Rheological Methods in Food Process Engineering (2nd Edition)

RHEOLOGICAL METHODS IN
FOOD PROCESS ENGINEERING

Second Edition
Jame s F. St e ffe , Ph. D. , P. E.
Professor of Food Process Engineering
Dept. of Food Science and Human Nutrition
Dept. of Agricultural Engineering
Michigan State University
Fre e man Pre s s
2807 Still Valley Dr.
East Lansing, MI 48823
USA
Prof. James F. Steffe
209 Farrall Hall
Michigan State University
East Lansing, MI 48824-1323
USA
Phone: 517-353-4544
FAX: 517-432-2892
E-mail: steffe@msu.edu
URL: www.egr.msu.edu/~steffe/
Copyright 1992, 1996 by James F. Steffe.
All rights reserved. No part of this work may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any
means, electronic, mechanical, photocopying, recording, or other-
wise, without the prior written permission of the author.
Printed on acid-free paper in the United States of America
Second Printing
Library of Congress Catalog Card Number: 96-83538
International Standard Book Number: 0-9632036-1-4
Fre e man Pre s s
2807 Still Valley Dr.
East Lansing, MI 48823
USA
Table of Cont e nt s
Preface ix
Chapter 1. Introduction to Rheology 1
1.1. Overview ...................................................................................... 1
1.2. Rheological Instruments for Fluids .......................................... 2
1.3. Stress and Strain .......................................................................... 4
1.4. Solid Behavior ............................................................................. 8
1.5. Fluid Behavior in Steady Shear Flow ....................................... 13
1.5.1. Time-Independent Material Functions ............................ 13
1.5.2. Time-Dependent Material Functions ............................... 27
1.5.3. Modeling Rheological Behavior of Fluids ....................... 32
1.6. Yield Stress Phenomena ............................................................. 35
1.7. Extensional Flow ......................................................................... 39
1.8. Viscoelastic Material Functions ................................................ 47
1.9. Attacking Problems in Rheological Testing ............................ 49
1.10. Interfacial Rheology ................................................................. 53
1.11. Electrorheology ......................................................................... 55
1.12. Viscometers for Process Control and Monitoring ................ 57
1.13. Empirical Measurement Methods for Foods ........................ 63
1.14. Example Problems .................................................................... 77
1.14.1. Carrageenan Gum Solution ............................................. 77
1.14.2. Concentrated Corn Starch Solution ................................ 79
1.14.3. Milk Chocolate .................................................................. 81
1.14.4. Falling Ball Viscometer for Honey ................................. 82
1.14.5. Orange Juice Concentrate ................................................ 86
1.14.6. Influence of the Yield Stress in Coating Food ............... 91
Chapter 2. Tube Viscometry 94
2.1. Introduction ................................................................................. 94
2.2. Rabinowitsch-Mooney Equation .............................................. 97
2.3. Laminar Flow Velocity Profiles ................................................ 103
2.4. Laminar Flow Criteria ................................................................ 107
2.5. Data Corrections ......................................................................... 110
2.6. Yield Stress Evaluation .............................................................. 121
2.7. Jet Expansion ............................................................................... 121
2.8. Slit Viscometry ............................................................................ 122
2.9. Glass Capillary (U-Tube) Viscometers .................................... 125
2.10. Pipeline Design Calculations .................................................. 128
2.11. Velocity Profiles In Turbulent Flow ....................................... 138
2.12. Example Problems .................................................................... 141
2.12.1. Conservation of Momentum Equations ........................ 141
2.12.2. Capillary Viscometry - Soy Dough ................................ 143
2.12.3. Tube Viscometry - 1.5% CMC ......................................... 146
2.12.4. Casson Model: Flow Rate Equation ............................... 149
2.12.5. Slit Viscometry - Corn Syrup .......................................... 150
2.12.6. Friction Losses in Pumping ............................................. 152
2.12.7. Turbulent Flow - Newtonian Fluid ................................ 155
2.12.8. Turbulent Flow - Power Law Fluid ................................ 156
v
Chapter 3. Rotational Viscometry 158
3.1. Introduction ................................................................................. 158
3.2. Concentric Cylinder Viscometry .............................................. 158
3.2.1. Derivation of the Basic Equation ...................................... 158
3.2.2. Shear Rate Calculations ...................................................... 163
3.2.3. Finite Bob in an Infinite Cup ............................................. 168
3.3. Cone and Plate Viscometry ....................................................... 169
3.4. Parallel Plate Viscometry (Torsional Flow) ............................ 172
3.5. Corrections: Concentric Cylinder ............................................. 174
3.6. Corrections: Cone and Plate, and Parallel Plate ..................... 182
3.7. Mixer Viscometry ....................................................................... 185
3.7.1. Mixer Viscometry: Power Law Fluids ............................. 190
3.7.2. Mixer Viscometry: Bingham Plastic Fluids ..................... 199
3.7.3. Yield Stress Calculation: Vane Method ........................... 200
3.7.4. Investigating Rheomalaxis ................................................ 208
3.7.5. Defining An Effective Viscosity ........................................ 210
3.8. Example Problems ...................................................................... 210
3.8.1. Bob Speed for a Bingham Plastic Fluid ............................ 210
3.8.2. Simple Shear in Power Law Fluids .................................. 212
3.8.3. Newtonian Fluid in a Concentric Cylinder ..................... 213
3.8.4. Representative (Average) Shear Rate .............................. 214
3.8.5. Concentric Cylinder Viscometer: Power Law Fluid ...... 216
3.8.6. Concentric Cylinder Data - Tomato Ketchup ................. 218
3.8.7. Infinite Cup - Single Point Test ......................................... 221
3.8.8. Infinite Cup Approximation - Power Law Fluid ........... 221
3.8.9. Infinite Cup - Salad Dressing ............................................ 223
3.8.10. Infinite Cup - Yield Stress Materials .............................. 225
3.8.11. Cone and Plate - Speed and Torque Range ................... 226
3.8.12. Cone and Plate - Salad Dressing ..................................... 227
3.8.13. Parallel Plate - Methylcellulose Solution ....................... 229
3.8.14. End Effect Calculation for a Cylindrical Bob ................ 231
3.8.15. Bob Angle for a Mooney-Couette Viscometer .............. 233
3.8.16. Viscous Heating ................................................................ 235
3.8.17. Cavitation in Concentric Cylinder Systems .................. 236
3.8.18. Mixer Viscometry .............................................................. 237
3.8.19. Vane Method - Sizing the Viscometer ........................... 243
3.8.20. Vane Method to Find Yield Stresses .............................. 244
3.8.21. Vane Rotation in Yield Stress Calculation .................... 247
3.8.22. Rheomalaxis from Mixer Viscometer Data ................... 250
Chapter 4. Extensional Flow 255
4.1. Introduction ................................................................................. 255
4.2. Uniaxial Extension ...................................................................... 255
4.3. Biaxial Extension ......................................................................... 258
4.4. Flow Through a Converging Die .............................................. 263
4.4.1. Cogswells Equations ......................................................... 264
4.4.2. Gibsons Equations ............................................................. 268
4.4.3. Empirical Method ............................................................... 271
4.5. Opposing Jets .............................................................................. 272
4.6. Spinning ....................................................................................... 274
4.7. Tubeless Siphon (Fano Flow) .................................................... 276
vi
4.8. Steady Shear Properties from Squeezing Flow Data ............. 276
4.8.1. Lubricated Squeezing Flow ............................................... 277
4.8.2. Nonlubricated Squeezing Flow ........................................ 279
4.9.1. Biaxial Extension of Processed Cheese Spread ............... 283
4.9.2. Biaxial Extension of Butter ................................................ 286
4.9.3. 45 Converging Die, Cogswells Method ........................ 287
4.9.4. 45 Converging Die, Gibsons Method ............................ 289
4.9.5. Lubricated Squeezing Flow of Peanut Butter ................. 291
Chapter 5. Viscoelasticity 294
5.1. Introduction ................................................................................. 294
5.2. Transient Tests for Viscoelasticity ............................................ 297
5.2.1. Mechanical Analogues ....................................................... 298
5.2.2. Step Strain (Stress Relaxation) .......................................... 299
5.2.3. Creep and Recovery ........................................................... 304
5.2.4. Start-Up Flow (Stress Overshoot) ..................................... 310
5.3. Oscillatory Testing ...................................................................... 312
5.4. Typical Oscillatory Data ............................................................ 324
5.5. Deborah Number ........................................................................ 332
5.6. Experimental Difficulties in Oscillatory Testing of Food ..... 336
5.7. Viscometric and Linear Viscoelastic Functions ...................... 338
5.8.1. Generalized Maxwell Model of Stress Relaxation ........ 341
5.8.2. Linearized Stress Relaxation ............................................. 342
5.8.3. Analysis of Creep Compliance Data ................................ 343
5.8.4. Plotting Oscillatory Data ................................................... 348
6. Appendices 350
6.1. Conversion Factors and SI Prefixes .......................................... 350
6.2. Greek Alphabet ........................................................................... 351
6.3. Mathematics: Roots, Powers, and Logarithms ....................... 352
6.4. Linear Regression Analysis of Two Variables ........................ 353
6.5. Hookean Properties .................................................................... 357
6.6. Steady Shear and Normal Stress Difference ........................... 358
6.7. Yield Stress of Fluid Foods ........................................................ 359
6.8. Newtonian Fluids ....................................................................... 361
6.9. Dairy, Fish and Meat Products ................................................. 366
6.10. Oils and Miscellaneous Products ........................................... 367
6.11. Fruit and Vegetable Products ................................................. 368
6.12. Polymer Melts ........................................................................... 371
6.13. Cosmetic and Toiletry Products ............................................. 372
6.14. Energy of Activation for Flow for Fluid Foods .................... 374
6.15. Extensional Viscosities of Newtonian Fluids ........................ 375
6.16. Extensional Viscosities of Non-Newtonian Fluids .............. 376
6.17. Fanning Friction Factors: Bingham Plastics .......................... 377
6.18. Fanning Friction Factors: Power Law Fluids ........................ 378
6.19. Creep (Burgers Model) of Salad Dressing ............................. 379
6.20. Oscillatory Data for Butter ...................................................... 380
6.21. Oscillatory Data Iota-Carrageenan Gel ................................. 381
6.22. Storage and Loss Moduli of Fluid Foods .............................. 382
vii
Nomenclature ......................................................................................... 385
Bibliography ........................................................................................... 393
Index ......................................................................................................... 412
viii
Pre fac e
Growth and devel opment of thi s work sprang from the need to
provi de educati onal materi al for food engi neers and food sci enti sts. The
fi rst edi ti on was concei ved as a textbook and the work conti nues to be
used i n graduate l evel courses at vari ous uni versi ti es. I ts greatest
appeal , however, was to i ndi vi dual s sol vi ng practi cal day-to-day prob-
l ems. Hence, the second edi ti on, a si gni fi cantl y expanded and revi sed
versi on of the ori gi nal work, i s ai med pri mari l y at the rheol ogi cal
practi ti oner (parti cul arl y the i ndustri al practi ti oner) seeki ng a broad
understandi ng of the subject matter. The overal l goal of the text i s to
present the i nformati on needed to answer three questi ons when faci ng
probl ems i n food rheol ogy: 1. What properti es shoul d be measured? 2.
What type and degree of deformati on shoul d be i nduced i n the mea-
surement? 3. How shoul d experi mental data be anal yzed to generate
practi cal i nformati on? Al though the mai n focus of the book i s food,
sci enti sts and engi neers i n other fi el ds wi l l fi nd the work a conveni ent
reference for standard rheol ogi cal methods and typi cal data.
Overal l , the work presents the theory of rheol ogi cal testi ng and
provi des the anal yti cal tool s needed to determi ne rheol ogi cal properti es
from experi mental data. Methods appropri ate for common food i ndustry
appl i cati ons are presented. Al l standard measurement techni ques for
fl ui d and semi -sol i d foods are i ncl uded. Sel ected methods for sol i ds are
al so presented. Resul ts from numerous fi el ds, parti cul arl y pol ymer
rheol ogy, are used to address the fl ow behavi or of food. Mathemati cal
rel ati onshi ps, deri ved from si mpl e force bal ances, provi de a funda-
mental vi ew of rheol ogi cal testi ng. Onl y a background i n basi c cal cul us
and el ementary stati sti cs (mai nl y regressi on anal ysi s) i s needed to
understand the materi al . The text contai ns numerous practi cal exampl e
probl ems, i nvol vi ng actual experi mental data, to enhance comprehen-
si on and the executi on of concepts presented. Thi s feature makes the
work conveni ent for sel f-study.
Speci fi c expl anati ons of key topi cs i n rheol ogy are presented i n
Chapter 1: basi c concepts of stress and strai n; el asti c sol i ds showi ng
Hookean and non-Hookean behavi or; vi scometri c functi ons i ncl udi ng
normal stress di fferences; model i ng fl ui d behavi or as a functi on of shear
rate, temperature, and composi ti on; yi el d stress phenomena, exten-
si onal fl ow; and vi scoel asti c behavi or. Effi ci ent methods of attacki ng
probl ems and typi cal i nstruments used to measure fl ui d properti es are
di scussed al ong wi th an exami nati on of probl ems i nvol vi ng i nterfaci al
ix
rheol ogy, el ectrorheol ogy, and on-l i ne vi scometry for control and mon-
i tori ng of food processi ng operati ons. Common methods and empi ri cal
i nstruments uti l i zed i n the food i ndustry are al so i ntroduced: Texture
Profi l e Anal ysi s, Compressi on-Extrusi on Cel l , Warner-Bratzl er Shear
Cel l , Bostwi ck Consi stometer, Adams Consi stometer, Amyl ograph,
Fari nograph, Mi xograph, Extensi graph, Al veograph, Kramer Shear
Cel l , Brookfi el d di sks and T-bars, Cone Penetrometer, Hoeppl er Vi s-
cometer, Zhan Vi scometer, Brabender-FMC Consi stometer.
The basi c equati ons of tube (or capi l l ary) vi scometry, such as the
Rabi nowi tsch-Mooney equati on, are deri ved and appl i ed i n Chapter 2.
Lami nar fl ow cri teri a and vel oci ty profi l es are eval uated al ong wi th data
correcti on methods for many sources of error: ki neti c energy l osses, end
effects, sl i p (wal l effects), vi scous heati ng, and hol e pressure. Tech-
ni ques for gl ass capi l l ary and sl i t vi scometers are consi dered i n detai l .
A secti on on pi pel i ne desi gn cal cul ati ons has been i ncl uded to faci l i tate
the constructi on of l arge scal e tube vi scometers and the desi gn of fl ui d
pumpi ng systems.
A general format, anal ogous to that used i n Chapter 2, i s conti nued
i n Chapter 3 to provi de conti nui ty i n subject matter devel opment. The
mai n foci of the chapter center around the theoreti cal pri nci pl es and
experi mental procedures rel ated to three tradi ti onal types of rotati onal
vi scometers: concentri c cyl i nder, cone and pl ate, and paral l el pl ate.
Mathemati cal anal yses of data are di scussed i n detai l . Errors due to
end effects, vi scous heati ng, sl i p, Tayl or vorti ces, cavi tati on, and cone
truncati ons are i nvesti gated. Numerous methods i n mi xer vi scometry,
techni ques havi ng si gni fi cant potenti al to sol ve many food rheol ogy
probl ems, are al so presented: sl ope and matchi ng vi scosi ty methods to
eval uate average shear rate, determi nati on of power l aw and Bi ngham
pl asti c fl ui d properti es. The vane method of yi el d stress eval uati on,
usi ng both the sl ope and si ngl e poi nt methods, al ong wi th a consi der-
ati on of vane rotati on duri ng testi ng, i s expl ored i n detai l .
The experi mental methods to determi ne extensi onal vi scosi ty are
expl ai ned i n Chapter 4. Techni ques presented i nvol ve uni axi al exten-
si on between rotati ng cl amps, bi axi al extensi onal fl ow achi eved by
squeezi ng materi al between l ubri cated paral l el pl ates, opposi ng jets,
spi nni ng, and tubel ess si phon (Fano) fl ow. Rel ated procedures,
i nvol vi ng l ubri cated and nonl ubri cated squeezi ng, to determi ne shear
fl ow behavi or are al sopresented. Cal cul ati ng extensi onal vi scosi ty from
fl ows through tapered convergences and fl at entry di es i s gi ven a
thorough exami nati on.
x
Essenti al concepts i n vi scoel asti ci ty and standard methods of
i nvesti gati ng the phenomenon are i nvesti gated i n Chapter 5. The ful l
scope of vi scoel asti c materi al functi ons determi ned i n transi ent and
osci l l atory testi ng are di scussed. Mechani cal anal ogues of rheol ogi cal
behavi or are gi ven as a means of anal yzi ng creep and stress rel axati on
data. Theoreti cal aspects of osci l l atory testi ng, typi cal data, and a
di scussi on of the vari ous modes of operati ng commerci al i nstruments
-strai n, frequency, ti me, and temperature sweep modes- are presented.
The Deborah number concept, and how i t can be used to di sti ngui sh
l i qui d from sol i d-l i ke behavi or, i s i ntroduced. Start-up fl ow (stress
overshoot) and the rel ati onshi p between steady shear and osci l l atory
properti es are al so di scussed. Conversi on factors, mathemati cal rel a-
ti onshi ps, l i near regressi on anal ysi s, and typi cal rheol ogi cal data for
food as wel l as cosmeti cs and pol ymers are provi ded i n the Appendi ces.
Nomencl ature i s conveni entl y summari zed at the end of the text and a
l arge bi bl i ography i s furni shed to di rect readers to addi ti onal i nfor-
mati on.
J.F. Steffe
June, 1996
xi
De dic at ion
To Susan, Justi nn, and Dana.
xiii
Chapt e r 1 . Int roduc t ion t o Rhe ology
1.1. Overview
The fi rst use of the word "rheol ogy"i s credi ted toEugene C. Bi ngham
(ci rca 1928) who al so descri bed the motto of the subject as
("panta rhei ," from the works of Heracl i tus, a pre-Socrati c Greek phi -
l osopher acti ve about 500 B.C.) meani ng "everythi ng fl ows" (Rei ner,
1964). Rheol ogy i s now wel l establ i shed as the sci ence of the deformati on
and fl ow of matter: I t i s the study of the manner i n whi ch materi al s
respond to appl i ed stress or strai n. Al l materi al s have rheol ogi cal
properti es and the area i s rel evant i n many fi el ds of study: geol ogy and
mi ni ng (Cri stescu, 1989), concrete technol ogy (Tattersal l and Banfi l l ,
1983), soi l mechani cs (Haghi ghi et al ., 1987; Vyal ov, 1986), pl asti cs
processi ng (Deal y and Wi ssburn, 1990), pol ymers and composi tes
(Nei l sen and Landel , 1994; Yanovsky, 1993), tri bol ogy (study of l ubri -
cati on, fri cti on and wear), pai nt fl ow and pi gment di spersi on (Patton,
1964), bl ood (Di ntenfass, 1985), bi oengi neeri ng (Skal ak and Chi en,
1987), i nterfaci al rheol ogy (Edwards et al ., 1991), structural materi al s
(Cal l i ster, 1991), el ectrorheol ogy (Bl ock and Kel l y, 1988), psychor-
heol ogy (Drake, 1987), cosmeti cs and toi l etri es (Laba, 1993b), and
pressure sensi ti ve adhesi on (Saunders et al ., 1992). The focus of thi s
work i s food where understandi ng rheol ogy i s cri ti cal i n opti mi zi ng
product devel opment efforts, processi ng methodol ogy and fi nal product
qual i ty. To the extent possi bl e, standard nomencl ature (Deal y, 1994)
has been used i n the text.
One can thi nk of food rheol ogy as the materi al sci ence of food. There
are numerous areas where rheol ogi cal data are needed i n the food
i ndustry:
a. Process engi neeri ng cal cul ati ons i nvol vi ng a wi de range of equi p-
ment such as pi pel i nes, pumps, extruders, mi xers, coaters, heat
exchangers, homogeni zers, cal enders, and on-l i ne vi scometers;
b. Determi ni ng i ngredi ent functi onal i ty i n product devel opment;
c. I ntermedi ate or fi nal product qual i ty control ;
d. Shel f l i fe testi ng;
e. Eval uati on of food texture by correl ati on to sensory data;
f. Anal ysi s of rheol ogi cal equati ons of state or consti tuti ve equati ons.
Many of the uni que rheol ogi cal properti es of vari ous foods have been
summari zed i n books by Raoand Steffe (1992), and Wei pert et al . (1993).

2 Chapter 1. Introduction to Rheology
Fundamental rheol ogi cal properti es are i ndependent of the i nstru-
ment on whi ch they are measured so di fferent i nstruments wi l l yi el d
the same resul ts. Thi s i s an i deal concept and di fferent i nstruments
rarel y yi el d i denti cal resul ts; however, the i dea i s one whi ch di sti n-
gui shes true rheol ogi cal materi al properti es from subjecti ve (empi ri cal
and general l y i nstrument dependent, though frequentl y useful )
materi al characteri zati ons. Exampl es of i nstruments gi vi ng subjecti ve
resul ts i ncl ude the fol l owi ng (Bourne, 1982): Fari nograph, Mi xograph,
Extensograph, Vi scoaml yograph, and the Bostwi ck Consi stometer.
Empi ri cal testi ng devi ces and methods, i ncl udi ng Texture Profi l e
Anal ysi s, are consi dered i n more detai l i n Sec. 1.13.
1.2. Rheological Instruments for Fluids
Common i nstruments, capabl e of measuri ng fundamental rheol og-
i cal properti es of fl ui d and semi -sol i d foods, may be pl aced i nto two
general categori es (Fi g. 1.1): rotati onal type and tube type. Most are
commerci al l y avai l abl e, others (mi xer and pi pe vi scometers) are easi l y
fabri cated. Costs vary tremendousl y from the i nexpensi ve gl ass capi l -
l ary vi scometer to a very expensi ve rotati onal i nstrument capabl e of
measuri ng dynami c properti es and normal stress di fferences. Sol i d
foods are often tested i n compressi on (between paral l el pl ates), tensi on,
or torsi on. I nstruments whi ch measure rheol ogi cal properti es are cal l ed
rheometers. Vi scometer i s a more l i mi ti ng term referri ng to devi ces
that onl y measure vi scosi ty.
Rotati onal i nstruments may be operated i n the steady shear (con-
stant angul ar vel oci ty) or osci l l atory (dynami c) mode. Some rotati onal
i nstruments functi on i n the control l ed stress mode faci l i tati ng the
col l ecti on of creep data, the anal ysi s of materi al s at very l ow shear rates,
and the i nvesti gati on of yi el d stresses. Thi s i nformati on i s needed to
understand the i nternal structure of materi al s. The control l ed rate
mode i s most useful i n obtai ni ng data requi red i n process engi neeri ng
cal cul ati ons. Mechani cal di fferences between control l ed rate and con-
trol l ed stress i nstruments are di scussed i n Sec. 3.7.3. Rotati onal sys-
tems are general l y used toi nvesti gate ti me-dependent behavi or because
tube systems onl y al l ow one pass of the materi al through the apparatus.
A detai l ed di scussi on of osci l l atory testi ng, the pri mary method of
determi ni ng the vi scoel asti c behavi or of food, i s provi ded i n Chapter 5.
1.2 Rheological Instruments for Fluids 3
Figure 1.1. Common rheological instruments divided into two major categories:
rotational and tube type.
There are advantages and di sadvantages associ ated wi th each
i nstrument. Gravi ty operated gl ass capi l l ari es, such as the Cannon-
Fenske type shown i n Fi g. 1.1, are onl y sui tabl e for Newtoni an fl ui ds
because the shear rate vari es duri ng di scharge. Cone and pl ate systems
are l i mi ted to moderate shear rates but cal cul ati ons (for smal l cone
angl es) are si mpl e. Pi pe and mi xer vi scometers can handl e much l arger
parti cl es than cone and pl ate, or paral l el pl ate, devi ces. Probl ems
associ ated wi th sl i p and degradati on i n structural l y sensi ti ve materi al s
are mi ni mi zed wi th mi xer vi scometers. Hi gh pressure capi l l ari es
operate at hi gh shear rates but general l y i nvol ve a si gni fi cant end
pressure correcti on. Pi pe vi scometers can be constructed to wi thstand
the ri gors of the producti on or pi l ot pl ant envi ronment.
Rotational Type
Parallel Plate Cone and Plate
Concentric Cylinder
Mixer
Tube Type
Pipe Glass Capillary High Pressure Capillary
Al l the i nstruments presented i n Fi g. 1.1 are "vol ume l oaded"devi ces
wi th contai ner di mensi ons that are cri ti cal i n the determi nati on of
rheol ogi cal properti es. Another common type of i nstrument, known as
a vi brati onal vi scometer, uses the pri nci pl e of "surface l oadi ng"where
the surface of an i mmersed probe (usual l y a sphere or a rod) generates
a shear wave whi ch di ssi pates i n the surroundi ng medi um. A l arge
enough contai ner i s used soshear forces donot reach the wal l and refl ect
back to the probe. Measurements depend onl y on abi l i ty of the sur-
roundi ng fl ui d to damp probe vi brati on. The dampi ng characteri sti c of
a fl ui d i s a functi on of the product of the fl ui d vi scosi ty (of Newtoni an
fl ui ds) and the densi ty. Vi brati onal vi scometers are popul ar as i n-l i ne
i nstruments for process control systems (see Sec. 1.12). I t i s di ffi cul t to
use these uni ts to eval uate fundamental rheol ogi cal properti es of non-
Newtoni an fl ui ds (Ferry, 1977), but the subjecti ve resul ts obtai ned often
correl ate wel l wi th i mportant food qual i ty attri butes. The coagul ati on
ti me and curd fi rmness of renneted mi l k, for exampl e, have been suc-
cessful l y i nvesti gated usi ng a vi brati onal vi scometer (Sharma et al .,
1989).
I nstruments used to eval uate the extensi onal vi scosi ty of materi al s
are di scussed i n Chapter 4. Pul l i ng or stretchi ng a sampl e between
toothed gears, sucki ng materi al i nto opposi ng jets, spi nni ng, or
expl oi ti ng the open si phon phenomenon can generate data for cal cu-
l ati ng tensi l e extensi onal vi scosi ty. I nformati on to determi ne bi axi al
extensi onal vi scosi ty i s provi ded by compressi ng sampl es between
l ubri cated paral l el pl ates. Shear vi scosi ty can al so be eval uated from
unl ubri cated squeezi ng fl ow between paral l el pl ates. A number of
methods are avai l abl e tocal cul ate an average extensi onal vi scosi ty from
data descri bi ng fl ow through a convergence i nto a capi l l ary di e or sl i t.
1.3. Stress and Strain
Si nce rheol ogy i s the study of the deformati on of matter, i t i s essenti al
to have a good understandi ng of stress and strai n. Consi der a rectan-
gul ar bar that, due to a tensi l e force, i s sl i ghtl y el ongated (Fi g. 1.2). The
i ni ti al l ength of the bar i s and the el ongated l ength i s where
wi th representi ng the i ncrease i n l ength. Thi s deformati on
may be thought of i n terms of Cauchy strai n (al so cal l ed engi neeri ng
strai n):
L
o
L
L L
o
+ L L
1.3 Stress and Strain 5
[1.1]
or Hencky strai n (al so cal l ed true strai n) whi ch i s determi ned by
eval uati ng an i ntegral from to :
[1.2]
Figure 1.2. Linear extension of a rectangular bar.
Cauchy and Hencky strai ns are both zero when the materi al i s
unstrai ned and approxi matel y equal at smal l strai ns. The choi ce of
whi ch strai n measure to use i s l argel y a matter of conveni ence and one
can be cal cul ated from the other:
[1.3]
i s preferred for cal cul ati ng strai n resul ti ng from a l arge deformati on.
Another type of deformati on commonl y found i n rheol ogy i s si mpl e
shear. The i dea can be i l l ustrated wi th a rectangul ar bar (Fi g. 1.3) of
hei ght . The l ower surface i s stati onary and the upper pl ate i s l i nearl y
di spl aced by an amount equal to . Each el ement i s subject to the same
l evel of deformati on so the si ze of the el ement i s not rel evant. The angl e
of shear, , may be cal cul ated as
c

L
L
o
L L
o
L
o
L
L
o
1
L
o
L
L
o
L
dL
L
ln(L/L
o
)
L L
0
h
ln(1 +
c
)
h
h
L

[1.4]
Wi th smal l deformati ons, the angl e of shear (i n radi ans) i s equal to the
shear strai n (al so symbol i zed by ), .
Figure 1.3. Shear deformation of a rectangular bar.
Figure 1.4. Typical stresses on a material element.
Stress, defi ned as a force per uni t area and usual l y expressed i n
Pascal (N/m
2
), may be tensi l e, compressi ve, or shear. Ni ne separate
quanti ti es are requi red to compl etel y descri be the state of stress i n a
materi al . A smal l el ement (Fi g. 1.4) may be consi dered i n terms of
tan()
L
h
tan
h
L
11
21
23
22
33
1
2
3
x
x
x
1.3 Stress and Strain 7
Cartesi an coordi nates ( ). Stress i s i ndi cated as where the fi rst
subscri pt refers to the ori entati on of the face upon whi ch the force acts
and the second subscri pt refers to the di recti on of the force. Therefore,
i s a normal stress acti ng i n the pl ane perpendi cul ar to i n the
di recti on of and i s a shear stress acti ng i n the pl ane perpendi cul ar
to i n the di recti on of . Normal stresses are consi dered posi ti ve when
they act outward (acti ng to create a tensi l e stress) and negati ve when
they act i nward (acti ng to create a compressi ve stress).
Stress components may be summari zed as a stress tensor wri tten
i n the form of a matri x:
[1.5]
A rel ated tensor for strai n can al so be expressed i n matri x form. Basi c
l aws of mechani cs, consi deri ng the moment about the axi s under
equi l i bri um condi ti ons, can be used to prove that the stress matri x i s
symmetri cal :
[1.6]
so
[1.7]
[1.8]
[1.9]
meani ng there are onl y si x i ndependent components i n the stress tensor
represented by Eq. [1.5].
Equati ons that show the rel ati onshi p between stress and strai n are
ei ther cal l ed rheol ogi cal equati ons of state or consti tuti ve equati ons. I n
compl ex materi al s these equati ons may i ncl ude other vari abl es such as
ti me, temperature, and pressure. A modul us i s defi ned as the rati o of
stress to strai n whi l e a compl i ance i s defi ned as the rati o of strai n to
stress. The word rheogram refers toa graph of a rheol ogi cal rel ati onshi p.
x
1
, x
2
, x
3

ij
11
x
1
x
1

23
x
2
x
3
ij

11

12

13
21

22

23
31

32

33
1
1
1
1
]
ij

ji
12

21
31

13
32

23
1.4. Solid Behavior
When force i s appl i ed to a sol i d materi al and the resul ti ng stress
versus strai n curve i s a strai ght l i ne through the ori gi n, the materi al i s
obeyi ng Hookes l aw. The rel ati onshi p may be stated for shear stress
and shear strai n as
[1.10]
where i s the shear modul us. Hookean materi al s do not fl ow and are
l i nearl y el asti c. Stress remai ns constant unti l the strai n i s removed
and the materi al returns to i ts ori gi nal shape. Someti mes shape
recovery, though compl ete, i s del ayed by some atomi sti c process. Thi s
ti me-dependent, or del ayed, el asti c behavi or i s known as anel asti ci ty.
Hookes l aw can be used to descri be the behavi or of many sol i ds (steel ,
egg shel l , dry pasta, hard candy, etc.) when subjected to smal l strai ns,
typi cal l y l ess than 0.01. Large strai ns often produce bri ttl e fracture or
non-l i near behavi or.
The behavi or of a Hookean sol i d may be i nvesti gated by studyi ng
the uni axi al compressi on of a cyl i ndri cal sampl e (Fi g. 1.5). I f a materi al
i s compressed so that i t experi ences a change i n l ength and radi us, then
the normal stress and strai n may be cal cul ated:
[1.11]
[1.12]
Figure 1.5 Uniaxial compression of a cylindrical sample.
12
G
G
22

F
A

F
(R
o
)
2
c

h
h
o
R
h
h
o
R R
h
Initial Shape Compressed Shape
o
o
1.4 Solid Behavior 9
Thi s i nformati on can be used to determi ne Youngs modul us ( ), al so
cal l ed the modul us of el asti ci ty, of the sampl e:
[1.13]
I f the deformati ons are l arge, Hencky strai n ( ) shoul d be used to
cal cul ate strai n and the area term needed i n the stress cal cul ati on
shoul d be adjusted for the change i n radi us caused by compressi on:
[1.14]
A cri ti cal assumpti on i n these cal cul ati ons i s that the sampl e remai ns
cyl i ndri cal i n shape. For thi s reason l ubri cated contact surfaces are
often recommended when testi ng materi al s such as food gel s.
Youngs modul us may al so be determi ned by fl exural testi ng of
beams. I n one such test, a canti l ever beam of known l ength (a) i s
defl ected a di stance (d) when a force (F) i s appl i ed to the free end of the
beam. Thi s i nformati on can be used to cal cul ate Youngs modul us for
materi al s havi ng a rectangul ar or ci rcul ar crossecti onal area (Fi g. 1.6).
Si mi l ar cal cul ati ons can be performed i n a three-poi nt bendi ng test (Fi g.
1.7) where defl ecti on (d) i s measured when a materi al i s subjected to a
force (F) pl aced mi dway between two supports. Cal cul ati ons are si ghtl y
di fferent dependi ng on wether-or-not the test materi al has a rectangul ar
or ci rcul ar shape. Other si mpl e beam tests, such as the doubl e canti l ever
or four-poi nt bendi ng test, yi el d comparabl e resul ts. Fl exural testi ng
may have appl i cati on to sol i d foods havi ng a wel l defi ned geometry such
as dry pasta or hard candy.
I n addi ti on to Youngs modul us, Poi ssons rati o ( ) can be defi ned
from compressi on data (Fi g. 1.5):
[1.15]
Poi ssons rati o may range from 0 to 0.5. Typi cal l y, vari es from 0.0 for
ri gi d l i ke materi al s contai ni ng l arge amounts of ai r to near 0.5 for l i qui d
l i ke materi al s. Val ues from 0.2 to 0.5 are common for bi ol ogi cal
materi al s wi th 0.5 representi ng an i ncompressi bl e substance l i ke potato
E
E

22
22

F
(R
o
+ R)
2

lateral strain
axial strain

R/R
o
h/h
o

fl esh. Ti ssues wi th a hi gh l evel of cel l ul ar gas, such as appl e fl esh, woul d
have val ues cl oser to 0.2. Metal s usual l y have Poi sson rati os between
0.25 and 0.35.
Figure 1.6 Deflection of a cantilever beam to determine Youngs modulus.
Figure 1.7 Three-point beam bending test to determine Youngs modulus (b, h,
and D are defined on Fig. 1.6).
E = 4Fa /(dbh )
E = 64Fa /(3d D )
Rectangular Crossection Circular Crossection
3 3 4 3
h
b D
d
F
a
E = Fa /(4dbh )
E = 4Fa /(3d D )
Rectangular Crossection Circular Crossection
4 3
3 3
F
d
a
1.4 Solid Behavior 11
I f a materi al i s subjected to a uni form change i n external pressure,
i t may experi ence a vol umetri c change. These quanti ti es are used to
defi ne the bul k modul us ( ):
[1.16]
The bul k modul us of dough i s approxi matel y Pa whi l e the val ue for
steel i s Pa. Another common property, bul k compressi bi l i ty, i s
defi ned as the reci procal of bul k modul us.
When two materi al constants descri bi ng the behavi or of a Hookean
sol i d are known, the other two can be cal cul ated usi ng any of the fol -
l owi ng theoreti cal rel ati onshi ps:
[1.17]
[1.18]
[1.19]
Numerous experi mental techni ques, appl i cabl e to food materi al s, may
be used to determi ne Hookean materi al constants. Methods i ncl ude
testi ng i n tensi on, compressi on and torsi on (Mohseni n, 1986; Pol a-
kowski and Ri pl i ng, 1966; Dal l y and Ri pl ey, 1965). Hookean properti es
of typi cal materi al s are presented i n the Appendi x [6.5].
Li near-el asti c and non-l i near el asti c materi al s (l i ke rubber) both
return to thei r ori gi nal shape when the strai n i s removed. Food may
be sol i d i n nature but not Hookean. A compari son of curves for l i near
el asti c (Hookean), el astopl asti c and non-l i near el asti c materi al s (Fi g.
1.8) shows a number of si mi l ari ti es and di fferences. The el astopl asti c
materi al has Hookean type behavi or bel ow the yi el d stress ( )but fl ows
to produce permanent deformati on above that val ue. Margari ne and
butter, at room temperature, may behave as el astopl asti c substances.
I nvesti gati on of thi s type of materi al , as a sol i d or a fl ui d, depends on
the shear stress bei ng above or bel ow (see Sec. 1.6 for a more detai l ed
di scussi on of the yi el d stress concept and Appendi x [6.7] for typi cal yi el d
stress val ues). Furthermore, to ful l y di sti ngui sh fl ui d from sol i d l i ke
behavi or, the characteri sti c ti me of the materi al and the characteri sti c
ti me of the deformati on process i nvol ved must be consi dered si mul ta-
K
K
change in pressure
volumetric strain

change in pressure
(change in volume/original volume)
10
6
10
11
1
E

1
3G
+
1
9K
E 3K(1 2) 2G(1 + )

3K E
6K

E 2G
2G
o
neousl y. The Deborah number has been defi ned to address thi s i ssue.
A detai l ed di scussi on of the concept, i ncl udi ng an exampl e i nvol vi ng
si l l y putty (the "real sol i d-l i qui d") i s presented i n Sec. 5.5.
Figure 1.8. Deformation curves for linear elastic (Hookean), elastoplastic and
non-linear elastic materials.
Food rheol ogi sts al so fi nd the fai l ure behavi or of sol i d food (parti c-
ul arl y, bri ttl e materi al s and fi rm gel s) to be very useful because these
data someti mes correl ate wel l wi th the concl usi ons of human sensory
panel s (Hamann, 1983; Montejano et al ., 1985; Kawanari et al ., 1981).
The fol l owi ng termi nol ogy (taken from Ameri can Soci ety for Testi ng and
Materi al s, Standard E-6) i s useful i n descri bi ng the l arge deformati on
behavi or i nvol ved i n the mechani cal fai l ure of food:
el asti c l i mi t - the greatest stress whi ch a materi al i s capabl e of sus-
tai ni ng wi thout any permanent strai n remai ni ng upon compl ete
rel ease of stress;
proporti onal l i mi t - the greatest stress whi ch a materi al i s capabl e of
sustai ni ng wi thout any devi ati on from Hookes Law;
compressi ve strength - the maxi mum compressi ve stress a materi al
i s capabl e of sustai ni ng;
shear strength - the maxi mum shear stress a materi al i s capabl e of
sustai ni ng;
tensi l e strength - the maxi mum tensi l e stress a materi al i s capabl e
of sustai ni ng;
yi el d poi nt - the fi rst stress i n a test where the i ncrease i n strai n
occurs wi thout an i ncrease i n stress;
o
Linear Elastic Elastoplastic Non-Linear Elastic
Permanent
Deformation
12 12 12
1.5.1 Time-Independent Material Functions 13
yi el d strength - the engi neeri ng stress at whi ch a materi al exhi bi ts
a speci fi ed l i mi ti ng devi ati on from the proporti onal i ty of stress to
strai n.
A typi cal characteri sti c of bri ttl e sol i ds i s that they break when gi ven
a smal l deformati on. Fai l ure testi ng and fracture mechani cs i n struc-
tural sol i ds are wel l devel oped areas of materi al sci ence (Cal l i ster, 1991)
whi ch offer much to the food rheol ogi st. Eval uati ng the structural
fai l ure of sol i d foods i n compressi on, torsi on, and sandwi ch shear modes
were summari zed by Hamann (1983). Jagged force-deformati on rel a-
ti onshi ps of crunchy materi al s may offer al ternati ve texture cl assi fi -
cati on cri teri a for bri ttl e or crunchy foods (Ul bri cht et al ., 1995; Pel eg
and Normand, 1995).
1.5. Fluid Behavior in Steady Shear Flow
1.5.1. Time-Independent Material Functions
Viscometric Functions. Fl ui ds may be studi ed by subjecti ng them
to conti nuous sheari ng at a constant rate. I deal l y, thi s can be accom-
pl i shed usi ng two paral l el pl ates wi th a fl ui d i n the gap between them
(Fi g. 1.9). The l ower pl ate i s fi xed and the upper pl ate moves at a
constant vel oci ty ( ) whi ch can be thought of as an i ncremental change
i n posi ti on di vi ded by a smal l ti me peri od, . A force per uni t area
on the pl ate i s requi red for moti on resul ti ng i n a shear stress ( ) on
the upper pl ate whi ch, conceptual l y, coul d al so be consi dered to be a
l ayer of fl ui d.
The fl ow descri bed above i s steady si mpl e shear and the shear rate
(al so cal l ed the strai n rate) i s defi ned as the rate of change of strai n:
[1.20]
Thi s defi ni ti on onl y appl i es to streaml i ne (l ami nar) fl ow between
paral l el pl ates. I t i s di rectl y appl i cabl e to sl i di ng pl ate vi scometer
descri bed by Deal y and Gi acomi n (1988). The defi ni ti on must be
adjusted to account for curved l i nes such as those found i n tube and
rotati onal vi scometers; however, the i dea of "maxi mum speed di vi ded
by gap si ze"can be useful i n esti mati ng shear rates found i n parti cul ar
appl i cati ons l i ke brush coati ng. Thi s i dea i s expl ored i n more detai l i n
Sec. 1.9.
u
L/t
21

d
dt

d
dt
L
h
_
,

u
h
Figure 1.9. Velocity profile between parallel plates.
Rheol ogi cal testi ng to determi ne steady shear behavi or i s conducted
under l ami nar fl ow condi ti ons. I n turbul ent fl ow, l i ttl e i nformati on i s
generated that can be used to determi ne materi al properti es. Al so, to
be meani ngful , data must be col l ected over the shear rate range
appropri ate for the probl em i n questi on whi ch may vary wi del y i n
i ndustri al processes (Tabl e 1.1): Sedi mentati on of parti cl es may i nvol ve
very l ow shear rates, spray dryi ng wi l l i nvol ve hi gh shear rates, and
pi pe fl ow of food wi l l usual l y occur over a moderate shear rate range.
Extrapol ati ng experi mental data over a broad range of shear rates i s
not recommended because i t may i ntroduce si gni fi cant errors when
eval uati ng rheol ogi cal behavi or.
Materi al fl ow must be consi dered i n three di mensi ons to compl etel y
descri be the state of stress or strai n. I n steady, si mpl e shear fl ow the
coordi nate system may be ori ented wi th the di recti on of fl ow sothe stress
tensor gi ven by Eq. [1.5] reduces to
[1.21]
h
VELOCITY PROFILE
FORCE
AREA
u
u = 0
1
2
x
x
ij

11

12
0
21

22
0
0 0
33
1
1
1
1
]
Table 1.1. Shear Rates Typical of Familiar Materials and Processes
Situation (1/s) Application
Sedimentation of particles in 10
-6
- 10
-3
Medicines, paints, spices in
a suspending liquid salad dressing
Leveling due to surface ten- 10
-2
- 10
-1
Frosting, paints, printing inks
sion
Draining under gravity 10
-1
- 10
1
Vats, small food containers,
painting and coating
Extrusion 10
0
- 10
3
Snack and pet foods, tooth-
paste, cereals, pasta, poly-
mers
Calendering 10
1
- 10
2
Dough Sheeting
Pouring from a bottle 10
1
- 10
2
Foods, cosmetics, toiletries
Chewing and swallowing 10
1
- 10
2
Foods
Dip coating 10
1
- 10
2
Paints, confectionery
Mixing and stirring 10
1
- 10
3
Food processing
Pipe flow 10
0
- 10
3
Food processing, blood flow
Rubbing 10
2
- 10
4
Topical application of creams
and lotions
Brushing 10
3
- 10
4
Brush painting, lipstick, nail
polish
Spraying 10
3
- 10
5
Spray drying, spray painting,
fuel atomization
High speed coating 10
4
- 10
6
Paper
Lubrication 10
3
- 10
7
Bearings, gasoline engines
Si mpl e shear fl ow i s al so cal l ed vi scometri c fl ow. I t i ncl udes axi al
fl ow i n a tube, rotati onal fl ow between concentri c cyl i nders, rotati onal
fl ow between a cone and a pl ate, and torsi onal fl ow (al so rotati onal )
between paral l el pl ates. I n vi scometri cfl ow, three shear-rate-dependent
materi al functi ons, col l ecti vel y cal l ed vi scometri c functi ons, are needed
to compl etel y establ i sh the state of stress i n a fl ui d. These may be
descri bed as the vi scosi ty functi on, , and the fi rst and second normal
stress coeffi ci ents, and , defi ned mathemati cal l y as
[1.22]
[1.23]
1
(
)
2
(
)
f (
)

21
1
f (
)

11

22
(
)
2

N
1
(
)
2
[1.24]
The fi rst ( ) and second ( ) normal stress di fferences are
often symbol i cal l y represented as and , respecti vel y. i s al ways
posi ti ve and consi dered to be approxi matel y 10 ti mes greater than .
Measurement of i s di ffi cul t; fortunatel y, the assumpti on that
i s usual l y sati sfactory. The rati oof , known as the recoverabl e shear
(or the recoverabl e el asti c strai n), has proven to be a useful parameter
i n model i ng di e swel l phenomena i n pol ymers (Tanner, 1988). Some
data on the val ues of fl ui d foods have been publ i shed (see Appendi x
[6.6]).
I f a fl ui d i s Newtoni an, i s a constant (equal to the Newtoni an
vi scosi ty) and the fi rst and second normal stress di fferences are zero.
As approaches zero, el asti c fl ui ds tend to di spl ay Newtoni an behavi or.
Vi scoel asti c fl ui ds si mul taneousl y exhi bi t obvi ous fl ui d-l i ke (vi scous)
and sol i d-l i ke (el asti c) behavi or. Mani festati ons of thi s behavi or, due
to a hi gh el asti c component, can be very strong and create di ffi cul t
probl ems i n process engi neeri ng desi gn. These probl ems are parti cu-
l arl y preval ent i n the pl asti c processi ng i ndustri es but al so present i n
processi ng foods such as dough, parti cul arl y those contai ni ng l arge
quanti ti es of wheat protei n.
Fi g. 1.10 i l l ustrates several phenomena. Duri ng mi xi ng or agi tati on,
a vi scoel asti cfl ui d may cl i mb the i mpel l er shaft i n a phenomenon known
as the Wei ssenberg effect (Fi g. 1.10). Thi s can be observed i n home
mi xi ng of cake or chocol ate browni e batter. When a Newtoni an fl ui d
emerges from a l ong, round tube i nto the ai r, the emergi ng jet wi l l
normal l y contract; at l ow Reynol ds numbers i t may expand toa di ameter
whi ch i s 10 to 15% l arger than the tube di ameter. Normal stress di f-
ferences present i n a vi scoel asti c fl ui d, however, may cause jet expan-
si ons (cal l ed di e swel l ) whi ch are two or more ti mes the di ameter of the
tube (Fi g. 1.10). Thi s behavi or contri butes to the chal l enge of desi gni ng
extruder di es toproduce the desi red shape of many pet, snack, and cereal
foods. Mel t fracture, a fl ow i nstabi l i ty causi ng di storted extrudates, i s
al soa probl em rel ated to fl ui d vi scoel asti ci ty. I n addi ti on, hi ghl y el asti c
fl ui ds may exhi bi t a tubel ess si phon effect (Fi g. 1.10).
2
f (
)

22

33
(
)
2

N
2
(
)
2
11

22

22

33
N
1
N
2
N
1
N
2
N
2
N
2
0
N
1
/
12
N
1
(

Figure 1.10. Weissenberg effect (fluid climbing a rotating rod), tubeless siphon
and jet swell of viscous (Newtonian) and viscoelastic fluids.
The recoi l phenomena (Fi g. 1.11), where tensi l e forces i n the fl ui d
cause parti cl es to move backward (snap back)when fl ow i s stopped, may
al so be observed i n vi scoel asti c fl ui ds. Other i mportant effects i ncl ude
drag reducti on, extrudate i nstabi l i ty, and vortex i nhi bi ti on. An excel -
l ent pi ctori al summary of the behavi or of vi scoel asti c pol ymer sol uti ons
i n vari ous fl ow si tuati ons has been prepared by Boger and Wal ters
(1993).
Normal stress data can be col l ected i n steady shear fl ow usi ng a
number of di fferent techni ques (Deal y, 1982): exi t pressure di fferences
i n capi l l ary and sl i t fl ow, axi al fl ow i n an annul us, wal l pressure i n
concentri ccyl i nder fl ow, axi al thrust i n cone and pl ate as wel l as paral l el
pl ate fl ow. I n general , these methods have been devel oped for pl asti c
mel ts (and rel ated pol ymeri c materi al s) wi th the probl ems of the pl asti c
i ndustri es provi di ng the mai n dri vi ng force for change.
VISCOUS FLUID VISCOELASTIC FLUID
WEISSENBERG
EFFECT
TUBELESS
SIPHON
JET SWELL
Cone and pl ate systems are most commonl y used for obtai ni ng
pri mary normal stress data and a number of commerci al i nstruments
are avai l abl e to make these measurements. Obtai ni ng accurate data
for food materi al s i s compl i cated by vari ous factors such as the presence
of a yi el d stress, ti me-dependent behavi or and chemi cal reacti ons
occurri ng duri ng processi ng (e.g., hydrati on, protei n denaturati on, and
starch gel ati ni zati on). Rheogoni ometer i s a term someti mes used to
descri be an i nstrument capabl e of measuri ng both normal and shear
stresses. Detai l ed i nformati on on testi ng vi scoel asti c pol ymers can be
found i n numerous books: Bi rd et al . (1987), Barnes et al . (1989), Bogue
and Whi te (1970), Darby (1976), Macosko (1994), and Wal ters (1975).
Figure 1.11. Recoil phenomenon in viscous (Newtonian) and viscoelastic fluids.
Vi scometri c functi ons have been very useful i n understandi ng the
behavi or of syntheti c pol ymer sol uti ons and mel ts (pol yethyl ene, pol y-
propyl ene, pol ystyrene, etc.). From an i ndustri al standpoi nt, the vi s-
cosi ty functi on i s most i mportant i n studyi ng fl ui d foods and much of
the current work i s appl i ed to that area. To date, normal stress data
for foods have not been wi del y used i n food process engi neeri ng. Thi s
i s partl y due tothe fact that other factors often compl i cate the eval uati on
of the fl ui d dynami cs present i n vari ous probl ems. I n food extrusi on,
for exampl e, fl ashi ng (vapori zati on) of water when the product exi ts the
START
STOP
RECOIL
VISCOUS FLUID VISCOELASTIC FLUID
START
STOP
di e makes i t di ffi cul t topredi ct the i nfl uence of normal stress di fferences
on extrudate expansi on. Future research may create si gni fi cant
advances i n the use of normal stress data by the food i ndustry.
Mathematical Modelsfor Inelastic Fluids. The el asti c behavi or of
many fl ui d foods i s smal l or can be negl ected (materi al s such as dough
are the excepti on) l eavi ng the vi scosi ty functi on as the mai n area of
i nterest. Thi s functi on i nvol ves shear stress and shear rate: the rel a-
ti onshi p between the two i s establ i shed from experi mental data.
Behavi or i s vi sual i zed as a pl ot of shear stress versus shear rate, and
the resul ti ng curve i s mathemati cal l y model ed usi ng vari ous functi onal
rel ati onshi ps. The si mpl est type of substance to consi der i s the New-
toni an fl ui d where shear stress i s di rectl y proporti onal to shear rate [for
conveni ence the subscri pt on wi l l be dropped i n the remai nder of the
text when deal i ng excl usi vel y wi th one di mensi onal fl ow]:
[1.25]
wi th bei ng the constant of proporti onal i ty appropri ate for a Newtoni an
fl ui d. Usi ng uni ts of N, m
2
, m, m/s for force, area, l ength and vel oci ty
gi ves vi scosi ty as Pa s whi ch i s 1 poi seui l l e or 1000 centi poi se (note: 1
Pa s = 1000 cP = 1000 mPa s; 1 P = 100 cP). Dynami c vi scosi ty and
coeffi ci ent of vi scosi ty are synonyms for the term "vi scosi ty"i n referri ng
to Newtoni an fl ui ds. The reci procal of vi scosi ty i s cal l ed fl ui di ty.
Coeffi ci ent of vi scosi ty and fl ui di ty are i nfrequentl y used terms.
Newtoni an fl ui ds may al so be descri bed i n terms of thei r ki nemati c
vi scosi ty ( ) whi ch i s equal to the dynami c vi scosi ty di vi ded by densi ty
( ). Thi s i s a common practi ce for non-food materi al s, parti cul arl y
l ubri cati ng oi l s. Vi scosi ty conversi on factors are avai l abl e i n Appendi x
[6.1].
Newtoni an fl ui ds, by defi ni ti on, have a strai ght l i ne rel ati onshi p
between the shear stress and the shear rate wi th a zero i ntercept. Al l
fl ui ds whi ch do not exhi bi t thi s behavi or may be cal l ed non-Newtoni an.
Looki ng at typi cal Newtoni an fl ui ds on a rheogram (Fi g. 1.12) reveal s
that the sl ope of the l i ne i ncreases wi th i ncreasi ng vi scosi ty.
Van Wazer et al . (1963)di scussed the sensi ti vi ty of the eye i n judgi ng
vi scosi ty of Newtoni an l i qui ds. I t i s di ffi cul t for the eye to di sti ngui sh
di fferences i n the range of 0.1 to 10 cP. Smal l di fferences i n vi scosi ty
are cl earl y seen from approxi matel y 100 to 10,000 cP: somethi ng at 800
cP may l ook twi ce as thi ck as somethi ng at 400 cP. Above 100,000 cP
i t i s di ffi cul t to make vi sual di sti ncti ons because the materi al s do not
21

/
Figure 1.12. Rheograms for typical Newtonian fluids.
pour and appear, to the casual observer, as sol i ds. As poi nts of reference
the fol l owi ng represent typi cal Newtoni an vi scosi ti es at room temper-
ature: ai r, 0.01 cP; gasol i ne (petrol ), 0.7 cP; water, 1 cP; mercury, 1.5
cP; coffee cream or bi cycl e oi l , 10 cP; vegetabl e oi l , 100 cP; gl ycerol , 1000
cP; gl yceri ne, 1500 cP; honey, 10,000 cP; tar, 1,000,000 cP. Data for
many Newtoni an fl ui ds at di fferent temperatures are presented i n
Appendi ces [6.8], [6.9], and [6.10].
A general rel ati onshi p to descri be the behavi or of non-Newtoni an
fl ui ds i s the Herschel -Bul kl ey model :
[1.26]
where i s the consi stency coeffi ci ent, i s the fl ow behavi or i ndex, and
i s the yi el d stress. Thi s model i s appropri ate for many fl ui d foods.
Eq. [1.26] i s very conveni ent because Newtoni an, power l aw (shear-
thi nni ng when or shear-thi ckeni ng when ) and Bi ng-
ham pl asti c behavi or may be consi dered as speci al cases (Tabl e 1.2, Fi g.
1.13). Wi th the Newtoni an and Bi ngham pl asti c model s, i s commonl y
cal l ed the vi scosi ty ( ) and pl asti c vi scosi ty ( ), respecti vel y. Shear-
thi nni ng and shear-thi ckeni ng are al so referred to as pseudopl asti c and
di l atent behavi or, respecti vel y; however, shear-thi nni ng and
0 5 10 15 20
0
0.5
1
1.5
2
2.5
3
Shear Rate, 1/s
40% fat cream, 6.9 cP
olive oil, 36.3 cP
castor oil, 231 cP
0 5 10 15 20
0
0.5
1
1.5
2
2.5
3
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
40% fat cream, 6.9 cP
olive oil, 36.3 cP
castor oil, 231 cP
K(
)
n
+
o
K n
o
0 < n < 1 1 < n <
K

pl
shear-thi ckeni ng are the preferred terms. A typi cal exampl e of a
shear-thi nni ng materi al i s found i n the fl ow behavi or of a 1% aqueous
sol uti on of carrageenan gum as demonstrated i n Exampl e Probl em
1.14.1. Shear-thi ckeni ng i s consi dered wi th a concentrated corn starch
sol uti on i n Exampl e Probl em 1.14.2.
Table 1.2. Newtonian, Power Lawand BinghamPlastic Fluids as Special Cases of
the Herschel-Bulkley Model (Eq. [1.26])
Fluid K n Typical Examples
Herschel-Bulkley > 0 0 < n < > 0 minced fish paste,
raisin paste
Newtonian > 0 1 0 water, fruit juice,
milk, honey, vegeta-
ble oil
shear-thinning > 0 0 < n < 1 0 applesauce, banana
(pseudoplastic) puree, orange juice
concentrate
shear-thickening > 0 1 < n < 0 some types of
(dilatent) honey, 40% raw
corn starch solution
Bingham plastic > 0 1 > 0 tooth paste, tomato
paste
An i mportant characteri sti c of the Herschel -Bul kl ey and Bi ngham
pl asti c materi al s i s the presence of a yi el d stress ( ) whi ch represents
a fi ni te stress requi red to achi eve fl ow. Bel ow the yi el d stress a materi al
exhi bi ts sol i d l i ke characteri sti cs: I t stores energy at smal l strai ns and
does not l evel out under the i nfl uence of gravi ty to form a fl at surface.
Thi s characteri sti c i s very i mportant i n process desi gn and qual i ty
assessment for materi al s such as butter, yogurt and cheese spread. The
yi el d stress i s a practi cal , but i deal i zed, concept that wi l l recei ve addi -
ti onal di scussi on i n a l ater secti on (Sec. 1.6). Typi cal yi el d stress val ues
may be found i n Appendi x [6.7].
o
Figure 1.13. Curves for typical time-independent fluids.
Figure 1.14. Rheogram of idealized shear-thinning (pseudoplastic) behavior.
Newtonian
Shear-Thickening
Shear-Thinning
Herschel-Bulkley
Bingham
S
h
e
a
r

S
t
r
e
s
s
,

P
a
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
Shear Rate, 1/s
Lower Region
Upper Region
Middle Region
Slope =
Slope =
Shear-thi nni ng behavi or i s very common i n frui t and vegetabl e
products, pol ymer mel ts, as wel l as cosmeti c and toi l etry products
(Appendi ces [6.11], [6.12], [6.13]). Duri ng fl ow, these materi al s may
exhi bi t three di sti nct regi ons (Fi g. 1.14): a l ower Newtoni an regi on
where the apparent vi scosi ty ( ), cal l ed the l i mi ti ng vi scosi ty at zero
shear rate, i s constant wi th changi ng shear rates; a mi ddl e regi on where
the apparent vi scosi ty ( ) i s changi ng (decreasi ng for shear-thi nni ng
fl ui ds) wi th shear rate and the power l aw equati on i s a sui tabl e model
for the phenomenon; and an upper Newtoni an regi on where the sl ope
of the curve ( ), cal l ed the l i mi ti ng vi scosi ty at i nfi ni te shear rate, i s
constant wi th changi ng shear rates. The mi ddl e regi on i s most often
exami ned when consi deri ng the performance of food processi ng equi p-
ment. The l ower Newtoni an regi on may be rel evant i n probl ems
i nvol vi ng l ow shear rates such as those rel ated to the sedi mentati on of
fi ne parti cl es i n fl ui ds. Val ues of for some vi scoel asti c fl ui ds are gi ven
i n Tabl e 5.4.
Numerous factors i nfl uence the sel ecti on of the rheol ogi cal model
used to descri be fl ow behavi or of a parti cul ar fl ui d. Many model s, i n
addi ti on to the power l aw, Bi ngham pl asti c and Herschel -Bul kl ey
model s, have been used torepresent the fl ow behavi or of non-Newtoni an
fl ui ds. Some of them are summari zed i n Tabl e 1.3. The Cross,
Rei ner-Phi l i ppoff, Van Wazer and Powel l -Eyri ng model s are useful i n
model i ng pseudopl asti c behavi or over l ow, mi ddl e and hi gh shear rate
ranges. Some of the equati ons, such as the Modi fi ed Casson and the
General i zed Herschel -Bul kl ey, have proven useful i n devel opi ng
mathemati cal model s to sol ve food process engi neeri ng probl ems (Ofol i
et al ., 1987) i nvol vi ng wi de shear rate ranges. Addi ti onal rheol ogi cal
model s have been summari zed by Hol dsworth (1993).
The Casson equati on has been adopted by the I nternati onal Offi ce
of Cocoa and Chocol ate for i nterpreti ng chocol ate fl ow behavi or. The
Casson and Bi ngham pl asti c model s are si mi l ar because they both have
a yi el d stress. Each, however, wi l l gi ve di fferent val ues of the fl ui d
parameters dependi ng on the data range used i n the mathemati cal
anal ysi s. The most rel i abl e val ue of a yi el d stress, when determi ned
from a mathemati cal i ntercept, i s found usi ng data taken at the l owest
shear rates. Thi s concept i s demonstrated i n Exampl e Probl em 1.14.3
for mi l k chocol ate.
o
Apparent Viscosity. Apparent vi scosi ty has a preci se defi ni ti on. I t
i s, as noted i n Eq. [1.22], shear stress di vi ded by shear rate:
[1.27]
Wi th Newtoni an fl ui ds, the apparent vi scosi ty and the Newtoni an
vi scosi ty ( ) are i denti cal but for a power l aw fl ui d i s
[1.28]
Table 1.3. Rheological Models to Describe the Behavior of Time-
independent Fluids
Model (Source) Equation
*
Casson (Casson, 1959)
Modified Casson (Mizrahi and Berk,
1972)
Ellis (Ellis, 1927)
Generalized Herschel-Bulkley (Ofoli et
al., 1987)
Vocadlo (Parzonka and Vocadlo, 1968)
Power Series (Whorlow, 1992)
Carreau (Carreau, 1968)
Cross (Cross, 1965)
Van Wazer (Van Wazer, 1963)
Powell-Eyring (Powell and Eyring, 1944)
Reiner-Philippoff (Philippoff, 1935)
*
and are arbitrary constants and power indices, respectively, determined
from experimental data.
f (

f (
)
K(
)
n
K(
)
n 1
0.5
(
o
)
0.5
+ K
1
(
)
0.5
0.5
(
o
)
0.5
+ K
1
(
)
n
1
K
1
+ K
2
()
n
1
n
1
(
o
)
n
1
+ K
1
(
)
n
2

