. 107–135).
Chichester, UK: John Wiley.
Akulova, V., Vasiliauskiene, D., and Talaliene D., 2002. Further insights into the persistence of trans-
ferred fibres on outdoor clothes, Sci. Justice, 42(3), 165–171.
Almazrooei, M., Hemmings, J., Robertson, J., Spindler, X., Conroy-Southey, D., de la Hunty, M.,
Maynard, P., and Roux, C., 2012. The evidential value of finding fibres on human hands. In
Proceedings of 20th ENFSI European Textile & Hair Group, Vienna.
Biermann, T. W., 2007. Title blocks of colour IV: The evidential value of blue and red cotton fibres,
Sci. Justice, 47(2), 68–87.
Biermann, T. W. and Grieve, M. C., 1996. A computerised database of mail order garments: A con-
tribution towards estimating the frequency of fibre types found in clothing. Part 1: The system
and its operation, Forensic Sci. Int., 77, 65–73.
tribution towards estimating the frequency of fibre types found in clothing. Part 3: The content
of the data bank—Is it representative? Forensic Sci. Int., 95, 117–131.
Bruschweiler, W., 1993. Moglichkeiten und Grenzen der Kriminaltechnik—Mikrospuren/textil-
fasern, Kriminalistik, 10/93, 647–652.
Bruschweiler, W. and Rey, P., 1997. Textilfasern und Haare als Mittel des Sachbeweises, Kriminalistik,
4/97, 265–271.
Cantrell, S., Roux, C., Maynard, P., and Robertson J., 2001. A textile fibre survey as an aid to the inter-
pretation of fibre evidence in the Sydney region, Forensic Sci. Int., 123, 48–53.
Cassista, A., and Peters, A. D., 1997. Survey of red, green and blue cotton fibres, Can Soc Forens. Sci.
J., 30, 225–231.
Chewning, D. D., Deaver, K. L., and Christensen, A. M., 2008. The persistence of fibers on Ski Masks
during transit and processing, Forensic Sci. Communic., 10(3); July.
Cook, R., Evett, I. W., Jackson, G., Jones, P. J., and Lambert, J. A., 1998a. A model for case assessment
and interpretation, Sci. Justice, 38, 151–156.
Cook, R., Evett, I. W., Jackson, G., Jones, P. J., and Lambert, J. A., 1998b. A hierarchy of propositions:
Deciding which level to address in casework, Sci. Justice, 38, 231–240.
Coxon, A., Grieve, M. C., and Dunlop, J., 1992. A method for assessing the fibre shedding potentialof
fabrics, J Forens Sci. Soc., 32, 151–158.
Coyle, T., Fairchild, J., Fielden, C., and Revell, D., 2007. Raman microspectroscopy and its place in
forensic fibre examination—The identification of man-made cellulosic fibres, Global Forensic
Science Today, 3.
Cresswell, S. L., Cunningham, D., and Nic Daeid N., 2003. Textile survey of cinema seats in Glasgow,
Forensic Sci. Int., 136(1), 117.
Crocker, J., 1986. Micrometric determination of fibre cross section modification ratios. In Proceedings
of the 33rd Meeting of the Canadian Forensic Science Society, Niagara Falls, Canada.
Deadman, H., 1984.Fibre evidence and the Wayne Williams Trial, Part 1, FBI Law Enforce. Bull.,
March, 13–20.
Dignan, S. J. and Murphy, K. J., 2002. Fibre evidence from fingernail clippings, Can. Soc. Forensic Sci.
J., 35(1), 17–21.
Dror, I. E., 2011. The paradox of human expertise: Why experts get it wrong, Paradoxical Brain,
177–188.
Dror, I. E., 2012. Cognitive forensics and experimental research about bias in forensic casework, Sci.
Justice, 52(2), 128–130.
ENFSI, ENFSI Guideline for Evaluative Reporting in Forensic Science, 9 March 2015, http://enfsi.eu/
sites/default/files/documents/external_publications/m1_guideline.pdf, last accessed on 16
May 2016.
Evett, I. W., 1990. The theory of interpreting scientific transfer vidence. In A. Maehly and R. Williams
(Eds), Forensic Science Progress (Vol. 4, pp. 141–180). Berlin, Germany: Springer Verlag.
Gaudette, B., 1988. The forensic aspects of textile fibre examination. In R. Saferstein, R (Ed.), Forensic
Science Handbook (Vol. 2, pp. 209–272). Englewood Cliffs, NJ: Prentice Hall.
Gianelli, P. C. and Imwinkelried, E. J., 1986. Scientific Evidence (pp. 1040–1060). Charlottesville, VA:
The Mitchie Co.
Grieve, M., 1990. Fibres and their examination in forensic science. In A. Maehly, and R. Williams
(Eds.), Forensic Science Progress (Vol. 4, pp. 44–125). Berlin, Germany: Springer Verlag.
Grieve, M., 1994. Fibers and forensic science—New ideas, developments and techniques, Forens Sci
Rev., 6, 59–80.
Grieve, M., 1995. Another look at the classification of acrylic fibres, using FTIR microscopy, Sci
Justice, 35, 179–191.
Grieve, M., 1998. A survey on the evidential value of fibres and on the interpretation of findings in
fibre transfer cases, Proceedings 6th European Fibres Group Meeting, Dundee.
Grieve, M. C., 2000. A survey on the evidential value of fibres and the interpretation of the findings
in fibre transfer cases. Part 2—Interpretation and reporting, Sci. Justice, 40, 201–209.
Grieve, M. C. and Biermann, T., 1997a. The population of coloured textile fibres on outdoor surfaces,
Sci. Justice, 37, 231–239.
Grieve, M. and Biermann, T., 1997b. Wool fibres—Transfer to vinyl and leather vehicle seats and-
some observations on their secondary transfer, Sci. Justice, 37, 31–39.
Grieve, M. C., and Biermann, T. W., 2003. The individuality of blue polyester fibers used to provide
forensic evidence. Forensic Sci. Int., 136(1), 121–122.
Grieve, M. C., Biermann, T. W., and Davignon, M., 2001. The evidential value of clack cotton fibres,
Sci. Justice, 41, 245–260.
Grieve, M. C., Biermann, T. W., and Davignon, M., 2003. The occurrence and individuality of orange
and green cotton fibres, Sci Justice, 43(1), 5–22.
Grieve, M., Biermann, T., and Schaub, K., 2005. The individuality of fibres used to provide forensic
evidence—Not all blue polyesters are the same, Sci. Justice, 45(1), 13–28.
Grieve, M., Biermann, T., and Schaub, K., 2006. The use of indigo derivatives to dye denim material,
Sci. Justice, 46(1), 15–24.
Grieve, M. C. and Dunlop, J., 1992. A practical aspect of the Bayesian interpretation of fibre evidence,
J. Forens. Sci. Soc., 32, 169–175.
Grieve, M., Dunlop, J., and Haddock, P. S., 1988. An assessment of the value of blue, red and black
cotton fibres as ‘target’ fibres in forensic science investigations, J. Forens. Sci., 33, 1332–1344.
Grieve, M., Dunlop, J., and Haddock, P. S., 1989. Transfer experiments involving acrylic fibres,
Forensic Sci. Int., 40, 267–277.
Grieve, M., Dunlop, J., and Haddock, P. S., 1990. An investigation of known blue, red and black dyes
used in the colouration of cotton fibres, J. Forens. Sci., 35, 301–315.
Hartshorne, A. and Laing, D. K., 1991. Microspectrofluorimetry of fluorescent dyes and brighteners
on single textile fibres. Part 1—Fluorescence emission spectra, Forensic Sci. Int., 51, 203–220.
Hellwig, J., 1997. The effect of textile construction on the shedding capacity of knitwear, Proceedings
of the 5th Meeting of the European Fibres Group (102–105), Berlin.
Hellwig J., 2011. The effect of textile construction on the shedding rate of knitwear. In Proceedings of
19th ENFSI European Textile & Hair Group, Riga.
Hong, S., H A., Kim S., Son D. and Min H., 2014. Transfer of fibres on the hands of living subjects
and their persistence during hand washing, Sci. Justice, 54(6), 451–458.
Jackson, G. and Lowrie C., 1987. Secondary transfer of fibres. In Proceedings of the 11th International
Association of Forensic Sciences Meeting, Vancouver.
Krauß, W. and Hildebrand, U., 1995. Fibre persistence on garments under open-air conditions. In
Proceedings of the 3rd European Fibres Group Meeting (pp. 32–36), Linkoping.
Krauß, W. and Doderer, U., 1997. Die Verweildauer von Textilfasern auf haut in Freien. (Persistence
of fibres on skin under open-air conditions). In Proceedings of the 9th German Forensic Textile
Fibres Symposium (pp. 102–105), Fulda.
Lazic, J., Caron, N., and Dufros, Y., 2012. The population of textile fibres in public places. In
Proceedings of 20th ENFSI European Textile & Hair Group, Vienna.
Lennard, C., 2013. Principles of quality assurance. In Jay A. Siegel, Pekka J. Saukko and Max M.
Houck (Eds.), Encyclopedia of Forensic Sciences (pp. 509–514). Waltham, MA: Academic Press.
Lepot, L. and Vanden Driessche, T., 2015. Fibre persistence on immersed garment—Influence of
water flow and stay in running water, Sci. Justice, 55(6), 431–436.
Lowrie, C. and Jackson, G., 1991. Recovery of transferred fibres, Forensic Sci. Int., 50, 111–119.
Lowrie, C. and Jackson, G., 1994.Secondary transfer of fibres, Forensic Sci. Int., 64, 73–82.
Marname, R., Elliot, D., and Coulson, S., 2006. A pilot study to determine the background popula-
tion of foreign fibre groups on a cotton/polyester T-shirt, Sci. Justice, 46(4), 215–220.
Massonnet, G. and Buzzini, P., 2008. Discrimination of coloured acrylic, cotton and wool fibres using
Raman spectroscopy. In Proceedings of the 16th European Fibres Group Meeting, Budapest,
Hungary: June 2008.
Mathieson, F. and Elliott, D., 1994. Direct and indirect transfer of wool fibres to underclothing and
their subsequent persistence. In Proceedings of the 12th ANZFSS Symposium, Auckland.
Merriam-Webster’s Online Dictionary (2016), http://www.merriam-webster.com/, last accessed on
9 May 2016.
Mitchell, E. and Holland, D., 1979a. An unusual case of identification of transferred fibres, J Forens
Sci. Soc., 19, 23–26.
Pailthorpe, M., 1990. Recent developments in the colouration of fibres encountered in forensic
examinations. In Proceedings of the 12th International Association of Forensic Sciences Meeting,
Adelaide.
Palenik, S. and Fitzsimmons, C., 1990a. Fiber cross sections—Part 1, Microscope, 38, 187–195.
Palenik, S. and Fitzsimmons, C., 1990b. Fiber cross sections—Part 2, A simple method for the
sectioning of single fibres, Microscope, 38, 313–320.
Palmer, R., 1997. The retention and recovery of transferred fibres following the washing of recipient
garments. In Proceedings of the 5th Meeting of the European Fibres Group (pp. 60–63), Berlin.
Palmer, R. and Banks, M., 2005. The secondary transfer of fibres from head hair, Sci. Justice, 45(3),
123–128.
Palmer, R. and Burch, H. J., 2009. The population, transfer and persistence of fibres on the skin of
living subjects, Sci. Justice, 49(4), 259–264.
Palmer, R., Hutchinson, W., and Fryer, V, 2009. The discrimination of (Non-Denim) Blue Cotton,
Sci. Justice, 49(1), 12–18.
Palmer, R. and Oliver, S., 2004. The population of coloured fibres in human head hair, Sci. Justice,
44(2), 83–88.
Palmer, R. and Polwarth, G., 2011. The persistence of fibres on skin in an outdoor deposition crime
scene scenario, Sci. Justice, 51(4), 187–189.
Parybyk, A. and Lokan, R., 1986. A study of the numerical distribution of fibres transferred from
blended fabrics. J. Forens. Sci. Soc., 26, 61–68.
Pounds, C., 1975. The recovery of fibres from the surface of clothing for forensic science examina-
tions, J. Forens. Sci. Soc., 15, 127–132.
Pounds, C. A. and Smalldon, K. W., 1975. The transfer of fibres between clothing materials during
simulated contacts and their persistence during wear. Part 2—Fibre persistence, J. Forens. Sci.
Soc., 15, 29–37.
Robertson, J., Kidd, C. B. M., and Parkinson, H. M. P., 1982. The persistence of textile fibres trans-
ferred during simulated contacts, J. Forens. Sci. Soc., 22, 353–360.
Robertson, J. and Lloyd, A., 1984. Observations on the redistribution of textile fibres, J. Forens. Sci.
Soc., 24, 3–7.
Robertson, J. and Olaniyan, D., 1986. The effect of garment cleaning on the recovery and distribution
of transferred fibres, J. Forens. Sci., 31, 73–78.
Robson, R., 1997. Spectral variation within red cotton dyes. In Proceedings of the 5th Meeting of the
European Fibres Group (66–75), Berlin.
Roux, C., Huttunen, J., Rampling, K., and Robertson, J., 2001. Factors affecting the potential
for fibre contamination in purpose-designed forensic search rooms, Sci. Justice, 41(3),
135–144.
Roux, C. and Margot, P., 1997. The population of textile fibres on car seats, Sci. Justice, 37, 25–31.
Roux, C. and Robertson, J., 2013a. Interpretation of fiber evidence. In Jay A. Siegel, Pekka J, Saukko
and Max M. Houck (Eds.), Encyclopedia of Forensic Sciences (pp. 155–160). Waltham, MA:
Academic Press.
Roux, C. and Robertson, J., 2013b. Transfer. In Jay A. Siegel, Pekka J. Saukko and Max M. Houck
(Eds.), Encyclopedia of Forensic Sciences (pp. 113–116). Waltham, MA: Academic Press.
Roux, C., Robertson, J., and Palmer, R., 2013. Persistence and recovery. In Jay A. Siegel, Pekka J.
Saukko and Max M. Houck (Eds.), Encyclopedia of Forensic Sciences (117–123). Waltham, MA:
Academic Press.
Roux, C., Talbot-Wright, B., Robertson, J., Crispino, F., and Ribaux, O., 2015. The end of the (foren-
sic science) world as we know it? The example of trace evidence, Philos. Trans. B, 370(1674),
doi:10.1098/rstb.2014.0260.
Rudram, D. A., 1996. Interpretation of scientific evidence, Sci. Justice, 36, 133–138.
Salter, M., Cook, R., and Jackson, A., 1987. Differential shedding from blended fabrics. Forens. Sci.
Int., 33, 155–164.
Starrs, J., 1990. But query: Forensic scientist? Plenary lecture. In Proceedings of the 12th International
Association of Forensic Sciences meeting, Adelaide.
Stoney, D. A., 1991. Transfer evidence. In Aitken, C. G. G. and Stoney, D. A. (Eds.), The Use of
Statistics in Forensic Science (pp. 107–138). Englewood Cliffs, NJ: Prentice Hall.
Stoney, D. A., 1994. Relaxation of the assumption of relevance and an application to one-trace and
two-trace problems, J. Forens. Sci. Soc., 34, 17–21.
Stoney, D. A., Bowen, A.M., and Stoney, P.L., 2015. Utilization of environmentally acquired very
small particles as a means of association, Forensic Sci. Int., 254, 26–50.
Stoney, D. A., Neumann, C., Mooney, K. E., Wyatt, J. M., and Stoney, P. L., 2015. Exploitation of very
small particles to enhance the probative value of carpet fibers, Forensic Sci. Int., 252, 52–68.
Szewcow, R., Robertson, and J., Roux, C., 2011. The influence of front-loading and top-loading wash-
ing machines on the persistence, redistribution and secondary transfer of textile fibres during
laundering, Aust. J. Forensic Sci., 43(4), 263–273.
Todd, N., 1991. The automation of basic dye HPLC analysis and the investigation of tiger-tail fibres.
MSc thesis, University of Strathclyde, Scotland.
Was-Gubala, J., 2004. Comparative population studies of fibres secured in Poland, Czech Republic
and Germany, Prob. Forensic Sci. LX 58–77.
Watt, R, Roux C., and Robertson, J., 2005. The population of textile fibres in domestic washing
machines, Sci. Justice, 45(2), 75–83.
Wiggins, K., 1994. Reactive dyes on cotton fibres. In Proceedings of the 2nd European Fibres Group
Meeting (7–9), Wiesbaden.
Willis, S., 2014. Accreditation—Straight belt or life jacket?, Sci. Justice, 54(6), 505–507.
10.2 Aids to Interpretation
CLAUDE ROUX
KENNETH G WIGGINS
10.2.1 Introduction
As discussed in other parts of this chapter, forensic scientists have a duty to help the
court by explaining the significance of their findings within the context of the case.
Section 10.1 discusses the various factors influencing the interpretation of fibre evidence
along with guidance with respect to quality assurance and report writing. Section 10.3
describes the use of a probabilistic framework based on likelihood ratios to assist with
this crucial task. This part presents data and other aids available to the forensic fibre
examiner to assist in interpreting the significance of fibre findings. This supporting body
of information has become increasingly important over the years because, in most juris-
dictions, providing expert opinion solely on the basis of personal experience is generally
no longer accepted.
When evaluating fibres evidence for the Court, the forensic scientist attempts to answer
a variety of questions that can generally be clustered in two fundamental ones:
• What is this fibre (or collective of fibres)?

• What does this finding mean?
Various sources of data and information can assist to address these questions and are
presented below.
10.2.2 Fibre Reference Collections

The first stage in any fibres case is to identify the materials involved, by definitive testing.
The ‘I’ve seen it before, therefore it must be acrylic’ approach, which many of forensic scien-
tists have been guilty of using at some time, usually achieves the correct result but carries
risks. For example, anyone can be caught out by an atypical example within a generic class
or a new type of fibre not seen before. A trilobal polypropylene seen for the first time can
be mistaken for nylon until infrared spectroscopy provides a different answer. It is easy
to jump to the wrong conclusion based on previous experience, i.e. a man-made fibre, tri-
lobal, with medium-range birefringence showing bright interference colours on the polar-
izing microscope is most likely to be nylon.
Therefore, all laboratories, regardless of experience, need a reference collection

as part of a systematic approach to fibre identification and training. Such collections
also need to be kept up to date. A comprehensive collection of samples with known
provenance is generally a prerequisite for accreditation, but the maintenance of such
a collection is both time-consuming and expensive. As accreditation to international
standards has become common practice throughout the world, there may be a need to
address such costs through, for example, collaborative efforts both within and between
countries.
One solution may be for only one or two laboratories in each country to hold a full
accredited collection. Other laboratories would have a reduced core collection taken from
this, containing only the fibre types commonly encountered in casework, but would have
access to the larger national collection when necessary. Obviously, this would be possible
for large organizations or the network of national and state forensic science laboratories,
but this approach may not work for commercial organizations, which will perceive such
information as a valuable resource that should not be given away to others competing for
the same work. The establishment of centralized reference collections also has the benefit
of reducing the demands made on manufacturers to provide samples.
Most collections start as ad hoc ones in response to the needs of particular proj-
ects. Historically, examples include the work of Culliford (1963), who produced a mul-
tiple entry card index system for the identification of man-made fibres, and Smalldon
(1973), whose collection of acrylic fibres led to improved discrimination by combining
physical characteristics with analytical data. The samples obtained by Culliford formed
the basis for the Metropolitan Police Forensic Science Laboratory (later Metropolitan
Laboratory, FSS) fibre collection. This has been irregularly added to over many years
to contain several thousand of fibres but has virtually no information regarding the
provenance of the samples. The fibre section of the Forensic Science Institute at the
Bundeskriminalamt (BKA), Wiesbaden also has an extensive collection. Sample collec-
tion was also required for various other projects designed to improve methodology for
the identification and discrimination within generic fibre classes (Grieve and Kotowski,
1977; Grieve and Cabiness, 1985; Grieve et al., 1988; Grieve, 1995, 1996; Clayson and
Wiggins, 1997). Collaborative Testing Services in Herndon, Virginia in conjunction
with the U.S. National Bureau of Standards marketed a collection of man-made fibres
between 1983 and 1987, but it contained only fibres manufactured in the United States,
and was not subsequently updated. The Forensic Fiber Reference Collection marketed
by Microtrace (Elgin, USA) and the McCrone Fiber Reference Set (Westmont, USA)
are both commonly used for training and casework. The former is composed of some
200 traceable, authenticated fibres, including synthetic fibres across 15 different generic
classes and represents a large variety of commonly encountered features including:
bicomponent fibres, hollow fibres, dyed fibres, pigmented fibres and those treated with
optical brighteners or flame retardants. The latter contains 100 prepared slides of natu-
ral and man-made fibres.
The volume of data in any reference collection makes it an obvious candidate for com-
puterization. Many such databases probably exist, but little has been published. Carroll et al.
(1988) used commercially available software to construct a collection based on reference
samples from the Collaborative Testing Services/U.S. National Bureau of Standards and case-
work materials. Tungol et al. (1990) created an FTIR spectral database containing 53 fibre
types, which was reliably used to identify all generic types and chemical subclasses with the
exception of the nylon sub classes. This collection was subsequently expanded to 86 records
stored in FBI’s version of the OMNIC library search software (Thermo Fisher Scientific Inc.).
The FBI Lab – Forensic Automobile Carpet Database is a collection of about 800 samples of
known automobile carpet fibres. Records include microscopic characteristics including cross-
section, diameter, colour as seen through microscope, FTIR microscopy and microspectrom-
etry. This is a centralized searchable database that is used to provide investigative information
on the make, model and year of a car. It is also worth noting that several companies produce
commercial fibre spectral databases, e.g. Thermo Scientific Nicolet and Sadtler.
10.2.3 Data Collections on Fibre Frequency
As explained in Section 10.1, in general, the more common the fibres, the less is the value of
the evidence. Conversely, very rare fibres will normally have strong evidential value even if
present in low numbers. An estimate of the frequency of occurrence of all fibre types, their
colour and their transfer and retention potentials would be an invaluable tool for the fibre
analyst, but is perhaps an ideal. Inevitably, no collection can contain all the data, but an
attempt can be made to obtain a representative sample. In the past the protocols for collection
have varied and produced different data sets. For example, one method used by fibres labora-
tories is to collect samples of garments which have yielded evidentially significant fibres. This
can be useful in assessing how common, for example, dark blue acrylic or red nylon is, but
will not accurately represent the total fibre population, which will contain a high proportion
of fibres that are colourless or do not transfer and will therefore not be treated as target fibres.
The first attempt to obtain information on the whole garment population was under-
taken by Textile Market Studies (TMS – an independent market research company) in
1981 for the UK Home Office forensic science laboratories. Members of the public were
interviewed about their clothing and garments were classified by type, wearer, fibre content
(indicated by the label) and a subjective assessment of the colour. This had the advantage of
quickly yielding a large amount of data, but suffered in not having any information about
fibre morphology or any objective assessment of colour.
The first comprehensive casework data collection was undertaken by Home and Dudley
(1980) who collected 10,034 fibres from 3,836 garments examined in eight UK forensic sci-
ence laboratories in two, two-month periods. The samples were classified by garment type
and fibre type and colour-coded using the Methuen Handbook of Colour (Kornerup and
Wanscher, 1967). Analysis of the data allowed quick estimations of the frequency of fibre
types and colours in particular garment types and also within the whole collection. The
second figure is perhaps most important, as during interpretation the scientist must con-
sider the possibility that the fibres have an alternative source to the garment in the case.
The usefulness of this data encouraged the forensic science laboratories in the UK to
embark on a much more ambitious project, and the collection protocol was improved by
Laing et al. (1987). This data collection contained 19,959 fibres from 7,367 garments col-
lected between 1982 and 1990 from 10 laboratories, and included information on micro-
scopic features, chemical subclass and mean complementary chromaticity coordinates.
The data were available for interrogation by all Forensic Science Service (FSS) scientists and
were widely used. As no new data were added after 1990, there was some anxiety that the
information was becoming ‘out of date’ and scientists were reluctant to use it. Therefore in
1995 a new small collection of 2000 fibres was compared with the original. The two sets of
data showed approximately similar distributions of garments, fibre types and colour, and
the authors of the draft report on the comparison consider that this revalidates the original
collection for continued use (Adams et al., personal communication, 1998).
A pilot project described by Jenne (1983) proposed a similar casework collection in
Germany, but it was never fully established due to the resources required and concerns
about exactly how representative the data would be in relation to the total fibre popula-
tion. Some further information was presented by Ritter (1997). Casework data collections
have been criticized as being unrepresentative because they are only a small sample of the
garment population and only come from materials submitted to forensic science laborato-
ries. Although not perfect, these data remain a useful aid to interpretation provided their
limitations are acknowledged.
Biermann and Grieve (1996a,b, 1998) have described a cheaper and innovative
approach using German mail-order catalogues as sources of fibre data. Information
is stored on garment type (divided into over 90 categories), fibre type and colour. This
approach is analogous to the original TMS survey described earlier. As colour is assessed
only on a subjective basis, and does not involve any measurement, a large amount of data
can be accumulated very quickly – 20,786 records were obtained in nine months and the
total number of records now stands at over 90,000. Comparison of the catalogue database
(CDB) with the FSS collection and world fibre production figures showed some differences,
but there was general agreement on the order of common fibre types and colours when
measured as a percentage of the total data. The CDB has some major advantages, mainly
that it is quick to accumulate data and therefore it is always up to date. It includes all types
of clothing in the general population and, as far as it is possible to check, has been found to
be representative, at least within Germany. Its principal disadvantages are that it does not
contain morphological information, the colour information is very basic and, in common
with other data collections, no way has yet been found to incorporate textile production
numbers. Morphological data, as mentioned above, tend to remain constant over a long
period, opening the possibility of combining existing data with the CDB. It would seem
possible to expand the concept of the CDB to online shopping.
Examples of the type of information which may be obtained from databases include:
• The frequency of morphological characters within a fibre type

• The frequency of polymer composition within a fibre type
• The frequency of uncommon fibre types in the general fibre population
• The frequency of usage of fibre types in different textiles
• The frequency of certain fibre type combinations in different textiles
Despite the different approaches, all types of frequency database are useful in
interpretation and they are complementary, each having different strengths and weaknesses.
Over the years, a number of organisations and laboratories have continued to develop
their own databases. In recent years, a number of research projects capitalised on the
combined advances in analytical chemistry and computing including chemometrics to
develop promising new relational databases often with web-based searching capabili-
ties. The aim of this body of work is to provide increased quantitative data to support
fibre discrimination and hence help interpretation. For example, in Australia, a database
of microspectrophotometric (MSP) spectra was developed at ChemCentre in Perth, in
response to a Western Australian ‘cold case’ investigation (Powell et al., 2015). Over 4,400
textile fibres were collected during the investigation, and no known donor garment was
identified. Currently, over 12,300 normalised and first derivative MSP spectra of casework,
validation and reference textile fibres are stored in the database, including approximately
900 samples from motor vehicle interior fabrics. The database allows the user to perform
objective MSP spectral comparisons by calculating modified Pearson correlation coeffi-
cients, and to visually compare images of fibres acquired via microscopy. As part of this
work, an interpretation approach for large cases was devised, which involves identifying
critical fibres in the casework data set (those exhibiting the largest number of close MSP
spectral matches) and using these critical fibres to develop preliminary fibre groups with
corresponding or similar MSP spectra. These preliminary groups were then confirmed via
bright-field and fluorescence comparison microscopy. The database has also successfully
been utilised for routine fibres casework and a proficiency trial.
Two additional examples are the collaborative work in the United States between the
University of South Carolina, Suffolk University in Boston and Indiana University–Purdue
University Indianapolis on UV/visible microspectrophotometry and FTIR (Morgan
et al., 2014) and the research at North Carolina State University building a Comparative
Finished Fiber Analytical Database (COMFFAD) based on dye analysis by liquid chroma-
tography and time-of-flight mass spectrometry (Beck et al., 2014). With rapid technology
developments, it is expected such databases will continue to grow in the future. The key
factor to confirm these early promises will reside in how well such databases will address
real casework questions in the future.
10.2.4 Target Fibre Studies

A simple defence to the presence of incriminating fibres evidence is to suggest that the
fibres on a suspect were present simply by chance and originated from another textile com-
posed of fibres coincidentally ‘matching,’ for example, those in the victim’s jumper. This
hypothesis was first tested by Cook and Wilson (1986), who searched 335 garments for
four fibre types from garments known from manufacturers’ data to be very common. Only
12 matching fibres were found on 10 garments, with a maximum of two fibres on any one
garment. A similar study by Jackson and Cook (1986) searched 108 front car seats for two
common fibres. A total of 45 matching fibres were found, with a maximum of 13 fibres on
any one seat and 20 in any car. In only two cases were there enough fibres (20 and 7 red
wools) to suggest primary contact, and in both cases the likely source of these fibres was
found by questioning the car owners. The higher incidence of matching fibres was attrib-
uted to longer persistence of fibres on car seats compared to clothing. In a follow-up to the
original publication, Cook et al. (1993) included a high-fashion colour and car seat fabric in
a new study of 100 garments. A total of 67 ‘matching’ fibres (62 blue wools) were found on
27 garments, with a maximum of 11 blue wools on any one item. Only four fibres ‘match-
ing’ the fashion colour were found, plus one blue cotton and no fibres ‘matching’ the car
seat fabric. Despite finding more blue wools, this study showed that finding coincidental
matches is still an uncommon event, with 73/100 garments showing no matching fibres.
Other studies have searched cinema and car seats (Palmer and Chinherende, 1996), cloth-
ing (Brüschweiler and Grieve, 1997; Wiggins et al., 2004; Jones and Coyle, 2010; Coyle et al.,
2013), head hair (Cook et al., 1997), seats in public houses (Kelly and Griffin, 1998) and a com-
bination of bus, pub and cinema seats (Palmer et al., 2015). Some details are given in Table 10.1.
It is expected reports on target fibre studies will continue to regularly appear in the literature.
Table 10.1 Summary of Target Fibre Studies

Number of Fibres
Year Researchers Items Searched Target Fibres Found
1986 Cook and Wilson 335 Garments Blue wool – type 1 9
Blue nylon 6.6 0
Blue acrylic 0
Red acrylic 2
Blue wool – type 2 1
1986 Jackson and Cook 108 Car seats Red wool 37
Brown polyester 8
1993 Cook et al. 100 Garments Blue wool 62
Pink cotton 4
Blue cotton 1
Grey polyester 0
1996 Palmer and 67 Cinema seats Red acrylic 14
Chinherende Green cotton 3
66 Car seats Red acrylic 0
Green cotton 6
1997 Brüschweiler and 435 Garments Red acrylic 2
Grieve
1997 Rothe 100 Men’s trousers Yellow-green 0
viscose
1997 Cook et al. 100 Head hair Blue wool 20
samples Grey acrylic – bean 15
Green acrylic 2
Grey acrylic 0
– round
1998 Kelly and Griffin 80 Pub seats Blue wool 9
2004 Wiggins et al. 58 Garments Blue wool 11
Black polyester 0
Grey polyester 1
Blue acrylic 4
2010 Jones and Coyle 100 Garments Black polyester 6
flock
Blue-grey nylon 0
flock
Grey-brown nylon 12
flock
Orange nylon flock 0
Green nylon flock 0
Black nylon flock 0
Grey nylon flock 0
2013 Coyle et al. 100 Garments Fluorescent 0
yellow polyester
(52 samples)
2015 Palmer et al. 30 Bus seats, Black acrylic 0
54 Pub seats and Blue polyester 0
53 Cinema seats
These data suggest that coincidental matches, where the fibres originate from a source
other than the putative one, are unlikely to occur in the majority of cases. The exceptions,
where alternative sources may be responsible, are with low numbers of matching fibres (<5)
or when very common fibres, e.g. blue wool, are involved. These studies have also shown
that the likelihood of two matching fibre types being present by chance is an extremely
rare event. Overall, these studies demonstrate the value of finding a collective of ‘matching’
fibres in casework.
10.2.5 Population Studies

A factor that must also be taken into account in evaluating fibres evidence is the total
population of extraneous fibres present on the surface of a garment from which the eviden-
tially significant ones have been selected. This information is important, as it firstly gives
a further estimation of which fibres are common, and secondly provides background data
which are essential if a likelihood ratio approach is to be used (see Section 10.3).
The first study to attempt to gather such information was undertaken by Fong and
Inami (1986), who examined 763 extraneous fibres from 40 garments. They effectively
performed over 280,000 comparisons in assessing the probability of chance matches
between garments. What they established was that most chance matches were blue fibres
(11 cotton, 1 wool, 1 polyester) together with 1 yellow acrylic, 2 red cotton and 3 near-
colourless polyester. This work also confirmed empirical observations about the common
fibre/colour combinations in this population. In order of magnitude these were: red acrylic,
blue cotton, blue polyester, blue wool, red cotton. The commonest colour was blue, then
red; the commonest fibre was acrylic, then cotton.
Because of the enormous amount of time required to gather this type of information,
relatively few population studies have been completed (see Table 10.2). Grieve and Dunlop
(1992) attempted to simplify the process by excluding colourless fibres, blue denim and
olive cotton and fibres under 1 mm in length in examining the foreign fibres present on
20 items of underclothing. An estimate of the number of foreign fibre groups present was
made, first using only stereomicroscopy and then by removing and mounting all fibres
except the above types and classifying them according to generic type, diameter, shape,
presence or absence of delustrant and (subjective) colour. The commonest fibre types were
cotton and acrylic and the commonest fibre was black cotton (including grey).
Rothe (1997) studied the fibre population found on 100 pairs of men’s trousers in the
early 1990s. Seventy-five pairs came from independent people and 25 came from casework.
Fibres on each garment were collected with adhesive tape from an area approx. 30 × 35 cm.
Recovered fibres (14,236) were categorized according to type, colour, delustrant, cross-sec-
tion, texturing and dyeing. This fibre population was also searched for fibres matching a
yellow-green viscose target fibre. None were found.
Some studies have looked at the population of fibres on car seats (Roux and Margot,
1997) and outdoor surfaces (Grieve and Biermann, 1997). Both have confirmed that the
commonest fibres were colourless, blue and black cotton and showed that man-made fibres
were relatively rare (13% and 14.3%). Grieve and Biermann additionally showed that many
fibres were <0.5 mm long and over 60% were 1 mm or less. Massonnet et al. (1998) exam-
ined the fibre population on white T-shirts under three different conditions and attempted
to use the information gained, in conjunction with that from a simple transfer study, in
the application of Bayes’ theorem to a simple case scenario. Additional fibre population
382
Table 10.2 Summary of Population Studies

Grieve Was-Gubala,
Fong and and Grieve and Roux and A. and Was- Palmer Palmer Marnane
Inami Dunlop Biermann Margot Watt et al. Massonnet Chocho Gubala Cantrell et and Oliver and Burch et al.
Reference (1986) (1992) (1997) (1997) (2005) et al. (1998) (2000)a (2004a,b)a al. (2001) (2004)a (2009)a (2006)
Colour Fibre
colourless Wool 2.9
Animal 1.4 4.8
grey-black Cotton 21.6 18.0 17.3 26.9 24.1 3.1 28.0 33.4 25.3 37.0 10.3
Wool 12.3 2.2 4.0 4.7 2.0 2.0 0.3
Man-made 3.4 12.2 1.8 16.0 5.8 8.4 8.0 2.6
Animal 1.1 4.8
Plant (other 1.1
than
cotton)
Blue Cotton 7.7 6.1 (exc. 27.8 16.4 20.2 14.2 17.2 17.0 29.6 22.0 18.0 26.4
denim) (denim);
10.0 (exc.
denim)
Wool 9.0 1.9 0.1 6.0 0.4
Man-made 8.1 (PAN) 1.1 7.1 6.6 4.6 8.8 4.0 6.5
Plant 4.0
Red Cotton 9.6 3.7 15.6 2.2 4.0 3.1 9.9 5.0 3.0
Wool 4.3 0.04
Man-made 4.7 (PES) 2.9 3.1 0.7
Animal 0.9
green Cotton 0.9 1.4 5.7 3.9 4.0 2.9
Wool 1.0 0.2
Man-made 5.3 2.1 7.5
Table 10.2 (Continued) Summary of Population Studies
Grieve Was-Gubala,
Fong and and Grieve and Roux and A. and Was- Palmer Palmer Marnane
Inami Dunlop Biermann Margot Watt et al. Massonnet Chocho Gubala Cantrell et and Oliver and Burch et al.
Reference (1986) (1992) (1997) (1997) (2005) et al. (1998) (2000)a (2004a,b)a al. (2001) (2004)a (2009)a (2006)
pinkb Cotton 2.1 5.0 2.1
wool 0.03
man-made 0.6
Interpretation of Fibre Evidence
purpleb Cotton 1.8

wool 1.8 0.03
man-made 0.6
brownc Cotton 3.0 3.0 4.0 4.8
wool 1.8 1.0 0.1
man-made 3.0 2.9
orangec Cotton 3.9
wool 0.3
man-made 1.1
cyan Cotton 14.1
wool 0.02
man-made 3.6
yellow Cotton 0.2
wool 1.2 0.03
man-made 0.6
Values provided in % of total fibre population of the considered surface.
a The values in these columns are not all mentioned in the cited work but are derived from tables or figures in this work by Vooijs et al. (2015).
b Pink and purple are combined in Cantrell et al. (2001) and in Palmer and Oliver (2004).
c Orange and brown are combined in Palmer and Oliver (2004).
383
studies include the following surfaces: bus seats (Was-Gubala and Chocho, 2000), cinema
seats (Cantrell et al., 2001), human head hair (Palmer and Oliver, 2004), domestic washing
machines (Watt et al., 2005), cotton/polyester T-shirt (Marnane et al., 2006) and skin of
living subjects (Palmer and Burch, 2009).
Some general conclusions may be drawn from these population studies, as follows:
• The majority of colour/morphology/generic type combinations (approximately

65%) are represented by a single fibre type only.
• When the most discriminating analytical techniques are used, it can be shown that
many colour/morphology/generic type combinations within a random population
are represented by single fibres.
• Even without colour measurement by microspectrophotometry, the chance of one
type of man-made fibre constituting > 1% of a random population is very small.
• Synthetic fibres form only a low percentage (13%–20%) of any population yet
studied.
• Synthetic fibres exhibit a very high degree of polymorphism.
These various data collections tend to confirm the subjective opinion of experienced
fibres experts that even the commonest types of fibre will not regularly be found in large
numbers by chance. Also, these provide an essential component for a more objective
assessment of evidential value, which can do much to enhance the value of this type of
forensic evidence.
10.2.6 Industrial Enquiries

Early examples of reports of industrial enquiries can be found in Mitchell and Holland
(1979b), Deadman (1984a,b) and Wiggins and Allard (1987), followed by Woltmann et al.
(1994), Brüschweiler and Grieve (1997), Deedrick (1996) and Adolf (1996). Since then, case
study reports have regularly emphasised the value of using fibres as an investigative tool
(as opposed to an evidential tool for court purposes only) in which some kind of indus-
trial enquiries had taken place. Recent typical examples include two English cases known
as ‘The Suffolk Strangler’ (Palmer, 2008) and ‘The Jigsaw Murder’ (Palmer, 2011). In both
cases, through industrial enquiries, fibres were crucial to lead the investigative team to
look for and ultimately find the sources of the fibres.
The time involved in tracing companies, and then carrying out additional technical
work, i.e. when continuity cards are obtained, makes it impossible to trace every relevant
garment in every case. Generally speaking, only major crimes, e.g. terrorism, armed
robbery, murder and rape would be the subject of industrial enquiries.
Before starting an enquiry, careful consideration must be given to both the case and
the garment/rope. If there is little chance of tracing the item, or if it is felt that any infor-
mation obtained would add little or nothing to the case findings, then the enquiry should
not be pursued. However, it is incorrect, for example, to dismiss an enquiry involving a
garment bought from a major chain store just because millions or more may have been
produced. There is a limit to the amount of fibre or number of garments that can be dyed
in a dye bath. If differences can be detected between samples from dye batches, the number
of garments that are identical would be limited. Designer clothing can be very useful for
enquiries as its production is limited, but garments/ropes manufactured in the Far East
are usually untraceable beyond the importer, at best, and evidentially this does not usually
help the case. Even bearing all these facts in mind, an enquiry that initially appears to be
straightforward may not turn out to be so. An enquiry carried out at the author’s labora-
tory concerned a pair of men’s wool/polyester trousers sold in a leading chain store. The
wool came from Australia, the polyester from Spain, the fabric manufacturer was in Israel
and the garment manufacturer and retailer in the United Kingdom. Hence the time taken
to complete such an enquiry is greatly increased by the distance and the language problems
involved.
A question that is often asked once an enquiry is complete is ‘who gives the evidence at
court?’ If the scientist is to give it, the problem of hearsay enters the situation. Evidence can
only be given, in normal circumstances, on something done, or known to be fact, by the
expert witness. If production figures for a garment were deemed to be of use in a case, an
employee of the company should be asked to provide a statement. As professional people,
they normally make excellent witnesses and more importantly, are seen to be impartial.
It is the authors’ opinion that these enquiries should always be carried out by a scientist
who is an expert in fibre analysis/examination. An information scientist can be the best
person to carry out initial enquiries, but when questions concerning production and
manufacture need to be asked the fibres expert should take over.
10.2.6.1 Value of Industrial Enquiries

At the fourth meeting of the European Fibres Group, Gillespie (1996) asked the question
‘Why do we need industrial enquiries?’, and decided that there were three reasons, as
follows:
• To attach a level of significance to the scientific findings in casework and assess the
strength of evidence for the courts. Gillespie pointed out that the finding of fibre
transfer evidence is of no value unless you can determine how significant it is. This
may be an overstatement, but when it is considered together with the conclusion in
Wiggins and Cook (1992) – ‘Enquiries will put the evidential value of the fibres into
perspective and enable the scientist to have a clearer understanding of the case’ – we
can begin to see how important these enquiries can be.
• Investigative purposes – to trace manufacturers and suppliers of items where this
may assist an enquiry. Textile items left at the scene of a crime may be traced
to a limited number of outlets, and a suspect may even be identified by a shop
worker as the person who purchased a particular item. Alternatively, as has been
the case at the Metropolitan Laboratory of the Forensic Science Service (ML-FSS),
stolen property seized by the police has been returned to its owner by tracing
the company from garment labels. Since then, this kind of enquiry has become
increasingly used with the advent of security cameras in many metropolitan areas
and shops and subsequent digital ‘traces’ of garments being recorded.
• To provide information on changes in fibre trends. Industrial enquiries are of the
utmost importance when new fibres, e.g. microfibres, appear in literature or in
casework in order to ascertain their production methods and general availability
(Clayson and Wiggins, 1997).
Not all textile material or ropes and cordage lend themselves to enquiries. If a garment
appears in a case without a label, it is virtually impossible to trace. Garments that have
been produced in a country distant to the scientist’s own laboratory and problems caused
by language barriers can also be reasons why some enquiries fail. However, with the forma-
tion and close working relationships of various textile and fibre groups around the world
and intensified international collaboration through regional networks (e.g. International
Forensic Strategic Alliance – IFSA, a multilateral partnership between the regional net-
works of operational forensic laboratories), enquiries can now be pursued in situations
where in the past they would not have been attempted. Indeed, two enquiries that were
initiated in one of the authors’ laboratory were successful only with the help of one scien-
tist working for the Gendarmarie in Paris, France and another employed by the Scientific
Service of the Zurich police in Switzerland.
Naturally, the making of industrial enquiries is highly dependent on building up a net-
work of contacts within the textile trade and rope industry. It is usually noticeable that as
more contacts have been made the enquiries themselves have become more complex. It
may no longer be sufficient just to trace the number of jumpers identical to that worn, for
example, by a suspect in an armed robbery; where possible, it becomes necessary to trace
the fibre back to its manufacturer and to find out how much has been produced and in how
many batches.
10.2.6.2 Methods of Tracing Manufacturers

Enquiries can be split into two categories: (1) ropes and cordage, and (2) clothing or other
textiles.
Ropes and cordage are traced using three sources of information: previous enquiries,
a rope collection consisting of approximately 1,000 samples which can be searched for
matching or similar ropes, and the manufacturers themselves. Wiggins (1995) outlines the
features which can be used to point to a particular manufacturer, and summarizes them in
the form of a check list. This information is also available in Chapter 3 of this book.
When clothing or other textiles need tracing, many more options are available. The
garment label or logo may be recognized by the scientist carrying out the enquiry or by a
colleague. It is sometimes necessary to search databases for company or brand names or
to screen trade directories for relevant information. Other options are the Patent Office or
textile associations. Nowadays, a wealth of information is available on the Internet. A non-
exhaustive list of relevant websites is presented in Section 10.1.3. However, very often the
starting point is a previous enquiry or a contact who has been used on a previous occasion.
If the contact cannot answer a particular question they will recommend another person
who will be able to assist. Dillinger (2008), through a casework example, illustrates the
investigative benefits of examining textile labels on items of clothing in order to obtain
useful information with regard to clothing evidence.
10.2.6.3 Type of Information Required

It is normally the retailer who will be the first point of contact in an enquiry. When
retailers are approached, it is essential that the scientist has all the relevant details available.
Garments often carry more than one label. Apart from the one that leads to the retailer,
others may carry details of country of origin, composition, cleaning instructions, printer’s
reference, manufacturing company, garment style number and date of production. This
information may not be obvious to the scientist, as it is often in a code or simply displayed
as a set of numbers. However, it may be vital to the retailer if he or she is going to give
the correct information. Other details that should be available are an accurate description
of the garment and a record of the fibre composition as identified by the laboratory. It is
important to establish beyond doubt that the retailer and scientist are talking about the
same item, and, to this end, the retailer may need to be shown either the garment or a clear
photograph. Often a small piece of material is sufficient for the retailer to recognize the
colour and weave or knit. It can then be established how many garments, taking account
of all sizes, were purchased, from whom and when. Any unsold garments also need to be
accounted for. Finally, an order or stock number needs to be obtained so that there will be
no misunderstanding when the supplier is approached.
The supplier will usually know whether the garment is unique to a particular retailer
or whether garments have been sold elsewhere. It is important to be aware that some may
have been substandard and sold on as ‘seconds’. Suppliers will also be able to say how many
garments they have supplied and who manufactured them.
If the information above is not available from the supplier, the manufacturer will know
how many were produced and where they went. He or she will know if other retailers were
involved and details of the yarn used in the garment production. It is important to know
how much yarn would be used in the production of the questioned garment, and if the
yarn is used in other garments manufactured by that manufacturer, as well as knowing
where it was purchased. In order to clarify the situation with the spinner it is useful to
know the identification code/number and batch number of the yarn. Continuity cards
may be obtained from the manufacturer. These are small pieces of material from different
batches of garments, and are useful for determining if any batch-to-batch variation is
present (Wiggins et al., 1988). Obviously if garments can be limited to having originated
from a particular batch or limited number of batches, then the evidential value of the case
findings will be increased.
The spinner can establish how much yarn of a particular batch number has been
supplied, and to whom, and whether any is still in stock. Once again continuity cards may
be available; this may also be the case if the dyer is contacted.
The dyer will be aware of any changes in dye recipes, and will be able to say if he or she
provided other spinners with specific batches of dyed yarn.
If all the above information is available, it should be possible to ascertain the total
amount of dyed fibre produced and how many garments would have been manufactured
that are identical to the questioned garment. It is almost inevitable that the amount of
fibre produced will be greater than that traced, because of wastage and loss at all stages of
production.
10.2.6.4 Examples of Industrial Enquiries

10.2.6.4.1 Rope In an English case, a request was made to trace a rope that was of interest
in a criminal investigation. It consisted of a black, polyethylene, plaited outer sheath and a
white polyester core. The diameter was approximately 8 mm.
A search was made of the database and, although no match was found, polyethylene
rope and cordage manufacturers were identified. A company, based on the south coast of
England, recognized the rope as its product. It was produced in 2,000 m lengths, used for
drawing fibre optic cables through pipes and discarded after use. The manufacturer said
that another company produced an identical product apart from the colour – it only made
it in orange. At the time of the offence it had been produced for three years, with by far the
greater part (20,000 m per week) going to one company. The additional amount (50,000 m
in total) went to other companies. Although it was not on sale to the general public, very
large quantities were produced.
Later the same year the orange rope (mentioned above) was the subject of another
enquiry, and within hours it was ascertained that it had been produced at a rate of
4–12,000,000 m per year over four to five years. A comparison of the case sample with that
obtained from the manufacturer confirmed a positive match. Furthermore, two years later
another case involved the same orange rope.
Although both the black and orange ropes were very common, the source was traced
and after the detailed information obtained from the first enquiry the subsequent enquiries
took very little time.
10.2.6.4.2 Twine One of three victims in a linked murder series was found with her
hands tied. The ‘twine’ used was 2.5 mm diameter, tightly twisted brown paper. The man-
ufacturer was quickly identified from the rope collection. Although not rare, the twine
was unusual and one of its main uses was in tying up bundles of clothes after they had
been laundered. A suspect was arrested and a quantity of the twine was discovered at his
mother’s house. Her place of work was a laundry. The suspect was found guilty of two mur-
ders and sentenced to life imprisonment.
10.2.6.4.3 Bank Security Bags A suspect was alleged to have slashed open bank secu-
rity bags to check that they contained money, during a robbery on a security van. When
arrested, the suspect was in possession of a Stanley knife, on the blade of which were found
73 fibres, microscopically similar to the fibres of the security bag. Microspectrophotometry
and thin layer chromatography were carried out on 15 of these fibres, and three were found
to have originated from the bag or an identical one. It was likely from the results, however,
that the bags were composed of fibres from two different dye batches. After submission of
a further six security bags, all 15 fibres were found to correspond in terms of the analytical
parameters.
It was requested that the origin of the bags be traced. The Bullion Centre gave the name
of the only firm with which it dealt. This firm described the bags as consisting of cross-
woven, mauve-dyed nylon, and it was found that all the material had been purchased from
one company and had been dyed exclusively for it. A dye batch would normally be 500 m,
equivalent to 2,000 bags. The dyers admitted that ‘topping up’ (Wiggins et al., 1988) of dye
batches happened, but thought that the same dye had been used for the six to seven years of
dyeing. However, from the continuity cards supplied, it was clear that two different recipes
had been used. As fibres from both dye batches had been found, it was concluded that the
knife in the suspect’s possession had been used to cut open the bank security bags. The
suspect was sentenced to 12 years’ imprisonment.
10.2.6.4.4 Label A small label was left at the scene of a murder. There was no obvious
source at the scene, and the police decided it had been necessary to trace the garment to
which it has been attached. The word on the label was ‘erredieci’.
The Italian trade centre fashion department identified the company as being based
in Milan. It produced high-quality silk ties, bow ties, pocket handkerchiefs, scarves and
dressing gowns. A visit to Italy by the police revealed that the label originated from a very
limited range of ties, and that the thread attached to the label was not the original sew-
ing thread. Two ‘erridieci’ ties were recovered from the suspect’s address; one had a label
missing. Unfortunately, the suspect had legitimate access to the murder scene so it didn’t
incriminate him. However, the police had spent four weeks trying to trace the garment: the
scientist took one afternoon.
10.2.6.4.5 Jacket A navy blue ‘bomber’ style jacket with a yellow polo player motif on
the outer left front was examined as part of a murder investigation. It became important to
establish the number of identical garments manufactured and, if possible, the amount of
blue cotton that had been dyed in an identical way. There were three labels of interest. The
first was inside the back at the neck region and read ‘Polo by Ralph Lauren’. The second was
inside the left seam and said ‘100% cotton, dry clean only, Made in England’. The third was
inside the right pocket and had the number ‘22’ printed in red.
The ‘manufacturer’ was contacted and stated that the jacket must be a copy. On a blue
jacket the motif is normally red, and genuine garments have woven labels complete with
washing instructions. The ‘22’ would not be on a genuine jacket, and the garments are
made in the USA, not England. The enquiry could not be pursued.
10.2.6.4.6 Overcoat It became necessary to trace the number of overcoats sold by a
leading chain store. The coat was only sold for a period of five months. Only 729 coats were
sold in five or six city-centre stores. The fabric and the overcoat itself were produced in two
areas of Italy. The coat was exclusive to the chain store, and although other garments would
have been made from the same fibre they would not have been sold in the UK. Despite the
fact that it was a chain store item, it was still quite unusual in the population of garments
in the UK.
10.2.6.4.7 Carpet The body of a murder victim was recovered from a trunk which
had been thrown into a river. Blue, polypropylene carpet type fibres were recovered
from the victim’s clothes. A suspect was arrested and it was thought that the murder
took place in his house. When the police visited the suspect’s house some of it had been
redecorated and new carpets laid. The suspect kept meticulous records and receipts of
all purchases but it was thought that one, relating to a carpet, was missing. Neighbours
said that a new blue carpet was thrown out on a date which would have coincided with
the murder. The police visited many carpet stores in the Croydon area and eventu-
ally found documents to prove that a blue polypropylene carpet had been laid at the
suspect’s house prior to the murder. When this information was viewed alongside the
suspect’s own records it was clear that the carpet was laid for only a few months before
being replaced. It was therefore important to ascertain the amount of carpet and identi-
cal fibre available at the time of the murder. Figure 10.2 outlines the route of the enquiry
and the connections that were made.
Connections:
1. Fibres recovered from the clothing of the victim were suspected of originating
from a carpet at the suspect’s house.
2. House 2, owned by people unrelated to the crime, had an identical carpet laid by
the same company at a similar time to that in the suspect’s house. The carpet fibres
formed an exact match to the fibres recovered from the victim’s clothing.
3. Invoices at the carpet store linked carpets laid in the suspect’s and house 2 to the
same batch.
4. The Halifax weavers deal with manufacture and distribution of the carpet.
The carpet store ordered a specific blue polypropylene carpet from the weavers.
VICTIM SUSPECTS HOUSE

Blue polypropylene
fibres
HOUSE 2
PATTERN BOOK
FROM CROYDON CARPET
CROYDON CARPET STORE LEWISHAM
STORE CARPET STORE
NEW MALDEN
CARPET STORE
CARPET DELIVERY CRAWLEY

COMPANY CARPET STORE
NEWCASTLE
CARPET STORE
BELGIUM, PRADO BELGIUM, RONSE MAIDSTONE/HINCKLEY ABERDEEN

STOREAGE DEPOT CARPET STORE
N. IRELAND SPINNERS HALIFAX WEAVERS
AUSTRIA DENMARK
ASOTA DANAKLON
Figure 10.2 An outline of the route followed and connections made in a carpet enquiry.
The order went via computer to Ronse, Belgium. The carpet was produced in
Prado, Belgium and cut to size in Ronse before despatch to either Hinckley
or Maidstone for storage. It would be collected by a carpet delivery company
and taken to the store in Croydon. The weavers stated that, in this case, a creel
would produce approximately 20 rolls of carpet, each being approximately 100
square metres. Other identical carpeting went to five stores in Lewisham (South
London), New Malden (Surrey), Crawley (Sussex), Newcastle (Northumberland)
and Aberdeen, Scotland.
5. The spinners in Northern Ireland said that 1440 cones would produce one creel
and the cones were delivered to Belgium for carpet manufacture.
6. The fibre was produced by Asota in Austria. The estimated date of the murder, as
ascertained by the pathologist, was taken into account when estimating the amount
of fibre available. After testing of the samples obtained from various sources, it was
clear that only two creels produced with Asota fibre could have been the source of
the fibres recovered from the victim’s clothing.
7. The spinners had previously used a company in Denmark that produced a similar
fibre under the name Danaklon. This fibre was identical to fibre in a pattern book
obtained from the Croydon carpet store. However, comparative tests revealed the
Danaklon fibre to be different to the fibres recovered from the victim.
Assuming that a room in a house was 12 metres square, then sufficient carpet would
have been produced to carpet a maximum of 333 rooms. However, some carpet would have
been wasted during fitting. This would be the total amount of carpeting in Great Britain
from which the fibres found on the victim could have originated.
Unfortunately, although the police were sure that the suspect’s house was the murder
scene, a number of people lived there. They all had a possible motive for the murder. Nobody
would admit it and they would not incriminate each other. As it could not be proved who
actually carried out the murder, the main suspect was charged only with perverting the
course of justice. He was imprisoned for five years.
10.2.6.4.8 Car Carpet In an Australian murder case, a total of five grey polypro-
pylene, five blue polypropylene, 26 black polyester and four purple wool fibres were
recovered from the right shoe of the deceased, and four grey polypropylene, seven blue
polypropylene and 24 black polyester from the left (Bennett et al., 2010). The grey poly-
propylene, blue polypropylene and black polyester fibres could not be discriminated from
the fibres taken from the carpet of the car owned by the victim’s former boyfriend, a 1991
Honda CRX. No potential source of the purple wool fibres could be found. Enquiries
made to the manufacturer confirmed the composition of this carpet as 60% off-black
(grey) polypropylene, 15% blue polypropylene and 25% black polyester. This particular
carpet was manufactured in Japan and was only used in 1990 and 1991 Honda CRX
vehicles. 296 of these vehicles had been sold in Australia. Affidavits were obtained from
the owners of 291 of these cars, each claiming not to have been in the area where the
deceased was found, from 29 March to 2 April 1995. Only five vehicles were unaccounted
for, one of which was the person of interest’s vehicle. Interestingly, in court, the analyti-
cal results combined with the industry enquiries and the police investigation led both
the prosecution and defence to accept that the combination of fibres recovered from the
shoes of the deceased came from the Honda CRX belonging to the defendant. The criti-
cal questions moved from the source to the activity level; i.e. when the fibres had been
deposited on the shoes and how long these would be retained. Data assisting the court
for these questions were addressed in Roux et al. (1999).
10.2.7 Conclusions
Aids currently available to the forensic fibre examiner to assist in interpreting the
significance of fibre findings are presented above. These include fibre reference collections,
data collections on fibre frequency, target fibre studies, population studies and industrial
enquiries. The value and also the limitations of such aids are discussed.
The value of these aids resides in the fact that they go a long way to address a require-
ment of modern practices in forensic science whereby fibre examiners need to support
their conclusions with reliable and, if possible, research-based data. Whilst it is unrealistic
to believe that all the required data will ever be available in an exhaustive manner for every
single case and set of circumstances, the body of information available to the forensic sci-
entist today is quite significant. Further, it is expected to continue to grow in the future.
Combined with the fibre examiner’s experience and the application of appropriate logical
rules, this information contributes to make forensic fibre examination more effective dur-
ing the investigative stage and more transparent in Court.
Acknowledgement
Some information in the section on industrial enquiries is taken from an article published
by Wiggins and Cook (1992) in Journal of the Forensic Science Society.
References
Adams, J., Coyle, T., Middlemist, A., Conder, A., Christensen, P., and McHugh, C., 1998. Personal
communication, draft FSS report.
Adolf, F.-P., 1996. The results of an industrial enquiry concerning a pair of U.S. Army camou-
flage fatigue trousers. In Proceedings of the 4th European Fibres Group Meeting (63–67),
London.
Beck, K., Griffis, D., Hinks, D., and Zhou, C., 2014. Method Development and Validation of
Comparative Finished Fiber Analysis Using Nano-Sampling Cryomicrotomy and Time-of
Flight Secondary Ion Mass Spectrometry, National Institute of Justice, Document No.: 244566 Date
Received: January 2014.
Bennett, S., Roux, C. P., and Robertson, J., 2010. The significance of fibre transfer and persistence—A
case study, Aust. J. Forensic Sci., 42(3), 221–228.
Biermann, T. W. and Grieve, M. C., 1996a. A computerised database of mail order garments: A con-
Biermann, T. W. and Grieve, M. C., 1996b. A computerised database of mail order garments: A con-
of the data bank and its statistical evaluation, Forensic Sci. Int., 77, 75–91.
of the data bank—Is it representative? Forensic Sci. Int., 95, 117–131.
Brüschweiler, W. and Grieve, M. C., 1997. A study on the random distribution of a red acrylic target
fibre, Sci. Justice, 37, 85–89.
Cantrell, S., Roux, C., Maynard, P., and Robertson J., 2001. A textile fibre survey as an aid to the inter-
pretation of fibre evidence in the Sydney region, Forensic Sci. Int., 123, 48–53.
Carroll, G. R., Lalonde, W. C., Gaudette, B. D., Hawley, S. L., and Hubert, R. S., 1988. A computerized
database for forensic textile fibres, Can. Soc. Forens. Sci. J., 21, 1–10.
Clayson, N. J. and Wiggins, K. G., 1997. Microfibres—A forensic perspective, J. Forens. Sci., 42,
842–845.
Cook, R., Salter, M. T. and O’Connor, A.-M., 1993. The significance of finding extraneous fibres on
clothing. In Proceedings of International Association of Forensic Sciences Meeting, Dusseldorf.
Cook, R., Webb-Salter, M. T., and Marshall, L., 1997. The significance of fibres found in head hair,
Cook, R. and Wilson, C., 1986. The significance of finding extraneous fibres in contact cases, Forensic
Sci. Int., 32, 267–273.
Coyle, T., Shaw, C., and Stevens, L., 2013. The evidential value of fibers used in ‘Hi-Vis’ workwear,
JASTEE, 4(2), 2–16.
Culliford, B. J., 1963. The multiple entry card index for the identification of synthetic fibres, J. Forens.
Sci. Soc., 4, 91–97.
Deadman, H., 1984a. Fibre evidence and the Wayne Williams trial, Part 1, FBI Law Enf. Bull., March,
13–20.
Deadman, H., 1984b. Fibre evidence and the Wayne Williams trial, Part 2, FBI Law Enf. Bull., 53,
10–19.
Deedrick, D., 1996. The search for the source: Automotive fibers in the O. J. Simpson case. In
Proceedings of the 4th European Fibres Group Meeting (pp. 52–60), London.
Dillinger, S., 2008. Investigation approaches based on textile labels, Global Forensic Science Today,
4, 6–7.
Fong, W. and Inami, H., 1986. Results of a study to determine the probability of chance match occur-
rences between fibres known to be from different sources, J. Forens. Sci., 31, 65–72.
Gillespie, C., 1996. Fibre related industrial enquiries. In Proceedings of the 4th European Fibres Group
Meeting (60–62), London.
Grieve, M. C., 1995. Another look at the classification of acrylic fibres, using FTIR microscopy, Sci.
Justice, 35, 179–190.
Grieve, M. C., 1996. New man-made fibres under the microscope—Lyocell fibres and nylon 6 block
co-polymers, Sci. Justice, 36, 71–80.
Grieve, M. C. and Biermann, T., 1997. The population of coloured textile fibres on outdoor surfaces,
Sci. Justice, 37, 231–239.
Grieve, M. C. and Cabiness, L. R., 1985. The recognition and identification of modified acrylic fibres,
Grieve, M. C. and Dunlop, J., 1992. A practical aspect of the Bayesian interpretation of fibre evidence,
J. Forens. Sci. Soc., 32, 169–175.
Grieve, M. C., Dunlop, J., and Kotowski, T. M., 1988. Bicomponent acrylic fibres – their characteriza-
tion in the forensic science laboratory, J. Forens. Sci. Soc., 28, 25–33.
Grieve, M. C. and Kotowski, T. M., 1977. The identification of polyester fibres in forensic science, J.
Forens. Sci., 22, 390–401.
Home, J. M. and Dudley, R. J., 1980. A summary of data obtained from a collection of casework
materials, J. Forens. Sci. Soc., 20, 253–261.
Jackson, G. and Cook, R., 1986. The significance of fibres found on car seats, Forensic Sci. Int., 32,
275–281.
Jenne, K., 1983. Eine neue, rechnergestutzte Moglichkeit zur Absicherung des Beweiswertes textile
Faserspuren, Arch. Krim., 168, 17–25.
Jones, J. and Coyle, T., 2010. Synthetic flock fibres: A population and target fibre study, Sci. Justice,
51(2), 68–71.
Kelly, E. and Griffin, R. M. E., 1998. A target fibre study on seats in public houses, Sci. Justice, 37, 39–44.
Kornerup, A. and Wanscher, J. H., 1967. Methuen Handbook of Colour (2nd ed.), London, UK:
Methuen and Co.
Laing, D. K., Hartshorne, A. W., Cook, R., and Robinson, G., 1987. A fibre data collection for forensic
scientists, collection and examination methods, J. Forens. Sci., 32, 364–369.
Marname, R., Elliot, D., and Coulson, S., 2006. A pilot study to determine the background popula-
tion of foreign fibre groups on a cotton/polyester T-shirt, Sci. Justice, 46(4), 215–220.
Massonnet, G., Schiesser, M., and Champod, C., 1998. Population of textile fibres on white T-shirts,
Proceedings of the 6th European Fibres Group Meeting, Dundee.
Mitchell, E. and Holland, D., 1979. An unusual case of identification of transferred fibres, J Forens
Sci. Soc., 19, 23–26.
Morgan, S. L., Birt, D. L., Bartick, E. G. and Goodpaster, J. V., 2014. Design of an extensible foren-
sic database for textile fibers. In: Evaluation of Statistical Measures for Fiber Comparisons:
Interlaboratory Studies and Forensic Databases, Morgan S.L., National Institute of Justice,
Document No. 248386. Date Received: October 2014, Award Number: 2010-DN-BX-K220,
80-91.
Palmer, R., 2008. The Suffolk serial killings. In Proceedings of the 16th European Fibres Group Meeting,
Budapest, Hungary, June.
Palmer, R., 2011. The ‘jigsaw’ murder, NIJ Trace Evidence Symposium, Kansas City, August.
Palmer, R. and Burch, H. J., 2009. The population, transfer and persistence of fibres on the skin of
living subjects, Sci. Justice, 49(4), 259–264.
Palmer, R., Burnett, W., Luff, N., Wagner, C., Carney, C., Sheridan, K., 2015. The prevalence of two
‘commonly’ encountered synthetic target fibres within a large urban environment, Sci. Justice,
55, 103–106.
Palmer, R. and Chinherende, V., 1996. A target fibre study using cinema and car seats as recipient
items, J. Forens. Sci., 41, 802–803.
Palmer, R. and Oliver, S., 2004. The population of coloured fibres in human head hair, Sci. Justice,
44(2), 83–88.
Powell, R., Priddis, C., Collins, P. and Coumbaros, J., 2015. Development of a microspectrophoto-
metric spectrum database for comparison of casework textile fibre samples to motor vehicle
interior fabrics. In AAFS 67th Annual Scientific Meeting, Orlando, Florida, Feb 16–21.
Ritter, T., 1997. A short report on the computerized data collection at the LKA Stuttgart. In Proceedings
of the 5th European Fibres Group Meeting, Berlin.
Rothe, M., 1997. Examination of foreign fibre populations, Proc. 5th European Fibres Group Meeting
(119–120), Berlin.
Roux, C. and Margot, P., 1997. The population of textile fibres on car seats, Sci. Justice, 37, 25–31.
Roux, C., Langdon, S., Waight, D., and Robertson, J. 1999. The transfer and persistence of automotive
carpet fibres on shoe soles, Sci. Justice, 39(4), 239–251.
Smalldon, K. W., 1973. The identification of acrylic fibres by polymer composition as determined by
infrared spectroscopy and physical characteristics, J. Forens. Sci., 18, 69–81.
Tungol, M. W., Bartick, E. G. and Montasser, A., 1990. The development of a spectral database for the
identification of fibres by infrared microscopy, Appl. Spectrosc., 44, 543–549.
Vooijs, C., Vergeer, P. and van der Weerd J., 2015. Towards source level evaluation of the evidential
value of fibre examinations, Forensic Sci. Int., 250, 57–67.
Was-Gubala, J. and Chocho, A., 2000. A population study of fibres found on bus seats in Cracow. In
Proceedings of the 8th Meeting European Fibres Group Meeting (82–86), Cracow.
Was-Gubala, J., 2004a. Comparative population studies of fibres secured in Poland, Czech Republic
and Germany, Prob. Forensic Sci. LX 58–77.
Watt, R, Roux C., and Robertson, J., 2005. The population of textile fibres in domestic washing
machines, Sci. Justice, 45(2), 75–83.
Wiggins, K. G., 1995. Recognition, identification and comparison of rope and twine, Sci. Justice, 53,
53–58.
Wiggins, K. G. and Allard, J. E., 1987. The evidential value of fabric car seats and car seat covers, J.
Forens. Sci. Soc., 27, 93–101.
Wiggins, K. G. and Cook, R., 1992. The value of industrial enquiries in cases involving textile fibres,
J. Forens. Sci. Soc., 32, 159–167.
Wiggins, K. G., Cook, R. and Turner, Y. J., 1988. Dye batch variation in textile fibres, J. Forens. Sci.,
33, 998–1007.
Wiggins, K., Drummond, P. and Champod, T. H., 2004. A study in relation to the random distribu-
tion of four fibre types on clothing—(incorporating a review of previous target fibre studies),
Sci. Justice, 44(3), 141–148.
Woltmann, A., Deinet, W. and Adolf, F.-P., 1994. Zur Bewertung von Faserspurbefunden mit Hilfe
von Wahrscheinlichkeitsbetrachtungen—ein Fallbeispiel, Arch. Krim., 194, 1–10.
10.3 A Probabilistic Approach to the Evaluation of Fibre Evidence
CHRISTOPHE CHAMPOD
FRANCO TARONI
10.3.1 Introduction
In 2006, Jackson et al. (2006) clarified the various roles a forensic scientist may be asked
to take during the conduct of an investigation (from the crime scene to court). They delin-
eated the contrast between the role of the forensic scientist as an ‘investigator’ and its role
as an ‘evaluator’. In the context of fibre examination, the investigative role can be when
the forensic scientist will guide, based on the examination of fibres, as to the potential
sources of the recovered material (carpets, scarf, synthetic fur, etc.) or the potential con-
tacts involved. That would be done generally without any control source available. The
purpose being here to provide the investigation with the most appropriate investigative
leads. That role is distinct from the situation when control material is made available,
comparative examinations carried out and will potentially lead the forensic scientist to
produce an evaluative statement to be used in a court of law. The latter is the evaluative role
and is the concern of this chapter.
Probabilistic theory has been offered as a coherent model for interpreting and evaluat-
ing forensic evidence (Aitken and Taroni, 2004). In the field of fibre transfer practitioners
commonly refer to an unstructured list of factors that are considered to be relevant for the
assessment of fibre evidence. These factors are (adapted here from Roux and Margot, 1994;
Scientific Working Group on Materials Analysis [SWGMAT], 1999; European Fibre Group
K. Wiggins and M. Grieve [Eds.], 2001, p. section 1.9):
• The location and number of the fibres (which depend upon the circumstances of
the case, time between offence and recovery)
• The number of fibres types involved
• The relative occurrence of these types and colour of fibres (which depends on the
extent of the analytical information obtained)
• The presence of a cross-transfer
The advantage of a probabilistic model is to structure the above list in a coherent