(
o
)
1/n
1
+ K
1
_
,
n
1
K
1
+ K
2
()
3
+ K
3
()
5
K
1
+ K
2

()
3
+ K
3

()
5
+ (
o

1 + (K
1
)
21
]
(n 1)/2

+

o

1 + K
1
(
)
n

o

1 + K
1
+ K
2
(
)
n
1
+
K
1
1
K
2
_
,
sinh
1
(K
3
+

o

1 + (()
2
/K
1
)
_
K
1
, K
2
, K
3
n
1
, n
2
Apparent vi scosi ti es for Bi ngham pl asti c and Herschel -Bul kl ey fl ui ds
are determi ned i n a l i ke manner:
[1.29]
[1.30]
decreases wi th i ncreasi ng shear rate i n shear-thi nni ng and Bi ngham
pl asti c substances. I n Herschel -Bul kl ey fl ui ds, apparent vi scosi ty wi l l
decrease wi th hi gher shear rates when , but behave i n the
opposi te manner when Apparent vi scosi ty i s constant wi th
Newtoni an materi al s and i ncreases wi th i ncreasi ng shear rate i n
shear-thi ckeni ng fl ui ds (Fi g. 1.15).
Figure 1.15. Apparent viscosity of time-independent fluids.
A si ngl e poi nt apparent vi scosi ty val ue i s someti mes used as a
measure of mouthfeel of fl ui d foods: The human percepti on of thi ckness
i s correl ated tothe apparent vi scosi ty at approxi matel y 60 s
-1
. Apparent
vi scosi ty can al sobe used to i l l ustrate the axi om that taki ng si ngl e poi nt
tests for determi ni ng the general behavi or of non-Newtoni an materi al s
may cause seri ous probl ems. Some qual i ty control i nstruments desi gned
for si ngl e poi nt tests may produce confusi ng resul ts. Consi der, for
f (
)
K(
) +
o
K +

o
f (
)
K(
)
n
+
o
K(
)
n 1
+

o
0<n <1.0
n >1.0.
Shear Rate, 1/s
A
p
p
a
r
e
n
t

V
i
s
c
o
s
i
t
y
,

P
a

s
Shear-Thickening
Shear-Thinning
Herschel-Bulkley
Bingham
Newtonian
Time-Independent Fluids
( 0 < n < 1.0 )
exampl e, the two Bi ngham pl asti c materi al s shown i n Fi g. 1.16. The
two curves i ntersect at 19.89 1/s and an i nstrument measuri ng the
apparent vi scosi ty at that shear rate, for each fl ui d, woul d gi ve i denti cal
resul ts: = 1.65 Pa s. However, a si mpl e exami nati on of the materi al
wi th the hands and eyes woul d show them to be qui te di fferent because
the yi el d stress of one materi al i s more than 4 ti mes that of the other
materi al . Cl earl y, numerous data poi nts (mi ni mum of twofor the power
l aw or Bi ngham pl asti c model s) are requi red to eval uate the fl ow
behavi or of non-Newtoni an fl ui ds.
Figure 1.16. Rheograms for two Bingham plastic fluids.
Solution Viscosities. I t i s someti mes useful to determi ne the vi sco-
si ti es of di l ute syntheti c or bi opol ymer sol uti ons. When a pol ymer i s
di ssol ved i n a sol vent, there i s a noti ceabl e i ncrease i n the vi scosi ty of
the resul ti ng sol uti on. The vi scosi ti es of pure sol vents and sol uti ons
can be measured and vari ous val ues cal cul ated from the resul ti ng data:
[1.31]
[1.32]
0 10 20 30 40 50 60
0
20
40
60
80
100
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
Yield Stress = 25.7 Pa
Plastic Viscosity = .36 Pa s
Plastic Viscosity = 1.35 Pa s
Bingham Plastic Fluids
relative viscosity
rel

solution
solvent
specific viscosity
sp

rel
1
1.5.2 Time-Dependent Material Functions 27
[1.33]
[1.34]
[1.35]
where i s the mass concentrati on of the sol uti on i n uni ts of g/dl or
g/100ml . Note that uni ts of reduced, i nherent, and i ntri nsi c vi scosi ty
are reci procal concentrati on (usual l y deci l i ters of sol uti on per grams of
pol ymer). The i ntri nsi c vi scosi ty has great practi cal val ue i n mol ecul ar
wei ght determi nati ons of hi gh pol ymers (Severs, 1962; Morton-Jones,
1989; Grul ke, 1994). Thi s concept i s based on the Mark-Houwi nk
rel ati on suggesti ng that the i ntri nsi c vi scosi ty of a di l ute pol ymer
sol uti on i s proporti onal to the average mol ecul ar wei ght of the sol ute
rai sed to a power i n the range of 0.5 to 0.9. Val ues of the proporti onal i ty
constant and the exponent are wel l known for many pol ymer-sol vent
combi nati ons (Progel f and Throne, 1993; Rodri quez, 1982). Sol uti on
vi scosi ti es are useful i n understandi ng the behavi or of some bi opol ymers
i ncl udi ng aqueous sol uti ons of l ocust bean gum, guar gum, and car-
boxymethyl cel l ul ose (Rao, 1986). The i ntri nsi c vi scosi ti es of numerous
protei n sol uti ons have been summari zed by Rha and Pradi pasena
(1986).
1.5.2. Time-Dependent Material Functions
I deal l y, ti me-dependent materi al s are consi dered tobe i nel asti cwi th
a vi scosi ty functi on whi ch depends on ti me. The response of the sub-
stance to stress i s i nstantaneous and the ti me-dependent behavi or i s
due to changes i n the structure of the materi al i tsel f. I n contrast, ti me
effects found i n vi scoel asti c materi al s ari se because the response of
stress to appl i ed strai n i s not i nstantaneous and not associ ated wi th a
structural change i n the materi al . Al so, the ti me scal e of thi xotropy may
be qui te di fferent than the ti me scal e associ ated wi th vi scoel asti ci ty:
The most dramati c effects are usual l y observed i n si tuati ons i nvol vi ng
short process ti mes. Note too, that real materi al s may be both ti me-
dependent and vi scoel asti c!
reduced viscosity
red

sp
C
inherent viscosity
inh

ln
rel
C
intrinsic viscosity
int

sp
C
1
1
]C 0
C
Figure 1.17. Time-dependent behavior of fluids.
Separate termi nol ogy has been devel oped to descri be fl ui ds wi th
ti me-dependent characteri sti cs. Thi xotropi c and rheopecti c materi al s
exhi bi t, respecti vel y, decreasi ng and i ncreasi ng shear stress (and
apparent vi scosi ty)over ti me at a fi xed rate of shear (Fi g. 1.17). I n other
words, thi xotropy i s ti me-dependent thi nni ng and rheopexy i s ti me-
dependent thi ckeni ng. Both phenomena may be i rreversi bl e, reversi bl e
or parti al l y reversi bl e. There i s general agreement that the term
"thi xotropy" refers to the ti me-dependent decrease i n vi scosi ty, due to
sheari ng, and the subsequent recovery of vi scosi ty when sheari ng i s
removed (Mewi s, 1979). I rreversi bl e thi xotropy, cal l ed rheomal axi s or
rheodestructi on, i s common i n food products and may be a factor i n
eval uati ng yi el d stress as wel l as the general fl ow behavi or of a materi al .
Anti -thi xotropy and negati ve thi xotropy are synonyms for rheopexy.
Thi xotropy i n many fl ui d foods may be descri bed i n terms of the
sol -gel transi ti on phenomenon. Thi s termi nol ogy coul d appl y, for
exampl e, to starch-thi ckened baby food or yogurt. After bei ng man-
ufactured, and pl aced i n a contai ner, these foods sl owl y devel op a three
di mensi onal network and may be descri bed as gel s. When subjected to
shear (by standard rheol ogi cal testi ng or mi xi ng wi th a spoon), structure
i s broken down and the materi al s reach a mi ni mum thi ckness where
Thixotropic
Time-Independent
Rheopectic
S
h
e
a
r

S
t
r
e
s
s
,

P
a
Time at Constant Shear Rate, s
Time-Dependent Behavior
Figure 1.18. Thixotropic behavior observed in torque decay curves.
they exi st i n the sol state. I n foods that show reversi bi l i ty, the network
i s rebui l t and the gel state reobtai ned. I rreversi bl e materi al s remai n
i n the sol state.
The range of thi xotropi c behavi or i s i l l ustrated i n Fi g. 1.18. Sub-
jected to a constant shear rate, the shear stress wi l l decay over ti me.
Duri ng the rest peri od the materi al may compl etel y recover, parti al l y
recover or not recover any of i ts ori gi nal structure l eadi ng to a hi gh,
medi um, or l ow torque response i n the sampl e. Rotati onal vi scometers
have proven to be very useful i n eval uati ng ti me-dependent fl ui d
behavi or because (unl i ke tube vi scometers) they easi l y al l ow materi al s
to be subjected to al ternate peri ods of shear and rest.
Step (or l i near) changes i n shear rate may al so be carri ed out
sequenti al l y wi th the resul ti ng shear stress observed between steps.
Typi cal resul ts are depi cted i n Fi g. 1.19. Actual curve segments (such
as 1-2 and 3-4) depend on the rel ati ve contri buti on of structural
breakdown and bui l dup i n the substance. Pl otti ng shear stress versus
shear rate for the i ncreasi ng and decreasi ng shear rate val ues can be
used togenerate hysteresi s l oops (a di fference i n the up and down curves)
for the materi al . The area between the curves depends on the ti me-
dependent nature of the substance: i t i s zero for a ti me-i ndependent
0
S
t
r
e
s
s
0
time
Rest
Period
S
h
e
a
r

R
a
t
e
Complete Recovery
Partial Recovery
No Recovery
Evidence of Thixotropy in Torque Decay Curves
fl ui d. Thi s i nformati on may be val uabl e i n compari ng di fferent
materi al s, but i t i s somewhat subjecti ve because di fferent step change
peri ods may l ead to di fferent hysteresi s l oops. Si mi l ar i nformati on can
be generated by subjecti ng materi al s to step (or l i near)changes i n shear
stress and observi ng the resul ti ng changes i n shear rates.
Figure 1.19. Thixotropic behavior observed from step changes in shear rate.
Torque decay data (l i ke that gi ven for a probl em i n mi xer vi scometry
descri bed i n Exampl e Probl em 3.8.22)may be used to model i rreversi bl e
thi xotropy by addi ng a structural decay parameter to the Herschel -
Bul kl ey model to account for breakdown (Ti u and Boger, 1974):
[1.36]
where , the structural parameter, i s a functi on of ti me. before the
onset of sheari ng and equal s an equi l i bri um val ue ( ) obtai ned after
compl ete breakdown from sheari ng. The decay of the structural
parameter wi th ti me may be assumed to obey a second order equati on:
[1.37]
time
S
h
e
a
r

S
t
r
e
s
s
S
h
e
a
r

R
a
t
e
1
2
3
4
Step Changes in Shear Rate
f (,
) (
o
+ K(
)
n
)
1

e
d
dt
k
1
(
e
)
2
for >
e
where i s a rate constant that i s a functi on of shear rate. Then, the
enti re model i s compl etel y determi ned by fi ve parameters: ,
and . and are determi ned under i ni ti al sheari ng condi ti ons
when and . I n other words, they are determi ned from the i ni ti al
shear stress i n the materi al , observed at the begi nni ng of a test, for each
shear rate consi dered.
and are expressed i n terms of the apparent vi scosi ty ( ) to
fi nd . Equati ng the rheol ogi cal model (Eq. [1.36]) to the defi ni ti on of
apparent vi scosi ty (whi ch i n thi s case i s a functi on of both shear rate
and the ti me-dependent apparent vi scosi ty) yi el ds an expressi on for :
[1.38]
Eq. [1.38] i s val i d for al l val ues of i ncl udi ng at , the equi l i bri um
val ue of the apparent vi scosi ty. Di fferenti ati ng wi th respect to ti me,
at a constant shear rate, gi ves
[1.39]
Usi ng the defi ni ti on of , Eq. [1.37] and [1.39] may be combi ned
yi el di ng
[1.40]
Consi deri ng the defi ni ti on of gi ven by Eq. [1.38], thi s may be rewri tten
as
[1.41]
Si mpl i fi cati on yi el ds
[1.42]
or
k
1
o
, K, n, k
1
(
e
K, n
o
1 t 0

e
/
k
1
o
+ K(
)
n

e

e
d
dt

d
dt
o
+ K(
)
n
_
,
d/dt
k
1
(
e
)
2
d
dt
o
+ K(
)
n
_
k
1
o
+ K(
)
n
_
o
+ K(
)
n
_
,
1
1
]
2
d
dt
o
+ K(
)
n
_
,
d
dt
k
1
o
+ K(
)
n
_
,
(
e
)
2
[1.43]
where
[1.44]
I ntegrati ng Eq. [1.43] gi ves
[1.45]
so
[1.46]
where i s the i ni ti al val ue of the apparent vi scosi ty cal cul ated from
the i ni ti al ( and ) shear stress and shear rate.
Usi ng Eq. [1.46], a pl ot versus , at a parti cul ar shear rate,
i s made to obtai n . Thi s i s done at numerous shear rates and the
resul ti ng i nformati on i s used to determi ne the rel ati on between and
and, from that, the rel ati on between and . Thi s i s the fi nal i nfor-
mati on requi red to compl etel y speci fy the mathemati cal model gi ven by
Eq. [1.36] and [1.37].
The above approach has been used to descri be the behavi or of
mayonnai se (Ti u and Boger, 1974), baby food (Ford and Steffe, 1986),
and buttermi l k (Butl er and McNul ty, 1995). More compl ex model s
whi ch i ncl ude terms for the recovery of structure are al so avai l abl e
(Cheng, 1973; Ferguson and Kembl owski , 1991). Numerous rheol ogi cal
model s to descri be ti me-dependent behavi or have been summari zed by
Hol dsworth (1993).
1.5.3. Modeling Rheological Behavior of Fluids
Model i ng provi des a means of representi ng a l arge quanti ty of
rheol ogi cal data i n terms of a si mpl e mathemati cal expressi on. Rheo-
grams, summari zed i n terms of the Herschel -Bul kl ey equati on (Eq.
[1.26]), represent one exampl e of model i ng. I n thi s secti on we wi l l
expand the i dea to i ncl ude temperature and concentrati on (or moi sture
content) effects i nto si ngl e empi ri cal expressi ons. Many forms of the
equati ons are possi bl e and one master model , sui tabl e for al l si tuati ons,
d
dt
a
1
(
e
)
2
a
1

k
1
o
+ K(
)
n
(
e
)
2
d

0
t
a
1
dt
1

e
o

e
+ a
1
t
o
t 0 1
1/(
e
) t
a
1
a
1
k
1

1.5.3 Modeling Rheological Behavior of Fluids 33

does not exi st. The equati ons covered here are acceptabl e for a l arge
number of practi cal probl ems i nvol vi ng homogeneous materi al s whi ch
do not experi ence a phase change over the range of condi ti ons under
consi derati on.
The i nfl uence of temperature on the vi scosi ty for Newtoni an fl ui ds
can be expressed i n terms of an Arrheni us type equati on i nvol vi ng the
absol ute temperature ( ), the uni versal gas constant ( ), and the energy
of acti vati on for vi scosi ty ( ):
[1.47]
and are determi ned from experi mental data. Hi gher val ues
i ndi cate a more rapi d change i n vi scosi ty wi th temperature. The energy
of acti vati on for honey i s eval uated i n Exampl e Probl em 1.14.4.
Consi deri ng an unknown vi scosi ty ( ) at any temperature ( ) and a
reference vi scosi ty ( ) at a reference temperature ( ), the constant ( )
may be el i mi nated from Eq. [1.47] and the resul ti ng equati on wri tten
i n l ogari thmi c form:
[1.48]
I n addi ti on to model i ng the vi scosi ty of Newtoni an fl ui ds, an Arrheni us
rel ati onshi p can be used to model the i nfl uence of temperature on
apparent vi scosi ty i n power l aw fl ui ds. Consi deri ng a constant shear
rate, wi th the assumpti on that temperature has a negl i gi bl e i nfl uence
on the fl ow behavi or i ndex, yi el ds
[1.49]
or
[1.50]
Eq. [1.50] can be used to fi nd at any temperature ( ) from appropri ate
reference val ues ( ). Acti vati on energi es and reference vi scosi ti es
for a number of fl ui d foods are summari zed i n Appendi x [6.14].
T R
E
a
f (T) A exp
E
a
RT
_
,
E
a
A E
a
T
r
T
r
A
ln
r
_
E
a
R
_
1
T

1
T
r
_
,
ln
r
_
,

E
a
R
1
T

1
T
r
_
r
exp
E
a
R
1
T

1
T
r
_
,
_
,
T
r
, T
r
The effect of temperature on vi scosi ty can al so be model ed usi ng a
rel ati onshi p known as the Wi l l i ams-Landel -Ferry (WLF) equati on
proposed by Wi l l i ams et al . (1955). A reference vi scosi ty at a reference
temperature, and the numeri cal val ue of two constants are needed to
speci fy the rel ati onshi p. The WLF equati on i s very useful i n model i ng
the vi scosi ty of amorphous foods above the gl ass transi ti on temperature
(Roos, 1992; Roos, 1995).
The effect of shear rate and temperature can be combi ned i nto a
si ngl e expressi on (Harper and El Sahri gi , 1965):
[1.51]
where i s an average val ue of the fl ow behavi or i ndex based on al l
temperatures. Eq. [1.51] can al so be expressed i n terms of shear stress:
[1.52]
The practi cal val ue of Eq. [1.52] i s demonstrated for concentrated
orange jui ce i n Exampl e Probl em 1.14.5. Parameters i n the model
( ) were determi ned usi ng a l i mi ted number of data sets taken at
a few speci fi c temperatures. The fi nal model (Eq. [1.52]), however, can
generate a rheogram at any temperature i n the range. Thi s i s useful
i n sol vi ng many food engi neeri ng probl ems such as those requi ri ng a
predi cti on of the fl ui d vel oci ty profi l e or pressure drop duri ng tube fl ow.
Effects of temperature and concentrati on ( ) on apparent vi scosi ty,
at a constant shear rate, can al so be combi ned i nto a si ngl e rel ati onshi p
(Vi tal i and Rao, 1984; Castal do et al . 1990):
[1.53]
The three constants ( ) must be determi ned from experi mental
data. Shear rate, temperature, and concentrati on (or moi sture content)
can al so be combi ned i nto a si ngl e expressi on (Mackey et al ., 1989):
[1.54]
where the i nfl uence of shear rate i s gi ven i n terms of a power l aw
functi on. The parameters ( , , and ) cannot be gi ven an exact
physi cal i nterpretati on because the sequence of steps used i n deter-
mi ni ng them i nfl uences the magni tude of the constants. Equati on
f (T,
) K
T
exp
E
a
RT
_
,
(
)
n 1
n
f (T,
) K
T
exp
E
a
RT
_
,
(
)
n
K
T
, n, E
a
C
f (T, C) K
T, C
exp
E
a
RT
_
,
C
B
K
T, C
, E
a
, B
f (T,
, C) K
, T, C
(
)
n 1
exp
E
a
RT
_
,
+ B(C)
1
1
]
K
, T, C
n E
a
B
1.6 Yield Stress Phenomena 35
parameters, for exampl e, may be determi ned usi ng stepwi se regressi on
anal ysi s wi th the assumpti on that i nteracti on effects (such as temper-
ature dependence of and )can be negl ected. i s a constant whi ch
combi nes the effects of shear rate, temperature and concentrati on.
Rheol ogi cal behavi or of fl ui d foods i s compl ex and i nfl uenced by
numerous factors. Eq. [1.54] al l ows predi cti on of apparent vi scosi ty on
the basi s of shear rate, temperature, and moi sture content. Ti me-
temperature hi story and strai n hi story may be added to form a more
comprehensi ve equati on (Dol an et al ., 1989; Dol an and Steffe, 1990;
Mackey et al ., 1989; Morgan et al ., 1989)appl i cabl e toprotei n and starch
based dough or sl urry systems.
The i nfl uence of temperature on the behavi or of pol ymeri c materi al s
may be model ed by determi ni ng a shi ft factor usi ng the pri nci pl e of
ti me-temperature superposi ti on (Bi rd et al ., 1987). Thi s techni que i s
one exampl e of the method of reduced vari abl es (outl i ned i n detai l by
Ferry, 1980)whi ch can be expanded toi ncl ude the effect of concentrati on
and pressure on rheol ogi cal behavi or. Ti me-temperature superposi ti on
equates the effect of ti me and temperature on rheol ogi cal properti es.
The useful consequence of the method i s that materi al behavi or can be
i nvesti gated i n ti me domai ns (usual l y very l ong or very short) that are
otherwi se unavai l abl e due to the experi mental l i mi tati ons. Ti me-
temperature superposi ti on has proven to be a val uabl e method i n stu-
dyi ng creep and stress rel axati on of syntheti c pol ymers (Nei l sen and
Landel , 1994); but the techni que, whi ch i s someti mes appl i cabl e to
bi ol ogi cal materi al s, has not been wi del y appl i ed to foods. Da Si l va et
al . (1994) found the ti me-temperature superposi ti on pri nci pl e was
i nappropri ate for model i ng the temperature dependence of storage and
l oss modul i of pecti n di spersi ons. The i dea, however, can provi de a
useful empi ri cal method for devel opi ng master-curves of rheol ogi cal
data for many fl ui d foods. The techni que i s i l l ustrated i n Exampl e
Probl em 1.14.5 where a shi ft factor i s cal cul ated for concentrated orange
jui ce.
1.6. Yield Stress Phenomena
A yi el d stress ( ) may be defi ned as the mi ni mum shear stress
requi red to i ni ti ate fl ow. The exi stence of a yi el d stress has been chal -
l enged (Barnes and Wal ters, 1985) usi ng the argument that everythi ng
fl ows gi ven suffi ci ent ti me or very sensi ti ve measuri ng equi pment. Thi s
concept i s expl ored i n more detai l i n Sec. 5.6. From a practi cal stand-
n b K
, T, C
o
poi nt, there i s l i ttl e doubt that i s an engi neeri ng real i ty (Hartnett
and Hu, 1989) whi ch may strongl y i nfl uence process engi neeri ng
cal cul ati ons.
Table 1.4. Methods of Determining Yield Stress
Method Description or Parameter Reference
Measured
Extrapolation Shear stress versus shear rate Ofoli et al. (1987)
curve extrapolated to zero Keentok (1982)
shear rate. Yoshimura et al. (1987)
Extrapolation Apparent viscosity versus Van Wazer et al. (1963)
shear stress curve extrapolated Kaletunc-Gencer and Peleg
to infinite apparent viscosity. (1984)
Stress Decay Residual stress on a bob. Lang and Rha (1981)
Stress Decay Residual stress in a back Steffe and Osorio (1987)
extruder (annular pump).
Stress to Initiate Flow Controlled stress rheometers James et al. (1987)
measure the minimum stress
required for flow in traditional
geometries: cone and plate, etc.
Stress to Initiate Flow Stress on a smooth, rough or Lang and Rha (1981);
grooved bob. VocadloandCharles (1971)
Stress to Initiate Flow Stress to move an immersed DeKee et al. (1980)
vertical plate.
Stress to Initiate Flow Stress to causemotioninatube Cheng (1986)
viscometer.
Stress to Initiate Flow Stress to create vane motion. Nguyen and Boger (1985)
Qiu and Rao (1988)
Yoo et al. (1995)
Stress to Initiate Flow Size of the plug flow radius in Lang and Rha (1981)
an annulus.
Stress to Initiate Flow Force to move material through Goodrich et al. (1989)
a finned, cylindrical cell.
Dynamic Testing Flat response of an oscillatory Cheng (1986)
input.
Vertical Plate Coating Amount of fluid remaining on a Lang and Rha (1981)
plate after withdrawal from Charm (1962)
sample.
Squeezing Flow Deformation between parallel Campanella and Peleg
circular disks. (1987a)
Cone Penetrometer Depth of penetration. Tanaka et al. (1971)
o
1.6 Yield Stress Phenomena 37
There are many ways to eval uate the yi el d stress for fl ui d l i ke
substances (Tabl e 1.4) and no si ngl e, "best"techni que can be i denti fi ed.
Di fferent appl i cati ons requi re di fferent methods. One common method
of obtai ni ng a yi el d stress val ue i s to extrapol ate the shear stress versus
shear rate curve back to the shear stress i ntercept at zero shear rate.
Val ues obtai ned usi ng thi s method wi l l be strongl y i nfl uenced by the
rheol ogi cal model (Bi ngham, Herschel -Bul kl ey, etc.) and shear rate
range sel ected to represent the data (Ofol i et al ., 1987). Thi s di ffi cul ty
i s demonstrated for mi l k chocol ate i n Exampl e Probl em 1.14.3. An
al ternati ve numeri cal procedure i s to pl ot apparent vi scosi ty versus
shear stress and determi ne from the poi nt (rel ated to zero shear rate)
where becomes i nfi ni te.
Reported yi el d stress val ues are actual l y defi ned by the rheol ogi cal
techni ques and assumpti ons used i n the measurement. An absol ute
yi el d stress i s an el usi ve property: I t i s not unusual for a yi el d stress
obtai ned by one techni que to be very di fferent from one found usi ng a
di fferent method. Cheng (1986) has wri tten an excel l ent revi ew of the
yi el d stress probl em and shown that the magni tudes of measured val ues
are cl osel y associ ated wi th creep, stress growth, thi xotropy, and the
characteri sti cti mes of these transi ent responses. He descri bed a concept
of stati c and dynami c yi el d stresses that has great practi cal val ue i n
rheol ogi cal testi ng of fl ui d foods.
Many foods, such as starch-thi ckened baby food (Steffe and Ford,
1985), thi cken duri ng storage and exhi bi t i rreversi bl e thi xotropi c
behavi or when sti rred before consumpti on. Chemi cal changes (e.g.,
starch retrogradati on)cause a weak gel structure toform i n the materi al
duri ng storage. Thi s structure i s sensi ti ve and easi l y di srupted by fl ui d
movement. The yi el d stress, measured i n an undi sturbed sampl e, i s the
stati c yi el d stress. The yi el d stress of a compl etel y broken down sampl e,
often determi ned from extrapol ati on of the equi l i bri um fl ow curve, i s
the dynami c yi el d stress (Fi g. 1.20). A stati c yi el d stress may be si g-
ni fi cantl y hi gher than the dynami c yi el d stress. I f the materi al recovers
i ts structure duri ng a short peri od of ti me (uncommon i n food products),
then a rate parameter may be uti l i zed to ful l y descri be rheol ogi cal
behavi or.
The i dea of a stati c and a dynami c yi el d stress can be expl ai ned by
assumi ng there are two types of structure i n a thi xotropi c fl ui d (Cheng,
1986). One structure i s i nsensi ti ve to shear rate and serves to defi ne
the dynami c yi el d stress associ ated wi th the equi l i bri um fl ow curve. A

Figure 1.20. Static and dynamic yield stresses.
second structure, the weak structure, forms over a certai n peri od of ti me
when the sampl e i s at rest. Combi ned, the two structures cause a
resi stance to fl ow whi ch determi nes the stati c yi el d stress. Behavi or
di scussed above has been observed, though not menti oned i n terms of
the stati c and dynami c yi el d stress, for many food products: appl e sauce,
banana and peach baby food, mustard, tomato ketchup (Barbosa
Canovas and Pel eg, 1983); and meat and yeast extracts (Hal omos and
Ti u, 1981). Al so, the same termi nol ogy (and essenti al l y the same
meani ng)was used by Pokorny et al . (1985)to i nterpret rheol ogi cal data
for margari ne. Yoo et al . (1995) defi ned a new di mensi onl ess number,
the yi el d number defi ned as the stati cyi el d stress di vi ded by the dynami c
yi el d stress, to di fferenti ate yi el d stresses.
An i mportant i ssue i n yi el d stress measurement, parti cul arl y from
a qual i ty control standpoi nt, i s reproduci bi l i ty of the experi mental data.
Thi s i s cri ti cal when compari ng the overal l characteri sti cs of products
made on di fferent producti on l i nes or i n di fferent pl ants. I n thi s si tu-
ati on the measurement i s cl osel y ti ed tothe appl i cati on and the absol ute
val ue of the yi el d stress may be uni mportant. A "true val ue (most l i kel y
the dynami c val ue)," of the yi el d stress may be essenti al to properl y
desi gn food processi ng systems l i ke those requi red i n tubul ar thermal
processi ng equi pment where fl ui d vel oci ty profi l es are cri ti cal . Typi cal
Equilibrium
Flow Curve
Dynamic Yield Stress
Static Yield Stress
S
h
e
a
r

S
t
r
e
s
s
Shear Rate
1.7 Extensional Flow 39
yi el d stresses, i ncl udi ng data showi ng di fferences between the stati c
and dynami c yi el d val ues, are summari zed i n Appendi x [6.7]. The vane
method, a si mpl e and practi cal means of measuri ng the yi el d stress, i s
presented i n detai l i n Sec. 3.7.3 and Exampl e Probl ems 3.8.19, 3.8.20,
and 3.8.21. Eval uati ng yi el d stresses usi ng control l ed stress rheometers
i s di scussed i n Sec. 3.7.3 and the rol e of the yi el d stress i n determi ni ng
the thi ckness of a food coati ng i s exami ned i n Exampl e Probl em 1.14.6.
1.7. Extensional Flow
Vi scometri c fl ow may be defi ned as that whi ch i s i ndi sti ngui shabl e
from steady si mpl e shear fl ow. Addi ti onal i nformati on may be obtai ned
from a di fferent type of fl ow: extensi onal fl ow that yi el ds an extensi onal
vi scosi ty. Pure extensi onal fl ow does not i nvol ve sheari ng and i s
someti mes referred to as "shear free" fl ow. I n publ i shed l i terature,
el ongati onal vi scosi ty and Trouton vi scosi ty are frequentl y used syn-
onyms for extensi onal vi scosi ty. Si mi l arl y, el ongati onal fl ow i s a syn-
onym for extensi onal fl ow.
Many food processi ng operati ons i nvol ve extensi onal deformati on
and the mol ecul ar ori entati on caused by extensi on, versus shear, can
produce uni que food products and behavi or. The reason shear and
extensi onal fl ow have a di fferent i nfl uence on materi al behavi or may
be expl ai ned by the way i n whi ch fl ow fi el ds ori ent l ong mol ecul es of
hi gh mol ecul ar wei ght. I n shear fl ow, the preferred ori entati on corre-
sponds to the di recti on of fl ow; however, the presence of a di fferenti al
vel oci ty across the fl ow fi el d encourages mol ecul es to rotate thereby
reduci ng the degree of stretchi ng i nduced i n mol ecul ar chai ns. The
tendency of mol ecul es to rotate, versus el ongate, depends on the mag-
ni tude of the shear fi el d: There i s rel ati vel y more el ongati on, l ess
rotati on, at hi gh shear rates. I n extensi onal fl ow, the si tuati on i s very
di fferent. The preferred mol ecul ar ori entati on i s i n the di recti on of the
fl ow fi el d because there are nocompeti ng forces tocause rotati on. Hence,
extensi onal fl ow wi l l i nduce the maxi mum stretchi ng of the mol ecul es
produci ng a chai n tensi on that may resul t i n a l arge (compared to shear
fl ow) resi stance to deformati on.
The nature of the mol ecul e, branched versus l i near, may si gni fi -
cantl y i nfl uence fl ow behavi or i n extensi on. I n comparabl e fl ui d systems
(i .e., hi gh-densi ty pol yethyl ene, a l i near mol ecul e, versus l ow-densi ty
pol yethyl ene, a branched mol ecul e) branched mol ecul es wi l l cause a
fl ui d to be l ess tensi on-thi nni ng then l i near mol ecul es. A si mi l ar
argument can be made i n compari ng the rel ati ve sti ffness of bi opol ymer
mol ecul es: Sti ffer mol ecul es are more qui ckl y ori ented i n an extensi onal
fl ow fi el d. Thi s phenomenon may be a factor i n the choi ce of a thi ckeni ng
agent for pancake syrup: Stri ngi ness can be reduced, whi l e mai ntai ni ng
thi ckness, when sti ffer mol ecul es are sel ected as addi ti ves. Reduced
stri ngi ness l eads to what can be cal l ed a cl ean "cut-off" after pouri ng
syrup from a bottl e. An exampl e of a sti ff mol ecul e woul d be the rod-l i ke
bi opol ymer xanthan compared to sodi um al gi nate or carboxymethyl -
cel l ul ose whi ch exhi bi t a randon-coi l -type conformati on i n sol uti on
(Padmanabhan, 1995).
Extensi onal fl ow i s an i mportant aspect of food process engi neeri ng
and preval ent i n many operati ons such as dough processi ng. Sheet
stretchi ng, as wel l as extrudate drawi ng, provi des a good exampl e of
extensi onal fl ow (Fi g. 1.21). Convergi ng fl ow i nto di es, such as those
found i n si ngl e and twi n screw extruders, i nvol ves a combi nati on of
shear and extensi onal fl ow; the extensi onal component of deformati on
i s i l l ustrated i n Fi g. 1.21. The anal ysi s of fl ow i n a convergi ng di e (see
Sec. 4.4) al l ows one to separate the pressure drop over the di e i nto the
shear and extensi onal components. Convergi ng fl ow may al so be
observed when fl ui d i s sucked i nto a pi pe or a straw, or when appl yi ng
a food spread wi th a kni fe.
One of the most common exampl es of extensi onal fl ow i s seen when
stretchi ng warm mozzarel l a cheese whi l e pul l i ng a sl i ce of pi zza away
from the servi ng pan. Someti mes thi s behavi or i s subjecti vel y referred
toas stri ngi ness. A si mi l ar observati on can be made when pul l i ng apart
a caramel fi l l ed candy bar or a pastry wi th frui t fi l l i ng. Extensi onal
deformati on i s al so present i n cal enderi ng (Fi g. 1.22), a standard
operati on found i n dough sheeti ng. Gravi ty i nduced saggi ng (Fi g. 1.22)
al so embodi es extensi onal deformati on. Thi s may be observed i n a
cut-off apparatus associ ated wi th frui t fi l l i ng systems for pastry prod-
ucts. Extensi onal fl ow i n thi s si tuati on i s undesi rabl e because i t may
contri bute to i nconsi stent l evel s of fi l l or an unsi ghtl y product
appearance due to smeared fi l l i ng. Bubbl e growth from the producti on
of carbon di oxi de gas occurri ng duri ng dough fermentati on, extrudate
expansi on from the vapori zati on of water, and squeezi ng to achi eve
product spreadi ng i nvol ve extensi onal deformati on (Fi g. 1.23). Exten-
si onal fl ow i s al so a factor i n di e swel l and mi xi ng, parti cul arl y dough
mi xi ng wi th ri bbon bl enders.
Figure 1.21. Extensional flow found in sheet stretching (or extrudate drawing)
and convergence into an extruder die.
Figure 1.22. Extensional flow in calendering and gravity induced sagging.
Sheet Stretching Extruder Die
Calendering Sagging
Figure 1.23. Extensional flow found in bubble growth and squeezing flow
between lubricated plates.
Al though extensi onal vi scosi ty i s cl earl y a factor i n food processi ng,
our use of thi s rheol ogi cal property i n engi neeri ng desi gn of processes
and equi pment i s sti l l at an earl y stage of devel opment. Extensi onal
fl ow i s al so an i mportant factor i n the human percepti on of texture wi th
regard to the mouthfeel and swal l owi ng of fl ui d foods and fl ui d drugs.
I n addi ti on, many pl asti cmanufacturi ng operati ons i nvol ve extensi onal
fl ow: compressi on moul di ng, thermoformi ng, bl ow moul di ng, fi ber
spi nni ng, fi l m bl owi ng, i njecti on moul di ng, and extrusi on.
Extensi onal vi scosi ty has been measured for vari ous food products.
Lei ghton et al . (1934) used the saggi ng beam method devel oped by
Trouton (1906) to measure the extensi onal vi scosi ty of i ce cream.
Resul ts were presented i n terms of apparent vi scosi ty by usi ng the wel l
known Trouton rati oshowi ng that extensi onal vi scosi ty i s equal tothree
ti mes the shear vi scosi ty (see Eq. [1.78]). Thi s appears to be the fi rst
reported measurement of the extensi onal fl ow of a food product. Moz-
zarel l a cheese has been tested i n uni axi al tensi on by Ak and Gunase-
karan (1995). Bi axi al extensi onal fl ow, produced by squeezi ng materi al
between paral l el pl ates, has been used i n eval uati ng cheese
(Campanel l a et al ., 1987; Casi raghi et al ., 1985), wheat fl our doughs
(Huang and Koki ni , 1993; Wi kstrm et al ., 1994), gel s (Bagl ey et al .
1985; Chri sti anson et al . 1985), and butter (Rohn, 1993; Shuka et al .,
1995). Data from the Chopi n Al veograph, a common dough testi ng
devi ce where a spheri cal bubbl e of materi al i s formed by i nfl ati ng a
sheet, can be i nterpreted i n terms of bi axi al extensi onal vi scosi ty (Fari di
Bubble Growth Squeezing
and Rasper, 1987; Launay and Bur, 1977). Thi s techni que requi res an
accurate determi nati on of the sampl e geometry before and duri ng
i nfl ati on. Doughs have al so been eval uated by subjecti ng them to
uni axi al extensi on (de Brui jne et al ., 1990).
The spi nni ng test (al so cal l ed extrudate drawi ng) was appl i ed to
measure the stretchabi l i ty of mel ted Mozzarel l a cheese (Cavel l a et al .,
1992). Entrance pressure drop from convergi ng fl ow i nto a di e has been
used to eval uate an extensi onal vi scosi ty for corn meal dough (Bhat-
tacharya et al ., 1994; Padmanabhan and Bhattacharya, 1993; See-
thamraju and Bhattacharya, 1994) and bread dough (Bhattacharya,
1993). Addi ti onal methods have been proposed for eval uati ng the
extensi onal behavi or of pol ymeri c materi al s (Ferguson and Kemb-
l owski , 1991; James and Wal ters, 1993; Jones et al ., 1987; Macosko,
1994; Petri e, 1979; Ti rtaamadja and Sri dhar, 1993; Wal ters, 1975):
bubbl e col l apse, stagnati on fl ow i n l ubri cated and unl ubri cated di es,
open si phon (Fanofl ow), fi l ament stretchi ng, spi nni ng drop tensi ometer,
and convergi ng jets. Extensi onal vi scosi ti es for some Newtoni an and
non-Newtoni an fl ui ds are presented i n Appendi ces [6.15] and [6.16],
respecti vel y. Measurement methods, and exampl e probl ems, are di s-
cussed i n Chapter 4.
TypesofExtensional Flow. There are three basi ctypes of extensi onal
fl ow (Fi g. 1.24): uni axi al , pl anar, and bi axi al . Duri ng uni axi al extensi on
materi al i s stretched i n one di recti on wi th a correspondi ng si ze reducti on
i n the other two di recti ons. I n pl anar extensi on, a fl at sheet of materi al
i s stretched i n the di recti on wi th a correspondi ng decrease i n thi ckness
( decreases)whi l e the wi dth ( di recti on) remai ns unchanged. Bi axi al
extensi on l ooks l i ke uni axi al compressi on, but i t i s usual l y thought of
as fl ow whi ch produces a radi al tensi l e stress.
Uniaxial Extension. Wi th a constant densi ty materi al i n uni axi al
extensi on (Fi g. 1.24), the vel oci ty di stri buti on i n Cartesi an coordi nates,
descri bed wi th the Hencky strai n rate, i s
x
1
x
2
x
3
Figure 1.24. Uniaxial, planar, and biaxial extension.
[1.55]
[1.56]
[1.57]
where . Si nce thi s fl ow i s axi symmetri c, i t may al so be descri bed
i n cyl i ndri cal coordi nates (i t may be hel pful to vi sual i ze thi s si tuati on
wi th the posi ti ve axi s al i gned wi th the axi s, Fi g. 1.24):
[1.58]
[1.59]
[1.60]
Pure extensi onal fl ow does not i nvol ve shear deformati on; therefore, al l
the shear stress terms are zero:
[1.61]
AFTER DEFORMATION
BEFORE DEFORMATION
uniaxial extension
planar extension
biaxial extension
z
r
1
2
3
x
x
x
u
1

h
x
1
u
2

h
x
2
2
u
3

h
x
3
2
h
> 0
z x
1
u
z

h
z
u
r

h
r
2
u
12

13

23

r

rz

z
0
Stress i s al so axi symmetri c:
[1.62]
resul ti ng i n one normal stress di fference that can be used to defi ne the
tensi l e extensi onal vi scosi ty:
[1.63]
Materi al s are consi dered tensi on-thi nni ng (or extensi onal -thi nni ng) i f
decreases wi th i ncreasi ng val ues of . They are tensi on-thi ckeni ng
(extensi onal -thi ckeni ng) i f i ncreases wi th i ncreasi ng val ues of .
These terms are anal ogous toshear-thi nni ng and shear-thi ckeni ng used
previ ousl y (Sec. 1.5.1) to descri be changes i n apparent vi scosi ty wi th
shear rate.
Biaxial Extension. The vel oci ty di stri buti on produced by uni axi al
compressi on causi ng a bi axi al extensi onal fl ow (Fi g. 1.24) can be
expressed i n Cartesi an coordi nates as
[1.64]
[1.65]
[1.66]
where . Si nce , bi axi al extensi on can actual l y be vi ewed as
a form of tensi l e deformati on. Uni axi al compressi on, however, shoul d
not be vi ewed as bei ng si mpl y the opposi te of uni axi al tensi on because
the tendency of mol ecul es to ori ent themsel ves i s stronger i n tensi on
than compressi on. Axi al symmetry al l ows the above equati ons to be
rewri tten i n cyl i ndri cal coordi nates (Fi g. 1.24) as
[1.67]
[1.68]
[1.69]
Bi axi al extensi onal vi scosi ty i s defi ned i n terms of the normal stress
di fference and the strai n rate:
[1.70]
Planar Extension. I n pl anar extensi on (Fi g. 1.24), the vel oci ty
di stri buti on i s
rr

22

33
E
f (
h
)

11

22

11

33

zz

rr
h
u
1

B
x
1
u
2
2

B
x
2
u
3

B
x
3
B
> 0

h
2
B
u
z
2
B
z
u
r

B
r
u
B
f (
B
)

11

22

33

22

rr

zz
2(
rr

zz
)
h
[1.71]
[1.72]
[1.73]
Thi s type of fl ow produces two di sti nct stress di fferences: and
. Pl anar extensi onal vi scosi ty i s defi ned i n terms of the most
easi l y measured stress di fference, :
[1.74]
I t i s di ffi cul t to generate pl anar extensi onal fl ow and experi mental tests
of thi s type are l ess common than those i nvol vi ng tensi l e or bi axi al fl ow.
Relation Between Extensional and Shear Viscosities. The fol -
l owi ng l i mi ti ng rel ati onshi ps between extensi onal and shear vi scosi ti es
can be expected for non-Newtoni an fl ui ds at smal l strai ns (Deal y, 1994;
Wal ters, 1975; Petri e, 1979):
[1.75]
[1.76]
[1.77]
Rel i abl e rel ati onshi ps for non-Newtoni an fl ui ds at l arge strai ns have
not been devel oped. The above equati ons may be preci sel y defi ned for
the speci al case of Newtoni an fl ui ds:
[1.78]
[1.79]
[1.80]
Eq. [1.78], [1.79], and [1.80] can be used to veri fy the operati on of
extensi onal vi scometers. Cl earl y, however, a Newtoni an fl ui d must be
extremel y vi scous to mai ntai n i ts shape and gi ve the sol i d-l i ke
appearance requi red i n many extensi onal fl ow tests. Extensi onal
behavi or of l ow vi scosi ty fl ui ds can be eval uated wi th the method of
opposi ng jets (Sec. 4.5), by spi nni ng (Sec. 4.6), or by i nvesti gati ng
tubel ess si phon behavi or (Sec. 4.7).
u
1

h
x
1
u
2

h
x
2
u
3
0
11

22
11

33
11

22
P
f (
h
)

11

22
h
lim
h
0
E
(
h
) 3 lim
0
(
)
lim
B
0
B
(
B
) 6 lim
0
(
)
lim
h
0
P
(
h
) 4 lim
0
(
E
3
B
6
P
4
1.8 Viscoelastic Material Functions 47
Trouton establ i shed a mathemati cal rel ati onshi p between tensi l e
extensi onal vi scosi ty (he cal l ed i t the coeffi ci ent of vi scous tracti on) and
shear vi scosi ty (Trouton, 1906). Presentl y, data for extensi onal and
shear vi scosi ti es are often compared usi ng a di mensi onl ess rati o known
as the Trouton number ( ):
[1.81]
Si nce extensi onal and shear vi scosi ti es are functi ons of di fferent strai n
rates, a conventi onal method of compari son i s needed to remove
ambi gui ty. Based on a consi derati on of vi scoel asti c and i nel asti c fl ui d
behavi or, Jones et al . (1987) advocated the fol l owi ng conventi ons i n
computi ng the Trouton numbers for uni axi al and pl anar extensi onal
fl ow:
[1.82]
[1.83]
meani ng that shear vi scosi ti es are cal cul ated at shear rates equal to
or for uni axi al or pl anar extensi on, respecti vel y. Usi ng the
si mi l ar consi derati ons, Huang and Koki ni (1993) showed that the
Trouton number for case of bi axi al extensi on shoul d be cal cul ated as
[1.84]
The Trouton rati o for a Newtoni an fl ui d may be determi ned from Eq.
[1.78], [1.79], and [1.80]: i n tensi l e extensi on i t i s equal to 3; i t i s 6 and
4, respecti vel y, i n bi axi al and pl anar fl ow. Departure from these
numbers are due to vi scoel asti c materi al behavi or. Experi mental
resul ts may produce consi derabl y hi gher val ues.
1.8. Viscoelastic Material Functions
Fl ui ds that have a si gni fi cant el asti ccomponent may exhi bi t unusual
behavi or (Fi g. 1.10; Fi g. 1.11): Wei ssenberg effect (rod cl i mbi ng),
tubel ess si phon, jet expansi on, and recoi l . El asti c behavi or may be
eval uated usi ng vi scometri c methods to determi ne the normal stress
di fferences found i n steady shear fl ow. Al ternati vel y, vi scoel asti c
N
Tr
N
Tr

extensional viscosity
shear viscosity
(N
Tr
)
uniaxial

E
(
h
)
(
h
)
(N
Tr
)
planar

P
(
h
)
(2
h
)
h
2
h
(N
Tr
)
biaxial

B
(
B
)
(
12

B
)
materi al functi ons may be determi ned from experi ments i nvol vi ng the
appl i cati on of unsteady state deformati ons. General l y, these dynami c
testi ng techni ques may be di vi ded i nto two major categori es: transi ent
and osci l l atory. Transi ent methods i ncl ude tests of start-up fl ow, ces-
sati on of steady shear fl ow, step strai n, creep, and recoi l . I n osci l l atory
testi ng, sampl es are deformed by the appl i cati on of harmoni cal l y
varyi ng strai n whi ch i s usual l y appl i ed over a si mpl e shear fi el d. A
comprehensi ve revi ew of l i near and nonl i near vi scoel asti c materi al
functi ons i s summari zed i n Chapter 5. Steady shear and l i near vi s-
coel asti c materi al functi ons can be rel ated (see Sec. 5.7) usi ng vari ous
empi ri cal rel ati onshi ps such as the Cox-Merz rul e, the Gl ei ssel es mi rror
rel ati on, and Launs rul e (Bi rd et al ., 1987).
Creep and Step-Strain (Stress Relaxation). I n a creep test,
materi al i s subjected to a constant stress and the correspondi ng strai n
i s measured as a functi on of ti me, . The data are often pl otted i n
terms of the shear creep compl i ance,
[1.85]
versus ti me. I n a step-strai n test, commonl y cal l ed a stress rel axati on
test, a constant strai n i s appl i ed to the test sampl e and the changi ng
stress over ti me i s measured, . The data are commonl y presented
i n terms of a shear stress rel axati on modul us,
[1.86]
versus ti me. Data from creep and stress rel axati on tests can al so be
descri bed i n terms of mechani cal (spri ng and dashpot) anal ogs (Moh-
seni n, 1986; Sherman, 1970; Barnes et al ., 1989; Pol akowski and Ri pl i ng
(1966)) whi ch wi l l be consi dered i n more detai l i n Chapter 5. Creep and
stress rel axati on experi ments can be conducted i n shear, compressi on,
or tensi on. Shear creep parameters for vari ous creamy styl e sal ad
dressi ngs are gi ven i n Appendi x [6.19].
Oscillatory Testing. The vi scoel asti c behavi or of fl ui ds can be
determi ned from dynami c testi ng where sampl es are subjected to
osci l l atory moti on when hel d i n vari ous contai nment systems, usual l y
a cone and pl ate or a paral l el pl ate apparatus. Typi cal l y, a si nusoi dal
strai n i s appl i ed to the sampl e causi ng some l evel of stress to be
transmi tted through the materi al . The magni tude and the ti me l ag of
the transmi ssi on depend on the vi scoel asti cnature of the test substance.
(t )
J f (t )

constant
(t )
G f (t )

constant
1.9 Attacking Problems in Rheological Testing 49
I n vi scous (more l i qui d l i ke) materi al s, much of the stress i s di ssi pated
i n fri cti onal l osses; i t i s mostl y transmi tted i n hi ghl y el asti c materi al s.
Li kewi se, the ti me l ag (al so cal l ed the phase l ag) i s l arge for hi ghl y
vi scous substances but smal l for materi al s exhi bi ti ng a hi gh degree of
el asti ci ty. I nvesti gati ng thi s type of phenomena l eads to the defi ni ti on
of vari ous materi al functi ons: compl ex vi scosi ty ( ), dynami c vi scosi ty
( ), compl ex modul us ( ), l oss modul us ( ), and storage modul us ( ).
These functi ons are di scussed i n detai l i n Chapter 5. Someti mes,
osci l l atory testi ng i s referred to as "smal l ampl i tude osci l l atory testi ng"
because smal l deformati ons must be empl oyed to mai ntai n l i near vi s-
coel asti c behavi or. Typi cal osci l l atory data for vari ous food products
are summari zed i n Appendi ces [6.20], [6.21], and [6.22].
1.9. Attacking Problems in Rheological Testing
Attacki ng rheol ogi cal probl ems i nvol ves a cri ti cal judgement
regardi ng the type of fl ow behavi or i nvol ved and a careful determi nati on
of the appropri ate i nstruments and techni ques to use i n fi ndi ng a
sol uti on. A si mpl e cl assi fi cati on of materi al behavi or (Fi g. 1.25)provi des
a useful framework toapproach rheol ogi cal testi ng of an unknown fl ui d.
Behavi oral extremes woul d be that of pure Hookean behavi or (i deal l y
el asti c materi al ) and pure Newtoni an behavi or (i deal l y vi scous mate-
ri al ); hence, these categori es have been pl aced on the upper ri ght and
l eft extremi ty of the fi gure. Thi s symbol i zes the fact that al l real
materi al s exhi bi t both vi scous and el asti c behavi or al though one type
of behavi or i s frequentl y domi nant. Water, for i nstance, i s consi dered
Newtoni an but wi l l show some degree of el asti ci ty under condi ti ons
i nvol vi ng a very short process ti me, e.g., when a hi gh vel oci ty object
i mpacts a body of water. I n eval uati ng sol i ds, one i s typi cal l y l ooki ng
at a stress-strai n rel ati onshi p as opposed to a fl ui d where a shear
stress-shear rate rel ati onshi p i s studi ed.
When i nvesti gati ng the behavi or of a new fl ui d, one must fi rst
determi ne i f the materi al can be consi dered i nel asti c (purel y vi scous)
meani ng that behavi or associ ated wi th el asti ci ty (di e swel l , rod cl i m-
bi ng, etc.)i s not i mportant i n the appl i cati on. I f purel y vi scous, the next
questi on deal s wi th ti me-dependency and i nvol ves i ssues of structural
stabi l i ty or breakdown when subjected to a shear force. Wi th materi al s
that are ti me-i ndependent, a rheogram may be devel oped and di fferent
mathemati cal equati ons (power l aw, Bi ngham, Herschel -Bul kl ey, or
any of those presented i n Tabl e 1.3) consi dered to fi nd a model that
accuratel y descri bes fl ow behavi or. Vi scoel asti c fl ui ds, those showi ng
G
*
G

Figure 1.25. Simple classification of rheological behavior.
si gni fi cant l evel s of both vi scous and el asti c behavi or, may be tested
usi ng dynami c methods or stati c techni ques (creep or stress rel axati on)
wi th vari ous mechani cal model s (di scussed i n Sec. 5.2) bei ng used to
i nterpret resul ts: Kel vi n, Maxwel l or Burgers model s. Normal stress
di fferences can be determi ned from vi scometri c fl ow. Extensi onal
deformati on may be used to determi ne addi ti onal materi al functi ons
associ ated wi th stretchi ng.
Cl assi fyi ng fl ui ds i s a val uabl e way to conceptual i ze fl ui d behavi or;
however, i t i s not meant to i mpl y that the types of behavi or noted i n
Fi g. 1.25 are mutual l y excl usi ve. A materi al showi ng el asti c behavi or
(such as dough) may si mul taneousl y be shear-thi nni ng and ti me-
dependent! Other factors, l i ke agi ng, may al so i nfl uence rheol ogi cal
behavi or. Tomato ketchup, for exampl e, may be properl y descri bed as
a ti me-i ndependent, shear-thi nni ng, fl ui d i mmedi atel y after manufac-
ture but agi ng often gi ves the materi al a weak gel l i ke structure causi ng
the product to exhi bi t thi xotropi c behavi or when used by the consumer.
Thi s expl ai ns why agi tati ng the ketchup, by sti rri ng or shaki ng i n the
bottl e, makes the condi ment more pourabl e. Cl earl y, the abi l i ty to
Fluid (Viscous Behavior) Solid (Elastic Behavior)
Newtonian
Non-Newtonian
Time-Dependent
Power Law Bingham Herschel-Bulkley
Rheopectic Thixotropic
Viscoelastic
Hookean Non-Hookean
Fluid-Solid
Other Models
Time-Independent
Non-Linear Elastic
Kelvin Maxwell Burgers
Structural Models
(time-independent)
1.9 Attacking Problems in Rheological Testing 51
conceptual i ze di fferent types of rheol ogi cal behavi or i s very i mportant
i n the devel opment, or i mprovement, of many new food products and
processes.
Estimating Shear Rates in Practical Applications. I t i s very
i mportant that steady shear data be col l ected over the appropri ate shear
rate range. The mi ni mum shear rate i s often zero due to the presence
of stati onary equi pment surfaces. Esti mates of the maxi mum shear
rate found i n many processi ng systems can be obtai ned from a cri ti cal
eval uati on of the equi pment.
I n systems where fl ui d i s cl osel y contai ned between movi ng
machi nery parts, the maxi mum shear rate can be esti mated from the
vel oci ty di fference di vi ded by the separati on di stance. An exampl e of
thi s can be found wi th an anchor i mpel l er turni ng i n a mi xi ng vessel
(Fi g. 1.26) where the maxi mum shear rate i s cal cul ated as the ti p speed
of the i mpel l er di vi ded by the gap between the i mpel l er and the mi xi ng
tank:
[1.87]
Spreadi ng (butter or margari ne) or brushi ng (frosti ng or pai nt) opera-
ti ons are frequentl y found i n the food i ndustry. I n thi s case (Fi g. 1.26),
the maxi mum shear rate can be esti mated from the vel oci ty of the brush
(or kni fe) di vi ded by the thi ckness of the coati ng:
[1.88]
Esti mati ng maxi mum shear rates i n systems wi th wi del y spaced
movi ng parts presents a di fferent probl em. Consi der, for exampl e, a
paddl e mi xer i n a vessel where so the i nfl uence of the wal l i s
negl i gi bl e (Fi g. 1.27). Experi mental data, taken a smal l di stance from
the i mpel l er, can be used to obtai n the vel oci ty profi l e perpendi cul ar to
the axi s of rotati on. I f these data are avai l abl e, the maxi mum shear rate
can be esti mated from the di fferenti al vel oci ty and the hei ght of the
bl ade (Fi g. 1.27):
max

d
D d
max
u/z
D d
Figure 1.26. Maximum shear rates found in mixing with an anchor impeller
and brushing.
[1.89]
I n the absence of experi mental data, the ti p speed of the i mpel l er ( )
can be used i n pl ace of but thi s procedure wi l l gi ve an over
esti mate of the maxi mum shear rate because due to the
fact that fl ui d momentum i s rapi dl y di ssi pated i n the mi xi ng vessel .
Data of fl ui d vel oci ty i n the vi ci ni ty of mi xi ng bl ades have been col l ected
for a radi al fl at-bl ade turbi ne type mi xer (Ol dshue, 1983; Koutsakos
and Ni enow, 1990).
The maxi mum shear rate for fl ui d fl ow i n a tubul ar geometry (Fi g.
1.27)can be determi ned from the vol umetri c fl ow rate ( )and the i nsi de
radi us of the tube:
[1.90]
Thi s cal cul ati on i s exact for Newtoni an fl ui ds but must be modi fi ed for
power l aw materi al s to i ncl ude the fl ow behavi or i ndex:
[1.91]
d
D
Anchor Impeller in Mixing Vessel
u
z
Brushing
max
= u/z
= d
max
/(D-d)
mixing
blade
max
(u
b/2
u
b
)/(b/2)
d/2
(u
b/2
u
b
)
(u
b/2
u
b
) < (d/2)
Q
max