way and to specify their relationships. Such a model has been applied by (Roux, 1997;
Monard, 2007).
Even though a probabilistic approach has been proposed for fibre evidence for some
years (see Champod and Taroni, 1999 and related references), the approach is still viewed
with unjustified scepticism and as a novelty in the field of trace evidence as shown by the
survey carried out by Grieve (2000), despite its greatest advantage of providing ‘a model
for incorporating the most relevant objective knowledge currently available into evidence
evaluation and forces the examiner to consider other factors relating to the case other than
just the results of the laboratory examination’ (Grieve and Wiggins, 2001, p. 841). It is
increasingly seen as the most appropriate approach to evaluate fibre evidence (Roux, 2009).
The aim of this chapter is to present an updated formal probabilistic framework, that
ought to be useful for the evaluation of fibre transfer cases. We first propose a general
presentation of the framework including the likelihood ratio and the concept of hierarchy
of propositions.* We then apply it to various scenarios involving fibre evidence taken from
car seats or recovered from garments. These scenarios have been chosen in order to show
the impact of probabilistic assignments of parameters on the values of the likelihood ratio.
Finally these scenarios have been incorporated in probabilistic graphical models generally
known as Bayesian networks (Taroni et al., 2014).
10.3.2 Interpretation Process

In this section we aim at presenting how scientific information can be used to update the
knowledge about a case under investigation. We refer to this process as ‘interpretation’. The
Bayesian approach is especially useful here (Robertson and Vignaux, 1995). The interpreta-
tion process takes also advantage of the forensic data whose value is expressed through a
likelihood ratio. When scientists focus on the likelihood ratio only, we will say that they
‘evaluate’ the evidence. When the trier of fact uses that piece of forensic evidence in the
context of the case, we will talk about ‘interpretation’. Essentially, this section will lead to
the distinction of two roles: the role of the forensic scientist focused on the evaluation of
evidence and the role of the factfinder focused on the interpretation of the case. That will
condition the questions forensic scientists can address.
The evidence, say E, is generally the combination of results gathered from two materials.
One of unknown origin (i.e. material recovered on the crime scene) and control
material of known origin (i.e. the ‘victim’s’ garment).
Forensic scientists should give the court an evaluation which illustrates the convincing
force of the results. This is inevitably linked with probability as measure of uncertainty, and
therefore a model to interpret the value of evidence is essential. Bayes’ theorem permits the
revision based on new information (E) of a measure of uncertainty about the truth or oth-
erwise of an issue (H1 or H2). It shows how to combine prior, or background information (I)
with new data to give posterior probabilities for particular outcomes or issues. Let O(H1) =
P(H1)/P(H2) be the ratio of the prior probability of an issue H1 to the prior probability of its
alternative H2. This is commonly called ‘prior odds’. The model allows one to alter given
‘prior’ odds in favour of H1, in the light of new information, to obtain ‘posterior’ odds,
O(H1|E) = P(H1|E)/P(H2|E), on the issue, through simple multiplication by the likelihood
ratio (LR), P(E|H1)/P(E|H2). Prior information, I, may be accounted for by conditioning on
I throughout. More explicitly:
P ( E | H1 , I )
O ( H1 | I , E ) = ⋅ O ( H1 | I )
P ( E | H 2 , I )
Posterior odds Prior odds
Likelihood ratio
The likelihood ratio measures the value of the evidence in terms of a pair of hypotheses,
indicating if the given set of observations supports one hypothesis more than the other. The
concept of evidence is therefore relative: it shows how observations should be interpreted
as evidence for H1 vis-à-vis H2, but it makes no mention of how those observations should
be interpreted as evidence in relation to H1 alone (Royall, 1997). This way of reporting
* We will use the term ‘proposition’ or ‘hypothesis’ synonymously to express the matter of interest for a
Court of Justice.
does not provide an answer on the truthfulness of the hypotheses themselves, but only
an answer on the degree of support for one hypothesis versus another. Likelihood ratios
can also be expressed using a verbal scale phrasing conclusions like ‘the evidence (slightly,
strongly, very strongly, etc.) supports H1 against H2’.
Scientists are not generally in a position (nor is it their duty) to assess the odds in favour
of an issue, because a complete assessment must combine both the forensic statement (E)
and background information (I). The scientist does not usually have access to background
information available to a member of a jury or to a judge. Most of the time the scientist does
not know the particular circumstances of a case and thus is not able to justify any assess-
ment of the prior odds. This means that the numerical statement (or the opinion) given by
the scientist is not sufficient alone to assign the odds on H1. Scientists are concerned solely
with the likelihood ratio. Jurists deal with the odds on an issue. If scientists give their
opinion on the truthfulness of hypothesis (e.g. ‘It is very probable that the questioned fibres
originated from the known textile’), they commit what has been coined the transposed
conditional fallacy or the prosecutor’s fallacy (Evett, 1995). Scientists should focus on their
results, e.g. my findings are more probable given that the questioned fibres originated from
the known textile, than given they originated from an unknown textile.
This distinction of roles is generally not respected in literature and practice. Several
examples of experts’ conclusions in fibre examination were presented in the first edition of
this chapter. Unfortunately they are still found in practice.
These conclusions offer the court an incorrect answer if they are based only on scientific
examination (the opinion of the forensic scientist on an issue) to a question the court
is interested in. These conclusions are inappropriate for two main reasons. Firstly, the
scientist has usurped the role of the judge by making a probabilistic statement about the
issue. Secondly, the scientist has assessed the posterior probability without knowledge of
the background information on the specific case.
Some examiners solve this problem by refusing to qualify their findings phrasing
their conclusions like ‘the fibres recovered are consistent with the suspect’s garment’ or
‘the fibres recovered match the victim’s pullover’. Such phasing is unfortunately the rule
in the results presented from proficiency tests (see for example Test No. 11–539: Fibers
Analysis, http://www.ctsforensics.com/assets/news/3139_web.pdf) and recommended by
the SWGMAT (Scientific Working Group on Materials Analysis) guidelines*. In reference
to such patterns of similarities, phrases such as ‘similar to’, and ‘consistent with’ are used
to describe – and not evaluate – results. However the non-scientist cannot be expected to
pick up the subtle nuances and thus may give the results more weight than is deserved.
The statement in People v. Giangrande made more than 20 years ago remains actual: ‘We
believe use of the word “match”, “consistent”, “could have originated”, misrepresented
the evidence.’ (People v. Giangrande 101 Ill. App. 3d 397, 1991). In wrongful conviction
of Guy Paul Morin in Canada, the public inquiry noted that the use by experts of vague
conclusions such as ‘the hair might come from . . .’, ‘the fibres are consistent with . . .’ or ‘the
* ‘To say that the questioned fiber originated from the known textile, it either had to be the only fabric of
its type ever produced or now existing, or the transfer of fibers was directly observed. As neither of these
situations is likely to occur, fiber examiners must conclude that because the questioned fibers exhibit
the same results in all tested properties as the fibers from the known sample, the questioned fibers are
consistent with originating from the source textile.’ (Scientific Working Group on Materials Analysis
(SWGMAT), 1999, p. 4).
fibres match’, was one of the causes leading to the attribution of disproportionate probative
value to evidence (Kaufman, 1998).
In conclusion Bayes’ Theorem appears to be a useful framework that invites scientists
to assess the value of scientific evidence under two competing hypotheses and clarifies the
respective roles of scientists and of members of the court. It leads to the establishment of
three general principles for evidence evaluation that are summarized below (Champod
and Evett, 2009) and applied in subsequent scenarios.
1. Scientific evidence must be evaluated within a framework of circumstances:
The alleged circumstances dictate the propositions at hand. Any change in the
circumstances may impact their choice. In addition, these alleged circumstances
are part of background information I that will condition the evaluation of the
evidence (E). What matters only here is the domain-relevant information that has
an impact on the specification of the textiles involved* or the activities associated
with these textiles.† As suggested by Grieve (Grieve, 1992, pp. 240–241), evaluation
requires a cooperation between the scientist and the case agent so as to be aware of
the alleged circumstances of the case.
2. Scientific evidence can only be evaluated by considering at least two propositions: In
order for forensic scientists to provide a balanced view on the contribution of the
scientific evidence in a specific case, there is a need for them to consider the find-
ings in respect to the allegations of both sides. That ensures balance between both
parties, because scientists are invited not to consider the evidence under one side
only (Jackson, 2000)‡. The issues are set before the evaluation of the evidence (LR).
Consequently the definition of the propositions themselves is outside the duties of
the experts, it is the matter for the court that is dealing with activities.
3. It is necessary for the scientist to address questions of the kind ‘what is the prob-
ability of the evidence, given the proposition?’: Whereas the factfinder necessarily
addresses questions of the kind ‘what is the probability of a given proposition
given the evidence?’, a proper interpretation framework dictates that the scientist
must consider the probability of the evidence, given the propositions set by the
parties.
From these three principles, we can derive formulation and statement-writing
guidelines (Evett, Jackson, Lambert, et al., 2000; Association of Forensic Science Providers,
2009; Willis et al., 2015; Champod et al., 2016).
10.3.3 Likelihood Ratios for the Evaluation of Trace Evidence

In this section we will move into the evaluative process and, following some definitions,
show how the case circumstances allows to set the relevant propositions in scenarios
involving fibre evidence. We have chosen two scenarios: one where the questioned material
* For example, a witness account of the garment worn by offender.

† For example, a sexual assault carried out around 2 am on a naked woman on her bed; the lining is seized
on the following morning; a suspect wearing a blue tack suit is arrested two days after the event; he
denies having ever been in the home of the woman.
‡ If the evidence is reported against one proposition only (for example, because only one side of the story is
disclosed), then the evidence is not evaluated.

is recovered on the crime scene, and the other where the questioned material is recovered
on the person of interest (or suspect).
The following terms need to be defined:
T Traces or extraneous material of forensic interest recovered on the receptor R.

These traces could have one or more sources.
R The object or person (receptor) on which traces have been recovered.
CS The object or person (control source) which could be the source (or one of the
sources) of the traces, and that is at the origin of the material defined as control
material.
The following two examples give an illustration of the definitions and show that
traces T are not always related to the crime scene, but can be found in association
with a suspect. The argument is also valid for the control source which is not always
associated with the suspect, but can come from the victim or be an object from the
scene.
1. Example 1: Material recovered on the crime scene: An offender entered the rear of
a house through a hole which he cut in a metal grille. The offender attempted to
force the entry but failed, the security alarm went off. He left the scene. About ten
minutes after the offence, a person of interest, wearing a red pullover, is appre-
hended in the vicinity of the house following information from an eyewitness who
testified that he saw a man wearing a red pullover running away from the scene.
At the scene, a tuft of red fibres was found on the jagged end of one of the cut edges
of the grille. The trace T is the tuft of fibres from the grille which is the receptor
R, whereas the control source CS is the suspect’ s pullover. Fibres taken from this
pullover will constitute the control material.
2. Example 2: Material recovered on the suspect/person of interest: A victim came to
the police to report that she had been raped by one of her old boyfriends. The sus-
pect denies any recent contact with the victim. His T-shirt is taken for examina-
tion. Foreign fibres are collected from his garment (which is the receptor R), they
constitute the traces, T. In this case, the victim’s garment is the control source(s),
CS, which will produce the control material.
After technical examination, the traces and control material can be described by their
respective sets of attributes which are:
y: y represents the relevant forensic attributes describing the traces T, which can be
the sum of extrinsic features (physical attributes such as quantity, number, posi-
tions on the receptor, etc.) and intrinsic features (chemical or physical descriptors
such as analytical results).
x: x represents the set of relevant forensic features describing the control material
produced by CS. As a general rule, intrinsic features contribute principally to this
set of data; however, following control experiments of transfer according to the
alleged activities, extrinsic characteristics may be added.
E: E = {x, y} represents the evidence, or the overall forensic findings, provided by the
characteristics examined.
Note that in the above definitions, there is no mention of a so-called ‘match’. One could
either adopt a two-stage approach (comparison and then significance) or a continuous
approach. The two-stage approach is common in forensic science, however flaws have been
identified (Robertson and Vignaux, 1995, pp. 118–120). Various publications deal with con-
tinuous measurements where the evidence E is evaluated without adopting a comparison
stage (using a match criteria or a threshold value) (Lindley, 1977; Aitken and Taroni, 2004).
The examples developed in this chapter only consider categorical data (e.g. chemical types
of fibres or colour categories, presence or absence of delustrant) to highlight the global
architecture. Managing continuous data (such as for example colour measurement using
microspectrophotometry) requires more sophisticated statistical techniques but are based
on the same framework (Taroni et al., 2010).
In most transfer cases, more than one single trace element is recovered especially for
fibres. Hence, it is convenient to adopt a grouping approach for the recovered traces. In
the same way, groupings are possible for the control material originating in the control
source. A group is defined as a set of material (from the traces and the control material)
which shares the same forensic attributes. Moreover, for traces, a group is declared only
if there is sufficient specificity in the shared features to link reasonably these traces with
a unique source. Most of the time for fibres, these grouping decisions can be made only
with difficulty through complete numerical demonstration (for glass evidence statistical
grouping approaches based on refractive index measurements are available (Triggs et al.,
1997)), but there can, and should be, logically qualified opinions. As an example of this
kind of reasoning, we may consider the following: When a certain number of extraneous
fibres are recovered on the seat of a stolen car, and the forensic analysis indicates an agree-
ment between all fibres recovered, then the scientist can reasonably admit that all the fibres
came from the same source. A group can be reasonably declared and this entire group will
be described by the set y.
From these definitions, we need to move to the questions that the forensic scientist will
help to address. The concept of hierarchy of propositions* is critical here and is detailed
in (Cook et al., 1998a; Evett, Jackson, and Lambert, 2000). We will use an example to
stress upon the need to assess forensic findings considering activity level propositions in
fibre cases.
Let us consider the case where it is alleged that around 24h00, a woman, Ms Johnson,
has been raped at her home, in her bed, by an acquaintance of hers who was wearing a blue
tracksuit. The bed lining has been secured at 2 am. Mr Smith, wearing a blue tracksuit, is
arrested at his home at 3 am. His garments are seized. He says he has nothing to do with
the incident and has not seen the victim for at least three weeks. He does not deny that she
was assaulted, but it was not by him. The lining and the suspect’s garments are submitted
for fibre examination.
In this case, the prosecution will present the fibre evidence as a result of the alleged
aggression on the bed of the victim by the suspect wearing the seized garments. This
prosecution hypothesis will be noted Hp. The defence will provide an alternative account
* Propositions can be positioned at respectively on the source of recovered material, on the alleged activi-
ties involving the textiles or on the qualification of the offence.
for the recovered material on both items.* At a particular moment in a trial, the context is
generally restricted to two competitive hypotheses, the one proposed by the prosecution,
and the one retained by the defence. In this case, the defence hypothesis, denoted Hd, is
that he has nothing to do with the assault.
The evaluation of the forensic results consists in the assessment of the probabilities of
the observations (x, y) under the two competitive views (prosecution’s and defence’s).
In the area of fibre evidence, we claim that the issues are systematically in relation to
the alleged activities in association with textiles more than in relation to the sources of
the material. Hence activity-level propositions should prevail over source-level proposi-
tions. Consider the following development of the case scenario to argue that point. Let us
assume that in that case, among all recovered fibres, one pink cotton fibre, matching the
lining, has been found on Mr Smith’s tracksuit. If evaluated at source level,† regardless
of the circumstances, this finding will provide support for the prosecution view, support
of strength that is proportional to the rarity of the fibre (based only on the intrinsic fea-
tures). The same conclusion would have been reached had a thousand matching fibres been
recovered. The same conclusion would have been reached if the matching fibre had been
found in the tracksuit pocket or if a thousand matching fibres was recovered on the front
of the tracksuit bottom. The same value is given to two scenarios that bring drastically
opposed extrinsic features that tell something about the alleged activities. Finding a single
matching fibre is not expected under the alleged aggression proposition, whereas is in line
with the defence hypothesis. Conversely, finding a thousand matching fibre is expected
under the alleged aggression proposition, whereas is not in line with the defence hypothe-
sis. A fit-for-purpose evaluation of the findings should concentrate on the alleged activities
in order to bring to the factfinder the appropriate strength associated with all the forensic
observations: i.e. both the instrinsic and extrinsic features of the evidence. If we recall the
list of factors given in introduction, the significance of the number (including the absence
of matching fibres), location and cross transfer of fibres is meaningful only when activities
are considered. It is worth insisting that if we address source level propositions, then factors
such as the number of fibres that were found, their location, transfer, persistence, nature
of the contacts cannot be taken into account, even if these factors appear in the SWGMAT
guidelines, under the paragraph source determination.
Sometimes, when there is a lack of information on what activities are alleged by both
parties, experts are tempted to address propositions such as ‘The clothing of Mr Smith has
been in recent contact with the clothing of Ms Johnston’ versus ‘The clothing of Mr Smith
has NOT been in recent contact with the clothing of Ms Johnston’. Although intuitively
appealing, there are several problems with this approach (Evett et al., 2000). This type of
proposition has been named a pseudo-activity proposition. Indeed, it is very difficult to
assign transfer probabilities, if there is no indication on the type of contact is alleged by
the prosecution. If there has been contact, then the probability of recovering the fibres
could range from 0 to 1 depending on the type of contact. One needs to define this contact
in sufficient detail, so that one could perform transfer experiments or/and assign transfer
* Note that the defence may suggest more than one alternative proposition. The evaluation framework of
this chapter can be used also in such situation.
† Hp: the recovered fibre originates from the victim’s bed lining versus Hd: the recovered fibre originates
from another unknown textile.
probabilities. This definition often passes by describing the activities (i.e. the offender
pinned the victim to the bed; or the offender drove the car). One can therefore argue that
it is simpler to address the activity.
Some experts are reluctant to address activity level propositions, because they say that
with fibres, one cannot, for example, know if the person drove the car or not. And, this is
perfectly correct, the scientist evaluates the number, the emplacement, and type of fibres
given that the suspect drove the car (for example), but does not evaluate the proposition
according to which the suspect drove the car (this is a classical example of the transposed
conditional as we have seen previously).
Another difficulty with the term contact, is the definition of the alternative: what do we
mean by there has not been a recent contact? What is recent? Do we mean that the fibres
are present because of an unknown activity? Do we mean they are present as background?
Again, one needs to be transparent and describe what is the view of the defence.
If we do not define what is meant by ‘contact’, then factors such as transfer and
background cannot be evaluated appropriately. Therefore, the value given to the evidence
will not be the same given genuine activity level propositions, and given unclearly defined
activity level propositions. As the court will not be aware of this, this approach is not
balanced or transparent.
The whole evaluation process should then deal only with assessing the expectation of the
findings given the alleged activities proposed by the parties. That position was already sug-
gested for the evaluation of glass evidence (Hicks Champod, 2007). Leaving the evaluation
at source level is inappropriate and may lead to misleading representation of the evidence.
As a consequence for statement writing we would like to stress that evaluative statements
in relation to fibre evidence should deal with matters at activity level only. As the reader
will realise in the next sections, the interpretative factors at activity level are complex and
require specialist forensic knowledge that needs to be explicit in the statement. We do not
recommend the approach of producing, aside of any case circumstances, a report stating
the factual results (maybe complemented by a ‘consistent with’ or ‘cannot exclude’) and
then arguing that it will be up to the parties during examination and cross-examination
to provoke a discussion regarding the interpretation of the findings in the context of
the alleged activities. Firstly, the task is too complex to be tackled in court without any
preparation. Secondly, leaving the burden of raising such questions to the parties gives the
misleading impression that it constitutes their duty and that can be done in absentia with-
out asking for expert knowledge.
We may have given the impression that the evaluation of the findings is a step that
follows some examinations on recovered items. In reality the steps of obtaining the
propositions precede the forensic examinations. In addition much can be gained by
anticipating on the expected results to optimize value for money and assess the rel-
evance of the forensic examination. This is known as case assessment and is driven by
the concept of likelihood ratio (Cook et al., 1998b, 1999; Jackson and Jones, 2009), as
recently points out by the ENFSI Guidelines for evaluative reporting in forensic science
(Willis et al., 2015), that put forward the necessity to assess scientific findings under
‘activity level’ propositions to meet the needs of the mandating authorities (judges,
prosecutors, defence teams) and help address the key issues in the case. Similar recom-
mendations have been expressed in relation to minute traces of DNA (Taroni et al.,
2013; Champod, 2014).
10.3.4 Likelihood Ratios in Various Scenarios

In the introduction, various factors related to the observed fibres have been recognized as
important in the assessment of the evidential value, hence should impact on the likelihood
ratio. Evett (1984) gave the structure to combine the rarity of the fibres’ characteristics with
considerations of the location of the fibres, their quantity as well as the prevalence of fibres
as background.
Activities involving textiles may lead to multiple transfers. Even if the scientist
faces multiple transfer cases, it is easier here, for didactic reasons, to consider them
individually. A multiple transfer evaluation (cross-transfer) will be developed in a
final example.
10.3.4.1 Evidence Left by the Offender(s)

10.3.4.1.1 Scenario 1: Material Recovered on a Crime Scene (Cut Metal Grille) Let us
recall the example 1, where an offender entered the rear of a house through a hole which he
cut in a metal grille. Here, the offender attempted to force the entry but failed, the security
alarm went off. He left the scene. About ten minutes after the offence, a suspect,wearing
a red pullover, is apprehended in the vicinity of the house following information from an
eyewitness who testified that he saw a man wearing a red pullover running away from
the scene. At the scene, a tuft of red fibres was found on the jagged end of one of the cut
edges of the grille. The trace T is the tuft of fibres from the grille which is the receptor R,
whereas the control source CS is the suspect’s pullover. Fibres taken from this pullover are
the control material.
The likelihood ratio is formally expressed as follows:
P ( E ⏐H p , I ) P ( x , y ⏐H p , I )
LR = = (10.1)
P ( E ⏐H d , I ) P ( x , y ⏐H d , I )
where Hp is the proposition set out by the prosecution that the suspect, wearing this
pullover,* is the person who entered the premises through the grille and Hd the alternative
given by the defence that the suspect has nothing to do with the incident and was not near
the scene.
Equation 10.1 can be expanded using the law of multiplication of probabilities:
P ( x , y ⏐H p , I ) P ( y ⏐x , H p , I ) P ( x ⏐H p , I )
LR = = ⋅ (10.2)
P ( x , y ⏐H d , I ) P ( y ⏐x , H d , I ) P ( x ⏐H d , I )
* Note that this chapter focuses on likelihood ratio based evaluations of fibre evidence in cases in which
there is no uncertainty about whether or not the reference item—typically an item taken from the
suspect or seized at his home —was actually worn at the time of the offence. In situations where this
relationship is uncertain, the reader can refer to the formal treatment presented in Taroni & al. (2012).
It is reasonable to assume that the probability of the characteristics of the suspect’s pullover,
x, do not depend on whether or not the suspect is the offender. So, P(x|Hp,I) = P(x|Hd,I) and
P(y|x,Hd, I) = P(y|Hd, I), thus (2) becomes:
P ( y ⏐x , H p , I )
LR = (10.3)
P ( y ⏐H d , I )
The numerator represents the probability of the observed characteristics of the recovered
fibres (y) given that the suspect is indeed the offender. If we admit that the fibres on the
grille have been left by the offender, this probability, P ( y ⏐x , H p , I ), is equal to 1.*
The denominator asks for the probability on y if the suspect is not the offender (someone
else, wearing another garment, is the offender). The scientist is then interested in assigning
a probability of finding these trace characteristics in a population of potential sources as
defined by the defence proposition. Given that fibres cannot be left at the position for a
reason unconnected to the crime, this probability can be informed after a survey using the
relative occurrence of the compared characteristics among relevant sources, which will be
denoted by γ.
Then the likelihood ratio (3) is expressed by:
P ( y ⏐x , H p , I ) 1
LR = = (10.4)
P ( y ⏐H d , I ) γ
The relevant population is defined by the hypothesis proposed by the defence (Robertson
and Vignaux, 1995, pp. 36–37) and by background information related to the case. To illus-
trate these phenomena let us consider possible lines adopted by the defence:
• If the defence strategy is to argue that the suspect had never been present on
the scene, then the potential sources are defined as red upper garments worn by
burglars (here we are admitting that the eyewitness has seen the perpetrators).
• If the defence proposes the hypothesis that the suspect has been correctly identified
by the witness, but he has not entered the premises through the grille, then the
relevant population is defined by the potential sources without any distinction in
respect to the colour of the garment.
Hence, depending on the strategy of the defence, the relevant population can be modified
and it affects the value of γ. Moreover, the inquiry (part of the background information
related to the case) can affect the definition of the relevant population. It can be illustrated
as follows:
• In the absence of any eyewitness, if the suspect has been apprehended following the
observation by the forensic scientist that the tuft of fibres is red coloured, then the
relevant population is defined by the potential perpetrators wearing red garments.
* This probability set to 1 also means that it is assumed that every time fibres from such a source are
submitted, the laboratory will correctly detect these intrinsic characteristics. Remember that data
considered here are only categorical.
• In the absence of any eyewitness, if the suspect has been apprehended independently
of the forensic attributes of the tuft, then the relevant population is defined by the
potential perpetrators without any distinction in respect to the colour of the garment.
Some examiners argue that it is not necessary to work through this likelihood ratio
development to end up, in this case, with the result that the value of the evidence is dictated
by the relative occurrence of the fibres’ characteristics. The added value of the likelihood
ratio framework here relies in the necessity to specify the hypotheses and background
information available that dictate the relative occurrence of interest.
Moreover, we will illustrate by modifying the hypotheses and then modifying the
relation between the traces and the offence, that the likelihood ratio may be more subtle
and complex than 1/γ and need to incorporate data to demonstrate that the nature of the
result is predictable with what is known about the various phenomena associated with fibre
transfer (e.g. transfer, persistence, tenacity, prevalence).
10.3.4.1.2 Scenario 2: Material Recovered on a Crime Scene (Cut Metal Grille); k
Offenders A change in the competitive hypotheses is adopted; it is related to other pub-
lished case interpretations (Evett, 1993): the scenario is the same as in 1, but instead of one
offender described by a witness, a certain number of offenders, say k, were viewed entering
through the hole which they cut in the metal grille. The suspect has been arrested alone.
The LR is still expressed by Equation 10.3, but now the two main hypotheses (given by
the and prosecution) are:
• Hp: The suspect is one of the k offenders.

• Hd: The suspect is not one the k offenders and has nothing to do with the offence.
The probability P(y|x,Hp, I) must take into account two possible explanations: either
the tuft has been left by the suspect (A) or the tuft has been left by another one of the k
offenders ( A ). Then invoking the law of total probability:
LR = P ( y ⏐x , H p , I ) = P ( y ⏐x , H p , I , A ) P ( A ⏐x , H p , I ) + P ( y ⏐x , H p , I , A ) P ( A ⏐x , H p , I )
In the absence of prior knowledge about the modus operandi of the team of offenders,
we will assume that each offender is equally likely to be the source of the tuft, hence
P ( A ⏐x , H p , I ) = 1 k and P ( A ⏐x , H p , I ) = ( k − 1) k. It is clear that P ( y ⏐x , H p , I , A ) = 1 and
P ( y ⏐x , H p , I , A ) = γ . The denominator implies that the suspect was not on the scene, hence
the number of offenders has no effect on the probability P(y|Hd,I) which is given by γ.
Finally, the LR becomes:
1 ⎛ k −1⎞
+γ⎜
k ⎝ k ⎟⎠ 1 k − 1
LR = = + (10.5)
γ kγ k
Equation 10.5 can be approximated by:
1
LR ≈ (10.6)
kγ
As demonstrated in (4) and (6) the LRs very much depend on the hypotheses tested. Note
also that both P ( y ⏐H d , I ) and P ( y ⏐x , H p , I , A ) equal γ because the conditioning informa-
tion directs, in both hypotheses, to the same relevant population.
10.3.4.2 Evidence Possibly Left by the Offender(s)

In the scenario presented above, it is clear that the recovered material is associated with the
offence. However, in most fibre cases, the recovered traces may either be associated with
the offence or may have been present for reasons unconnected with the offence. We intro-
duce here the concept of the relevance of the traces. If the recovered n fibres are considered
to have come from one source (as defined by the grouping approach), there are two main
explanations for the evidence:
• Either the recovered group of n fibres was transferred, has persisted, and has been
successfully recovered on R, in which case, the receptor did not have this group of
n fibres before. This event will be denoted Ti = n.
• Or the recovered group of n fibres was not transferred, persisted or recovered and
the group of fibres is unconnected with the offence. This event will be denoted Ti = 0.
The likelihood ratio in this case is
P( y | x, Hp ) P ( y | x ,Ti = n , H p ) ⋅ P (Ti = n | x , H p ) + ( y | x ,Ti = 0 , H p ) ⋅ P (Ti = 0 | x , H p )

LR = = (10.7)
P ( y | Hd ) P ( y | Ti = n , H d ) ⋅ P (Ti = n | H d ) + P ( y | Ti = 0 , H d ) ⋅ P (Ti = 0 | H d )
From the equation above and for the remainder of the chapter, the notation of I (back-
ground information) is omitted for sake of simplicity. Keep in mind, however, that this
information still conditions the probabilities at hand.
10.3.4.2.1 Scenario 3: Material Recovered on a Crime Scene (Stolen Car) A stolen car
is used in a robbery on the day of its theft. One hour later, the car is abandoned. During
the night the police find the stolen vehicle. On the polyester seats, which have recently been
cleaned with a car vacuum cleaner, extraneous textile fibres are collected by taping. The
car owner lives alone and has never lent his vehicle to anyone. The owner wears nothing
but cotton.
The day following the robbery a suspect is apprehended, his red woollen pullover and
his denim pants are confiscated and transmitted to the laboratory. He says he has nothing
to do with the incident. He does not know the owner of the car and has never driven his car.
On the driver’s seat (receptor R) three groups of foreign fibres have been collected.
These groups consist of: 150 white cotton fibres (G1), 200 blue cotton fibres (G2), 170 red
wool fibres (G3). Following laboratory examinations, the fibres from G1 and G2 are found
to correspond to clothing of the owner; the presence of these fibres can be easily explained
and therefore only for sake of simplicity they will be ignored. We will only consider group
G3 for the evaluation.
The forensic evidence can be formulated as follows:
• y: The group of 170 red woollen fibres recovered is described by a set y of extrinsic
and intrinsic characteristics.
• x: The red woollen pullover of the suspect (control source) generates control fibres
described by a set x of characteristics.
The suspect has been apprehended independently of the evidence recovered.

The competitive hypotheses considered by the parties could be as follows:
• Hp: The suspect, wearing that jumper, drove the stolen car.
• Hd: The suspect has nothing to do with the incident.
The LR is given by Equation 10.7.
P ( y | x, Hp ) P ( y | x ,Ti =170 , H p ) ⋅ P (Ti =170 | x , H p ) + ( y | x ,Ti =0 , H p ) ⋅ P (Ti =0 | x , H p )

LR = = (10.8)
P ( y | Hd ) P ( y | Ti =170 , H d ) ⋅ P (Ti =170 | H p ) + P ( y | Ti =0 , H d ) ⋅ P (Ti =0 | H d )
In this type of transfer, the probability of the event Ti=170 depends on the hypotheses
considered (Hp or Hd ). The probability of recovering 170 fibres following this offence
depends on the source of the traces. Here, in car seat cases, the Ti probabilities (i = 0, 1,
2, 3,…, 170,…) represent the number of fibres that would be recovered on R following
this offence). They depend more on the sheddability of the garment than on variables
such as persistence or recovering methods. The sheddability is not only a factor of the
composition of the garment (intrinsic characteristics analysed), but also depends on
variables such as its construction. Both variables are assumed controlled in the hypoth-
esis Hp (indeed the police have the pullover of the author), but they are unknown if it is
someone else who drove the car, as the real offender and his pullover are not available
(Hd ). Hence:
P (Ti =170 | x , H p ) ≠ P (Ti =170 | H d )
P (Ti = 0 | x , H p ) ≠ P (Ti = 0 | H d )
Following (10.8), the LR becomes:
1γ b
b0
t170
b t0

LR =
( ) ( )
P y x ,Ti =170 , H p ⋅ P (Ti =170 x , H p ) + P y x ,Ti =170 , H p ⋅ P (Ti =170 x , H p )
=
b0t170 + b1γ bt0
(

)

(

)
P y Ti =170 , H d ⋅ P (Ti =170 H d ) + P y Ti = 0 , H d ⋅ P (Ti = 0 H d ) b0 γt '170 + b1γ bt '0
b0 γ t '170 b1 γ b t '0
(10.9)
where: t170, t’170, t0 and t’0 are called transfer probabilities
t170 is the probability that these 170 fibres G3 have been transferred from the suspect’s
upper garment to the car seat. They have remained and have been recovered. This prob-
ability is assigned according to the hypothesis Hp. t’170 is the probability that these 170 fibres
G3 have been transferred from the offender’s upper garment (and not from the suspect’s
garments) to the car seat. They have remained and have been recovered. This probability is
assigned given the hypothesis Hd. t0 is the probability that no fibres have been transferred
(remained or recovered) from the suspect’s upper garment to the car seat. This probability
is assigned according to the hypothesis Hp. t’0 is the probability that no fibres have been
transferred (remained or recovered) from the offender’s upper garment to the car seat. This
probability is assigned according to the hypothesis Hd.
b0 and b1 are called background probabilities:
b0 is the probability of recovering no foreign fibre group (FFG) by ‘chance’ on the
driver’s car seat. These fibres can be distinguished from the garments of the habitual user(s)
of the car. b1 is the probability of having by ‘chance’ the occurrence of one foreign fibres
group (compatible number) on the driver’s car seat. These fibres can be distinguished from
the garments of the habitual user(s) of the car, b1 ≤ 1 – b0.
γ and γb are the estimated occurrences of the compared characteristics from y in
respectively the population of alternative sources (garments from potential offenders) and
similar sized extraneous groups of fibres found as part of the background on car seats.
It is important to emphasise that b0 and b1 are mutually exclusive partitions of a total
event which could be called, ‘having on the driver’s seat, zero, one or more foreign fibre
group (compatible number) which can be distinguished from the garments of the habitual
user(s) of the car’. Considering that the number of groups range from 0 to infinity:
i =∞
1= ∑b
i =0
i
Then:
i =∞
b1 = 1 − b0 − ∑b ≤ 1− b
i =2
i 0
Average background probabilities (b0 or b1) can be estimated from data obtained through
surveys where groups of extraneous fibres (foreign fibre groups or FFGs) were counted on
relevant surfaces (driver’s seats or garments). Such surveys have been made for car seats by
Roux and Margot (1997). The values b0 or b1 on car seats can be more accurately assigned if,
in the survey, the groups of fibres corresponding to the owners’ garments are specified and
subsequently excluded. Roux and Margot obtained average values such as: b0 = 0.05, b1 =
0.27, b2 = 0.55, b3 = 0.09 and b4 = 0.05. The use of average probabilities could be challenged
when specific information about the surface under consideration shows a departure from
the average surface (e.g. the inquiry shows that the seats of the stolen car have just been
cleaned with a specific car vacuum cleaner). In the above scenario, as the stolen vehicle is
regularly cleaned and the car is solely used by the owner wearing cotton-made garments,
the probability b0 can be assumed here to be close to 1. Therefore b1 is close to 0.
For assigning a probability to γ, following (Champod et al., 2004), an appropriate
source of data could be a survey of upper garments worn arrested individuals suspected of
similar offence. For γb, survey such as (Roux and Margot, 1997) can be also used.
Transfer probabilities can be obtained through controlled experiments involving the
surfaces in question. Only simulations – which try to reproduce the alleged actions – can
provide useful data because of the high number and complexity of the variables involved in
transfer, persistence and recovery phenomenon in general. However, when the variables are
more limited, as for the transfer onto car seats, average data can be obtained from surveys
described in the literature (Roux et al., 1996). Here, it will be assumed that on average a
large number of fibres are transferred and recovered in the particular circumstances of the
case. Therefore t170 will be much larger than t0. In the same way, controlled experiments
involving woollen garments coming from potential offenders and the driver’s seat of the
stolen car would yield a valid estimation for t’170 and t’0. Here, it will be assumed again that
on average a large number of fibres are transferred and recovered in the particular circum-
stances of the case. Therefore t’170 will be much larger than t’0.
Thus, the likelihood ratio (9) can be approximated as follows:
Negligible in this scenario
bt + b1γ bt0 t170

LR = 0 170 ≈ (10.10)
b0 γt '170 + b1γ bt '0 γt '170

Negligible in this scenario
If we admit that the transfer characteristics of the suspect’s pullover do not differ from the
transfer characteristics of the possible offenders’ garment, then (10) is approximated by:
1
LR = (10.11)
γ
This formulation of the LR is the same to the LR developed in scenario 1 where the rel-
evance of the recovered traces was assumed established. One could argue that the value
of the evidence is driven by the relative occurrence (γ) whether we consider source level
propositions or activity level propositions, and that therefore we need not to worry about
activity. We do not agree, it is not because the LR formula is the same that we did not con-
sider the factors transfer, persistence and background. These factors were evaluated and
assigned to be of the same order of magnitude using our expert judgment: this task has not
been left to the Court.
10.3.4.2.2 Scenario 4: Material Recovered on a Crime Scene (Stolen Car) The fourth
scenario is very similar to the third except that the number of recovered fibres is low
(n = 10) compared to the number expected in this kind of scenario. Considering this fact,
the transfer and background probabilities must be reassessed.
The equation of LR remains analogous as in scenario 3:
b0t10 + b1γ bt0

LR = (10.12)
b0 γt '10 + b1γ bt '0
The background and transfer probabilities could be estimated as previously described,

but as the data available to assess the significance of such a small number of fibres are
limited in literature, we propose to study the robustness of the LR in function of these
probabilities (γ and γb being set for illustrative purposes to 0.01). This is an analysis of the
sensitivity of the LR.
For the transfer probabilities, we will assume that they do not differ under Hp or Hd.
Let us imagine that we have been asked before we started the examination how many
fibres we expected to be transferred from the suspect’s (or offender’s) garment onto the
driver’s seat and imagine that we had replied ‘about 30’, the mean obtained from controlled
experiments. We are not committing ourselves to 30 and only 30 – there is a vagueness

which recognises non-zero probabilities for fewer or greater numbers of fibres. It appears
reasonable to model the probability distribution for the number of fibres recovered after
a given time according to transfer, persistence and recovery phenomenon. Curran et al.
(1998) suggested a modelling technique for glass evidence which we believe is also useful
for assessing fibres transfers. Simulations (100,000) have been carried out given that, in
our case, the number of fibres does not decrease in function of time (excellent persistence)
(Curran et al., 1998). The quality of the recovery technique (adhesive efficiency) has been
set between 90% and 95% (Chable et al., 1994) and the other variables specific to glass have
been discarded (distance between offender and window, fragments stuck in the pockets,
cuffs or seams). The empirical distribution for the number of fibres (mean = 30) is shown
in Figure 10.1.
This distribution gives us a t0 = t’0 of about 0.00009 and a t10 = t’10 of about 0.053. As con-
trolled experiments are time and resource consuming, we are interested to obtain transfer
probabilities for various mean transferred numbers of fibres as shown in Table 10.3.
For background probabilities, as shown in Table 10.4, b0 has been changed to simulate
three situations of ascending order of probabilities. The value of b1 is approximated by 1- b0.
With these parameters’ estimates, according to Equation 10.12, we can represent
(Figures 10.3 and 10.4) the LR as a function of the mean number of fibres:
′
Table 10.3 Empirical Probabilities Obtained for t0 = t0
′
and t10 = t10 in Function of the Estimated Mean Number
of Transferred Fibres in the Alleged Scenario
Mean Number of Transferred Fibres t0 = t0′ t10 = t10′

1 0.407 0.00001
5 0.02 0.0178
10 0.006 0.097
15 0.0003 0.0658
20 0.0002 0.028
25 0.0001 0.0116
30 0.00009 0.0053
35 0.00009 0.004
40 0.00009 0.0013
45 0.00009 0.0012
50 0.00009 0.0011
Table 10.4 Estimates Given to b0 and b1 in

Three Different Situations
Situation 1 (LR1) b0 0.01
High background b1 0.99
Medium background b1 0.5
Low background b1 0.01
The mean number of recovered fibres under H2 has been set to 60.
20000
18000
16000
14000
12000
Likelihood ratio
LR1
10000 LR2
LR3
8000
6000
4000
2000
0
5 10 15 20 25 30 35 40 45 50
Expected number of recovered fibres under H1
Figure 10.3 Evolution of the LRs in function of the mean number of fibres in three situa-
tions of background probabilities (and) when the expected number under Hd is higher than the
expected numbers under Hp..
This figure helps to emphasize the importance of factors other than the relative
occurrence in the evaluation of the LR. In function of the transfer probabilities, asso-
ciative evidence may lead to likelihood ratios near 1, implying no support for either
hypothesis. Moreover, it helps the scientist to have an idea on the sensitivity of the LR
as a function of the uncertainty of the parameters. Hence, in that case, even if we assess
the evidence under the more conservative scenario in terms of background probabilities
(b0 = 0.01 and b1 = 0.99), the LR remains above 20 for mean number of expected fibres
ranging from 10 to 40.
It could be deduced from the above example the LR has its minimum at 1 and its
maximum at 100 (corresponding to 1/γ). This will hold only if t0 = t’0 and t10 = t’10 but these
are rarely reasonable assumptions. Figure 10.5 shows how the LRs will vary if the transfer
probabilities differ under Hp and Hd.
Therefore, this analysis leads to the following conclusions:
• We have recovered 10 fibres. If 10 fibres is the average number expected under the
hypothesis of implication of the suspect and if on average an potential offender’s
garment will leave an average of 60 fibres, then the LR far exceeds 100 without any
significant influence of the background probabilities.
The mean number of recovered fibres under H2 has been set to 5.

1000.00
100.00
10.00
Likelihood ratio
1.00
5 10 15 20 25 30 35 40 45 50
0.10 LR1
LR2
LR3
0.01
Expected number of recovered fibres under H1
Figure 10.4 Evolution of the LRs in function of the mean number of fibres in three situations
of background probabilities (b0 and b1) when the expected number under Hd is lower than the
expected numbers under Hp.
• 10 fibres have been recovered, but this number does not correspond to the aver-
age number obtained with the suspect’s pullover (>25). Moreover, if the expected
number for an average offender’s garment is compatible with the recovered num-
ber (5 for example), then the LR may be under 1 and then support the alternative
hypothesis.
10.3.4.2.3 Scenario 5: Material Recovered on the Suspect’s Belongings (Car) We may

consider another scenario where the car belongs to a man who is suspected of abducting
a woman and attempting rape. The victim was wearing a red woollen pullover and denim
pants. According to the suspect, nobody but his wife ever sits in the passenger’s seat; fur-
thermore, the car seats have been recently vacuumed. On the passenger’s seat three groups
of foreign fibres have been collected. These groups consist of a large number (170) of white
cotton fibres (G1), blue cotton fibres (G2) and red wool fibres (G3). Following laboratory
examinations, the fibres G1 and G2 are similar to the wife’s clothing. As the presence of
these fibres can be explained, they will be ignored (simplification).
Then the forensic evidence can be formulated as follows:
• y: The recovered group of 170 fibres is described by a set y of characteristics.
• x: The red woollen pullover of the victim (control source) generates control fibres
described by a set x of characteristics.
60
50
40
Likelihood ratio
30
20
10
0
0 0.08 0.15 0.23 0.3 0.38 0.45 0.53 0.6 0.68 0.75 0.83 0.9 0.98
Conditional transfer probability u(20)
Figure 10.5 Evolution of the LRs in function of the mean number of fibres in three situations
of background probabilities (b0 and b1) when the expected number under Hd is higher than the
expected numbers under Hp.
Considering that the suspect denies that the victim has ever been in his car, the competi-
tive hypotheses are:
• Hp: The victim sat on the passenger’s seat of the suspect’s car as alleged.
• Hd: The victim has never been on the passenger’s seat of the suspect’s car.
The LR is expressed as follows:
1γ b
b0
t170
b t0

LR =
( ) ( )
P y x ,Ti =170 , H p ⋅ P (Ti =170 x , H p ) + P y x ,Ti =170 , H p ⋅ P (Ti =170 x , H p )
=
b0t170 + b1γ bt0
P( y H ) d b1γ b

b1 γ b
(10.13)
Logically, under the hypothesis Hd (meaning no offence), the probability that the traces are
related to the offence is zero. Considering the circumstances of the offence, the number of
recovered fibres and the sheddability of the pullover, it will be assumed that on average a
large number of fibres are transferred and recovered in the particular circumstances of the
case. Therefore t170 will be much larger than t0 and the expression b1γbt0 is negligible: (13)
then becomes:
b0t170
LR = (10.14)
b1γ b
It is important to note that the denominator of the LR now includes b1: indeed, if Hd is
accepted, then this foreign group is present for some unknown reason (i.e. background),
b1 being equal or smaller than 1 – b0, if b0 is assessed as close to 1 (considering the use and
the cleaning procedures of the car) then b1 becomes smaller than 1 and the LR is strongly
increased compared to a standard 1/γb likelihood ratio.
In casework it is not rare that the number of extraneous groups of fibres (large
number of fibres in each group) is greater than 1, let us say N > 1. Data related to the
traces are noted y1 to yN. Nevertheless, only one group is compatible with the victim’s
pullover. In this case the LR is developed considering that the evidence is the set {x1,
y1, y2 ,…, yN}. The equation of the LR is developed in the same way as proposed for glass
evidence (Curran et al., 2000):
bN −1 ⋅γ 2 , … , N ⋅( N −1)!
t170

Negligible in this case

P ( y1, y 2 , … y N ⏐x1 ,Ti =170 , H p ) ⋅ P (Ti =170 | x1 , H P ) + P ( y1 , y 2 … , y N | x1 ,Ti = 0 , H P ) ⋅ P (Ti = 0 | x1 , H P ))
LR =
P ( y1 , y , … , y N | H d )

bN ⋅γ 1 ⋅γ 2,… ,N ⋅ N!
10.15
where:
• bN-1 the probability of recovering as background N-1 group of foreign fibres (com-
patible number) on the passenger’s car seat.
• γ2,…,N the probability of observing the analytical characteristics of the groups 2 to
N among the foreign groups recovered as background on car seats.
According to Fong and Inami (1986), when N is increased, bN is equal bN-1, thus (14)
becomes:
t170
LR ≈ (10.16)
γ1 ⋅N
This simple result demonstrates that, in evaluating a match, the scientist must consider
concording and non-concording elements. Pre-assessing the case and setting out our
expectations before looking at the suspect’s garments (for material recovered on the crime
scene or victim) or before looking at the victim’s garments (for material recovered on the
suspect) helps avoid focusing on matching fibres only. It is important not only to focus on
the fibres that match, but also to consider other groups of fibres compatible with the alleged
facts. A similar analysis is proposed by Causin et al. (2004).
10.3.4.2.4 Scenario 6: Material Recovered on the Crime Scene and on the Person
of Interest This scenario is the same as the third (car theft), except that two pieces of
evidence are considered, E1 and E2:
• E1: On the driver’s seat one group of relevant foreign fibres has been collected
(a group of fibres other than cotton). It consists of a large number (170) of red wool
fibres. This forensic evidence E1 will be denoted {y1, x1}, where y refers to recovered
traces and x refers to control material (from the suspect’s red woollen pullover).
• E2: On the suspect’s pullover and denim jeans among N FFG, a group of 20 extra-
neous black polyester fibres have been collected. They share the same intrinsic
features with the fibres of the driver’s seat and this evidence E2 is noted {y2, x2}.
It is generally accepted in literature that the presence of a cross-transfer in fibres is
a factor that appears to strengthen the value of the evidence (Grieve, 1992). When two
individuals (or an individual and an object) are in contact during a criminal action, a
reciprocal transfer of material is usually involved. In that case, the two sets of recovered
traces have to be considered as dependent. If a transfer has occurred in one direction, and
the expert has recovered traces characterising this transfer, then the expert would generally
expect to find trace evidence characterising the transfer in the other direction. The pres-
ence of one set of transfer evidence gives information about the presence of the other set
of transfer evidence. The final likelihood is the multiplication of two dependant likelihood
ratios (LR1 and LR 2|1), the first analogous to Equation 10.11, the second to Equation 10.16
as the propositions and assumptions about background are similar*. This means, in this
scenario, that, under Hp, the number of fibres transferred onto the car seat gives an infor-
mation about the ‘intensity’ of the contact, which influences the amount of fibres recovered
on the suspect’s garment given the view that he is the offender.
1 u20
LR = ⋅ (10.17)
γ1
Nγ 2
LR1 LR 2
where:
• u20 is the conditional transfer probability that the 20 recovered fibers on the
suspect’s clothing (y2) have been transferred (remained and recovered) from the
car seat of the stolen car to the suspect’s upper garment. This probability is condi-
tional because it is assigned given that the suspect has sat on this seat (given all the
specific circumstances pertaining to the alleged fact) and that 170 fibres (y1) have
been found on the seat.
• γ1 is the probability of observing the analysed characteristics in the relevant
population. Here, if the fibres do not come from the suspect, they have been left
by the real offender’s upper garment. Therefore we would assign the rarity of the
observed characteristics among upper garments worn by persons stealing cars.
• γ2 is the probability associated to the instrinsic feature from y2 in similar sized
extraneous groups of fibres (background) found on clothing of potential offenders.
* By relaxing this assumption, an alternative model has been proposed in Taroni & al., 2014.
From this last formula, we constructed Figure 10.5 to investigate the sensibility of the
LR in function of the variation on with fixed values for γ1 and γ2 set to 5% and N equals to 6.
We can observe that the strength of the ‘link’ between the car seat and the suspect’s
garment increases (compared to the value of LR1 = 20) only in situations where the con-
ditional transfer probability (u20 ) is greater than 0.3. Hence, if the two transfers are not
compatible (u20 < 0.3), the likelihood ratio is reduced. Therefore, contrary to intuition, a
cross-transfer does not necessarily mean an increase of the strength of the evidence. To
assess correctly the strength of a link between two objects, it is essential to consider all
potential transfers. This necessity leads to considering missing evidence as pointed out by
Schum (1994, p. 96):
It is often common to focus on evidence regarding the occurrence of events and easy to over-
look evidence regarding the non-occurrence of events. In any inferential context it is just as
important to inquire about what did not happen as it is to inquire about what did happen.
In the forensic context, missing evidence may occur in two ways. First, there is the
situation in which potential receptors have been searched and evidence has not been found.
Secondly there is the situation in which potential receptors have not been searched or the
results of the search have not been reported. Both situations define the concept of missing
evidence. A probabilistic approach to missing evidence has been developed by (Lindley
and Eggleston, 1983).
The framework presented here enables us to deal with missing evidence. Suppose that
instead of recovering 20 polyester black fibres (E2), no matching fibres have been recovered
on the suspect’s upper garment. The likelihood ratio LR 2|1 becomes:
u0
LR2 1 = = u0 (10.18)
1
The denominator represents the probability of finding no matching fibres if the suspect
never sat on the car seat, this probability is clearly 1. The numerator represents the condi-
tional probability that no fibres have been transferred (remained and recovered) from the
car seat of the stolen car on the suspect’s upper garment. This probability remains condi-
tional because it is assigned given that the suspect has sat on this seat and that 170 fibres
(y1) have been found on the seat. As used in scenario 4, the value of u0 can be estimated
by controlled experiments or modelling techniques. Here, the value of u0 will be closed to
0. It can be shown that if u0 < 0.015 then the likelihood ratio is below 1 and sustains the
alternative hypothesis.
The evaluation of fibre evidence recovered during the investigation of criminal cases
is commonly approached without properly taking into account the influence of the phe-
nomenon of cross-transfer between the two parties involved in the action (e.g. the vic-
tim and the aggressor). For example, fibres recovered on a victim’s jumper are scompared
with control material coming from a suspect’s clothing. Furthermore, if forensic scientists
find fibres similar to those of the victim’s jumper on the suspect’s clothing, it is generally
accepted that this new event increases the value of the evidence and favours the pros-
ecution’s view. It has been shown that the value of the link between the victim and the
suspect does not necessarily increase, but it could sustain the alternative hypothesis that
he never aggressed her. In this situation, a relevant role is played by the so-called missing
evidence which allows the scientist to take into consideration information on the absence
of expected evidence.
It has been strongly emphasised that it is necessary to consider cross-transfer evidence
in order to assess correctly the value of an association. However, cases exist where the
information is not available. Therefore, some questions remain open. For example, how
do we assess the value of the evidence when investigators did not look for cross-transfer
evidence? Were cross-transfer traces effectively present but not collected or was there no
recovered evidence?
10.3.5 Bayesian Networks

Bayesian networks (BNs, for short) have been developed as a framework that should assist
scientists and practitioners in applying the theory of probability to inference problems
of more substantive size and, thus, to more complex problems. Therefore they constitute
a substantial development for the management of fibre cases and their application will
expand in the future. There is now a considerable body of scientific literature that reports
on BNs as a tool that can be used to study, develop and implement probabilistic proce-
dures for evaluating the probative value of particular items of scientific evidence in forensic
science (Taroni et al., 2014).
BNs can roughly be defined as a pictorial representation of the dependencies and influ-
ences (represented by arcs) among variables (represented by nodes) deemed to be relevant
for a probabilistic inference problem. BNs are therefore compact model representations for
reasoning under uncertainty. In a BN, usually, the variables are discrete, with a finite set
of mutually exclusive states. Directed edges between nodes are used to express assumed
relationships among the variables. Both nodes and edges are combined in order to form a
directed acyclic graph.
Assuming discrete variables, the strength of the relationship between the variables is
quantified by conditional probability distributions associated with each node: for a vari-
able B with parents A1,…,An, there is a conditional node probability table P(B|A1,…,An),
whereas a table containing unconditional probabilities P(A) is assigned to a variable A that
has no parents. If the conditional relationships implied by the structure of a BN hold for a
set of variables (A1,…,An), then the joint probability distribution, P(A1,…,An), is given by
the product of all specified ‘conditional’ probabilities:
P ( A1 , … , An ) = ∏ P ( A par ( A ))
i i
i
The term par(Ai) represents the set of parental variables of Ai. The previous equation for-
mally defines what a BN means: a representation of the joint probability distribution for
all the variables.
For the building of BNs, it important to note that they do not represent the flow of
information, but serve as a direct representation of a part of the real world. This means that
through the use of BNs, experts can graphically and numerically articulate their personal
view of the real world (Lindley, 2000). The task of modelling will principally be influenced
by both, the properties and experts’ individual view, perception and ultimately, extent of
understanding of a domain of interest.
The aim in this section is to show how previously discussed formulae, for example (9)
and (13) can be translated into a simple graphical model. Such a model can then be further
expanded to deal with more complex case. Dedicated software is available to construct the
structure, the associated probability tables and compute all relevant probabilities.*
From the likelihood ratio formula (9) and the propositions given, one can see that
y (the characteristics of the recovered material) is conditioned on x (the characteristics of
the control material), T (the transfer) and H (the hypotheses) and that T is conditioned on
x and H. Therefore, one can propose the following structure (Figure 10.7).
The variables in the BN shown in Figure 10.8 and specified in Section 4.2 are defined
as follows:
• H: This variable accounts for the activity level propositions Hp ‘The suspect, wear-
ing that jumper, drove the stolen car at the time of the offence’ and Hd ‘the suspect
has nothing to do with the incident or the stolen car’.
• T: This variable represents the occurrence of transfer, persistence and recovery of
a group of n fibres. There are three states, Ti=n (i.e. a transfer of a group of n fibres),
Ti=0 (i.e. no transfer (persisted or recovery) of fibres) and Tother (transfer of a group
of fibres of size other than n and 0).†
• X: This variable represents the intrinsic characteristic of the control source (i.e.
the red woollen suspect’s pullover) and has two states x (i.e. the observed intrinsic
fibres characteristic) and not-x (i.e. a fibre characteristic other than x).
• Y: This variable denotes the extrinsic and intrinsic characteristics of the recovered
fibres. There are two states, y, the observed fibre characteristics, and not-y, other
fibre characteristics (different from y).
X1 X2
T U
H
Y1 Y2
Figure 10.6 Evolution of the LR in function of, u20 f1 and f2 being set to 5% and N to 6.
X T
H
Y
Figure 10.7 Structure of a BN for evaluating transfer evidence using four variables H, T, X and
Y as defined previously in the text.
* Freeware: GeNIe available at http://genie.sis.pitt.edu// Commercial software: HUGIN available at www.