4Q
R
3
max

3n + 1
4n
_
,
4Q
R
3
1.10 Interfacial Rheology 53
Figure 1.27. Maximum shear rate in a mixer when (where is the vessel
diameter), and flow of a Newtonian fluid in a tube.
Eq. [1.91] i s very hel pful i n esti mati ng the shear rate range needed to
determi ne rheol ogi cal data for pi pel i ne desi gn cal cul ati ons because
many fl ui d foods exhi bi t shear-thi nni ng behavi or, . These
rel ati onshi ps can al so be used i n consi deri ng fl ow through tube type
systems such as oi ntment and tooth paste contai ners, frosti ng tubes,
and spray nozzel s. The ori gi ns of Eq. [1.90] and [1.91] wi l l be cl ari fi ed
i n a l ater di scussi on of tube vi scometry gi ven i n see Sec. 2.2.
1.10. Interfacial Rheology
Both the shear and extensi onal vi scosi ti es di scussed i n precedi ng
secti ons of thi s chapter are bul k materi al properti es. Fl ow behavi or at
materi al i nterfaces, however, can be very di fferent than bul k fl ow
behavi or. I nterfaci al rheol ogy i s a fi el d of study that i nvesti gates
deformati ons occurri ng at fl ui d i nterfaces. Many practi cal probl ems
may i nvol ve i nterfaci al rheol ogi cal phenomena: formati on, stabi l i ty and
processi ng of foams and emul si ons; sprayi ng and atomi zati on; sel ecti on
of surfactants; and fi l m formati on. Bul k and i nterfaci al vi scosi ti es can
be rel ated usi ng the Boussi nesq number:
u=0
u
b/2
b
u
b
Velocity Profile: Mixer Impeller
(D>>d)
max
= ( - )/(b/2)
u
b/2 b
u
d
Pump
Flow in a Tube
R
= (4 Q)/( R )
max
3
D d D
0 < n < 1
[1.92]
I nterfaci al vi scosi ty (usual l y reported i n uni ts of surface poi se, sp,
equi val ent to1 g s
-1
)strongl y affects bul k fl ow behavi or at fl ui d i nterfaces
when (Edwards et al ., 1991).
Figure 1.28. Disk surface viscometer to measure interfacial shear viscosity.
Vari ous techni ques are avai l abl e to measure i nterfaci al rheol ogi cal
behavi or. These can be di vi ded i nto i ndi rect methods i nvol vi ng the
exami nati on of vel oci ty profi l es or di rect methods i nvol vi ng the mea-
surement of i nterfaci al torsi on. The di sk surface vi scometer i s a cl assi cal
(di rect) method of measuri ng i nterfaci al shear vi scosi ty (Fi g. 1.28). I n
thi s system the cup i s turned and the resul ti ng torque i s measured on
the fi xed pl ate. To determi ne the i nterfaci al vi scosi ty, the torque con-
tri buti on i s consi dered i n two parts: 1) one part, rel ated to the bul k
vi scosi ty, due to fl ui d contact under the pl ate; 2) one part, rel ated to the
i nterfaci al shear vi scosi ty, due to the fi l m i nduced tracti on al ong the
ri m of the pl ate. I nterfaci al shear vi scosi ty may exhi bi t Newtoni an or
non-Newtoni an behavi or. Addi ti onal i nstruments to study i nterfaci al
rheol ogy i ncl ude the deep-channel surface vi scometer, bi coni cal i nter-
faci al vi scometer, and vari ous types of kni fe edge vi scometers. Edwards
et al . (1991) gi ves a detai l ed summary of numerous measurement
N
Bo

interfacial viscosity
bulk viscosity (length scale)
N
Bo
>1
fluid
interface
1.11 Electrorheology 55
techni ques for the determi nati on of i nterfaci al shear vi scosi ty. Future
research i n i nterfaci al rheol ogy wi l l l ead to many i mproved products
and processes i n the food i ndustry.
1.11. Electrorheology
The phenomenon of el ectrorheol ogy refers to changes i n the rheo-
l ogi cal behavi or due to the i mposi ti on an el ectri c fi el d on a materi al . I t
i s someti mes cal l ed the Wi nsl ow effect after W.M. Wi nsl ow who
di scovered i t i n 1947 (Wi nsl ow, W.M. 1947. U.S. Patent Speci fi cati on
2417850). There has been a great deal of i nterest i n el ectrorheol ogi cal
(ER)fl ui ds for use i n vari ous mechani cal devi ces (Bl ock and Kel l y, 1988):
cl utches, brakes, hydraul i c val ves, acti ve or tunabl e damper systems,
wi de-band-hi gh-power vi brators, chucks, exerci se equi pment, and
roboti c control systems. I n the future, ER fl ui ds wi l l have a strong
i mpact on the automoti ve i ndustry and may eventual l y l ead toi mproved
desi gns for food manufacturi ng machi nery. I t may al so be possi bl e to
devel op uni que processi ng schemes and new products for foods that
exhi bi t an ER effect. The search for i ndustri al l y vi abl e el ectrorheo-
l ogi cal fl ui ds has been hi ndered by the abrasi veness and chemi cal
i nstabi l i ty of candi date materi al s. Si mi l ar possi bi l i ti es, and probl ems,
are found wi th magnetorheol ogi cal fl ui ds where fl ow behavi or may be
changed wi th the i mposi ti on of a magneti c fi el d.
ER fl ui ds are di spersi ons of sol i d parti cul ates, typi cal l y 0.1 to100 m
i n di ameter, i n an i nsul ati ng (non-conducti ng) oi l . At l ow shear rates,
i n the absence of an el ectri c fi el d, parti cl es are randoml y di stri buted
(Fi g. 1.29) and many ER fl ui ds wi l l show nearl y Newtoni an behavi or.
Wi th the appl i cati on of an el ectri c fi el d, parti cl es become pol ari zed,
causi ng parti cl e al i gnment across the el ectrode gap creati ng an
enhanced fi ber-l i ke structure. Thi s al i gnment, associ ated wi th i nherent
el ectri cal charges on the parti cl es, causes ER substances to thi cken
dramati cal l y. Appl i cati on of a vol tage causes some materi al s to devel op
hi gh yi el d stresses characteri sti c of Bi ngham pl asti c behavi or. I n fact,
the yi el d stresses can be so hi gh that fl ow ceases, effecti vel y trans-
formi ng the materi al from a l i qui d to a sol i d. Resul ts are i nfl uenced by
many factors: nature (al ternati ng or di rect current) and strength of the
el ectri cfi el d, temperature, composi ti on and vol ume fracti on of parti cl es,
shear rate, rheol ogi cal and di el ectri c properti es of the di spersi ng oi l .
Extensi ve experi mental work i s needed to eval uate the fl ow behavi or of
a parti cul ar ER fl ui d. To ful l y understand thi s probl em i nvol ves a

Figure 1.29. Particle alignment with and without an electric field (voltage per
unit length) applied across a narrow gap.
careful study of the i nterpl ay between vi scous, thermal , and pol ari zati on
forces (Zukoski , 1993). Comments si mi l ar to the above may be made
for magnetorheol ogi cal fl ui ds where the magneti c domai ns of the par-
ti cl es rotate unti l they l i ne up wi th the appl i ed magneti c fi el d.
There are a number of food rel ated ER fl ui ds such as cel l ul ose or
sodi um carboxymethyl cel l ul ose i n l i qui d paraffi n, and starch or gel ati ne
i n ol i ve oi l . Mi l k chocol ate i s al so known to be an ER fl ui d. Daubert
and Steffe (1996) observed an ER response i n thi s materi al : mi l k
chocol ate rheograms were shi fted upward wi th an i ncrease i n el ectri c
fi el d strength. A typi cal exampl e of thi s i s i l l ustrated i n Fi g. 1.30. The
yi el d stress, defi ned by the Casson equati on, al soi ncreased wi th vol tage.
Temperature changes had a very i nteresti ng effect on ER behavi or. I n
the presence of a vol tage, i ncreasi ng the temperature caused mi l k
chocol ate to thi cken produci ng an upward shi ft i n the rheogram. I t
appears that parti cl e pol ari ty was enhanced at hi gher temperatures.
I n the absence of a vol tage the usual trend, a decrease i n apparent
vi scosi ty wi th temperature i ncreases, was observed.
No Electric Field Applied Electric Field
+
+
+
+
-
-
-
-
1.12 Viscometers for Process Control and Monitoring 57
Figure 1.30 Typical influence of electric field strength (DC Volts/mm) on the
flow behavior of molten milk chocolate.
1.12. Viscometers for Process Control and Monitoring
The goal of thi s secti on i s to provi de an overvi ew of the pri mary
measurement concepts and i ssues i nvol ved i n the use of vi scometers for
conti nuous process control or moni tori ng. Al l process vi scometers used
for food products must conform to appropri ate sani tary standards and
accepted practi ces such as the 3-A and Egg 3-A standards publ i shed by
the I nternati onal Associ ati on of Mi l k, Food and Envi ronmental Sani -
tari ans (Des Moi nes, I A). Many process control vi scometers were not
desi gned for food appl i cati ons and cannot be modi fi ed for acceptabl e
sani tary operati on. Vi scometers di scussed here are typi cal of i ndus-
tri al l y avai l abl e uni ts whi ch are general l y acceptabl e for use i n the food
i ndustry. The current focus i s on uni ts that eval uate a steady shear
vi scosi ty. An i ndustri al system to determi ne extensi onal vi scosi ty,
based on fl ow through an ori fi ce (see Sec. 4.4.3), i s al so avai l abl e.
0 0.5 1 1.5 2
0
2
4
6
8
10
12
14
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
0 DC Volts 150 DC Volts 300 DC Volts
Milk Chocolate
In-line or On-line Installation. Vi scometers for process control or
moni tori ng may be i nstal l ed usi ng vari ous measurement schemes.
"I n-l i ne"systems are i nstal l ed di rectl y i n the process l i ne, usual l y i n a
pi pe. These systems are subject to process vari ati ons, such as changes
i n sampl e temperature, whi ch may si gni fi cantl y i nfl uence sensor out-
put. "On-l i ne"uni ts make measurements on a product si de stream, al so
cal l ed a by-pass l oop, taken from the mai n process fl ow l i ne. One
advantage of thi s type of system i s that sampl e vari abl es (i ncl udi ng fl ow
rate, temperature, and pressure) may be control l ed duri ng testi ng. A
thi rd type of process control vi scometer i s the i mmersi on system
desi gned for use i n process vessel s, parti cul arl y mi xi ngtanks. The above
uni ts provi de al ternati ves to "off-l i ne" measurements where a smal l
sampl e i s removed from the process l i ne and eval uated i n a standard
l aboratory i nstrument.
Figure 1.31. Capillary, on-line viscometer using side stream flow.
Measurement Concepts. A si de stream capi l l ary vi scometer i s
i l l ustrated i n Fi g. 1.31. Sampl e fl ow rate i s determi ned by the speed of
the gear pump. The system may be operated i n two modes: constant
fl ow rate or constant pressure. I n the constant pressure mode, the gear
pump i s dri ven at the speed requi red to mai ntai n a set pressure at the
entrance to the capi l l ary. I n the constant fl ow rate mode, the pump
main flow
side stream
gear pump
capillary
pressure
transducer
speed i s fi xed and the pressure drop requi red to force materi al through
the capi l l ary i s measured. Regardl ess of the mode of operati on, vi scosi ty
i s di rectl y proporti onal to the pressure drop across the capi l l ary di vi ded
by the fl ow rate through the system. A commerci al versi on of the si de
stream capi l l ary vi scometer, desi gned pri mari l y for pol ymer mel ts, i s
produced by Goettfert, I nc. (Rock Hi l l , SC).
Figure 1.32. Concentric cylinder type in-line viscometer.
An i n-l i ne concentri c cyl i nder vi scometer measures the torque, on
the i nner cyl i nder, generated by the outer cyl i nder movi ng at a fi xed
speed (Fi g. 1.32). Vi scosi ty i s proporti onal to the torque di vi ded by the
speed of the outer cyl i nder. The vi scometer rel i es on perforati ons i n the
bob and cup to achi eve conti nuous fl ow throughout the annul ar gap
where fl ui d properti es are measured. An al ternati ve rotati onal vi s-
cometer i s i l l ustrated i n Fi g. 1.33. I n thi s case, an off-set cyl i ndri cal
el ement i s attached to a rotati ng shaft causi ng a gyratory moti on of the
sensor. A conti nuous fl ow of test fl ui d moves through a perforated
sheath duri ng testi ng. Torque requi red to mai ntai n a fi xed speed of
rotati on i s measured. Vi scosi ty i s proporti onal to the torque requi red
to mai ntai n a constant speed of rotati on. Brookfi el d Engi neeri ng
Laboratori es (Stoughton, MA), and C.W. Brabender I nstruments (S.
Hackensack, NJ), respecti vel y, produce i nstruments l i ke those i l l us-
trated i n Fi g. 1.32 and 1.33.
outlet inlet
transducer
stator
rotor
drive motor
torque
perforation
Figure 1.33. I n-line viscometer using an off-set rotating element.
Figure 1.34. In-line viscometer using a vibrating rod.
torque
transducer
flow
rotating
sensor
perforated
sheat
flow
vibrating rod
transducer
Figure 1.35. In-line viscometer using a vibrating sphere.
Vi brati onal vi scometers are consi dered surface l oaded systems
because they respond to a thi n l ayer of fl ui d at the surface of the sensor.
I n-l i ne uni ts may i nvol ve rod (Fi g. 1.34) or spheri cal (Fi g. 1.35) sensors.
I n each case, the sensor i s dri ven at a fi xed frequency and the power
requi red to mai ntai n a preci se ampl i tude i s measured. Si nce the
vi brati ng probe accel erates the fl ui d, power i nput i s proporti onal to
product of vi scosi ty and densi ty. Densi ty compensati on can be i ncor-
porated al l owi ng a di rect computati on of Newtoni an vi scosi ty. Spheri cal
type vi brati onal vi scometers are manufactured by Nameter (Metuchen,
NJ).
An i mmersi on type, fal l i ng body vi scometer i s i l l ustrated i n Fi g. 1.36.
A pi ston i s peri odi cal l y rai sed al l owi ng sampl e to fi l l the cyl i ndri cal
contai ner. Duri ng testi ng, sampl e i s expel l ed from the cyl i nder by the
pi ston whi ch i s al l owed to fal l under the i nfl uence of gravi ty. Usi ng
reference data for standard Newtoni an fl ui ds, the fal l i ng ti me i s cor-
rel ated tosampl e vi scosi ty. Thi s concept can al sobe appl i ed toan on-l i ne
processi ngsystem. Vari ous fal l i ng pi ston vi scometers are manufactured
by Norcross Corporati on (Newton, MA).
Practical Considerations in Selecting a Process Control Vis-
cometer. I t i s i mportant to have a good understandi ng of the fl ui d
under consi derati on before sel ecti ng a process control vi scometer. The
i nfl uence of temperature, i ngredi ent formul ati on, and processi ng con-
di ti ons on fl ow behavi or must be ascertai ned before qual i ty control or
set-poi nts can be accuratel y establ i shed. Temperature has such a strong
flow
transducer
vibrating sphere
Figure 1.36. Falling piston, immersion type: (A) loading phase, (B) end of mea-
surement.
i nfl uence on rheol ogi cal behavi or that i t i s usual l y necessary to ei ther
careful l y control i t when conducti ng measurements or compensate for
i t when maki ng fi nal cal cul ati ons.
Process control vi scometers are general l y desi gned for Newtoni an
fl ui ds where the vi scosi ty i s not a functi on of shear rate. To eval uate
non-Newtoni an fl ui ds, mul ti pl e data poi nts, taken at di fferent shear
rates are requi red. Hence, operati onal shear rates must be establ i shed
and matched to the capabi l i ti es of the process control vi scometer. The
shear rates can be easi l y esti mated i n the capi l l ary vi scometers (Fi g.
1.31)or concentri ccyl i nder systems (Fi g. 1.32), but they are very di ffi cul t
to cal cul ate i n others such as the uni t wi th the offset rotati onal el ement
i l l ustrated i n Fi g. 1.33. Determi ni ng fundamental rheol ogi cal proper-
ti es when shear rates cannot be eval uated i s very compl ex; hence, some
measuring tube
piston
fluid level
tube opening
flow
A
B
1.13 Empirical Measurement Methods for Foods 63
process control vi scometers can onl y generate comparati ve fl ow
behavi or data (not absol ute rheol ogi cal properti es) for non-Newtoni an
fl ui ds.
Process control vi scometers may provi de a stri ctl y empi ri cal
parameter, an apparent vi scosi ty, or a fl ow curve i f sampl es are
deformed at mul ti pl e shear rates. Thi s i nformati on must be correl ated
to speci fi c processi ng factors, such as the amount of cocoa butter i n
chocol ate or the amount of water i n tomatopaste, before a control scheme
can be i ni ti ated. I f an on-l i ne i nstrument i s used for qual i ty control ,
then qual i ty must be careful l y defi ned and di rectl y rel ated to the rhe-
ol ogi cal property bei ng measured. Al so, the acceptabl e vari ati on i n
qual i ty must be known to establ i sh the proper l i mi ts (or set-poi nts)
requi red i n devel opi ng a control strategy. Al l process control vi scome-
ters must be cal i brated regul arl y and careful l y observed to ensure
sati sfactory l ong term performance.
1.13. Empirical Measurement Methods for Foods
The food i ndustry uses many empi ri cal i nstruments (Tabl e 1.5) to
measure the fl ow behavi or of food products. These devi ces are not used
to determi ne fundamental rheol ogi cal properti es, but resul ts may fi nd
di verse appl i cati ons: qual i ty control , correl ati on to sensory data, or even
serve as offi ci al standards of i denti ty. Food engi neers may fi nd i t
necessary to repl ace empi ri cal devi ces, l i ke the Bostwi ck Consi stometer
used for pureed foods, wi th more fundamental i nstruments to achi eve
engi neeri ng objecti ves rel ated to process control . Wi th the excepti on of
the mel t fl ow i ndexer for mol ten pol ymers, al l the i nstruments di scussed
i n thi s secti on are used for food products.
I t i s i mportant to recogni ze the fact that numerous foods are so
compl ex i t i s not practi cal , and i n many cases not possi bl e, to measure
thei r fundamental rheol ogi cal properti es. Carrots, peanuts, peas, or
beans (for exampl e) are non-homogeneous, noni sotropi c materi al s wi th
compl ex geometri es. Onl y empi ri cal testi ng devi ces, capabl e of mea-
suri ng composi te materi al behavi or, provi de a sui tabl e means of char-
acteri zi ng these foods. Empi ri cal i nstruments are a val uabl e and wel l
establ i shed part of the food i ndustry. Si nce they do not measure
fundamental properti es, they may appropri atel y be cal l ed i ndexers.
Some of the most common uni ts are descri bed i n thi s secti on. Consul t
Bourne (1982) and Brennan (1980) for addi ti onal i nformati on on eval -
uati ng food texture.
Table1.5. Typical Empirical TestingI nstruments and Applications for
Food Products (Summarized fromBourne, 1982)
Device Common Application
Adams Consistometer consistency of semifluid food purees
Armour Tenderometer beef tenderness
Baker Compressimeter staleness of bread
Ballauf Pressure Tester puncture testing of fruit and vegetables
BBIRA Biscuit Texture Meter hardness of cookies and crackers
Bloom Gelometer puncture test of gelatins and gelatin jellies
Bostwick Consistometer flow of baby foods and similar purees
Chatillon Pressure Tester puncture testing of fruit and vegetables
Effi-Gi Pressure Tester puncture testing of fruit and vegetables
Extensigraph behavior of wheat dough
Farinograph baking quality of wheat flour
FMC Pea Tenderometer quality and maturity of fresh green peas
FTC Texture Test System attachments for many foods
GF Texturometer attachments for many foods
Haugh Meter egg quality
Hilker-Guthrie Plummet firmness of cultured cream
Instron Universal Testing Machine attachments for many foods
Kramer Shear Press tenderness of peas and other particulate foods
Magness-Taylor Pressure Tester puncture testing of fruit and vegetables
Marine Colloids Gel Tester puncture test marine extract gels
Mixograph baking quality of wheat flour
Ottawa Pea Tenderometer quality and maturity of fresh green peas
Ottawa Texture Measuring System attachments for many foods
Pabst Texture Tester firmness of particulate foods
Penetrometer firmness of butter and margarine
Plint Cheese Curd Torsiometer setting of cheese curd
Resistograph baking quality of wheat flour
Ridgelimiter stiffness of pectin and fruit jellies
Stevens Compression Response attachments for many foods
Analyzer
Succulometer maturity and quality of fresh sweet corn
SURDD Hardness Tester hardness of fats and waxes
Torry Brown Homogenizer toughness of fish
USDA Consistometer consistency of semifluid food purees
Van Dorran Pressure Tester puncture testing of butter
Warner-Bratzler Shear toughness of meat
Dough Testing Equipment (Farinograph, Mixograph, Extensi-
graph, Alveograph). Dough - a combi nati on of cereal (usual l y wheat)
fl our, water, yeast, sal t, and other i ngredi ents - i s probabl y the most
compl ex materi al faci ng the food rheol ogi st; hence, i t i s not surpri si ng
that many empi ri cal i nstruments have been devel oped to eval uate the
fl ow behavi or of dough. The i nstruments may be di vi ded i nto two major
groups: those whi ch measure the power i nput duri ng dough devel op-
ment caused by a mi xi ng acti on, and those whi ch subject prepared
(devel oped) dough to an extensi onal deformati on. The fol l owi ng
di scussi on wi l l descri be the major i nstruments found i n each group.
Standard methods for operati ng these i nstruments and i nterpreti ng
data are publ i shed by the Ameri can Associ ati on of Cereal Chemi sts (St.
Paul , MN), and the I nternati onal Associ ati on for Cereal Chemi stry and
Technol ogy.
One of the most wi del y used dough mi xers i s the Farinograph
(DAppol oni a and Kunerth, 1984). Thi s i nstrument combi nes dough
i ngredi ents usi ng two Z-shaped mi xi ng bl ades that rotate, at di fferent
speeds, i n opposi te di recti ons. Mi xi ng i s i ni ti ated wi th dry fl our and
water i s added from a ti trati ng buret duri ng testi ng. A dynamometer
i s used to record torque on the dri ve shaft of the mi xi ng bl ades. Output
i s gi ven as a fari nogram: a pl ot of an i nstrument-dependent parameter
proporti onal to torque, expressed as a Brabender uni t (BU, al so cal l ed
consi stency), versus ti me. The shape of the fari nogram i s i nterpreted
i n terms of factors rel ated to fl our qual i ty and the behavi or of the dough
i n the bakery: dough devel opment ti me, stabi l i ty, mi xi ng tol erance, and
degree of softeni ng. The amount of water requi red to gi ve a consi stency
of 500 BU toa 14%moi sture content (wet basi s)fl our i s al soan i mportant
fl our parameter, known as the fari nograph water adsorpti on, deter-
mi ned usi ng the Fari nograph.
An al ternati ve to the Fari nograph i s the Mixographwhi ch i nvol ves
a pl anetary rotati on of verti cal pi ns (l owered i nto the dough) about
stati onary verti cal pi ns attached to the mi xi ng bowl . Torque i s recorded
whi l e mi xi ng a fi xed amount of fl our and water. Resul ts are gi ven i n
terms of a mi xogram whi ch i s i nterpreted i n a manner anal ogous to that
di scussed for the fari nogram.
The Extensigraph (Rasper and Preston, 1991) general l y conducts
tests on doughs prepared i n the Fari nograph. A speci al mol di ng devi ce
shapes the dough i nto a cyl i ndri cal speci men whi ch i s pl aced hori zon-
tal l y i ntoa support system. The ends are cl amped fi rml y i n pl ace l eavi ng
the mi ddl e secti on free for testi ng. A hook contacts the mi ddl e of the
sampl e and stretches i t whi l e movi ng downward at a constant speed.
The force on the sampl e, caused by the downward moti on of the hook,
i s recorded. Resul ts are gi ven as an extensogram, a pl ot of force (i n
Brabender uni ts) versus ti me and extensi on, whi ch provi des val uabl e
qual i ty control i nformati on for the dough. I t i s i mportant to note that
materi al s wi th si mi l ar fari nograms may have very di fferent extenso-
grams. The effects of oxi di zi ng agents and enzymes on dough behavi or,
for exampl e, can often be eval uated wi th extensograms.
The Alveograph(Shuey and Ti ppl es, 1980), al so cal l ed the Chopi n
Extensi graph, measures dough behavi or when subjecti ng i t to an
extensi onal deformati on. I n thi s i nstrument, a ci rcul ar di sk i s cut from
a sheet of dough and cl amped, around i ts ci rcumference, to the base
pl ate of the test apparatus. Ai r fl owi ng through the base pl ate causes
the dough to expand i nto a spheri cal l y shaped bubbl e whi ch eventual l y
ruptures compl eti ng the test. The ai r pressure i n the bubbl e over ti me
i s recorded and pl otted as an al veogram. I n routi ne testi ng, the maxi -
mum hei ght, overal l l ength, and the area under the curve are the pri -
mary parameters taken off the al veogram. Al veograph data have been
used to cal cul ate bi axi al extensi onal vi scosi ty (Fari di and Rasper, 1987;
Launay and Bur, 1977).
Cone Penetrometer. Sti ff materi al s --l i ke butter, peanut butter, or
margari ne-- are often assessed for "spreadabi l i ty" usi ng cone penetra-
ti on data. Thi s i nstrument consi sts of a wei ghted cone that i s posi ti oned
verti cal l y over the fl at surface of the test sampl e. Cone angl es of 20 or
45 degrees are typi cal . I n standard testi ng, the cone i s rel eased i nto the
sampl e and the depth of penetrati on, after a fi xed peri od of ti me, i s
measured. Si nce test materi al s have a hi gh yi el d stress, the cone comes
torest qui ckl y. Resul ts may be presented i n terms of a yi el d val ue whi ch
i s di rectl y proporti onal to the wei ght of the cone assembl y and i nversel y
proporti onal to the depth of penetrati on (Hai ghton, 1959; Sone, 1972).
Operati ng cone penetrometers wi th a constant downward speed, i nstead
of a constant wei ght, i s al so an effecti ve method of obtai ni ng exper-
i mental data (Tanaka et al ., 1971). Standard methods for testi ng
l ubri cati ng greases i nvol ve doubl e angl e cones: one cone, wi th a smal l
angl e, mounted on a second cone wi th a l arger angl e.
Warner-Bratzler Shear. The basi c part of thi s i nstrument i s made
from a 1 mm thi ck steel bl ade wi th a tri angul ar hol e cut from i t. Metal
bars, one l ocated on each si de of the bl ade, serve two functi ons: they
ori ent the bl ade and push the test sampl e i nto the V shaped notch of
the tri angul ar openi ng. Maxi mum force to cut through the sampl e i s
recorded. Hi gher l evel s of cutti ng force are associ ated wi th i ncreasi ng
sampl e toughness. The Warner-Bratzl er Shear i s extensi vel y used to
eval uate the texture of raw and cooked meats. I n these experi ments,
careful sampl e preparati on and ori entati on are essenti al for obtai ni ng
reproduci bl e resul ts. Other foods such as carrots, cel ery, rhubarb, and
asparagus have al so been tested i n thi s devi ce.
Bostwick Consistometer. The i nstrument i s a si mpl e devi ce used to
eval uate the fl ow characteri sti cs of pureed foods i ncl udi ng appl esauce,
catsup, and numerous baby-food products. These uni ts are usual l y made
of stai nl ess steel and consi st of two abutti ng compartments connected
wi th a common fl oor but separated by a spri ng-l oaded gate. The fi rst
compartment i s 5 X 5 X 3.8 cm when the gate i s l owered. Thi s secti on
i s l oaded wi th fl ui d at the begi nni ng of the test whi ch i s i ni ti ated by
pressi nga tri gger that rel eases the gate. Fl ui d fl ows, under the i nfl uence
of gravi ty, i nto the second compartment consi sti ng of an i ncl i ned trough
whi ch i s 5 cm wi de, 24 cm l ong and approxi matel y 2.5 cm hi gh. The
fl oor of the trough i s graduated i n 0.5 cm i ncrements and movement
down the trough refl ects fl ui d properti es. Measurements are taken after
a speci fi ed ti me (typi cal l y 5 to 30 s)and reported as centi meters of travel
from the starti ng gate. I f fl ui d moti on produces a curved surface, the
travel di stance of the l eadi ng edge i s reported. The Bostwi ck Consi s-
tometer i s sti l l wi del y used as a qual i ty control tool by the food i ndustry.
I t i s di ffi cul t to preci sel y rel ate Bostwi ck readi ngs to fundamental
rheol ogi cal behavi or (Vercrusse and Steffe, 1989); however, si gni fi cant
progress has been made for Newtoni an and power l aw fl ui ds usi ng a
gravi ty current anal ysi s (McCarthy and Seymour (1994) whi ch showed
Bostwi ck measurements to be l i nearl y rel ated to apparent vi scosi ty
di vi ded by densi ty rai sed to the -0.2 power: .
Adams Consistometer. Thi s devi ce measures fl ow, due to gravi ty,
over a hori zontal pl ate made of gl ass, metal , or steel . The pl ate has a
seri es of concentri c ci rcl es, l ocated one-quarter i nch apart, radi ati ng out
from the geometri c center of the pl ate. A truncated cone i s centered on
the pl ate and l oaded wi th sampl e, then verti cal l y rai sed by hand
al l owi ng materi al to fl ow radi al l y outward over the fl at surface. After
(/)
0.2
a fi xed peri od of ti me (10 to 30 s i s typi cal ), the di stance travel ed i n each
quadrant i s measured and the average val ue i s recorded (i n i nches) as
the "Adams Consi stency." The Adams Consi stometer i s commonl y used
to eval uate cream styl e corn and si mi l ar products.
Zhan Viscometer. Thi s devi ce i s a wel l known exampl e of an ori fi ce
vi scometer. I t consi sts of a smal l cup-shaped vessel wi th a hol e (ori fi ce)
i n the bottom. The cup i s fi l l ed wi th sampl e whi ch i s al l owed to di s-
charged through the ori fi ce. Ti me i s measured from the begi nni ng of
di scharge unti l the steady stream comi ng out of the cup begi ns to dri p.
Di scharge ti me i s correl ated to vi scosi ty. Zhan type vi scometers are
used i n many qual i ty control appl i cati ons.
Visco-Amylograph. The Vi sco-Amyl ograph (Shuey and Ti ppl es, 1980)
was desi gned to eval uate the behavi or of starch sol uti ons duri ng gel a-
ti ni zati on. I t consi sts of a rotati ng bowl wi th ei ght verti cal pi ns and a
matchi ng, suspended el ement, wi th seven verti cal pi ns. Torque i s
recorded on the upper el ement duri ng rotati on of the bowl . The system
i ncl udes a thermoregul ator whi ch al l ows the sampl e to be heated (the
standard rate i s 1.5 C per mi nute) duri ng testi ng. When an aqueous
suspensi on of starch i s heated above the gel ati ni zati on temperature,
the fl ui d thi ckens dramati cal l y. A compl ete amyl ograph test usual l y
i nvol ves four di sti nct thermal peri ods whi l e the bowl i s rotated at a
constant speed: heati ng, hol di ng, cool i ng, and hol di ng. Resul ts are
presented as an amyl ogram whi ch i s a pl ot of torque (gi ven as vi scosi ty
i n Brabender uni ts) versus ti me. Amyl ograms have proven useful i n
eval uati ng the qual i ty of starch and i ts behavi or as a thi ckeni ng agent
i n many food systems.
Rapid Visco Analyser. Thi s i nstrument generates data si mi l ar to
that provi ded by the Vi sco-Amyl ograph. Smal l sampl es, typi cal l y 3 to
4 grams of starch i n water, are heated i n a smal l mi xi ng vessel wi th a
pi tched paddl e i mpel l er. The al umi num mi xi ng vessel and pl asti c
i mpel l er are both di sposabl e. Sampl es are subjected to user program
changes i n temperature (heati ng, hol di ng, and cool i ng) i ntended to
match processi ng condi ti ons found i n a parti cul ar appl i cati on. Torque,
or i nstrument vi scosi ty, are measured over ti me whi l e the sampl e i s
agi tated and programmed temperature changes are executed. Thi s
i nstrument, ori gi nal l y i ntended to eval uate the qual i ty of Austral i an
wheat, may be used to exami ne the qual i ty of a wi de vari ety of food
starches.

Brookfield Rotating Disks and T-Bars. Brookfi el d Engi neeri ng
Laboratori es (Stoughton, MA) manufactures a rotati onal i nstrument
(cal l ed the Brookfi el d Vi scometer) that i s extensi vel y used i n the food
i ndustry. The most common sensors are fl at di sks (spi ndl es) attached
to the i nstrument wi th a verti cal shaft. Di sks are avai l abl e i n vari ous
si zes and may be rotated at di fferent speeds. Torque requi red to
mai ntai n constant rotati on i s measured. Thi s devi ce can read the vi s-
cosi ty of Newtoni an fl ui ds di rectl y because i t i s cal i brated wi th New-
toni an materi al s (si l i cone oi l s).
Si nce a thorough anal ysi s of shear rate on the di sk i s compl ex
(Wi l l i ams, 1979), i t i s di ffi cul t to use the Brookfi el d Vi scometer for the
determi nati on of non-Newtoni an fl ui d properti es. Si mpl i fi ed
approaches suggested by Mi tschka (1982) and Durguei l (1987) are
appl i cabl e to some foods. These techni ques use numerous constants to
convert torque and angul ar vel oci ty data i nto shear stress and shear
rate val ues, respecti vel y. Bri ggs (1995) successful l y determi ned the
shear-thi nni ng behavi or of banana puree, sal ad dressi ng, chocol ate
syrup, enchi l ada sauce, and pancake syrup usi ng Brookfi el d spi ndl es
and the Mi tschka (1982) method of anal ysi s.
Di sk sensors can be very useful i n obtai ni ng a rel ati ve i ndex of food
thi ckness for the purpose of compari ng products or maki ng qual i ty
control judgments. I n addi ti on to di sks, T-shaped bars are made for the
same purpose. I nstruments, equi pped wi th the T-bars, can al so be
attached to the Brookfi el d Hel i path Stand whi ch al l ows the enti re
i nstrument to be l owered duri ng testi ng. Thi s causes the sensor to take
a spi ral path through the sampl e whi l e torque data are obtai ned. The
Hel i path Stand i s typi cal l y used for thi ck pastes and gel s where a
rotati ng di sk woul d be di ffi cul t to i nsert or create a channel i ng effect
duri ng measurement.
Falling Ball Viscometer. Equati ons for the fal l i ng bal l vi scometer
are deri ved i n Exampl e Probl em 1.14.4. Thi s type of vi scometer i nvol ves
a verti cal tube where a bal l i s al l owed to fal l , under the i nfl uence of
gravi ty, through a Newtoni an fl ui d. Vi scosi ty i s cal cul ated on the basi s
of the ti me taken to fal l a fi xed di stance. I f the vessel di ameter i s 10
ti mes the bal l di ameter, wal l effects can be negl ected.
The ri si ng bubbl e vi scometer represents another appl i cati on of the
fal l i ng bal l concept. I n thi s case, a bubbl e of ai r i s al l owed to ascend
through a col umn of sampl e. Ri si ng ti me over a set di stance i s correl ated
to Newtoni an vi scosi ty.
Hoeppler Viscometer. The Hoeppl er Vi scometer i s a vari ati on of the
fal l i ng bal l concept. I n thi s case, a bal l havi ng a di ameter sl i ghtl y
smal l er than the cyl i ndri cal vessel contai ni ng the test fl ui d, fal l s
through a tube posi ti oned wi th a 10 degree i ncl i nati on from the verti cal
posi ti on. Ti me of fal l over a set di stance i s correl ated to the vi scosi ty of
Newtoni an fl ui ds. Uni ts of thi s type are someti mes cal l ed rol l i ng bal l
vi scometers because descendi ng spheres may rol l al ong the wal l of the
vi scometer.
Brabender-FMC Consistometer. Thi s uni t was ori gi nal l y desi gned
toeval uate cream styl e corn but i t has al sobeen used for ketchup, tomato
paste, baby food, and si mi l ar products. I t i s desi gned to l ower a thi n,
rectangul ar shaped, paddl e i nto a sampl e hel d i n a stai nl ess steel cup.
The cup i s rotated at a si ngl e speed of 78 rpm. Thi s moti on creates a
torque on the paddl e that i s read from a di al l ocated on the top of the
i nstrument. Paddl es are avai l abl e i n vari ous si zes: 5.08 cm (2 i nch) by
3.56 cm (1.4 i nch) i s typi cal .
Figure 1.37. Compression-extrusion and Kramer shear cells commonly used to
evaluate the behavior of particulate foods.
blade holder
shear blades
sample holder
Kramer Shear Cell
Compression-Extrusion Cell
annulus
Compression-Extrusion Cell. Thi s test apparatus (Fi g. 1.37) i s a
common devi ce used to measure the behavi or of parti cul ate foods
(Bourne, 1982). I n the compressi on-extrusi on cel l (al so cal l ed a back
extrusi on cel l ), sampl e compressi on causes materi al to fl ow through the
annul us formed between the pl unger and the cyl i ndri cal contai ner.
Sampl e data consi st of a curve rel ati ng the force on the pl unger versus
the di stance of pl unger travel (or travel ti me). Curves from di fferent
sampl es, i nvol vi ng di fferent treatments or vari eti es, are compared to
establ i sh di fferences i n product texture.
Kramer Shear Cell. The Kramer shear cel l (Fi g. 1.37) i s a wel l
establ i shed tool for eval uati ng the composi te fl ow behavi or of parti cul ate
foods. A typi cal system contai ns 10 shear bl ades whi ch are 3.2 mm thi ck
and separated by a di stance equal to the thi ckness (3.2 mm). Bars form
matchi ng sl i ts i n the top and bottom of the sampl e hol der. The sampl e
box i s approxi matel y 65 mm wi de, l ong, and deep. Duri ng testi ng, the
sampl e hol der i s fi l l ed wi th food and the shear bl ades (properl y al i gned
wi th the bars i n the top)are forced through the materi al unti l they pass
through the bars i n the bottom of the sampl e hol der. Force on the ram
hol di ng the bl ades i s measured over ti me and correl ated to product
fi rmness.
SimpleCompression. Bi ol ogi cal materi al s may be eval uated i n terms
of a bi oyi el d poi nt and a rupture poi nt (Mohseni n, 1984). A curve such
as the one i l l ustrated i n Fi g. 1.38 i s typi cal for sol i d foods, l i ke frui ts
and vegetabl es, when a cyl i ndri cal sampl e i s tested i n si mpl e com-
pressi on. The i ni ti al porti on of the curve (a-b) i s a strai ght l i ne up to
the l i near l i mi t (b). Youngs modul us may be cal cul ated from the stress
and strai n at that poi nt: . Asecant or a tangent modul us, defi ned
at a parti cul ar strai n, may be cal cul ated i f the l i ne i s curved (Mohseni n,
1986). The sl ope of the i ni ti al porti on of the curve i s often taken as an
i ndex of fi rmness. When stress and strai n cannot be cal cul ated, data
may be si mpl y pl otted i n terms of force and deformati on.
The bi oyi el d poi nt (c) i s rel ated to a fai l ure i n the mi crostructure of
the materi al associ ated wi th an i ni ti al di srupti on of cel l ul ar structure
(Fi g. 1.38). I t i s observed at a stress and strai n of and , respecti vel y.
The rupture poi nt (d) of the materi al , defi ned by and , correl ates to
the macroscopi c fai l ure i n the sampl e. Wi th more bri ttl e materi al s the
rupture poi nt may be very cl ose to the bi oyi el d poi nt: These poi nts may
be wi del y separated i n tough materi al s. The Ameri can Soci ety of
b
/
b
E
c

c
d

d
Figure 1.38. Generalized compression curve for a biological solid.
Agri cul tural Engi neers (ASAE, St. Joseph, MI ) has establ i shed a
standard method for the compressi on testi ng of food materi al s of convex
shape (ASAE Standard S368.2). I t i s si mi l ar to the si mpl e compressi on
test descri bed here.
TextureProfileAnalysis. Texture refers to the human sensati on of
food deri ved from i ts rheol ogi cal behavi or duri ng masti cati on and
swal l owi ng. Obtai ni ng a quanti tati ve descri pti on of texture usi ng
i nstrumental data i s very compl i cated because no i nstrument can
dupl i cate human capabi l i ti es. From an engi neeri ng perspecti ve, the
mouth can be consi dered an i ntri cate mechani cal system and chemi cal
reactor that can crush, wet, enzymati cal l y degrade, pressuri ze, heat or
cool , pump, chemi cal l y sampl e for taste, and sense force and tempera-
ture. I n addi ti on, thi s "eati ng machi ne" has a sophi sti cated feedback
control system. I ni ti al l y there i s open l oop, feed forward control to set
pri mary parameters: si ze of mouth openi ng, surface sel ecti on for fi rst
bi te (i nci sors or mol ars), etc. Once the food i s i n the mouth, there i s an
adapti ve feed back control system wi th a vari abl e gai n -hi gh wi th
unfami l i ar foods, l ow wi th everyday foods- that depends on bol us
a
b
c
d
d
c
b
b
d
c ,
Strain ( )
S
t
r
e
s
s

(

)
bioyield point
devel opment duri ng masti cati on. Thi s process i s i nfl uenced by many
factors: vol ume of the oral cavi ty, rate of addi ti on of sal i va, chemi cal
composi ti on of sal i va, chemi cal and physi cal i nteracti on of sal i va and
the food, rate of chewi ng, total number of chews, surface area i n contact
wi th the food, movement of the l i ps and cheeks, dynami c vol ume of the
oral cavi ty duri ng masti cati on, resi dence ti me of the bol us, i ni ti al vol -
ume of the bol us, and parti al fl ui d removal (by swal l owi ng) duri ng
chewi ng. Gi ven the above, i t i s not surpri si ng that l i ttl e progress has
been made i n correl ati ng fundamental rheol ogi cal properti es to the
human percepti on of texture. There has been l i mi ted success wi th some
fl ui d (Chri stensen, 1987; Koki ni and Cussl er, 1987; Sherman, 1988)and
sol i d foods (Hamann and Lani er, 1987; Montejano et al ., 1985).
Overal l , there are two methods to eval uate food texture: sensory and
i nstrumental . The sensory method of devel opi ng a texture profi l e
(Muoz et al . 1992) uti l i zes a human taste panel and provi des the
ul ti mate test whi ch, as di scussed above, cannot be compl etel y dupl i cated
by any i nstrumental procedure. I nstrumental methods, however, are
much l ess costl y and ti me consumi ng than sensory tests. Moreover,
they often correl ate to cri ti cal sensory attri butes whi ch al l ow some
measure of consumer acceptabi l i ty. I t i s, however, rare for them tostand
al one as a compl ete test. I n any event, they can certai nl y be very
val uabl e when used i n conjuncti on wi th sensory panel s. Generati ng
and i nterpreti ng texture profi l e i nformati on, wi th i nstrumental or
sensory means, i s cal l ed Texture Profi l e Anal ysi s. A doubl e compressi on
test, the most recogni zed i nstrumental means of characteri zi ng the
texture of sol i d and semi -sol i d foods, i s di scussed bel ow.
The i dea of texture profi l i ng food was proposed i n 1963 (Fri edman
et al ., 1963; Szczesnai k et al . 1963) and conducted usi ng an i nstrument
known as the General Foods Texturometer. Bourne (1968 and 1974)
adopted, and extended, the techni que to the I nstron Uni versal Testi ng
Machi ne where a food sampl e (bi te si ze pi eces of food, usual l y a 1 cm
cube) i s compressed, two ti mes, usual l y to 80 percent of i ts ori gi nal
hei ght. Compressi on i s achi eved usi ng paral l el pl ates where one pl ate
i s fi xed and the other pl ate moves wi th a reci procati ng l i near cycl i cal
moti on. Si nce thi s test i s i ntended to refl ect the human percepti on of
texture, the fi rst and second compressi on cycl es are referred to as the
fi rst bi te and second bi te, respecti vel y. A general i zed texture profi l e
curve i s i l l ustrated i n Fi g. 1.39. Vari ous textural parameters may be
determi ned from the curve (Bourne, 1978; Szczesnai k et al . 1963):
Figure 1.39. Generalized texture profile curve obtained from the Instron Univer-
sal Testing Machine (from Bourne et al., 1978).
Figure 1.40. Texture profile curves (1 = first bite; 2 = second bite) for four food
products (from Bourne, 1978).
0
F
o
r
c
e
Time
Area 1
Area 3
Fracturability
Hardness 1
Stringiness
First Bite Second Bite
downstroke upstroke
Area 2
Hardness 2
Springiness
downstroke upstroke
0
F
o
r
c
e
Time
APPLE
0
F
o
r
c
e
Time
FRANKFURTER
0
F
o
r
c
e
Time
PRETZEL STICK
0
F
o
r
c
e
Time
CREAM CHEESE
1
1
1
1
2
2
2
2
Fracturabi l i ty: force at the fi rst major drop i n force curve. Popul ar
terms descri bi ng fracturabi l i ty are crumbl y, crunchy, and bri ttl e.
Hardness 1: force at maxi mum compressi on duri ng fi rst bi te.
Popul ar terms descri bi ng hardness are soft, fi rm, and hard.
Area 1 (area under the sol i d l i ne up to the dashed l i ne i n the fi rst
compressi on cycl e): work done on the sampl e duri ng the fi rst bi te.
Adhesi veness: Area 3 (area under the zero force l i ne) representi ng
the work, caused from a tensi l e force, needed to pul l food apart and
separate i t from the compressi on pl ates. A si mi l ar adhesi veness
characteri sti c i s present i n the second compressi on cycl e. Popul ar
terms descri bi ng adhesi veness are sti cky, tacky, and gooey.
Adhesi ve Force: maxi mum negati ve force.
Stri ngi ness: di stance food extends before i t breaks away from the
compressi on pl ates.
Hardness 2: force at maxi mum compressi on duri ng second bi te.
Area 2 (area under the sol i d l i ne up to the dashed l i ne i n the second
compressi on cycl e): work done on the sampl e duri ng the second bi te.
Spri ngi ness: di stance or l ength of compressi on cycl e duri ng the
second bi te. Popul ar terms descri bi ng spri ngi ness are pl asti c and
el asti c.
Cohesi veness: the rati o of Area 2 di vi ded by Area 1.
Gummi ness: the product of Hardness (fi rst peak cal l ed Hardness
1 i n Fi g. 1.39) ti mes Cohesi veness. Popul ar terms descri bi ng
gummi ness are short, meal y, pasty, and gummy.
Chewi ness: the product of Gummi ness ti mes Cohesi veness ti mes
Spri ngi ness whi ch i s equi val ent to Gummi ness ti mes Spri ngi ness.
Popul ar terms descri bi ng chewi ness are tender, chewey, and tough.
Though Chewi ness and Gummi ness are si mi l ar, they are mutual l y
excl usi ve. The same product cannot exhi bi t both Chewi ness and
Gummi ness: Chewi ness refers to sol i d foods and Gummi ness refers
to semi -sol i d foods (Szczesnai k, 1995).
Typi cal texture profi l e curves for appl e ti ssue, frankfurter, cream
cheese, a pretzel sti ck are shown i n Fi g. 1.40. Observi ng these curves
makes i t cl ear that many foods donot exhi bi t al l the textural parameters
defi ned above. Texture Profi l e Anal ysi s has proved to be a very useful
techni que for exami ni ng food products whi ch fracture; however, texture
profi l e curves are subject to many vari ati ons maki ng the proper i nter-
pretati on of experi mental data chal l engi ng. More fundamental anal yses
can be attempted usi ng the sci ence of fracture mechani cs.
Figure 1.41. Simple melt flow indexer for molten polymers.
Melt FlowIndexer for Polymers. The mel t fl ow i ndexer i s a wei ght
dri ven capi l l ary i nstrument used as a qual i ty control devi ce to charac-
teri ze the behavi or of mol ten pol ymers. Al though the devi ce i s not
empl oyed for food, use of the i nstrument i n the thermopl asti cs i ndustry
i s so wi despread that anyone i nterested i n fl ui d rheol ogy shoul d have
a general awareness of how i t operates.
A si mpl e mel t fl ow i ndexer (Fi g. 1.41) i s contai ned wi thi n a thermal
jacket whi ch control s sampl e temperature. A known mass i s pl aced on
the pl unger whi ch dri ves a pi ston downward causi ng the sampl e to be
extruded through the di e. Exact di mensi ons and operati ng procedures
mass
piston
sample
die
jacket
extrudate
1.14.1 Carrageenan Gum Solution 77
are descri bed by Ameri can Soci ety for Testi ng and Materi al s (ASTM)
standards. Typi cal di mensi ons i nvol ve a di e di ameter of 2.1 mm, and
a rati o of di e l ength to di e di ameter equal to 4. I n a common testi ng
procedure known as "condi ti on E," a sampl e (such as pol yethyl ene) i s
al l owed to come to an equi l i bri um temperature of 190 C. Then, a 2.16
kg mass i s pl aced on the pi ston and the mol ten pol ymer i s extruded from
the di e. The wei ght, i n grams, of the extrudate produced i n 10 mi nutes
i s the mel t fl ow i ndex, or si mpl y the mel t i ndex, of the pol ymer. Val ues
of the mel t fl ow i ndex refl ect the vi scosi ty of the materi al : Low vi scosi ty
materi al s, correspondi ng to a l ow mol ecul ar wei ght, have a hi gh mel t
fl ow i ndex. Thi s i nformati on i s very val uabl e i n the thermopl asti cs
i ndustry because i t correl ates wel l to the qual i ty and processi ng char-
acteri sti cs of numerous pol ymers.
1.14. Example Problems
1.14.1. Carrageenan Gum Solution
Rheol ogi cal data for a 1%aqueous sol uti on of carrageenan gum at 25 C
are avai l abl e (Tabl e 1.6). Determi ne the power l aw parameters and pl ot
the apparent vi scosi ty curve.
Table 1.6. Steady Shear, Rheological Data for a 1% Aqueous Solution of Carra-
geenan Gum at 25 C (Data from Prentice and Huber, 1983)
(s
-1
) (Pa) (s
-1
) (Pa)
9.88 2.61 58.8 8.20
11.4 2.97 75.4 9.08
12.9 2.81 104.1 11.63
14.1 3.44 110.4 10.65
17.6 3.80 120.5 12.75
26.3 4.85 136.5 13.10
42.0 6.61 145.8 14.90
48.6 6.19 187.1 15.85
49.3 5.89 210.2 12.70
55.5 7.22 270.0 20.50
The data were pl otted (Fi g. 1.42) and an excel l ent fi t was obtai ned
wi th l i near regressi on anal ysi s usi ng the power l aw model : = 0.66 Pa
s
n
and = 0.60. Other rheol ogi cal model s coul d be used to fi t the
experi mental data. Some (l i ke the Newtoni an model ) woul d resul t i n

K
n
greater stati sti cal vari ati on whi l e others (l i ke the Herschel -Bul kl ey
model ) woul d i mprove the stati sti cal accuracy wi th whi ch the equati on
represents the data.
Figure 1.42. Rheogram for a carrageenan gum solution at 25 C.
Figure 1.43. Apparent viscosity of a carrageenan gum solution at 25 C.
Carrageenan Gum
1% Aqueous Solution
S
h
e
a
r

S
t
r
e
s
s
,

P
a
Shear Rate, 1/s
K = 0.66 Pa s
n
n = 0.60
20
15
10
5
0
0 50 100 150 200
0 50 100 150 200

0
0.05
0.1
0.15
0.2
0.25
0.3
Shear Rate, 1/s
A
p
p
a
r
e
n
t

V
i
s
c
o
s
i
t
y
,

P
a

s
Carrageenan Gum
1% Aqueous Solution
K = 0.66 Pa s
n
n = 0.60
1.14.2 Concentrated Corn Starch Solution 79

Apparent vi scosi ty i s cal cul ated as and pl otted as a functi on of
shear rate (Fi g. 1.43). Mathemati cal l y, the curve can be descri bed i n
terms of the power l aw fl ui d model usi ng and val ues determi ned
from the shear stress versus shear rate data:
The data presented i n Tabl e 1.6 have al so been eval uated i n terms of
the Casson, Bi ngham, and Herschel -Bul kl ey model s i n Appendi x 6.4
where the method of l i near regressi on anal ysi s i s expl ai ned.
1.14.2. Concentrated Corn Starch Solution
Exami ne the rheol ogi cal i nformati on presented (Fi g. 1.44) for a con-
centrated (53% wt/wt) sol uti on of raw corn starch and water. Steady
shear data were col l ected usi ng a conventi onal cone (4 degree, 60 mm
di ameter) and pl ate apparatus.
Figure 1.44. Rheogram of a 53% (wt/wt) solution of raw corn starch and water
at 25 C.
/
K n
K(
)
n 1
.66(
)
.4
0 2 4 6 8 10
0
2
4
6
8
10
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s

(
P
a
)

o
r

A
p
p
a
r
e
n
t

V
i
s
c
o
s
i
t
y

(
P
)
shear stress apparent viscosity
53% Corn Starch and Water

The corn starch sol uti on i s a concentrated suspensi on of starch
parti cl es i n water. Exami nati on of the apparent vi scosi ty curve (Fi g.
1.44) shows i ni ti al shear-thi nni ng fol l owed by strong shear-thi ckeni ng
behavi or. At l ow shear rates ( s
-1
), the water has a l ubri cati ng
effect between the parti cl es and fl ow i s rel ati vel y unhi ndered. Wi th
hi gher shear rates ( s
-1
), i ncreased resi stance from parti cl e to
parti cl e i nteracti on causes a si gni fi cant i ncrease i n apparent vi scosi ty.
Curve fi tti ng the data, above 4.5 s
-1
, to the power l aw model yi el ds:
where = 0.131 Pa s and = 1.72. A fl ow behavi or i ndex equal to 1.72,
a val ue si gni fi cantl y greater than 1.0, i s a numeri cal i ndi cati on of a l arge
shear-thi ckeni ng effect. The above equati on i s i l l ustrated as the l i ne
pl otted i n Fi g. 1.44.
Table 1.7. Rheological Data for SwedishCommercial Milk Chocolate at 40 C(Data
from Prentice and Huber, 1983)
(s
-1
) (Pa) (s
-1
) (Pa)
0.099 28.6 6.4 123.8
0.140 35.7 7.9 133.3
0.199 42.8 11.5 164.2
0.390 52.4 13.1 178.5
0.790 61.9 15.9 201.1
1.60 71.4 17.9 221.3
2.40 80.9 19.9 235.6
3.90 100.0
A concentrated corn starch sol uti on can provi de an excel l ent vi sual
and tacti l e exampl e of shear-thi ckeni ng behavi or. Fi rst, a 50 to 55%
(wt/wt) sol uti on consi sti ng of raw corn starch (avai l abl e i n most grocery
stores) and water i s requi red. Next, one must be wi l l i ng to exami ne the
materi al wi th bare hands. I f you sl owl y move a fi nger through the
materi al , the sol uti on feel s and appears l i ke a l i qui d. I f the fi nger i s
moved qui ckl y, however, thi s substance provi des much greater
resi stance and shows sol i d-l i ke behavi or by fracturi ng and separati ng
at the hi gher shear rate! The materi al qui ckl y reverts to a l i qui d-l i ke
appearance at the cessati on of movement.
0 <

< 4.5
4.5
K(
)
n
.131(
)
1.72
K n

1.14.3 Milk Chocolate 81
1.14.3. Milk Chocolate
Rheol ogi cal data for mi l k chocol ate at 40 C are avai l abl e (Tabl e 1.7).
Determi ne the Casson and Bi ngham pl asti c model parameters for thi s
materi al .
Figure 1.45. Regression analyses of 40 C milk chocolate applying the Bingham
plastic model over different shear rate ranges.
Data are present for l ow shear stresses and a pl ot (Fi g. 1.45) of thi s
i nformati on suggests the presence of a yi el d stress. Casson and Bi ng-
ham pl asti cmodel parameters were cal cul ated over three di fferent shear
rate ranges (Tabl e 1.8). The resul ts cl earl y i ndi cate that the model and
shear rate range covered i n the anal ysi s have a strong i nfl uence on the
yi el d stress (the dynami c yi el d stress) cal cul ated from the i ntercept of
the regressi on curve. When data at the l ower shear rates are empha-
si zed, the cal cul ated yi el d stress decreases. Thi s resul ts i n Bi ngham
yi el d val ues rangi ng from 35.1 to 52.3 Pa.
Shear Rate, 1/s

S
h
e
a
r

S
t
e
s
s
,

P
a
0 - 20 1/s

0 - 1.6 1/s
1.6 - 20 1/s
Data Range
Milk Chocolate
250
200
150
100
50
0
0 5 10 15 20 25

Table 1.8. Constants for Bingham Plastic ( ) and Casson
( ) Models Used to Describe the Behavior of 40 C Milk
Chocolate Over Three Shear Rate Ranges
Casson Model Bingham Plastic
r
2
r
2
(s
-1
) (Pa
.5
s
.5
) (Pa) (Pa s) (Pa)
0 - 20 2.21 29.7 .99 8.82 62.3 .99
1.6 - 20 2.14 32.3 .99 9.80 48.7 .98
0 - 1.6 3.04 17.8 .89 25.5 35.1 .84
1.14.4. Falling Ball Viscometer for Honey
The effect of temperature on the vi scosi ty of a Newtoni an fl ui d can be
i l l ustrated by consi deri ng data from a fal l i ng bal l vi scometer. Deri ve
(Part a) the fal l i ng bal l vi scometer equati ons and use them to eval uate
(Part b) the vi scosi ty of honey from the data gi ven i n Tabl e 1.9. Al so,
eval uate the i nfl uence of temperature on honey vi scosi ty usi ng the
Arrheni us equati on.
Part a. Consi der a sphere (radi us= ; densi ty= ) droppi ng through a
Newtoni an fl ui d (densi ty= ) otherwi se at rest. The sphere, travel i ng
downward at termi nal vel oci ty ( ), i s subject to three forces: buoyancy
( ), drag ( ) and gravi ty ( ). A force bal ance
yi el ds
[1.93]
where the drag term comes from Stokes l aw (Trans. Cambri dge Phi l .
Soc. Vol . 8, 1845 and Vol . 9, 1851). Si mpl i fi cati on of Eq. [1.93] gi ves
[1.94]
The ti me ( ) requi red for the bal l to fal l a fi xed di stance ( ) i s
[1.95]
so the termi nal vel oci ty i s

o
+
pl
(
)
n
0.5
(
o
)
0.5
+ K
1
(
)
0.5
K
1

o

pl

o
R
s
l
u
t
(4R
3
l
g)/3 (4R
3
s
g)/3 6Ru
t
4
3
R
3
_
,
g
l
+ 6Ru
t

4
3
R
3
_
,
g
s
s

l
2R
2
g
9u
t
t L
t
L
u
t
1.14.4 Falling Ball Viscometer for Honey 83
Table 1.9. Falling Ball Viscometer Data for Honey at Six Temperatures
Ball #1 Ball #2 Ball #3
ball diameter ( ), m .00475 .00678 .00792
ball density, kg/m
3
7900 7900 7900
length of fall, m 0.20 0.20 0.20
honey density, kg/m
3
1400 1400 1400
container diameter ( ), m 0.036 0.036 0.036
= d/D 0.132 0.188 0.220
Faxen correction factor 0.727 0.617 0.559
at 6.5 C, s 270.0 152.5 122.0
at 12.5 C, s 108.3 58.5 46.8
at 20.0 C, s 31.7 17.8 14.4
at 21.5 C, s 25.7 14.4 11.4
at 38.0 C, s 3.0 1.8 1.4
at 48.0 C, s 1.5 1.0 0.9
at 6.5 C, Pa s 107.79 124.04 135.41
at 12.5 C, Pa s 43.22 47.58 51.94
at 20.0 C, Pa s 12.67 14.51 15.98
at 21.5 C, Pa s 10.26 11.67 12.65
at 38.0 C, Pa s 1.20 1.46 1.55
at 48.0 C, Pa s 0.60 0.81 0.99
at 6.5 C, Pa s 78.38 76.59 75.68
at 12.5 C, Pa s 31.43 29.38 29.03
at 20.0 C, Pa s 9.21 8.96 8.93
at 21.5 C, Pa s 7.46 7.20 7.07
at 38.0 C, Pa s 0.87 0.90 0.87
at 48.0 C, Pa s 0.44 0.50 0.56
at 6.5 C, s
-1
0.94 1.16 1.24
at 12.5 C, s
-1
2.3 3.0 3.2
at 20.0 C, s
-1
7.9 9.9 10.5
at 21.5 C, s
-1
9.8 12.3 13.3
at 38.0 C, s
-1
84.2 98.3 108.2
at 48.0 C, s
-1
168.9 176.9 168.4
[1.96]
Substi tuti ng Eq. [1.96] i nto Eq. [1.94] yi el ds
d
D
max

u
t

L
t
[1.97]
so
[1.98]
where i s the fal l i ng bal l vi scometer constant defi ned as
[1.99]
Our focus i n thi s secti on i s on the fal l i ng bal l vi scometer. Eq. [1.94],
however, can al so be used as a basi s for cal cul ati ng settl i ng vel oci ti es
of parti cl es i ncl udi ng pi gments i n pai nt, and chocol ate i n mi l k.
Usi ng Eq. [1.98] and [1.99] the vi scosi ty of a Newtoni an fl ui d can be
cal cul ated di rectl y. I t i s al so possi bl e to cal cul ate the vi scosi ty usi ng a
reference l i qui d, a Newtoni an l i qui d of known vi scosi ty, to el i mi nate
errors caused from i naccurate determi nati on of the bal l radi us or drop
l ength. i s equal for the unknown and reference l i qui ds, therefore,
[1.100]
where the subscri pts 1 and 2 refer to the test and reference l i qui d,
respecti vel y. The maxi mum shear rate, l ocated at the equator of the
sphere, i s (Sherman, 1970):
[1.101]
Eq. [1.100] i s acceptabl e for and cases where the
sphere di ameter i s l ess than 1/10 the vessel di ameter. I f necessary, one
may mathemati cal l y account for the presence of the wal l when .
Faxen (Arki v. Mat. Astron. Fyzi k. 17(27)1: 1922-1923) showed that the
vi scosi ty cal cul ated usi ng Stokes l aw coul d be corrected for wal l effects
usi ng the fol l owi ng equati on:
[1.102]
s

l
2R
2
gt
9L
s

l
kt
k
k
2R
2
g
9L
k
1
(
s

1
)t
1

2
(
s

2
)t
2
max

3u
t
R
N
Re
u
t
d/ < 1.0
d/D >0.1
c
[1 2.104() + 2.09()
3
.95()
5
]
1.14.4 Falling Ball Viscometer for Honey 85
where i s equal to the sphere di ameter ( ) di vi ded by the contai ner
di ameter ( ) and i s the corrected vi scosi ty. The term i n brackets, the
Faxen correcti on factor, i s appl i cabl e up to = 0.32. An al ternati ve
expressi on (Chhabra, 1992) can be used when :
[1.103]
Part b. Raw data for a test i nvol vi ng three spheres, one cyl i ndri cal
contai ner and honey at si x di fferent temperatures are presented i n Tabl e
1.9. The Faxen correcti on factor, as wel l as uncorrected (from Eq. [1.97])
and corrected (from Eq. [1.102]) vi scosi ti es were cal cul ated from the
experi mental data (Tabl e 1.9).
Corrected honey vi scosi ti es, taken from Tabl e 1.9, were averaged at
each temperature (Tabl e 1.10)and the data fi t (r
2
=0.99)tothe Arrheni us
equati on (Eq. [1.47]):
yi el di ng
where: i s the temperature i n degrees Kel vi n; = 1.987 cal /(g-mol e K);
= 21,801 cal /(g-mol e); = 5.58 (10
-16
) Pa s. Resul ts may al so be
expressed i n terms of a reference vi scosi ty. Usi ng a val ue of 3.77 Pa s,
cal cul ated at 300 K wi th the previ ous rel ati onshi p, yi el ds
Table 1.10. Average of Corrected Viscosities for Honey
T T
( C) (K) (Pa s)
6.5 278.7 76.88
12.5 285.7 29.95
20.0 293.2 9.03
21.5 294.7 7.25
38.0 311.2 0.88
48.0 321.2 0.50
d
D
c
<0.97
1
1 .475
_
,
4
f (T) A exp
E
a
RT
_
,
5.58(10
16
) exp
21801
RT
_
,
T R
E
a
A
3.77 exp
21801
1.987
1
T

1
300
_
,
_
,
3.77 exp
11, 126
1
T

1
300
_
,
_

The above equati ons provi de conveni ent expressi ons for cal cul ati ng the
vi scosi ty of honey at any temperature from 6.5 C to 48 C.
1.14.5. Orange J uice Concentrate
Rheol ogi cal data for concentrated orange jui ce, at four di fferent tem-
peratures, are gi ven i n Tabl e 1.11. Descri be thi s data usi ng the rheo-
l ogi cal model expressed by Eq. [1.52]:
Al so, make a master-curve of the data usi ng 9.5 C as the reference
temperature.
A rheogram (Fi g. 1.46), usi ng the power l aw model , i s determi ned
at each temperature generati ng a fl ow behavi or i ndex and a consi stency
coeffi ci ent for each curve (Tabl e 1.12). Usi ng thi s data, the average fl ow
behavi or i ndex may be cal cul ated:
Table 1.11. Rheological Data for Concentrated Orange Juice (65 Brix, 5.7%Pulp)
Made from Perna Oranges (Data from Vitali and Rao, 1984)
T = -18.8 C T = -5.4 C T = 9.5 C T = 29.2 C
(Pa) (s
-1
) (Pa) (s
-1
) (Pa) (s
-1
) (Pa) (s
-1
)
14.4 0.5 4.3 0.6 2.6 1.1 3.6 8
24.3 1 6.5 1 10.3 8 7.6 20
141.9 10 38.4 10 17.1 15 13.1 40
240.4 20 65.4 20 29.5 30 17.5 60
327.2 30 88.7 30 50.3 60 31.2 120
408.0 40 111.1 40 69.4 90 54.5 240
483.9 50 131.9 50 103.3 150 94.4 480
555.9 60 151.7 60 153.8 250 141.7 800
635.2 70 171.3 70 199.8 350 170.0 1000
692.5 80 189.4 80 242.8 450 183.2 1100

f (T,
) K
T
exp
E
a
RT
_
,
(
)
n
n
.764 + .772 + .762 + .797
4
.774
1.14.5 Orange Juice Concentrate 87

Figure 1.46. Rheograms of orange juice concentrate (Perna oranges: 65 Brix,
5.7% pulp) at four different temperatures.
Temperature dependency of the consi stency coeffi ci ent (data i n
Tabl e 1.12, pl otted i n Fi g. 1.47) may be determi ned from regressi on of
Table 1.12. Power Law Fluid Properties for Concentrated Orange Juice at Four
Temperatures
T T
( C) (K) (1/K) (Pa s
n
) (-)
-18.8 254.4 0.003931 24.37 0.764
-5.4 267.8 0.003734 6.45 0.772
9.5 282.7 0.003538 2.25 0.762
29.2 302.4 0.003307 0.69 0.797
0.3 1 3 10 30 100 300 1,000
1
2
5
10
20
50
100
200
500
1,000
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
Orange Juice Concentrate
-18.8 C -5.4 C 9.5 C 29.3 C
1/T K n

Figure 1.47. Variation of the consistency coefficient of concentrated orange juice
with temperature.
[1.104]
showi ng that Pa and K. Hence, the
compl ete model , gi vi ng shear stress as a functi on of temperature and
shear rate, may be expressed as
or, i n terms of apparent vi scosi ty, as
Si nce the shear stress and apparent vi scosi ty equati ons may be used
over the enti re temperature and shear rate range, they coul d be very
useful i n sol vi ng many food process engi neeri ng desi gn probl ems rel ated
to the producti on of orange jui ce concentrate. A compari son (Fi g. 1.48)
of the ful l predi cti on equati on wi th the ori gi nal data i ndi cate reasonabl y
0.0032 0.0033 0.0034 0.0035 0.0036 0.0037 0.0038 0.0039 0.004
0.5
1
2
5
10
20
50
K
,

P
a

s
n
1/T, K
-1
Effect of Temperature on K
K K
T
exp
E
a
RT
_
,
K
T
4.65(10
9
) s
n
E
a
/R 5668.25
4.646(10
9
) exp
5668.3
T
_
,
(
)
.774
4.646(10
9
) exp
5668.3
T
_
,
(
)
.774 1.0
1.14.5 Orange Juice Concentrate 89
good resul ts when compared to the case where separate rheograms are
generated for each data set (Fi g. 1.46). Thi s l evel of accuracy i s
acceptabl e i n sol vi ng most food engi neeri ng probl ems.
Figure 1.48. Comparison of the raw data and prediction equation for the full
shear rate-temperature model of concentrated orange juice.
A reference temperature of 9.5 C wi l l be used i n devel opi ng a
master-curve of the experi mental data. Note that any val ue over the
range of temperatures consi dered coul d be sel ected as the reference
temperature. Devel opi ng a master curve requi res a hori zontal shi fti ng
of the data at -18.8 C, -5.4 C, and 29.3 C to the 9.5 C curve (Fi g. 1.46).
A di mensi onl ess shi ft factor ( ) i s numeri cal l y found to account for the
movement of each curve. Shear stress versus shear rate di vi ded by the
shi ft factor ( ) are pl otted to produce the master-curve.
I n thi s exampl e, a shear stress of 100 Pa wi l l be used as the basi s
for determi ni ng . Shear rates at Pa are cal cul ated usi ng the
constants provi ded i n Tabl e 1.12. At -18.8 C, for i nstance,
0.1 0.3 1 3 10 30 100 300 1,000
1
2
5
10
20
50
100
200
500
1,000
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
-18.8 C -5.4 C 9.5 C 29.3 C

a
T
/a
T
a
T
100

Table 1.13. Shear Rates (at = 100 Pa) and Shift Factors
T at 100 Pa
( C) (1/s) ( - )
-18.8 6.35 6.35 / 145.4 = 0.0437
-5.4 34.83 34.83 / 145.4 = 0.2395
9.5 145.4 145.4 / 145.4 = 1.000
29.3 514.0 514 / 145.4 = 3.536
and at 9.5 C, the shear rate i s cal cul ated as
Wi th thi s i nformati on, the shi ft factor at -18.8 C i s found usi ng the shear
rate at the reference temperature of 9.5 C:
Cal cul ated shear rates and shi ft factors for each curve are summari zed
i n Tabl e 1.13.
Table 1.14. Shifted Rheological Data for Concentrated Orange Juice (65 Brix,
5.7%Pulp) Used to Produce the Master-Curve
T=-18.8 C T=-5.4 C T=9.5 C T=29.2 C
(Pa) (s
-1
) (Pa) (s
-1
) (Pa) (s
-1
) (Pa) (s
-1
)
14.4 11.4 4.3 2.5 2.6 1.1 3.6 2.3
24.3 22.8 6.5 4.2 10.3 8.0 7.6 5.7
141.9 228.8 38.4 41.8 17.1 15.0 13.1 11.3
240.4 457.7 65.4 83.5 29.5 30.0 17.5 17.0
327.2 686.5 88.7 125.3 50.3 60.0 31.2 33.9
408.0 915.3 111.1 167.0 69.4 90.0 54.5 67.9
483.9 1144.2 131.9 208.8 103.3 150.0 94.4 135.7
555.9 1373.0 151.7 250.5 153.8 250.0 141.7 226.2
635.2 1601.8 171.3 292.3 199.8 350.0 170.0 282.8
692.5 1830.7 189.4 334.0 242.8 450.0 183.2 311.1
a
T
K
_
,
1/.764
100
24.37
_
,
1/.764
6.35 s
1
K
_
,
1/.762
100
2.25
_
,
1/.762
145.4 s
1
a
T
6.35/145.4 0.0437

/a
T

/a
T

/a
T

/a
T
1.14.6 Influence of the Yield Stress in Coating Food 91
Figure 1.49. Plot of shear stress versus providing a master-curve of con-
centrated orange juice having a reference temperature of 9.5 C.
Raw data, shi fted usi ng appropri ate val ues, are gi ven i n Tabl e
1.14 and pl otted to produce a master-curve (Fi g. 1.49). Hori zontal
shi fti ng causes the data to overl ap on a si ngl e l i ne. Addi ti onal anal ysi s
coul d be performed to determi ne as a functi on of temperature by
pl otti ng the i nformati on provi ded i n Tabl e 1.13. Master-curves can be
very useful i n compari ng data from di fferent products such as concen-
trated orange jui ce made from di fferent vari eti es of oranges.
1.14.6. Influence of the Yield Stress in Coating Food
Consi der the rol e of the fl ui d yi el d stress ( )i n coati ng food by exami ni ng
potenti al fl ow and coati ng thi ckness on an i ncl i ned pl ane.
Assumi ng the shear stress of the ai r on the free surface i s negl i gi bl e,
a force bal ance on a fl ui d on an i ncl i ned pl ane (Fi g. 1.50) yi el ds an
equati on gi vi ng shear stress as a functi on of (Churchi l l , 1988):
1 3 10 30 100 300 1,000 3,000
1
2
5
10
20
50
100
200
500
1,000
Shear Rate / a , 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
Master-Curve
T
-18.8 C -5.4 C 9.5 C 29.2 C
/ a
T
a
T
a
T
o
y
Figure 1.50. Coating material on an inclined plane.
[1.105]
where i s the di stance perpendi cul ar to the pl ane, i s the hei ght or
thi ckness of the coati ng fl ui d l ayer, i s the densi ty of the fl ui d, i s the
angl e of i ncl i nati on, and i s the accel erati on due to gravi ty. Thi s
equati on shows that the shear stress i s a maxi mum on the surface of
the wal l (the i ncl i ned pl ane) and decreases i n movi ng toward the free
surface of the coati ng; hence, the maxi mum shear stress may be
expressed as
[1.106]
When the maxi mum shear stress exceeds the yi el d stress ( ),
gravi ty al one wi l l cause fl ow down the wal l produci ng a gross saggi ng
phenomenon known as sl umpi ng i n the pai nt i ndustry (Patton, 1964).
Si mi l ar probl ems can be observed i n food coati ngs such as mi l k
chocol ate, vani l l a frosti ng, and barbecue sauce.
I f , the fl ui d wi l l remai n on the surface of the object bei ng
coated. The maxi mum coati ng thi ckness possi bl e, wi thout materi al
fl owi ng off the surface, may be cal cul ated i n terms of the yi el d stress
and the angl e of i ncl i nati on:
[1.107]
h
y
f ( y) g(h y) sin
y h

g
max
f (0) gh sin
max
>
o
o

max
h
max

o
gsin
1.14.6 Influence of the Yield Stress in Coating Food 93
The yi el d stress concept i s a very useful tool for exami ni ng the thi ckness
of food coati ngs. Equati ons gi ven above al so appl y to a verti cal wal l
where resul ti ng i n
[1.108]
sinsin90 1
h
max

o
g
Chapt e r 2 . Tube Vis c ome t ry
2.1. Introduction
Tube vi scometers are very useful i n col l ecti ng rheol ogi cal data.
These i nstruments may be pl aced i nto three basi c categori es: gl ass
capi l l ari es (Fi g. 2.1), often cal l ed U-tube vi scometers because of thei r
resembl ance to the l etter U; hi gh pressure capi l l ari es (Fi g. 2.2); and pi pe
vi scometers (Fi g. 2.3). Al l establ i sh a pressure di fference to create fl ow.
The major di fference between a capi l l ary and a pi pe vi scometer i s the
di ameter of the tube. Al though there i s no cl earl y defi ned si ze at whi ch
a tube shoul d be cal l ed a capi l l ary or a pi pe, some gui del i nes can be
offered.
Figure 2.1. Ostwald and Canon-Fenske glass capillary (U-tube) viscometers.
Di ameters i n commerci al capi l l ary i nstruments typi cal l y range from
0.1 to 4 mm wi th a vari ati on i n entrance angl es (Fi g. 2.2) of 15 to 90
degrees. Pi pe vi scometers are usual l y bui l t "on-si te" so si ze vari es
wi del y. Some may be as smal l as 7mm i n di ameter but val ues greater
V
L
1
2
Etched Lines
Capillary
Efflux Vessel
Receiving Vessel
Cannon-Fenske Ostwald
2.1 Introduction 95
than 12 mm (typi cal l y 12 to 32mm) are not uncommon i n food appl i -
cati ons. val ues i n tube vi scometers range from 2 to 400: The smal l er
val ues are found i n the capi l l ary uni ts but are sel dom seen i n pi pe
systems.
Figure 2.2. High pressure capillary viscometer.
I n typi cal operati on, the U-tube vi scometer i s fi l l ed by i nverti ng i t
i nto a sampl e and sucki ng (i nto the si de wi th the capi l l ary) fl ui d i nto
the fi xed sampl e bul b. The vi scometer i s turned upri ght, then pl aced
i n a temperature control l ed bath and al l owed to reach thermal equi -
l i bri um. After a certai n peri od of ti me, usual l y 5 to 10 mi nutes, the fl ui d
i s al l owed to fl ow down through the capi l l ary. A stopwatch i s started
when fl ui d passes the upper etched l i ne and stopped when the fl ui d
surface passes the l ower etched l i ne. The resul ti ng ti me i s consi dered
the effl ux ti me for fl ui d di scharge from the bul b and fl ui d vi scosi ty i s
cal cul ated from thi s val ue.
U-tube vi scometers are desi gned as gravi ty operated i nstruments.
Hi gh pressure capi l l ari es, whi ch may al so be constructed from gl ass
tubes but are not "U-shaped," are typi cal l y pi ston dri ven (Fi g. 2.2) or
gas operated. A pump or gas (Fi g. 2.3) system can be used to create a
L/D
L
R
Piston
Entrance Angle
96 Chapter 2. Tube Viscometry
dri vi ng force i n pi pe vi scometers. These uni ts can be operated at el e-
vated pressures such as those found i n asepti c food processi ng equi p-
ment (Dai l and Steffe, 1990a and 1990b).
Figure 2.3. Gas driven pipe viscometer.
Raw data for tube vi scometers are pressure drop and vol umetri c fl ow
rate. The pressure drop i s determi ned from pressure transducers or, i n
the case of U-tube vi scometers, the hei ght of fl ui d above a reference
poi nt. I n hi gh pressure capi l l ari es, fl ow rates are cal cul ated from the
assumpti on that vol umetri c fl ow i n the pi ston (or barrel ) and the cap-
i l l ary are equi val ent. Vol umetri c fl ow rate may be determi ned from the
mass fl ow rate measured i n pi pe systems usi ng a mass fl ow meter or a
wei ght tank. Densi ty i s requi red for thi s cal cul ati on because the vol u-
metri c fl ow rate equal s the mass fl ow rate di vi ded by the densi ty. The
quanti ty of fl ow i n a gl ass capi l l ary i s fi xed by the vol ume ( ) of the
effl ux vessel (Fi g. 2.1).
The mai n focus of the current chapter i s ti me-i ndependent fl ui ds.
Rotati onal i nstruments are superi or for the i nvesti gati on of ti me-
dependent materi al s because the sampl e can be mai ntai ned i n the test
chamber duri ng peri odi c eval uati on. Li ttl e attenti on i s gi ven to
eval uati ng the el asti c behavi or of fl ui ds from capi l l ary data. Theoret-
i cal l y, i t i s possi bl e to determi ne normal stress di fferences wi th tube (or
Air Supply
Pressure Regulator
L X
E
R
Pressure Transducers
Mixing Paddle
Temperature
Sensors
V
2.2 Rabinowitsch-Mooney Equation 97
sl i t)vi scometers from data col l ected usi ng vari ous methods such as axi al
thrust, resi dual exi t pressure, or jet expansi on (Wal ters, 1975; Whorl ow,
1992). General l y, these appl i cati ons have focused on pol ymers wi th
mi nor consi derati on gi ven to food products. I n addi ti on, commerci al l y
avai l abl e tube vi scometers requi re speci al modi fi cati ons to make these
measurements. Normal stress i nformati on can be more easi l y cal cu-
l ated from axi al force data generated on cone and pl ate vi scometers.
Tubul ar systems i nvol vi ng back extrusi on (al socal l ed annul ar pumpi ng)
have al sobeen devel oped toeval uate power l aw fl ui ds (Osori oand Steffe,
1986; Steffe and Osori o, 1987).
2.2. Rabinowitsch-Mooney Equation
Derivation of the Basic Equation. Numerous assumpti ons are
requi red i n devel opi ng the Rabi nowi tch-Mooney equati on: fl ow i s l am-
i nar and steady, end effects are negl i gi bl e, fl ui d i s i ncompressi bl e,
properti es are not a functi on of pressure or ti me, temperature i s
constant, there i s nosl i p at the wal l of the tube meani ng that the vel oci ty
of the fl ui d i s zero at the wal l -fl ui d i nterface, and radi al and tangenti al
vel oci ty components are zero.
The starti ng poi nt i n the deri vati on of a tube vi scometer equati on i s
a force bal ance. Consi der a fl ui d fl owi ng through a hori zontal tube of
l ength ( )and i nsi de radi us ( ). A pressure drop ( )over a fi xed l ength
( )i s causi ng fl ow. The force bal ance, equati ng the shear stress causi ng
fl ow to the shear stress resi sti ng fl ow (i .e., the fl ui d), over a core of fl ui d
(Fi g. 2.4) wi th radi us and l ength yi el ds
[2.1]
whi ch can be sol ved for the shear stress:
[2.2]
Eq. [2.2] can al so be obtai ned by starti ng from the general conservati on
of momentum equati ons (Bi rd et al ., 1960; Brodkey and Hershey, 1988;
Darby, 1976; Denn, 1980) as di scussed i n Exampl e Probl em 2.12.1. Eq.
[2.2] depi cts the shear stress varyi ng over the pi pe from zero at the
center ( ) to a maxi mum at the wal l ( ) where the equati on may
be wri tten as
L R P
L
r L
(P)r
2
2rL
f (r)
(P)r
2
2rL

(P)r
2L
r 0 r R
Figure 2.4. Core of fluid in tube flow geometry.
[2.3]
To devel op the shear rate equati ons, a di fferenti al fl ow el ement ( )
must be eval uated. Thi s can be expressed by consi deri ng the steady,
l ami nar fl ow of fl ui d movi ng through an annul us l ocated between the
core, wi th radi us , and the posi ti on :
[2.4]
where i s the l i near vel oci ty at . The total vol umetri c fl ow rate i s found
by i ntegrati ng Eq. [2.4] over the radi us:
[2.5]
Recogni zi ng that , so that becomes the vari abl e of i ntegra-
ti on, al l ows Eq. [2.5] to be wri tten as
[2.6]
The ri ght hand si de of thi s equati on can be i ntegrated by parts:
[2.7]
and si mpl i fi ed by appl yi ng the no sl i p boundary condi ti on whi ch
sti pul ates that the fl ui d vel oci ty i s zero at the wal l of the pi pe or,
mathemati cal l y, :
R
r
L
Flow
Flow fluid core
w

(P)R
2L
dQ
r r + dr
dQ u2rdr
u r
Q

0
Q
dQ
r 0
r R
u2r dr
2rdr dr
2
r
2
Q
r
2
0
r
2
R
2
(u) dr
2
Q ur
2
r
2
0
r
2
R
2
r
2
0
r
2
R
2
r
2
du
u 0 at r R
[2.8]
Tofurther eval uate Eq. [2.8], a number of i tems must be noted. Fi rst,
by assumi ng steady l ami nar fl ow, we know the shear rate i s some
functi on of the shear stress:
[2.9]
or
[2.10]
The negati ve si gn i s requi red i n Eq. [2.9] because we assume, as
i ndi cated i n Eq. [2.1], the posi ti ve di recti on of to be opposi te the
di recti on of fl ow. Second, Eq. [2.2] and [2.3] can be combi ned to gi ve
[2.11]
whi ch, when di fferenti ated, yi el ds
[2.12]
Taki ng the expressi on for gi ven by Eq. [2.10], and i nserti ng Eq. [2.12]
for , gi ves
[2.13]
Usi ng Eq. [2.13] and noti ng, from Eq. [2.11], that al l ows
Eq. [2.8] to be rewri tten as
[2.14]
Observe the change i n the l i mi ts of i ntegrati on: goes from 0 to as
goes from 0 to . Si mpl i fyi ng thi s expressi on gi ves the fi nal general
equati on rel ati ng shear stress and shear rate:
[2.15]
Eq. [2.15] may be eval uated by di fferenti ati on usi ngLei bni tzrul e, whi ch
al l ows an i ntegral of the form
Q [(0)R
2
u(0)
2
]
r
2
0
r
2
R
2
r
2
du
r
2
0
r
2
R
2
r
2
du
du
dr
f ()
du f () dr
w
R
dr

w
_
,
d
du
dr
du f () dr f ()
w
_
,
d
r
2
()
2
R
2
/(
w
)
2
Q
()
2
(
w
)
2
_
,
R
2
f ()
R
w
_
,
d

w
R
2
r
Q
R
3

1
(
w
)
3
w
()
2
f () d
[2.16]
to be wri tten as
[2.17]
to di fferenti ate the i ntegral . Wri ti ng, Eq. [2.15] as
[2.18]
then, appl yi ng Lei bni tz rul e on the ri ght hand si de, and di fferenti ati ng
both si des wi th respect to (whi ch i s ) gi ves
[2.19]
Sol vi ng Eq. [2.19] for the shear rate at the wal l ( )yi el ds the wel l known
Rabi nowi tsch-Mooney equati on:
[2.20]
where the deri vati ve i s eval uated at a parti cul ar val ue of . Appl i cati on
of thi s equati on i s demonstrated for soy dough i n Exampl e Probl em
2.12.2.
Eq. [2.20] can al so be expressed i n terms of the apparent wal l shear
rate, :
[2.21]
where . Further mani pul ati on gi ves
[2.22]
or
[2.23]
that can be rewri tten i n the fol l owi ng si mpl i fi ed format:
[2.24]
where
d
dz
'
0
z
z
2
f (z) dz
(z)
2
f (z)
Q
R
3
_
,
(
w
)
3
w
()
2
f () d
w
z
(
w
)
3
d(Q/(R
3
))
d
w
1
1
]
+
Q
R
3
_
,
3(
w
)
2
(
w
)
2
f (
w
)
w
f (
w
)

3Q
R
3
_
,
+ (
w
)
d(Q/(R
3
))
d
w
1
1
]
w
f (
w
)

3
4
_
,
+
w
4
_
d
d
w
_
,
4Q/(R
3
)
w
f (
w
)

3
4
_
,
+
w
4
_
d
d
w
_
,
1
1
]
w
f (
w
)

3
4
_
,
+
1
4
_
d(ln )
d(ln
w
)
_
,
1
1
]
3n + 1
4n
1
1
]

[2.25]
showi ng that i s the poi nt sl ope of the versus . I f the fl ui d
behaves as a power l aw materi al , the sl ope of the deri vati ve i s a strai ght
l i ne and . Eq. [2.24] i s a conveni ent form of the Rabi nowi tsch-
Mooney equati on because power l aw behavi or i s frequentl y observed
wi th fl ui d foods. Al so, sl i ght curvature i n the l ogari thmi c pl ot can often
be i gnored. Appl i cati on of Eq. [2.24] and [2.25] i s i l l ustrated for a 1.5%
sol uti on of sodi um carboxymethyl cel l ul ose i n Exampl e Probl em 2.12.3.
Newtonian Fluids. I n devel opi ng the Rabi nowi tsch-Mooney equati on
a general expressi on rel ati ng shear stress to shear rate, Eq. [2.15], was
devel oped:
Thi s can be sol ved for a Newtoni an fl ui d by i nserti ng the Newtoni an
defi ni ti on for shear rate, :
[2.26]
I ntegrati on of Eq. [2.26] gi ves
[2.27]
Substi tuti ng the shear stress at the wal l ( ) i nto Eq. [2.27]
resul ts i n the Poi seui l l e-Hagen equati on:
[2.28]
Eq. [2.28] i ndi cates that the radi us has a very strong i nfl uence on
the behavi or of the system si nce i t i s rai sed to the power four. Al so, i f
thi s equati on i s wri tten i n terms of the defi ni ti on of a Newtoni an fl ui d
( ), then the formul a for the shear rate at the wal l may be
determi ned:
[2.29]
n
d(ln
w
)
d(ln )
n ln(
w
) ln()
n n
Q
R
3

1
(
w
)
3
w
()
2
f () d
f () /
Q
R
3

1
(
w
)
3
w
()
2
(/) d
1
(
w
)
3
()
3
,
d
Q
R
3

1
(
w
)
3
_
()
4
4
_
,
1
1
10

w
4
w
(P)R/(2L)
Q
(P)R
4
8L
w

4Q
R
3
Thi s expressi on i s i denti cal to the one gi ven i n Eq. [1.90] (and Fi g. 1.27)
for esti mati ng the maxi mum shear rate of a Newtoni an fl ui d i n tube
fl ow.
Power LawFluids. Eq. [2.15] can be sol ved for a power l aw fl ui d by
i nserti ng i nto the equati on:
[2.30]
I ntegrati on and substi tuti on of gi ves
[2.31]
I f thi s equati on i s wri tten i n terms of the defi ni ti on of a power l aw fl ui d
( ), then the formul a for the shear rate at the wal l may be
determi ned:
[2.32]
Eq. [2.32] i s an exact sol uti on for a power l aw fl ui d and al so useful as
an esti mate of the maxi mum shear rate i n tube fl ow (as i ndi cated i n Eq.
[1.91]) for a wi de range of fl ui d foods.
BinghamPlastic Fluids. I n a Bi ngham pl asti c fl ui d, the shear rate
i s defi ned i n terms of the pl asti c vi scosi ty and the yi el d stress:
. Thi s functi on i s di sconti nuous because there i s no
sheari ng fl ow at poi nts i n the tube near the center where the shear
stress i s bel ow the yi el d stress. Mathemati cal l y, for i n
the central pl ug regi on and for i n the outer
porti on of the tube. Gi ven the above, i t i s cl ear that Eq. [2.15] must be
i ntegrated for each regi on of the tube to determi ne the total vol umetri c
fl ow rate:
[2.33]
Si nce shear rate i s zero when the shear stress i s bel ow the yi el d stress,
Eq. [2.33] can be si mpl i fi ed to
[2.34]
f () (/K)
1/n
Q
R
3

1
(
w
)
3
1
1
]
w
()
2
K
_
,
1/n
d

1
(
w
)
3
K
1/n
1
1
]
w
()
2 + 1/n
d
w
(P)R/(2L)
Q
(P)
2LK
_
,
1/n
n
3n + 1
_
,
R
(3n + 1)/n
w
K(
w
)
n
3n + 1
4n
_
4Q
R
3
_
f () (
o
)/
pl
f () 0 0 < <
o
f () (
o
)/
pl

o
<
w
Q
R
3

1
(
w
)
3
1
1
]
o
()
2
f () d+
1
(
w
)
3
1
1
]
w
()
2
pl
_
,
d
Q
R
3

1
(
w
)
3
1
1
]
w
()
2
pl
_
,
d
2.3 Laminar Flow Velocity Profiles 103
whi ch, when i ntegrated, yi el ds the Bucki ngham-Rei ner equati on for
fl ow of a Bi ngham pl asti c materi al i n a pi pe:
[2.35]
The same techni que di scussed for Bi ngham pl asti cs i s used to deri ve
the fl ow rate equati on for Casson fl ui ds i n Exampl e Probl em 2.12.4.
Herschel-Bulkley Fluids. The vol umetri c fl ow rate for a Herschel -
Bul kl ey fl ui d ( ), i s found usi ng the same method di scussed
for the Bi ngham pl asti c materi al :
[2.36]
2.3. Laminar Flow Velocity Profiles
Rheol ogi cal properti es have a strong i nfl uence on fl ui d vel oci ty
profi l es i n tube fl ow. Understandi ng these profi l es i s i mportant i n
devel opi ng a cl ear pi cture of i nstrument performance and i n maki ng
vari ous food process engi neeri ng desi gn cal cul ati ons such as deter-
mi ni ng the appropri ate hol d tube l ength for a thermal processi ng sys-
tem. Si nce tube vi scometers operate i n the l ami nar fl ow regi me, onl y
l ami nar fl ow vel oci ty profi l es are presented here. Equati ons descri bi ng
the turbul ent vel oci ty of Newtoni an and power l aw fl ui ds are di scussed
l ater i n Secti on 2.11.
Combi ni ngthe shear stress rel ati onshi p (Eq. [2.2])and the defi ni ti on
of a Newtoni an fl ui d (Eq. [1.25]) wi th the shear rate wri tten as
yi el ds
[2.37]
I ntegrati ng thi s equati on, usi ng the no sl i p boundary condi ti on
( ) to determi ne the unknown constant, gi ves an expressi on
for the vel oci ty profi l e of a Newtoni an fl ui d:
[2.38]
I f the defi ni ti on of a power l aw fl ui d i s used, Eq. [2.37] becomes
[2.39]
Q

R
4
(P)
8
pl
L
_
4
3
_
w
_
,
+
1
3
_
w
_
,
4
1
1
]
K(
)
n
+
o
Q

R
3
256
_
4n
3n + 1
_
w
K
_
,
1/n
1

o
w
_
,
1/n
1
(
o
/
w
)
2n + 1
1 +
2n
n + 1
w
_
1 +
n
o
w
_
,
1
1
]
1
1
]
du/dr
du
dr
_
,

(P)r
2L
u 0 at r R
u f (r)
(P)
4L
(R
2
r
2
)
K
du
dr
_
,
n
(P)r
2L
The vel oci ty profi l e i s found after i ntegrati on and appl i cati on of the no
sl i p boundary condi ti on:
[2.40]
By combi ni ng Eq. [2.40] wi th the vol umetri c fl ow rate equati on (Eq.
[2.31]), the rel ati onshi p between the vel oci ty at and the vol umetri c
average vel oci ty ( ) may be cal cul ated:
[2.41]
Lower val ues of the fl ow behavi or i ndex resul t i n a fl atter vel oci ty profi l e
and hi gher val ues maxi mi ze the di fference between the sl owest and
fastest fl ui d el ements (Fi g. 2.5). The extreme case occurs wi th at
where . The curve (Fi g. 2.5) for a Newtoni an fl ui d i s
found at .
Figure 2.5. Laminar velocity profiles for power law fluids with different values
of the flow behavior index.
The maxi mum vel oci ty, l ocated at the center l i ne where , may
be determi ned from Eq. [2.41] for any val ue of the fl ow behavi or i ndex:
u f (r)

(P)
2LK
_
,
1/n
n
n + 1
_
,
[R
(n + 1)/n
r
(n + 1)/n
]
r
u Q/(R
2
)
u/( u )

3n + 1
n + 1
_
r
R
_
,
(n + 1)/n
1
1
]
n
r/R 0 u/ u 3.0
n 1.0
0
0.5
1
1.5
2
2.5
1 0.5 0 0.5 1
n=2.5
n=2.0
n=1.0
n=0.7
n=0.4
n=0.1
r/R
Power Law Fluid
Velocity Profiles
u

/

u
r 0
2.3 Laminar Flow Velocity Profiles 105
[2.42]
Cl earl y, the fl ow behavi or i ndex has a very strong i nfl uence on the
vel oci ty profi l e. Vel oci ty profi l es for power l aw fl ui ds i n turbul ent fl ow
are very fl at by compari son (see Secti on 2.11, Tabl e 2.7).
I n the case of a Bi ngham pl asti c fl ui d, the force bal ance equati on i s
[2.43]
whi ch i s val i d i n the outer regi ons of the pi pe where the fl ui d i s subject
to shear. Wi th the no sl i p boundary condi ti on, Eq. [2.43] may be
i ntegrated to gi ve an expressi on for the vel oci ty profi l e between the pl ug
and the wal l of the tube:
[2.44]
for where . The cri ti cal radi us ( ), whi ch defi nes the
outer boundary of the pl ug, may be cal cul ated from the yi el d stress:
[2.45]
The maxi mum vel oci ty i n the pi pe, whi ch i s the vel oci ty of the pl ug i n
the center secti on of the pi pe, i s found by substi tuti ng i nto Eq.
[2.44]:
[2.46]
whi ch i s val i d for where .
Combi ni ng Eq. [2.44] or Eq. [2.46] wi th the vol umetri c fl ow rate
equati on (Eq. [2.35]), the rel ati onshi p between the vel oci ty at and the
vol umetri c average vel oci ty ( ) may be cal cul ated as
[2.47]
i n the sheared porti on of the fl ui d ( ), and
[2.48]
u
max
/( u )
3n + 1
n + 1
pl
du
dr
_
,
+
o

(P)r
2L
u f (r)
(P)R
2
4
pl
L
r
R
_
,
2
2R
o
R
1
r
R
_
,
1
1
]
R
o
r R
o
R
o
R
o

o
2L
(P)
r R
o
u
max

(P)R
2
4
pl
L
1
R
o
R
_
,
2
1
1
]
0 < r R
o

o
r
u Q/(R
2
)
u/( u )
2(1 2c + 2cr/R (r/R)
2
)
1 4c/3 + c
4
/3
1 r/R c
u/( u )
2(1 c)
2
1 4c/3 + c
4
/3
i n the pl ug ( ). The val ue of i s defi ned i n terms of the yi el d stress
or the cri ti cal radi us: . Pl otti ng these equati ons show
that i ncreasi ng the yi el d stress enl arges the radi us of the pl ug fl owi ng
down the center of the pi pe (Fi g. 2.6). The curve for a Newtoni an fl ui d
i s i l l ustrated by the l i ne wi th
Figure 2.6. Laminar velocity profiles for Bingham plastic fluids with different
values of .
The same approach descri bed for Bi ngham pl asti c fl ui ds can be used
to determi ne the vel oci ty profi l e for a Herschel -Bul kl ey fl ui d:
[2.49]
Vel oci ty of the pl ug may be determi ned by usi ng Eq. [2.49] wi th ,
cal cul ated from Eq. [2.45].
r/R c c
c
0
/
w
R
o
/R
c 0.
0
0.5
1
1.5
2
1 0.5 0 0.5 1
r/R
u

/

u
c=0.0
c=0.2
c=0.4
c=0.6
c=0.8
Bingham Plastic Fluid
Velocity Profiles
c
o
/
w
u f (r)
2L
(P) (1 + 1/n)K
1/n
(
w

o
)
1 + 1/n
(P)r
2L

o
_
,
1 + 1/n
1
1
]
r R
o
2.4 Laminar Flow Criteria 107
2.4. Laminar Flow Criteria
To obtai n meani ngful rheol ogi cal data, tube vi scometers must
operate i n the l ami nar fl ow regi me. Turbul ence i s rarel y i nvol ved i n
hi gh pressure capi l l ari es but may be a probl em i n pi pe vi scometers wi th
l arge di ameter tubes or when i nvesti gati ng l ow vi scosi ty fl ui ds.
The accepted cri teri on for l ami nar fl ow of Newtoni an fl ui ds i n tube
fl ow i s to mai ntai n a Reynol ds number l ess than 2100: where
. Transi ti on away from l ami nar fl ow, however, has been
observed for val ues between 1225 and 3000. I n addi ti on, l ami nar
fl ow, under i deal condi ti ons, has been observed for numbers as hi gh
as 50,000 (Grovi er and Azi z, 1972). Si mi l ar probl ems exi st for non-
Newtoni an fl ui ds. Hence, the fol l owi ng di scussi on can onl y offer general
gui del i nes to determi ne i f the fl ow regi me i s l ami nar or turbul ent.
As noted above, fl ow i s consi dered l ami nar for Newtoni an fl ui ds
when the Reynol ds number i s bel ow the cri ti cal val ue of 2100:
[2.50]
Wi th power l aw fl ui ds, l ami nar fl ow occurs when (Hanks, 1963):
[2.51]
where the term on the ri ght si de of the i nequal i ty i s the cri ti cal power
l aw Reynol ds number. The power l aw Reynol ds number i s defi ned as
[2.52]
Pl otti ng Eq. [2.51] shows the cri ti cal Reynol ds number i ncreasi ng
sharpl y toa maxi mum near , then droppi ng off wi th greater val ues
of the fl ow behavi or i ndex (Fi g. 2.7). Al so, for the speci al case of a
Newtoni an fl ui d ( ) i t i s equal to the expected val ue of 2100.
Eq. [2.51] may gi ve a conservati ve esti mate of the cri ti cal Reynol ds
number. Work by Campos et al . (1994) shows that cri ti cal Reynol ds
numbers i n the range of 4000 (at ) to 3000 (at ) may be
determi ned usi ng a stati sti cal eval uati on procedure. Hi gher numbers
are al sopredi cted, parti cul arl y for , usi ng the equati on deri ved
by Mi shra and Tri pathi (1973):
N
Re
< 2100
N
Re
D u /
N
Re
N
Re
N
Re
< 2100 (N
Re
)
critical
N
Re, PL
<
6464n(2 + n)
(2 + n)/(1 + n)
(1 + 3n)
2
(N
Re, PL
)
critical
N
Re, PL

D
n
( u )
2 n
8
n 1
K
_
4n
3n + 1
_
,
n
n 0.4
n 1
n 0.46 n 0.78
0 < n < 0.6
[2.53]
Eq. [2.53] predi cts gradual l y decreasi ng val ues of cri ti cal Reynol ds
numbers from 3480 at to 2357 at . Over , the
Mi shra and Tri pathi (1973) equati on predi cts val ues (Fi g. 2.7) very
si mi l ar to those gi ven by Eq. [2.51].
Figure 2.7. Variation of the critical power law Reynolds number with .
Wi th tube fl ow of a Bi ngham pl asti c fl ui d, steady-state l ami nar fl ow
can be expected (Hanks, 1963) when
[2.54]
The term on the ri ght si de of Eq. [2.54] i s the cri ti cal Bi ngham Reynol ds
number and , the cri ti cal val ue of , i s defi ned as
[2.55]
(N
Re, PL
)
critical

2100(4n + 2) (5n + 3)
3(1 + 3n)
2
n 0.1 n 0.60 0.65 < n < 1.0
0 0.2 0.4 0.6 0.8 1
1,000
1,500
2,000
2,500
3,000
3,500
4,000
Flow Behavior Index
C
r
i
t
i
c
a
l

P
o
w
e
r

L
a
w

R
e
y
n
o
l
d
s

N
u
m
b
e
r
Eq.[2.51] from Hanks (1963)
Eq.[2.53] from Mishra and Tripathi (1973)
n
N
Re, B

N
He
8c
c
(1 4c
c
/3 + c
c
4
/3) (N
Re, B
)
critical
c
c
c
c
c
(1 c
c
)
3

N
He
16, 800
2.4 Laminar Flow Criteria 109
The Hedstrom number and the Bi ngham Reynol ds number are defi ned
by the fol l owi ng equati ons:
[2.56]
[2.57]
Cri ti cal val ues of versus (Eq. [2.55]), and the cri ti cal Bi ngham
Reynol ds number versus the Hedstrom number (Eq. [2.54]) are pl otted
i n Fi g. 2.8 and 2.9, respecti vel y. These i l l ustrati ons i ndi cate the
di ffi cul ty i n achi evi ng turbul ence wi th materi al s havi ng a si gni fi cant
yi el d stress. Note that the Hedstrom number i ncreases wi th l arger
val ues of the yi el d stress. Lami nar fl ow cri teri on i s al so avai l abl e (but
not experi mental l y veri fi ed)for Herschel -Bul kl ey materi al s (Garci a and
Steffe, 1987).
Figure 2.8. Variation of with the Hedstrom number.
N
He

D
2
(
pl
)
2
N
Re, B

Du
pl
c N
He
10 100 1,000 10,000 100,000 1,000,000 10,000,000
0
0.2
0.4
0.6
0.8
1
C
c
N
He
c
c
2.5. Data Corrections
There are numerous measurement errors whi ch may occur i n usi ng
tube vi scometers (Tabl e 2.1). Some are general l y appl i cabl e and others
onl y appl y to speci fi c systems. Data correcti ons requi red for hi gh
pressure and pi pe vi scometers are si mi l ar, and consti tute the focus of
thi s secti on. Gl ass capi l l ari es (gravi ty operated U-tube vi scometers)
have speci al requi rements whi ch are di scussed i n Sec. 2.9.
Figure 2.9. Variation of the critical Bingham Reynolds number with the Hed-
strom number.
Kinetic Energy Losses. There i s a pressure l oss, due to a di fference
i n ki neti c energy, i n hi gh pressure capi l l ari es caused by the accel erati on
of the fl ui d from the barrel vel oci ty to the capi l l ary vel oci ty. Thi s
pressure l oss ( ) may be expressed i n terms of the ki neti c energy
correcti on factor ( ):
[2.58]
100 300 1,000 3,000 10,000 30,000 100,000 300,000 1,000,000
0
5,000
10,000
15,000
20,000

N
(
N

)
He
R
e
,
B
c
r
i
t
i
c
a
l
P
(( u
2
)
2
( u
1
)
2
)
2.5 Data Corrections 111
where and are the average capi l l ary and barrel vel oci ti es,
respecti vel y. Equati ons for cal cul ati ng are avai l abl e (see Tabl e 2.4 i n
Sec. 2.10). Losses due to ki neti c energy are general l y smal l and very
di ffi cul t toseparate from entrance pressure l osses. Hence, i t i s accepted
practi ce to assume that ki neti c energy l osses are accounted for i n the
entrance effect correcti on.
End Effects: Entrance Correction. Energy l osses due to fl ui d
di vergence at the end of a capi l l ary are smal l and usual l y negl ected but
entrance l osses can be very si gni fi cant and must be eval uated.
Table 2.1. Sources of Error in Operating Tube Viscometers (from Van Wazer et
al., 1963)
Factor Cause Applicability
Kinetic energy losses Loss of effective pressure because General
of the kinetic energy in the issu-
ing stream.
End effects Energy losses due to viscous or General
elastic behavior when a fluid con-
verges or diverges at the ends of a
capillary
Elastic energy Energy loss by elastic deforma- Viscoelastic materials
tion of the fluid not recovered
during flow in the capillary
Turbulence Departure from laminar flow General
Pressure losses prior Sticking of the piston or energy Cylinder-piston
to the capillary dissipated in the flow of the mate- viscometers
rial within the cylinder before
entering the capillary.
Drainage Liquid adhering to the wall of the Glass capillary
viscometer reservoir. viscometers
Surface-tension Variations in surface tension from Glass capillary
effects one test substance to another. viscometers
Heat effects Conversion of pressure energy High-shear viscometers
into heat energy through flow.
Wall effects Surface phenomena at the Polyphase fluids
fluid-wall interface.
Effect of time- Variations in the residence time Thixotropic and rheo-
dependent properties in the capillary. pectic materials
The entrance effect correcti on accounts for excess pressure l oss
occurri ng at the openi ng of the tube vi scometer from sudden changes i n
geometry caused by a convergence, and l osses i n ki neti c energy. Thi s
probl em may be experi mental l y eval uated usi ng a number of tubes
u
2
u
1

Figure 2.10. Bagley plot illustrating entrance effect pressure corrections deter-
mined at three flow rates.
havi ng di fferent rati os (Bagl ey, 1957). Data of total pressure drop
versus fl ow rate are col l ected for each tube. Pressure drop versus
data are pl otted at each fl ow rate (or apparent wal l shear rate, )
and the l i nes are extrapol ated to equal to zero. These fi gures (Fi g.
2.10) are someti mes cal l ed "Bagl ey pl ots." The resul ti ng pressure drop
i s the entrance effect pressure l oss at a parti cul ar fl ow rate and pi pe
di ameter. To achi eve the greatest l evel of accuracy, the same procedure
must be fol l owed for each di ameter under consi derati on by usi ng a
number of tubes (at l east three) havi ng the same di ameter but di fferent
l engths.
Pressure l oss at the entrance has al so been eval uated by subtracti ng
ori fi ce di e data from capi l l ary data. These cal cul ati ons are made wi th
an ori fi ce and capi l l ary of the same di ameter usi ng data taken at an
equi val ent vol umetri c fl ow rate. Resul ts wi th soy protei n i sol ate show
excel l ent agreement between thi s method and the Bagl ey procedure
di scussed previ ousl y (Hyashi et al ., 1991).
Entrance pressure l oss i s used tocorrect the measured pressure drop
val ues:
[2.59]
L/D
P
Q
Q
Q
Q
Q
1
2
3
P
1
Entrance Pressure Loss at Q1
P
3
L/D
L/D
4Q/(R
3
)
L/D
P P
m
P
en
where i s the measured pressure drop, and i s the entrance
pressure l oss. Corrected val ues of are used i n cal cul ati ng the shear
stress at the wal l , Eq. [2.3]. The above techni que i s demonstrated for
soy dough i n Exampl e Probl em 2.12.2.
Toms (1958) descri bed an al ternati ve method of el i mi nati ng end
effects. The concept i nvol ves a l ong and short tube of equi val ent
di ameter and suffi ci ent l ength so both have an i nternal secti on where
fl ow i s ful l y devel oped. I f the vol umetri c fl ow rate i s the same i n both
tubes, then the exi t pressure l oss ( ) and the entrance pressure l oss
( ) are equal for the l ong (subscri pt L) and short (subscri pt S) tubes.
Wi th thi s i dea, the pressure drop across each tube may be wri tten as
[2.60]
and
[2.61]
where and are the pressure l osses i n tube secti ons where fl ow
i s ful l y devel oped. At constant fl ow rate, the pressure gradi ents i n the
ful l y devel oped secti ons are equal :
[2.62]
Subtracti ng Eq. [2.61] from Eq. [2.60], and usi ng Eq. [2.62] yi el ds a
pressure gradi ent term that i s free of end effects:
[2.63]
Eq. [2.63] can be used tocal cul ate the gradi ent term ( )i n cal cul ati ng
the wal l shear stress from Eq. [2.3]. I n practi ce, numerous tubes woul d
be used and care taken to be sure i s l arge enough to avoi d
si gni fi cant experi mental errors.
EntranceLength. Wi th l ong tubes, the entrance correcti on can often
be negl ected. Pressure transducers can be strategi cal l y pl aced i n pi pe
vi scometers so that the entrance regi on does not i nfl uence experi mental
data. Al so, entrance l ength i nformati on may be needed to desi gn, or
eval uate the performance of, tube vi scometers.
P
m
P
en
P
P
ex
P
en
P
L
P
en
+ P
ex
+ P
fd, L
P
S
P
en
+ P
ex
+ P
fd, S
P
fd, L
P
fd, S
P
fd, L
L
L
P
fd, S
L
S
P
fd, L
L
L
P
L
P
S
L
L
L
S
P/L
P
L
P
S
There i s some publ i shed i nformati on for esti mati ng the entrance
l ength ( ) requi red to obtai n 98%of ful l y devel oped fl ow whi ch shoul d
be consi dered a conservati ve gui del i ne i n the desi gn of tube vi scometers.
An approxi mate equati on tocal cul ate for Newtoni an fl ui ds i n l ami nar
fl ow, based on theoreti cal and experi mental studi es (Boger, 1982), i s
[2.64]
The constant, 0.55, accounts for the entrance effect at very l ow val ues
of the Reynol ds number. Entrance l ength equati ons gi ven for power
l aw and Bi ngham pl asti c fl ui ds di scussed subsequentl y do not have an
anal ogous term.
Col l i ns and Schowal ter (1963) presented entrance l ength data for
pseudopl asti c fl ui ds i n l ami nar fl ow. Resul ts were gi ven i n terms of a
modi fi ed power l aw Reynol ds number ( ). There was a l i near
rel ati onshi p from that may be expressed as
[2.65]
Usi ng Eq. [2.52], the modi fi ed power l aw Reynol ds number may be
wri tten i n terms of as
[2.66]
whi ch can be substi tuted i nto Eq. [2.65] yi el di ng
[2.67]
Pl otti ng Eq. [2.67], assumi ng constant val ues, shows
i ncreasi ng wi th i ncreasi ng val ues of the fl ow behavi or i ndex (Fi g. 2.11)
when .
Entrance l ength data have al so been publ i shed for Bi ngham pl asti c
fl ui ds i n l ami nar fl ow (Mi chi yosi et al ., 1966). Thei r data may be
summari zed wi th the fol l owi ng equati on:
X
E
X
E
X
E
D
0.55 + 0.055 (N
Re
)
D
n
( u )
(2 n)
/K
0.1 n 1.0
2X
E
D
(.250n + .350)
D
n
( u )
(2 n)
K
_
,
N
Re, PL
D
n
( u )
(2 n)
K
_
,
N
Re, PL
(8
n 1
)
3n + 1
4n
_
,
n
X
E
DN
Re, PL
(.125n + .175) (8
n 1
)
3n + 1
4n
_
,
n
N
Re, PL
X
E
/D
0.1 < n < 1.0
Figure 2.11. Entrance length of power law fluids.
[2.68]
where, recal l , . Pl otti ng Eq. [2.68] i l l ustrates (Fi g. 2.12)
decreasi ng entrance l ength requi rements wi th greater val ues of the
yi el d stress.
Wi th short or medi um l ength tubes, entrance effects must be
experi mental l y eval uated; however, when tubes are suffi ci entl y l ong
(easi l y desi gned thi s way wi th pi pe vi scometers), the entrance correcti on
can be negl ected. I f i s i n the order of 0.01 probl ems shoul d be
mi ni mal but tubes may be l ong. Dervi sogl u and Koki ni (1986) found,
through tri al and error experi mentati on, that an entrance l ength of 90
di ameters was suffi ci ent to mi ni mi ze entrance effect errors i n studyi ng
vari ous food products i ncl udi ng mustard, ketchup, appl esauce, and
tomato paste. Care shoul d be exerci sed wi th el asti c fl ui ds because the
entrance l ength may be si gni fi cantl y hi gher than predi cted by equati ons
0.2 0.4 0.6 0.8
0.02
0.03
0.04
0.05
0.06
E
R
e
,
P
L
X
D
N
n
X
E
DN
Re, B
.0476exp(5.125c)
c
o
/
w
X
E
/L
Figure 2.12. Entrance length of Bingham plastic fluids.
such as those presented i n thi s secti on (Whorl ow, 1992). A tube vi s-
cometer, where the end effect may be negl ected due to hi gh val ues,
i s di scussed i n Exampl e Probl em 2.12.3.
Wall Effects- SlipCorrection. Sl i p occurs when a thi n l ayer of fl ui d,
havi ng a vi scosi ty l ower than that of the fl ui d, forms at the wal l of the
tube (or the wal l of any vi scometer). Thi s may be a probl em i n food
suspensi ons l i ke frui t and vegetabl e purees. Theoreti cal l y, the probl em
may be attacked by addi ng an addi ti onal term, representi ng added fl ow,
to the overal l fl ow rate term. The expressi on descri bi ng the vol umetri c
fl ow rate (Eq. [2.15]) may be wri tten as
[2.69]
At constant val ues of , the above i ntegral term i s constant so a sl i p
vel oci ty may be i ntroduced to account for vari ati ons i n the measured
val ues of fl ow rate:
0 0.2 0.4 0.6 0.8
0
0.005
0.01
0.015
0.02
0.025
0.03
0.035
c
X
D
N
E
R
e
,
B
L/D
Q
without slip
Q
ws

R
3
(
w
)
3
w
()
2
f () d
w
[2.70]
where i s the effecti ve sl i p vel oci ty whi ch i s assumed to be a functi on
of the shear stress at the wal l . I n the absence of sl i p, .
Eq. [2.70] can be expanded by defi ni ng an effecti ve sl i p coeffi ci ent
( ) such that
[2.71]
or
[2.72]
Si mpl i fi cati on yi el ds
[2.73]
whi ch i s a form of the equati on often seen i n the l i terature. Jastrzebski
(1967), however, found that the sl i p coeffi ci ent was a functi on of the wal l
shear stress and al so vari ed i nversel y wi th tube radi us. To account for
thi s fi ndi ng, a corrected sl i p coeffi ci ent ( ) was defi ned to gi ve a better
mathemati cal representati on of experi mental data:
[2.74]
Usi ng , the vol umetri c fl ow rate expressi on i s wri tten as
[2.75]
may be eval uated from capi l l ary tube measurements usi ng vari ous
tubes (at l east three) havi ng di fferent radi i . Experi mental data of
versus are pl otted at di fferent val ues of . From thi s pl ot,
val ues of at di fferent val ues of and constant are obtai ned.
Then, by pl otti ng versus , at constant , can be
determi ned as the sl ope of the l i ne (Eq. [2.73]) at each val ue of . The
Q
measured
Q
m
Q
ws
+ R
2
u
s
u
s
u
s
0
u
s
/
w
Q
m
Q
ws
w
+
R
2
u
s
w
Q
m
Q
ws
w
+ R
2
Q
m
R
3
Q
ws
R
3
w
+

R
c

c
R
c
Q
m
R
3
Q
ws
R
3
w
+

c
R
2
Q
m
/(R
3
w
)
w
R
Q
m
/(R
3
w
) R
w
Q
m
/(R
3
w
) 1/R
w

w
resul ti ng i nformati on can be used to eval uate as a functi on of . I f
desi red, the same procedure can be fol l owed to fi nd by consi deri ng
i n pl ace of .
The effecti ve sl i p coeffi ci ent, or the corrected sl i p coeffi ci ent, i s used
to correct the vol umetri c fl ow rate data:
[2.76]
or
[2.77]
i s used i n the Rabi nowi tsch-Mooney equati on (Eq. [2.20]) to deter-
mi ne the shear rate at the wal l .
I ntroduci ng an effecti ve sl i p coeffi ci ent may be val uabl e i n deal i ng
wi th suspensi ons such as frui t and vegetabl e purees but can gi ve
physi cal l y meani ngl ess resul ts (those produci ng negati ve val ues) for
dense suspensi ons where sl i di ng fri cti on may be a si gni fi cant factor. I n
these cases i t may be more appropri ate to model materi al movements
as pl ug fl ow i nstead of vi scous fl ow. Thi s approach was taken by Jasberg
et al . (1979) i n exami ni ng the fl ow of defatted soy fl akes i n the screw
channel of an extruder.
Koki ni and Pl utchok (1987) have reported val ues of the corrected
sl i p coeffi ci ent for appl esauce (Tabl e 2.2) and sl i p vel oci ti es for potato
paste are presented i n Hal l i day and Smi th (1995). Corrected sl i p
coeffi ci ents were al so successful l y used by Shukl a and Ri zvi (1995) to
eval uate the fl ow behavi or of butter.
Viscous Heating. Experi mental tests shoul d be desi gned to avoi d
si gni fi cant temperature i ncreases due to vi scous heat di ssi pati on. Vi s-
cous heati ng occurs when any fl ui d i s sheared. Fortunatel y, seri ous
experi mental errors are general l y found onl y wi th fl ui ds havi ng a very
hi gh vi scosi ty or i n i nstruments operated at very hi gh shear rates.
To determi ne the temperature ri se due to vi scous heati ng, the
equati ons of energy and momentum, coupl ed by the temperature
dependence of vi scosi ty, must be sol ved. I t i s not possi bl e to obtai n an
exact anal yti cal sol uti on to thi s probl em (Warren, 1988). Reasonabl e
esti mates are possi bl e usi ng si mpl i fyi ng assumpti ons and a nomograph
outl i ni ng a sol uti on of thi s type i s avai l abl e i n Deal y (1982) and
Mi ddl eman (1968).

w
c
1/R
2
1/R
Q
ws
Q
m

w
R
2
Q
ws
Q
m

c
w
R
Q
ws

Table 2.2. CorrectedSlipCoefficients for Applesauce Calculatedwith Tubes having
=65 and Radii of 0.4425 cm, 0.3235 cmand 0.2150 cm(Data fromKokini and
Plutchok, 1987)
Pa m
2
/(Pa s)
40 .0030
54 .0051
68 .0076
81 .010
95 .013
109 .016
123 .019
136 .022
150 .025
A si mpl e method to eval uate the extent of the probl em i s di scussed
by Whorl ow (1992). He shows that a pressure drop ( )causi nga vol ume
( ) of materi al to fl ow resul ts i n work done per second equal to .
Thi s energy causes the mean temperature of the vol ume to i ncrease by
:
[2.78]
where i s the densi ty and i s the speci fi c heat of the fl ui d. Many fl ui d
foods are aqueous systems havi ng a densi ty and speci fi c heat of
approxi matel y 1000 kg m
-3
and 4000 J kg
-1
K
-1
, respecti vel y. I f you
assume that a temperature ri se of l ess than 1 C has a negl i gi bl e effect
on rheol ogi cal testi ng, then a pressure drop l ess than 4,000 kPa
(approxi matel y 40 atmospheres) wi l l not cause a probl em due to vi scous
heati ng! I t i s i mportant toreal i ze that thi s cal cul ati on wi l l overesti mate
the magni tude of the probl em. The effects of vi scous heati ng probl ems
may al so be eval uated i n terms of the Nahme number defi ned as the
temperature rai se due to vi scous heati ng di vi ded by the temperature
change necessary to al ter vi scosi ty (Macosko, 1994).
HolePressureError. Someti mes the pressure i n a tube i s measured
by a transducer communi cati ng wi th the fl ui d i n the pi pe through a
fl ui d wel l connected to the pi pe. Thi s practi ce causes curvature i n the
fl ow streaml i nes whi ch may create errors i n the pressure measurement.
Wi th pol ymer sol uti ons, i n parti cul ar, the presence of a normal stress
L/D
w

c
P
V PV
T
T
PV
Vc
p
P
c
p
c
p

tends to l i ft the fl ui d out of the hol e maki ng the measured pressure at
the wal l l ess than the true wal l pressure (Bi rd et al . 1987; Barnes et al .,
1989). Thi s phenomenon i s wel l establ i shed and an i nstrument,
desi gned to measure normal stress di fferences, has been devel oped
usi ng the concept (Lodge, 1988). I t i s di ffi cul t to appl y the concept to
fl ui ds wi th a si gni fi cant yi el d stress because these materi al s woul d tend
to fi l l recessed hol es hi nderi ng the proper transmi ssi on of the stress
si gnal .
I n desi gni ng capi l l ary vi scometers, the hol e pressure error can be
el i mi nated by usi ng i denti cal pressure transducers at each sensi ng
l ocati on. Then, when pressure di fferences are cal cul ated, the hol e
pressure errors cancel out. The probl em can al sobe el i mi nated by usi ng
fl ush mounted (di aphragm type) pressure transducers. Hol e pressure
errors are mi ni mal for most food materi al s.
Data Correction Summary. Performi ng a compl ete rheol ogi cal
anal ysi s usi ng tube vi scometry requi res a great deal of data. A typi cal
sequence of the steps i s i l l ustrated i n Fi g. 2.13. To eval uate the entrance
effect, tubes havi ng the same radi us but di fferent l engths are requi red.
The entrance effect shoul d be eval uated for each tube radi us consi dered
before the measured pressure drop can be corrected and the shear stress
at the wal l cal cul ated. Data from tubes of the same l ength, havi ng
di fferent radi i , are needed to eval uate the sl i p coeffi ci ent so the fl ow
rate data can be corrected. Next, the Rabi nowi tsch-Mooney equati on i s
appl i ed and the rheogram devel oped. As menti oned previ ousl y, vi scous
heati ng and hol e pressure errors are usual l y not a probl em wi th food
materi al s.
Accounti ng for sl i p adds si gni fi cantl y to the computati onal
requi rements. One shoul d note that sl i p probl ems wi l l decrease wi th
i ncreasi ng tube si zes. Al so, before correcti ng for sl i p, one shoul d see i f
i t i s present. Thi s can be done by checki ng resul ts from data col l ected
usi ng tubes of di fferent radi i - at l east two tubes wi th si gni fi cantl y
di fferent radi i . After correcti ng for the end effect, rheograms are
compared. Al l owi ng for di fferences i n the shear rate range, the rheo-
grams for ti me-i ndependent fl ui ds shoul d be i denti cal i n the absence of
sl i p.
2.6 Yield Stress Evaluation 121
Figure 2.13. Typical sequence of steps required for the analysis of tube viscome-
ter data for time-independent fluids.
2.6. Yield Stress Evaluation
Yi el d stress may be determi ned i n a tube vi scometer from the stress
to i ni ti ate fl ui d movement (Cheng, 1986). Usi ng thi s method, the
mi ni mum pressure ( ) requi red to cause fl ow i n a hori zontal tube
vi scometer i s measured. The yi el d stress i s cal cul ated from a force
bal ance on the fl ui d (Eq. [2.2]) as
[2.79]
I n practi ce, the pressure drop i s sl owl y i ncreased unti l fl ow i s observed.
I f the structure of the materi al causes the yi el d stress to exhi bi t
ti me-dependent characteri sti cs, then the rate of change of the appl i ed
pressure may i nfl uence resul ts.
2.7. J et Expansion
Jet expansi on (al so cal l ed di e swel l or extrudate swel l ) can be esti -
mated from the pri mary normal stress di fference. Conversel y, the
normal stress di fference can be esti mated from the jet expansi on.
Assumi ng the phenomenon i s caused by el asti c recoi l , due to the sudden
removal of the tube, then swel l i ng can be esti mated as (Tanner, 1988)
determine entrance correction for each radius
determine slip coefficient
correct pressure drop data
correct flow rate data
calculate shear stress at the wall
calculate shear rate at the wall
construct rheogram
P
min
o