hugin.com
† This variable is not characterised by two states T and T . A potential use of two states for nodes T implies
n 0
that, in the conditional probability table, the sum of the two probabilities associated with these states need
to add to 1. This is not correct in a strict sense because the value for, say t0, is not 1–tn = 170. Practically, an
additional state called, for example Tother, and which represents all possibilities for transfer other than t0 and
tn would be needed. This would allow the sum of all associated probabilities to sum up to 1.
A probability table is associated to each variable (node). For nodes H and X, the tables are
presented here below (Table 10.5):
Note that relevant states of the nodes H and X will be instantiated when using the
Bayesian network. Therefore, the initial probabilities assigned to these nodes are not
relevant (in previous table are set to 0.5). The only requirement is that the values sum
up to one. In fact, when a likelihood ratio is calculated using the general expression
( )
LR = P y x , H p / P ( y x , H d ), all information on the right hand side of the vertical line
(conditioning bar) is assumed known, thus, in the BN, their states are instantiated (speci-
fied as known) and the probability associated to each of this state is automatically set to 1.
The probability table for node T is more complex (Table 10.6). The node T has entering
arcs from x and H. The table thus has the structure shown below. Note that inspecting the
likelihood ratio formula can provide relevant indications on how to complete the prob-
ability table of the node T.
Note that:
• When Hd is true, then knowledge of x does not affect the belief about the prob-
ability of transfer, so values t’0 and t’170 are assigned. These refer to the probability
of transfer from the true offender’s garment which is different from that of the
suspect, described as x (see Section 4.2.1).
X1" X2"
T" U"
H"
Y1" Y2"
Figure 10.8 Structure of a BN for evaluating cross-transfer evidence using seven variables. H
for the propositions and respectively T, X1 and Y1 for the evidence found on the car seat and U,
X 2 and Y 2 for the evidence found on the suspect’s garment.
Table 10.5 Probability Tables for Node H

and X Respectively
State Probability
Hp 0.5
Hd 0.5
State Probability
x 0.5
not-x 0.5
Table 10.6 Probability Table for Node T

Hp Hd
x not-x x not-x
T0 t0 1/3 t’0 t’0
Tn t170 1/3 t’170 t’170
Tother 1- t0- t170 1/3 1- t’0- t’170 1- t’0- t’170
• Under Hp it is assumed that the suspect’s garment was in contact with the car seat
(while driving). This is why only probabilities for transfer assuming x are consid-
ered. Probabilities for transfer given not-x (and Hp) do not enter the calculations
(default values of 1/3 can be assigned).
Following Figure 10.6, the node Y depends on the three (parents) nodes H, T and X. The
probability tables for the node Y (respectively given Hp and Hd) are as follows (Table 10.7
and Table 10.8):
Note that there are also some default values of 0.5 which have been used. The reason for
this is the same as that given previously (i.e. probabilities must sum to one); the values are
not relevant because they do not enter the calculations of the likelihood ratios.
Note that it is assumed to be impossible to observe y if a number of fibres other than
n = 170 has been transferred, that is a case in which when Tother is assumed to hold. This is
also why the extension to Tother is not written in the likelihood ratio formula.
The BN presented in Figure 10.8 and described throughout probability tables in the
previous tables offers the same results as Equations 10.9 and 10.13. The only difference
between the two equations is that in Equation 10.13 no transfer is admitted under hypoth-
eses Hd, this means P (Tn =170 | x , H d ) = 0.
The BN presented here can be easily extended to consider more complex situations like
cross-transfer as presented in Section 4.2.4 or in cases involving more than one relevant
group of fibres (Figure 10.8).
In cases involving ‘cross-transfer’, please note – as seen in Figure 10.7 – two key additions.
An arrow is used to link the transfer nodes T and U. Indeed, if a transfer has occurred in one
direction and the scientist has recovered fibres characterizing this transfer, then the scientist
may expect to find fibres characterizing the transfer in the other direction. The two recovered
traces should be considered as dependent. Arrows are also used to specify that the physical
attributes of the textiles involved will influence on transfer, persistence and recovery of fibres
in either direction, from the offender to the car seat and vice-versa. That is to say that vari-
ables such as fibres length and sheddability of the suspect’s garment influence the transfer of
fibres to the car’s seat. And conversely the physical characteristics of the textile composition
Table 10.7 Probability Table for Node Y under the Prosecution

Hypothesis Hp
Hp
T0 T170 Tother
x not-x x not-x x not-x
y b1 γb 0.5 b0 0.5 0 0
not-y 1-(b1 γb) 0.5 1- b0 0.5 1 1
Table 10.8 Probability Table for Node Y under the Prosecution

Hypothesis Hd
Hd
T0 T170 Tother
x not-x x not-x x not-x
y b1 γb b1 γb b0 γ b0 γ 0 0
not-y 1-(b1 γb) 1-(b1 γb) 1-(b0 γ) 1-(b0 γ) 1 1
of the car seat influence the transfer of fibres to the suspect’s or offender’s garments. Physical
characteristics also affect the persistence of the transfer of fibres (Taroni et al., 2014).
How can we obtain a numerical value for the likelihood ratio using Bayesian networks?
We want to update our beliefs in the light of new information and because a BN is a
complete model for the variables and their relationships, it can be used to answer proba-
bilistic queries about them. The network can be used to obtain updated knowledge of the
state of a subset of variables when other variables are observed. So, the idea is to instanti-
ate the nodes corresponding to the known variables (i.e. variables that are located on the
right side of the conditioning bar in the likelihood ratio formulae); the software propagates
the effects of this information to all nodes and a new probability distribution is obtained.
Practically speaking, the numerator of the LR, P(y|x,Hp), is obtained by instantiating nodes
X and H to states x and Hp, respectively. We then observe the value corresponding to the
state y of node Y. Similarly, for the denominator of the LR, P(y|x,Hd): its value is obtained
by instantiating state Hd of node H. Note that – given the independence between y and x
under Hd - then information x does not alter the probability of y. An advantage of such a
tool is that the computation of the posterior probability on hypotheses of interest can be
obtained by instantiating the evidence (i.e. states y and x of nodes Y and X, respectively)
and observe their effect on the distribution of node H.
Bayesian networks can be used to evaluate a given outcome as presented above but
also to carry out pre-assessment of the expected likelihood ratios for a range of possible
outcomes. The only amendment required is to extend the list of states of the variables T and
Y. T could be the following ranges: T0, T1–5, T6–20, T>20. Conversely for Y: y 0, y1–5, y6–20, y>20.
Using that structure, we can reproduce pre-assessment calculations presented in Jackson
and Jones (2009).
The role of Bayesian networks is then advisory; they provide assistance and support in
the rigorous evaluation of evidence according to the laws of probability, which guarantee
rationality in reasoning under uncertainty. Bayesian networks can take an important part,
notably as (1) cognitive tools for the structuring and logical guiding of mental processes,
(2) a coherent inferential concept for drawing inferences to hypotheses based upon evi-
dence, and (3) a flexible and dynamic approach allowing knowledge to be transparently
revised upon newly acquired information.
10.3.6 The Need for Data

Use of statistical data in the area of fibre evidence is viewed with scepticism if not a blind
refusal even to an attempt. This situation is more salient in the United States. For Houck
the major problem is that ‘trace material populations may change at least as fast, if not
sometimes faster, than they can reasonably be sampled.’ (Houck, 1999, p. 4). The recent
response from SWGMAT reaffirms Houck’s position focused on the perceived impossibil-
ity to assign a probability to relative occurrence of a given characteristic of interest (i.e. γ).*
That approach can only lead to a reporting scheme of the kind ‘consistent with or
could have come from’, that has been already critically discussed in section 2. As stated by
Garrett and Neufeld in their analysis of wrongful convictions in the United States (Garrett
and Neufeld, 2009, p. 19): ‘To say that two items are “consistent” without being able to tell
* http://www.swgmat.org/SWG%20response%20to%20RDT.doc
the jury that consistency is rare or common, renders the evidence potentially mislead-
ing and hence raises questions whether it is inadmissible as both irrelevant and unduly
prejudicial.’
This refusal of some examiners to articulate their opinion in probabilistic suggests
to us that there is a major misconception of the role of inferential models as a support
to decision making. There is no other way to structure the list of factors influencing the
significance of fibre evidence (spelled out in the introduction) than to use a formal devel-
opment. The model helps to understand the relationship between the factors, their relative
importance and realise that the relative occurrence (γ) represents only one factor among
others.
Despite the complexity of the task to assign probabilities to the factors (γ, γb, ti, bi) we
have outlined in the previous sections, forensic scientists cannot shy away from this duty.
Assigning probabilities of that nature is what constitute expertise. To help in that task, as
Grieve and Wiggins put it (2001, p. 843): ‘[…] some objective data is better than none.’ Reviews
of relevant papers helping to assign these probabilities can be found in the present monog-
raphy or in the tri-annual Interpol reviews (Robertson and Roux, 2004; Palmer, 2010, 2011).
The evaluation framework developed here shows another benefit to this discussion:
The question is not about gathering data from textile productions, but it is about gathering
data in relation to the relevant populations at hand (for γ and γb) and the alleged activities
(for ti, bi). Data acquisition should be driven by the case circumstances and may require
case-specific experimentations. Recent examples are given in (De Wael et al., 2011; Palmer
and Polwarth, 2011).
Note that it would be an illusory exercise to seek precision in the assignment of
probabilities to the parameters of the model. The model itself is an approximation of
the reality. What matters is the order of magnitude of each parameters and their relative
importance (as specified by the model) more than the decimal precision of the estimate.
We can see a lot of benefits in undertaking sensitivity analysis as presented previously.
10.3.7 Conclusion
Following an explicit definition of objects and attributes involved in trace transfer, a gen-
eral expression of the likelihood ratio has been derived. This framework has permitted
to approach different case scenarios involving fibre evidence where the direction of the
transfer and the amount of the recovered material were varied. The variations around the
scenarios allowed the identification and evaluation of the dominant parameters and their
respective effect on the likelihood ratio.
Throughout this chapter, our aim is to recommend a logical approach to fibre evalua-
tion through a systematic assignment of likelihood ratios to assist the forensic scientist in
the evaluation of scientific evidence. The main points can be summarized as follows:
1. Evaluation of the evidential value of forensic evidence is a matter of probability.

2. For the assessment of the strength of scientific evidence (E) it is necessary to
consider the probability of the evidence under at least two given competing
hypotheses for its occurrence, respectively presented by the prosecutor and by the
defence (Hp and Hd). The value of the evidence is estimated using a likelihood
ratio LR = P(E| Hp, I)/P(E|Hd, I). Hence, the likelihood ratio is not only defined by
the evidence, but also by the strategy chosen either by the prosecution or by the
defence, as well as by the forensically relevant case circumstances.
3. The purpose of fibre examination, offered to court as associative evidence, is to pro-

vide guidance as to the precise alleged activities in the case at hand. Hence the only
appropriate set of propositions in this area has to tackle the problem at activity level.
4. Evidence (recovered trace and control material) may be to some extent incompat-
ible with the offence. Hence P(E|Hp, I) is not always equal to 1, because of their
number and position, and because of the phenomena of transfer, persistence and
recovery. There is a need to consider transfer probabilities.
5. When traces (recovered evidence and control material) are supposed to be asso-
ciated with the offence, we have also to allow for the probability of their absence
before the commission of the offence. The corollary of this is that if traces are sup-
posed to be unrelated to the offence, the probability of their presence by ‘chance’
(i.e. due to an unknown activity) has also to be considered. There is a need to con-
sider background probabilities.
6. The evaluation of trace evidence has to consider all potential recovered traces and
not only the concordant evidence. This number of groups has to be considered.
7. The number of declared offenders has a significant impact on the likelihood ratio.
8. The definition of Hd excludes the implication of the defendant. Thus the scientist
has to specify what is the relevant population in which adequate forensic surveys
have to be made.
9. In all investigation and evaluation of transfer traces, all possible exchanges have to
be investigated including extreme situations in which expected evidence has either
not been found or not produced.
10. Bayesian networks represent a powerful graphical approach for evidence evalua-
tion in judicial contexts. They represent a major improvement to deal coherently
with complex fibre cases.
References
Aitken, C. G. G. and Taroni, F., 2004. Statistics and the Evaluation of Evidence for Forensic Scientists
(2nd ed.). Chichester, UK: John Wiley & Sons Ltd.
Association of Forensic Science Providers. 2009. Standards for the formulation of evaluative forensic
science expert opinion, Science & Justice, 49, 161–164.
Causin, V., Schiavone, S., Marigo, A., and Carresi, P. 2004. Bayesian framework for the evaluation of
fibre evidence in a double murder—A case report, Forensic Science International, 141, 159–170.
Chable, J., Roux, C., and Lennard, C. 1994. Collection of fiber evidence using water-soluble cello-
phane tape, Journal of Forensic Sciences, 39, 1520–1527.
Champod, C. 2014. DNA transfer: Informed judgment or mere guesswork? In A. Biedermann, J.
Vuille and F. Taroni (Eds.), DNA, Statistics and the Law: A Cross-Disciplinary Approach to
Forensic Inference (Vol. 2, pp. 22–24). Lausanne, CH: Frontiers Media S.A.
Champod, C. and Evett, I. W. 2009. Evidence interpretation: A logical approach. In A. Moenssens
and A. Jamieson (Eds.), Wiley Encyclopedia of Forensic Science (Vol. 2, pp. 968–976). Chichester,
UK: John Wiley & Sons.
Champod, C., Evett, I. W., and Jackson, G. 2004. Establishing the most appropriate databases for
addressing source level propositions, Science and Justice, 44, 153–164.
Champod, C., and Taroni, F. 1999. Interpretation of fibres evidence—The Bayesian approach. In M.
Grieve and J. Robertson (Eds.), Forensic Examination of Fibres (2nd ed., pp. 379–398). London:
Taylor & Francis Ltd.
Cook, R., Evett, I. W., Jackson, G., Jones, P. J., and Lambert, J. A., 1998a. A model for case assessment
and interpretation, Sci. Justice, 38, 151–156.
Cook, R., Evett, I. W., Jackson, G., Jones, P. J., and Lambert, J. A., 1998b. A hierarchy of propositions:
Deciding which level to address in casework, Sci. Justice, 38, 231–240.
Cook, R., Evett, I. W., Jackson, G., Jones, P. J., and Lambert, J. A. 1999. Case pre-assessment and
review in a two-way transfer case, Science and Justice, 39, 103–111.
Curran, J., Triggs, C. M., Buckleton, J. S., Walsh, K. A. J., and Hicks, T. 1998. Assessing transfer prob-
abilities in a Bayesian interpretation of forensic glass evidence, Science and Justice, 38, 15–21.
Curran, J. M., Hicks, T. N., and Buckleton, J. S. 2000. Forensic Interpretation of Glass Evidence, Boca
Raton, FL: CRC Press.
De Wael, K., Baes, C., Lepot, L., and Gason, F. 2011. On the frequency of occurrence of a peculiar
polyester fibre type found in blue denim textiles, Science and Justice, 51, 154–162.
European Fibre Group, K. Wiggins and M. Grieve (Eds.). 2001. The Manual of Best Practice for the
Forensic Examination of Fibres, Wiesbaden, Germany: Bundeskriminalamt.
Evett, I. W. 1984. A quantitative theory for interpreting transfer evidence in criminal cases, Applied
Statistics, 33, 25–32.
Evett, I. W. 1993. Establishing the evidential value of a small quantity of material found at a crime
scene, Journal of the Forensic Science Society, 33, 83–86.
Evett, I. W. 1995. Avoiding the transposed conditional, Science and Justice, 35, 127–131.
Evett, I. W., Jackson, G., and Lambert, J. A. 2000. More in the hierarchy of propositions: Exploring
the distinction between explanations and propositions, Science and Justice, 40, 3–10.
Evett, I. W., Jackson, G., Lambert, J. A., and McCrossan, S. 2000. The impact of the principles of
evidence interpretation on the structure and content of statements, Science and Justice, 40,
233–239.
Fong, W. and Inami, S. H. 1986. Results of a study to determine the probability of chance match
occurrences between fibers known to be from different sources, Journal of Forensic Sciences,
31, 65–72.
Garrett, B. L. and Neufeld, P. J. 2009. Invalid forensic science testimony and wrongful conviction,
Virginia Law Review, 95, 1–97.
Grieve, M. C. 1992. Information content: The interpretation of fibres evidence. In J. Robertson (Ed.),
Forensic Examination of Fibres (pp. 239–262), New York: Ellis Horwood Limited.
Grieve, M. C. 2000. A survey on the evidential value of fibres and the interpretation of the findings
in fibre transfer cases. Part 2 - interpretation and reporting, Science and Justice, 40, 201–209.
Grieve, M. C. and Wiggins, K. G. 2001. Fibers under fire: Suggestions for improving their use to
provide forensic evidence, Journal of Forensic Sciences, 46, 835–843.
Hicks Champod, T. 2007. Addressing activity levels when interpreting glass or trace evidence: What
else? Global Forensic Science Today, 16–21.
Houck, M. M. 1999. Statistics and trace evidence: The tyranny of numbers, Forensic Science
Communications, 1, http://www.au.af.mil/au/awc/awcgate/fbi/houck.htm.
Jackson, G. 2000. The scientist and the scale of justice, Science and Justice, 40, 81–85.
Jackson, G. and Jones, P. J. 2009. Case assessment and interpretation. In A. Moenssens and A.
Jamieson (Eds.), Wiley Encyclopedia of Forensic Science (Vol. 2, pp. 483–496). Chichester, UK:
John Wiley & Sons.
Kaufman, F. 1998. Report of the Kaufman Commission on Proceedings Involving Guy Paul Morin,
Toronto: Ministry of the Attorney General, Ontario, http://www.attorneygeneral.jus.gov.on.ca/
english/about/pubs/morin/.
Lindley, D. V. 1977. A problem in forensic science, Biometrika, 64, 207–213.
Lindley, D. V. 2000. The philosophy of statistics, Journal of the Royal Statistical Society. Series D (The
Statistician), 49, 293–337 (with comments).
Lindley, D. V. and Eggleston, R. 1983. The problem of missing evidence, The Law Quaterly Review,
99, 86–99.
Monard, F. 2007. Etude des Mécanismes de Transfert des Fibres en Sciences Forensiques. PhD thesis,
Université de Lausanne, Institut de police scientifique, Lausanne.
Palmer, R. 2010. The forensic examination of fibres. In N. Nic Daeid and M. Houck (Eds.), Interpol’s
Forensic Science Review (pp. 2–18). Boca Raton, FL: CRC Press.
Palmer, R. 2011. Forensic examination of fibres—A review: 2007 to 2010. In Proceeding of the 16th
Interpol Forensic Science Symposium (Vol. to appear), Boca Raton, FL: CRC Press.
Palmer, R. and Polwarth, G. 2011. The persistence of fibres on skin in an outdoor deposition crime
scene scenario, Science and Justice, 51, 187–189.
Robertson, B. and Vignaux, G. A. 1995. Interpreting Evidence—Evaluating Forensic Science in the
Courtroom, Chichester, UK: John Wiley & Sons Inc.
Robertson, J. and Roux, C. 2004. Forensic examination of fibres—A review: 2001 to 2004. Proceeding
of the 14th Interpol Forensic Science Symposium, http://www.interpol.int/Public/Forensic/IFSS/
meeting14/ReviewPapers.pdf.
Roux, C. 1997. La valeur indiciale des fibres textiles découvertes sur un siège de voiture: Problèmes et
solutions. PhD thesis, Université de Lausanne, Institut de police scientifique et de criminologie,
Lausanne.
Roux, C. 2009. Fibers evidence, interpretation of. In A. Moenssens and A. Jamieson (Eds.), Wiley
Encyclopedia of Forensic Science (Vol. 3, pp. 1277–1282). Chichester, UK: John Wiley & Sons.
Roux, C., Chable, J., and Margot, P. 1996. Fibre transfer experiments onto car seats, Science and
Justice, 36, 143–151.
Roux, C. and Margot, P. 1994. L’estimation de la valeur indiciale des fibres textiles découvertes en
relation avec une affaire criminelle—Utopie ou réalité? Revue internationale de criminologie et
de police technique, 67, 229–241.
Roux, C. and Margot, P. 1997. The population of textile fibers on car seats, Science and Justice, 37,
25–30.
Royall, R. M. 1997. Statistical evidence—A likelihood paradigm London: Chapman Hall.
Schum, D. A. 1994. Evidential Foundations of Probabilistic Reasoning, New York: John Wiley &
Sons.
Scientific Working Group on Materials Analysis (SWGMAT). 1999. Forensic fiber examination
guidelines, Forensic Science Communications, 1, http://www.fbi.gov/about-us/lab/forensic-sci-
ence-communications/fsc/april1999/houcktoc.htm.
Taroni, F., Bozza, S., Biedermann, A., and Garbolino, P. 2010. Data Analysis in Forensic Science:
A Bayesian Decision Perspective, Chichester, UK: John Wiley & Sons, Ltd.
Taroni, F., Biedermann, A., Bozza, S., Comte, J. and Garbolino, P. 2012. Uncertainty about the
true source. A note on the likelihood ratio at the activity level, Forensic Science International:
Genetics, 7, 467–470.
Taroni, F., Biedermann, A., Vuille, J., and Morling, N. 2013. Whose DNA is this? How relevant a
question? (a note for forensic scientists), Forensic Science International, 220, 173–179.
Taroni, F., Biedermann, A., Bozza, S., Garbolino, P., and Aitken, C. 2014. Bayesian Networks for
Probabilistic Inference and Decision Analysis in Forensic Science (2nd ed.), Chichester, UK: John
Wiley & Sons, Ltd.
Triggs, C. M., Curran, J. M., Buckleton, J. S., and Walsh, K. A. J. 1997. The grouping problem in foren-
sic glass analysis: A divisive approach, Forensic Science International, 85, 1–14.
Willis, S. M., et al. 2015. ENFSI Guideline for Evaluative Reporting in Forensic Science, Dublin:
European Network of Forensic Science Institutes, http://enfsi.eu/sites/default/files/documents/
external_publications/m1_guideline.pdf.
Future Trends for Forensic
Fibre Examination
KORNELIA NEHSE
11
SANDRA KOCH
MICHAEL GRIEVE
MAX M HOUCK
JAMES ROBERTSON
Contents
11.1 New Fibre Developments 428
11.1.2 Smart Textiles – Intelligent Textiles 429
11.1.2.1 Thermal Properties 429
11.1.2.2 Moisture Regulating Properties 430
11.1.2.3 Antibacterial Properties – Health Sector 430
11.1.2.4 Chromic (Colour Changing) Properties 430
11.1.2.5 Shape Memory Properties 431
11.1.2.6 Conductive Properties 431
11.1.2.7 Stain Repellent Properties 432
11.1.2.8 UV Protection Properties 432
11.1.2.9 Enhanced Properties via Plasma Treatment 432
11.1.3 Nanomaterials 432
11.1.4 Green Materials 433
11.1.4.1 Cellulosic Fibres 434
11.1.4.2 Poly Lactic Acid (PLA) Fibres 434
11.1.4.3 Protein Fibres 434
11.1.5 Composite and Bicomponent Fibres 435
11.1.6 Recycled Material 436
11.1.6.1 Pre-Consumer Textile Waste Material 436
11.1.6.2 Post-Consumer Textile Waste Material 437
11.1.6.3 Other Post-Consumer Waste Material 437
11.1.7 Security Measures 437
11.1.7.1 Production Tracking 437
11.1.7.2 Body Armour/Protective Textiles (e.g. Zylon, Kevlar, Nomex) 438
11.1.8 Fibre Updates 438
11.1.8.1 Poly Trimethylene Terephthalate
(PTT – Triexta, Subclass of Polyester) 438
11.1.8.2 Poly Lactic Acid (PLA) 438
References 440
Online References 440
11.2 Emerging Technologies and Management Considerations 441
427

11.2.2 Improving Analytical Capability 442
11.2.3 Colour Analysis 444
11.2.4 Case Management 444
11.2.5 Interpretation 446
11.2.6 International Cooperation 448
11.2.7 Manufacturing Information 449
11.2.8 Improving Standards 449
11.2.9 Use of Textiles and Fibres as Intelligence in Investigations 450
References 453
11.1 New Fibre Developments
KORNELIA NEHSE
SANDRA KOCH
11.1.1 Introduction
Textile production has been and continues to be a very traditional process. The fibre and
textile markets have changed significantly in the past 20 years but some of the new fibres are
merely reengineered fibres or of a type that has seen a revival and new marketing opportu-
nities. Today, natural and man-made fibres are equally used for the production of clothes
and other textiles. These fibres are often found in casework and may represent a contact
between two or more people and/or items. Therefore, it is necessary to be aware of new
developments in the market as these materials may be encountered in forensic cases. The
focus of this chapter is on fibres and fibre developments which affect routine investigations.
The use of new resources to manufacture man-made fibres and the so called intelligent
textiles or smart clothes will likely play a dominant role regarding the next generation
of fibres, fabrics and ready made textile products. In addition to the recycling of textile
materials, updated production methods that create fibres by environmentally conscious
methods are also entering the market. Improved security measures for clothing and tex-
tiles have also emerged to be able to track clothing in production, delivery and use.
Considerable development started several years ago in the area of functional textiles
for sportswear and outdoor clothing to address the extended need for protection and
humidity transport. These systems were mainly based on modified fibre types and textile
layers with capillary capability to regulate temperature.
Over the past few years, the situation in textile industries has changed further due to
several influences. A new strategy has been to look for external partners and to create com-
pletely new solutions. This is a new way of thinking for the textile industry which has lead
to co-operation between material sciences, engineering, chemistry, design, process develop-
ment and clothing physiology and may influence our work in forensic fibre analysis over the
coming years. In particular, the developments in nanotechnology have introduced fibre types
with added values. Today communication and other technical devices play a significant role
in our lives which the textile industry is now taking advantage of. This has lead to a focus on
an increasing number of solutions that will be implemented in textiles and clothing over the
next several years regarding communication purposes and wearable biofeedback systems.
The driving forces can be found in healthcare, sports, military and public safety areas.
Future Trends for Forensic Fibre Examination 429
While some of these fibres and fabrics have been announced and have been avail-
able for many years, the Federal Trade Commission and the International Bureau for the
Standardization of Man-Made Fibres have been slow to declare new fibre types such as poly
trimethylene terephthalate (PTT) or poly lactic acid (PLA) as separate fibre types. Research
is always ahead of the market and what the market can sell often dictates how older fibres
are brought back under different advertising or reused in different ways.
11.1.2 Smart Textiles – Intelligent Textiles

Smart or intelligent textiles are those which ‘think’ or ‘act’ for themselves and enhance
the functionality of regular textile materials and refine basic functions like protection,
safety and convenience (Blüchel and Malik, 2006; Matilla, 2006). What makes these high
performance fibres function are often additives to the fibre or polymer (e.g. coatings on
the fibre or fabric or implementations into the fibres material) or the cross-sectional
shape and inner structures. Additional innovations have been introduced in electronic
monitoring through fabrics which have been incorporated for health, fitness, military
and security uses (e.g. communication technique).
11.1.2.1 Thermal Properties

Occasionally we find that our clothing is not appropriate to keep us warm or cool in
a particular environment. The textile industry has worked to create clothing that provides
a balanced system to trap heat or cool by releasing heat. The goal of these smart textiles
are to have the same textile material keep us warm or cool as needed, in different environ-
ments and with different activities.
11.1.2.1.1 Phase Change Materials Phase change materials (PCM’s) can be found in out-
door and sport clothing (e.g. Comfortemp®, Outlast®). These materials have the capability to
alter their state of aggregation. They are contained in durable microcapsules that melt and
store heat during times of higher activity. As temperature decreases, the contents of the cap-
sules crystallize and release heat. During the melting and heating processes the temperature of
the PCM itself doesn’t change. During these processes the encapsulated paraffin materials are
either in a solid or a liquid state and are capable of charging or recharging. To reach a suitable
thermoregulation result within a textile, a complex system of textile and layer construction in
combination with the PCM is necessary. The general technique was already developed in the
60s for NASA space suits, but only with the introduction of nanotechnologies has this effect
become available to a wider market. This micro-encapsulated PCM is also known as mPCM.
There are several options to attach the PCM material to the fabric. The PCM capsules
can be found locked within fibres (e.g. acrylic), in foams (e.g. polyurethane) or the material
might be coated with the PCM capsules. Under the last conditions the PCM capsules can
be found on the surface of a fabric and both between the fibres as well as attached to them
(Figure 11.1).
(a) (b) (c)
Figure 11.1 Phase Change Material: (a) attached, (b) coated, (c) implemented.
11.1.2.1.2 Blended Materials Recently a polyester smart textile based on blended fibre
cross sections (e.g. Thermo-cool™) was developed to produce better heat and water transfer.
These fabrics are a blend of hollow core fibres with fibres that have a channeled surface. The
channeled surface fibres serve to wick away moisture and allow evaporation and breath-
ability while the hollow core fibres allow for regulation of heat either to transfer heat away
from the body or allow air in to cool down the wearer. These fibres are used in a wide range
of products (e.g. outdoor garments, sleeping bags, bedclothes, underwear, socks, shoes).
11.1.2.1.3 Ceramic Implementation Ceramic particles are also used to enhance thermal
effects in textiles. The particles are fused into the fibres (e.g. polyester or polypropylene)
and can be found locked within the fibres. Ceramic particles store body heat which is then
reflected through long wave heat radiation. Aluminosilicates are the primary component
of ceramic fibres and along with continuous filament fibreglass a cellulose fibre or poly-
propylene homopolymer are used to make a variety of woven textiles, ropes and tape for
industrial use (e.g. CeraTex™, Nextel™ and Fiberfrax®). The ceramic fibres are typically
used as a substitute for asbestos in insulation or sealing products as they are made to be
resistant to chemical corrosion, thermal shock and mechanical vibration, have low thermal
conductivity and provide insulation. They are also used in functional textiles and medical
textile applications. As ceramic fibres generally have industrial applications, they would
only rarely appear in forensic contexts. They are used in functional textiles and medical
textile applications.
11.1.2.2 Moisture Regulating Properties

Multi-layer systems have been in use since the 1980s. Textiles that utilize a layering system
to wick away moisture incorporate fibres in the fabric with water absorbing polymer
fibres. New developments have improved these textiles with the introduction of super
water-absorbing polymer fibres blended into fabric. Hydrophilic (water absorbing) and
hydrophobic (water repellent) fibres combined into one fabric have improved the per-
formance of moisture regulation and enhanced the thermoregulation and protective
properties (e.g. Hydroweave®).
11.1.2.3 Antibacterial Properties – Health Sector

Products made with chitin have been found to have antibacterial properties and may help
with wound healing. Fibres made with chitin are of great interest to the medical field and
may appear in items received from hospitals, socks, undergarments and exercise clothing
(e.g. Crabyon®).
Silver particles may be added to fibre surfaces for antibacterial purposes. One example
is the cellulosic fibre Sea Cell Active®. In addition to the antibacterial properties, there
are odor control aspects to adding silver to fibres and fabrics. When viewed under a light
microscope up to 400x, there is no apparent difference between the regular and active
versions. However; using SEM-EDX, small opaque dots have been found on the active ver-
sion of the fibre which LA-ICP-MS analysis identified as silver. (Koch in GFST issue 1 2007)
11.1.2.4 Chromic (Colour Changing) Properties

The needs of the military have also led to the rapid development of intelligent textiles.
Textiles with improved camouflage properties are needed in the military in the same
way as the previously mentioned protective and regulating functionalities. For example,
uniforms that change colour are desirable in varying surroundings. A reversible, change-
able material can offer a desired chameleon effect. Materials which remove and change
colour are dependent on external stimulation.
• Photochromic: External stimuli energy is light.

• Those which emit the colour when activated by the visible light, most likely
organic materials
• Those which emit the colour when activated by ultraviolet radiation (e.g.
activated in the light or by the loss of light), most likely inorganic material
(approximately 10% of fluorescent dye added to the spinning mass)
• Thermochromic: External stimuli energy is heat.