P
min
R
2L
[2.80]
where i s the di ameter of the capi l l ary and i s the fi nal di ameter of
the extrudate. The fi rst normal stress di fference di vi ded by the shear
stress ( ), a term cal l ed the recoverabl e shear (Tanner, 1988),
i s eval uated at the tube wal l . Al though thi s equati on i s adequate for
esti mati on purposes, i t excl udes some i mportant factors such as
extensi onal vi scosi ty and vapori zati on (fl ashi ng)of moi sture, whi ch may
strongl y i nfl uence di e swel l of extruded foods.
Newtoni an fl ui ds may exhi bi t jet expansi on havi ng val ues
rangi ng from 1.12 at l ow val ues to 0.87 at hi gh val ues (Mi ddl e-
man, 1977). Pol ymer mel ts may behave si mi l arl y at l ow shear rates but
have numbers from 2 to 4 at hi gh shear rates (Tadmor and Gogos,
1979).
2.8. Slit Viscometry
The devel opment of equati ons for sl i t vi scometers i s anal ogous to
the methods used i n devel opi ng tube vi scometer equati ons. Sl i ts can
be constructed at food processi ng faci l i ti es wi th rel ati ve ease. Food
engi neers have successful l y used them to determi ne the rheol ogi cal
properti es of vari ous extruded food materi al s: mai ze gri ts and potato
fl our (Senouci and Smi th, 1988a and 1988b); corn meal (Bhattacharya
and Padmanabhan, 1992), ri ce fl our dough (Al tomare et al ., 1992), and
wheat fl our dough (Gogos and Bhakuni , 1992). Thi s i nformati on can be
useful i n moni tori ng product qual i ty and i s requi red for vari ous engi -
neeri ng desi gn cal cul ati ons.
Sl i t si ze i s descri bed (Fi g. 2.14) i n terms of the l ength ( ) and aspect
rati o ( ) where and are the wi dth and hei ght of the sl i t, respec-
ti vel y. To negl ect edge effects, the aspect rati o shoul d be greater than
10: . The vel oci ty profi l e for a Newtoni an fl ui d i n the sl i t i s
[2.81]
D
e
D

1 +
1
8
11

22
12
_
,
2
1
1
]
1/6
D D
e
(
11

22
)/
12
D
e
/D
N
Re
N
Re
D
e
/D
L
w/h w h
w/h > 10
u f (x
2
)

3Q
2hw
_
1 4
x
2
h
_
,
2
1
1
]
2.8 Slit Viscometry 123
where at the center of the sl i t and at the outer edge. The
shear rate and shear stress at the wal l ( ) for a Newtoni an fl ui d
are
[2.82]
and
[2.83]
where i s the pressure drop across the sl i t of l ength . The shear
stress equati on, Eq. [2.83], i s val i d for any ti me-i ndependent fl ui d.
Newtoni an vi scosi ty may be cal cul ated as
[2.84]
To cal cul ate the shear rate for non-Newtoni an fl ui ds, a general
equati on rel ati ng the vol umetri c fl ow rate and the shear rate i n the sl i t
i s requi red:
[2.85]
Figure 2.14. Slit viscometer.
x
2
0 x
2
h/2
x
2
h/2
w

6Q
h
2
w
w

Ph
2L
P L

w
/
P
L
_
h
3
w
12Q
_
,
Q
w
2
w
_
,
2
w
f ()d
L
w
h
Flow
1
2
3
x
x
x
Eq. [2.85] i s anal ogous to Eq. [2.15] that was devel oped i n deri vi ng the
Rabi nowi tsch-Mooney equati on. I t may be sol ved di rectl y for power l aw
and Bi ngham pl asti c fl ui ds:
[2.86]
and
[2.87]
respecti vel y, wi th the sti pul ati on that for Eq. [2.87].
I f an unknown fl ui d i s bei ng tested, a general sol uti on to Eq. [2.85]
(comparabl e to the Rabi nowi tsch-Mooney sol uti on) i s requi red:
[2.88]
where the correcti on factor i s defi ned as
[2.89]
I f the fl ui d i s Newtoni an, and, i f power l aw, . The
term i s cal l ed the apparent wal l shear rate for sl i t fl ow. Note, that after
accounti ng for geometri cal di fferences between sl i ts and tubes, Eq.
[2.88] and [2.89] are very si mi l ar to Eq. [2.24] and [2.25].
The above equati ons assume ful l y devel oped fl ow i n the sl i t. I f
correcti ons to experi mental data are needed, the same methods outl i ned
for hi gh pressure capi l l ary vi scometers are appropri ate. I t i s very
i mportant to eval uate the need for an entrance effect correcti on when
usi ng short sl i ts and taki ng as the pressure drop over the enti re sl i t.
A superi or method of obtai ni ng experi mental data i s to use fl ush
mounted pressure transducers i nstal l ed di rectl y on the sl i t. The
transducers (at l east three, preferabl y four) are pl aced a suffi ci ent
di stance from the entrance and exi t of the sl i t so the observed pressure
drop per uni t l ength ( ), requi red i n Eq. [2.83], i s l i near (Han, 1988).
Exi t pressure data from a sl i t have been used tocharacteri ze pri mary
normal stress di fferences i n mol ten pol ymers (Han, 1988). Bhatta-
charya (1993)used thi s method to exami ne the i nfl uence of gl uten l evel s
on the rheol ogi cal behavi or of bread dough. Thi s procedure i nvol ves
Q
wh
2
2(2 + 1/n)
w
K
_
,
1/n
Q
h
2
w
w
6
pl
1
3
2
w
_
,
+
1
2
w
_
,
3
1
1
]
w

o
w
f (
w
)

2n + 1
3n
_
6Q
h
2
w
_
,
n
d ln(Ph/(2L))
d ln(6Q/(wh
2
))

d ln(
w
)
d ln(6Q/(wh
2
))
6Q/(wh
2
) n 1.0 n n
P
P/L
2.9 Glass Capillary (U-Tube) Viscometers 125
l i nearl y extrapol ati ng pressure profi l es through the exi t of the sl i t to
determi ne the exi t pressure ( ). The fi rst normal stress di fference i s
rel ated to the rate of change of the exi t pressure wi th respect to the
shear rate at the wal l :
[2.90]
Obtai ni ng good experi mental data usi ng the exi t pressure method
i s di ffi cul t because smal l pressure di fferences are measured. Han (1988)
argues that the exi t pressure method shoul d produce acceptabl e resul ts
when i s above a cri ti cal val ue of approxi matel y 25 kPa. Unfortu-
natel y, the exi t pressure method i s probl emati c for many foods because
fl ow does not remai n ful l y devel oped through the exi t of the rheometer
vi ol ati ng the assumpti on that the pressure gradi ent i s constant over the
enti re i nstrument (Ofol i and Steffe, 1993). The anal ysi s of sl i t vi s-
cometer data for corn syrup i s gi ven i n Exampl e Probl em 2.12.5.
2.9. Glass Capillary (U-Tube) Viscometers
Gl ass capi l l ary vi scometers (U-tube vi scometers) are desi gned to be
gravi ty operated, and general l y l i mi ted to use wi th Newtoni an fl ui ds
havi ng vi scosi ti es i n the range of 0.4 to 20,000 mPa s. Si mpl e confi gu-
rati ons are seen i n Ostwal d and Cannon-Fenske type uni ts (Fi g. 2.1).
Someti mes, they are dri ven wi th appl i ed external pressure to i ncrease
the vi scosi ty range of the i nstrument. I f the magni tude of the external
pressure makes the magni tude of the stati c pressure head i nsi gni fi cant
i n compari son, the i nstruments may be used for non-Newtoni an fl ui ds
by appl yi ng the Rabi nowi tsch-Mooney equati on to the data. Thi s,
however, i s a l abori ous procedure and shoul d not be consi dered "stan-
dard practi ce"for gl ass capi l l ary vi scometers.
The dri vi ng force i n gravi ty operated gl ass capi l l ary vi scometers i s
the hydrostati c head that vari es duri ng di scharge. Thi s vari ati on i n
pressure causes a vari ati on i n the shear rate duri ng testi ng whi ch i s
the mai n reason gravi ty operated uni ts are unsui tabl e for non-
Newtoni an fl ui ds. Wi th Newtoni an fl ui ds the starti ng poi nt for anal ysi s
i s the Poi seui l l e-Hagen equati on (Eq. [2.28]) wri tten i n terms of the
vol umetri c average vel oci ty, :
[2.91]
P
ex
11

22
P
ex
+
w
dP
ex
d
w
_
,
P
ex
1 +
d(lnP
ex
)
d(ln
w
)
_
w
u Q/R
2
u
(P)R
2
8L
where and are the radi us and l ength of the capi l l ary. Pressure drop
over the capi l l ary i s generated by a hei ght ( ) of l i qui d:
[2.92]
may be the average or maxi mum fl ui d hei ght, but shoul d be the same
for al l measurements. Substi tuti on of Eq. [2.92] i nto Eq. [2.91] yi el ds
[2.93]
whi ch can al so be wri tten as
[2.94]
The effl ux ti me, defi ned as the ti me to di scharge a fi xed vol ume of
l i qui d ( ) from the capi l l ary bul b (Fi g. 2.1), i s
[2.95]
so the average vel oci ty may be expressed i n terms of the di scharge ti me:
[2.96]
Substi tuti on of Eq. [2.96] i nto Eq. [2.94] yi el ds
[2.97]
whi ch can be expressed i n terms of a constant as
[2.98]
where , the gl ass capi l l ary vi scometer constant, i s defi ned as
[2.99]
The val ue of may be computed di rectl y from Eq. [2.99] i f an accurate
physi cal descri pti on of the vi scometer i s provi ded. One probl em wi th
thi s method i s that smal l vari ati ons i n geometry, parti cul arl y vari ati ons
i n the radi us due towi de manufacturi ng tol erances or fi l m bui l d up after
repeated use, may si gni fi cantl y i nfl uence the numeri cal val ue of .
Hence, i t i s not surpri si ng that the most common way to use Eq. [2.98]
i s to determi ne the properti es of an unknown fl ui d from the known
properti es of a reference fl ui d. Thi s experi mental method may be
i mpl emented by noti ng that i s the same for both fl ui ds, so
R L
h
P gh
h
u
ghR
2
8L

ghR
2
8Lu
V
t
V
u( R
2
)
u
V
R
2
t

gh R
4
t
8LV
kt
k
k
gh R
4
8LV
k
k
k
2.9 Glass Capillary (U-Tube) Viscometers 127
[2.100]
or
[2.101]
where subscri pts 1 and 2 refer to the unknown and known fl ui ds,
respecti vel y. A common reference materi al i s water but many com-
merci al l y produced standards, mai nl y si l i cone oi l s, are al so avai l abl e.
The above method of determi ni ng vi scosi ty i s usual l y wi thi n acceptabl e
l i mi ts of accuracy but does not account for the vari ous experi mental
errors di scussed bel ow.
A general worki ng equati on whi ch i ncorporates errors associ ated
wi th smal l vari ati ons i n capi l l ary radi us, end effects, and ki neti c energy
may be wri tten i n terms of the ki nemati c vi scosi ty ( ) as (Kawata
et al ., 1991)
[2.102]
where (wi th uni ts of m
2
/s
2
) and (wi th uni ts of m
2
) are constants for
a parti cul ar capi l l ary. When the ki nemati c vi scosi ti es ( ) and fl ow
ti mes ( )of twostandard l i qui ds are known, the i nstrument constants
may be cal cul ated:
[2.103]
and
[2.104]
I f the vi scosi ty range of a gl ass capi l l ary vi scometer i s i ncreased by
operati ng the i nstrument wi th appl i ed external pressure, addi ti onal
correcti on factors are needed (Kawata et al ., 1991; Van Wazer et al .,
1963).
1
t
1
2
t
2
2
_
t
1
t
2
_
,
/
c
1
t
c
2
t
c
1
c
2
1
,
2
t
1
, t
2
c
1

1
t
1

2
t
2
(t
1
)
2
(t
2
)
2
c
2

(
1
t
1

2
t
2
)t
1
t
2
(t
1
)
2
(t
2
)
2
2.10. Pipeline Design Calculations
The purpose of thi s secti on i s to provi de the practi cal i nformati on
necessary to predi ct pressure drop and power requi rements for homo-
geneous, non-ti me-dependent materi al s i n fl ui d handl i ng systems.
Rheol ogi cal properti es have a strong i nfl uence on the cal cul ati ons and
thi s i nformati on i s needed tosel ect proper pumps and rel ated equi pment
when desi gni ng l arge scal e tube (pi pe) vi scometers or commerci al fl ui d
handl i ng systems (Steffe and Morgan, 1986; Steffe and Garci a, 1987).
Al though rheol ogi cal properti es can onl y be eval uated from data taken
i n the l ami nar fl ow regi me, the case of turbul ent fl ow i s al so presented
to provi de a thorough anal ysi s of pi pel i ne desi gn probl ems commonl y
encountered by food process engi neers.
Mechanical EnergyBalance. The mechani cal energy bal ance for an
i ncompressi bl e fl ui d i n a pi pe may be wri tten as
[2.105]
where , the summati on of al l fri cti on l osses, i s
[2.106]
and subscri pts 1 and 2 refer to two speci fi c l ocati ons i n the system. The
fri cti on l osses i ncl ude those from pi pes of di fferent di ameters and a
contri buti on from each i ndi vi dual val ve, fi tti ng, and si mi l ar parts.
i s the work output per uni t mass and the power requi rement of the
system i s found by cal cul ati ng the product of and the mass fl ow rate.
A negati ve val ue of i ndi cates that work i s bei ng put i nto the system
whi ch i s the normal functi on of a pump.
Rheol ogi cal properti es are requi red to eval uate the mechani cal
energy bal ance equati on. Al though there are many mathemati cal
model s avai l abl e to descri be fl ow behavi or (Tabl e 1.3), few can be con-
si dered practi cal for maki ng pressure drop cal cul ati ons i nvol vi ng pi pe
fl ow. Most pumpi ng probl ems i nvol vi ng fl ui d foods can be sol ved usi ng
the Newtoni an, power l aw or Bi ngham pl asti c model s. Over si mpl i fi -
cati on, however, can cause si gni fi cant cal cul ati on errors (Steffe, 1984).
FanningFrictionFactor. The Fanni ng fri cti on factor ( ) i s defi ned,
from consi derati ons i n di mensi onal anal ysi s, as the rati o of the wal l
shear stress i n a pi pe to the ki neti c energy per uni t vol ume:
( u
2
)
2
( u
1
)
2
,
+ g(z
2
z
1
) +
P
2
P
1
+ F + W 0
F
F
2f ( u )
2
L
D
+
k
f
( u )
2
2
W
W
W
f
2.10 Pipeline Design Calculations 129
[2.107]
Substi tuti ng the defi ni ti on of the shear stress at the wal l of a pi pe (Eq.
[2.2]) i nto Eq. [2.107] gi ves
[2.108]
where . Hence, the energy l oss per uni t mass (needed i n the
mechani cal energy bal ance) may be expressed i n terms of :
[2.109]
Some engi neers cal cul ate the fri cti on l osses wi th the Darcy fri cti on
factor whi ch i s equal to four ti mes the Fanni ng fri cti on factor. Pressure
drop cal cul ati ons may be adjusted for thi s di fference. Fi nal resul ts are
the same usi ng ei ther fri cti on factor. Cal cul ati ons i n thi s text deal
excl usi vel y wi th the Fanni ng fri cti on factor.
I n l ami nar fl ow, val ues can be determi ned from the equati ons
descri bi ng the rel ati onshi p between pressure drop and fl ow rate for a
parti cul ar fl ui d. Consi der, for exampl e, a Newtoni an fl ui d ( ). Usi ng
Eq. [2.28], the vol umetri c average vel oci ty for thi s materi al , i n l ami nar
tube fl ow, may be expressed as:
[2.110]
Si mpl i fi cati on gi ves an expressi on for the pressure drop per uni t l ength:
[2.111]
Substi tuti ng Eq. [2.111] i nto Eq. [2.108] yi el ds the fri cti on factor:
[2.112]
whi ch i s a common equati on appropri ate for predi cti ng fri cti on factors
for Newtoni an fl ui ds when . Usi ng the same approach, l ami nar
fl ow fri cti on factors for power l aw and Bi ngham pl asti c fl ui ds may be
cal cul ated, respecti vel y, from the fol l owi ng equati ons:
[2.113]
f
2
w
( u )
2
f
(P)R
L( u )
2

(P)D
2L( u )
2
P P
2
P
1
f
P

f 2L( u )
2
D
f

u
Q
R
2

1
R
2
(P)R
4
8L
_
,

(P)D
2
32L
P
L

32u
D
2
f

P
L
1
1
]
D
2( u )
2
_
32( u )
D
2
1
1
]
D
2( u )
2
_
,

16
N
Re
N
Re
< 2100
f
16
N
Re, PL
and
[2.114]
The above equati ons are appropri ate when the l ami nar fl ow cri teri a,
presented i n Secti on 2.4, are sati sfi ed. Eq. [2.114], pl otted i n Fi g. 2.15,
i s an approxi mati on for the Fanni ng fri cti on factor based on the
assumpti on that (Heywood, 1991a). Thi s fi gure i l l ustrates
how the fri cti on factor decreases wi th l arger val ues of the Bi ngham
Reynol ds number, and i ncreases wi th l arger val ues of the Hedstrom
number. A l arger versi on of Fi g. 2.15, more conveni ent for probl em
sol vi ng, i s gi ven i n Appendi x [6.17].
Figure 2.15. Fanning friction factors (from Eq. [2.114]) for Bingham plastic flu-
ids in laminar flow at different values of the Hedstrom Number.
I n turbul ent fl ow, fri cti on factors may be determi ned from empi ri cal
equati ons (Tabl e 2.3) formul ated from experi mental data (Grovi er and
Azi z, 1972). The equati ons are onl y appl i cabl e to smooth pi pes whi ch
i ncl ude sani tary pi pi ng systems for food. I t may be very di ffi cul t to
accuratel y predi ct transi ti on from l ami nar to turbul ent fl ow i n actual
f
16(6N
Re, B
+ N
He
)
6N
Re, B
2
(
o
/
w
)
4
1
100 200 300 500 1,000 2,000 3,000 5,000 10,000
0.001
0.002
0.005
0.01
0.02
0.05
0.1
0.2
0.5
1
F
a
n
n
i
n
g

F
r
i
c
t
i
o
n

F
a
c
t
o
r
100 1,000 5,000 10,000 25,000 50,000 100,000 200,000
N
He
N
Re,B
Table 2.3. Fanning Friction Factor Equations for Turbulent Flowin Smooth Tubes
Fluid Fanning Friction Factor
Newtonian
where:
Power Law
where:
Bingham Plastic
where:
and
processi ng systems and the equati ons gi ven here are onl y i ntended for
use i n esti mati ng the power requi rements for pumpi ng. Curves for
power l aw fl ui ds i n turbul ent fl ow are pl otted i n Fi g. 2.16 (a l arger
versi on of the same pl ot i s gi ven i n Appendi x [6.18]). Newtoni an fl ui ds
are represented by the curve wi th = 1.0.
Kinetic Energy Evaluation. The ki neti c energy term i n the
mechani cal energy bal ance can be eval uated i f the ki neti c energy cor-
recti on factor ( ) i s known. I n turbul ent fl ow of any fl ui d, .
Expressi ons to compute val ues for vari ous fl ui ds i n l ami nar fl ow are
summari zed i n Tabl e 2.4. These equati ons may be gi ven i n terms of ,
the fl ow behavi or i ndex, or whi ch i s defi ned as the rati o of the yi el d
stress ( ) to the shear stress at the wal l ( ). Equati ons provi ded for
the Bi ngham pl asti c and Herschel -Bul kl ey cases are approxi mati ons.
1
f
4.0log
10
(N
Re
f ) 0.4

N
Re

Du
1
f

4
n
0.75
_
,
log
10
(N
Re, PL
)f
(1 ( n/2))1
]

0.4
n
1.2
_
,
K(
)
n
N
Re, PL

D
n
( u )
2 n
8
n 1
K
_
4n
3n + 1
_
,
n
1
f
4.53 log
10
(1 c) + 4.53 log
10
((N
Re, B
)f ) 2.3

pl
+
o
N
Re, B

Du
pl
c

o
2
o
f ( u )
2
n
2
n
c
o

w
Figure 2.16. Fanning friction factors (equation given in Table 2.3) for power law
fluids in turbulent flow at different values of the flow behavior index.
An exact, but cumbersome, mathemati cal equati on for the ki neti c
energy correcti on factor of a Herschel -Bul kl ey fl ui d has been publ i shed
by Osori o and Steffe (1984). Val ues of , determi ned from thi s equati on,
are pl otted i n Fi g. 2.17. Thi s fi gure reveal s some i nteresti ng features
of the ki neti c energy correcti on factor: val ues go to 2 as the yi el d stress
approaches the wal l shear stress for al l val ues of ; val ues i ncrease
wi th decreasi ng val ues of . Overal l , the numeri cal val ue of ranges
from 0.74 to 2 for Herschel -Bul kl ey fl ui ds (Osori o and Steffe, 1984). The
mi ni mum val ue of occurs at as approaches i nfi ni ty. KE
di fferences are usual l y smal l and often i gnored i n eval uati ng power
requi rements when sel ecti ng pumps.
2,000 4,000 6,000 8,000 10,000
0.002
0.004
0.006
0.008
0.01
0.012
0.014
F
a
n
n
i
n
g

F
r
i
c
t
i
o
n

F
a
c
t
o
r
N
Re,PL
n = 1.2
n = 1.0
n = 0.8
n = 0.6
n = 0.4
n = 0.3
n = 0.2
Power Law Fluids in Turbulent Flow
n
n
0.74 c 0 n
Table 2.4. Kinetic Energy Correction Factors for Laminar Flow in Tubes
Fluid , dimensionless
Newtonian 1.0
Power Law
+
Bingham Plastic
+
Herschel-Bulkley
for
and
for
+
Solution for the Bingham plastic material is within 2.5% of the true solution
(Metzner, 1956). Errors in using the Herschel-Bulkley solution are less than 3% for
0.1 1.0 but as high as 14.2% for 0.0 0.1 (Briggs and Steffe, 1995).
FrictionLosses: Valves, Fittings, andSimilar Parts. Fri cti on l oss
coeffi ci ents ( )must be determi ned from experi mental data. I n general ,
publ i shed val ues are for the turbul ent fl ow of water taken from Crane
(1982). An adequate summary of these numbers may be found i n
Saki adi s (1984). Lami nar fl ow data are much more l i mi ted. Some are
avai l abl e for vari ous fl ui ds: Newtoni an (Ki ttredge and Rowl ey, 1957),
shear-thi nni ng (Banerjee et al ., 1994; Lewi cki and Ski erkowski , 1988;
Steffe et al ., 1984) and shear-thi ckeni ng (Gri skey and Green, 1971).
Overal l , the quanti ty of engi neeri ng data requi red to predi ct pressure
l osses i n val ves and fi tti ngs for fl ui ds, parti cul arl y non-Newtoni an
fl ui ds, i n l ami nar fl ow i s i nsuffi ci ent.
Gi ven thi s si tuati on, a "rul e of thumb" esti mati on procedure i s
needed. Fi rst some general observati ons shoul d be made: a) The
behavi or of val ues for Newtoni an and non-Newtoni an fl ui ds i s si mi l ar
(Metzner, 1961; Skel l and, 1967), b) val ues decrease wi th i ncreasi ng
pi pe di ameter (Crane, 1982) -- they may drop as much as 30% i n goi ng
from 3/4 to 4 i nch (1.9 to 10.2 cm) pi pe, c) val ues sharpl y i ncrease wi th

2(2n + 1)(5n + 3)
3(3n + 1)
2
K(
)
n

2
2 c

pl
+
o
exp(0.168 c 1.062 n c 0.954 n
.5
0.115 c
.5
+ 0.831)
0.06 n 0.38
K(
)
n
+
o
exp(0.849 c 0.296n c 0.600 n
.5
0.602 c
.5
+ 0.733)
0.38 < n 1.60
c c
k
f
k
f
k
f
k
f
Figure 2.17. Kinetic energy correction factors for the laminar flow of Herschel-
Bulkley fluids (from Osorio and Steffe, 1984).
decreasi ng Reynol ds numbers (Cheng, 1970; Ki ttredge and Rowl ey,
1957; Lewi cki and Ski erkowski , 1988; Steffe et al ., 1984) i n the l ami nar
fl ow regi me but are constant i n the turbul ent fl ow regi me (Saki adi s,
1984) and show l i ttl e change above (Ki ttredge and Rowl ey,
1957), d) Entrance pressure l osses for power l aw fl ui ds i n l ami nar fl ow
decrease wi th smal l er val ues of the fl ow behavi or i ndex (Col l i ns and
Schowal ter, 1963), e) Entrance l osses for Bi ngham pl asti c fl ui ds
decrease wi th i ncreasi ng val ues of the yi el d stress when the wal l shear
stress ( )i s constant (Mi chi yosi et al ., 1966), f) Resi stance to fl ow
of non-Newtoni an fl ui ds i n l ami nar fl ow, through si mi l ar val ves, can be
expected tobe up to133 percent hi gher than that observed for Newtoni an
fl ui ds (Ury, 1966).
Fri cti on l oss coeffi ci ents for many val ves and fi tti ngs are summa-
ri zed i n Tabl es 2.5 and 2.6. val ues for the sudden contracti on, or
expansi on, of a Newtoni an fl ui d i n turbul ent fl ow, may be found i n Crane
(1982). The l oss coeffi ci ent for a sudden contracti on i s cal cul ated i n
terms of the smal l and l arge pi pe di ameters:
0 0.2 0.4 0.6 0.8 1
0.6
0.8
1
1.2
1.4
1.6
1.8
2
c = Yield Stress / Shear Stress at the Wall
n=0.1 n=0.3 n=0.5 n=0.7 n=0.9 n=2.0
Flow Behavior Index
N
Re
500
PR/(2L)
k
f
Table 2.5. Friction Loss Coefficients ( Values) for the Laminar Flowof Newtonian
Fluids throughValves and Fittings (fromSakiadis (1984) withOriginal Data from
Kittredge and Rowley, 1957)
Type of Fitting or Valve =1000 500 100
90-deg. elbow, short radius 0.9 1.0 7.5
Tee, standard, along run 0.4 0.5 2.5
Branch to line 1.5 1.8 4.9
Gate valve 1.2 1.7 9.9
Globe valve, composition disk 11 12 20
Plug 12 14 19
Angle valve 8 8.5 11
Check valve, swing 4 4.5 17
[2.115]
Losses for a sudden enl argement, or an exi t, are determi ned as
[2.116]
The l argest vel oci ty whi ch i s the mean vel oci ty i n the smal l est di ameter
pi pe, shoul d be used for both contracti ons and expansi ons i n cal cul ati ng
the fri cti on l oss term ( ) found i n Eq. [2.106].
After eval uati ng the avai l abl e data for fri cti on l oss coeffi ci ents i n
l ami nar and turbul ent fl ow, the fol l owi ng "rul e-of-thumb" gui del i nes,
conservati ve for shear-thi nni ng fl ui ds, are proposed for esti mati ng
val ues:
1. For Newtoni an fl ui ds i n l ami nar or turbul ent fl ow use the data of
Ki ttredge and Rowl ey (1957) or Saki adi s (1984), respecti vel y
(Tabl es 2.5 and 2.6).
2. For non-Newtoni an fl ui ds above a Reynol ds number ( )
of 500, use data for Newtoni an fl ui ds i n turbul ent fl ow (Tabl e 2.6).
3. For non-Newtoni an fl ui ds i n the Reynol ds number range of
use the fol l owi ng equati on:
k
f
N
Re
k
f
.5
D
small
D
large
_
,
2
_
,
k
f

D
small
D
large
_
,
2
_
,
2
k
f
u
2
/2
k
f
N
Re, PL
or N
Re, B
20 N 500
Table 2.6. Friction Loss Coefficients for the Turbulent Flow of Newtonian Fluids
through Valves and Fittings (from Sakiadis, 1984)
Type of Fitting of Valve
45-deg. elbow, standard 0.35
45-deg. elbow, long radius 0.2
90-deg. elbow, standard 0.75
Long radius 0.45
Square or miter 1.3
180-deg. bend, close return 1.5
Tee, standard, along run, branch blanked off 0.4
Used as elbow, entering run 1.0
Used as elbow, entering branch 1.0
Branching flow 1.0
Coupling 0.04
Union 0.04
Gate valve, open 0.17
3/4 open 0.9
1/2 open 4.5
1/4 open 24.0
Diaphragm valve, open 2.3
3/4 open 2.6
1/2 open 4.3
1/4 open 21.0
Globe valve, bevel seat, open 6.0
1/2 open 9.5
Composition seat, open 6.0
1/2 open 8.5
Plug disk, open 9.0
3/4 open 13.0
1/2 open 36.0
1/4 open 112.0
Angle valve, open 2.0
Y or blowoff valve, open 3.0
Plug cock =0 (fully open) 0.0
=5 0.05
=10 0.29
=20 1.56
=40 17.3
=60 206.0
Butterfly valve =0 (fully open) 0.0
=5 0.24
=10 0.52
=20 1.54
=40 10.8
=60 118.0
Check valve, swing 2.0
Disk 10.0
Ball 70.0
Foot valve 15.0
Water meter, disk 7.0
Piston 15.0
Rotary (star-shaped disk) 10.0
Turbine-wheel 6.0
k
f

[2.117]
where i s dependi ng on the type of fl ui d i n questi on.
The constant, , i s found for a parti cul ar val ve or fi tti ng (or rel ated
i tems l i ke contracti ons and expansi ons) by mul ti pl yi ng the tur-
bul ent fl ow fri cti on l oss coeffi ci ent by 500:
[2.118]
Val ues of for many standard i tems may be cal cul ated from the
val ues provi ded i n Tabl e 2.6. The val ue of 20 was arbi trari l y
set as a l ower l i mi t of i n Eq. [2.117] because very l ow val ues of
the Reynol ds number i n that equati on wi l l generate unreasonabl y
hi gh val ues of the fri cti on l oss coeffi ci ent. Val ues of wi l l
cover most practi cal appl i cati ons for fl ui d foods. Eq. [2.117] and
[2.118] may al so be used for Newtoni an fl ui ds when i s used
for .
The above gui del i nes are offered wi th cauti on and shoul d onl y be used
i n the absence of actual experi mental data. Many factors, such as a
hi gh extensi onal vi scosi ty, may si gni fi cantl y i nfl uence val ues.
GeneralizedPressureDropCalculation. Metzner (1956) presents
a general i zed approach to rel ate fl ow rate and pressure drop for ti me-
i ndependent fl ui ds i n l ami nar fl ow. The governi ng equati on i s
[2.119]
where
[2.120]
and are easi l y determi ned from a pl ot of the experi mental data.
There i s a strong si mi l ari ty wi th the above equati on and those descri bi ng
the fl ow of power l aw fl ui ds i n pi pes. I n fact, for a power l aw fl ui d,
[2.121]
k
f

A
N
N N
Re, PL
or N
Re, B
A
A (k
f
)
turbulent
(500)
A
k
f
N
N 20
N
Re
N
k
f
(P)R
2L
K
4Q
R
3
_
,
n
n
d ln((P)R/(2L))
d ln(4Q/(R
3
))

d(ln
w
)
d(ln)
K n
n n and K K
3n + 1
4n
_
,
n
Wi th the general sol uti on, may vary wi th the shear stress at the wal l
and must be eval uated at the parti cul ar val ue of i n questi on.
Eq. [2.119] and [2.120] have practi cal val ue when consi deri ng di rect
scal e-up from data taken wi th a smal l di ameter tube or for cases where
a wel l defi ned equati on (power l aw, Bi ngham pl asti c or Herschel -
Bul kl ey) i s not appl i cabl e. A si mi l ar method i s avai l abl e for scal e-up
probl ems i nvol vi ng the turbul ent fl ow of ti me-i ndependent fl ui ds (Lord
et al ., 1967).
Sl i p and ti me-dependent behavi or may be a probl em i n predi cti ng
pressure l oss i n pi pes. One sol uti on i s to i ncorporate these effects i nto
the consi stency coeffi ci ent. Houska et al . (1988) gi ve an exampl e of thi s
techni que for the transport of mi nced meat i n pi pes where i ncorpo-
rates the property changes due to the agi ng of the meat and wal l sl i p
as a functi on of pi pe di ameter. An exerci se i n pi pel i ne desi gn i s
presented i n Exampl e Probl em 2.12.6.
2.11. Velocity Profiles In Turbulent Flow
Vel oci ty profi l es, cri ti cal i n thermal processi ng systems (parti cul arl y
i n hol d tubes), are strongl y i nfl uenced by rheol ogi cal properti es.
Accuratel y predi cti ng vel oci ty profi l es for fl ui ds i n turbul ent fl ow i s
di ffi cul t. Rel ati onshi ps for Newtoni an fl ui ds are rel i abl e. Those for
power l aw fl ui ds are avai l abl e but they have not recei ved adequate
experi mental veri fi cati on for fl ui d foods. Approxi mate mean, di vi ded
by maxi mum, vel oci ti es are summari zed for some condi ti ons i n Tabl e
2.7. A detai l ed di scussi on of l ami nar fl ow profi l es i s presented i n Sec.
2.3.
Newtonian Fluids. Semi -theoreti cal predi cti on equati ons for the
vel oci ty profi l e of Newtoni an fl ui ds i n turbul ent fl ow are wel l establ i shed
and presented i n terms of three di sti nct regi ons of the pi pe (Brodkey
and Hershey, 1988):
for the vi scous subl ayer (cal l ed l ami nar subl ayer i n some l i terature)
[2.122]
for the transi ti on zone where turbul ent fl uctuati ons are generated
[2.123]
and for the turbul ent core
n
w
K
u
+
y
+
y
+
5
u
+
3.05 + (11.513) log
10
( y
+
) 5 < y
+
< 30
2.11 Velocity Profiles In Turbulent Flow 139
Table 2.7. Approximate Average and Maximum Velocities for Newtonian and
Power Law Fluids in Tube Flow
Newtonian Fluids
< 2100 4000 10
4
10
5
10
6
10
7
10
8
0.5 0.790 0.811 0.849 0.875 0.893 0.907
Power Law Fluids: Laminar Flow
n 0.0 0.3 0.4 0.5 1.0 2.0
1.00 0.68 0.64 0.60 0.50 0.43 0.33
Power Law Fluids: Turbulent Flow
0.2 0.2 0.2 0.5 0.5 0.5
10
4
10
5
10
6
10
4
10
5
10
6
0.92 0.94 0.95 0.86 0.89 0.92
[2.124]
where
[2.125]
and
[2.126]
and
[2.127]
N
Re
u/u
max
u/u
max
(n + 1)/(3n + 1)
u/u
max
n
N
Re, PL
u/u
max
u
+
5.5 + (5.756) log
10
( y
+
) 30 y
+
u
+
u
u
*
y
+
yu
*
u
*
f
2
The di stance from the pi pe wal l , , i s defi ned as
[2.128]
Note that the ori gi n of the coordi nate system i s (by conventi on) l ocated
at the wal l , . Therefore, the vel oci ty i s zero at where and
a maxi mum at the center of the pi pe where and . The combi ned
vel oci ty equati ons consti tute the uni versal vel oci ty profi l e.
Once the maxi mum vel oci ty has been determi ned, the power l aw
equati on may be used to approxi mate at other l ocati ons:
[2.129]
Eq. [2.129] does a reasonabl e job predi cti ng vel oci ty profi l es i n spi te of
the fact that i t i s not dependent on the Reynol ds number. Grovi er and
Azi z (1972) note that thi s equati on i s most appropri ate for
and . Al so the exponent may vary from at
to at . Maxi mum vel oci ty, of a turbul ent
Newtoni an fl ui d i n tube fl ow, i s cal cul ated i n Exampl e Probl em 2.12.7.
Power LawFluids. Dodge and Metzner (1959) deri ved equati ons to
descri be the vel oci ty profi l e of power l aw fl ui ds i n tube fl ow. Smal l errors
were corrected by Skel l and (1967) and the fi nal equati ons presented as
[2.130]
for the l ami nar subl ayer and
[2.131]
for the turbul ent core, where i ncorporates the fl ow behavi or i ndex
and the consi stency coeffi ci ent needed for the consi derati on of power
l aw fl ui ds:
[2.132]
Constants were obtai ned from fri cti on factor measurements so the
thi ckness of the l ami nar subl ayer was not obtai ned. These equati ons
are appl i ed i n Exampl e Probl em 2.12.8.
An al ternati ve equati on for predi cti ng vel oci ty i n the turbul ent core
for power l aw fl ui ds was presented by Cl app (1961):
y
y R r
r R r R y 0
r 0 y R
1/7
u/u
max
u
u
max
y
R
_
,
1/ 7
R r
R
_
,
1/ 7
0.1 < y/R < 1.0
3000 < N
Re
< 100, 000 1/6
N
Re
4, 000 1/10 N
Re
3, 200, 000
u
+
( y
+
)
1/n
u
+
5.66
n
.75
log
10
( y
+
)
0.566
n
1.2
+
3.475
n
.75
1.960 + 0.815n 1.628n log

10
3 +
1
n
_
,
1
1
]
y
+
y
+
y
n
(u
*
)
2 n
/K
2.12.1 Conservation of Momentum Equations 141
[2.133]
Thi s equati on correl ated wel l wi th experi mental data i n whi ch
and .
2.12.1. Conservation of Momentum Equations
Show that an equati on descri bi ng the shear stress for a fl ui d i n tube
fl ow may be determi ned from the conservati on of momentum pri nci pl e.
Compare the resul t to Eq. [2.2] whi ch was deri ved di rectl y from a force
bal ance. Thi s exampl e probl em i s i ncl uded to i l l ustrate an al ternati ve
method of obtai ni ng rheol ogi cal rel ati onshi ps.
Fi rst a number of assumpti ons must be stated: fl ow i s for an
i ncompressi bl e fl ui d under steady, l ami nar, i sothermal condi ti ons;
there are no entrance and exi t effects; and fl ow i s paral l el to the pi pe
wal l , i .e., there i s no radi al or ci rcumferenti al fl ow. Then, usi ng cyl i n-
dri cal coordi nates (Fi g. 2.18),
[2.134]
The conti nui ty (conservati on of mass) equati on may be wri tten as
[2.135]
Under steady condi ti ons wi th an i ncompressi bl e fl ui d ( ) and,
wi th the assumpti on that radi al and ci rcumferenti al vel oci ti es are zero,
thi s equati on reduces to
[2.136]
i ndi cati ng that conservati on of mass i s sati sfi ed and vel oci ty i s constant
al ong the (axi al ) di recti on. Consi deri ng the above assumpti ons the
momentum equati ons (Fi g. 2.18) col l apse to the fol l owi ng:
[2.137]
[2.138]
[2.139]
u
+
2.78
n
2.303log
10
( y
+
) +
3.80
n
0.698 < n < 0.813 5, 480 < N
Re, PL
< 42, 800
u
r
u
0 u
z
u
z
(r)
t
+
1
r
r
(ru
r
) +
1
r
(u
) +

z
(u
z
) 0
/t 0
du
z
dz
0
z
r component: 0
P
r
component: 0
1
r
P
z component: 0
P
z
+
1
r
r
(r
rz
)
Figure 2.18. Momentum equations in cylindrical coordinates. [Note: P is an
inclusive pressure term which ignores small variation in pressure due to varia-
tions in height ( )].
The and components woul d i ntegrate to a constant i ndi cati ng that
does not vary i n those di recti ons. I t fol l ows that i s a functi on of
onl y:
[2.140]
Then, we can repl ace wi th , and wi th
to wri te the component of the momentum equati on, Eq. [2.139],
as
[2.141]
or
[2.142]
where i s repl aced by the constant term, . Thi s i s al l owed
because the l eft hand si de of Eq. [2.141] i s a functi on of onl y, and the
z
r
r
u
r
t
+ u
r
u
r
r
+
u
r
u
r
2
r
+ u
z
u
r
z
_
,

P
r
+
1
r
r
(r
rr
) +
1
r
r
+
rz
z
+ g
r
t
+ u
r
u
r
+
u
r
u
+
u
r
u
r
+ u
z
u
z
_
,

1
r
P
+
1
r
2
r
(r
2
r
) +
1
r
z
z
+ g
u
z
t
+ u
r
u
z
r
+
u
r
u
z
+ u
z
u
z
z
_
,

P
z
+
1
r
r
(r
rz
) +
1
r
zz
z
+ g
z
gh
r
P P z
P P(z)
(1/r)(r
rz
)/r (1/r)d(r
rz
)/dr P/z
dP/dz z
1
r
d
dr
(r
rz
)
dP
dz
1
r
d
dr
(r
rz
)
( P)
L
dP/dz (P)/L
r
2.12.2 Capillary Viscometry - Soy Dough 143
ri ght hand si de of the equati on i s a functi on of onl y; hence, both si des
are equal to a constant (the ori gi n of the negati ve si gn wi l l be expl ai ned
shortl y). I ntegrati ng Eq. [2.142] yi el ds
[2.143]
gi vi ng
[2.144]
must be zero for the shear stress to be fi ni te at the centerl i ne of the
pi pe, therefore,
[2.145]
whi ch, wri tten i n one di mensi onal symbol i sm, i s l i ke Eq. [2.2] found
from a si mpl e force bal ance:
[2.146]
Eq. [2.146] and [2.2] are equi val ent, but appear di fferent, due to the
presence of a negati ve si gn i n Eq. [2.146]. Thi s refl ects the si gn
conventi on adopted i n devel opi ng the equati ons i l l ustrated i n Fi g. 2.18
and uti l i zed i n vari ous textbooks (Bi rd et al ., 1960; Denn, 1980; Brodkey
and Hershey, 1988). I n those equati ons i s negati ve so the negati ve
si gn i s requi red tomake the shear stress posi ti ve. I n thi s book, however,
was assumed to be a posi ti ve number i n deri vi ng our i ni ti al force
bal ance (Eq. [2.1]) maki ng a negati ve si gn unnecessary.
2.12.2. Capillary Viscometry - Soy Dough
Data for a defatted soy fl our dough from a capi l l ary vi scometer are
summari zed i n Tabl e 2.8. Determi ne the rheol ogi cal properti es of the
materi al .
The fi rst step i n the anal ysi s i s to correct the measured pressure l oss
for entrance effects. Pressure l osses are pl otted (Fi g. 2.19) at di fferent
val ues of for constant apparent wal l shear rates, .
Regressi on anal ysi s of these curves gi ves the requi red entrance l oss
correcti on val ues at =0. I n thi s parti cul ar probl em, entrance l osses
consti tute a very hi gh percentage, over 80% i n many tri al s, of the total
pressure drop across the capi l l ary (Tabl e 2.8).
z
d(r
rz
)

( P)
L
r
_
,
dr
rz

( P)
2L
r +
C
1
r
C
1
rz

( P)r
2L

( P)r
2L
P
P
4Q/(R
3
) L/D
L/D
Table 2.8. Capillary Viscometer (D=3.18mm) Data for Dough (34.7%) at Room
Temperature Made from Defatted Soy Flour Treated to Cause Protein Denatu-
ration (Data from Morgan, 1979)
(1/s) (1/s) (-) (MPa) (MPa) (MPa) (kPa) (1/s)
47.4 11.9 2 4.21 3.58 0.63 78.75 72.13
47.4 11.9 5 5.21 3.58 1.63 81.50 76.73
47.4 11.9 8 6.14 3.58 2.56 80.00 74.17
94.8 23.7 2 5.46 4.63 0.83 103.75 165.80
94.8 23.7 5 6.81 4.63 2.18 109.00 183.38
94.8 23.7 8 8.02 4.63 3.39 105.94 172.87
190.0 47.5 2 5.25 4.38 0.87 108.75 253.90
190.0 47.5 5 6.80 4.38 2.42 121.00 303.49
190.0 47.5 8 8.30 4.38 3.92 122.50 310.48
948.0 237.0 2 7.68 6.17 1.51 188.75 1457.45
948.0 237.0 5 10.12 6.17 3.95 197.50 1583.76
948.0 237.0 8 12.31 6.17 6.14 191.88 1500.96
*
Measured pressure drop;
**
Entrance loss pressure correction
***
Corrected pressure drop.
Figure 2.19. Measured pressure drop (including entrance loss) for capillary vis-
cometer data from defatted soy flour.
4Q/(R
3
) Q/(R
3
) (P)
m
*
(P)
en
**
(P)
***
L/D
w

w
1 2 3 4 5 6 7 8 9
2
4
6
8
10
12
14
L/D
P
r
e
s
s
u
r
e

D
r
o
p
,

M
P
a
(4 Q)/( R )
3
47.4 1/s 94.8 1/s 190.0 1/s 948.0 1/s
2.12.2 Capillary Viscometry - Soy Dough 145
Figure 2.20. Capillary data for soy dough corrected for pressure loss.
To determi ne the shear rate, the Rabi nowi tsch-Mooney equati on
(Eq. [2.20]), must be eval uated:
The rel ati onshi p between and was pl otted (Fi g. 2.20) and
determi ned (by curve fi tti ng) as a power functi on:
The deri vati ve of thi s equati on i s
whi ch can be i nserted i nto the Rabi nowi tsch-Mooney equati on to
determi ne the shear rate at the wal l :
60 80 100 120 140 160 180 200 220
0
50
100
150
200
250
Corrected Data for Soy Dough
Q
/
(

R

)
3
,

1
/
s
w
, kPa
w
f (
w
)

3Q
R
3
_
,
+ (
w
)
d(Q/(R
3
))
d
w
1
1
]
Q/ (R
3
)
w
Q
R
3
3.1(10
6
) (
w
)
3.44
d(Q/(R
3
))
d
w
1.06(10
5
) (
w
)
2.44
3Q
R
3
_
,
+
w
[1.06(10
5
) (
w
)
2.44
]
Resul ts of thi s computati on are summari zed i n Tabl e 2.8. After maki ng
these cal cul ati ons, the basi cel ements of the rheogram i l l ustrati ng shear
stress versus shear rate are avai l abl e for pl otti ng (Fi g. 2.21). I f the
materi al i s consi dered to be shear-thi nni ng over the shear rate range
consi dered, the fol l owi ng fl ui d parameters are cal cul ated: = 23.3 kPa
s
n
, = 0.29.
Figure 2.21. Rheogram for soy dough.
2.12.3. Tube Viscometry - 1.5% CMC
Data for a 1.5% aqueous sol uti on of sodi um carboxymethyl cel l ul ose
(CMC) were col l ected at room temperature usi ng a capi l l ary tube vi s-
cometer (Tabl e 2.9). Determi ne the rheol ogi cal properti es of thi s
materi al . Si nce the mi ni mum i s 248, entrance effects are assumed
to be i nsi gni fi cant. Assume CMC densi ty i s equal to 1003 kg/m
3
.
Regressi on anal ysi s of versus or l n( ) i n the form
of
K
n
0 500 1,000 1,500 2,000
60
80
100
120
140
160
180
200
220
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

k
P
a
Soy Dough
L/D
ln(4Q/(R
3
)) ln(
w
)
2.12.3 Tube Viscometry - 1.5% CMC 147
Table 2.9. Rheological Data for a 1.5% Aqueous Solution of Sodium Carboxy-
methylcellulose (CMC; =1003 kg/m
3
) Collected at RoomTemperature (Data from
Middleman, 1968)
(m) (m) (Pa) (m
3
/s) (Pa) (1/s) (1/s)
2.71E-03 0.944 1.379E+05 5.910E-07 98.97 302.47 409.50
2.71E-03 0.944 2.710E+05 2.950E-06 194.48 1509.78 2044.04
2.71E-03 0.944 4.103E+05 8.320E-06 294.43 4258.10 5764.89
2.71E-03 0.944 5.916E+05 2.100E-05 424.58 10747.61 14550.81
2.71E-03 0.674 1.034E+05 7.330E-07 103.96 375.14 507.89
2.71E-03 0.674 1.379E+05 1.260E-06 138.62 644.86 873.05
2.71E-03 0.674 2.724E+05 7.060E-06 273.77 3613.24 4891.84
2.71E-03 0.674 3.999E+05 1.760E-05 401.99 9007.52 12194.96
2.71E-03 0.674 5.482E+05 3.510E-05 551.00 17963.86 24320.64
1.82E-03 0.634 1.413E+05 1.920E-07 101.44 324.40 439.20
1.82E-03 0.634 1.931E+05 4.770E-07 138.55 805.94 1091.14
1.82E-03 0.634 2.586E+05 8.010E-07 185.56 1353.38 1832.29
1.82E-03 0.634 4.068E+05 2.580E-06 291.95 4359.19 5901.76
1.82E-03 0.634 5.482E+05 5.560E-06 393.39 9394.23 12718.51
1.82E-03 0.634 6.916E+05 9.340E-06 496.31 15780.95 21365.27
1.82E-03 0.452 6.206E+04 6.700E-08 62.47 113.20 156.26
1.82E-03 0.452 1.413E+05 3.500E-07 142.29 591.36 800.63
1.82E-03 0.452 2.137E+05 9.720E-07 215.16 1642.30 2223.45
1.82E-03 0.452 3.103E+05 2.720E-06 312.33 4595.74 6222.01
1.82E-03 0.452 4.103E+05 5.750E-06 412.98 9715.25 13153.14
1.82E-03 0.452 5.171E+05 9.840E-06 520.56 16625.76 22509.02
fi ts the data wel l (Fi g. 2.22) and yi el ds meani ng that
Note al so (Fi g. 2.22) that data for al l tube si zes overl ap veri fyi ng the
assumpti on of a negl i gi bl e entrance effect. Now, may be cal cul ated
usi ng the Rabi nowi tsch-Mooney equati on gi ven i n terms of Eq. [2.24]
and [2.25]:
4Q/(R
3
) D L P Q
w

w
ln(
w
) ln(Constant ) + n ln()
n 0.414
n
d(ln(
w
))
d(ln())
0.414
3n + 1
4n
_
3(.414) + 1
4(.414)
_
,
1.35
Figure 2.22. Shear stress versus apparent wall shear rate of
1.5% CMC at room temperature.
Figure 2.23. Rheogram for 1.5% CMC at room temperature.
100 300 1,000 3,000 10,000 30,000
50
100
200
300
500
1,000
Apparent Wall Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
1.5% CMC
( 4Q/(R
3
))
0 5,000 10,000 15,000 20,000 25,000 30,000
0
100
200
300
400
500
600
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
1.5% CMC
2.12.4 Casson Model: Flow Rate Equation 149
Resul ts are summari zed i n Tabl e 2.9 and pl otted i n Fi g. 2.23. The power
l aw model was used to represent fl ui d behavi or:
where = 8.14 Pa s
.414
and = 0.414. Note, wi th the power l aw model ,
, as expected.
shoul d be checked, tobe certai n fl ow i s l ami nar, usi ng Eq. [2.51]
or Eq. [2.53]. The "worst case,"at m and m
3
/s,
yi el ds m/s and . Wi th , ,
when cal cul ated usi ng Eq. [2.53]. Si nce 558 < 2581, one may concl ude
that fl ow i s l ami nar for al l fl ow rates under consi derati on.
I n thi s probl em i t was assumed the entrance effect was negl i gi bl e.
The actual entrance l ength can be esti mated from Eq. [2.67]:
Taki ng the experi mental val ues used i n checki ng for l ami nar fl ow yi el ds
Meani ng, for thi s parti cul ar case, the l ami nar fl ow vel oci ty (as descri bed
by Eq. [2.40]) was 98% ful l y devel oped i n the 0.271cm di ameter tube at
a di stance of approxi matel y 6.25cm (approxi matel y 23 di ameters) from
the entrance. Si nce the mi ni mum of thi s system i s 248, i t i s unl i kel y
the entrance l osses wi l l si gni fi cantl y i nfl uence resul ts.
2.12.4. Casson Model: Flow Rate Equation
Deri ve the rel ati onshi p between vol umetri c fl ow rate and pressure drop
for a Casson fl ui d i n l ami nar tube fl ow.
The starti ng poi nt i n the sol uti on i s Eq. [2.15] gi vi ng the general
fl ow rate rel ati onshi p i n tube fl ow:
The Casson equati on, gi ven i n Tabl e 1.3, i s
[2.147]
whi ch can easi l y be sol ved for the shear rate:
w
8.14(
w
)
.414
K n
n n
N
Re, PL
Q 3.510(10
5
) D 0.00271
u 6.09 N
Re, PL
558 n 0.414 (N
Re, PL
)
critical
2581
X
E
DN
Re, PL
(.125n + .175) (8
n 1
)
3n + 1
4n
_
,
n
X
E
.00271(558) (.125(.414) + .175) (8
.414 1
)
3(.414) + 1
4(.414)
1
1
]
.414
.0625 m
L/D
Q
R
3

1
(
w
)
3
w
()
2
f () d
.5
o
.5
+ K
1
(
)
.5
[2.148]
Eq. [2.148] may be substi tuted i nto the fl ow rate equati on: Recogni zi ng
that the presence of the yi el d stress makes the fl ow di sconti nuous gi ves
[2.149]
The fi rst i ntegral term i n Eq. [2.149] i s zerobecause i n the central
pl ug regi on of the tube where ; therefore,
[2.150]
Expansi on, i ntegrati on, and al gebrai c mani pul ati on yi el ds the fi nal
sol uti on:
[2.151]
Si nce , Eq. [2.151] provi des the rel ati onshi p between
vol umetri c fl ow rate and pressure drop for a Casson fl ui d.
2.12.5. Slit Viscometry - Corn Syrup
Data (Tabl e 2.10) for hi gh fructose corn syrup were col l ected at 28 C
usi ng a sl i t vi scometer. Determi ne the rheol ogi cal properti es of thi s
materi al . Pressure drops were measured wi th fl ush mounted trans-
ducers i n the regi on of the sl i t where fl ow was ful l y devel oped.
The apparent wal l shear rate ( ) and shear stress ( )
were cal cul ated (Tabl e 2.10) and pl otted (Fi g. 2.24) to fi nd from
Eq. [2.89]:
f ()
.5

o
.5
K
1
1
1
]
2
Q
w
3
K
1
2
R
3

2
f () d+
2
(
.5

o
.5
)
2
d
f () 0
0<<
o
Q
w
3
K
1
2
R
3

2
(
.5

o
.5
)
2
d
Q
R
3
K
1
2
w
4

4
o
.5
w
.5
7
+
o
3

o
4
w
4
_
1
84
_
,
_
w
PR/(2L)
6Q/(wh
2
) Ph/(2L)
n 0.97
n
d ln(
w
)
d ln(
w
)
0.97
2.12.5 Slit Viscometry - Corn Syrup 151
Table 2.10. Slit (Fig. 2.14: =0.3375 cm, = 8.89 cm, = 33.02 cm) Flow Data
of High Fructose Corn Syrup at 28 C
(m
3
/s) (1/s) (Pa/m) (kPa) (1/s)
6.16 E-6 44.3 2.28 3.63 44.8
8.16 E-6 54.6 3.01 4.78 55.2
9.65 E-6 64.6 3.55 5.63 65.3
10.0 E-6 73.4 4.02 6.39 74.2
12.2 E-6 81.6 4.39 6.98 82.4
13.8 E-6 92.6 4.86 7.71 93.6
15.1 E-6 101.4 5.30 8.41 102.4
16.2 E-6 108.6 5.58 8.86 109.8
Figure 2.24. Shear stress versus apparent wall shear rate for slit flow data of
high fructose corn syrup at 28 C.
A val ue of thi s cl ose to 1.0 i ndi cates the expected Newtoni an
behavi or. Usi ng Eq. [2.88], the shear rate was eval uated (Tabl e 2.10).
The rheogram was pl otted (Fi g. 2.25) and fi t wi th a l i near equati on:
h w L
6Q/(wh
2
) Q P/L
w

w
10 20 30 50 70 100
1,000
2,000
3,000
5,000
7,000
10,000
6Q/(wh ), 1/s
2
w
,

P
a
= .97
Corn Syrup
n
n
Showi ng the vi scosi ty of the sampl e i s 78.7 Pa s. The i ntercept, 374.6
Pa, i s a mathemati cal consequence of the stati sti cal procedure and
shoul d not be i nterpreted as an absol ute val ue of the yi el d stress.
Figure 2.25. Rheogram for high fructose corn syrup at 28 C.
2.12.6. Friction Losses in Pumping
Consi der the typi cal fl ow probl em i l l ustrated i n Fi g. 2.26. Assume the
pl ug di sk val ve i s open. The system has a 0.0348 m di ameter pi pe wi th
a vol umetri c fl ow rate of 0.00157 m
3
/s (1.97 kg/s) resul ti ng i n a vol u-
metri c average vel oci ty of 1.66 m/s. The fl ui d densi ty i s equal to 1250
kg/m
3
and the pressure drop across the strai ner i s 100 kPa. Determi ne
the fri cti on l osses i n the system, and cal cul ate the work i nput and
pressure drop across the pump for the fol l owi ng two cases i nvol vi ng
power l aw fl ui ds: Case 1) Assume = 5.2 Pa s
n
and = 0.45; Case 2)
Assume = 0.25 Pa s
n
and = 0.45.
Sol vi ng the mechani cal energy bal ance equati on, Eq. [2.105], for
work output yi el ds
w
374.6 + 78.7

w
40 50 60 70 80 90 100 110
3,000
4,000
5,000
6,000
7,000
8,000
9,000
10,000
Corn Syrup
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
K n
K n
2.12.6 Friction Losses in Pumping 153
Figure 2.26. Pumping system for a power law fluid.
Subscri pts 1 and 2 refer to the fl ui d l evel i n the i nput tank and the exi t
poi nt of the system, respecti vel y. The pressure at poi nts 1 and 2 i s equal
to one atmosphere, therefore, . As a worst case for pumpi ng,
assume a near empty tank ( ). Al so, assume a l arge di ameter
i nput tank maki ng . These consi derati ons si mpl i fy the above
equati on to
where ( ) represents the work i nput per uni t mass. The summati on
term, gi ven by Eq. [2.106], i ncl udes fri cti on l osses i n strai ght pi pe, fi t-
ti ngs, and the strai ner:
or
Pressure drop across the pump i s
1.0 m 2.0 m 1.0 m
3.0 m
3.0 m
0.5 m
Pump Strainer
Plug Disk Valve
W
( u
2
)
2
( u
1
)
2
1
+ g(z
2
z
1
) +
P
2
P
1
+ F
P
2
P
1
z
2
z
1
2.5
u
1
0
W g(z
2
z
1
) +
( u
2
)
2
2
+ F 9.81(2.5) +
(1.66)
2
2
+ F
W
F
2f ( u )
2
L
D
+
(k
f, entrance
) ( u )
2
2
+
(k
f, valve
) ( u )
2
2
+
3(k
f, elbow
) ( u )
2
2
+
100000
1250
F
2f ( u )
2
L
D
+ (k
f, entrance
+ k
f, valve
+ 3(k
f, elbow
))
( u )
2
2
+ 80.0
The power requi rement woul d be found by cal cul ati ng the product of
work i nput and mass fl ow rate.
SolutiontoCase1: =5.2Pas
n
and =0.45. Usi ng Eq. [2.52],
i s cal cul ated to be 323.9, a l ami nar fl ow val ue of . Fri cti on l oss
coeffi ci ents are determi ned from Eq. [2.115], Eq. [2.117], Eq. [2.118],
and Tabl e 2.6 yi el di ng:
and the fri cti on factor i s cal cul ated from Eq. [2.113]:
Usi ng the above val ues, the total fri cti on l oss i s
The ki neti c energy correcti on factor i s found from the l ami nar fl ow
equati on for power l aw fl ui ds (Tabl e 2.4):
Wi th thi s i nformati on, the cal cul ati ons for work i nput and pressure drop
can be compl eted:
and
(P)
p
(W)
K n N
Re, PL
N
Re, PL
k
f, entrance

(.5)500
323.9
0.77
k
f, valve

(9) (500)
323.9
13.89
k
f, elbow

(.45) (500)
323.9
.69
f
16
323.9
.0494
F
2(.0494) (1.66)
2
(10.5)
.0348
+ (0.77 + 13.89 + 3(.69))
(1.66)
2
2
+ 80.0 185.2 J/kg

2(2n + 1) (5n + 3)
3(3n + 1)
2

2(2(.45) + 1) (5(.45) + 3)
3(3(.45) + 1)
2
1.2
W 9.81(2.5) +
(1.66)
2
1.2
+ 185.2 212.0 J/kg
(P)
p
(212.0) (1250) 265 kPa
2.12.7 Turbulent Flow - Newtonian Fluid 155
Solution to Case 2: =0.25Pa s
n
and =0.45. Usi ng Eq. [2.52],
i s cal cul ated as 6,737. The cri ti cal val ue of , determi ned from
Eq. [2.51], i s
meani ng the fl ow i s turbul ent si nce 6,737 > 2,394. Fri cti on l oss coeffi -
ci ents may be determi ned from Eq. [2.115], and Tabl e 2.6: ;
; . The fri cti on factor i s found by i terati on of
(equati on from Tabl e 2.4)
yi el di ng = 0.0052. Then,
I n turbul ent fl ow, the ki neti c energy correcti on factor i s equal to2. Then,
the work i nput and pressure drop are cal cul ated as
and
2.12.7. Turbulent Flow - Newtonian Fluid
A common probl em faci ng food process engi neers i s to predi ct the
maxi mum vel oci ty found duri ng fl ow i n tubes. Determi ne, gi ven the
fol l owi ng data, the maxi mum vel oci ty i n the pi pe: Pa s;
m; m/s; kg/m
3
. Al so, cal cul ate the vel oci ty at
a poi nt hal fway between the center-l i ne and the wal l of the pi pe.
The Reynol ds number for a Newtoni an fl ui d i s cal cul ated as
whi ch i s suffi ci ent to concl ude that fl ow i s i n the turbul ent regi me. The
fri cti on factor equati on for Newtoni an fl ui ds i n turbul ent fl ow (see Tabl e
2.3) i s:
K n
N
Re, PL
N
Re, PL
(N
Re, PL
)
critical

6464(.45) (2 + .45)
(2 + .45)/(1 + .45)
(1 + 3(.45))
2
2, 394.
k
f, entrance
0.55
k
f, valve
9 k
f, elbow
0.45
1
4
(.45)
0.75
_
,
log
10
[(6736.6)f
(1 (.45/2))
]
0.4
(.45)
1.2
_
,
f
F
2(.0052) (1.66)
2
(10.5)
.0348
+ (.5 + 9 + 3(.45))
(1.66)
2
2
+ 80.0 103.6 J/kg
W 9.81(2.5) +
(1.66)
2
2
+ 103.6 129.5 J/kg
(P)
p
(129.5) (1250) 162 kPa
0.012
D 0.0348 u 1.66 1250
N
Re

Du

1250(.0348)1.66
0.012
6017.5
Sol vi ng for gi ves
The vel oci ty i s maxi mum at the center-l i ne where and . I t
must be cal cul ated from the fri cti on vel oci ty, Eq. [2.127]:
Cal cul ati ons proceed usi ng Eq. [2.126] and [2.124]:
and
The maxi mum vel oci ty may be found from the defi ni ti on (Eq. [2.125])
of the turbul ent vel oci ty, , as
Wi th thi s i nformati on, vel oci ti es at other l ocati ons can al sobe esti mated.
The 1/7 power l aw equati on (Eq. [2.129]) may be used to approxi mate
at the vel oci ty hal fway between the center-l i ne and the wal l
( ):
gi vi ng
2.12.8. Turbulent Flow - Power Law Fluid
Determi ne the maxi mum vel oci ty of a power l aw fl ui d i n a pi pe gi ven
the fol l owi ng i nformati on: Pa s
n
; m; m/s;
kg/m
3
; .
The Reynol ds number (Eq. [2.52]) i s
1
f
4.0log
10
(N
Re
f ) 0.4 4.0log
10
((6017.5)
f ) 0.4
f
f 0.0089
r 0 y R
u
*
u
f
2
1.66
.0089
2
.1107 m/s
y
+
yu
*

(.0348/2) (.1107)1250
.012
200.64
u
+
5.5 + 5.756log
10
( y
+
) 5.5 + 5.756log
10
(200.64) 18.753
u
+
u/u
*
u
max
u
+
u
*
(18.753).1107 2.075 m/s
u/u
max
r (.5)R
u
u
max
y
R
_
,
1/ 7
R r
R
_
,
1/ 7
(.5)
1/ 7
.906
u (2.075) (.906) 1.879 m/s
K 0.25 D 0.0348 u 1.66
1250 n 0.45
N
Re, PL

D
n
( u )
2 n
8
n 1
K
_
4n
3n + 1
_
,
n
.0348
.45
(1.66)
2 .45
1250
8
.45 1
(.25)
_
4(.45)
3(.45) + 1
_
,
.45
6736.6
2.12.8 Turbulent Flow - Power Law Fluid 157
al l owi ng to be cal cul ated (equati on presented i n Tabl e 2.3) from:
yi el di ng
Vel oci ty i s maxi mum at the center-l i ne where . The fri cti on
vel oci ty (Eq. [2.127]) i s
i s cal cul ated usi ng Eq. [2.132] as:
Then, i s determi ned from Eq. [2.131]:
The maxi mum vel oci ty i s found from the defi ni ti on (Eq. [2.125]) of the
turbul ent vel oci ty ( ) as
f
1
4
n
0.75
_
,
log
10
[(N
Re, PL
)f
(1 (n/2))
]
0.4
n
1.2
_
4
(.45)
0.75
_
,
log
10
[(6736.6)f
(1 (.45/2))
]
0.4
(.45)
1.2
_
,
f 0.0052
y R
u
*
u
f
2
1.66
.0052
2
.0846 m/s
y
+
y
+
y
n
(u
*
)
2 n
/K (.0348/2)
.45
(.0846)
2 .45
(1250)/.25 17.6
u
+
u
+
5.66
(.45)
.75
log
10
(17.6)
0.566
(.45)
1.2
+
3.475
(.45)
.75
1.960 + 0.815(.45) 1.628(.45) log

10
3 +
1
(.45)
_
,
1
1
]
22.75
u
+
u/u
*
u
max
u
+
u
*
(22.64).0846 1.92 m/s
Chapt e r 3 . Rot at ional Vis c ome t ry
3.1. Introduction
Tradi ti onal rotati onal vi scometers i ncl ude cone and pl ate, paral l el
pl ate, and concentri c cyl i nder uni ts operated under steady shear con-
di ti ons Fi g. 1.1). They may al so be capabl e of operati ng i n an osci l l atory
mode whi ch wi l l be consi dered i n the di scussi on of vi scoel asti ci ty,
Chapter 5. Cone and pl ate systems are someti mes capabl e of deter-
mi ni ng normal stress di fferences. Concentri c cyl i nder systems have
been used i n research toeval uate these di fferences (Padden and DeWi tt,
1954); however, commerci al i nstruments of thi s type are not avai l abl e.
Mi xer vi scometry, a "l ess tradi ti onal "method i n rotati onal vi scometry,
i s al so presented because i t has excel l ent uti l i ty i n sol vi ng many rheo-
l ogi cal probl ems found i n the food i ndustry.
3.2. Concentric Cylinder Viscometry
3.2.1. Derivation of the Basic Equation
The concentri c cyl i nder vi scometer i s a very common i nstrument
that wi l l operate i n a moderate shear rate range maki ng i t a good choi ce
for col l ecti ng data used i n many engi neeri ng cal cul ati ons. A number of
assumpti ons are made i n devel opi ng the mathemati cal rel ati onshi ps
descri bi ng i nstrument performance: fl ow i s l ami nar and steady, end
effects are negl i gi bl e, test fl ui d i s i ncompressi bl e, properti es are not a
functi on of pressure, temperature i s constant, there i s no sl i p at the
wal l s of the i nstrument, and radi al and axi al vel oci ty components are
zero. The deri vati on presented here i s based on a physi cal setup known
as the Searl e system where the bob rotates and the cup i s stati onary:
I t i s al so appl i cabl e to a Couette-type system i n whi ch the cup rotates
and the bob i s stati onary. Most concentri c cyl i nder vi scometers are
Searl e-type systems. Unfortunatel y, i t i s not uncommon for the word
"Couette"to be used i n referri ng to any concentri c cyl i nder system.
When the bob rotates at a constant speed and the cup i s stati onary
(Fi g. 3.1), the i nstrument measures the torque ( )requi red to mai ntai n
a constant angul ar vel oci ty of the bob ( ). The opposi ng torque comes
from the shear stress exerted on the bob by the fl ui d. A force bal ance
yi el ds
[3.1]
M
M 2rhr 2hr
2
3.2.1 Derivation of the Basic Equation 159

where i s any l ocati on i n the fl ui d, . Sol vi ng Eq. [3.1] for the
shear stress shows that decreases i n movi ng from the bob to the cup:
[3.2]
Uti l i zi ng Eq. [3.2], the shear stress at the bob ( ) can be defi ned as
[3.3]
Figure 3.1. Typical concentric cylinder testing apparatus (based in DIN 53018)
showing a bob with recessed top and bottom to minimize end effects.
To determi ne shear rate, consi der the l i near vel oci ty at i n terms
of the angul ar vel oci ty ( ) at :
[3.4]
The deri vati ve of the vel oci ty wi th respect to the radi us i s
[3.5]
r R
b
r R
c
f (r)
M
2hr
2
r R
b
b

M
2hR
b
2
h
R
b
R
c
r
r
u r
du
dr

r d
dr
+
160 Chapter 3. Rotational Viscometry
Si nce i s rel ated to the rotati on of the enti re body, i t does not rel ate to
i nternal sheari ng; therefore, Eq. [3.5] can be wri tten as
[3.6]
Usi ng the defi ni ti on of shear rate devel oped i n Eq. [2.9], may be defi ned
i n terms of :
[3.7]
To rel ate angul ar vel oci ty to shear stress, note that torque i s constant
wi th steady fl ow so an expressi on for may be determi ned from Eq.
[3.2]:
[3.8]
Di fferenti ati ng Eq. [3.8] wi th respect to the shear stress yi el ds
[3.9]
Substi tuti ng the val ue of torque defi ned by Eq. [3.1] i nto Eq. [3.9] gi ves
[3.10]
or, wi th si mpl i fi cati on,
[3.11]
The shear rate i s some functi on of the shear stress, hence,
[3.12]
Sol vi ng Eq. [3.12] for the di fferenti al of the angul ar vel oci ty yi el ds
[3.13]
whi ch can be expressed i n terms of by substi tuti ng Eq. [3.11] i nto
Eq. [3.13]:
[3.14]
du
dr

r d
dr

du
dr

r d
dr
r
r

M
2h
_
,
1/ 2
M
2h
_
,
1/ 2
()
1/ 2
dr
d

M
2h
_
,
1/ 2
1
2
_
,
()
3/ 2
dr
d

2hr
2
2h
_
,
1/ 2
1
2
_
,
()
3/ 2
r
2
dr
r

d
2
r
d
dr
f ()
d
dr
r
f ()
d
d
1
2
f ()
d
3.2.1 Derivation of the Basic Equation 161

I ntegrati ng Eq. [3.14] over the fl ui d present i n the annul us resul ts i n a
general expressi on for the angul ar vel oci ty of the bob ( ) as a functi on
of the shear stress i n the gap:
[3.15]
Note that the l i mi ts of i ntegrati on are an expressi on of the no sl i p
boundary condi ti on assumed i n the deri vati on: Angul ar vel oci ty i s zero
at the cup (the stati onary surface), and equal to at the bob (the movi ng
surface). The l eft hand si de of Eq. [3.15] i s easi l y i ntegrated resul ti ng
i n the fol l owi ng equati on rel ati ng angul ar vel oci ty to shear stress:
[3.16]
The sol uti on of Eq. [3.16] depends on whi ch i s di ctated by the
behavi or of the fl ui d i n questi on. I t can be sol ved di rectl y i f the functi onal
rel ati onshi p between shear stress and shear rate i s known. Eq. [3.15]
i s used as the starti ng poi nt i n Exampl e Probl em 3.8.5 tofi nd the vel oci ty
profi l e of a power l aw fl ui d i n a concentri c cyl i nder system.
Eq. [3.15] refl ects a general sol uti on for concentri c cyl i nder vi s-
cometers because the l i mi ts of the i ntegral coul d be easi l y changed to
the case where the bob i s stati onary and the cup rotates (torque i s equal
i n magni tude, but opposi te i n si gn i f measured on the cup) or even a
si tuati on where the bob and cup are both rotati ng. I t i s i mportant to
recogni ze the fact that Eq. [3.16] i s anal ogous to the general sol uti on
(Eq. [2.15]) devel oped for tube vi scometers. Both provi de an overal l
starti ng poi nt i n devel opi ng mathemati cal rel ati onshi ps for speci fi c
types of fl ui ds.
ApplicationtoNewtonian Fluids. The rel ati onshi p between shear
stress and shear rate for a Newtoni an fl ui d i s, by defi ni ti on,
[3.17]
Substi tuti ng thi s i nto the general expressi on for gi ven by Eq. [3.16]
yi el ds
[3.18]

0
d
1
2
c
f ()
d

1
2
c
f ()
d
f ()
f ()

1
2
c
f ()
d

1
2
,
d

1
2
c
d
then,
[3.19]
Usi ng Eq. [3.2] for shear stress al l ows Eq. [3.19] to be wri tten i n terms
of the system geometry and the torque response of the i nstrument:
[3.20]
Rearrangement gi ves a si mpl i fi ed expressi on, cal l ed the Margul es
equati on, descri bi ng the behavi or of a Newtoni an fl ui d i n a concentri c
cyl i nder system:
[3.21]
Thi s equati on cl earl y i ndi cates that experi mental data for Newtoni an
fl ui ds wi l l show torque to be di rectl y proporti onal to bob speed.
Application to Power Law Fluids. Wi th a power l aw fl ui d, the
rel ati onshi p between shear stress and shear rate i s
[3.22]
whi ch can be substi tuted i nto Eq. [3.16] yi el di ng
[3.23]
or, after i ntegrati on,
[3.24]
Usi ng Eq. [3.2], an al ternati ve expressi on for the power l aw fl ui d i s
obtai ned:
[3.25]
Eq. [3.25] reveal s that torque i s not di rectl y proporti onal to bob speed
because i t i s strongl y i nfl uenced by the fl ow behavi or i ndex.
Application to BinghamPlastic Fluids. A Bi ngham pl asti c fl ui d
has the fol l owi ng rel ati onshi p between shear stress and shear rate:

1
2
(
b

c
)

1
2
M
2hR
b
2

M
2hR
c
2
1
1
]

M
4h
1
R
b
2

1
R
c
2
1
1
]
f ()

K
_
,
1/ n

1
2
c
f ()
d

1
2
K
_
,
1/ n
d

n
2K
1/ n
[(
b
)
1/ n
(
c
)
1/ n
]

n
2K
1/ n
M
2hR
b
2
_
,
1/ n
M
2hR
c
2
_
,
1/ n
1
1
]
n
2K
1/ n
M
2hR
b
2
_
,
1/ n
R
b
R
c
_
,
2/ n
1
1
]
3.2.2 Shear Rate Calculations 163
[3.26]
Substi tuti ng Eq. [3.26] i nto Eq. [3.16] yi el ds
[3.27]
I ntegrati on and substi tuti on of Eq. [3.2] provi des the general rel ati on-
shi p (known as the Rei ner-Ri wl i n equati on)between the torque, angul ar
vel oci ty, and system geometry:
[3.28]
Thi s equati on i s val i d onl y when the yi el d stress i s exceeded at al l poi nts
i n the fl ui d meani ng that the mi ni mum shear stress must greater than
the yi el d stress:
[3.29]
i s the mi ni mum torque requi red to overcome the yi el d stress. I f
eval uati ng fl ui d behavi or near the l i mi ts descri bed by Eq. [3.29], the
yi el d stress shoul d be determi ned before conducti ng standard tests i n
Searl e-type concentri c cyl i nder vi scometers. Thi s can be accompl i shed
wi th vari ous techni ques, such as the vane method di scussed i n Sec.
3.7.3, Wi th that data, one can cal cul ate the mi ni mum rotati onal speed
of the bob requi red to i nsure sheari ng throughout the cyl i ndri cal gap
(see Exampl e Probl em 3.8.1). I n Couette systems, appl yi ng suffi ci ent
torque to rotate the cup assures shear fl ow i n the enti re annul us because
the mi ni mum shear stress occurs at .
3.2.2. Shear Rate Calculations
Numerous methods of esti mati ng shear rates i n concentri c cyl i nder
vi scometers have been proposed (Tabl e 3.1) and many of those tech-
ni ques are di scussed i n books by vari ous authors: Whorl ow (1992), Van
Wazer et al . (1963), and Darby (1976). A few of the more practi cal
approxi mati on techni ques are summari zed i n the current work.
f ()
(
o
)
pl

1
2
c
f ()
d

1
2
c
(
o
)
pl
d
M
4h
pl
1
1
]
1
R
b
2
1
R
c
2
1
1
]
pl
ln
R
c
R
b
_
min

M
min
2R
c
2
h
>
o
M
min
R
c
Table 3.1. Some Mathematical Techniques for Analyzing Data from Concentric
Cylinder Viscometers
Solution Source
Shear rate calculation, multiple bobs. Krieger, I.M. and S.H. Maron (1954)
Mean shear stress, small gaps. Mooney, M. (1931)
Shear rate at the bob. Krieger, I.M. and H. Elrod (1953)
Fluids following Ellis equation. Van Wazer et al. (1963)
Fluids following Herschel-Bulkley equation. Van Wazer et al. (1963)
Fluids following Eyring equation. Van Wazer et al. (1963)
Fluids following Casson equation. Murata, T. and S. Oka (1968)
Fluids following Vocadlo equation. Parzonka, W. and J. Vocadlo (1968)
The error i n usi ng approxi mati ons can be cal cul ated by eval uati ng
the exact versus the approxi mate sol uti on. Accuracy i s i mproved as the
approxi mati on becomes more sophi sti cated. When studyi ng fl ui d foods,
the si mpl e shear, Newtoni an or power l aw approxi mati ons are often
adequate.
Simple Shear Approximation. Wi th a very narrow annul us
( ), the curvature of the wal l s i s negl i gi bl e and the system
approaches si mpl e shear. Assumi ng a uni form shear rate across the
gap gi ves
[3.30]
where . When cal cul ati ng shear rates wi th Eq. [3.30] a
correspondi ng average shear stress shoul d be used:
[3.31]
The error i nvol ved i n usi ng Eq. [3.30] for power l aw fl ui ds i s i nvesti gated
i n Exampl e Probl em 3.8.2.
Newtonian Approximation. The shear rate at the bob for a New-
toni an fl ui d i s determi ned from the defi ni ti on of a Newtoni an fl ui d as
[3.32]
R
c
R
b
R
b
b

R
b
R
c
R
b

1
R
c
/R
b
a

1
2
(
c
+
b
)
1
2
M
2hR
c
2
+
M
2hR
b
2
1
1
]
M(1 +
2
)
4hR
c
2
b
(2)
2
1
_
,
Commerci al vi scometers frequentl y use thi s equati on to approxi mate
shear rate, often i n the form of a representati ve (average) shear rate.
The deri vati on of Eq. [3.32] i s presented i n Exampl e Probl em 3.8.3 and
the representati ve shear rate concept, devel oped usi ng thi s equati on, i s
exami ned i n Exampl e Probl em 3.8.4.
Power Law Approximation. Consi deri ng the defi ni ti on of a power
l aw fl ui d, an equati on for the shear rate at the bob can be deri ved as
[3.33]
The compl ete devel opment of Eq. [3.33] i s gi ven i n Exampl e Probl em
3.8.5.
Usi ng Eq. [3.33] requi res a numeri cal val ue of the fl ow behavi or
i ndex. I t may be determi ned di rectl y by consi deri ng the power l aw
equati on wi th Eq. [3.33] used as the expressi on for shear rate:
[3.34]
Taki ng the l ogari thm of each si de, Eq. [3.34] may be wri tten as
[3.35]
whi ch, by eval uati ng the deri vati ve wi th respect to , provi des a
si mpl e expressi on for the fl ow behavi or i ndex:
[3.36]
Si nce , Eq. [3.36] may al so be wri tten as
[3.37]
Hence, for power l aw fl ui ds, i s the sl ope (a strai ght l i ne) of , or
, versus . Once i s known, Eq. [3.33] can be easi l y eval uated.
Krieger Method. The general force bal ance on the bob gave the
rel ati onshi p between shear stress and shear rate as (Eq. [3.16])
2
n
_
2/ n
2/ n
1
1
1
]
b
K(
b
)
n

n
K
1/ n
2
n
_
2/ n
2/ n
1
1
1
]
_
,
n
ln(
b
) n ln() + n ln
K
1/ n
2
n
_
2/ n
2/ n
1
1
1
]
_
,
_
,
ln
n
d(ln
b
)
d(ln)
b
M/(2hR
b
2
)
n
d(lnM)
d(ln)
n ln(M)
ln(
b
) ln() n

1
2
b
f ()
d

Taki ng the deri vati ve of thi s expressi on wi th respect to the shear stress
at the bob yi el ds (Kri eger and Maron, 1952)
[3.38]
From Eq. [3.1]
[3.39]
so
[3.40]
and
[3.41]
Substi tuti on and si mpl i fi cati on, usi ng Eq. [3.40] and [3.41], transforms
Eq. [3.38] i nto
[3.42]
Asol uti on (a sui tabl e expressi on for shear rate at the bob)toequati on
Eq. [3.42] requi res the eval uati on of an i nfi ni te seri es. The best sol uti ons
are those whi ch i nvol ve a seri es where a good approxi mati on can be
obtai ned by eval uati ng a smal l number of terms. An excel l ent sol uti on
to Eq. [3.42], one whi ch truncates an i nfi ni te seri es after the fi rst term,
was devel oped by Kri eger (1968)and recommended by Yang and Kri eger
(1978):
[3.43]
where:
[3.44]
[3.45]
and
d
d
b
1
2
f (
b
)
f (
c
)
c
_
d
c
d
b
_
,
1
1
]
M
2h
r
2
constant
c
R
c
2

b
R
b
2
d
c
d
b
R
b
R
c
_
,
2
2
d
d
b
1
2
b
f (
b
)
f (
b
)
2
_
b
f (
b
)

2
s
_
2/ s
2/ s
1
1
1
]
(1 + s
2
sg)
1/s

,
d
d
b
d(ln)
d(ln
b
)

d(ln)
d(lnM)
s
b
d(1/s)
d
b
d(1/s)
d(ln
b
)

d(1/s)
d(lnM)
[3.46]
wi th , the argument of the functi on gi ven by Eq. [3.46], defi ned as
. I n concentri c cyl i nder vi scometers, i s typi cal l y i n the range
of 1.01 to 1.40 wi th smal l er val ues bei ng more common. Eq. [3.43] i s
acceptabl e over thi s range of , most accurate at smal l val ues of , and
good i n many cases for val ues up to 2.0 (Yang and Kri eger, 1978).
Figure 3.2. Plot of Eq. [3.46], , in Krieger (1968) solution for calculating
shear rate at the bob of a concentric cylinder viscometer.
The Kri eger sol uti on (Eq. [3.43]) i s very cl ose to the power l aw
approxi mati on. I n fact, wi th a power l aw fl ui d, and the Kri eger
equati on becomes the power l aw sol uti on gi ven by Eq. [3.33]. I t i s al so
i mportant to note that the maxi mum val ue of i s approxi matel y 0.1
(Fi g. 3.2) and, when cal cul ati ng the correcti on factor i n Eq. [3.43], thi s
number i s mul ti pl i ed by other smal l numbers. Hence, the power l aw
expressi on for shear rate, wi th found at each val ue of , i s an excel l ent
approxi mati on for the shear rate at the bob:
g f (x)
x(e
x
(x 2) + x + 2)
2(e
x
1)
2
x
2(ln)/s

0 1 2 3 4 5 6 7
0
0.02
0.04
0.06
0.08
0.1
0.12
x
g
(
x
)
g(x)
s n
g(x)
s
b
[3.47]
where i s defi ned by Eq. [3.44].
Other sol uti ons to Eq. [3.42] have been summari zed by Yang and
Kri eger (1978). Darby (1985) has di scussed the probl em of concentri c
cyl i nder vi scometer data reducti on for materi al s wi th a yi el d stress and
found the power l aw approxi mati on to gi ve acceptabl e resul ts i n the
majori ty of practi cal si tuati ons. Best resul ts, however, are obtai ned
wi th pri or knowl edge of the yi el d stress so compl ete sheari ng i n the
annul us can be assured (Nguyen and Boger, 1987). The uti l i ty of vari ous
shear rate approxi mati on equati ons i s i l l ustrated i n Exampl e Probl em
3.8.6. Thi s exampl e al sopresents the Kri eger and Maron (1952)sol uti on
to Eq. [3.42] for cal cul ati ng the shear rate at the bob i n a concentri c
cyl i nder system.
3.2.3. Finite Bob in an Infinite Cup
When , the case of a fi ni te bob i n an i nfi ni te cup, i s very
smal l so Eq. [3.42] becomes
[3.48]
Sol vi ng thi s expressi on for the shear rate at the bob and mul ti pl yi ng
numerator and denomi nator by , then si mpl i fyi ng, gi ves
[3.49]
Eq. [3.49] provi des an easy means of determi ni ng the shear rate at the
bob i n an i nfi ni te cup. The sol uti on may be useful i n tests i nvol vi ng
very l arge beakers or i ndustri al scal e food vats. I t i s cl earl y val i d for
any ti me-i ndependent materi al wi thout a yi el d stress. Remarkabl y, i t
i s al so val i d for fl ui ds wi th a yi el d stress (the yi el d stress probl em i s
di scussed l atter i n thi s secti on).
Shear rate equati ons devel oped for concentri c cyl i nder systems can
al so be used for the i nfi ni te cup case by maki ng the appropri ate
al l owance for geometry. Thi s i dea i s i l l ustrated for a power l aw fl ui d,
tomato ketchup, i n Exampl e Probl em 3.8.7. Errors i nvol ved i n usi ng
the i nfi ni te cup approxi mati on depend on geometry as wel l as fl ui d
2
s
_
2/ s
2/ s
1
1
1
]
s
R
c
R
b
1/
d
d
b
f (
b
)
2
b
f (
b
)

b
(2
b
)
d
d
b
2
b
,
d
d
b
(2)
d(ln)
d(ln
b
)
3.3 Cone and Plate Viscometry 169
properti es (see Exampl e Probl em 3.8.8). A general probl em i l l ustrati ng
the i nfi ni te cup cal cul ati on techni que i s gi ven for sal ad dressi ng i n
Exampl e Probl em 3.8.9.
Eq. [3.49] may al so be used to cal cul ate an exact sol uti on for the
shear rate at the bob i n concentri c cyl i nder systems (Nguyen and Boger,
1987) when the fl ui d i n the annul us i s onl y parti al l y sheared, i .e., when
Eq. [3.29] i s vi ol ated and the mi ni mum shear stress i s l ess than the
yi el d stress. I n thi s si tuati on, the yi el d stress ( ) may be substi tuted
for i n Eq. [3.16], then di fferenti ated to yi el d an expressi on si mi l ar to
Eq. [3.38]:
[3.50]
Si nce the second term wi thi n the brackets i s zero, Eq. [3.50] becomes
Eq. [3.48] whi ch can be si mpl i fi ed to Eq. [3.49]. Therefore, the sol uti on
for a si ngl e cyl i nder rotati ng i n an i nfi ni te medi um and the sol uti on for
the case of a bob rotati ng i n an annul us wi th parti al l y sheared fl ui d are
i denti cal ! Thi s surpri si ng resul t means that Eq. [3.49] can be used i n
a vat (or beaker) contai ni ng materi al wi th a yi el d stress i f the bob i s
pl aced suffi ci entl y far i nto the contai ner so wal l effects are not a source
of error. The appropri ate di stance from the wal l can be esti mated from
Eq. [3.2]. Thi s i dea i s i l l ustrated i n Exampl e Probl em 3.8.10.
3.3. Cone and Plate Viscometry
Usi ng a cone and pl ate apparatus (Fi g. 3.3), the shear stress versus
shear rate curve may usual l y be obtai ned di rectl y so the cal cul ati ons
are qui te si mpl e. The i nstrument i s a moderate shear rate devi ce whi ch
i s i nappropri ate for fl ui ds wi th l arge parti cl es because the cone angl e
( ) i s smal l , preferabl y l ess than 0.09 rad (5 degrees). I n operati ng a
cone and pl ate vi scometer, the apex of the cone al most touches the pl ate
and fl ui d fi l l s the gap. The cone i s rotated at a known angul ar vel oci ty
( )and the resul ti ngtorque ( )i s measured on the fi xed pl ate or through
the cone. Some i nstruments are desi gned wi th rotati ng pl ates and fi xed
cones.
Fl ow i n a cone and pl ate vi scometer can be very compl ex requi ri ng
a l abori ous sol uti on of the fundamental equati ons of moti on (Wal ters,
1975). However, when usi ng a smal l cone angl e (l ess than 5 degrees),
suffi ci entl y l ow rotati onal speeds, and wi th no errors due to surface
c
d
d
b
1
2
f (
b
)
f (
o
)
o
_
d
o
d
b
_
,
1
1
]
M
tensi on effects at the free fl ui d surface (surface shoul d be spheri cal i n
shape wi th a radi us of curvature equal to the cone radi us), the shear
rate at may be cal cul ated as
[3.51]
i ndi cati ng that the shear rate i s constant throughout the gap. Thi s i s
one of the mai n advantages of a cone and pl ate vi scometer. Wi th the
smal l angl es found i n typi cal fi xtures, .
To devel op an expressi on for shear stress, consi der the di fferenti al
torque on an annul ar ri ng of thi ckness dr:
[3.52]
Figure 3.3. Cone and plate (left), and parallel plate (right) sensors.
Eq. [3.52] i s i ntegrated over the radi us to fi nd the total torque response:
[3.53]
Si nce the shear rate i s constant i n the gap, the shear stress i s al so
constant i n that area so . Then, Eq. [3.53] can be si mpl i fi ed to
[3.54]
hence,
[3.55]
r

r
r tan

tan
tan
dM (2rdr)r
R
CONE ANGLE
R
0
M
dM

0
R
(2r
2
) dr
f (r)
M 2
0
R
r
2
dr

3M
2R
3
3.3 Cone and Plate Viscometry 171
Thi s resul t shows that , l i ke , i s constant throughout the gap. Usi ng
Eq. [3.51] and [3.55], shear rate and shear stress can be easi l y cal cul ated.
By varyi ng the angul ar vel oci ty, cone angl e and cone radi us, a wi de
vari ety of condi ti ons can be tested. I f a speci fi c model i s sel ected,
rheol ogi cal properti es can be cal cul ated di rectl y. The fol l owi ng equa-
ti on, for exampl e, woul d appl y to power l aw fl ui ds:
[3.56]
Figure 3.4. Cone and plate system showing pressure distribution on a plate for
a viscoelastic fluid.
Fl ui ds whi ch have a si gni fi cant el asti c component wi l l produce a
measurabl e pressure di stri buti on i n the di recti on perpendi cul ar to the
shear fi el d (Fi g. 3.4). Some cone and pl ate vi scometers al l ow mea-
surement of the resul ti ng normal (axi al di recti on) force on the cone
maki ng i t possi bl e to cal cul ate the fi rst normal stress di fference, noted
i n Eq. [1.23], as (Wal ters, 1995):
[3.57]
3M
2R
3
K
tan
_
,
n
plate
test fluid
cone
N
1
(
11

22
)
2(F
normal
)
R
2
The normal force di fference i ncreases wi th the shear rate for vi scoel asti c
fl ui ds: I t i s equal tozerofor Newtoni an fl ui ds. Expected torque response
and data anal yses for cone and pl ate systems are i nvesti gated i n
Exampl e Probl ems 3.8.11 and 3.8.12, respecti vel y.
3.4. Parallel Plate Viscometry (Torsional Flow)
Shear rate i n a paral l el pl ate apparatus (Fi g. 3.3) i s a functi on of
[3.58]
so the shear rate at the ri m of the pl ate ( ) i s
[3.59]
Shear stress must be determi ned from the torque response of the
i nstrument whi ch i s eval uated by constructi ng a force bal ance equati on
on the di sk and i ntegrati ng over the radi us. The same procedure was
fol l owed i n the previ ous secti on for the cone and pl ate vi scometer.
Wi th the cone and pl ate system, the shear stress was constant
al l owi ng Eq. [3.53] to be easi l y eval uated. I n a paral l el pl ate system,
however, the shear stress i s a functi on of the radi us maki ng the i nte-
grati on more compl i cated. Eq. [3.53] may be wri tten i n a si mpl i fi ed form
as
[3.60]
Usi ng Eq. [3.58], the vari abl e of i ntegrati on may be changed from to
. Maki ng the appropri ate substi tuti ons ( and ),
and eval uati ng the i ntegral from 0 to , yi el ds
[3.61]
Di vi di ng each si de of Eq. [3.61] by and si mpl i fyi ng the resul t gi ves
[3.62]
or, si nce ,
r
f (r)
r
h
R

R
h
0
M
dM 2
0
R
(r
2
) dr
r
r
2
(
h/)
2
dr (h/)d
0
M
dM 2
,
3
R
(
)
2
d
R
3
M
2R
3

1
(
R
)
3
R
(
)
2
d
f (
)
3.4 Parallel Plate Viscometry (Torsional Flow) 173
[3.63]
Di fferenti ati ng Eq. [3.63] wi th respect to , usi ng Lei bni tz rul e (Eq.
[2.16] and [2.17]) on the ri ght hand si de, gi ves an i ndependent term for
:
[3.64]
Si mpl i fi cati on of Eq. [3.64] provi des an expressi on for the shear stress
at the ri m of the pl ate ( ):
[3.65]
or
[3.66]
where, recal l , . Thi s expressi on i s si mi l ar i n form to the
Rabi nowi tsch-Mooney equati on, Eq. [2.20]. Appl i cati on of Eq. [3.66] i s
demonstrated for a 3% hydroxypropyl methyl cel l ul ose sol uti on i n
The rel ati onshi p between and can be eval uated di rectl y for
parti cul ar types of behavi or. Wi th a Newtoni an fl ui d
[3.67]
whi ch can be substi tuted i nto Eq. [3.60]:
[3.68]
yi el di ng, after i ntegrati on and si mpl i fi cati on,
[3.69]
When wri tten i n thi s form, i t i s cl ear that for Newtoni an
fl ui ds. The same procedure can be fol l owed for power l aw fl ui ds gi vi ng
M
2R
3
(
R
)
3
R
(
)
2
f (
) d
R
f (
R
)
(
R
)
3
d(M/(2R
3
))
d
R
+
M
2R
3
_
,
3(
R
)
2
(
R
)
2
f (
R
)
R
f (
R
)
3M
2R
3
+
R
d(M/(2R
3
))
d
R
f (
R
)
M
2R
3
3 +
d ln(M)
d ln(
R
)
1
1
]
R
R/h
M

r
h
0
M
dM 2
0
R
r
3
h
dr
2M
R
3

R
h
R
2M/(R
3
)
[3.70]
showi ng that the shear stress at the ri m for a power l aw fl ui d depends
on the numeri cal val ue of the fl ow behavi or i ndex:
[3.71]
Consi deri ng the above rel ati onshi ps, one can see that the deri vati ve
term i n Eq. [3.66] i s equal to one for Newtoni an fl ui ds and for power
l aw fl ui ds.
Paral l el pl ate systems, usi ng axi al thrust data, can be used to cal -
cul ate the second normal stress di fference (noted i n Eq. [1.24]) provi ded
the fi rst normal stress di fference has been determi ned from cone and
pl ate measurements (Wal ters, 1975):
[3.72]
I f i s known, thi s equati on can be used to determi ne . However,
si nce i s very smal l compared to , i t i s often reasonabl e to assume
that the normal stress determi ned from paral l el pl ate data i s a good
approxi mati on of .
3.5. Corrections: Concentric Cylinder
End Correction. I t i s i mportant to account for the i nfl uence of the
bottom of the cyl i nder on the torque response of the system. Thi s surface
i s i n contact wi th the fl ui d but not taken i ntoaccount i n the force bal ance
gi ven by Eq. [3.1].
To determi ne the end correcti on, torque (or i nstrument scal e di vi -
si on) i s measured at a fi xed rate of rotati on when the annul us i s fi l l ed
to vari ous hei ghts (Fi g. 3.5). Resul ti ng data are pl otted (Fi g. 3.6) as
torque versus the hei ght of fl ui d i n contact wi th the i mmersed l ength
of the bob. The curve shoul d be l i near wi th the sl ope equal to the torque
requi red tomai ntai n the fi xed rate of rotati on per uni t l ength of cyl i nder.
Effecti ve hei ght ( ) i s determi ned from the i ntercept by extrapol ati ng
to a val ue of zero torque (Fi g. 3.6). Thi s techni que i s i l l ustrated i n
Exampl e Probl em 3.8.14 for the tapered bob of a Hercul es hi gh-shear
vi scometer.
M(3 + n)
2R
3
K
R
h
_
,
n
R

M(3 + n)
2R
3
n
N
1
N
2

2F
normal
R
2
_
1 +
1
2
_
,
d(lnF
normal
)
d(ln
R
)
1
1
]
N
1
N
2
N
2
N
1
N
1
h
o
3.5 Corrections: Concentric Cylinder 175
Figure 3.5. Illustration of values used in determining end correction.
Figure 3.6. End correction for a concentric cylinder system using a graphical
technique to determine .
Effecti ve hei ght val ues are used i n the previ ous equati ons devel oped
for concentri c cyl i nder systems. The Margul es equati on (Eq. [3.21]), for
exampl e, woul d be expressed as
h
h
h
1
2
3
h
-10 0 10 20 30 40 50
T
O
R
Q
U
E
,

S
C
A
L
E

U
N
I
T
S
x
x
x
-h
0
Height of Bob in Contact with Fluid, mm
h
o
[3.73]
The val ue of gi ven i n Eq. [3.1] i s repl aced by whi ch, together,
may be thought of as the effecti ve hei ght of the bob. An end correcti on
cal cul ated for a parti cul ar bob wi th a standard Newtoni an fl ui d provi des
a general approxi mati on for To obtai n maxi mum accuracy, the end
correcti on shoul d be eval uated for each fl ui d and rotati onal speed under
consi derati on. Thi s procedure, however, i s very l abori ous and not
consi dered standard practi ce.
The end correcti on can al so be eval uated i n terms of an equi val ent
torque ( )generated by a fl ui d i n contact wi th the bottom of the sensor.
Thi s i dea i s i l l ustrated for three di fferent speeds i n Fi g. 3.7. can be
pl otted as a functi on of to determi ne the rel ati onshi p between the two
parameters. The torque correcti on i s subtracted from the measured
torque i n cal cul ati ng the shear stress at the bob:
[3.74]
Correcti ng for end effects wi th or shoul d yi el d i denti cal resul ts.
Vari ous bob desi gns have been devel oped to mi ni mi ze end effects.
Bobs can be made wi th a reservoi r at the top and a recessed bottom.
Thi s bob desi gn, shown i n Fi g. 3.1, i s based on a German standard (DI N
53018) devel oped by the German I nsti tute for Standardi zati on
(Deutsches I nsti tut fr Normung). End effect probl ems can al so be
reduced by desi gni ng the bottom wi th a sl i ght angl e (cal l ed a Mooney-
Couette bob, Fi g. 3.8) i n an effort to make the shear rate at the bottom
equi val ent to the shear rate i n the annul us. The proper angl e ( ) can
be cal cul ated by equati ng the annul ar shear rate to the shear rate i n
the gap (see Exampl e Probl em 3.8.15). Probl ems wi th l arge parti cul ates
and sensor al i gnment l i mi t the useful ness of Mooney-Couette systems.
Other methods, such as usi ng a mercury i nterface at the bottom of the
bob (Pri ncen, 1986), have al so been proposed.

M
4(h + h
o
)
1
R
b
2
1
R
c
2
1
1
]
h h + h
o
h
o
M
e
M
e
b

M M
e
2hR
b
2
M
e
h
o

Figure 3.7. End correction for a concentric cylinder system using a graphical
technique to determine a torque correction at different speeds.
Figure 3.8. Mooney-Couette bob design.
ViscousHeating. Temperature i ncrease i n a fl ui d duri ng rheol ogi cal
testi ng can be caused by the vi scous generati on of heat. I t may be a
seri ous probl em i n some experi ments because rheol ogi cal properti es are
strongl y i nfl uenced by temperature. The purpose of thi s secti on i s to
provi de a means of determi ni ng i f a si gni fi cant temperature i ncrease
1
2
e
e
Torque Due to Bottom of Bob at
3
M
M
1
h h h
1 2 3
Height of Fluid in Contact with Bob
T
o
r
q
u
e
1
may occur duri ng testi ng. I f the probl em i s seri ous (i t wi l l al ways exi st)
appropri ate acti on must be taken. Most vi scometers are desi gned wi th
effecti ve temperature control systems that mi ni mi ze vi scous heati ng
probl ems by rapi dl y removi ng the excess heat generated duri ng testi ng.
To address vi scous heati ng, the case of uni form sheari ng between
paral l el pl ates (Fi g. 1.9) may be consi dered (Deal y, 1982). A concentri c
cyl i nder system can be approxi mated usi ng thi s i dea when the gap i s
narrow ( ). Thi s i s a one di mensi onal probl em where i t i s
assumed that pl ates are separated by a di stance , wi th at the
bottom pl ate and at the top pl ate. Al so, assume the fl ui d i s
Newtoni an wi th a vi scosi ty that does not vary wi th temperature. I n
thi s case the di fferenti al equati on rel ati ng temperature and l ocati on,
under steady state condi ti ons, i s
[3.75]
where i s the fl ui d thermal conducti vi ty. i s the vi scous energy
generated per uni t ti me per uni t vol ume expressed i n uni ts of J s
-1
m
-3
.
Shear rate i s consi dered to be uni form throughout the gap. The sol uti on
to Eq. [3.75] i s
[3.76]
where and are constants whi ch depend on the boundary condi ti ons
of the probl em bei ng consi dered. Sol uti ons for twodi fferent cases fol l ow.
The vari ous scenari os shoul d be vi sual i zed i n terms of the propensi ty
of heat to move through the cup and (or) the bob surfaces duri ng
sheari ng.
Both surfaces are mai ntai ned at the same temperature ( ). I n thi s
si tuati on the boundary condi ti ons are at , and at ,
whi ch al l ow the constants i n Eq. [3.76] to be determi ned: ,
. Substi tuti ng these val ues back i nto Eq. [3.76] al l ows the
temperature tobe expressed as a functi on of posi ti on between the pl ates:
[3.77]
R
c
R
b
R
b
s x
2
0
x
2
s
k
d
2
T
dx
2
2
(
)
2
(
)
2
k
T f (x
2
)
(
)
2
2k
x
2
2
+ C
1
x
2
+ C
2
C
1
C
2
T
o
T T
o
x
2
0 T T
o
x
2
s
C
2
T
o
C
1

()
2
s/(2k)
T f (x
2
) T
o
+

()
2
x
2
2k
(s x
2
)
meani ng temperature di stri buti on i s parabol i c and the maxi mum
temperature occurs at the mi dpl ane:
[3.78]
Hence, the temperature ri se i n the gap i s equal to i ndi cati ng
that mi ni mi zi ng the si ze of the gap ( ), anal ogous to havi ng a
smal l er val ue of , i s benefi ci al i n reduci ng vi scous heati ng probl ems.
One surface i s adi abati c and the other surface i s mai ntai ned at . I n
thi s case the boundary condi ti ons are at and at
al l owi ng determi nati on of the constants: and .
Substi tuti on of these val ues i nto Eq. [3.76] yi el ds the temperature
di stri buti on functi on:
[3.79]
The di stri buti on i s parabol i c wi th the maxi mum temperature occurri ng
at the adi abati c surface where :
[3.80]
Compari ng thi s resul t to the case where both surfaces are mai ntai ned
at i ndi cates the temperature vari ati on i n a sampl e may be four ti mes
greater when one surface i s consi dered adi abati c. Eq. [3.80] has been
appl i ed to tomato ketchup i n Exampl e Probl em 3.8.16.
Effect of temperature vari ati on on vi scosi ty. Probl ems associ ated wi th
vi scous heati ng wi l l depend on the extent towhi ch rheol ogi cal properti es
are sensi ti ve to temperature. Deal y (1982) gave an exampl e i n whi ch
vi scosi ty was expressed as an exponenti al functi on of temperature:
[3.81]
or
[3.82]
T
max
f (s/2) T
o
+

()
2
s
2
8k

()
2
s
2
/(8k)
R
c
R
b
s
T
o
T T
o
x
2
0 dT/dx
2
0
C
1

()
2
s/k x
2
s C
2
T
o
T f (x
2
) T
o
+

()
2
x
2
2k
(2s x
2
)
x
2
s
T f (s) T
max
T
o
+

()
2
s
2
2k
T
o
f (T)
o
e
(b(T T
o
)/T
o
)
ln
,
b
T T
o
T
o
_
,
where i s the vi scosi ty at and i s a constant, numeri cal l y dependent
on the fl ui d i n questi on. Assumi ng i s the wal l temperature, the rati o
of the maxi mum to the mi ni mum vi scosi ty i s
[3.83]
Consi deri ng the case when both wal l s are mai ntai ned at , Eq. [3.78]
can be substi tuted i nto Eq. [3.83] yi el di ng:
[3.84]
Meani ng, for exampl e, i f the vari ati on i n vi scosi ty duri ng testi ng i s to
be l ess than 10%, then
[3.85]
maki ng i t necessary to mai ntai n the fol l owi ng i nequal i ty:
[3.86]
I n the si tuati on where one wal l i s mai ntai ned at and the other wal l
i s adi abati c, the vi scosi ty rati omay be eval uated by combi ni ng Eq. [3.83]
and [3.80]:
[3.87]
To mai ntai n a vi scosi ty vari ati on of l ess than 10%
[3.88]
Al though the above cal cul ati ons are not quanti tati vel y exact for
non-Newtoni an fl ui ds, they do i l l ustrate the rel ati ve i mportance of
di fferent experi mental vari abl es. Anal yti cal sol uti ons for power l aw
fl ui ds i n couette fl ow -where the consi stency coeffi ci ent i s expressed as
a power seri es of temperature and the fl ow behavi or i ndex i s assumed
to be i ndependent of temperature- are cumbersome, but avai l abl e
(Mi ddl eman, 1968).
o
T
o
b
T
o
ln
max
min
_
,
b
T
max
T
o
T
o
_
,
T
o
ln
max
min
_
,
b
T
o
+

()
2
s
2
/(8k) T
o
T
o
_
,

b

()
2
s
2
8kT
o
8(ln(1.10)) .76
b

()
2
s
2
kT
o
< .76
T
o
ln
max
min
_
,

b

()
2
s
2
2kT
o
b

()
2
s
2
kT
o
< 2(ln(1.10)) .19
Wall Effects (Slip). Wal l effects due to separati on i n mul ti phase
materi al s may cause errors i n concentri c cyl i nder systems si mi l ar to
those di scussed for tube vi scometers i n Sec. 2.5. Ol droyd (1956) sug-
gested that sl i p may be consi dered i n terms of the general expressi on
for angul ar vel oci ty (Eq. [3.16]) by addi ng a sl i p vel oci ty ( ) that i s a
functi on of wal l shear stress at the bob and the cup:
[3.89]
I n the absence of sl i p, the sl i p vel oci ty i s zero and thi s equati on reduces
toEq. [3.16]. Usi ng the method of Mooney (1931), i t i s possi bl e to correct
for sl i p i n concentri c cyl i nder vi scometers. The method requi res
numerous bobs because measurements are requi red at di fferent val ues
of .
A si mpl e sl i p eval uati on method, requi ri ng two seri es of measure-
ments i n two di fferent measuri ng sets, that have di fferent gap wi dths,
has al so been suggested (Ki l janski , 1989). Cheng and Parker (1976)
presented a method of determi ni ng wal l -sl i p based on the use of a smooth
and a rough bob. They al so urged cauti on i n i nvesti gati ng sl i p because
fl ui ds exhi bi ti ng that phenomenon may have accompanyi ng parti cul ate
behavi or whi ch may mask sl i p and compl i cate data treatment. A rel ated
procedure was proposed by Yoshi mura and Prudhomme (1988). I f sl i p
i s a seri ous probl em, mi xer vi scometry shoul d be eval uated as an
al ternate experi mental method.
The Mooney techni que was used by Qi u and Rao (1989) to eval uate
sl i p i n appl e sauce, a typi cal food di spersi on of sol i d parti cl es i n a l i qui d.
Thi s work represents one of the few thorough studi es deal i ng wi th sl i p
i n a food product. The i nvesti gators found that the wal l sl i p correcti on
di d not si gni fi cantl y i nfl uence the fl ow behavi or i ndex (the average val ue
for appl esauce was 0.253), but i ncreased the consi stency coeffi ci ent: An
average consi stency coeffi ci ent equal to37.53 Pa s
n
was found for typi cal
appl esauce and the sl i p correcti on caused thi s val ue to i ncrease by an
average of 5% to a val ue of 39.40 Pa s
n
. Qui and Rao (1989) al so made
a very i nteresti ng observati on when they sai d "Due to the fortui tous
opposi te effects of correcti on for non-Newtoni an behavi or (i t i ncreases
the magni tudes of shear rates)and correcti on for sl i p effects (i t decreases
the magni tudes of shear rates), i t appears that for food suspensi ons
Newtoni an shear rates uncorrected for sl i p may be cl oser to the shear
u
s

1
2
c
f ()
d
+
(u
s
)
bob
R
b
+
(u
s
)
cup
R
c
R
c
/R
b
rates corrected for both non-Newtoni an behavi or and for wal l sl i p." They
cauti oned that thi s conjecture shoul d be veri fi ed before appl yi ng i t to
any parti cul ar product.
Secondary Flow. Equati ons devel oped for the anal ysi s of rheol ogi cal
data assume that the streaml i nes are ci rcul ar, i .e., fl ow i s l ami nar.
When an i nner cyl i nder rotates i n a concentri c cyl i nder system, the fl ui d
near the i nner surface tri es to move outward due to centri fugal forces.
Thi s movement may create non-streaml i ne fl ow due to the presence of
"Tayl or vorti ces"(G.I . Tayl or (1923. Phi l . Trans. Roy. Soc. (London), Ser.
A 223: 289). Such vorti ces may occur for Newtoni an fl ui ds when
(Whorl ow, 1992)
[3.90]
I n a Couette type system where the outer surface (the cup) i s rotated,
the i nerti al forces have a stabi l i zi ng effect and fl ow i s l ami nar at much
hi gher shear rates. Consul t Larson (1992)for a detai l ed anal ysi s of fl ow
i nstabi l i ti es i n concentri c cyl i nder systems. Appl i cati on of Eq. [3.90] i s
presented i n Exampl e Probl em 3.8.6 and 3.8.17.
Cavitation. The formati on and col l apse of vapor cavi ti es, known as
cavi tati on, may occur i n a hi gh shear envi ronment when the radi al
pressure drop i s suffi ci ent to cause parti al vapori zati on of the sampl e.
By consi deri ng the Bernoul l i equati on i n terms of the mechani cal energy
bal ance (Eq. [2.105]), one fi nds (Saki adi s, 1984)that cavi tati on wi l l occur
when where i s the l i near vel oci ty of the bob or cup,
whi chever i s greater. The l eft hand si de of the equati on ( ) i s rel ated
to the pressure drop ( ) across the gap: . I f present, cavi -
tati on may cause erroneous torque responses i n a concentri c cyl i nder
vi scometer. Cavi tati on i s not a si gni fi cant probl em i n food rheol ogy
because i t i s usual l y not present when l ami nar fl ow condi ti ons are
mai ntai ned. The cavi tati on probl em i s exami ned i n Exampl e Probl em
3.8.17.
3.6. Corrections: Cone and Plate, and Parallel Plate
Sources of error i n cone and pl ate, and paral l el pl ate systems are
si mi l ar and i ncl ude the fol l owi ng (Deal y, 1982): vi scous heati ng, sec-
ondary fl ow, shear rate nonuni formi ty due to l arge angl es, edge effects,
and non-i deal geometry (al so a probl em i n concentri c cyl i nder systems).
R
b
(R
c
R
b
)
> 41.3
R
c
R
c
R
b
u >
2(P
atm
P
vap
)/ u
u
u (2P/)
1/2
P
3.6 Corrections: Cone and Plate, and Parallel Plate 183
I t i s di ffi cul t, however, to numeri cal l y quanti fy these probl ems. Devi -
ati on from i deal geometry, i nvol vi ng eccentri ci ty or i ncorrect angl es,
may be moni tored by vi sual i nspecti on. The temperature i ncrease i n
cone and pl ate systems can be esti mated i n terms of the Bri nkman
number (Powel l , 1988) defi ned as the rate of heat generated by vi scous
di ssi pati on di vi ded by the rate of heat conducti on to the surface of the
fl ui d contai nment system. Vi scous heati ng i s rarel y a probl em when
testi ng bi ol ogi cal materi al s i n cone and pl ate, or paral l el pl ate, systems.
Edge Effects. One type of edge effect, sampl e ski n formati on from
dehydrati on, can be mi ni mi zed by appl yi ng a thi n coati ng of oi l on the
outer surface of the sampl e. Some rheometers come equi pped wi th a
sol vent trap to reduce l oss of vol ati l es and subsequent edge effects. A
second edge effect, known as edge fai l ure, may be observed wi th thi ck
foods. Even at rather l ow shear rates, the sampl e may appear to be
recessed at the center but fl owi ng out on the top and bottom surfaces of
the materi al . Typi cal l y, a sharp drop i n torque i s observed at the onset
of edge fai l ure. Thi s probl em may be the governi ng factor i n establ i shi ng
maxi mum shear rates i n cone and pl ate, and paral l el pl ate testi ng.
Slip. Sl i p correcti on methodol ogy for paral l el pl ate systems, based on
a compari son of shear stress versus shear rate at di fferent gap setti ngs,
has been presented by Yoshi mura and Prudhomme (1988). Thi s
anal ysi s i s si mi l ar to that presented i n Sec. 2.5 for eval uati ng a sl i p
correcti on factor for capi l l ary vi scometers. I n the absence of sl i p, pl ots
of torque ( ) versus the apparent shear rate at the ri m ( ), deter-
mi ned usi ng a si ngl e pl ate (constant radi us) but di fferent gap hei ghts,
wi l l yi el d i denti cal curves. Thi s fact i s very useful i n checki ng for sl i p
because data sets can be col l ected, on the same sampl e, by runni ng rate
sweeps at successi vel y smal l er val ues of .
Truncated Cones. Errors associ ated wi th truncated cones (Fi g. 3.9)
may be i nvesti gated as a devi ati on from i deal geometry. Fortunatel y,
the maxi mum error i ntroduced by the truncati on can be easi l y esti -
mated. Torque i n the coni cal secti on i s cal cul ated from Eq. [3.54] usi ng
, the radi us of the truncated porti on, as the l ower l i mi t of the i ntegral :
M R/h
h
R
T
Figure 3.9. Ideal and truncated cone design.
[3.91]
The percent maxi mum error i n torque i n the coni cal secti on i s
[3.92]
After substi tuti ng the appropri ate expressi ons, Eq. [3.55] for the i deal
torque and Eq. [3.91] for the approxi mate torque, Eq. [3.92] becomes
[3.93]
or, wi th si mpl i fi cati on,
[3.94]
R
R
Ideal Cone
Truncated Cone
T
M 2
R
T
R
r
2
dr (2/3)(R
3
R
T
3
)
% Max. Error

1
approximate torque
ideal torque
1
1
]
100
% Max. Error

1
(2/3)(R
3
R
T
3
)
(2/3)(R
3
)
1
1
]
100
% Max. Error

1
R
3
R
T
3
R
3
1
1
]
100
3.7 Mixer Viscometry 185
Usi ng Eq. [3.94], the maxi mum error can be cal cul ated. I f, for exampl e,
i s or , then the percent maxi mum error i s 0.1 and 0.8,
respecti vel y. I n actual operati on, the true error i s l ess than the predi cted
val ue because the truncated secti on of the cone al so contri butes some
torque duri ng measurement.
3.7. Mixer Viscometry
Extensi ve work has been conducted on mi xer vi scometry (Castel l -
Perez and Steffe, 1992). The techni que has been used mostl y for non-
reacti ng bi ol ogi cal materi al s but al so appl i ed i n eval uati ng the
workabi l i ty of fresh concrete (Tattersal l and Banfi l l , 1983), and to
chemorheol ogi cal studi es i nvol vi ng starch gel ati ni zati on (Dol an and
Steffe, 1990; Steffe et al ., 1989). Mi xer vi scometry may be useful to the
food engi neer i n eval uati ng di ffi cul t fl ui ds l i ke those exhi bi ti ng sl i p or
ti me-dependent behavi or, and those havi ng l arge parti cl es or parti cl e
settl i ng probl ems. Some concepts, such as the matchi ng vi scosi ty
method of determi ni ng the mi xer vi scometer constant, are al so useful
i n devel opi ng model s to si mul ate the shear hi story found i n compl ex
food processi ng equi pment such as scrape-surface heat exchangers.
An unusual type of mi xer vi scometer, known as the hel i cal screw
rheometer, was proposed by Krayni k et al . (1984) and successful l y used
for tomato products (Tamura et al ., 1989). The i nstrument consi sts of
a hel i cal screw i n a ti ght fi tti ng barrel and resembl es a si ngl e screw
extruder wi th a cl osed di scharge. Screw rotati on mai ntai ns parti cl e
suspensi on, and rheol ogi cal properti es are correl ated to pressure drop
over the l ength of the screw. The hel i cal screw rheometer i s not com-
merci al l y avai l abl e but has potenti al for future appl i cati ons i n on-l i ne
vi scometry.
I n past years, si gni fi cant advances have been made i n our under-
standi ng of commerci al mi xi ng and rel ated processes (Harnby et al .,
1985; Hol l and and Chapman, 1966; Nagata, 1975; Ol dshue, 1983;
Si l vester, 1985; Skel l and, 1983; Sweeney, 1978; Ul brecht and Patterson,
1985; Uhl and Gary, 1986). Ori gi nal i deas i n mi xer vi scometry came
from concepts devel oped toaddress i ndustri al probl ems soa short revi ew
of commerci al mi xi ng i s appropri ate before exami ni ng mi xer vi scometry.
Commercial Mixing. The terms mi xi ng, bl endi ng and di spersi ng are
someti mes used i nterchangeabl y. Mi xi ng can be defi ned as a uni t
operati on whi ch i nvol ves the i ntermi ngl i ng of two or more di ssi mi l ar
R
T
0.1R 0.2R
materi al s to obtai n a desi red degree of uni formi ty. I t i s usual l y
accompl i shed by mechani cal agi tati on whi ch creates moti on i n the
materi al bei ng processed.
I n mi xi ng, an agi tator i nduces materi al fl ow by i mparti ng i nerti al
forces to the fl ui d whi ch, i f not si gni fi cantl y dampened by vi scous forces,
cause fl ui d moti on at some di stance away from the i mpel l er. When
materi al s are so thi ck that fl ui d cannot be convected away from the
sti rrer, mi xi ng i s accompl i shed by the bul k movement of materi al due
to physi cal di spl acement by the agi tator.
Mi xer i mpel l ers for l ow vi scosi ty fl ui ds may be di vi ded i ntoaxi al and
radi al fl ow types. Wi th axi al fl ow i mpel l ers, such as the mari ne type
and fi xed bl ade turbi nes, top-to-bottom moti on i s promoted by pl aci ng
bl ades at an angl e of l ess than 90 degrees wi th the angl e of rotati on.
Radi al fl ow i mpel l ers, i ncl udi ng fl at and curved bl ade turbi nes, have
bl ades whi ch are mounted paral l el to the verti cal axi s of the dri ve shaft.
Anchor type agi tators (Fi g. 1.26) are commonl y used for hi gh vi scosi ty
materi al s. They may have a cl ose-cl earance between the i mpel l er and
the tank to enhance heat transfer. Extremel y hi gh vi scosi ty materi al s
i ncl udi ng pastes, dough, and meat emul si ons rel y on hel i cal ri bbons,
screws or kneaders for mi xi ng.
Dimensional Analysis. Mi xi ng i s a compl ex process that does not
l end i tsel f to ri gorous anal yti cal treatment. Studi es on the subject, and
most practi cal resul ts, have come from di mensi onal anal ysi s and si m-
i l ari ty theory (Gupta, 1984; Langharr, 1980; Murphy, 1950). The most
common techni que i nvol ves the Bucki ngham pi theorem di scussed
bel ow.
I f physi cal vari abl es i nfl uence a process, then the basi c equati on
rel ati ng the vari abl es may be wri tten as
[3.95]
The Bucki ngham pi theorem states that a rel ati onshi p may be found
between i ndependent di mensi onl ess groups of vari abl es, cal l ed pi
groups, whi ch have fewer terms than the basi c equati ons:
[3.96]
where wi th defi ned as
[3.97]
The pi groups are
m
f
1
(x
1
, x
2
, x
3
, x
4
, , x
m
) 0
f
2
(
1
,
2
,
3
, ,
i
) 0
i < m i
i m j
[3.98]
[3.99]
[3.100]
where , , and are constants determi ned from the
di mensi ons of the physi cal vari abl es. I n most cases, i s the number of
fundamental di mensi ons (such as l ength, mass and ti me) i nvol ved i n
the probl em. Eq. [3.96] i s determi ned from experi mental data. Reduci ng
the number of vari abl es by di mensi onal anal ysi s has an obvi ous
advantage of reduci ng the amount of data requi red to model mi xi ng
systems.
Power ConsumptioninFluidMixing. I n fl ui d mi xi ng, di mensi onal
anal ysi s shows that the power number i s a functi on of many di men-
si onl ess vari abl es:
[3.101]
where:
[3.102]
[3.103]
[3.104]
[3.105]
[3.106]
The power i s equal tothe product of torque and angul ar vel oci ty: .
and are the vi scosi ty functi on (Eq. [1.22])and the fi rst normal stress
coeffi ci ent (Eq. [1.23]), respecti vel y. Numerous geometri cal di men-
si onl ess numbers may be consi dered for a typi cal mi xer (Fi g. 3.10): ,
, , , , , wi dth of baffl e di vi ded by the i mpel l er
di ameter, number of i mpel l er bl ades, i mpel l er pi tch, and number of
1
x
1
a
1
x
2
a
2
x
3
a
3
x
m
a
m
2
x
1
b
1
x
2
b
2
x
3
b
3
x
m
b
m
3
x
1
c
1
x
2
c
2
x
3
c
3
x
m
c
m
a
1
. a
m
b
1
. b
m
c
1
. c
m
j
N
Po
f (N
Re, I
, N
Fr
, N
We
, N
Wi
, Geometric Dimensionless Numbers)
N
Po