• Those whose colour changes as a result of reaction to heat
The colour change is produced by a change in the refractive index of the liquid crystal
or a rearrangement of the molecular structure of the dye. In both cases, this is a result of
the change in temperature.
Upon heating the colour changes, upon cooling the process is reversed
• Solvatechromic: External stimuli energy is liquid.

• Those whose colour changes as a result of contact with liquid (e.g. water)
Examples of water activated colour change can be found in swim wear.
Chromic changes are produced by dyes that can be found incorporated in microcap-
sules. When directly added to the spinning mass it becomes difficult to detect the dye.
11.1.2.5 Shape Memory Properties

Shape memory properties (SMPs) in textiles are intended to provide better insulation. They
respond to extreme heat or cold to activate the shape memory and increase the tiny air gaps
within a garment or seal them through small springs (Hu, 2007). Cuprous-zinc alloys or nickel-
titanium alloys are used for the springs. This process is reversible when using cuprous-zinc.
Shape memory polymers, like polyurethane films and sensitive polymer gels, have
the same effect as the Ni Ti alloys and will potentially be more compatible with textiles,
ensuring greater comfort for the wearer. The trigger temperature for activating the shape
memory polymer is close to body temperature.
11.1.2.6 Conductive Properties

Fabrics that incorporate electrodes or other sensing devices to ‘read’ signals from the body,
whether responding to the heartbeat, pressure, exposure to heat, motion, communications
etc., are being developed. Wires incorporated into a yarn, nickel, copper and silver coated
fibres or even yarns made of conductive polymers incorporating the conductive particles
within the fibres are bringing technology into clothing design for specific monitoring pur-
poses. (e.g. medical, military and safety professions) Conductive fibres braided into a shield
offer superior performance against electromagnetic interference. In 2000, the Nobel Prize
in Chemistry was awarded to Alan Heeger, Alan MacDiarmid and Hideki Shirakawa for
the discovery and development of conductive polymers.
Figure 11.2 Teflon coating.
11.1.2.7 Stain Repellent Properties

Coatings (e.g. Teflon®) or other surface protectants cover fibres and fabrics to prevent stains
from setting in by adding an extra protective resin layer (e.g. Scotchguard™) (Figure 11.2).
11.1.2.8 UV Protection Properties

UV blockers have been added as a coating to some fabrics as a sunblock (e.g. zinc) but these
often wash off after many uses. Marketing a garment as sun protective is allowed by the
FDA without much oversight; however ASTM 6603 and ASTM D6544 are more specific.
Generally the use of the term Ultraviolet Protection Factor (UPF) is merely a marketing
label for tightly woven, dark fabrics that block UV rays from penetrating the garment.
Denim fabric is one type of fabric that meets this basic definition; however, it is often too
hot to be worn in the hot summer months when UV protection is most sought nor is it
marketed for sun protection. Fabrics that mimic this construction in a lightweight garment
are able to be labelled as sun-protective, highlighting the benefits especially in locations
where sun protection is most sought.
11.1.2.9 Enhanced Properties via Plasma Treatment

Plasma treatment can be used to alter the surface to enhance the dye uptake properties
or to make the fibres hydrophobic, hydrophilic, dirt-repellant, shrink resistant etc. By
altering the chemical structure and the morphology of the surface, different properties can
be enhanced or achieved (Höcker, 2002). This treatment can be used both on natural fibres
and manufactured fibres.
11.1.3 Nanomaterials
Very often nature offers effects and systems which can be mimicked with the help of modern
design and engineering systems into new technical solutions. This method is called Bionics
or Biomimetics (Kesel, 2005; Blüchel, 2006; Passino, 2005). One of the most commonly
known effects is called the lotus effect. The lotus plant developed a surface system where
dirt particles are rinsed off by moving water droplets as a method of self cleaning. This
Figure 11.3 Scheme of distribution (e.g. of silver particles, different in size).
effect has been replicated in fabrics through nanomaterials. Other examples are the use of
nanoparticles in airbag coatings as a means to prevent friction or the application of silver
particles to enhance an antimicrobial effect. These particles are often applied to the surface
of a fibre or textile and distributed in different ways (Nehse, 2007).
In general nanomaterials can be used to enhance most of the before mentioned
properties (see Section 1.2). Some solutions are only possible through the use of nanoparticles
(e.g. implementation of phase change capsules). Nevertheless regarding solutions with
nanomaterials, much depends on (Figure 11.3). the compatibility with the fibre. Wet
chemical processes, mechanical processes, form-in-place processes and gas phase synthesis
are used. Via these methods a wide range of properties can be offered including conductivity,
magnetism, piezoelectric effects, colour changes, water repellence and anti-microbial effects
(Sargent, 2006; Ratner and Ratner, 2003). The processes control the surface chemistry and
nanomaterials can be incorporated into the fibres or bound to the surface including ultra fine
coatings. The particles can be of polymeric or metallic origin. Nanoparticles range from 1 to
100 nanometers (definitions may differ slightly between countries) and mainly are created to
add specific functions or modify the fibre functions. Different methods are guarded closely
by the companies and more and more solutions will be found in the market.
The durability of the particles, layers and coatings depends on the type of material, and
the kind of fixation or incorporation. This is closely observed by official organizations (e.g.
The Federal Institute for Risk Assessment [BfR] in Germany) which recommend manufac-
turers to refrain from using nanoparticles on textile surfaces until they are in a position to
reliably rule out potential health risks.
11.1.4 Green Materials

Green materials also termed ‘eco-friendly’ are produced from plant and algae products to
make cellulosic fibres. As natural polymers, they are biocompatible and biodegradable, they
should be non-toxic and there is a nearly unlimited supply of the raw materials; however, the
term eco-friendly does not necessarily mean that the production methods are also green.
Green manufacturing aims at reducing the amount of harsh chemicals used to produce the
fibres and the amount of air pollution that is subsequently released during the production.
Some green production methods may still be chemically intense processes including the use
of dyestuffs and additives. Recycling of those chemicals may reduce the overall amount of
the chemicals that could potentially enter the environment as waste. One caution for cellu-
losic and protein based fibres is that they are often microscopically indistinguishable despite
their different basic constituents (e.g. bamboo rayon vs. cupramonium rayon).
11.1.4.1 Cellulosic Fibres

11.1.4.1.1 Bamboo Makers of rayon fibres using bamboo as the basic cellulosic con-
stituent are now banned from marketing those fibres as bamboo fibres based on a 2009
complaint to the FTC. The FTC found that one company used deceptive marketing to
imply that their rayon fibres were more green than the actual manufacturing process was
because the fibres were made from a renewable resource like bamboo (Figure 11.4).
11.1.4.1.2 Alginate Alginate fibres are produced on the base of cellulose. As natural
polymers, they are biocompatible, biodegradable, nontoxic and nearly in unlimited supply.
Alginate fibres have a rich supply of trace elements which are skin friendly and protective
to the skin. The fibres are smooth and comfortable to wear and offer good physical char-
acteristics of the clothing material as breathability and moisture transport. With applied
silver particles the fibres even offer an additional antimicrobial effect (Figure 11.5).
11.1.4.2 Poly Lactic Acid (PLA) Fibres

11.1.4.2.1 Corn Corn is used as a source of dextrose, which is then fermented to form
lactic acid. The lactic acid is converted catalytically to form lactide monomer, which can
undergo polymerization to poly(lactic acids) (PLAs). These PLAs can be converted into
fibres via melt spinning. Corn fibre is relatively new on the market and fibres are mainly
produced in the United States and Japan. Corn fibres have a silky touch and luster. They
have a relatively low ability to absorb moisture (comparable to polyester fibre). The fibre
is tear proof, elastic and has a good UV resistance but a tendency to high shrinkage and
wrinkle. The fibres are used in clothes, geo- and agrotextiles and as a filling material in
pillows and bedcovers. Despite being a renewable resource and biodegradable, the use of a
food plant as a resource for fibre production is questioned.
11.1.4.3 Protein Fibres

11.1.4.3.1 Soy Fibres made from soy protein (SPF) were invented by chemists Robert A.
Boyer and Frank Calvert in the 1930s and introduced to the public by Henry Ford but they
did not have a commercial breakthrough. In 1998, soy fibres were reengineered. These fibres
are often compared to silk in their colour, feel and luster. The cross sectional shape appears
similar to a dogbone or a flattened bilobal. Soy fibres are able to be dyed with acid, neutral and
Figure 11.4 Cellulosic fibre produced from bamboo.

20 μm Mag = 412 X WD = 14 mm Signal A = QBSD Date :8 Mar 2006

EHT = 20.00 kV Photo No. = 224 Time :16:03:17
Figure 11.5 Sea CellR active with silver particles embedded in fibre.
reactive dyes and have good colourfastness. Manufacturers claim the moisture absorbency
is high as well as the ability of soy protein fibres to transmit moisture or ventilate the fabrics
(Li, 2004). Additionally, fabrics made with soy fibres are heat resistant so they will not shrink
much, have a high mechanical strength so they do not break easily and have a good resistance
to sunlight damage making them good UV protectors (Figures 11.6 and 11.7) (Li, 2004).
11.1.4.3.2 Casein/ Milk Fibres made from milk protein were invented in the 1930s but
the fibres now manufactured from casein protein are stronger than those original fibres.
These fibres can be dyed by acid, reactive or cationic dyes and have good colourfastness.
They are lustrous and strong and are often compared to silk in feel (Figure 11.8).
11.1.5 Composite and Bicomponent Fibres

Composite fibres are fibres that have particles dispersed within the polymer matrix that
can enhance the properties of the fibre while retaining the feel and flexibility of traditional
textile fibres. They differ from bicomponent fibres which are a mixture of two different
polymer materials that are extruded together from a spinnerette.
Figure 11.6 PLA fibre from corn.

Figure 11.7 Botanic protein fibre from soy.
Figure 11.8 Cross sectional shape encountered in casein fibres.
Bi-component fibres also offer enhanced properties which don’t exist in a single poly-
mer. The polymers are transported in a molten state in separate extruder lines and combined
when they pass the spinneret. Multiple new cross sectional shapes or so-called formations
can be produced. The combination of different polymers allows the development of fibres
with completely new features. The cross sectional shape may give some indication regard-
ing the use of these fibres in specific products.
11.1.6 Recycled Material

11.1.6.1 Pre-Consumer Textile Waste Material
This material is recovered from textile industries and fibre manufacturing plants as
a by-product from production. The fibrous materials are used for different purposes
(e.g. insulation for automotives, upholstery for furniture and mattresses, coarse yarns) and
are often less mechanically affected than post consumer waste material.
11.1.6.2 Post-Consumer Textile Waste Material
11.1.6.2.1 Textile Materials Used in a Different Context There is a new design philosophy
developing in Europe and North America. Parts of used heavy duty textiles are converted
into new items. Old truck tarpaulins combined with used airbags, bicycle inner tubes and
old seat belts are given a fashionable new life as messenger bags, badgeholders or buckles, etc.
11.1.6.2.2 Textile Materials Used in Textile Context Fibres or yarns are being created
from recycled textiles. When waste portions of carpets and textiles are sold to recycling
plants, they may end up in rags, cushion fillings, blended into inexpensive non-woven
blankets or melted down into a polymer to be made into new fibres or molded plastics.
Insulation fillings and upholstery may also include recycled fibres. Cordage that includes
recycled fibres may change over the length of the cordage. Recycled fibres may have different
properties than the original product if they are over-dyed, crushed, melted or otherwise
retextured and may also be encountered in clothes (e.g. wool garments). Effects like an
extremely high variability of fibres in colour and shade in very dark to black garments and
mechanically damaged fibres may point to recycling.
This is especially important to recognize in cases when an analyst is requested to
provide possible product source information for investigative leads.
11.1.6.2.3 Carpet Fibres Carpet Recycling Europe (CRE) was founded in 1998 as a pilot
project. The CRE plant ‘was able to sort up to 3.5 tons of carpet per hour into different
mono-fractions’. The pilot plant was closed by the end of 2002 since the economic condi-
tions were not in favour of carpet recycling at that time. Another plant was closed in 2003.
Despite this set back the acquired know-how is being used to start up new co-operations
with the European authorities.
In North America, the Carpet America Recovery Effort (CARE) began in 2002. This
was an attempt to reduce the amount of carpeting ending up in landfills as well as redirect
some of the waste products from the production process back into manufacturing.
While much has been done in the industry to reduce the amount of waste in production
and to reclaim old carpets for new materials, the effect of the recycling on the forensic sig-
nificance of these fibres must be taken into account. Non-uniform carpet backings and
junk fibre assemblages may be a result of recycling processes.
11.1.6.3 Other Post-Consumer Waste Material

Fibres are being created from other recycled materials (e.g. soda bottles from PET), Soda
bottles are being used as a secondary resource, shredded, melted down, sometimes mixed
with new PET pellets and spun into fibre filaments. These fibres offer the same features as
completely new material only with the exception that coloured PET bottles have to be used
in dark or black fibres material.
11.1.7 Security Measures

11.1.7.1 Production Tracking
Production tracking is a means to prevent fraud and pirating of brand products. This
technique may also offer some added value for forensics. The possibility to prove the
authenticity of a textile may offer some new aspects with textile intelligence work.
Tracer codes are unique codes and can be found within synthetic fibres. They can
be supplied as staple or filaments fibres. These fibres are blended into the surrounding
materials (e.g. fabrics, sewing yarns, labels) in a certain ration (e.g. 1:3000 with Verify TT™)
so that they cannot be detected with the unaided eye. These ratios are also necessary to
prevent any influence on the textiles regular properties. Tracers can be used in garments,
bags, backpacks, shoes, carpets and other products.
Instant detection at any point in the chain of production is possible by using an elec-
tronic reader.
11.1.7.2 Body Armour/Protective Textiles (e.g. Zylon, Kevlar, Nomex)

Protective clothes can be found in numerous areas. They are used in certain fields of
occupation (e.g. police, firefighters), sports (e.g. ice hockey, football), free time (e.g.
motorcycling). The goal is to prevent, e.g. chemical, thermal and mechanical influ-
ences to affect the human body. Fibre types like Zylon™, Kevlar® and Nomex® are very
well in use in this area. Zylon [poly(p-phenylene-2,6-benzobisoxazole) is a synthetic
fibre produced by Toyobo in Japan. The fibre offers a high tear strength, which is fac-
tor 2 of aramid, in combination with a better elasticity. With a melting point of 650 °C
and a necessary oxygen content of 68% it is nearly incombustible. The material is also
far lighter then aramid (e.g. Kevlar® and Nomex®). A problem occurs with moisture.
Body armour ages and the lifespan warranty decreased, so that Zylon was decertified
for use in its originally approved ballistic vests for law enforcement organizations in
the United States and in Germany. This was due to U.S. and European government’s
activities and advice. Aramid fibres like Kevlar® and Nomex ® are heavier in weight but
very effectively used in bicycle tires, racing sails, constructive ropes for bridges and
body armor. This is due to the fact that Kevlar® is still ‘Five times stronger than steel
on an equal weight basis . . .’.
Nomex® is also an aramid fibre whereas it is more in use for thermal protection because
of its good flame resistance. Para- (e.g. Kevlar®) and meta-aramid fibres (e.g. Nomex®) can
already be differentiated with the help of UV/VIS microspectrophotometry, since the
para- and meta-aramid fibres allow the transmission of light starting at different wave-
lengths (Figure 11.9) (Kueck, 2008).
11.1.8 Fibre Updates

11.1.8.1 PolyTrimethylene Terephthalate (PTT – Triexta,
Subclass of Polyester)
In 2008 ‘triexta’ was officially recognized by the FTC as a new generic subclass of polyester
for fibres made with PTT (Figure 11.10).
According to Mohawk Industries, which sought the separate definition under FTC
rules, ‘the PTT fibre, while having the general chemical composition of polyester, has
distinctive features related to durability, resilience, softness, and ability to stretch with
recovery’ that made it different enough to enable a subclass recognition.
11.1.8.2 PolyLactic Acid (PLA)

In 2002 PLA was officially designated as a new generic fibre type by the FTC. This is a
synthetic fibre; however, made with naturally produced sugars such as the dextrose from
vegetable starches or corn (see also Section 11.1.4.2.1); which is seen as a renewable resource.
100.00
95.00 Nomex®
Kevlar®
90.00
85.00
80.00
75.00
70.00
65.00
60.00
Transmission %T
55.00
50.00
45.00
40.00
35.00
30.00
25.00
20.00
15.00
10.00
5.00
0.00
275.0 300.0320.0340.0360.0380.0400.0420.0440.0460.0480.0500.0520.0540.0560.0 580.0600.0620.0640.0660.0680.0700.0 720.0740.0760.0
Wavelength [nm]
Figure 11.9 UV/VIS Spectra from PTT and PET fibres.
1356
PTT
933
1326
1383 1036
PET
1370
1040
4000.0 3000 2000 1500 1000 650.0

cm–1
Figure 11.10 FTIR Spectra from PTT and PET fibres.

Keeping current with the innovations and developments that are happening in the
fibre industry is not always possible for caseworking forensic examiners; however, presen-
tations, literature and industry information sheets can serve to notify fibre examiners of
what to expect in future casework and if any of these innovations may be apparent under a
microscope thus providing further characteristics for comparison.
References
Blüchel, K. G. and Malik, F., 2006. Faszination Bionik, Die Intelligenz der Schöpfung. München:
MCB-VERLAG.
Höcker, H., 2002. Plasma treatment of textile fibers, Pure Appl. Chem., 74(3), 423–427.
Hu, J., 2007. Shape Memory Polymers and Textiles. Abington Hall, UK: Woodhead Publishing Limited.
Kesel, A. B., 2005. Bionik, Wie kann Technik von der Natur lernen. Frankfurt: Fischer Verlag.
Koch, S., 2007. Beyond cotton, nylon and polyester: The emergence of new fibres and yarns. Global
Forensic Science Today, Issue 1.
Kueck, A., 2008. Ansatz zur Materialdifferenzierung mittels UV/VIS-Spektroskopie—Approach
to material differentiation via UV/VIS spectroscopie. Proceedings of 17th Symposium Textile
Spuren—Textile Traces. Mainz.
Li, Y-Y., 2004. The soybean protein fibre—A healthy and comfortable fibre for the 21st century, Fibres
& Textiles in Eastern Europe 12, Issue (2, 46), 8–9.
Matilla, H. (Ed.), 2006. Intelligent Textiles and Clothing. Abington Hall, UK: Woodhead Publishing
Limited.
Nehse, K., 2007. New developments in textile industries. Proceedings of 15th European Fibres Group
Meeting, Madrid, Spain.
Passino, K. M., 2005. Biomimicry for Optimization, Control and Automation. London, UK: Springer.
Ratner, M. A. and Ratner, D., 2003. Nanotechnology: A Gentle Introduction to the Next Big Idea.
Upper Saddle River, NJ: Prentice Hall PTR.
Sargent, T., 2006. The Dance of Molecules: How Nanotechnology Is Changing Our Lives. New York, NY:
Thunder’s Mouth Press.
Online References
http://ftc.gov/opa/2009/10/bamboosa.shtm
http://www.3m.com/market/industrial/ceramics/materials/nextel.html
http://www.unifrax.com
http://www.ceramicfiber.net
http://www.ftc.gov/opa/2008/04/text.shtm
http://www.textronicsinc.com/
http://www.outlast.com/index.php?id=95&L=0
http://www.schoeller-textiles.com/en/technologies/c-change
http://www.swicofil.com/soybeanproteinfiber.html
http://euroflax.com/products_imports%20of_textiles.htm
http://www.carpetrecovery.org
https://www.freitag.ch/en/fabric%20
http://www.sigmaaldrich.com/materials-science.html
http://www.policeone.com/Zylon/
http://www2.dupont.com/Kevlar/en_US/
http://www2.dupont.com/Nomex/en_US/
11.2 Emerging Technologies and Management Considerations
MICHAEL GRIEVE
MAX M HOUCK
JAMES ROBERTSON
11.2.1 Introduction
Microscopical examination of trace evidence is a traditional part of forensic science and
the use of fibres as evidence has been steadily developing over the last 50 years. Increased
awareness on the part of police, attorneys, and the courts has lead to an improvement
in collection, analysis, and interpretation of fibre evidence. Fibres have played a role in
assisting investigations, obtaining convictions, and exonerating individuals in major cases
reported in the national press in different countries, bringing this form of trace evidence
to the attention of public as well as professional stakeholders. Fibres, and many other types
of trace evidence, have the advantage of being personal, intimate, and understandable in
our daily lives—it is not just a blob on an electrophoretic gel – and this helps the jurors
appreciate the evidence, sparking their imagination, and making an impression. Fibres are
a canonical form of polymeric forensic evidence (Causin, 2010).
Every new edition of this book has seen advances in fibre technology, methods, and
resources. International groups of specialists have established guidelines for fibre train-
ing and analysis: In Europe, the European Fibres Group (EFG), in the United States, the
Fibre Subgroup of the Scientific Working Group on Materials Analysis (SWGMAT), and
in Australia, the Chemical Criminalistics Scientific Advisory Group (SAG) have purview
over fibre resources, many of which have been published or are otherwise available on the
Internet. These groups have solidified the foundation for fibre examinations and provide
impetus to ensure a quality service incorporating the latest techniques and understand-
ing. The objectives of the groups are to provide a means to facilitate a rapid and open
exchange of information throughout the forensic community, establish guidelines for fibre
examination and work toward international standardization, and to promote collaborative
research. The groups meet annually and these meetings include presentations, reviews,
case reports and document generation. SWGMAT produced the first formal set of guide-
lines and the other groups have followed that lead, adjusting the documents for analytical
and jurisdictional needs. Proficiency or competency tests are available from some of the
groups, which are offered annually. In addition, the National Institute of Justice supported
three international symposia on Trace Evidence from 2007 to 2011, all of which included
fibres as a central topic. Together, these professional working groups offer a wealth of
resources, documentation and collegiality for the forensic fibre examiner.
Less positive for the fibre examiner has been the incessant drumbeat for DNA as
the so-called gold standard in forensic science. This has driven forensic laboratories to
drive resources to DNA and away from other still viable forensic methods, including
fibre examinations. The DNA myopia of policy makers, managers and the courts shifted
attention away from the ‘other 95% of the laboratory’ and has, it can be argued, weakened
the comprehensiveness of forensic services offered. The promotion of any one discipline
at the expense of another results in a lack of adequate space and equipment, insufficient
personnel and improper training, all of which contribute to lower quality performance.
More fundamentally, it is also contrary to the holistic concept of ‘trace’ whose informa-
tional content should be exploited according to the hypotheses being investigated in a
case and not according to predetermined models*. The consequences can be disastrous,
as exemplified in the Proceedings of the Morin Commission of Enquiry in Canada (1998)
among many other examples of forensic malfeasance or misconduct. Such episodes result
in a loss of trust and confidence in forensic science and the reputation of the offending
laboratory can be quickly destroyed. Repairing that damage can take years, if it can be
fixed at all.
The lesson here is that it is dangerous for laboratories to remain in the field of fibre
examination if their management is not willing or able to provide the necessary resources
in terms of personnel, equipment and facilities. Fibre examination is not a trivial or
part-time pursuit and it should not have to depend on some catastrophe to occur before
management ensures that the highest standards are being applied to fibre examinations
carried out in their organization. Nevertheless, it should also not be discarded because of
perceived weaknesses or cost issues; fibres can provide excellent evidence (Houck, 2001,
2003a, 2003b) and be examined cost-effectively with a proper process of triage, intelligence
and methodology.
Considerations about the future of forensic fibre examinations can be divided into
six areas:
• Improving analytical capability

• Colour analysis
• Case management considerations
• Improving the capability to interpret examination results
• International cooperation
• Manufacturing information
• Improving standards
• Use of textiles and fibres as intelligence in investigations
Each of these considerations will be taken up individually.
11.2.2 Improving Analytical Capability

The existing methods used for fibre examination and comparison outlined in detail in this
book and elsewhere are without doubt highly discriminatory. Working groups around the
world have published analytical guidelines for operational laboratories and these are more
than suitable for the bench analyst. Training guidelines also exist to bring new fibre exam-
iners up to competency for casework (for example, the SWGMAT fiber training guidelines
are quite extensive). Methods used to examine fibres should also be
• Applicable to very small sample size

• Non-destructive
• Economically viable, in dollars and personnel hours
* It is interesting to note that Edmond Locard, himself, identified the risk of focusing on one type of trace
(fingerprints at that stage) at the cost of another; Locard, E. (1960) ‘Entretiens avec Edmond Locard’ by L.
Baissat, http://www.ina.fr/video/CPF86607035/le-professeur-edmond-locard-criminalogiste-video.html.
The role of the discrimination power of morphological differences and the methods
used to examine them has been less well documented, although it is known that the
discriminating power of bright-field and comparison microscopy alone is 0.97 (Adolf, 1993).
Despite the fact that the methods may be highly discriminating, the extent of inter-sample
and intra-sample variation has not been sufficiently researched to make interpretation on
a casework basis standardized. The polymeric morphology and structure of fibres has not
been as extensively exploited as it could be and this could open new avenues of specificity
(Causin, 2010). As a blue-ocean topic, prospective researchers could do far worse than
explore textiles in a forensic perspective.
This raises the question, however: Is there any need to introduce new methods unless
they are demonstrably superior to those that they replace or augment? To replace existing
techniques, the old adage of ‘twice’ should be invoked: The new method will have to be
twice as fast, twice as accurate or twice as cheap to replace current methods. In addition,
the introduction of any new method carries with it a significant cost in meeting validation
and quality assurance requirements. In the opinion of the authors, the core of existing
analytical methods for forensic fibre identification and comparison are now mature and
will continue to be used well into this century. It is important to realize that within the
pool of tried and tested methods, there are those which may be particularly suited to
certain circumstances and samples. It is not necessary to use all these methods in every
examination but it is important that the discriminating power of the existing methods be
emphasized through additional research.
The research needed in forensic fibre examinations, therefore, must be more aspirational
than mere methodology. Research is needed in:
• Specificity of sourcing
• Interpreting the wide variances of textile science, applications and products
• Modelling the textile environment given its wildly changing landscape, content
and pace*
• Placing these results in a context that affords validation, connection with related
science and clear communication to non-scientists
Alternate areas of research in related sciences, such as epidemiology or social

networking (Watts and Strogatz, 1998), may offer assistance in framing new methods of
understanding textiles and fibres for forensic analysis and interpretation. Epidemiology,
like forensic science and other disciplines, deals in transfers. The analogy could be made
from host-disease-infected to source-transfer-persistence (see Webb et al., 2005 for a
primer). Other transactional disciplines, like economics (Merton and Bodie, 1995), also
might offer insights.
Our message for analytical capabilities is twofold. First, perhaps, there needs to be an
increased focus on the textile rather than the fibre and, second, whilst we should keep an
open mind about new analytical techniques applied to fibres, the existing core methodolo-
gies are likely to outlast this edition of the Forensic Examination of Fibres.
* For example, Zara, a Spanish clothing manufacturer and retailer with a large number of outlets world-
wide, can take a clothing product from design table to retail store shelf in just two weeks Tiplady, R.
(2006). Zara: Taking the lead in fast-fashion. Bloomberg Businessweek. http://www.businessweek.com/
globalbiz/content/apr2006/gb20060404_167078.htm, Bloomberg.
11.2.3 Colour Analysis

Colour analysis is critical to many forensic analyses although forensic scientists are often
woefully under-educated on colour, colour analysis and the applications of colour to the
materials they analyse (Thornton, 1997). Understanding colour vision and how it works
(and deceives) is crucial to good textile colour analysis (Kuehni, 2006).
Given the complexity of modern product supply chains (Swamidass, 2000) and the
overall complexity of the textile industry’s network of fibre, yarn and fabric producers,
dyers and finishers (Nakamura, 2000) not to mention post-production distribution, both in
retail and after-market uses (for example, see Rivoli, 2009), colour is a key discriminator of
fibres and should be made more central to the analysis of fibre evidence. Furthermore, the
specificity of dye chemistry and colour choice to consumer products provides even greater
discrimination of potential sources for fibres. For example, direct dyes are typically used
on cellulosic fibres in products requiring good lightfastness but not good washfastness,
like window coverings, upholstery, or heavy bedding where the textiles are rarely washed
or are dry cleaned; direct dyes are good for light to medium dark colour range (Esche,
2004). This knowledge leads the fibre examiner to a different set of analytical properties
and interpretations, directly affecting the specificity of the evidence (SWGMAT, 1999a;
Goodpaster, 2009).
Beyond the chemical aspects of the fibre and dye combination, many perceptual,
fashion and social factors go into colour production, including colour specifiers, colour
laboratories, the actual dyeing and finishing, colour quality control and sorting and cutting
and sewing (Azoulay, 2005). Post-purchase use and wear can also affect colours through
staining, washing, as well as routine and irregular use (Hatch, 1993). Sampling is an issue
here and the fibre examiner is encouraged to better understand how sampling can affect
their colour analysis (Levy and Lemeshow, 1991; SWGMAT, 1999b).
Currently, Raman spectroscopy holds much promise in fibre analysis, but also with
the spectroscopy of dyes in situ (Massonnet et al., 2005). Raman microprobe spectroscopy
can characterize the dyes in natural and synthetic fibres, providing information on dye
states, dye concentrations (in cotton), and can discriminate between synthetic fibres of
the same polymer type from different manufacturers (Kokot et al., 1997; Keen et al., 1998;
Jochem and Lehnert, 2002); even fibre additives, such as delustrants, can be analysed (Vann
et al., 2009). In addition to the chapter in the current volume, Goodpaster’s recent review
of textile dye analysis is an excellent overview of the topic (Goodpaster, 2009) and is highly
recommended.
11.2.4 Case Management

Despite the domination of DNA analysis in forensic science, there will always be cases
where there is no DNA evidence and where fibre transfer is crucial, such as:
• Homicides, in which fibres on a body (naked or clothed) originated from the last
contacts and may provide useful investigative leads
• Cases of sexual assault where no body fluids are shed or transferred (because of a
condom, for example) and no consent is stated or implied
• Cases involving vehicles, for example hit-and-run accidents or cases where it is
necessary to establish who was driving the vehicle at the time of an accident
As Robertson and Roux (2010) have pointed out, good trace evidence work starts at the
crime scene. If the investigating police and/or the crime scene examiner fail to recognise the
potential for fibre trace materials to be present they will not be recorded and they will not be
recovered. Crime scene examination is still largely the responsibility of law enforcement agen-
cies. Putting aside the issues of appropriate background, qualification and training of crime
scene examiners, the future will see a move of what were formerly laboratory-based activi-
ties to the field, or, at least, the responsibility of the investigating agency. Case management
and triage/evidence selection will increasingly be done in the police domain. Unless there is
improved awareness of the value of fibre evidence provided in the training of crime scene and
evidence technicians, the future of fibres could be bleak (Robertson and Roux, 2010).
Fibre laboratory examinations can be time-consuming and, in cases where the police
must pay for the analysis, therefore expensive. Lack of understanding of the evidential
potential has possibly contributed to a situation where they may be felt to represent poor
value, either investigatively or financially. More effective case management can help to
remedy this situation, as discussed by Grieve (Grieve, 1996) who provided the following
the following suggestions:
• Reduce the time taken for analysis by using new equipment.