P
3
d
5
N
Re, I

d
2
N
Fr

2
d
g
N
We

2
d
3
st
N
Wi

P M

1
d/D
(Z
1
+ Z
2
+ h)/d Z
1
/d Z
2
/d h/d W/d
baffl es. To be consi stent wi th the previ ous work on rotati onal vi sco-
metry, the angul ar speed i s gi ven i n uni ts of rad/s i nstead of rev/s or
revs/mi n whi ch are frequentl y used i n i nvesti gati ng commerci al mi xi ng.
Figure 3.10. Typical mixer viscometer apparatus with paddle type impeller and
water jacket for temperature control.
I n geometri cal l y si mi l ar systems, Eq. [3.101] may be si mpl i fi ed to
[3.107]
where the power number ( ) represents the rati o of the appl i ed force
to the opposi ng i nerti al force. The i mpel l er Reynol ds number ( )
represents the rati o of the i nerti al to the opposi ng vi scous force i n the
mi xi ng system and defi nes l ami nar, transi ti onal , and turbul ent fl ow i n
the mi xi ng vessel . I t i s general l y accepted that l ami nar fl ow exi sts for
and turbul ent fl ow i s assured at . [Note: i f i s
formul ated wi th speed expressed as rev/s i nstead of rad/s, l ami nar and
turbul ent fl ows are found at and , respecti vel y.
Di fferent uni ts, rev/s versus rad/s, cause the numbers to be di fferent by
h
D
Z
d
W
1
Z
2
water jacket
N
Po
f (N
Re, I
, N
Fr
, N
We
, N
Wi
)
N
Po
N
Re, I
N
Re, I
< 63 N
Re, I
> 63, 000 N
Re, I
N
Re, I
< 10 N
Re, I
> 10, 000
a factor or .] The transi ti on regi on i s l arge and depends on the
parti cul ar system i n questi on. I n some cases, turbul ent fl ow may be
present wi th an i mpel l er Reynol ds number as l ow as 1900.
The Froude number ( )refl ects the rati o of i nerti al to gravi tati onal
forces and i s used to account for the effect of vortexi ng on the power
number. I t may be i mportant i n unbaffl ed systems operati ng at hi gh
i mpel l er Reynol ds numbers but can be i gnored at l ow speed, and i n most
baffl ed systems. The Weber number ( ) represents the rati o of the
i nerti al force to surface tensi on force. I t shoul d be consi dered i n systems
where i nterfaci al effects are i mportant such as those found i n two phase
di spersi ons. Vi scoel asti c behavi or i s characteri zed by the Wei ssenberg
number ( )defi ned as the rati oof the pri mary normal stress coeffi ci ent
(Eq. [1.23]) ti mes the angul ar vel oci ty, di vi ded by the apparent vi scosi ty
functi on (Eq. [1.22]). The i nfl uence of l i qui d el asti ci ty on power con-
sumpti on i s not cl ear but thought to be smal l (Ul brecht and Carreau,
1985).
Wi th a si ngl e phase fl ui d operati ng at l ow speed, or i n a baffl ed
system, the power number can be expressed wi th the i mpel l er Reynol ds
number al one:
[3.108]
or
[3.109]
The general functi onal rel ati onshi p may be stated as
[3.110]
where and depend on the geometry of the system and the fl ow regi me
present duri ng mi xi ng. when , and when
. The val ues of and i n the i ntermedi ate regi on wi l l
depend on the parti cul ar mi xi ng system under consi derati on.
Hol l and and Chapman (1966) provi de power versus Reynol ds
number sol uti ons for a wi de range of mi xi ng systems. Rel ati onshi ps for
some standard mi xers are provi ded i n Saki adi s (1984). Power curves
are i ndependent of scal e and depend onl y on the geometry of the system.
When the power curve i s avai l abl e for a parti cul ar confi gurati on, i t may
2
N
Fr
N
We
N
Wi
N
Po
f (N
Re, I
)
P
3
d
5
f

d
2
,
P
3
d
5
A
d
2
,
B
A B
B 1 N
Re, I
< 63 B 0
N
Re, I
> 63, 000 A B
be used to cal cul ate the power requi rement gi ven vari ous agi tator
speeds, l i qui d vi scosi ti es, and densi ti es. Other references (Ul brecht and
Carreau, 1985; Nagata, 1975) may al so be consul ted for power con-
sumpti on i nformati on rel ated to mi xi ng wi th hel i cal screws, ri bbons,
and draught tubes.
3.7.1. Mixer Viscometry: Power Law Fluids
Mi xer vi scometry, typi cal l y conducted usi ng fl ag or pi tched paddl e
i mpel l ers (Fi g. 3.11), has been used extensi vel y for power l aw fl ui ds
(Castel l -Perez and Steffe, 1990; Castel l -Perez and Steffe, 1992).
Techni ques have been devel oped for esti mati ng power l aw fl ui d prop-
erti es that are useful for qual i ty control and engi neeri ng desi gn appl i -
cati ons. Shear rate esti mati on methods have been devel oped to al l ow
fl ui d properti es to be cal cul ated over an appropri ate shear rate range.
They are al so useful i n determi ni ng the degree of mi xi ng acti on present
when eval uati ng process performance characteri sti cs i nvol vi ng parti cl e
moti on and gas di spersi on.
Power Consumption and AverageShear Rate
Mi xer vi scometry i s an i mportant tool for i nvesti gati ng power l aw
fl ui ds. Worki ng equati ons are devel oped by fi rst consi deri ng a New-
toni an fl ui d i n l ami nar fl ow ( ). When , and may be
negl ected (surface tensi on, el asti c, and vortexi ng effects are
i nsi gni fi cant), the power consumpti on equati on (Eq. [3.110]) may be
wri tten as
[3.111]
or, i ncorporati ng Eq. [3.102] and [3.103], as
[3.112]
Eq. [3.112] may be used for power l aw fl ui ds ( ) i f the Newtoni an
fl ui d vi scosi ty ( ) i s repl aced by an apparent vi scosi ty ( )
eval uated at an average shear rate ( ) defi ned as
[3.113]
where i s the mi xer vi scometer constant havi ng uni ts of 1/rad (uni ts
of 1/rev are found i n some publ i shed works). i s uni que for any
parti cul ar physi cal system and must be determi ned from experi mental
data.
N
Re, I
< 63 N
We
N
Wi
N
Fr
N
Po

A
N
Re, I
P
d
5

A
d
2
n
K(
a
)
n 1
a
k
k
k
3.7.1 Mixer Viscometry: Power Law Fluids 191
I n mi xer vi scometry, there are two pri mary techni ques for deter-
mi ni ng : the "sl ope method" and the "vi scosi ty matchi ng method."
These methods are presented i n the fol l owi ng di scussi on.
Figure 3.11. Typical impellers used in mixer viscometry.
Evaluation of
SlopeMethod. Substi tuti on of Eq. [3.113], and the power l aw apparent
vi scosi ty for the Newtoni an vi scosi ty, i ntoa si mpl i fi ed form of Eq. [3.112]
gi ves
[3.114]
or, after si mpl i fyi ng and taki ng l ogari thms,
[3.115]
Usi ng power l aw fl ui ds wi th di fferent known val ues of and , a pl ot
of versus i s used to determi ne from the sl ope
of the l i ne whi ch i s equal to . I f the pl ot i s a strai ght l i ne, then
the approxi mati on proposed ( ) i n Eq. [3.113] i s val i d.
k
0.127 m
.01 m dia
.008 m dia
15 degree pitch blade
0.04143 m
0.02692 m
Haake Pitched Paddle
Flag
Paddle
k
P
d
3
2
A AK(
a
)
n 1
AK(k)
n 1
log
10
P
K
n + 1
d
3
_
,
log
10
(A) (1 n) log
10
(k)
K n
log
10
(P/[K
n + 1
(d)
3
]) (1 n) k
log
10
(k)
a
k
One must exerci se care i n usi ng the sl ope method for cal cul ati ng the
mi xer vi scometer constant. Smal l errors i n determi ni ng the sl ope of the
l i ne wi l l resul t i n l arge errors i n because of the l ogari thmi c rel a-
ti onshi p between the two numbers. By conventi on, base 10 l ogari thms
were used i n Eq. [3.115]: I denti cal resul ts woul d be obtai ned usi ng base
. Determi nati on of usi ng the sl ope method i s i l l ustrated i n Exampl e
Probl em 3.8.18.
MatchingViscosityMethod. Thi s techni que i nvol ves the compari son
of power curves for Newtoni an and non-Newtoni an fl ui ds usi ng the i dea
of matchi ng vi scosi ti es. The phrase "matchi ng vi scosi ti es"refers to the
assumpti on that the average shear rate for a non-Newtoni an fl ui d i s
equal to the average shear rate for a Newtoni an fl ui d when the New-
toni an vi scosi ty equal s the apparent vi scosi ty of the non-Newtoni an
fl ui d. The techni que i s excel l ent for determi ni ng the average shear rate
i n a mi xer and i s al so useful to food process engi neers i n eval uati ng the
performance of commerci al equi pment (e.g., a scraped-surface heat
exchanger) havi ng poorl y defi ned shear fi el ds.
Usi ng Newtoni an fl ui ds such as si l i cone oi l or corn syrup i n the mi xer
vi scometer, the constant i s determi ned from the sl ope of experi mental
data presented i n the form of Eq. [3.111]. Wi th a routi ne vi scometer
(l i ke a concentri c cyl i nder or cone and pl ate system) and tradi ti onal
rheol ogi cal techni ques, the properti es ( and ) of a power l aw fl ui d are
determi ned. Thi s "standard"(or reference) power l aw materi al i s then
pl aced i n the mi xer vi scometer, and mi xed at a constant speed. i s
determi ned and the resul ti ng "average vi scosi ty" cal cul ated usi ng the
power number expressi on for the Newtoni an fl ui ds. Thi s equati on,
found by combi ni ng Eq. [3.111] and [3.103], i s
[3.116]
Next, the matchi ng vi scosi ty assumpti on i s appl i ed and an average shear
rate ( ) cal cul ated. Fi rst, assume,
[3.117]
then, substi tute Eq. [3.117] i nto Eq. [3.116],
[3.118]
k
e k
A
K n
N
Po

(d
2
) (N
Po
)
A
a
K(
a
)
n 1
K(
a
)
n 1
(d
2
) (N
Po
)
A
Figure 3.12. Matching viscosity method to calculate average shear rate.
and sol ve for the average shear rate:
[3.119]
The matchi ng vi scosi ty procedure to eval uate the average shear rate
i s summari zed i n Fi g. 3.12:
Step 1. Power number rel ati onshi p i s establ i shed for Newtoni an
fl ui ds and the numeri cal val ue of i s determi ned usi ng Eq.
[3.111];
Po
N
1
2
NNN
Re,I
3
Newtonian
Non-Newtonian
4
=
d
2
5 =
= K
n-1
a
a
=
K
6
K,n
Po
N
NNN
Po
.
.
.
A
1/(n-1)
d
2
(N
Po
)
KA
1
1
]
1/(n 1)
A
Step 2. At a steady i mpel l er speed, the power number i s measured
for a non-Newtoni an fl ui d;
Step 3. Newtoni an and non-Newtoni an power numbers are set equal
and the correspondi ng i mpel l er Reynol ds number, from the
Newtoni an fl ui d data, i s eval uated;
Step 4. "Vi scosi ty"i s cal cul ated from the i nformati on generated i n
Step 3 usi ng Eq. [3.116];
Step 5. Matchi ng vi scosi ty assumpti on i s appl i ed so the Newtoni an
and apparent vi scosi ti es are set equal as i ndi cated i n Eq. [3.117];
Step 6. Taki ng the apparent vi scosi ty and power l aw fl ui d param-
eters, average shear rate i s cal cul ated usi ng Eq. [3.119] wri tten
i n terms of where .
i s cal cul ated as . Steps 2-6 shoul d be repeated at numerous
i mpel l er speeds to determi ne how may vary wi th . An average val ue
of may be taken to represent the constant requi red i n Eq. [3.113].
Comparison of Calculation Methods. There are vari ous
advantages and di sadvantages toconsi der before usi ngthe sl ope method
(SM) or the matchi ng vi scosi ty method (MVM). SM and MVM both
requi re power l aw fl ui d "standards" (reference fl ui ds) to determi ne
that, wi th an appropri ate sel ecti on of fl ui ds, wi l l cover a sui tabl e range
of fl ow behavi or i ndi ces. Fewer reference fl ui ds may be needed usi ng
the MVM. The mai n advantage of the SM i s the si mpl i ci ty of the cal -
cul ati ons; however, probl ems wi th accuracy may occur due to the hi gh
sensi ti vi ty of wi th the sl ope of a l ogari thmi c pl ot. I n compari son to
SM, the MVM i s somewhat l abori ous due to the l arger amount of data
handl i ng. Resul ts from the SM and MVM are si mi l ar. The effecti ve
shear rates of fl ag and star i mpel l ers were eval uated usi ng both methods
by Rao and Cool ey (1984). Resul ts from that work showed the two
methods to be i n good agreement. Bri ggs (1995) reached a si mi l ar
concl usi on worki ng wi th fl ag i mpel l ers. MVM woul d be the method of
choi ce for engi neers attempti ng to eval uate average shear rates i n
compl ex processi ng equi pment.
Values and Variables Influencing . Skel l and (1983) has sum-
mari zed val ues of mi xer vi scometer constants ( ) and power curves for
many commerci al mi xers. Usi ng the average shear rate concept to
cal cul ate an apparent vi scosi ty val ue, Nagata (1975) determi ned power
correl ati ons of Bi ngham pl asti c materi al s for turbi ne, anchor, and ri b-
d
2
N
Po
/A
k

a
/
k
k
k
k
k
k k
k
bon mi xers. Bowen (1986) has presented an i nformati ve di scussi on
rel ated to the determi nati on of average and maxi mum shear rates for
radi al fl ow turbi ne mi xers. A mi xer vi scometer constant has been
determi ned for the Brabender Vi scograph by Wood and Goff (1973), and
for a fl ag i mpel l er rotated i n a number 303 can by Rao (1975).
Usi ng the sl ope method, Steffe and Ford (1985) found = 4.47 rad
-1
for the Haake MV cup and pi tched paddl e i mpel l er (Fi g. 3.11). The
pi tched paddl e i mpel l er can be parti cul arl y useful i n mai ntai ni ng par-
ti cl e suspensi on i n sampl es exhi bi ti ng sedi mentati on probl ems. Raw
data, and anal yti cal procedures for the determi nati on of , used by
Steffe and Ford (1985) are presented i n Exampl e Probl em 3.8.18.
Castel l -Perez and Steffe (1990) studi ed shear rate eval uati on usi ng
paddl e i mpel l ers (Fi g. 3.10)for power l aw fl ui ds. The effects of numerous
factors on the determi nati on of were consi dered. Resul ts showed that
was hi gher wi th l ess shear-thi nni ng fl ui ds, i ncreases wi th a
decrease i n rotati onal speed and reaches an al most constant val ue when
operati ng at speeds greater that 0.33 rev/s (20 rev/mi n), and tends to
decrease wi th an i ncrease i n the rati o of the i mpel l er to cup di ameter
( ). The effort resul ted i n a number of recommendati ons for practi cal
work i n mi xer vi scometry i nvol vi ng paddl e-type i mpel l ers: i f possi bl e,
sel ect a system wi th a smal l gap, > 0.709; i f the system to be used
has a l arge gap, determi ne the effect of fl ui d properti es on ; and use a
mi ni mum i mpel l er speed of 0.33 rev/s (20 rev/mi n).
Bri ggs (1995) found val ues for the fl ag i mpel l er descri bed i n Fi g.
3.13. Fl ui d sampl es were hel d i n a Brookfi el d (Brookfi el d Engi neeri ng
Laboratori es, Stoughton, MA) smal l sampl e cup havi ng a hei ght of 6.48
cm, and an i nsi de di ameter of 1.90 cm maki ng . Data were
col l ected over speeds rangi ng from 0.5 to 100 rev/mi n (0.5 to 10.5 rad/s).
Both the sl ope and matchi ng vi scosi ty methods were found to produce
si mi l ar resul ts for shear-thi nni ng fl ui d foods. The sl ope anal ysi s pro-
duced a constant val ue of the mi xer vi scometer constant: rad
-1
.
Thi s val ue (and )i s recommended for routi ne anal ysi s. The smal l
sampl e cup and fl ag i mpel l er provi de a si mpl e, yet powerful , tool for the
rheol ogi cal i nvesti gati on of fl ui d foods.
Determination of Non-Newtonian Fluid Properties
Evaluating versus . The rel ati onshi p between power
and angul ar vel oci ty for a power l aw fl ui d was establ i shed i n Eq. [3.114]:
k
k
k
k k
k
d/D
d/D
k
k
d/D 0.79
k 2.92
A 4.84
log
10
(M) log
10
()
Figure 3.13. Flag impeller used by Castell-Perez et al. (1993) and Briggs
(1995).
Substi tuti ng , and sol vi ng for the torque yi el ds
[3.120]
or, i n terms of l ogari thms,
[3.121]
The fl ow behavi or i ndex can be found as the sl ope of versus
. The constant ( ) i s not requi red for the deter-
mi nati on of : I t may be used to esti mate , i f and are known. The
val ue of , however, may be found wi thout by appl yi ng Eq. [3.120] to
two fl ui ds, one wi th unknown properti es (fl ui d i ndi cated wi th subscri pt
) and a reference fl ui d wi th known properti es (fl ui d i ndi cated wi th
subscri pt ):
[3.122]
h
d
blade length
blade shaft diameter
3.0
1.5
0.5
0.5
Flag Impeller
d
e
1.0
[ dimensions in cm ]
0.1
P
d
3
2
AK(k)
n 1
P M
M d
3
AK(k)
n 1
n
log
10
(M) log
10
(d
3
AK(k)
n 1
) + n log
10
()
log
10
(M)
log
10
(d
3
AK(k)
n 1
) log
10
()
n K A k
K A
x
y
M
x
d
3
AK
x
(k)
n
x
1
x
n
x
and
[3.123]
Di vi di ng by , Eq. [3.122] by Eq. [3.123], gi ves
[3.124]
whi ch may be sol ved for :
[3.125]
The properti es of the known fl ui d ( and ) are determi ned usi ng
conventi onal rheol ogi cal methods. i s found as the sl ope of Eq. [3.121]
and i s determi ned usi ng the sl ope or matchi ng vi scosi ty method.
Under some condi ti ons the cal cul ati on of may be greatl y si mpl i fi ed:
I f and i s cl ose to , then i s approxi matel y . Mi xer
vi scometry techni ques di scussed above are used to determi ne power l aw
fl ui d properti es i n Exampl e Probl em 3.8.18.
Applying the Matching Viscosity Assumption. Power l aw fl ui d
properti es may al so be esti mated usi ng the matchi ng vi scosi ty
assumpti on. An average apparent vi scosi ty i s cal cul ated usi ng Eq.
[3.116] at an average shear rate defi ned by Eq. [3.113], . Then,
fl ui d properti es are found by regressi on anal ysi s of wi th an
appropri ate set of experi mental data. Thi s techni que i s i l l ustrated i n
One advantage of usi ng the matchi ng vi scosi ty assumpti on i s that
i t easi l y al l ows the i nvesti gati on of a wi de array of non-Newtoni an
behavi or. Al though the average shear rate determi ned usi ng power l aw
fl ui ds can be consi dered "exact" for these materi al s, i t may al so be
consi dered a shear rate approxi mati on for other types of fl ui ds. Thi s i s
anal ogous to the way i n whi ch Eq. [3.32], the Newtoni an approxi mati on,
may be used to esti mate the shear rate of a non-Newtoni an fl ui d i n a
concentri c cyl i nder apparatus.
M
y
d
3
AK
y
(k)
n
y
1
y
n
y
M
x
M
y
M
x
M
y
K
x
x
n
x
(k)
n
x
K
y
y
n
y
(k)
n
y
K
x
K
x
K
y
M
x
M
y
_
y
n
y
(k)
n
y
x
n
x
(k)
n
x
_
,
K
y
n
y
n
x
k
K
x
y

x
n
y
n
x
K
x
K
y
M
x
/M
y
a
k
K(
a
)
n 1
UsingtheConcentric Cylinder Analogy. Fl ui d properti es may be
determi ned from shear stress and shear rate approxi mati ons devel oped
from a concentri c cyl i nder anal ogy. Mi xi ng systems (i denti cal to the
one i l l ustrated i n Fi g. 3.10) usi ng di fferent cyl i ndri cal cups and paddl e
i mpel l ers were consi dered by Castel l -Perez et al . (1991) i n devel opi ng
rheograms for power l aw fl ui ds. The research assumed that mi xers
were anal ogous to concentri c cyl i nder systems. Thi s work showed that
the average shear stress and average shear rate coul d be esti mated from
geometri cal parameters and the fl ow behavi or i ndex usi ng the fol l owi ng
equati ons:
[3.126]
and
[3.127]
for and wi th and
. The fl ow behavi or i ndex i s found as the sl ope of Eq. [3.121].
Castel l -Perez et al . (1993) determi ned shear stress and shear rate
esti mates for power l aw fl ui ds usi ng a fl ag i mpel l er (Fi g. 3.13):
[3.128]
and
[3.129]
where , the equi val ent di ameter, i s equal to the l ength of the i mpel l er
bl ade pl us the di ameter of the bl ade shaft. Di mensi ons of the i mpel l er
used i n the Castel l -Perez et al . (1993) study are gi ven i n Fi g. 3.13: The
equi val ent di ameter for that geometry i s equal to 1.0 cm. Bl ade
thi ckness was 0.1 cm. Val ues of were l i mi ted to the range of 0.27
to 0.59; however, the i nfl uence of di ameter was mi ni mal i n that range
maki ng i t acceptabl e to assume i n Eq. [3.129]. Wi th thi s
assumpti on, i t i s possi bl e to obtai n a reasonabl e esti mate of the average
d
3
2
h
d
+
1
3
1
1
]
1
1
]
1
M
a
2
(D/d)
(2 n)/n
(D/d)
2n
1
1
1
]
d
h
_
,
n/2
d/D 0.709 0.36 < d/h < 1.80 1.0 cm h 5.0 cm
0.5 n 0.9
a

2M
d
e
2
h
a
2
D
d
_
,
n/2
d
e
d/D
d/D 1
3.7.2 Mixer Viscometry: Bingham Plastic Fluids 199
shear rate wi thout knowi ng the fl ow behavi or i ndex. The greatest l evel
of accuracy, however, i s obtai ned usi ng the fl ow behavi or i ndex cal cu-
l ated from the rel ati onshi p gi ven i n Eq. [3.121].
3.7.2. Mixer Viscometry: Bingham Plastic Fluids
Power l aw fl ui ds have been studi ed extensi vel y usi ng mi xer vi sco-
metry, but l i ttl e work has been done wi th materi al s exhi bi ti ng a yi el d
stress (Castel l -Perez and Steffe, 1992). Nagata (1975) i nvesti gated
power consumpti on i n mi xi ng Bi ngham pl asti cs and some success has
been reported usi ng mi xer vi scometry methods on fresh concrete
showi ng Bi ngham pl asti c behavi or (Tattersal l and Banfi l l , 1983).
Determination of Fluid Properties. Usi ng the Newtoni an rel a-
ti onshi p between and , assumi ng , and usi ngthe defi ni ti on
of apparent vi scosi ty for a Bi ngham pl asti c fl ui d ( ), yi el ds
[3.130]
Substi tuti ng the defi ni ti on of (Eq. [3.102]) i nto Eq. [3.130], recog-
ni zi ng that , then si mpl i fyi ng the resul ti ng equati on gi ves an
expressi on rel ati ng torque and speed i n a mi xer vi scometer:
[3.131]
whi ch, after addi ti onal mani pul ati on, becomes
[3.132]
Col l ecti ng mi xer vi scometer data of torque versus angul ar vel oci ty for
a Bi ngham pl asti c and pl otti ng the resul t wi l l provi de a sl ope ( )
and an i ntercept ( ) that refl ect the rheol ogi cal properti es of the
fl ui d. I f and are known, pl asti c vi scosi ty and the yi el d stress can
be cal cul ated.
Evaluation of . Theoreti cal l y, can be determi ned wi th the same
matchi ng vi scosi ty techni que di scussed for power l aw fl ui ds. Usi ng the
rel ati onshi p for Newtoni an fl ui ds, Eq. [3.111], i s determi ned from data
i nvol vi ng standard Newtoni an materi al s. Taki ng data wi th a tradi -
N
Po
N
Re, I

a
k

o
/
+
pl
N
Po

A
N
Re, I
A
d
2

A
d
2

A(
o
/
a
+
pl
)
d
2

A(
o
/(k) +
pl
)
d
2
N
Po
P M
M
d
5
o
kd
2
+

pl
d
2
,
M
Ad
3
o
k
+ Ad
3
pl
Ad
3
pl
Ad
3
o
/k
A k
k k
A
ti onal i nstrument, such as a concentri c cyl i nder vi scometer, the prop-
erti es of a "standard"(or reference)Bi ngham pl asti c fl ui d are eval uated.
Thi s materi al i s pl aced i n a mi xer vi scometer and mi xed at constant
speed ( ) al l owi ng determi nati on of . Usi ng that i nformati on, the
"vi scosi ty"i s cal cul ated from Eq. [3.116]:
[3.133]
The matchi ng vi scosi ty assumpti on ( ) i s appl i ed:
[3.134]
and the average shear rate determi ned:
[3.135]
Usi ng Eq. [3.113], i s found as . I f the behavi or of di fferent reference
fl ui ds i s eval uated, the i nfl uence of angul ar vel oci ty and fl ui d properti es
on can be i nvesti gated.
The presence of a yi el d stress i n a mi xer means there wi l l be areas
i n the system where . Thi s creates pl ug fl ow and "dead spots"that
may adversel y i nfl uence resul ts. Work by Nagata et al . (1970) i ndi cates
that may be a functi on of the yi el d stress.
3.7.3. Yield Stress Calculation: Vane Method
Yi el d stress can be determi ned usi ng the same basi c equi pment
requi red i n mi xer vi scometry. I n the vane method, the stress to i ni ti ate
fl ow from a vane i mmersed i n test materi al i s measured (Nguyen and
Boger, 1985). Vane and vessel di mensi ons (Fi g. 3.14)shoul d stay wi thi n
speci fi ed l i mi ts (Steffe, 1992): ; ; or
i f the vane i s compl etel y i mmersed i n the sampl e; where i s
the di ameter of the contai ner i f ci rcul ar, or the mi ni mum crossecti onal
di mensi on i f some other shape i s used. Vanes wi th 4 (Nguyen and Boger,
1985), 6 and 8 bl ades (Qi u and Rao, 1988) have been used. Al l produce
si mi l ar resul ts.
N
Po

d
2
N
Po
A

o
/
a
+
pl

d
2
N
Po
A
a

o
d
2
N
Po
A

pl
_
,
1
k

a
/
k
<
o
k
1.5 h/d 4.0 Z
2
/d 0.5 Z
1
0.0 Z
1
/d 1.0
D/d 2.0 D
3.7.3 Yield Stress Calculation: Vane Method 201
I f you assume the test materi al yi el ds al ong a cyl i ndri cal surface
(shaft excl uded), then the total torque ( ) to overcome the yi el d stress
of the fl ui d i s
[3.136]
where i s the shear stress on the end surfaces (top and bottom).
Si mpl i fi cati on of Eq. [3.136] yi el ds
[3.137]
Assume vari es wi th the radi us accordi ng to a power rel ati onshi p:
[3.138]
where i s a constant. Thi s assumpti on i s somewhat arbi trary, but
reasonabl e. Substi tuti ng Eq. [3.138] i nto [3.137] yi el ds
[3.139]
whi ch, wi th si mpl i fi cati on, gi ves
Figure 3.14. Vane, with 4 blades, and vessel for yield stress determination.
M
o
M
o
( dh)
d
2
_
o
+ 2
0
d/ 2
2 r
2
e
dr
e
M
o

hd
2
2

o
+ 4
0
d/ 2
r
2
e
dr
e
f (r)

2 r
d
_
,
m
o
m
M
o

hd
2
2

o
+ 4
0
d/ 2
r
2
2r
d
_
,
m
o
dr
d
h
z
z
1
2
D
h
[3.140]
or, i f sol ved for the yi el d stress:
[3.141]
Errors i n usi ng Eq. [3.141] for decrease wi th l arger val ues of
. I f , errors l ess than or equal to 3.7 percent may be obtai ned
when (Steffe, 1992). The assumpti on that i s usual l y
sati sfactory. I t i s certai nl y acceptabl e when maki ng a qual i ty control
compari son between products or when si mpl e, qui ck sol uti ons are
needed. To el i mi nate error due to the upper end effect, the top surface
of the vane may be pl aced even wi th the fl ui d ( , Fi g. 3.14) gi vi ng
(assumi ng ) the fol l owi ng torque equati on at the yi el d poi nt:
[3.142]
Thi s expressi on was used by Yoshi mura et al . (1987) to eval uate the
yi el d stresses of vari ous non-food emul si ons.
Error associ ated wi th the stress di stri buti on over the ends can be
avoi ded i f Eq. [3.137] i s used to determi ne the yi el d stress: A pl ot of
versus i s made usi ng data col l ected from vanes havi ng the same
di ameter but di fferent l engths. The yi el d stress i s cal cul ated from the
sl ope of the resul ti ng l i ne whi ch i s equal to . Thi s procedure
shoul d be used i f there i s any doubt regardi ng the val i di ty of the
assumpti on that or when more preci se val ues of the yi el d stress
are needed.
When measuri ng the stati c yi el d stress (di scussed i n Sec. 1.6), vanes
shoul d be careful l y pl aced i n the sampl e to mi ni mi ze di srupti on to
surroundi ng materi al . I n qual i ty control work, i t may be possi bl e to use
the food contai ner i tsel f as the test vessel tomi ni mi ze sampl e di srupti on.
When measuri ng the dynami c yi el d stress, sampl e di srupti on i s not a
probl em because the weak structure i s al ready destroyed from pri or
handl i ng operati ons such as mi xi ng or pumpi ng. Thi s assumpti on woul d
be i nval i d i f the materi al was abl e to rapi dl y rebui l d i nternal structure.
Rapi d redevel opment of structure i s unusual i n food products.
M
o

d
3
2
h
d
+
1
m + 3
_
o

2M
o
d
3
h
d
+
1
m + 3
_
,
1
m > 1
h/d m 1
h/d > 2 m 0
Z
1
0.0
m 0
M
o

d
3
2
h
d
+
1
6
_
o
M
o
h
o
d
2
/2
m 0
Figure 3.15. Typical torque time response when operating the vane in the con-
trolled rate mode.
Modes of Operation. The vane may be operated i n two modes when
col l ecti ng experi mental data: control l ed rate or control l ed stress. A
typi cal curve for testi ng i n the control l ed rate mode (Fi g. 3.15) shows a
steady i ncrease i n torque up to a peak val ue ( ) fol l owed by a gradual
decl i ne unti l reachi ng an equi l i bri um l evel . I n a sampl e compl etel y
broken down pri or to testi ng, the peak torque woul d represent the
dynami c yi el d stress. I n an undi sturbed sampl e taken from storage,
the peak torque woul d represent the stati c yi el d stress; however, i n thi s
type of control l ed rate test, some weak structural bonds may be di s-
rupted before the peak torque i s reached perhaps causi ng the measured
val ue to be l ower than that found i n a control l ed stress experi ment.
I n the control l ed rate mode, vari ous speeds of operati on have been
consi dered. Speeds as l ow as possi bl e shoul d be used, preferabl y no
hi gher than 1.0 rpm (Steffe, 1992). Vel oci ti es of greater magni tude wi l l
al ter resul ts for many products. The angl e of rotati on at whi ch yi el d
occurs, needed to determi ne the strai n at fai l ure, depends on many
factors: type of materi al bei ng tested, si ze of the vane, angul ar vel oci ty
of the vane, and the wi nd-up characteri sti cs of the vi scometer. Cal cu-
l ati ng yi el d stress usi ng the vane method i n the control l ed rate mode i s
Time at Constant Speed
T
o
r
q
u
e
M
o
M
o
i nvesti gated i n Exampl e Probl ems 3.8.19 and 3.8.20. Wi nd-up
characteri sti cs of rotati onal vi scometers are di scussed l atter i n thi s
secti on.
Vanes can al so be operated i n the control l ed stress mode. I n
experi mentati on, the torque (or stress) i s i ncreased unti l fl ow i s
observed. Data may show the angl e of rotati on (rel ated to strai n) as a
functi on of ti me wi th stepped i ncreases i n torque. When the yi el d stress
i s exceeded, the strai n wi l l i ncrease rapi dl y wi th ti me. I n a typi cal pl ot,
the torque at yi el d ( ) i s between torque l evel s M4 and M5 (Fi g. 3.16).
Figure 3.16. Typical response curve when operating the vane in the controlled
stress mode.
Testi ng i n the control l ed stress mode has the i mpl i ci t assumpti on
(whi ch i s easi l y tested experi mental l y) that resul ts are not ti me-
dependent so the amount of ti me spent at each stress l evel has no
i nfl uence on the outcome. Fi g. 3.16 shows changes i n strai n i ncreasi ng
sl i ghtl y (but remai ni ng constant) wi th each stepped i ncrease i n torque
bel ow the fai l ure torque. I n a ti me-dependent materi al the pl ot may
exhi bi t a gradual i ncrease i n strai n, i n addi ti on to the step i ncrease,
duri ng the hol di ng peri od at torque l evel s M1 through M4 (Fi g. 3.16).
M
o
S
t
r
a
i
n
Time at Constant Torque (or Stress)
M1 M2 M3 M4 M5
M1
M2
M3
M4
M5
increasing levels of torque
M4 < M < M5
o
Control l ed stress rheometers may al l ow the appl i cati on of a ramped
(l i near) i ncrease i n stress. Thi s wi l l produce a strai n response that i s
uni que for the materi al bei ng tested. Resul ts may be pl otted to show
changes i n strai n over ti me, or the correspondi ng changes i n strai n wi th
torque or stress (Fi g. 3.17). Behavi or i s very di fferent dependi ng on
whether or not the materi al i s demonstrati ng sol i d or fl ui d-l i ke behavi or.
There woul d be a l i near rel ati onshi p between stress and strai n wi th
Hookean behavi or. When the yi el d stress i s exceeded, the onset of fl ow
causes a dramati c change i n strai n. I f the strai n response (Fi g. 3.17) i s
consi dered i n terms of two curves, one for the case where and the
other for , the poi nt of i ntersecti on may be defi ned as the peak
torque ( ) requi red i n the yi el d stress cal cul ati on. The actual yi el d
stress woul d be found at the i ntersecti on poi nt i f stress, i nstead of torque,
i s pl otted. When a break poi nt i s evi dent, thi s i s a conveni ent method
of determi ni ng the yi el d stress from control l ed stress data. Logari thmi c
pl ots may be useful i n ampl i fyi ng di fferences.
Figure 3.17. Strain-stress curve showing the location of the yield stress when a
linear increase in stress is applied to the sample.

o
>
o
M
o
S
t
r
a
i
n
Torque or Stress
o
o
o
solid behavior
fluid behavior
M
o
or
The i dea of i nvesti gati ng the rel ati onshi p between strai n and stress
todetermi ne a yi el d stress can be appl i ed tosampl es hel d wi thi n fi xtures
other than vanes: concentri c cyl i nder, cone and pl ate, and paral l el pl ate.
Data for Col gate Toothpaste Gel , col l ected at room temperature usi ng
a cone and pl ate apparatus, are presented i n Fi g. 3.18. The i nformati on
was obtai ned by subjecti ng the sampl e toa l i near change i n shear stress,
from 1 Pa to 5 Pa, over a peri od of 500 s. Strai n i s hel d near zero unti l
the appl i ed stress exceeds the yi el d stress (approxi matel y 2.1 Pa)where
a substanti al i ncrease i n strai n i s observed (Fi g. 3.18). Many sampl es
donot produce such cl earl y defi ned di fferences. Data i n the earl y porti on
of the strai n - stress curve (Fi g. 3.17) can al so be used to cal cul ate a
shear modul us.
Figure 3.18. Strain-stress data for presheared Colgate Toothpaste Gel, at 25 C,
collected on a cone (60mm diameter, 4 ) and plate rheometer.
Numerous means of eval uati ng the yi el d stress are noted i n Tabl e
1.4. Most cannot measure a stati c yi el d stress because sampl e structure
i s broken down duri ng testi ng. Al so, the experi mental di ffi cul ti es of
many approaches make them unattracti ve. Overal l , the vane method
has much to offer someone i n need of a yi el d stress measurement: The
1 1.5 2 2.5 3 3.5 4
Stress, Pa
S
t
r
a
i
n
,

%
Colgate Toothpaste Gel
1
0.8
0.6
0.4
0.2
0
-0.2

techni que i nvol ves a si mpl e theory that can produce qui ck, reproduci bl e
resul ts. Operati ng the vane i n the control l ed rate, versus control l ed
stress, mode offers some advantages because i t produces unequi vocal
resul ts ( i s preci sel y defi ned i n the control l ed rate mode) whi ch are
the most sensi ti ve to di fferences between stati c and dynami c yi el d
stresses (Yoo et al ., 1995). I n addi ti on, control l ed rate i nstruments are
general l y l ess expensi ve and more durabl e than control l ed stress uni ts.
Wind-up Characteristics of Rotational Viscometers. I t i s
i mportant to understand the physi cal characteri sti cs of the torque
sensor when consi deri ng yi el d stress measurements usi ng a rotati onal
vi scometer. Two typi cal torque measuri ng concepts, used i n control l ed
rate i nstruments, are i l l ustrated i n Fi g. 3.19. I n one system, force i s
transferred through the sampl e from the bottom pl ate causi ng the
defl ecti on of a coi l ed spri ng. Spri ng deformati on i s measured and cor-
rel ated to torque. Ful l scal e wi nd-up (how much the upper pl ate must
rotate before the ful l scal e torque i s measured) i s hi gh for thi s type of
system, often reachi ng as much as 80 degrees.
Figure 3.19. Rotational viscometers with coil spring (high wind-up) and torsion
bar (low wind-up) torque sensing elements.
M
o
High Wind-Up System Low Wind-Up System
coil spring torsion bar
sample
deflection indicator
rotating member
A system wi th l ow wi nd-up i s al so i l l ustrated i n Fi g. 3.19. I n thi s
system the defl ecti on of a torsi on bar i s correl ated to torque. Ful l scal e
defl ecti on i s achi eved wi th very l ow wi nd-up val ues often i n the range
of 1 to 2 degrees. The wi nd-up characteri sti cs of two vi scometers, one
wi th a spri ng type system and the other wi th a torsi on bar, are con-
si dered i n Exampl e Probl em 3.8.21.
Figure 3.20. Parallel plate viscometer with air bearing sensor support.
Control l ed stress i nstruments use an ai r beari ng to provi de "fri c-
ti onl ess"movement of the upper fi xture (Fi g. 3.20)and a drag cup motor
to generate control l ed torque on the shaft. A posi ti on i ndi cator i s used
to measure di spl acement. Wi nd-up i n control l ed stress systems i s
essenti al l y zero.
3.7.4. Investigating Rheomalaxis
Mi xer vi scometry can be a useful , and si mpl e, method of subjecti vel y
quanti fyi ng rheomal axi s (i rreversi bl e structural breakdown) because
damage i s mi ni mi zed duri ng l oadi ng. The i mpel l er of a mi xer vi scometer
i s pl aced i n an undi sturbed sampl e and torque decay data are col l ected
at vari ous speeds (at l east three). A new, undi sturbed, sampl e i s
requi red for each test. Raw data are pl otted as shown i n Fi g. 3.21. A
cross pl ot of the data i s made: fi rst, one fi nds torque at ti me zero (start
air air
air air
sample
position
air bearing
fixed plate
drag cup
motor
indicator
3.7.4 Investigating Rheomalaxis 209
of test) and equi l i bri um ti me (after the curve becomes hori zontal ) for
each angul ar vel oci ty, then torque ( ) versus angul ar vel oci ty ( ) i s
pl otted. The area between the resul ti ng i ni ti al and equi l i bri um ti me
curves refl ects the degree of rheomal axi s i n the sampl e. I f the i ni ti al
and equi l i bri um torque curves are model ed as functi ons and ,
respecti vel y, the area ( ) between the curves may be cal cul ated as
[3.143]
has the uni ts of power, N m s
-1
. I f i s l arge, the power requi red to
mi x an undi sturbed sampl e i s hi gh compared to that requi red for a ful l y
broken down sampl e. I f the materi al i s not ti me-dependent, . Si nce
thi s i s an empi ri cal testi ng method, i denti cal equi pment and testi ng
procedures must be fol l owed to compare di fferent sampl es. The above
concept i s i l l ustrated for strai ned apri cots i n Exampl e Probl em 3.8.22.
Figure 3.21. Raw and cross plotted data to quantify rheomalatic behavior.
Torque decay data can provi de a useful qual i ty control parameter
for compari ng the structure of di fferent food products by consi deri ng
energy l evel s requi red for mi xi ng. Power ( )versus ti me, at a constant
val ue of , i s pl otted so the area under the resul ti ng curve represents
the mechani cal energy i nput to the sampl e. Sampl es requi ri ng l ow
energy l evel s for mi xi ng refl ect structural breakdown i n the materi al
when compared to si mi l ar, but non-ti me-dependent, fl ui ds. One can
M
f
1
() f
2
()
3
(f
1
() f
2
())d

0
M M
Time
1
2
3
1 2 3
Initial
Equilibrium
area between curves
Raw Data Cross Plotted Data
M

i nvesti gate recovery of sampl e structure by i ntroduci ng a rest peri od,
then repeati ng the test procedure. To make meani ngful compari sons,
the vol ume of sampl e subjected to agi tati on duri ng testi ng must be
constant. Thi s energy anal ysi s concept i s exami ned i n Exampl e Probl em
3.8.22.
Mi xer vi scometry data can al so be used to eval uate a structural
parameter for rheomal ati c materi al s, such as defi ned by Eq. [1.37]
(Ford and Steffe, 1986), or to model torque decay wi th a si mpl e expo-
nenti al equati on (Steffe and Ford, 1985).
3.7.5. Defining An Effective Viscosity
A si mpl e approach can be used to characteri ze non-ti me-dependent
fl ui ds for the purpose of qual i ty control . I n thi s case absol ute rheol ogi cal
properti es are not needed or di ffi cul t to obtai n: Onl y a representati ve
fl ow curve, whi ch can be compared to a reference fl ow curve, i s requi red.
The reference fl ow curve i s determi ned from products consi dered to be
acceptabl e for the i ntended market.
An effecti ve vi scosi ty may be defi ned i n terms of mi xi ng parameters:
[3.144]
where i s equal tothe di ameter of the i mpel l er. Si nce effecti ve vi scosi ty
i s a rel ati ve val ue, other parameters, such as the vol ume of the mi xi ng
vessel or the sampl e vol ume swept through by the i mpel l er, coul d be
used i n pl ace of . Equi pment, sampl e vol ume and i mpel l er ori entati on
shoul d be constant for al l tests. Eq. [3.144] i s si mi l ar to Eq. [3.116] wi th
the assumpti on that and the recogni ti on that .
One can assume that effecti ve vi scosi ty i s di rectl y proporti onal to
apparent vi scosi ty, and angul ar vel oci ty i s di rectl y proporti onal toshear
rate. These val ues can be pl otted to determi ne effecti ve rheol ogi cal
properti es to characteri ze sampl es. Sui tabl e qual i ty control cri teri a can
be devel oped usi ng thi s concept. The effecti ve vi scosi ty i dea i s appl i ed
to strai ned apri cots i n Exampl e Probl em 3.8.18.
3.8.1. Bob Speed for a Bingham Plastic Fluid
Consi der testi ng a Bi ngham pl asti c fl ui d (yi el d stress = 13 Pa ; pl asti c
vi scosi ty = 1.7 Pa s) i n a Searl e-type (the bob rotates and the cup i s
effective

M
d
3
d
d
3
N
Po
M/(d
5
2
) A 1.0
3.8.1 Bob Speed for a Bingham Plastic Fluid 211
stati onary) concentri c cyl i nder vi scometer wi th val ues ( ) of
1.1, 1.3, and 1.5. I n each case, esti mate the mi ni mum angul ar vel oci ty
of the bob requi red to achi eve fl ow at the wal l of the cup, thus, assuri ng
compl ete fl ow i n the annul us.
Eq. [3.29] states that
Therefore, the mi ni mum val ue of torque to achi eve fl ow at i s
[3.145]
The equati on rel ati ng torque to angul ar vel oci ty i n thi s system,
assumi ng fl ow throughout the annul us, i s gi ven by Eq. [3.28]:
Usi ng Eq. [3.145], the mi ni mum torque expressi on, the mi ni mum
angul ar vel oci ty may be cal cul ated:
[3.146]
or, gi ven the above expressi on for mi ni mum torque,
[3.147]
Si mpl i fi cati on, and substi tuti on of , yi el ds a general expressi on
for the mi ni mum angul ar vel oci ty of the bob:
[3.148]
Thi s equati on i ndi cates the cri ti cal i mportance of gap si ze i n deter-
mi ni ng mi ni mum angul ar vel oci ty. Usi ng the speci fi ed yi el d stress and
the pl asti c vi scosi ty gi ves
[3.149]
R
c
/R
b
M
min
2R
c
2
h
>
o
R
c
M
min
2R
c
2
h
o

M
4h
pl
1
1
]
1
R
b
2
1
R
c
2
1
1
]
pl
ln
R
c
R
b
_
min

M
min
4h
pl
1
1
]
1
R
b
2
1
R
c
2
1
1
]
pl
ln
R
c
R
b
_
min

2R
c
2
h
o
4h
pl
1
1
]
1
R
b
2
1
R
c
2
1
1
]
pl
ln
R
c
R
b
_
,
R
c
/R
b
min

o
2
pl
[
2
1 2ln]
min

13
2(1.7)
[
2
1 2ln]
Substi tuti ng the speci fi ed val ues of i nto Eq. [3.149] produces the
fol l owi ng resul ts:
, rad/s
1.1 0.074 (0.71 rpm)
1.3 0.63 (6.04 rpm)
1.5 1.67 (15.95 rpm)
2.0 6.16 (58.82 rpm)
The mi ni mum bob speed to mai ntai n fl ow of a Bi ngham pl asti c fl ui d
i n the annul us i ncreases rapi dl y as the si ze of the gap i ncreases.
Therefore, smal l gaps, appropri ate for gi ven parti cl e si ze l i mi tati ons,
are preferred to l arge gaps.
3.8.2. Simple Shear in Power Law Fluids
Determi ne the error i nvol ved i n usi ng the si mpl e shear approxi mati on
for power l aw fl ui ds at di fferent val ues of .
To sol ve thi s probl em, both the approxi mate (Eq. [3.30]) and actual
(Eq. [3.33]) sol uti ons must be consi dered:
[3.150]
[3.151]
The "% Error"may be cal cul ated as
[3.152]
Substi tuti ng Eq. [3.150] and Eq. [3.151] i nto Eq. [3.152] yi el ds
[3.153]
whi ch descri bes the error expected for di fferent val ues of and .
Resul ts are best seen i n graphi cal form (Fi g. 3.22). To keep error to a
mi ni mum, smal l er gaps (refl ected i n smal l er val ues of ) are needed for
R
c
/R
b

min
R
c
/R
b
(
b
)
approximate

1
(
b
)
actual

2
n
_
2/n
2/n
1
1
1
]
% Error

1
(
b
)
approximate
(
b
)
actual
1
1
]
100
% Error

1
n
2( 1)
2/n
1
2/n
_
,
1
1
]
100
n
3.8.3 Newtonian Fluid in a Concentric Cylinder 213

l ower val ues of the fl ow behavi or i ndex. I f and , error
i nvol ved i n usi ng the si mpl e shear approxi mati on wi l l be l ess than 6%:
At and , the error wi l l exceed 10%.
3.8.3. Newtonian Fluid in a Concentric Cylinder
Determi ne an expressi on for the shear rate as a functi on of the radi us
for a Newtoni an fl ui d i n the annul us of a concentri c cyl i nder vi scometer.
Usi ng thi s expressi on, determi ne the shear rate at the bob and cup.
A Newtoni an fl ui d i s defi ned (Eq. [1.25]) as
and the equati on gi vi ng the shear stress i n the annul us (Eq. [3.2]) i s
Figure 3.22. Error (Eq [3.153]) in shear rate calculation when using the simple
shear approximation for a power law fluid in a concentric cylinder system
( ).
n 0.4 1.0 < < 1.02
n 0.2 > 1.02
f (r)
M
2hr
2
1.01 1.02 1.03 1.04 1.05
0
2
4
6
8
10
%

E
r
r
o
r
n=.2
n=.4
n=.6
n=.8
n=1.0
n=1.2
R
c
/R
b
for . Combi ni ng the above equati ons, the shear rate may be
wri tten as
[3.154]
Eq. [3.21] states that
whi ch can be sol ved for torque as
[3.155]
where, recal l , . Substi tuti ng Eq. [3.155] i nto Eq. [3.154] gi ves
the shear rate as a functi on of the radi us for any constant val ue of the
angul ar vel oci ty:
[3.156]
where . The expressi ons for the shear rate at the bob ( , gi ven
previ ousl y as Eq. [3.32]) and cup ( ) can be easi l y cal cul ated from Eq.
[3.156]:
[3.157]
3.8.4. Representative (Average) Shear Rate
Fi g. 3.23 shows a popul ar bob desi gn based on a German standard (DI N
53019) devel oped by the German I nsti tute for Standardi zati on,
Deutsches I nsti tut fr Normung, known by the DI N acronym. The
fol l owi ng restri cti ons are gi ven for di mensi onal consi derati ons:
; ; ; ; . Preferred
di mensi ons are al sostated: ; ; ; ; .
No preferred di mensi on i s gi ven for . Note that i s the apex angl e
of the cone at the bottom of the i nner cyl i nder, and the summati on of
val ues ( ) i s equal to the fl ui d l evel i n the cup.
R
b
r R
c
f (r)
M
2hr
2

M
4h
1
R
b
2
1
R
c
2
1
1
]
M
4hR
c
2
2
1
R
c
/R
b
f (r)
M
2hr
2
1
2hr
2
1
1
]
4hR
c
2
2
1
1
1
]
2
R
c
2
/r
2
2
1
_
,
R
b
r R
c

b
f (R
b
) 2
2
1
_
c
f (R
c
) 2
2
1
_
,
R
c
/ R
b
1.1 h/R
b
3 h/R
b
1 R
s
/R
b
0.3 90 150
1.0847 h/R
b
3 h/R
b
1 h/R
b
1 120
R
s
/R
b

h
h + h + h
3.8.4 Representative (Average) Shear Rate 215
Figure 3.23. Bob and cup design based on German standard, DIN 53018.
DI N 53019 states that the bob and cup combi nati on i l l ustrated i n
Fi g. 3.23 can be used i n conjuncti on wi th representati ve val ues (average
val ues)of the shear stress and the shear rate i n determi ni ng rheol ogi cal
behavi or of Newtoni an and non-Newtoni an fl ui ds. The representati ve
shear stress, defi ned as the average shear stress between the bob and
the cup, was presented i n Eq. [3.31]:
where has been added to account for both the top and bottom end
effects. The DI N standard states that for the preferred geo-
metri cal rel ati ons al l owi ng the worki ng equati on for average shear
stress to be wri tten as
[3.158]
R
c
R
s
R
b
h
h
h
a

M(1 +
2
)
4(h + h
o
)R
c
2
h
o
h
o
0.1h
a

M(1 +
2
)
4(h + .1h)R
c
2

M(1 +
2
)
(4.4) hR
c
2
Question: Taki ng the Newtoni an approxi mati on for shear rate (Eq.
[3.32]), determi ne the representati ve shear rate that woul d be appro-
pri ate i n eval uati ng the rheogram.
The representati ve shear rate i s the average shear rate ( ) between
the bob and the cup:
[3.159]
Usi ng the val ues of and (Eq. [3.32] and Eq. [3.157])gi ven i n Exampl e
Probl em 3.8.3, Eq. [3.159] can be wri tten as
[3.160]
Mani pul ati on yi el ds a si mpl e equati on to cal cul ate the representati ve
shear rate:
[3.161]
A rheogram can be devel oped from representati ve val ues of shear
stress and shear rate. These equati ons are often used i n the computer
software cal cul ati ons found i n commerci al l y avai l abl e concentri c cyl -
i nder vi scometers. Thi s i dea i s appl i cabl e to most concentri c cyl i nder
systems wi th narrow gaps ( ). When usi ng the average shear
rate equati on for bobs havi ng an end geometry di fferent than that
i l l ustrated i n Fi g. 3.23, the average shear stress equati on woul d need
to be adjusted for di fferent val ues of . I f appl i ed to the bob shown i n
Fi g. 3.1, for exampl e, the val ue of woul d be smal l er than the val ue of
speci fi ed by the DI N standard. Theoreti cal justi fi cati on for usi ng
representati ve shear stress and shear rate val ues may be found i n
Gi esekus and Langer (1977).
3.8.5. Concentric Cylinder Viscometer: Power Law Fluid
Devel op an expressi on for the shear rate at the bob for a power l aw fl ui d
i n a concentri c cyl i nder vi scometer. Al so deri ve an equati on descri bi ng
the fl ui d vel oci ty profi l e found i n thi s system. Assume the vi scometer
i s a Searl e system where the bob rotates and the cup i s stati onary.
b
+
c
2
2
1
_
,
+ 2
2
1
_
,
1
1
]
/ 2
a

(
2
+ 1)
2
1
R
c
/R
b
1.1
h
o
h
o
0.1h
3.8.5 Concentric Cylinder Viscometer: Power Law Fluid 217
Thi s probl em can be sol ved usi ng the same approach taken i n
Exampl e Probl em 3.8.3. By consi deri ng the defi ni ti on of a power l aw
fl ui d ( ) and the force bal ance on a bob (Eq. [3.2]), the fol l owi ng
expressi on for the shear rate i n the annul us i s obtai ned:
[3.162]
Then, the shear rate at the bob i s
[3.163]
However, from Eq. [3.25],
Sol vi ng thi s equati on for and substi tuti ng the resul t i nto the above
expressi on for yi el ds
[3.164]
or, usi ng , Eq. [3.164] becomes
whi ch provi des the fi nal sol uti on gi ven earl i er as Eq. [3.33].
To deri ve an equati on for the vel oci ty profi l e of a power l aw fl ui d i n
a concentri c cyl i nder vi scometer, start wi th Eq. [3.15]:
Changi ng the l i mi ts of i ntegrati on, and substi tuti ng , yi el ds
[3.165]
or, after i ntegrati on,
K
f (r)

M
2hr
2
K
_
,
1/n
b
f (R
b
)

M
2hR
b
2
K
_
,
1/n

n
2K
1/n
R
b
R
c
_
,
2/n
_
M
2hR
b
2
1
1
]
1/n
M
b

2
n
R
c
2/n
R
c
2/n
R
b
2/n
_
,
R
c
/R
b
b

2
n
2/n
2/n
1
_

0
d
1
2
c
f ()
d
f () (/K)
1/n
0
d
1
2
K
_
,
1/n
d

[3.166]
Usi ng Eq. [3.2], Eq. [3.166] can be wri tten as a functi on of the radi us:
[3.167]
Si mpl i fi cati on gi ves
[3.168]
whi ch can be used to generate an expressi on for a di mensi onl ess vel oci ty
profi l e i n the gap:
[3.169]
where . As expected, at , and at .
3.8.6. Concentric Cylinder Data - Tomato Ketchup
Gi ven rawdata for tomatoketchup (Tabl e 3.2), determi ne the rheol ogi cal
properti es of the fl ui d. Use the power l aw approxi mati on for shear rate
presented i n Eq. [3.33]:
Al so eval uate the al ternati ve equati on for gi ven by Heywood (1991b).
The fi rst step i n fi ndi ng the shear rate at the bob i s to cal cul ate the
fl ow behavi or i ndex. Regressi on anal ysi s of versus yi el ds
meani ng, from Eq. [3.37], that
Wi th thi s resul t and the equati on for , the shear rate at each val ue of
may be cal cul ated. Shear stress at the bob i s determi ned from Eq.
[3.3]:
f ()
n
2K
1/n
[
1/n

c
1/n
]
f (r)
n
2K
1/n
M
2hr
2
_
,
1/n
M
2hR
c
2
_
,
1/n
1
1
]
f (r)
n
2K
1/n
M
2hR
c
2
_
,
1/n
R
c
r
_
,
2/n
1
1
1
]
f (r)
f (R
b
)
(1/r
2
)
1/n
(1/R
c
2
)
1/n1
]
(1/R
b
2
)
1/n
(1/R
c
2
)
1/n1
]
[(R
c
/r)
2/n
1]
[(R
c
/R
b
)
2/n
1]
R
b
r R
c
r R
b
0 r R
c
2
n
_
2/n
2/n
1
1
1
]
b
ln(M) ln()
ln(M) (.307) ln() 5.0529
n
d ln(M)
d ln()
0.307
3.8.6 Concentric Cylinder Data - Tomato Ketchup 219