• Eliminate examinations where it is obvious from the outset that any findings will
have little or no evidential value.
• Screen exhibits to concentrate on searching for target fibers that are easiest
to identify and recover (strong colours, for example), which will have better
discriminatory value, which have been transferred to the most probative locations.
• Use the analytical methods that have the highest discriminating power appropriate
to the samples in question.
• Limit the number of fibres recovered and reduce the number of examples on
which confirmatory tests are run; statistical justification for this course of action
is required but easily mounted.
It is also important for fibre examiners to educate stakeholders so they appreciate the
value fibres can have as an investigative tool as opposed to evidence in court. A typical
example would be, in the absence of comparison material, the identification of the type of
garment (including end-use) to search for at the early stages of the investigation. Palmer
(2009) presented such a case (The Suffolk Serial Killings); a case that also contains many
elements of a successful case management*.
Finally, equally important in improving the understanding of the potential value of
fibre evidence will be to engage with social science researchers to measure the effectiveness
of forensic science and the role of fibre work. One such study is the FORESIGHT Project†, a
business benchmarking study that provides forensic service providers with data to allocate
their resources, streamline costs and improve effectiveness (Houck et al., 2009). Despite
aspirations ‘justice for all’, public services, like forensic science, have costs associated with
them and are not free. The allocation of resources – forensic examinations – is key to
providing the best science to a criminal justice system (Speaker, 2009a, b). Trace evidence
* NIJ Trace Evidence Symposium, Clearwater Beach, 2-7 August 2009, http://projects.nfstc.org/trace/2009/
presentations/5-palmer-ipswich.pdf
† www.be.wvu.edu/forensic/foresight
examinations, including fibre exams, are costly and personnel-intensive; ignoring or get-
ting rid of them altogether may be more expensive in the long run. Another such example
is a study being carried out in Australia (Julian and Kelty, 2009) looking at the down-
stream value of various forensic evidence. Working groups will continue to be an impor-
tant source of information and collaboration in the future of fibre examination. It is the
authors’ contention that they will become more important and eventually a reliable fixture
as a resource to the forensic science community in the next decade.
11.2.5 Interpretation
The amount of information and literature to assist the fibre analyst with the interpretation
of fibre results has increased enormously over the last few decades. However, almost
without exception, all examiners polled in a survey on the evidential value of fibres and
on the interpretation of the findings in fibre transfer cases said that they would welcome
more workshops or seminars on the topic (Grieve, 1998). This is probably due to the
inherent complexity of fibers and textiles, first and foremost, as a designed and manu-
factured material. A product must be designed in restrictive terms that define the con-
straints under which the designer must work. Each additional constraint complicates the
design process. Parameters are prioritized and become the arbiters for trade-offs, and
economics. It is easier to design a simple product, like a drinking glass, than one that
is more complicated, like an automobile (Coates, 2003). The challenges with a material
even as deceptively simple as cotton fabric are daunting; even pre-industrial dyers faced
many constraints:
To cheat nature, dyers had to create a bridge between substances that naturally repelled one
another. The first step . . . was to repeatedly soak cotton in solutions that broke down its
waxy structure. Bleaching with lemon juice or sour milk helped convert the grayish raw crop
to white fibers . . . dyers learned to then ‘animalize’ their cotton fiber [with] buffalo milk,
goat urine, and camel dung, and sometimes blood and albumen. The proteins in the animal
excretions did the bulk of the work in making the fiber receptive to the dyes. Next came the
mordants . . . faintly colored metallic salts obtained from natural sources, most commonly
alum and iron . . . The only important exception was the popular blue indigo, a pigment that
chemically fixes to cotton fiber around its core and needs no chemical catalyst.
(Yafa, 2005, p. 28)
Modern textiles are more complex and, accordingly, require greater study and testing
to characterize them. Interpretation of fibres, in terms of frequency and location, will only
become harder to characterize with current methods of statistical patterning and modelling.
And the variables of transfer and persistence and the statistics or probabilities multiply
the complexities of an already complex material. The variables to be tested (and controlled
for) are many, with potentially wide ranges and multiple effects. Live tests are often used
for determination of fibre transfer potential or sheddability. The results of these tests can
often influence interpretation of casework findings. As discussed elsewhere in this book,
further research is needed to improve the existing test procedure or to develop a simple and
practical alternative that would take into account more of the variables involved. It would
be extremely beneficial to have a standard system to assess sheddability; a preliminary
attempt was offered by Walbridge (Walbridge, 2003) and more recently De Wael et al (De
Wael et al., 2010) also addressed the issue.
With reference to fibre frequency databasing, it should be feasible to factor in spectral

frequencies to take the influence of colour into account. The original fibre data base work
of Laing et al. (1987) was checked for relevancy in 1998 and was found to still be viable
in that year (Adams et al. 1998). Fibre data such as this could be combined with spectral
colour data and information from other textile data base information, such as that from
clothing catalogues (Biermann and Grieve, 1996a, 1996b, 1996c) or similar. It must be
realized that there are certain problems with databases, including the following:
• They do not take into account fibre product.

• The validity of the data is limited to geographical, temporal and seasonal samplings.
• The sampling must be shown to be representative and relevant to items dealt with
in forensic casework.
A policy of centralization has often been suggested for trace evidence analysis. All
samples would be sent to a specialist laboratory that contains the latest and best analytical
equipment and highly trained scientists. The potential drawback is that although the
analysis may be excellent, unless experienced specialists are present to triage the original
submission, the ability to assess different aspects of the case and to make the appropri-
ate decisions may be lacking in those less well trained in that trace evidence specialty.
Nonetheless, the authors’ view that such centres of specialization may be the inevitable
consequence of the diminishing capacity or unwillingness of many laboratories to provide
fibre sections with adequate resources. This will almost certainly be the outcome in many
cases as the levels of acceptable standards become raised through accreditation processes.
Trace evidence is an expensive type of analysis, with high capital overhead and personnel
hours, and centralization of resources may well help keep forensic investigations becoming
a one-DNA-trick pony. While there are dangers in going down this path, it can be argued
that consolidation is preferable to laboratories continuing to do fibre examinations in an
unsatisfactory and unsupported manner.
It is vital that fibre examiners keep abreast of developments relating to fibre fre-
quency and evidence interpretation, and use the data and information which is now at
their disposal and, further, that they continue to build on this solid foundation. They
must be convinced of the value of their evidence, and must strive to communicate this
to police, legal personnel, and the public more effectively than has been done in the
past. This must include a serious attempt to make clear the exact relevance and worth of
the fibre evidence (or lack of it) within the context of any particular set of case circum-
stances. The use of a Bayesian statistical approach to evaluating evidence will remain a
topic of considerable interest. This is dealt with in Chapter 10 of this book. Suffice to say
that recent appeal court ruling in the UK* pointed out the need for ‘a sufficient database’
if an attempt was to be ‘realistically made in the generality of cases to use a formula to
calculate the probabilities’. This case related to a footwear database but the principles are
universal. If a Bayesian approach is to be used to assist in the interpretation of fibre find-
ings, serious efforts will need to be made to first define what databases are needed and,
second, to populate such databases.
* R v T [2010] EWCA Crim 2439

11.2.6 International Cooperation

In the so-called information age, symposia, conferences, meetings, and colleagues are still
significant resources for information and should not be avoided. Much of the output of
these physical meetings makes its way to the Internet and thus to the rest of the world. Some
organizations hold their meetings only every two years, such as the European Academy of
Forensic Sciences or the Australian–New Zealand Forensic Science Society, while others
are offered every three years, such as the Interpol Forensic Science Managers Symposium
or the International Association of Forensic Sciences. In the United States, the rule is
a meeting every year, such as the American Academy of Forensic Sciences. Special meet-
ings, like the FBI Trace Evidence Symposium, are held occasionally as funding allows.
The Internet has revolutionized information but not necessarily the way it is used. Even
the most savvy web searcher can be quickly overwhelmed by the amount of information
available with a simple search for ‘fibres’ or ‘textile science’*. Manufacturers are more easily
located today than ever before because of the Internet and this provides fibre examiners
with a greater appreciation for the design and production of their samples. Textile trade
shows are also a valuable source of contacts, information, and samples. Gaining the trust
of textile manufacturers can be difficult – textiles are a competitive industry (Singleton,
1997) – but a kind face with an official business card can open many doors.
Working groups in Australia, Europe, and the United States have contributed
significantly to the wealth of resources for the forensic fibre examiner, including:
• Providing dissemination of current information concerning fibre examinations

to a wider audience faster than was previously possible, even just a few years ago.
• Accumulating information relating to fibres examination by means of surveys
which identify problems requiring attention and highlighting topics where
research is necessary.
• Avoiding duplication of work and providing personnel for joint research activities
which would have been previously impossible. The time and effort required for
research, especially that needed to provide information relevant to evidence inter-
pretation, was previously beyond the resources of individual laboratories.
• Broadening the awareness of current literature and resources and assistance
related to current cases.
However, much of the work of specialist groups has remained relatively informal. It is
almost certain that there will be greater regulation of the forensic industry. The challenge
for specialist groups and managers will be to turn guidelines into standards and to ensure
these are international through international cooperation. For those of us who have expe-
rience of developing formal ISO standards this simple statement belies a complex, time
consuming and sometimes difficult journey.
* ‘Fibres’ yielded 34.2 million hits and ‘textile science’ gave 366,000 hits in a Google search on 17 May
2016
11.2.7 Manufacturing Information

A supply chain is the system of organizations, people, suppliers, intermediate proces-
sors, activities and resources involved in moving a product or service from supplier to
customer. It involves all of the direct and indirect stages in fulfilling a customer request.
Supply chains include the manufacturer, suppliers, transporters, warehouses, retailers
and even the customers themselves. Supply chain activities transform natural resources,
raw materials and components into a finished product that is delivered to an end cus-
tomer. From an economist’s point of view, supply chains link value chains (the processes
that a product passes through, gaining value at each step) in the production of that good
or service. Production is the conversion of one supply chain component into another,
such as raw materials into processed goods or those into finished goods. Inventory is
everything from raw materials to processed goods to finished goods held throughout the
supply chain by manufacturers, distributors and retailers. Knowledge of the fibre and
textile manufacturing industry is therefore central to a good understanding of fibres,
textiles and their meaning as evidence. The textile industry and its production meth-
ods – including supply chains – should be part of a forensic fibre examiner’s baseline
knowledge.
11.2.8 Improving Standards

The last 20 years have seen the increased development of quality systems and their
application to forensic providers. While some would argue we still have some way to go
in the path forward, (NAS, 2009) there can be little argument that overall standards are
better today than when the first edition of this book was written in the 1980s. However, our
industry should not and cannot be complacent. As stated earlier, more formal regulation
of forensic science providers will be a future reality. Of course, developing improved
standards that underpin quality systems and formal accreditation are in themselves to be
desired. However, a word of caution is needed. Many forensic scientists, and managers,
would argue that even our existing requirements are imposing an unacceptable burden
on the delivery of forensic work. This burden can be measured in many ways but it would
include time to process and analyse evidence, delays to investigations, cost, inhibiting
innovation and ultimately not effectively service the justice process.
The future challenge is to re-invent and re-engineer existing quality documentation
and processes to ensure a correct balance is achieved between risk, consequence and
likelihood (Robertson, Metz et al., 2010). As Butler (2006) comments, ‘fear of auditors
rather than scientific reasoning governs the collection of large numbers of data points . . .
thus, the applications of a new technology for solving cases more quickly can be delayed
because an overzealous number of validation experiments are performed’. As Robertson
and Roux (2010) have stated, the purpose of a good quality system is to drive improve-
ment through innovation, not to stifle innovation by overcooked risk management. The
key challenge for our industry is to evolve into a genuine profession where appropriate
exercise of professional judgment is a recognized and key characteristic. Fibre examiners
cannot expect to escape the calls for a slimmed-down quality system and the potentially
competing call for even more standards. Getting the balance right will require a sophisti-
cated understanding of what is a good standard and an acceptance of personal professional
responsibility (Robertson, 2011).
11.2.9 Use of Textiles and Fibres as Intelligence in Investigations

Searching the environment of a suspect, victim, or crime scene for extraneous fibres is
based on the well-supported propositions that:
• Textiles and fibres are widely varying due to design and manufacturing
requirements (Rivoli, 2008; Swamidass, 2000).
• People choose products based on design, seasonality, price, and color, among other
criteria (Underhill, 2008).
• Unless the things in question have had previous contact, it is rare to find analyti-
cally indistinguishable fibres in common (Houck, 2003b; Cantrell, et al., 2001).
The utility of fibres in leading investigations is well known and documented in the
forensic sciences (for example, see Palmer, 2003; Deedrick, 2001; and Cantrell, et al., 2001,
among many others) but not always utilized by law enforcement. Despite case after case
being aided in the forensic phase (post-event), the use of textiles and fibres in the investiga-
tory phase is employed less than it should.
11.2.10 Conclusions
It is interesting to look back on the comments made by Gaudette about ‘the future’ of fibre
examinations and to see how many of the suggested aims have been achieved (Gaudette,
1993):
• New methods for the recovery of fibre evidence have been developed.
• Quality control procedures have been improved.
• Studies have been made on the discriminating power of the analytical methods
used.
• New analytical techniques have been introduced.
• Considerable advances have been made in the investigation of fibre frequencies
and Populations and several further studies have been made on fibre transfer and
persistence in relation to the significance of fibre evidence.
All these things have served to put fibre evidence on an even firmer basis and to
strengthen its value. The benefits of the dramatic increase in international cooperation
can be clearly seen. There has been a flood of research publications concerning fibres since
the last edition of this volume. Not only has this helped focus attention on important
undertakings necessary for the future but it has also helped promote awareness of the
importance of fibres evidence and how it can be used. The Internet is now the main vehicle
that rapidly communicates standards, guidelines and communications about fibres.
For fibre examinations to continue to play a valuable and, indeed, expanding role in
commercial forensic science, laboratories have to be able to attract their customers (usually
the police) by offering a quality service which can be provided in a timely and effective man-
ner. Increasing efficiency in casework is only part of the key to future success. In addition,
the evaluation of positive results has to be such that the customers feel that they are getting
valuable information – that the effort of collecting and submitting the evidence has been
worthwhile. It is no good getting a quick result if it is of little value or if its significance
cannot be properly expressed. This means that experts must hone their ability to interpret
the evidential value of analytical results. Five things are necessary to achieve this:
• Fibre analysts must be encouraged and given time to build up their experience in
a specialist discipline that requires extreme patience and concentration.
• In the past, there has been a severe lack of adequate training on the subject of
evidence interpretation. This needs to be remedied by offering teaching work-
shops, suggested educational programs and continuing professional development.
• Management must be prepared to allow its analysts to take advantage of these.
• More data are needed on fibre production, manufacturing methods and qual-
ity tolerances to delineate batch-to-batch variation – the basis for evidential
significance.
• Guidelines must be converted to formal standards.
These aims can be realized and it is hoped that the necessary management support will
be forthcoming to ensure that this area of forensic science truly fulfils its potential.
Addendum: Mike Grieve (1942–2002)

Mike won the European Network of Forensic Science Institutes’ Distinguished Scientist
Award posthumously in 2003. His biography from the ENFSI website is reproduced below.
Mike Grieve was born in Buxton, Derbyshire, the son of a railway executive. Mike
obtained a BSc (Hons) degree from the University of Durham and went on to work at
the Metropolitan Police Forensic Science Laboratory (now the FSS London Laboratory)
between 1964 and 1967 where he started the fibre section. A short period of working for the
Pharmaceutical Society of Great Britain on the editorial staff of the Journal of Pharmacy
and Pharmacology followed between 1968 and1972. Quickly tiring of reading papers
about cannabis and amphetamines, he returned to forensic science, moving to the United
States Army Criminal Investigation laboratory in Frankfurt, Germany where he again
established a fibres section.
After obtaining a Graduate Certificate of Education in 1976 at the University of London,
he survived three years teaching chemistry and biology in a notorious south London
school, reported by a reputable national newspaper to be ‘one of the worst in Britain’. He
said he was saved by the intervention of an arsonist that incinerated the science block. Two
of the few good things that came out of his teaching career in south London was his love of
reggae music and his ability to lead, a quality that came to the fore in his later life.
Mike returned to Frankfurt in 1980 to run the fibre section and remained there until
1992. It was during this period that he learnt the intricacies of microspectrophotometry
and FTIR microscopy and was also awarded the United States Army Achievement Medal
for Civilian Service.
The Frankfurt laboratory was threatened with closure due to the changing politi-
cal situation in Eastern Europe and, in April 1992, Mike started working for yet another
Government. This time he joined the Forensic Science Institute of the Bundeskriminalamt
in Wiesbaden, Germany and it was in the fibre section where he carried out
case-applicable R&D.
It was in the latter part of his career in Frankfurt that Mike – together with Ken
Wiggins – felt the need to establish a European Fibre Group (EFG). The dream came to
fruition in November 1993 with the first meeting in The Netherlands with 21 scientists
attending. With Ken and Mike in the roles of Chairman and General Secretary respec-
tively, the group has flourished ever since. It was also in 1993 that Mike was invited to join
the Technical Working Group for Fibre Examination (TWGFIBE), this eventually became
the Scientific Working Group for Materials (SWGMAT).
Mike wasn’t content to raise the profile of fibres in Europe alone; he decided America
was also a target. They loved him, a typical English gent with a booming voice who got
things done by leading by example. He served on the committee for colour analysis
thus helping to complete their Forensic Fibre Examination Guidelines in 1998. In 2001
Mike was particularly pleased when the EFG produced The Manual of Best Practice for
Forensic Examination of Fibres. The document was probably only finished because of the
enthusiasm, drive and determination that Mike had to complete any task that he started.
The EFG and SWGMAT meant that Mike had even more opportunity to follow his
love of travelling. He has lectured across the length and breadth of the United States,
throughout Europe, in Asia and Australia. After the lectures and meetings had finished he
would always add on days of his holiday to see more of the country. If he couldn’t get to the
more exotic places through work he would take holidays to such places as China, Jamaica,
Trinidad and Tobago.
Mike spent most of his forensic career trying to draw attention to the potential of fibres
to provide excellent evidence in criminal cases. His particular interests were: everything
related to improving our knowledge about the evidential value of fibres and interpretation
of casework findings; working toward international recognition of minimum standards in
fibre examination; FTIR-microscopy (particularly of acrylic and modacrylic fibres); and
reference collections of fibre standards. Nothing ever got thrown away be it fibre samples or
information, he was a great believer in the phrase ‘you never know when you will need it’.
One of his main life goals was to achieve his target of having had 50 papers accepted
for publication; luckily he achieved this and more. It wasn’t a question of wanting to write
more papers than others in the field but simply because he had so much to say about his
work and wanted others to benefit from it. Probably his greatest legacy to fibre examiners
will be the book Forensic Examination of Fibres (second edition) which he co-edited with
James Robertson (ISBN 0-7484-0816-9, Taylor & Francis Ltd, 1999). It is already referred
to as the ‘Fibre Examiners Bible’ and acknowledged worldwide as one of the best, if not the
best, forensic fibre books ever produced. It was typical of Mike that only the best would do,
so he went out and recruited practising scientists who were acknowledged experts in their
field to write the individual chapters, so how could he fail (incidentally he wrote a lot of it
himself).
Mike was still working on a number of fibre-related projects right up to the day
he suddenly passed away in September 2002. Luckily his colleagues in Germany are
completing these outstanding projects; he would have hated all his effort to be wasted.
He had already once received the P.W. Allen award from the Forensic Science Society (in
1998) and just after his death it was announced that he had won it again in 2002 with his
colleague Thomas Biermann.
The most important improvement for forensic sciences is that Michael Grieve provided
a huge amount of basic information concerning the value of textile traces including, e.g.
frequencies of fibres, frequencies of fibre dyes and that of fibre type colour combinations.
Without his research fibres would not have such a high standard as trace evidence. Any
forensic scientist, from no matter what discipline, could learn from him.
Between 90 and 100 people attended his funeral which took place in Hanau, Germany
with European Fibre Group colleagues travelling from England, France and Switzerland
as well as from all corners of Germany. He was the best forensic fibre expert in the world
and was respected and liked by many. Two citations from colleagues to illustrate the feel-
ings for Mike:
‘Mike Grieve, a Golden Thread in the Tapestry of Our Lives.’ SWGMAT September 2002
‘Mike, The Best Colleague, Friend and Mentor We Had.’ EFG September 2002
References
Adolf, F. P., 1993. Discrimination between polyester fibres using thickness, delusterant and fluores-
cence. Internal Research Report. Wiesbaden, Bundeskriminalamt.
Azoulay, J., 2005. The devil in the details: The challenge of color, AATCC, 5(2), 9–13.
Biermann, T. and Grieve, M. C., 1996a. A computerized data base of mail order garments: A con-
Biermann, T. and Grieve, M.C., 1996b. A computerized data base of mail order garments: A contri-
bution towards estimating the frequency of fibre types found in clothing. Part 2: The content of
the data bank and its statistical evaluation, Forensic Sci. Int., 77, 75–91.
Biermann, T. and Grieve, M.C., 1996c. A computerized data base of mail order garments: A contri-
bution towards estimating the frequency of fibre types found in clothing. Part 3: The content of
the data bank—Is it representative? Forensic Sci. Int., 95, 117–131.
Butler, J. M., 2006. Debunking some urban legends surrounding validation within the forensic DNA
community, Promega Profiles in DNA, 3–6.
Cantrell, S., Roux, C., Maynard, P., and Roberston, J., 2001. A textile fibre survey as an aid to the
interpretation of fibre evidence in the Sydney region, Forensic Sci. Int., 123(1), 48–53.
Causin, V., 2010. Polymers on the crime scene: How can analytical chemistry help to exploit the
information from these mute witnesses? Anal. Methods, 2, 792–804.
Coates, D., 2003. Watches Tell More Than Time. New York, NY: McGraw-Hill.
De Wael, K., Lepot, L., Lunstroot, K., and F. Gason., 2010. Evaluation of the shedding potential of
textile materials, Sci. Justice, 50, 192–194.
Deedrick, D., 2001. Fabric processing and ‘nubs’. In M. Houck (Ed.), Mute Witnesses. Amsterdam:
Elsevier.
Esche, S., 2004. Direct dyes–their application and uses, AATCC, 4(10), 14–17.
Gaudette, B. D., 1993. Forensic fiber analysis. In R. Safetstein (Ed.), Forensic Science Handbook
(vol. 3). Englewood Cliffs, NJ: Prentice-Hall.
Goodpaster, J., 2009. Forensic analysis of dyed textile fibers, Anal. Bioanal. Chem. 394, 2009–2018.
Grieve, M. C., 1996. The perspective for forensic fibre examinations in the year 2000. Proceedings of
the FBI International Symposium on Forensic Examination of Trace Evidence in Transition, San
Antonio, TX.
Grieve, M. C., 1998. Survey on the evidential value of fibres and on the interpretation of the findings
in fibre transfer cases. 6th European Fibres Group Meeting, Dundee, Scotland.
Hatch, K., 1993. Textile Science. St. Paul, MN: West Publishing.
Houck, M. M., 2001. Mute Witnesses. San Diego, CA: Elsevier, Academic Press.
Houck, M. M., 2003a. Intercomparison of unrelated fiber evidence, Forensic Sci. Int., 135, 146–149.
Houck, M. M., 2003b. Trace Evidence Analysis. San Diego, CA: Elsevier, Academic Press.
Houck, M. M., Riley, R., Speaker, P. J., and T. S Witt., 2009. FORESIGHT: A business approach to
improving forensic science services, Forensic Sci Pol. Manag., 1, 85–95.
Jochem, G. and Lehnert, R., 2002. On the potential of Raman microscopy for the forensic analysis of
coloured textile fibres, Sci. Justice, 42(4), 215–221.
Julian, R. and Kelty, S., 2009. The effectiveness of forensic science in criminal investigations, Aust.
Policing, 1(2), 10–15.
Keen, I., White, G., and Fredericks, P., 1998. Characterization of fibers by Raman microprobe spec-
troscopy, J. Forensic Sci., 43, 82–89.
Kokot, S., Nguyen, A., and Rintoul, T., 1997. Discrimination of reactive dyes on cotton fabric by
Raman spectroscopy and chemometrics, Appl. Spectrosc., 51(3), 387–395.
Kuehni, R., 2006. Color vision: How and why? AATCC, 6(1), 27–30.
Levy, P. and Lemeshow, S., 1991. Sampling of Populations. New York, NY: John Wiley & Sons.
Massonnet, G., Buzzini, P., Jochem, G., Stauber, M., Coyle, T., Roux, C., Thomas, J., Leijenhorst, H.,
Van Zanten, Z., Wiggins, K., Russell, C., Chabli, S., and Rosengarten, A., 2005. Evaluation of
Raman spectroscopy for the analysis of colored fibers: A collaborative study, J. Forensic Sci.,
50(5), 1028–1038.
Merton, R. and Bodie, Z., 1995. A conceptual framework for analyzing the financial environment.
In D. Crane (Ed.), The Global Financial System (pp. 3–31). Cambridge, MA: Harvard Business
Press.
Nakamura, A., 2000. Fiber Science and Technology. Enfield, NH: Science Publishers.
NAS. 2009. Strengthening Forensic Science in the United States: A Path Forward. Washington, DC:
National Academies Press.
Palmer, R., 2003. An intelligence led investigation using trace evidence, In M. Houck (Ed.), Trace
Evidence Analysis: More Cases in Mute Witnesses. Amsterdam: Elsevier.
Rivoli, P., 2009. The Travels of a T-shirt in the Global Economy. New York, NY: John Wiley and Sons.
Robertson, J. and Roux, C., 2010. Trace evidence: Here today, gone tomorrow? Sci. Justice, 50, 18–22.
Robertson, J., 2011. Forensic science: An enabler or dis-enabler for criminal justice? Aust. J. Forensic
Sci., 43, in press.
Robertson, J., Metz, H., Scudder, N., and Hodgson, V., 2010. A quality system review: Australian
Federal Police and data centres, Forensic Sci. Policy Manag. 1, 209–213.
Singleton, J., 1997. World Textile Industry. New York, NY: Routledge.
Speaker, P., 2009a. The decomposition of return on investment for forensic laboratories, Forensic Sci.
Policy Manage., 1, 96–102.
Speaker, P., 2009b. Key performance indicators and managerial analysis for forensic laboratories,
Forensic Sci. Policy Manag. 1(1), 32–42.
Swamidass, P., 2000. Encyclopedia of Production and Manufacturing Management. New York, NY:
Springer.
SWGMAT, 1999a. Forensic fiber analysis, Forensic Scie. Commun., 1(1).
SWGMAT, 1999b. Trace evidence recovery guidelines, Forensic Sci. Commun., 1(3): Available at
www.fbi.gov.
Thornton, J., 1997. Visual color comparisons in forensic science, Forensic Sci. Rev., 9, 37–67.
Tiplady, R., 2006. Zara: Taking the lead in fast-fashion. Bloomberg Businessweek. http://www.busi-
nessweek.com/globalbiz/content/apr2006/gb20060404_167078.htm, Bloomberg.
Underhill, P., 2008. Why We Buy. New York, NY: Simon and Schuster.
Vann, B., Angel, S., Henrix, J., Bartick, E., and S. Morgan., 2009. Analysis of titanium dioxide in syn-
thetic fibers using Raman microspectroscopy, Appl. Spectrosc. 63(4), 407–411.
Walbridge, S., 2003. Determining the Sheddability of Various Fiber Types. School of Criminal Justice.
East Lansing, MI: Michigan State University. MS.
Watts, D. J. and Strogatz, S. H., 1998. Collective dynamics of ‘small-world’ networks, Nature 393,
440–442.
Webb, P., Bain, C. and Pirozzo, S., 2005. Essential Epidemiology: An Introduction for Students and
Health Professionals. Cambridge, UK: Cambridge University Press.
Yafa, S., 2005. Cotton: The Biography of a Revolutionary Fiber. New York, NY: Penguin.
Index
A B
Abrasion damage, 71 Background probabilities, 408, 410, 413
Acetylation, 25 Balanced plain weave fabrics, 43
Acid dyes, 225 Balance, reporting requirements, 364
Acrylic fibres, 34–35, 352 Bamboo fibres, 23–24, 434
Acrylonitrile, polymerisation of, 34 Bank security bags, 388
Addition polymers, 26, 27 Basic dyes, 226
Adhesive bonding, 54 Bast fibres, 7, 12, 150–152
AFID, see Alkali flame ionization detector Bayesian approach, 346, 347, 396, 447
Agava bourcroydes, 15 Bayesian networks (BNs), 417–421
Agave sisalana, 15 Bayes’ theorem, 396, 398
Airy disc, 257, 258 Beam polarization, 265–266
Alginate fibres, 26, 434, 435 Bi-component fibres, 38–39, 436
Alkali cellulose, 22 Blade irregularities, 77
Alkali flame ionization detector (AFID), 315 Blended materials, 430
Alkaline hydrolysis Blue denim cotton, 351
with sodium hydroxide, 28 Blunt implements, 77
of wool fibres, 233 BNs, see Bayesian networks
Alpaca fibre, 18 Body armour, 438
Aluminosilicates, 430 Boehmeria nivea, 13, 133, 150, 152
Amphibole, 20 Bombyx mori, 19, 153
Analytical capabilities, 442–443 Branched chain polymers, 27
Anaphe silk, 19 Bright-field microscopy, morphological features
Angle of lay, 90, 91 silk fibres, 153
Angora fibre, 19 synthetic and regenerated fibres, 153–157
Angstrom Sun Technologies Inc., 183 vegetable fibres, 149–152
Animal fibres, 131, 135–139 wool and fur fibres, 152–153
silk, 19–20
sub-classification of, 16 C
wool, 16–19
Anthophyllite, 20 Camel hair fibre, 18
Antibacterial properties, fibres with, 430 Camelus bactrianus, 18
Apertures, 269–270 Cannabis sativa, 14, 132, 150
Aramides, 36–37 Capillary electrophoresis (CE), 331–332
Aramid polymer fibres, 20 advantages, 338–339
AS 5388 standard, 82, 103, 362, 363 capillary zone electrophoresis, 333–334
Ashing, 93 CE-ESI-MS, 338
Assessment, types of, 62–63 disadvantages, 338–339
Atactic polymers, 27 for dye analysis, 337–338
Atactic polypropylene, 32 mass spectrometry, 335–336
Atlas of Textile Damage, 74 micellar electrokinetic capillary chromatography,
Attenuated total reflectance (ATR) spectroscopy, 334–335
271–274 principles, 332–333
Australian–New Zealand Forensic Science Society, 448 separation of analytes in, 335
Australian Standard 5338, 82, 103, 362, 363 Capillary zone electrophoresis (CZE), 333–334
Automobile carpet fibres, 377 Capra hircus laniger, 18
Auxiliary telescope, 191 Carbon fibres, 38, 269
Azoic dyes, 225–226 Carbonising process, 41
455
456 Index
Car carpet, 391 Comprehensive approach, 70

Carded yarns, 8 Condensation polymers, 26, 28
Carpet America Recovery Effort (CARE), 437 Conditional transfer probability, 415
Carpet fibres, 437 Conductive fibres, 431
Carpet Recycling Europe (CRE), 437 Conjugated focal planes, 190
Carpets, 389–391 Continuous filament yarns, 4
Casein fibres, 26, 435, 436 Continuous flow IRMS (CF-IRMS), 318
Case management, 444–446 Continuous-interference filters, 193
Casework data collections, 378 COOLMAX®, 29, 30
Cashmere fibre, 18 Copolymerisation, 34
Catalogue database (CDB), 374 Corchorus spp., 13, 152
CCC, see Complementary chromaticity coordinates Cordage manufacturers, 386, 387
CCD detectors, see Charged coupled device detectors Core-sheath process, 39
CDB, see Catalogue database Corn fibres, 434, 435
CE, see Capillary electrophoresis Corrected-grain leathers, 56
CE-ESI-MS, 338 Cortex layer, 16, 135
Celba pentandra, 11 Cotton fibres, 8
Cellulose, 7 dyes classification from, 231
in bamboo, 24 mature, immature and dead fibres, 10, 11
esters of, 25 scanning electron micrographs of, 9–10
in ramie, 13 Cotton twines, 93
Cellulose acetate, 25, 285 Cotton yarns, 40
Cellulosic fibres, 434 CRE, see Carpet Recycling Europe
Centralization policy, in trace evidence analysis, 447 Crêpe yarns, 41
Ceramic fibres, 37, 430 Crime scene
CF-IRMS, see Continuous flow IRMS case management considerations, 118–119
Chamois leather, 56 fibre persistence, 113–116
Charged coupled device (CCD) detectors, 183, 195 fibre transfer, 109–113
Chemical bonding, 54 first responder indicators, 64
Chemical Criminalistics Scientific Advisory Group interpreting textile damage, 65
(SAG), 441 microscopic comparison, 126–129
Chemical retting, 12 recovery of fibres, 116–118
Chemometrics, 279 Crime scene examiners (CSE), 101–102, 104, 109
Chlorine compounds, 7 Criminal justice system (CJS), 100
Chlorofibres, 35 Crimp, 17, 155
Christiansen Effect, 274 Crocidolite fibres, 20
Chrome mordant method, 226 Cross sectional shape
Chromic properties, textiles with, 430–431 casein fibres, 436
CIE L*a*b* (CIELAB), 217–218, 354 nylon fibres, 31, 33
CIE system, 216–219 refraction and, 255–256
CI number, see Colour index number Croydon carpet, 389–391
CJS, see Criminal justice system Crysotile fibres, 20
Coarser fibres, 11 CSE, see Crime scene examiners
Cocos nucifera, 152 ‘Cuban sisal’, 15
Coir fibre, 11 Cuprammonium hydroxide, 19
Colorimetry, 199, 216 Curie-point pyrolyzers, 312
Colour analysis, 444 CZE, see Capillary zone electrophoresis
Colour-coded materials, 95
Colour index (CI) number, 224–225
D
Combed yarns, 8
Combing/brushing, recovery of fibres, 106, 107 Dam retting, 12
Comparative Finished Fiber Analytical Database Dead cotton fibres, 10, 11
(COMFFAD), 379 Decomposition, 74
Comparison microscope, 148 Decorative knife, 79
Complementary chromaticity coordinates (CCC), Degree of crystallinity, 21, 280
219–220 Degree of polymerisation (DP), 7, 22, 25
Composite fibres, 435 De-gumming, 13, 19
Index 457
Denim fabric, 432 Electrospray ionisation (ESI), 335

Density of fabric, 49, 95, 163 Elemental analysis of fibres
Depth of penetration, 271, 272 inductively coupled plasma techniques, 324–325
Detailed examinations, 73–74 inorganic constituents, 322–323
Detailed photography, 82 neutron activation analysis, 324
Detectors X-ray diffraction, 324
in HPLC system, 330 X-ray fluorescence techniques, 325–326
in microspectrometers, 248–249 ENFSI, see European Network of Forensic Science
in pyrolysis technique, 315 Institutes
Deuterated triglycine sulphate (DTGS) detector, 248 EOF, see Electro-osmotic flow
Dew retting, 12 Epicutical region, wool fibre, 16
Diamond cell, 267–269 EPMA, see Electron probe microanalysis
Dichroism, 169–172 ‘Erredieci’, 388
Differential shedding, 110, 360 ESI, see Electrospray ionisation
Diffraction, photometric inaccuracy factor Esters, of cellulose, 25–26
definition of, 257 ETHG, see European Textile and Hair Group
flattened Enkron fibre, 262 European Academy of Forensic Sciences, 448
at high contrast edge, 259 European Fibres Group (EFG), 184, 208, 295, 385,
point spread function, 265 441
slit-like aperture, 260, 261, 263, 264 European Network of Forensic Science Institutes
wavelength and, 259, 264 (ENFSI)
DI-IRMS, see Dual inlet IRMS Guidelines for evaluative reporting in forensic
N, N-Dimethylformamide (DMF), 282–283 science, 363, 402
Direct dyes, 226 reporting requirements, 364
Direct pyrolysis mass spectrometry (DP-MS), 316 European Textile and Hair Group (ETHG), 184, 295
Discriminating power (DP), 208 Exotic fibres, 17
Disperse dyes, 226 Extraneous fibres, population of, 381
DNA analysis, 100, 138, 139, 441 Eye-brain system, 180, 181
Documentation, 65
Double cloths, 46
F
Double knits, 50
DP, see Discriminating power Fabric structures, 41
DP-MS, see Direct pyrolysis mass spectrometry Fancy yarns, 41
Dried retted flax, 12 FBI Lab – Forensic Automobile Carpet Database, 377
Dry spinning, 21 Fibre colouration, 32
DTGS detector, see Deuterated triglycine sulphate Fibre cross-section, 26
detector Fibre/dye combinations, 228–229
Dual inlet IRMS (DI-IRMS), 318 Fibre frequency, 351–353
DuPont, 20, 32, 164 data collections on, 377–379, 447
Dye analysis, 221 Fibre level morphological information, 74
capillary electrophoresis for, 337–338 Fibre persistence, 113–116
classification of dyes, 224–228, 230–232 Fibres, 2–5
fibre/dye combinations, 228–229 characteristics, 4–5
high-perfomance liquid chromatography for, classification, 3–4
330–331 new fibres, 39
by thin layer chromatography, see Thin layer production, 4
chromatography Fibres forming polymers, 5
Fibre transfer, 109–113
Fibre transfer examinations, 350
E
casework findings, examples of, 366–370
EFG, see European Fibres Group characteristics, comparison of, 355–358
EI, see Electron ionization frequency, 351–353
Elastomeric fibres, 35, 36 known/unknown facts, 348
Electromagnetic radiation spectrum, 185 matching fibres recovered, number of, 358–361
Electron ionization (EI), 316 quality assurance, 361–363
Electron probe microanalysis (EPMA), 325 report writing, 363–366
Electro-osmotic flow (EOF), 334 target fibres and case strategy, 353–354
458 Index
Fibrillation, 25 Henequen fibre, 15

Fibroin fibre, 19 Herring bone weave, 47
FID, see Flame ionization detector Hibiscus cannabinus, 13
Filament pyrolyzers, 313 High-perfomance liquid chromatography (HPLC),
Filament yarns, 28 328
‘First principles’ guidelines detection system, 330
dyes and pigments, 285–286 for dye analysis, 330–331
nitrile absence and carbonyl presence, 284–285 instrumentation, 328–329
nitrile and carbonyl absorption, absence of, types of, 329
279–280 Home Office Central Research Establishment
nitrile and carbonyl presence, 281–284 (HOCRE), 109
nitrile presence and carbonyl absence, 280–281 Homicides, 444
polymer identification from, 281–283 Hot stage microscope, 149
First responder indicators, 64 House markers, 95
1st level examination, 68–73 HPLC, see High-perfomance liquid chromatography
Flame ionization detector (FID), 315 Hydro entanglement, 55
Flame photometric detector (FPD), 315 Hydrophilic fibres, 430
Flattening fibres, 267 Hydrophobic epicuticle, 17
apertures, setting-up of, 269–270 Hydrophobic fibres, 430
diamond cell, 267–269
rolled fibres, 269
I
transmission-reflection, 270
Flax plant, 12–13 ICP-AES, see Inductively coupled plasma atomic
Fluorescence microscopy, 92, 149, 172–175 emission spectrometry
Fluorescence phenomenon, 292–293 ICP-MS, see Inductively coupled plasma mass
Fluorofibres, 35 spectrometry
Focal plane array (FPA), 275 Immature cotton fibres, 10, 11
Forensic fibre microscopy Impact damage, 71
comparison of, 146–147 Implements/weapons, of textile damage
fibre end-use, 147 Individual fibres, 14, 74
instruments for, 148–149 Inductively coupled plasma atomic emission
Forensic practitioner, 83 spectrometry (ICP-AES), 324
FORESIGHT Project, 445 Inductively coupled plasma mass spectrometry (ICP-
FPA, see Focal plane array MS), 324–325
FPD, see Flame photometric detector Industrial enquiries
FTIR microscopes, 355 examples of, 387–391
Fur fibres, 152–153 required information, 386–387
tracing manufacturers, 386
value of, 385–386
G
Informative appendix, 82
Get it first time (GIFT) principle, 116 Infrared microspectroscopy, 95
Glass fibres, 37 flattening fibres, 267
Global textile fibre production, 4, 5 apertures, setting-up of, 269–270
Glucose, 7 diamond cell, 267–269
Gossypium barbadense, 8 rolled fibres, 269
Gossypium hirsutum, 8 transmission-reflection, 270
Grating, Raman analysis, 301 IR spectrum interpretation
Green materials, 433–435 computer-assisted process, 279
first principles guidelines, see ‘First
principles’ guidelines
H
micro-ATR spectroscopy, 271–274
Hair fibres, 17–19 photometric inaccuracy factors
Halogenated polymers, 280 beam polarization, 265–266
Handpicking, 104 cross-sectional shape, 255–256
Hemi-cellulose, 7 diffraction, see Diffraction, photometric
Hemp, 12, 14 inaccuracy factor
abaca, 14–15 refraction, 255–256
Index 459
scattering, 254–255 Laser micropyrolysis GC-MS, 314

specimen flatness, 252–254 Lasers, in Raman spectrometer, 294–296, 300
specimen thickness, 252 Laundering, 67
stray light, 256–257 LC-MS, see Liquid chromatography-mass
polarized, 275–278 spectrometry
Raman spectroscopy vs., 291 Leaf fibres, 14–15, 150–152
spectral imaging, 275 Leather, 55–56
spectral photometric accuracy, 250 Length of lay, 90
working principle, 246–250 Lepitoptera, 19
Ingrain dyes, 228 Light source
Inorganic additives, 322–323 multichannel spectrophotometer, 194
Inorganic fibres, 3 scanning spectrophotometer, 192–193
Instrument calibration, 197–199 Lignin, 7
Intelligence indicators, 64 Lignosulphonic acid, 22
Internal reflection element (IRE), 271–272, 274 Likelihood ratios
Internal reflection spectroscopy (IRS), 271 evidence left by offenders, 403–406
International cooperation, 448 evidence possibly left by offenders, see Stolen car
Interpretation, 302–304 scenario
of damage, 82–83 for trace evidence evaluation, 398–402
of fibre results, 446–447 Linum usitatissimum, 12, 150
Investigations, fibres in, 450 Liquid chromatography-mass spectrometry (LC-
INVISTA CORPORATION, 29, 35 MS), 331
IRE, see Internal reflection element Liquid silk, 19
IRMS, see Isotope ratio mass spectrometry Logic report, 364
IRS, see Internal reflection spectroscopy Lotus effect, nanomaterials, 432
Isotactic polymers, 28 Lumen, 13
Isotactic polypropylene, 32 Lustrous fibre, 18
Isotope ratio mass spectrometry (IRMS), 317–319 black carbon, 38
Isotope ratios, 317–318 Lycra®, 31, 51
Isotropic refractive index, 159, 163–164 Lyocell fibres, 24–25, 279
Italian trade centre, 388
M
J
Maceration, 93
Jackets, 368, 389 Macroscopic examination, 91
Jacquard weaves, 45 Manila hemp, 14–15
‘Jigsaw Murder, The’, 384 Man-made fibres, 3, 199
Jute plant, 13 composition, 91
inorganic, 37–38
optical properties of
K
birefringence, 161–162
Kangaroo leather, 55 fibre identification, 164–165
Kapok tree, 11 isotropic refractive index, 163–164
Kevlar®, 36, 438 refractive indices, 159–160
Knitted fabrics, 47–52 sign of elongation, 162–163
Knives, 76–77 Standort diagram, 164, 165
KROY™, 17 regenerated from natural polymers
alginate fibres, 26
bamboo fibre, 23–24
L
esters of cellulose, 25–26
LA-ICP-MS, see Laser ablation inductively coupled improved forms of viscose, 24–25
plasma mass spectrometry viscose rayon, 21–23
Laid ropes, 90 ropes, 93–95
Lama pacos, 18 synthetic fibres, 26–28
Lambert–Beer law, 197 aramides, 36–37
Laser ablation inductively coupled plasma mass polyamides, 30–32
spectrometry (LA-ICP-MS), 324, 325 polyesters, 28–30
460 Index
polyolefins, 32–33 spectral measurement, 199–205

polyurethane fibres, 35–36 theory of, 184–189, 190
polyvinyl alcohol fibres, 37 validation, 196–197
polyvinyl fibres, 34–35 Military fabrics, 11
Manufacturing information, 449 Milk (protein) fibres, 435
Mass spectrometer (MS), 315, 335–336 Mineral fibres, 20
Mature cotton fibres, 10, 11 Modacrylic fibre, 34, 162
MCT detector, see Mercury cadmium telluride Modern terminology, 61
detector Mohair fibre, 18
Mechanical spinning, 40 Moisture regulating properties, 430
Medulla region, wool fibre, 16 Molecular chain flexibility, 5
MEKC, see Micellar electrokinetic capillary Monochromatic detectors, 330
chromatography Monochromator, scanning spectrophotometer, 193
Melting/burning, fibre, 74 Monomer reversion mechanism, 312, 315
Melting point determination, 95 Morphological data, 378
Melt spinning, 21 MS, see Mass spectrometer
Mercerising process, 9, 10 MSP, see Microspectrophotometry
Mercury cadmium telluride (MCT) detector, 248 Multichannel microspectrophotometer, 194–195
Metachrome method, 226 measurement conditions of, 203
Metallised yarns (Lurex)-polyester film, 38 Multi-layer systems, 430
Metallized dyes, 226 Musa textilis, 14
Metamerism, 181
Methacrylic acid, 34
N
Metropolitan Laboratory of the Forensic Science
Service, 92 NAA, see Neutron activation analysis
Micellar electrokinetic capillary chromatography Nanomaterials, 432–433
(MEKC), 334–335 National Institute of Standards and Technology
Micro-ATR spectroscopy, 271–274 (NIST), 197
Microbial damage, 67 Natural fibres, 3, 129, 355
Microchemical testing, 175 animal fibres, 131, 135
Microfibres, 29 non-human animal hair, 135–139
Micro furnace, 312 vegetable fibres, 129–131
Microscope, in Raman spectrometer, 296–297, 300 leaf fibres, 14–15
Microscopic comparison, 126–129 seed fibres, 8–11
Microscopic examination of fibres, 91 viscose rayon, 21–23
colour observation, 168, 169 Needling, 54
definition of, 146 Neochrome process, 358
dichroism, 169–172 Neutron activation analysis (NAA), 324
fluorescence, 172–175 New Zealand hemp, 14–15
microchemical testing, 175 NIST, see National Institute of Standards and
polarizing microscope, 158 Technology
thermal, 167–168 Nitrile groups, 160
Microspectrophotometry (MSP), 296, 353, 356, 388 Nomex™, 20, 36, 438
absorption measurement, 200–202 Non-discriminating pyrolysis-GC, 316
colour, 180–182 Non-extractable dyes, 241
data control, processing and recording, 196 Non-fatal stabbing, 79
discriminating power, 207–208 Non-human animal hair, 135–139
fluorescence emission measurement, 203–204 Non-reactive fibre dyes, 234–239
history of, 182–184 Non-woven textile material, 53–56
instrument calibration, 197–199 Normal phase HPLC (NP-HPLC), 329
measurement procedure, 199 Numerical colour coding
microscope, 190–192 complementary chromaticity coordinates,
multichannel microspectrophotometer, 194–195 219–220
quantitative and qualitative limitations, 206–207 instrumental measurements, 216–217
sampling and sample preparation, 205 measurement approaches, 213
scanning spectrophotometer, 190 visual comparison, 214–216
spectral comparison, 208–213 Nylon, 4
Index 461
Nylon 6, 30–32 Polyacrylonitrile fibres, dyes classification from,

pyrolysis profiles of, 312 231
Nylon 6.6, pyrolysis profiles of, 312 Polyamide fibres, 30–32
Nylon 66, 30, 32 dyes classification from, 232
Polybenzimadazole fibres, 280
Polychromator, 194
O
Polyester fibres, 28–30, 53, 356
Ongoing chemical reactions, 66 dyes classification from, 232
Organic synthetic polymer fibres, 4 thermal damage, 75
Overcoat, 389 Polyethylene (polyethene), 32, 95
fibres, 280
random chain scission of, 313
P
Polyethylene terephthalate (PET) fibres, 28, 267, 439
Packaging, of textile damage examination, Polygalaturonic acid, 7
65–67 Poly lactic acid (PLA) fibres, 285, 434, 438–440
PAN, see Polyacrylonitrile Polymerisation
PCM, see Phase change material of acrylonitrile, 34
PDA detector, see Photodiodearray detector of p-phenylene diamine, 36
Pectins, 7 of tetrafloroethene gas, 35
People v. Giangrande, 397 Polymers, fibres forming, 5
Persistence, fibre, 113–116 Polymethylmethacrylate, monomer reversion
Personal protective equipment (PPE), 107 mechanism of, 315
PET fibres, see Polyethylene terephthalate fibres Polymorphism, in synthetic fibres, 355
Phase change material (PCM), 429 Polyolefin fibres, 32–33, 95
m-Phenylene diamine, 36 stress marks in, 92
p-Phenylene diamine, 36 Polypropylene fibres, 32–33, 280
Phormium tenax, 15 carpet type fibres, 389–391
Photochromic property, 431 dyes classification from, 232
Photodiode array (PDA) detector, 182, 183, 194, 196, properties of, 35
330 Polytrimethylene terephthalate (PTT), 39, 438, 439
Photometric inaccuracy, 250 Polyurethane fibres, 35–36
instrumental contributions Polyvinyl alcohol fibres, 37
beam polarization, 265–266 Polyvinyl chloride, side-group scission of, 314
diffraction, see Diffraction, photometric Polyvinyl fibres
inaccuracy factor acrylic fibres, 34
stray light, 256–257 chlorofibres, 35
specimen contributions Population studies, 381–384
flatness, 252–254 Post-consumer textile waste material, 437
refraction and cross-sectional shape, PPE, see Personal protective equipment
255–256 Pre-consumer textile waste material, 436–437
scattering, 254–255 Preservation, of textile damage examination, 65–67
thickness, 252 Prism monochromators, 192
Photomultiplier detector, 193 Probabilistic approach, fibre evidence evaluation, 395
Photomultiplier tubes (PMT), 196 Bayesian networks, 417–421
Pigments, 285–286 interpretation process, 396–398
Pile fabrics, 45 likelihood ratios, see Likelihood ratios
Pilling, 52–53 need for data, 421–422
Pima type cotton, 8 Probability table, 419, 420
PLA fibres, see Poly lactic acid fibres Proceedings of the Morin Commission of Enquiry in
Plain weave fabrics, 43, 44 Canada (1998), 442
Plasma treatment, 432 Production tracking, 437–438
Pleochroism, 169, 201 Prosecution hypothesis, 400, 420
PMT, see Photomultiplier tubes Protective textiles, 438
Polarized infrared microspectroscopy, 275–278 Protein fibres, 20
Polarized light microscopy, 93 casein, 435, 436
Polarizing microscope, 148, 158 soy, 434, 436
Polyacrylonitrile (PAN), 34, 160, 280 Pseudo-activity proposition, 401
462 Index
PTT, see Polytrimethylene terephthalate Recovery of fibres

Puncture damage, 71 combing/brushing, 106, 107
Puncture type weapons, 64 cutting, 107
Purl stitches, 49 handpicking, 104
Py-MS, see Pyrolysis-mass spectrometry search methodologies, guidance on, 103
Pyrolysis-gas chromatograph (pyrolysis-GC), 311, tape-lifts, 104–106
315 trace materials, 108–109
disadvantages, 316–317 vacuuming, 106–107
non-discriminating, 316 Recycled material, 436–437
Pyrolysis-mass spectrometry (Py-MS), 315–316 Regenerated fibres, 3, 153–157
Pyrolysis techniques, fibre identification by Resonance Raman scattering (RR), 301
detectors, types of, 315 Reversed phase HPLC (RP-HPLC/RPC), 329
laser micropyrolysis GC-MS, 314 Rib knit fabrics, 50
pyrolysis mechanisms, 311–315 Ring spun yarns, 40
pyrolyzers, 312–314 Robustness report, 364
Pyrolyzers, types of, 312–314 Rolled fibres, 269
Ropes/twines, 90
comparison microscopy, 92
Q
manufacturers, 386, 387
QA, see Quality assurance markers and colour comparison, 95–96
Qiana, pyrolysis profile of, 312 natural fibres, 355
Quadrupole time-of-flight (QTOF) mass polypropylene rope production, 96
spectrometer, 336 structure of, 90
Quality assurance (QA), 361–363 Rotor spun yarns, 53
Quality system, improving standards, 449
S
R
Sample focus, in Raman analysis, 301
Rachel knit, 51, 52 Sateen weave fabrics, 45
Raman scattering, 290 Satin weave fabrics, 44, 47
Raman spectroscopy, 289, 290, 444 Savitzky–Golay method, 211
data handling, 302 Scale-cell layer, 16
fibre mounting conditions, 297–300, 300 Scanning electron micrographs, 9
fluorescence phenomenon, 292–293 Scanning electron microscopy (SEM), 74
instrumentation elemental analysis of fibres, see Elemental
considerations, 293 analysis of fibres
instrument set-up, 293–294 as imaging tool, 319–321
lasers, 294–296 Scanning spectrophotometer, 190
microscope, 296–297 measurement conditions of, 202
interpretations, 302–304 Scattering, photometric inaccuracy factor, 254–255
principle of, 290–292 Scientific Working Group Material Sciences
Raman analysis conditions, 300–301 Advisory Group (SWGMAT), 118,
Raman spectrum, 291, 292 184, 441
sub-types of, 301 Scissors, 76, 77
Raman spectrum, 291, 292 Scotching, 12
of dyed fibre, 304 SeaCell®, 280, 430
Ramie fibres, 13–14 Secondary cuts, 77
Random chain scission mechanism, 311, 313 2nd level examination
Rapid population growth, 2 detailed, 73–74
Raw cellulose, 24 fibre level, 74
Raw fibre, 11 limitations and cautions, 81
Raw silk, 19 simulation experiment, 78–79
Raw wool fibre, 16 textile effects, 77–78
Rayleigh scattering, 291 weapon/implement effects, 75–77
Reactive dyes, 227 Security measures, 437–438
thin layer chromatography of, 239–241 Seed fibres, 8–11, 150
Recognition level examination, 63–68 SEM-EDX, 358
Index 463
Semi-worsted system, 41 likelihood ratio, formulation of, 405, 407, 409,

Sericin, 19 413, 414, 416
Serpentine, 20 likelihood ratio, representation of, 410, 413, 418
SERRS, see Surface enhanced resonance Raman recovered fibres, empirical distribution for, 406
scattering transfer probabilities, 408, 409
SERS, see Surface enhanced Raman scattering Stray light, 246, 256–257
Sexual assault , cases of, 444 Stress marks, in polyolefin, 92
Shape memory properties (SMPs), 431 S twist yarns, 42
Sheepskins, 56 ‘Suffolk Strangler, The’, 384
Side-by-side extrusion, 38 Sulphur dyes, 227
Side-group scission mechanism, 311, 314 Sun protective clothing, 432
Sign of elongation, man-made fibres, 162–163 Supply chain, 449
Silk fibre, 153 Surface enhanced Raman scattering (SERS), 301
Silkworm, 19–20 Surface enhanced resonance Raman scattering
Silver particles, application of, 430, 433, 435 (SERRS), 301
Simulations, 78–79 Surface hydrolysis, 28
Sisal plant, 15 Surface lustre, 9
Slash type damage, 73 Sweeping, 106
Smart textiles, 429 Syndiotactic polymers, 28
antibacterial properties, 430 Syndotactic polypropylene, 32
chromic properties, 430–431 Synthetic fibres, 3, 4, 26–38
conductive properties, 431 condensation polymers, 26
thermal properties, 429–430 extreme variability of, 354–355
UV protection properties, 432 inorganic man-made fibres, 37–38
SMPs, see Shape memory properties morphology of, 6, 153–157
Sodium alginate solution, 26 polyamides, 30–32
Sodium carbonate, 8 polyesters, 28–30
Sodium chlorite, 7 polyolefins, 32–33
Sodium hydroxide, 8 polyurethane fibres, 35–36
Sodium hypochlorite, 7, 22 polyvinyl alcohol fibres, 37
Solvatechromic property, 431 sub-classification of, 27
Sorona, 39 Systematic approach, 70
Soy fibres, 26, 434, 436
Spandex, 35
T
Specialist testing, 63
Spectral photometric accuracy, 250 Tank retting, 12
Spectral presentation, 302 Tanning method, 56
Spectrophotometers, 362 Tape-lifts, 104–106
Spider silk, 20 Target fibres, 353–355, 379–381
Spinning solution, 22 Tear type damage, 69
Spun bonded fabrics, 54 Technical review, 83
Spun yarn, 42 Teflon coating, 432
Stabbing angle, 77 Tencel, 24
Stain repellent properties, of fabrics, 432 Tensile testers, 82
Standort diagram, 164, 165 Tent fabric, 75
Stem fibres Textile damage examination, 61
flax plant, 12–13 briefing, 67–68
hemp, 14 causes of, 64–65
jute plant, 13 documentation, 65
ramie, 13–14 1st level examination, 68–73
Stereomicroscopy, 74, 148, 381 interpretation, reporting and standards, 82–83
Stitch bonding fabrics, 54 2nd level examination, see 2nd level
Stolen car scenario examination
background probabilities, 408, 410 packaging and preservation, 65–67
conditional transfer probability, 415 testing, 81–82
cross-transfer evidence, 416, 417 Textile effects, 77–78
forensic evidence, formulation of, 412 Textile fibres, classification of, 3
464 Index
Textile Institute, fibre definition, 4 Ultraviolet protection factor (UPF), 432

Textile manufacturing industry, 8 Unbalanced plain weave fabrics, 43
Textile Market Studies, 377 UPF, see Ultraviolet protection factor
Thermal bonding, 54 UV protective clothing, 432
Thermal damage, 72 UV-vis spectroscopy, 184–187, 208
Thermal microscopy, 167–168
Thermal properties, of clothing fabrics, 429–430
Thermal sources, 64 V
Thermochromic property, 431 Vacuuming, 106–107
THERMOLITE®, 29, 31 Validation, 361
Thiazine, 225 Vat dyes, 227
Thin layer chromatography (TLC), 233–234, 328, 338 Vegetable fibres, 5–15, 129–131, 382
non-extractable dyes, 241 Very small particles (VSPs), 358
non-reactive fibre dyes, 234–239 ν4 dependency, 294, 295
reactive fibre dyes, 239–241 Vicuna fibre, 19
TIDAS MSP 400™, 183 Viscose fibres, 279
TIDAS MSP 800™, 183 dyes classification from, 231
TLC, see Thin layer chromatography improved forms of, 24–25
‘Topping-up’ process, 181 Viscose rayon, 21–23
Tossa jute, 13 Visual assessment, 62
Total reflection X-ray fluorescence spectrometry VSPs, see Very small particles
(TXRF), 326, 327
Trace biological material, 68
Trace fibre evidence W
case scenario, 119
fibre examination, 119, 120 Warp fibres, 42
low magnification microscopy, 121 Warp-knitted polypropylene, 33
questioned and recovered fibre, 126 Warp knitting, 47, 51–52
tape-lifts, 123, 125 Water insolubility, 37
target fibres, 124 Water-soluble PVOH fibres, 37
Trace materials, 108–109 Wavelengths range, Raman analysis, 300
Tracer codes, 438 Weapon/implement effects, 75–77
Tracing manufacturers methods, 386 Weft knitting, 47, 48–50
Transfer probabilities, 408, 409 Weft yarns, 43
Transmission-reflection, 270 Wet spinning, 21
Transparency report, 364 White jute, 13
Trevira CS, 29 Wild silk fibres, 19
Triacetate, 25, 26 Witness statement indicators, 64
Triexta, 438 Woody bundles, 12
Trilobal polypropylene, 375 Wool fibres, 152–153
Triple quadrupole mass spectrometer, 336 crimp, 17
True synthetic fibres, 20 dyes classification from, 230
Tuck stitch, 49 regions, 16
Tufting, 46 speciality, 17–19
Turquoise viscose fibres, 366 Woollen system, 41
Tussah silk, 19 Worsted system, 40
Twill weave fabrics, 43 Woven fabrics, 42–47
Twines, 90, 92–93, 388, see also Ropes/twines
TXRF, see Total reflection X-ray fluorescence X
spectrometry
Type 704EF, 29, 30 Xanthate process, 279
Type 727EF, 29, 31 X-ray diffraction (XRD), 324
X-ray fluorescence microanalysis (micro XRF/
XRFM), 325, 327
U
X-ray fluorescence spectrometry (XRF), 325,
Ultimates, 93, 129 325–326
Ultra fine steel fibres, 38 Xylose, 7
Index 465
Y Z
Yarns, 40–41 Zeiss interference microscope, 160
morphological information, 71–73 Zeiss MPM 03™, 183
spinning system, 21 Zeiss MPM 800™, 191
types of, 41 Z twist yarns, 41
‘Yucatan sisal’, 15

(James Robertson, Claude Roux, Kenneth G. Wiggins) (B-Ok - Xyz)

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

(James Robertson, Claude Roux, Kenneth G. Wiggins) (B-Ok - Xyz)

Hochgeladen von

Copyright:

Verfügbare Formate

Forensic Examination

Forensic Examination of Hair

Forensic Examination of Fibres, 2nd edition

Forensic Examination of Glass and Paint: Analysis and Interpretation

Forensic Speaker Identification

The Practice of Crime Scene Investigation

Forensic Computer Crime Investigation

Analytical and Practical Aspects of Drug Testing in Hair

Nonhuman DNA Typing: Theory and Casework Applications

Forensic Science in Wildlife Investigations

Ethics and the Practice of Forensic Science

Introduction to Data Analysis with R for Forensic Scientists

Forensic Investigation of Explosions, Second Edition

Firearms, the Law, and Forensic Ballistics, Third Edition

The Neuroscience of Handwriting: Applications for Forensic Document Examination

Fingerprints and Other Ridge Skin Impressions, Second Edition

Forensic Examination of Fibres, Third Edition

© 2018 by Taylor & Francis Group, LLC

No claim to original U.S. Government works

Printed on acid-free paper

International Standard Book Number-13: 978-1-4398-2864-9 (Hardback)

Library of Congress Cataloging-in-Publication Data

Visit the Taylor & Francis Web site at

and the CRC Press Web site at

Preface to second edition ix

2 Textile Damage Interpretation 61

3 Ropes and Cordages 89

4 From Crime Scene to Laboratory 99

5 Microscopic Examination of Fibres 145

6 Colour Analysis of Fibres 179

7 Infrared Microspectroscopy of Fibres 245

8 Raman Spectroscopy of Fibres 289

9 Other Instrumental Approaches to Fibre Examination 309

10 Interpretation of Fibre Evidence 345

11 Future Trends for Forensic Fibre Examination 427

Thomas W Biermann is currently with the Forensic Science Institute in Wiesbaden,

Max M Houck is an international forensic expert with over 25 years of experience.

K Paul Kirkbride is currently Professor of Forensic Science at Flinders University in

1.6.2.4 Polyvinyl Fibres 34

Table 1.1 Use of Textiles in Various Industries

1.2.1 Classification of Textile Fibres

Man-made fibres are manufactured and fall into three classes:

• Regenerated fibres (those formed from naturally occurring fibre-forming poly-

Animal Vegetable Organic Inorganic

Transformation of natural Synthetic polymers

Figure 1.1 Broad classification of natural and man-made fibres.

1.2.2 Fibre Production

1.2.3 Fibre Characteristics

* Textile Terms and Definitions, 11th edition, Textile Institute.

Table 1.2 Global Textile Fibre

1.2.4 Properties of Fibre Forming Polymers

1. A degree of crystallinity of the molecular chains to give longitudinal strength

1.3 Natural Fibres – Vegetable Fibres

Animal Vegetable Organic Inorganic

Seed Stem (bast) Leaf

Figure 1.3 Sub-classification of vegetable fibres.

Table 1.3 Approximate (%) Chemical Composition of Cellulosic Fibres

* Retting is the bacterial decomposition of lignan and hemicellulose.

1.3.1 Seed Fibres – Cotton, Kapok and Coir

Figure 1.5 Cross-section of a cotton fibre showing the major features.

Mature fibre with a

Dead fibre with very

This is particularly so when scouring using sodium hydroxide. Partial mercerisation