Table 3.2. Data and Results for a Rheological Test for Tomato Ketchup at 25 C
Conducted using a Concentric Cylinder Viscometer: =60.00 mm; =21.00 mm;
=20.04 mm; =1050 kg/m
3
.
RPM
(rad/s) (N m) (Pa) (1/s)
1 0.105 -2.256 0.00346 -5.666 22.85 2.60
2 0.209 -1.563 0.00398 -5.526 26.29 5.19
4 0.419 -0.870 0.00484 -5.331 31.97 10.39
8 0.838 -0.177 0.00606 -5.106 40.03 20.77
16 1.676 0.516 0.00709 -4.949 46.83 41.55
32 3.351 1.209 0.00848 -4.770 56.01 83.10
64 6.702 1.902 0.01060 -4.547 70.01 166.20
128 13.404 2.596 0.01460 -4.227 96.43 332.39
256 26.808 3.289 0.01970 -3.927 130.12 664.78
Resul ts of these cal cul ati ons are presented i n Tabl e 3.2.
Assumi ng a power l aw model to represent the rheogram of thi s
materi al gi ves an excel l ent representati on of the data and a val ue for
the consi stency coeffi ci ent ( ) equal to 15.73 Pa s
n
. Si nce thi s materi al
exhi bi ts power l aw behavi or, the resul ts are al most i denti cal to those
found usi ng the Kri eger approxi mati on (Eq. [3.47]) because i s
essenti al l y equal to . I n addi ti on, one can observe that si mpl e and
Newtoni an shear rate equati ons (Eq. [3.30] and [3.32]) gi ve, at best,
gross approxi mati ons. Consi deri ng the shear rate at 256 rpm (26.808
rad/s) for both cases gi ves
and
Al though, these val ues are the same order of magni tude as the power
l aw approxi mati on of 664.78 s
-1
, the error woul d be unacceptabl e i n most
cases.
h R
c
R
b

ln() M ln(M)
b

b

M
2hR
b
2
K
1/s
1/n
(
b
)
simple shear

1

26.808
1.048 1
558.5 s
1
(
b
)
Newtonian
(2)
2
1
1
1
]
2(26.808)
(1.048)
2
(1.048)
2
1
1
1
]
599.0 s
1
Tocheck for potenti al probl ems due tosecondary fl ow Eq. [3.90] must
be eval uated:
Recal l that Tayl or vorti ces may be expected when thi s i nequal i ty i s
sati sfi ed. Al though the equati on was establ i shed for Newtoni an fl ui ds,
a reasonabl e assessment of the current probl em can be obtai ned by
substi tuti ng the apparent vi scosi ty ( ) for . Taki ng the worst case
(hi ghest speed) the cal cul ati on can be made as
whi ch i s cl earl y much l ess than
meani ng errors due to the presence of secondary fl ow are negl i gi bl e.
Heywood (1991b) offered the fol l owi ng equati on to determi ne the
shear rate at the bob i n a concentri c cyl i nder system:
[3.170]
where
[3.171]
and
[3.172]
Eq. [3.170] i s a si mpl i fi ed form of an i nfi ni te seri es sol uti on of Eq [3.42]
gi ven by Kri eger and Maron (1952). To eval uate Eq. [3.170] for the
speci al case of a power l aw fl ui d one must recogni ze, by observi ng Eq.
[3.37], that and . Maki ng these substi tuti ons, Eq. [3.170]
becomes
[3.173]
I n the case of tomato ketchup, and .
Substi tuti ng these val ues i nto the Kri eger and Maron sol uti on yi el ds
R
b
(R
c
R
b
)
> 41.3
R
c
R
c
R
b
b
/
b

(26.81) (.02004) (.02100 .02004) (1050)
(130.12/664.78)
2.77
41.3
.02100
.02100 .02004
193.16
b

ln
1 +
ln
a
+
1
3
ln
a
_
,
2
(1 a) + .
1
1
]
a
d(lnM)
d(ln)
a
d(a)
d(ln)
a n a 0
b

ln
1 +
ln
n
+
1
3
ln
n
_
,
2
1
1
]
n 0.307 21.00/20.04 1.048
3.8.7 Infinite Cup - Single Point Test 221
The exact power l aw sol uti on i s
meani ng the above equati ons to cal cul ate shear rate are equi val ent to
the second deci mal pl ace.
3.8.7. Infinite Cup - Single Point Test
A vi scometer wi th a si ngl e bob, and no cup, i s to be used i n devel opi ng
a si ngl e poi nt qual i ty control test for tomato ketchup (at 25 C) when
hel d i n a l arge vat. I f the i nstrument i s run at 50 rpm, what apparent
vi scosi ty woul d be expected?
Previ ous work (Exampl e Probl em 3.8.6) wi th a concentri c cyl i nder
i nstrument yi el ded power l aw parameters for the fl ui d: = 15.73 Pa s
n
and =0.307 . The shear rate at the bob for thi s materi al i n a concentri c
cyl i nder system (Eq. [3.33]) i s
I n a vat, whi ch equal s i s l arge sothe term i n brackets i s effecti vel y
one. Therefore, the shear rate may be expressed as
al l owi ng the shear rate at 50 rpm to be cal cul ated:
Wi th thi s val ue of the shear rate, the expected apparent vi scosi ty may
be determi ned:
3.8.8. Infinite Cup Approximation - Power Law Fluid
Determi ne the error i nvol ved i n usi ng the i nfi ni te cup approxi mati on
for power l aw fl ui ds at di fferent val ues of .
b

ln(1.048)
1 +
ln(1.048)
.307
+
1
3
ln(1.048)
.307
_
,
2
1
1
]
24.75
2
.307
_
1.048
2/.307
1.048
2/.307
1
1
1
]
24.75
K
n
2
n
_
2/n
2/n
1
1
1
]
R
c
/R
b
b

2
n
b

2(50) (2)
(.307)60
34.11 s
1
K(
b
)
n 1
15.73(34.11)
.307 1
1.36 Pa s
R
c
/R
b
Figure 3.24. Error (Eq. [3.177]) in using the infinite cup solution for power law
fluids in a concentric cylinder system ( ).
The approxi mate sol uti on i s found from Eq. [3.49] by noti ng (Eq.
[3.36]) that for a power l aw fl ui d:
[3.174]
The actual sol uti on, provi ded Eq. [3.33], i s
[3.175]
The "% Error"may be cal cul ated as
[3.176]
Substi tuti ng Eq. [3.174] and Eq. [3.175] i nto Eq. [3.176] yi el ds
[3.177]
2 4 6 8 10 12
0
1
2
3
4
5
%

E
r
r
o
r
n=.2
n=.4
n=.6
n=.8
n=1.0
n=1.2
R
c
/R
b
d(ln)/d(ln
b
) 1/n
(
b
)
approximate

2
n
(
b
)
actual

2
n
_
2/n
2/n
1
1
1
]
% Error

1
(
b
)
approximate
(
b
)
actual
1
1
]
100
% Error

2/n
1
2/n
_
,
1
1
]
100
3.8.9 Infinite Cup - Salad Dressing 223
whi ch gi ves the error expected for di fferent val ues of and . Pl otti ng
the resul ts (Fi g. 3.24) makes i t cl ear that l arger contai ners are requi red
(l arger ) to keep error to acceptabl e l evel s as the fl ow behavi or i ndex
i ncreases. Wi th , must be greater than 3 to keep the error bel ow
5%. The techni que of error determi nati on used above (and i n Exampl e
Probl em 3.8.2) can be very hel pful i n eval uati ng the assumpti ons made
i n the "automati c"cal cul ati ons of commerci al i nstruments.
3.8.9. Infinite Cup - Salad Dressing
Rheol ogi cal data (Tabl e 3.3) for Kraft French sal ad dressi ng were col -
l ected usi ng a concentri c cyl i nder geometry wi th the fol l owi ng di men-
si ons: bob radi us =20.04 mm; cup radi us =73.00 mm; bob hei ght =60.00
mm. The bob has a recessed top and bottom l i ke the one i l l ustrated i n
Fi g. 3.1 so we wi l l assume end effects are negl i gi bl e. Usi ng the i nfi ni te
cup assumpti on, determi ne the rheol ogi cal properti es of thi s ti me-
i ndependent materi al .
Table 3.3. Rheological Data for Kraft French Salad Dressing at 22 C [Concentric
Cylinder System: =20.04 mm, =73.00 mm, =60.00 mm]
(rad/s) (N m) (Pa) (1/s)
0.146 0.000609 4.02 0.79
0.512 0.000998 6.59 2.77
1.036 0.001264 8.35 5.60
2.087 0.001623 10.72 11.28
4.163 0.002033 13.43 22.50
6.276 0.002430 16.05 33.92
8.359 0.002708 17.89 45.18
10.490 0.002970 19.62 56.70
12.590 0.003149 20.80 68.05
14.680 0.003335 22.03 79.35
16.770 0.003509 23.18 90.65
Shear stress at the bob i s cal cul ated from Eq. [3.3] as
The shear rate may be determi ned from Eq. [3.49] as
n
n 0.6
R
b
R
c
h
M
b

b

M
2hR
b
2
b
(2)
d(ln)
d(ln
b
)
Figure 3.25. Logarithmic plot of angular velocity versus shear stress for Kraft
French salad dressing.
Figure 3.26. Rheogram for Kraft French salad dressing at 22 C.
3 5 10 20 30
0.1
0.2
0.5
1
2
5
10
20
Shear Stress, Pa
A
n
g
u
l
a
r

V
e
l
o
c
i
t
y
,

r
a
d
/
s
Kraft French Dressing
0 20 40 60 80 100
0
5
10
15
20
25
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
Kraft French Dressing
3.8.10 Infinite Cup - Yield Stress Materials 225

Regressi on of the data i ndi cates a good fi t of the l ogari thmi c pl ot (Fi g.
3.25) wi th a sl ope of 2.73:
Knowi ng the sl ope, the shear rate can be cal cul ated as
Wi th the shear stress and shear rate data computed (Tabl e 3.3), the
rheogram may be pl otted (Fi g. 3.26), power l aw constants determi ned
(more regressi on anal ysi s) and the fi nal model presented:
Note, the sl ope of the l ogari thmi c pl ot i s the reci procal of the fl ow
behavi or i ndex ( ) because, as stated i n the previ ous
exampl e probl em, the i nfi ni te cup sol uti on for a power l aw fl ui d i s
Al so, i n consul ti ng Fi g. 3.24, one can see the error i n usi ng the i nfi ni te
cup approxi mati on for and shoul d be l ess
than 0.5%.
3.8.10. Infinite Cup - Yield Stress Materials
Under what condi ti ons coul d Eq. [3.49], an expressi on to determi ne the
shear rate at the bob i n an i nfi ni te cup, be used for a vat contai ni ng a
materi al wi th a yi el d stress?
Knowi ng the yi el d stress, the cri ti cal radi us ( ) can be cal cul ated
by consi deri ng Eq. [3.2]:
[3.178]
where i s the maxi mum torque generated by the i nstrument duri ng
testi ng. defi nes the sheared area of the sampl e i n the vat when fl ui d
moti on i s due onl y to bob rotati on.
d(ln)
d(ln
b
)
2.73
b
(2)2.73
4.43(
)
.37
1/n 1/0.37 2.7
b

2
n
73.00/20.04 3.64 n 0.37
R
o
R
o

M
max
2h
o
M
max
R
o
Figure 3.27. Bob placement, showing sheared area, in a vat full of fluid having a
yield stress.
The i nstrument shoul d be posi ti oned (Fi g. 3.27) so the center of the
bob i s, at a mi ni mum, a di stance equal to from the wal l of the vat.
Under these condi ti ons, the physi cal si tuati on i n the vat i s anal ogous
tohavi ng a concentri ccyl i nder system, wi th a parti al l y sheared annul us;
hence, Eq. [3.49] shoul d gi ve sati sfactory resul ts.
3.8.11. Cone and Plate - Speed and Torque Range
Assume tomato ketchup wi l l be tested i n a cone and pl ate vi scometer
so the resul ts can be compared wi th those found i n the concentri c cyl -
i nder tests. Usi ng a cone wi th an angl e of 0.0524 rad (3 degrees) and a
di ameter of 5.0 cm, what speed and torque response woul d be needed
to cover an experi mental shear rate range of 1 to 100 s
-1
?
I n a cone and pl ate system, the rel ati onshi p between angul ar vel oci ty
and shear rate i s gi ven by rearrangi ng Eq. [3.51]:
Usi ng thi s expressi on, the mi ni mum and maxi mum vel oci ti es may be
cal cul ated as
R
o
vat wall
bob
sheared area
R
o

(tan)
3.8.12 Cone and Plate - Salad Dressing 227
and
To cover the shear rate range of 1 to 100 s
-1
the i nstrument shoul d have
a mi ni mum speed range of 0.5 to 50 rpm. The requi red torque capabi l i ty
of the i nstrument may be consi dered from Eq. [3.55]:
Rearrangi ng thi s expressi on, and i ncorporati ng the defi ni ti on of the
shear stress for a power l aw fl ui d yi el ds
Substi tuti ng the appropri ate numbers (from Exampl e 3.8.6:
Pa s
n
, ) and cal cul ati ng the resul ts gi ves
and
To test sampl es over the shear rate range of 1 to 100 s
-1
the i nstrument
torque response must i ncl ude the range from 5,150 to 21,2000 dyne cm.
3.8.12. Cone and Plate - Salad Dressing
Rheol ogi cal data (Tabl e 3.4) were col l ected for Kraft French sal ad
dressi ng usi ng a cone and pl ate system. Determi ne i f a power l aw model
wi l l adequatel y descri be the behavi or of thi s materi al .
Usi ng Eq. [3.51] and [3.55], the shear rate and shear stress can be
cal cul ated di rectl y from the raw data:
and
min
1(tan(0.0524)) 0.0524 rad/s(0.50 rpm)
max
100(tan(0.0524)) 0.524 rad/s(5.00 rpm)

3M
2R
3
M
2R
3
3

2R
3
K(
)
n
3
K 15.73
n 0.307
M
min

2(.05/2)
3
(15.73) (1)
.307
3
0.000515 N m 5, 150.0 dyne cm
M
max

2(.05/2)
3
(15.73) (100)
.307
3
0.00212 N m 21, 000 dyne cm

tan

3M
2R
3
Table 3.4. Cone and Plate Data ( =25mm; =0.02 rad) for Kraft French Salad
Dressing at 22 C
rad/s N m 1/s Pa
0.002 1.34E-4 0.10 4.09
0.005 1.65E-4 0.25 5.04
0.013 2.16E-4 0.63 6.60
0.025 2.59E-4 1.25 7.91
0.040 2.98E-4 1.99 9.11
0.063 3.42E-4 3.16 10.45
0.100 3.97E-4 5.01 12.13
0.159 4.58E-4 7.94 14.00
0.252 5.28E-4 12.59 16.13
0.399 6.24E-4 19.95 19.07
0.632 7.33E-4 31.62 22.40
1.002 8.70E-4 50.11 26.59
Figure 3.28. Rheogram for Kraft French salad dressing (22 C) determined from
cone and plate data.
R
M

0 10 20 30 40 50 60
0
5
10
15
20
25
30
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
French Dressing
3.8.13 Parallel Plate - Methylcellulose Solution 229

Regressi on anal ysi s of the resul ts (Tabl e 3.4 and Fi g. 3.28) yi el ds a
stati sti cal l y acceptabl e fi t to the experi mental data as
where = 7.64 Pa s
n
and = 0.303. These resul ts are di fferent than
those found i n Exampl e Probl em 3.8.9 when eval uati ng Kraft French
sal ad dressi ng usi ng the i nfi ni te cup assumpti on wi th a concentri c
cyl i nder vi scometer. Di fferences may be due to vari ous factors such as
anal yti cal assumpti ons, natural bi ol ogi cal vari abi l i ty, and di fferences
i n shear rate coverage. More experi mental data woul d be requi red to
resol ve these i ssues.
3.8.13. Parallel Plate - Methylcellulose Solution
Data for a 3% hydroxypropyl methyl cel l ul ose sol uti on have been col -
l ected usi ng a paral l el pl ate vi scometer (Tabl e 3.5). Generate a rheo-
gram for thi s materi al .
Table 3.5. Parallel Plate Data ( =25mm; =0.70mm) for a 3%Aqueous Solution
of Hydroxypropyl Methylcellulose (Methocel K4M, Dow Chemical Co.) at 24.2
C
ln ln
N m 1/s Pa
0.116E-4 0.0127 -11.36 -4.37 0.5
0.211E-4 0.0198 -10.77 -3.92 0.8
0.334E-4 0.0317 -10.31 -3.45 1.3
0.442E-4 0.0503 -10.03 -2.99 1.7
0.807E-4 0.0797 -9.42 -2.53 3.2
1.259E-4 0.1262 -8.98 -2.07 4.9
2.029E-4 0.1999 -8.50 -1.61 7.9
2.979E-4 0.3166 -8.12 -1.15 11.7
4.536E-4 0.5016 -7.70 -0.69 17.8
6.687E-4 0.7945 -7.31 -0.23 26.2
9.343E-4 1.258 -6.98 0.23 36.6
12.900E-4 1.994 -6.65 0.69 50.5
17.270E-4 3.158 -6.36 1.15 67.6
22.700E-4 5.003 -6.09 1.61 88.8
29.260E-4 7.925 -5.83 2.07 114.5
37.320E-4 12.55 -5.59 2.53 146.0
The shear rate at the ri m of the pl ate was determi ned, by consi deri ng
angul ar vel oci ty and geometry, from Eq. [3.59] as
7.64(
)
.303
K n
R h
R
M

R

R
and shear stress was cal cul ated wi th Eq. [3.66]:
Usi ng l i near regressi on, the fol l owi ng rel ati onshi p was determi ned from
the torque and shear rate data (Fi g. 3.29):
whi ch i denti fi es the sl ope term as
Figure 3.29. Torque versus shear rate for 3% aqueous solution of hydroxypropyl
methylcellulose at 24.2 C.
Thi s resul t can be used to eval uate the shear stress at the ri m for each
torque val ue:
R

R
h
R
f (
R
)
M
2R
3
3 +
d ln(M)
d ln(
R
)
1
1
]
ln(M) 7.12 + .934ln(
R
)
d ln(M)
d ln(
R
)
.934
-6 -4 -2 0 2 4
-12
-11
-10
-9
-8
-7
-6
-5
ln
l
n

(
M
)
R
3.8.14 End Effect Calculation for a Cylindrical Bob 231

Shear stress val ues cal cul ated from thi s equati on are presented i n Tabl e
3.5 and pl otted i n Fi g. 3.30. Usi ng the power l aw model , val ues of the
consi stency coeffi ci ent and the fl ow behavi or i ndex were determi ned:
= 25.3 Pa s
n
and = 0.83.
Figure 3.30. Flow behavior of a 3% aqueous solution of hydroxypropyl methyl-
cellulose (Methocel K4M, Dow Chemical Co.) at 24.2 C.
3.8.14. End Effect Calculation for a Cylindrical Bob
Determi ne the bottom end effect ( ) for the bob ( = 1.95 cm) and cup
( = 2.00 cm) combi nati on i l l ustrated i n Fi g. 3.31.
R

M
2R
3
[3 + .934]
K
n
0 5 10 15 20 25 30 35
0
50
100
150
200
250
Shear Rate, 1/s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
h
o
R
b
R
c
Figure 3.31. Bob and cup combination tested for end effect correction.
Figure 3.32. End effect data for the "A-bob" of the Hercules high-shear
viscometer.
5.00 cm
0.60 cm
1.50 cm
4.00 cm
3.90 cm
0 1 2 3 4 5
0
100
200
300
400
500
600
h, cm
M
,

k
d
y
n
e

c
m
600 rpm 900 rpm 1100 rpm
h
17 degrees
3.8.15 Bob Angle for a Mooney-Couette Viscometer 233
The end correcti on for the bottom of the bob was eval uated from
torque data taken at three fi xed bob speeds wi th di fferent val ues of
(Fi g. 3.32) usi ng the techni que presented i n Sec. 3.5. Note that i s
measured from the top of the cone shaped projecti on of the bob. Al l tests
were run at room temperature wi th a Dow Corni ng 200 fl ui d whi ch i s
a di methyl -si l i cone materi al havi ng the fol l owi ng characteri sti cs at
25 C: Newtoni an behavi or wi th = 95.63 cP , and = 0.965 g cm
-3
.
Torque data were pl otted (Fi g. 3.32) as at three speeds:
600 rpm, 900 rpm, and 1100 rpm. val ues were determi ned (Tabl e
3.6) from l i near regressi on coeffi ci ents: at , . Correl ati on
coeffi ci ents were 0.99 for each curve. Resul ts showed the magni tude of
the end effect i ncreasi ng wi th i ncreasi ng speed goi ng from 0.141 cm at
600 rpm to 0.182 cm at 1100 rpm. An average val ue of = 0.158 cm i s
recommended for maki ng practi cal end effect correcti ons over the rpm
range consi dered.
Table3.6. Linear Regression Parameters ( ) Relating Torque
( ) and Length ( ) of Bob I mmersed in Fluid to Evaluate End Effect
Correction ( )
600 rpm 900 rpm 1100 rpm
, dyne 61.74 92.48 113.72
, dyne cm 8.68 14.04 20.67
, cm 0.141 0.152 0.182
3.8.15. Bob Angle for a Mooney-Couette Viscometer
The Mooney-Couette vi scometer i s a concentri c cyl i nder system con-
structed sothe bob has a coni cal bottom whi ch al most touches the bottom
surface of the cup (Fi g. 3.8). I f the system i s desi gned so the shear rate
i n the bottom gap i s equal to the shear rate i n the annul us, the end effect
can be si gni fi cantl y reduced. Determi ne the bottom angl e requi red to
h
h

M ah + b
h
o
M 0 h
o
b/a
h
o
M ah + b
M h
h
o
a
b
h
o
redesi gn a concentri c cyl i nder bob so the shear rate i n the coni cal and
annul ar gaps are equal . Al so expl ai n how the end effect can be i ncor-
porated i nto the shear stress cal cul ati on. Assume the bob under con-
si derati on has the fol l owi ng di mensi ons: , ,
.
The Newtoni an approxi mati on (Eq. [3.32]) can provi de a reasonabl e
esti mate of the shear rate i n the gap:
where:
One can sol ve for the requi red cone angl e by equati ng the shear rates
i n the coni cal secti on, gi ven by Eq. [3.51], to the shear rate i n the gap:
Sol vi ng for yi el ds
so
Therefore, maki ng a bob wi th a bottom angl e (Fi g. 3.8) of 2.56 degrees
wi l l resul t i n an apparatus where the shear rate i n the annul us and
bottom gap are approxi matel y equal . I f the si mpl e shear approxi mati on
(Eq. [3.30]) i s used i n the determi nati on, a sl i ghtl y l arger angl e of 2.75
degrees i s cal cul ated.
The end effect can be i ncorporated i nto the cal cul ati on of the shear
stress at the bob by fi ndi ng a numeri cal val ue for . When i gnori ng the
end effect, shear stress at bob i s gi ven by Eq. [3.3]:
and the shear stress i n the coni cal secti on i s taken from Eq. [3.55] as
R
c
21.00 mm R
b
20.04 mm
h 60.00 mm
b
2
2
1
_
,

R
c
R
b
21.00
20.04
1.048
tan()
2
2
1
_
,
tan()
tan()
1
2
2
1
2
_
,

1
2
1.048
2
1
1.048
2
_
,
0.0447
.0447 radians .0447
360
2
_
,
2.56 degrees
h
o
b

M
2hR
b
2
3.8.16 Viscous Heating 235
Wi th the Mooney-Couette sensor the radi us of the cone and the bob are
equal , , so the above equati on may be wri tten as
[3.179]
Assumi ng the shear stress contri buti on i n the coni cal secti on can be
cal cul ated i n terms of , the shear stresses for the bob and the cup can
be equated:
[3.180]
Sol vi ng for the effecti ve hei ght yi el ds
[3.181]
Usi ng Eq. ]3.181], the overal l shear stress i n the Mooney-Couette system
can be cal cul ated i n terms of the effecti ve hei ght:
[3.182]
3.8.16. Viscous Heating
Consi deri ng a previ ous probl em (Exampl e Probl em 3.8.6) deal i ng wi th
concentri c cyl i nder data for tomato ketchup, esti mate the extent of
vi scous heati ng that may occur duri ng data col l ecti on.
Toconsi der the worst case, take the hi ghest shear rate gi ven, 664.78
s
-1
. Al so, assume the bob surface i s adi abati c and the cup wal l i s
mai ntai ned at a constant temperature equal to . Then, the maxi mum
temperature di fference found at the surface of the bob may be esti mated
as (Eq. [3.80])
Thi s expressi on can be eval uated after substi tuti ng an apparent vi s-
cosi ty functi on for the Newtoni an vi scosi ty and the si ze of the annul us
for :

3M
2R
3
R R
b

3M
2R
b
3
h
o
M
2h
o
R
b
2

3M
2R
b
3
h
o

R
b
3
b

M
2(h + h
o
)R
b
2

M
2(h + R
b
/3)R
b
2
T
o
T
max
T
o

()
2
s
2
2k
s
Assumi ng a thermal conducti vi ty of 0.516 W/(m K) and, usi ng rheo-
l ogi cal properti es determi ned i n Exampl e Probl em 3.8.6 ( = 15.73 Pa
s
n
and = 0.307), yi el ds
A temperature i ncrease of thi s magni tude wi l l usual l y have a negl i gi bl e
i nfl uence on experi mental resul ts.
3.8.17. Cavitation in Concentric Cylinder Systems
Gi ven a concentri c cyl i nder system l i ke the one i l l ustrated i n Fi g. 3.31
( , , ), what bob speed i s requi red to
produce cavi tati on i n 40 C water?
At atmospheri c pressure (101.35 kPa), water at 40 C has the fol -
l owi ng properti es: = 7.38 kPa, = 992.2 kg m
-3
, . Con-
si deri ng the cavi tati on cri teri on gi ven i n Sec 3.5 yi el ds
When > 13.76 m/s, bob speed i s 6738 rpm and cavi tati on may be
present.
I n checki ng the thermodynami c properti es of water, one fi nds that
the vapor pressure of water vari es from 1.4 kPa at 12 C to 19.9 kPa at
60 C. Therefore, the speed requi red to achi eve cavi tati on i n water i s
hi gher at l ower temperatures. Mi ni mum speeds requi red to produce
cavi tati on can be i ncreased i n pressuri zed systems: They wi l l decrease
i n systems hel d under a vacuum.
I t i s i nteresti ng to compare the speed requi red for cavi tati on and the
formati on of Tayl or vorti ces. Vortex formati on i s expected when the
i nequal i ty expressed by Eq. [3.90] i s sati sfi ed:
or
T
max
T
o

K(
)
n 1
(
)
2
(R
c
R
b
)
2
2k
K
n
T
max
T
o

15.73(664.78)
.307 1
(664.78)
2
(.02100 .02004)
2
2(.516)
.069 C
R
b
1.95 cm R
c
2.00 cm h 5.00 cm
P
vap
0.6529 cP
u >
2(P
atm
P
vap
)
2(101.35 7.38) (1000)

992.2
13.76 m/s
u
R
b
(R
c
R
b
)
> 41.3
R
c
R
c
R
b
3.8.18 Mixer Viscometry 237
Meani ng that Tayl or vorti ces can be expected when bob speeds exceed
168.3 rpm. I n thi s exampl e, the mi ni mum bob speed requi red for vortex
formati on i s much l ower than the mi ni mum speed needed for the onset
of cavi tati on. I t woul d be unusual to devel op cavi tati on probl ems when
testi ng food products. I n general , the l ami nar fl ow assumpti on woul d
be vi ol ated before the onset of cavi tati on.
3.8.18. Mixer Viscometry
A mi xer vi scometer was used to determi ne the rheol ogi cal properti es of
a starch-thi ckened baby food product, strai ned apri cots at 25 C. The
materi al was mi xed suffi ci entl y to remove ti me-dependent effects.
Mi xer vi scometer data were col l ected usi ng a rotati onal vi scometer
equi pped wi th a cup (i nsi de radi us =0.021 m), and Haake pi tched-paddl e
i mpel l er (bl ade di ameter = 0.04143 m, bl ade hei ght = 0.02692 m)
i l l ustrated i n Fi g. 3.11. Rheol ogi cal properti es ( and )stated i n Tabl e
3.7 were determi ned usi ng concentri c cyl i nder data, wi th Eq. [3.3] to
cal cul ate shear stress, the Kri eger method to cal cul ate shear rate (Eq.
[3.47]), and standard stati sti cal methods to eval uate the rheogram.
Questions: Part a. Taki ng the raw data for corn syrup and vari ous
aqueous sol uti ons found i n Tabl e 3.7, determi ne the mi xer vi scometer
constant ( )usi ng the sl ope method. Partb.Usi ng the data for strai ned
apri cots gi ven i n Tabl e 3.8, cal cul ate the fl ow behavi or i ndex and the
consi stency coeffi ci ent of the product. Compare di fferent methods of
eval uati on. Part c. Usi ng the data for strai ned apri cots (Tabl e 3.8),
determi ne an effecti ve vi scosi ty fl ow curve for thi s materi al whi ch may
be used for qual i ty control purposes.
Part a. Fi rst, the raw data are mani pul ated i nto the appropri ate form
(Tabl e 3.7) and pl otted (Fi g. 3.33). Regressi on anal ysi s of
versus ( ) as speci fi ed i n Eq. [3.115],
(.0195) (.02 .0195)992.2
.0006529
> 41.3
.02
.02 .0195
> 17.63 rad/s or 168.3 rpm
K n
k
log
10
(P/(K
n + 1
d
3
)) 1 n
Table 3.7. Raw(first three columns) and Manipulated Data used to Determine the
Mixer Viscometer Constant. [ ] Data from: Steffe and
Ford (1985)
Fluid
(Pa s
n
) (-) (N m/s) (-)
2.5% 16.55 0.513 0.0585 0.487 3.0882 0.4897
Hydroxypropyl 15.56 0.520 0.0573 0.480 3.1762 0.5019
Methylcellulose 16.61 0.505 0.0577 0.495 3.0799 0.4885
15.90 0.513 0.0562 0.487 3.0881 0.4897
2.0% 7.29 0.574 0.0300 0.426 3.2141 0.5071
Hydroxypropyl 6.94 0.586 0.0311 0.414 3.4236 0.5345
7.06 0.576 0.0308 0.424 3.3948 0.5308
1.5% 2.10 0.676 0.0129 0.324 3.9776 0.5996
Hydroxypropyl 2.16 0.672 0.0127 0.328 3.8352 0.5838
2.10 0.680 0.0129 0.320 3.9484 0.5964
1.0% 1.84 0.560 0.0072 0.440 3.1532 0.4988
Hydroxypropyl 1.26 0.605 0.0058 0.395 3.4136 0.5332
1.83 0.504 0.0063 0.496 3.0385 0.4827
Corn Syrup 2.70 1.000 0.0488 0.000 6.4520 0.8097
2.84 0.992 0.0474 0.008 6.0462 0.7815
2.64 1.000 0.0482 0.000 6.5175 0.8141
1.5% Guar Gum 30.98 0.158 0.0328 0.842 1.7762 0.2495
31.65 0.159 0.0339 0.841 1.7936 0.2537
27.39 0.169 0.0316 0.831 1.8968 0.2780
1.0% Guar Gum 7.28 0.275 0.0113 0.725 2.1002 0.3223
4.93 0.292 0.0083 0.708 2.2159 0.3455
7.99 0.266 0.0121 0.734 2.0832 0.3187
8.62 0.237 0.0123 0.763 2.0704 0.3160
10.76 0.224 0.0143 0.776 1.9749 0.2955
gi ves a good strai ght l i ne fi t ( ) wi th , and a sl ope
equal to -0.6503 meani ng that
so
60 rpm 6.283 rad/s
K n P M 1 n P
K
n +1
d
3
log
10
P
K
n +1
d
3
_
,
log
10
P
K
n + 1
d
3
_
,
log
10
(A) (1 n) log
10
(k)
r
2
0.99 log
10
A 0.803
log
10
(k) 0.6503
Al so si nce , rad
-1
.
Figure 3.33. Plot of data needed to determine the mixer viscometer constant ( )
using the slope method.
Part b. Raw data are mani pul ated i nto the appropri ate form (Tabl e
3.8). Eval uati on of Eq. [3.121],
usi ng l i near regressi on anal ysi s yi el ds and
wi th . may be found wi thout usi ng the numeri cal
val ue of , as , from Eq. [3.125],
Taki ng data at 60 rpm (mi d-range val ue) and the average val ues of the
1%hydroxypropyl methyl cel l ul ose as the known fl ui d gi ves: =0.00533
k 4.47 rad
1
log
10
0.803 A 6.35
1-n
Mixer Viscometer Data
0.8
0.9
0.7
0.6
0.5
0.4
0.3
0.2
0 0.2 0.4 0.6 0.8 1.0
l
o
g
1
0
[
P
/
(
K

d

)
n
+
1
3
k
log
10
(M) log
10
(d
3
AK(k)
n 1
) + n log
10
()
log
10
[d
3
AK(k)
n 1
] 2.597
r
2
0.99 n 0.378 K
A K
x
K
x
K
y
M
x
M
y
_
y
n
y
(k)
n
y
x
n
x
(k)
n
x
_
,
M
x
Table 3.8. Raw (first two columns) and Manipulated Mixer Viscometer Data for
Strained Apricots at 25 C
Speed
(rpm) (rad/s) (N m) (Pa s) (1/s) (Pa s)
20 2.094 0.00332 0.321 -2.479 3.511 9.36 22.30
40 4.189 0.00429 0.622 -2.368 2.268 18.72 14.40
60 6.283 0.00533 0.798 -2.273 1.879 28.09 11.93
80 8.378 0.00556 0.923 -2.255 1.470 37.45 9.33
100 10.47 0.00607 1.020 -2.217 1.284 46.81 8.15
120 12.57 0.00655 1.099 -2.184 1.154 56.17 7.33
N m; = 6.283 rad/s; = 0.378; = 4.47 rad
-1
; = 0.588; =
0.000984 N m; = 1.45 Pa s
n
. Substi tuti ng these val ues i nto the above
equati on yi el ds
Each "standard"or reference fl ui d used i n determi ni ng gi ves a sl i ghtl y
di fferent val ue of . Vari ati ons of 5 to 10% have been observed wi th
thi s method of cal cul ati on (Steffe and Ford, 1985).
The consi stency coeffi ci ent may al sobe found usi ng the known val ue
of and the i ntercept determi ned from our eval uati on of Eq. [3.121]:
Substi tuti ng known val ues gi ves
whi ch can be sol ved for the consi stency coeffi ci ent:
Thi s compares reasonabl y wel l wi th the val ue of 15.77 Pa s
n
found
previ ousl y, and probabl y gi ves a more accurate i ndi cati on of sampl e
behavi or.
Wi th a known val ue of , power l aw fl ui d properti es can al so be
determi ned by appl yi ng the matchi ng vi scosi ty assumpti on to the raw
data. An average apparent vi scosi ty may be found usi ng Eq. [3.116]
wri tten as
M log
10
log
10
M

a
M
1
d
3
x

y
n
x
k n
y
M
y
K
y
K
x

(.00533) (1.45) (6.28)
.588
(4.47)
.588
(.000984) (6.28)
.378
(4.47)
.378
15.77 Pa s
n
k
K
x
A
log
10
[d
3
AK(k)
n 1
] 2.597
log
10
[(.04143)
3
(6.35)K(4.47)
.378 1
] 2.597
K 14.22 Pa s
n
k
Si nce was found to be 6.35 rad
-1
and the di ameter was gi ven ( =
0.04143 m), apparent vi scosi ty can be di rectl y cal cul ated from the
torque:
The mi xer vi scometer constant was determi ned to be 4.47 rad
-1
so the
average shear rate i s
and , eval uated usi ng the above equati ons, are gi ven i n Tabl e 3.8.
Assumi ng power l aw fl ui d behavi or, these terms are rel ated as
or
Regressi on anal ysi s of the data, usi ng thi s rel ati onshi p, yi el ds
and Pa s
n
. Si nce and are di rectl y proporti onal to and ,
respecti vel y, resul ts are the same as those found previ ousl y. Note that
average shear stress coul d al so be cal cul ated as , and regressi on
anal ysi s coul d be preformed on versus .
Part c. An effecti ve vi scosi ty can be cal cul ated on the basi s of Eq.
[3.144]:
Numeri cal val ues of thi s term are gi ven i n Tabl e 3.8 and pl otted as a
functi on of angul ar vel oci ty i n Fi g. 3.34. Assumi ng effecti ve vi scosi ty
i s di rectl y proporti onal to apparent vi scosi ty and angul ar vel oci ty i s
di rectl y proporti onal to shear rate, the vi scosi ty may be model ed as a
power l aw functi on:

d
2
A
(N
Po
)
d
2
M
d
5
,

M
Ad
3
A d

M
6.35(.04143)
3
a
k 4.47
a
K(
a
)
n 1
ln ln K + (n 1) ln
a
n 0.378
K 14.22

a
M
/
effective

M
d
3
Figure 3.34. Effective viscosity of strained apricots at 25 C determined from
mixer viscometry data.
[3.183]
where may be defi ned as an effecti ve consi stency coeffi ci ent (Pa )
and as an effecti ve fl ow behavi or i ndex (di mensi onl ess). Performi ng
a regressi on anal ysi s on the data gave
meani ng = 35.59 Pa and = 0.378. Thi s expressi on i s pl otted as
the l i ne di spl ayed i n Fi g. 3.34. Due tothe defi ni ti on of effecti ve vi scosi ty,
the val ue of i s uni que to the fl ui d i n questi on wi th respect to the
experi mental equi pment empl oyed; , however, i s numeri cal l y the same
as the val ue of the fl ow behavi or i ndex found i n Part b. These resul ts
coul d be compared to those from a reference fl ui d (product found
acceptabl e i n the market pl ace) and used i n maki ng qual i ty control
deci si ons i n a commerci al food processi ng operati on.
0 2 4 6 8 10 12
0
5
10
15
20
25
Angular Velocity, rad/s
E
f
f
e
c
t
i
v
e

V
i
s
c
o
s
i
t
y
,

P
a

s
effective
K( )
n 1
s
n
K
n
effective
35.59( )
.623
s
n
K n
K
n
3.8.19 Vane Method - Sizing the Viscometer 243
3.8.19. Vane Method - Sizing the Viscometer
The vane method wi l l be used to i nvesti gate vari ous food products that
have yi el d stresses rangi ng from 10 to 150 Pa. Vane speed i s 0.1 rpm
(or as l ow as possi bl e) and the di mensi ons of the vane are = 26.15 mm
and = 1.92. I t i s proposed that one of the fol l owi ng Brookfi el d vi s-
cometers be used for thi s task (Note, 1 N m = 10
7
dyne cm):
I nstrument Model Ful l Scal e Torque Range (dyne cm)
LV Seri es 673.7
RV Seri es 7,187
HAT Seri es 14,374
HBT Seri es 57,496
Part a. Determi ne the feasi bi l i ty of usi ng one of the above vi scometers
for the proposed yi el d stress measurement. Part b. Desi gn an appro-
pri ate sampl e contai ner for yi el d stress measurement and speci fy the
mi ni mum sampl e si ze requi red for conducti ng each experi ment.
Part a. The torque range requi red to determi ne the speci fi ed yi el d
stresses must be eval uated. Assumi ng the vane i s ful l y i mmersed i n
the sampl e, and , the peak torque i s eval uated from Eq. [3.141] as
Therefore the range of maxi mum torque response wi l l be
or
or
Accordi ng to the torque capabi l i ti es of the i nstruments, none are suf-
fi ci ent to handl e the job. The HBT, whi ch has the maxi mum torque
capaci ty, can measure a yi el d stress wi th the fol l owi ng maxi mum val ue:
d
h/d
m 0
M
o

o
d
3
2
_
h
d
+
1
3
_
,
1
1
]

o
(.02615)
3
2
_
1.92 +
1
3
_
,
1
1
]
6.329(10
5
)
o
6.329(10
5
) (10) < M
o
< 6.329(10
5
) (150)
6.329(10
4
) N m < M
o
< 9.494(10
3
) N m
6, 329 dyne cm < M
o
< 94, 940 dyne cm
57496
6.329(10
5
)
_
,
10
7
90.85 Pa
One sol uti on tothi s probl em i s to work wi th Brookfi el d to "speci al order"
an i nstrument wi th a l arger spri ng constant. Al ternati vel y, a smal l er
vane coul d be uti l i zed to reduce the torque requi rements of the
i nstrument.
Part b. Typi cal contai ner di mensi ons, suggested i n Sec. [3.7.3] and
i l l ustrated i n Fi g. 3.14, may be cal cul ated from the si ze of the vane. I f
m and , then m. The mi ni mum contai ner
si ze can be cal cul ated from thi s i nformati on:
m
m
m
Wi th these resul ts, the mi ni mum sampl e vol ume may be determi ned:
A mi ni mum sampl e vol ume of 192.1 cm
3
i s needed to perform the test.
3.8.20. Vane Method to Find Yield Stresses
Cal cul ate the stati c and dynami c yi el d stresses, usi ng both the si ngl e
poi nt and sl ope methods, wi th the data for tomato ketchup found i n
Tabl e 3.9. Data for cal cul ati ng stati c yi el d stresses were col l ected usi ng
undi sturbed sampl es. Dynami c stress data were taken from sti rred
sampl es where gentl e agi tati on el i mi nated the ti me-dependent struc-
ture of the materi al . Vanes were compl etel y i mmersed i n the sampl e
fol l owi ng the gui del i nes for vane and vessel di mensi ons gi ven i n Sec.
3.3.7 and summari zed i n the previ ous exampl e probl em.
Eq. [3.141] i s needed to cal cul ate the yi el d stress usi ng the si ngl e
poi nt method:
d 0.02615 h/d 1.92 h 0.05021
D/d 2.0 so D 2.0(d) 0.0523
Z
1
/d 1.0 so Z
1
1.0(d) 0.02615
Z
2
/d 0.5 so Z
1
0.5(d) 0.01308
min. sample
D
2
_
,
2
(z
1
+ h + z
2
)
min. sample
0.0523
2
_
,
2
(0.02615 + 0.05021 + 0.01308) 1.921(10
4
) m
3
192.1 cm
3
3.8.20 Vane Method to Find Yield Stresses 245
Table 3.9. Raw Data, and Yield Stresses, for Tomato Ketchup at 21 C Taken with
a Brookfield Viscometer Having a Full Scale Torque of 0.00575 NmUsing a Vane
with 4 Blades having a Diameter of 2.5 cm.
Stress Torque Torque Vane Height Yield Stress
1
(% total) (N m) (cm) (Pa)
Static 30.00 0.00173 5.0 30.12
Static 29.25 0.00168 5.0 29.37
Static 27.50 0.00158 5.0 27.61
Static 26.00 0.00150 5.0 26.11
Static 34.00 0.00196 6.6 26.79
Static 37.25 0.00214 6.6 29.35
Static 38.00 0.00219 6.6 29.94
Static 47.50 0.00273 9.0 28.29
Static 48.00 0.00276 9.0 28.59
Static 48.75 0.00280 9.0 29.04
Average 28.52
SDEV 1.33
Dynamic 22.50 0.00129 5.0 22.59
Dynamic 22.50 0.00129 5.0 22.59
Dynamic 21.00 0.00121 5.0 21.08
Dynamic 20.00 0.00115 5.0 20.08
Dynamic 25.33 0.00146 6.6 19.96
Dynamic 26.50 0.00152 6.6 20.87
Dynamic 27.50 0.00158 6.6 21.67
Dynamic 37.00 0.00213 9.0 22.04
Dynamic 35.00 0.00201 9.0 20.85
Dynamic 36.00 0.00207 9.0 21.44
Dynamic 34.75 0.00200 9.0 20.70
Average 21.26
SDEV 0.90
1
Yield Stress calculated by the single point method, Eq. [3.141].
Assumi ng , thi s expressi on can be reduced to
o

2M
o
d
3
h
d
+
1
m + 3
_
,
1
m 0
o

6M
o
d
2
(3h + d)
Atypi cal cal cul ati on can be used toi l l ustrate the method. I f, for exampl e
(see fi rst data l i ne, Tabl e 3.9), , ,
and , then . Resul ts for each tri al , as wel l as
average val ues, are gi ven i n Tabl e 3.9.
Eq. [3.137] i s requi red to cal cul ate the yi el d stress usi ng the sl ope
method:
Noti ng that the i ntegral term i s constant, versus may be pl otted
as
[3.184]
The sl ope ( ) of the best fi tti ng l i ne through the data poi nts wi l l i ncl ude
the yi el d stress:
[3.185]
therefore,
[3.186]
Thi s techni que must be performed separatel y for stati c and dynami c
data sets. Data are pl otted (Fi g. 3.35), and sl opes are found from a
standard l i near regressi on program, al l owi ng the stati c and dynami c
yi el d stresses to be cal cul ated from Eq. [3.186]:
and
Reasonabl y good agreement was found between yi el d stresses cal cul ated
usi ng the sl ope and si ngl e poi nt methods (Tabl e 3.10). Si ngl e poi nt yi el d
stresses are averaged val ues taken from Tabl e 3.9.
M
o
0.3(0.00575) 0.00173 N m d 0.025 m
h 0.050 m
o
30.12 Pa
M
o

hd
2
2

o
+ 4
0
d/ 2
r
2
e
dr
M
o
h
M
o
a h + constant
a
a
d
2
o
2
o

2a
d
2
(
o
)
static

2a
d
2

2(0.0285)
(0.025)
2
29.03 Pa
(
o
)
dynamic

2a
d
2

2(0.0205)
(0.025)
2
20.88 Pa
3.8.21 Vane Rotation in Yield Stress Calculation 247
Figure 3.35. Plot of the raw data needed to determine the yield stress using the
slope method.
Table 3.10. Comparison of Yield Stresses Calculated for Tomato Ketchup at 21 C.
Method Static Yield Stress Dynamic Yield Stress
(Pa) (Pa)
Single Point 28.52 21.26
Slope 29.03 20.88
3.8.21. Vane Rotation in Yield Stress Calculation
Typi cal data, for undi sturbed tomato ketchup at 23 C, col l ected to
determi ne the yi el d stress usi ng the vane method, are i l l ustrated i n Fi g.
3.36 and 3.37. The data i n each fi gure are taken wi th two i nstruments
havi ng di fferent wi nd-up characteri sti cs: the Brookfi el d uses a coi l ed
spri ng and the Haake i nstrument uses a torsi on bar type system (Fi g.
3.19). Both systems have rotati ng vanes and stati onary sampl e hol ders.
Gi ven the characteri sti cs of each i nstrument (Tabl e 3.11), determi ne
the amount of vane rotati on (Fi g. 3.38) whi ch has occurred between the
start of the test and the ti me the peak torque val ue ( ) i s observed.
Al so, compare yi el d stresses cal cul ated from each curve (Fi g. 3.36 and
Fi g. 3.37) usi ng the si ngl e poi nt method.
0
0.0005
0.001
0.0015
0.002
0.0025
0.003
h (m)
M

(
N

m
)
Static Dynamic
M = 0.0285 h + 0.000199
M = 0.0205 h + 0.000192
o
o
o
.04 .05 .06 .07 .08 .09 0.1
M
o
Figure 3.36. Torque response for 23 C ketchup obtained using the Brookfield
HBDT viscometer and a vane with 4 blades.
Figure 3.37. Torque response for 23 C ketchup obtained using the Haake VT500
viscometer and a vane with 4 blades.
0 5 10 15 20 25 30 35
0
0.05
0.1
0.15
0.2
Time, s
T
o
r
q
u
e
,

N

c
m
Instrument: Brookfield HBDT
Speed = 0.5 rpm
Vane: h=6.6 cm, d=2.5 cm
0 1 2 3 4 5 6 7
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Time, s
T
o
r
q
u
e
,

N

c
m
Instrument: Haake VT500
Speed = 2.0 rpm
Vane: h=6.0 cm, d=4.0 cm
3.8.21 Vane Rotation in Yield Stress Calculation 249

Figure 3.38. Movement of a 4-bladed vane during yield stress testing.
Table 3.11. Vane Rotation Data and Instrument Characteristics for Two Vis-
cometers.
Haake Brookfield
VT500 HBDT
Vane height ( ), cm 6.0 6.6
Vane diameter ( ), cm 4.0 2.5
Full Scale Torque, N cm 2.0 0.575
Full Scale Wind-up, degree 1.0 80
Time to Peak Torque, s 1.3 18
Peak Torque ( ), N cm 0.605 0.184
Test Speed, rev/min (degree/sec) 2 (12) 0.5 (3.0)
Zero-Load Rotation at Peak Torque, degree 15.6 54.0
Wind-up at Peak Torque, degree 0.3 25.6
Actual Rotation at Peak Torque, degree 15.3 28.4
Yield Stress ( ) from Eq. [3.141], Pa 32.8 28.4
Initial Vane Position
Vane Position at Peak Torque
Angle of Rotation
h
d
M
o
o
The maxi mum torque measured i n the VT500 test was 0.605 N cm.
Si nce the ful l scal e torque of thi s i nstrument i s 2 N cm and the ful l scal e
wi nd-up i s 1 degree, the wi nd-up i n thi s experi ment was approxi matel y
0.605 (1) /2 = 0.3 degrees. The test was run wi th an i mpel l er speed of
2 rpm or 12 degrees /s. Taki ng the ti me to achi eve peak torque as 1.3
s means that an unrestri cted i mpel l er woul d travel 1.3 (12) = 15.6
degrees. Thi s i s referred to as the zero-l oad rotati on. After accounti ng
for wi nd-up, the actual VT500 vane rotati on can be cal cul ated as 15.6-0.3
=15.3 degrees. The comparabl e val ue for the Brookfi el d i s 28.4 degrees
(Tabl e 3.11).
Stati c yi el d stresses were cal cul ated usi ng Eq. [3.141] wi th the
assumpti on that = 0. The val ue measured wi th the Brookfi el d ( =
25.1 Pa) i s l ower than the val ue measured by the VT500 ( = 32.8 Pa).
Thi s i s due, i n part, to the di fferent wi nd-up characteri sti cs of each
vi scometer: At ful l scal e torque, the VT500 wi nd-up i s 1 degree and the
Brookfi el d wi nd-up i s approxi matel y 80 degrees. The combi nati on of
vane speed and wi nd-up caused a greater amount of vane rotati on i n
the Brookfi el d duri ng testi ng. Thi s produced greater fl ui d moti on whi ch
subsequentl y reduced peak torque val ues. Di fferences i n the wi nd-up
characteri sti cs of rotati onal vi scometers shoul d al ways be taken i nto
consi derati on when eval uati ng yi el d stresses determi ned usi ng the vane
method. These di fferences may be parti cul arl y i mportant when the
yi el d stresses under consi derati on exhi bi t strong ti me-dependent
behavi or.
3.8.22. Rheomalaxis from Mixer Viscometer Data
Torque decay data for starch thi ckened strai ned apri cots were col l ected
at vari ous angul ar vel oci ti es (Tabl e 3.12, Fi g. 3.39). The materi al was
rheomal acti c, i .e., no structural recovery was evi dent i n a reasonabl e
peri od of ti me (three hours). Eval uate the rel ati ve degree of i rreversi bl e
thi xotropy i n thi s materi al by cal cul ati ng the area between the i ni ti al
and equi l i bri um torque curves. Al so determi ne the energy i nput to the
sampl e when the i mpel l er i s rotated at 3.14 rad/s for approxi matel y 10
mi nutes. (Note: data i n Tabl e 3.12 coul d al so be used to eval uate the
parameters i n the thi xotropy model gi ven by Eq. [1.36].)
I ni ti al and equi l i bri um torque val ues, obtai ned from the raw data,
are pl otted on Fi g. 3.40 and summari zed i n Tabl e 3.13. Fi tti ng these
curves to a power equati on (any equati on whi ch adequatel y fi ts the data
woul d be acceptabl e) yi el ds:
m
o
o
3.8.22 Rheomalaxis from Mixer Viscometer Data 251
Table 3.12. Torque Decay Data for Starch Thickened, Strained Apricots at 22 C
Collected with a Rotational Viscometer using a Cup ( =21.00 mm) and Pitched
Paddle Impeller (Fig. 3.11)
Torque, N m
Time (min.) 1.05 rad/s 1.57 rad/s 2.09 rad/s 2.62 rad/s 3.14 rad/s
0.42 0.00301 0.00324 0.00369 0.00404 0.00424
1.223 0.00259 0.00292 0.00324 0.00348 0.00369
2.023 0.00240 0.00280 0.00310 0.00330 0.00354
2.821 0.00238 0.00272 0.00299 0.00322 0.00340
3.623 0.00230 0.00266 0.00293 0.00312 0.00332
4.421 0.00224 0.00260 0.00287 0.00306 0.00326
5.223 0.00219 0.00255 0.00284 0.00303 0.00321
6.023 0.00214 0.00251 0.00279 0.00298 0.00315
6.821 0.00210 0.00245 0.00273 0.00294 0.00311
7.623 0.00205 0.00242 0.00270 0.00292 0.00309
8.421 0.00201 0.00238 0.00267 0.00287 0.00304
9.223 0.00198 0.00237 0.00265 0.00286 0.00301
10.02 0.00194 0.00234 0.00262 0.00282 0.00298
10.83 0.00192 0.00230 0.00260 0.00280 0.00297
11.63 0.00190 0.00228 0.00258 0.00277 0.00295
12.43 0.00188 0.00227 0.00255 0.00275 0.00293
13.23 0.00186 0.00224 0.00254 0.00273 0.00289
14.03 0.00184 0.00223 0.00253 0.00270 0.00288
14.83 0.00182 0.00220 0.00250 0.00269 0.00286
15.63 0.00180 0.00218 0.00248 0.00267 0.00286
and
The area between the two curves i s found from Eq. [3.143]:
R
c
M
initial
.00290()
.33
M
equilibrium
.00179()
.42

low
high
[M
initial
M
equlibrium
]d
Figure 3.39. Torque decay data for strained apricots at 22 C.
Figure 3.40. "Initial" and "equilibrium" torque curves for apricots at 22 C.
0 5 10 15
0.0015
0.002
0.0025
0.003
0.0035
0.004
0.0045
Time, min.
T
o
r
q
u
e
,

N

m
3.14 2.62 2.09 1.57 1.05
Strained Apricots
1 1.5 2 2.5 3 3.5

0.0015
0.002
0.0025
0.003
0.0035
0.004
0.0045
T
o
r
q
u
e
,

N

m
.42 min. 15.6 min.
Strained Apricots
3.8.22 Rheomalaxis from Mixer Viscometer Data 253

Eval uati ng the i ntegral wi th the appropri ate torque curves and l i mi ts
yi el ds
i s a rel ati ve measure of thi xotropy; when none i s present
Table3.13. "Initial"and "Equilibrium"Torquevalues, for Strained Apricots, at Different
Angular Velocities
Torque, N m
"Initial" "Equilibrium"
(rad/s) (after 0.42 min.) (after 15.6 min.)
1.05 .00301 .00180
1.57 .00324 .00218
2.09 .00369 .00248
2.62 .00404 .00267
3.14 .00424 .00286
Torque decay curves al so provi de a practi cal way of eval uati ng
thi xotropy by consi deri ng the energy i nput to the sampl e. Data (Tabl e
3.12)taken at an i mpel l er angul ar vel oci ty of 3.14 rad/s may be repl otted
i n terms of power ( ) versus ti me. The area under the resul ti ng
power decay curve (Fi g. 3.41) represents the mechani cal energy i nput
to the sampl e.
The power decay data were fi t to a si mpl e mathemati cal model
yi el di ng
whi ch i s the l i ne pl otted on Fi g. 3.41. The exponent i n thi s equati on i s
an i ndex of thi xotropy: More negati ve val ues i ndi cate a rapi d rate of
structural breakdown, and a val ue of zero means the sampl e i s not
ti me-dependent. Usi ng the above equati on, the energy i nput over the
peri od of ti me (Tabl e 3.12) under consi derati on (t = 0.42 mi n = 25.2 s to
t = 10.02 mi n = 601.2 s) i s easi l y cal cul ated:

1.05
3.14
(.00290()
.33
.00179()
.42
)d .00261 N m s
1
0.
P M
P .0186(t )
.108
Energy Input

t 25.2
t 601.2
.0186(t )
.108
dt 5.91 N m
I n contrast, i f there was no thi xotropy, the energy requi red to mai ntai n
= 3.14 rad/s over the same ti me woul d be approxi matel y
0.00424(3.14)(601.2 - 25.2) = 7.67 N m. Thi s cal cul ati on assumes the
power l evel i s constant and equal to the i ni ti al val ue when no ti me-
dependent behavi or i s present.
Figure 3.41. Power decay curves obtained for strained apricots at 22 C with a
constant impeller velocity of 3.14 rad/s.
Torque or power decay data curves can provi de a useful measure of
thi xotropy. Resul ts depend on the parti cul ar mi xi ng system and sampl e
vol ume used i n the study. These factors must be kept constant to make
meani ngful sampl e to sampl e compari sons. Temperature i s al so a
si gni fi cant factor whi ch shoul d be careful l y control l ed duri ng testi ng.
0 100 200 300 400 500 600

0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
Time, s
P
o
w
e
r
,

N

m

/

s
Strained Apricots
Angular Velocity = 3.14 rad/s
Chapt e r 4 . Ext e ns ional Flow

4.1. Introduction
A basi c knowl edge of extensi onal vi scosi ty i s essenti al for under-
standi ng many fl ow si tuati ons found i n the food i ndustry. The three
basi c types of extensi onal fl ow - uni axi al , bi axi al , and pl anar - al ong
wi th numerous appl i cati ons of extensi onal deformati on are descri bed
i n Secti on 1.7. Experi mental methods todetermi ne extensi onal vi scosi ty
are di scussed i n thi s chapter. Techni ques i nvol vi ng tensi on (stretchi ng
or pul l i ng), squeezi ng fl ow (fl ow between paral l el pl ates bei ng pushed
together), convergi ng fl ow i nto a di e, fl ow i nto opposi ng jets, spi nni ng,
and the tubel ess si phon phenomenon (Fano fl ow) are presented. I n
addi ti on, al ternati ve i nterpretati ons of squeezi ng fl ow data are con-
si dered because of the rel ati ve ease wi th whi ch thi s experi ment can be
conducted i n a l aboratory. Bi axi al extensi on, produced by l ubri cated
squeezi ng fl ow, i s consi dered i n terms of steady shear properti es (Sec.
4.8.1). Si mi l ar consi derati ons are made i n the case of non-l ubri cated
squeezi ng fl ow (Sec. 4.8.2) whi ch generates a combi ned shear and
extensi onal deformati on.
4.2. Uniaxial Extension
Consi der the uni axi al extensi on (Fi g. 1.24) of a materi al where one
end i s stati onary and the other end i s movi ng at some vel oci ty, . Thi s
can be accompl i shed experi mental l y i f one end i s fi xed (by gl ui ng or
cl ampi ng) and the other end i s pul l ed by a movi ng cl amp, wrappi ng i t
around a rod, or pul l i ng i t between rotati ng gears. The di fferenti al
Hencky strai n (Eq. [1.2]: ) descri bi ng di spl acement i s
[4.1]
and the strai n rate i s
[4.2]
Si nce i s the vel oci ty at the end of the sampl e, the strai n rate can
be expressed as
[4.3]
u
1
h
ln(L/L
o
)
d
h

dL
L
h

d
h
dt

1
L
dL
dt
dL/dt
h

u
1
L
256 Chapter 4. Extensional Flow
I f the vel oci ty i s hel d constant ( )duri ng experi mentati on, then the
strai n rate woul d conti nual l y decrease duri ng el ongati on due to
i ncreasi ng val ues of . At very l ow speeds, a constant vel oci ty test may
approxi mate a constant strai n rate; however, to mai ntai n a trul y
constant val ue of the strai n rate, the vel oci ty of the movi ng end of the
sampl e must be i ncreased duri ng testi ng.
I ntegrati ng Eq. [4.2] for a constant strai n rate ( ), usi ng as the
i ni ti al sampl e l ength, gi ves
[4.4]
resul ti ng i n the fol l owi ng expressi on for as a functi on of ti me:
[4.5]
To produce a constant strai n rate, Eq. [4.3] shows that .
Combi ni ng thi s expressi on and Eq. [4.5] demonstrates that the vel oci ty
at the movi ng end of the sampl e must exponenti al l y i ncrease over ti me
to mai ntai n a constant strai n rate:
[4.6]
Assumi ng the materi al i s i ncompressi bl e, the vol ume i s constant so
[4.7]
where i s the i ni ti al crossecti onal area. Combi ni ng Eq. [4.5] and [4.7]
shows the crossecti onal area of the sampl e must decrease exponenti al l y
duri ng testi ng as a consequence of the exponenti al change i n l ength:
[4.8]
The normal stress di fference over the materi al defi nes the net tensi l e
stress:
[4.9]
i s the force requi red to stretch the sampl e. Usi ng Eq. [4.7], the
stretchi ng stress can be expressed as
[4.10]
I n an i deal stretchi ng experi ment the strai n i s i ncreased to a
constant val ue ( ) i nstantaneousl y:
[4.11]
u
1
u
o
L
ho
L
o
0
t
ho
dt

L
o
L
dL
L
L
L f (t ) L
o
exp(
ho
t )
L u
1
/
ho
u
1
f (t )

ho
L
o
exp(
ho
t )
AL A
o
L
o
A
o
A f (t ) A
o
exp(

ho
t )
E

11

22

F
A
F
E

FL
A
o
L
o
ho
ho
for t 0
4.2 Uniaxial Extension 257
whi ch establ i shes the vel oci ty functi on descri bed by Eq. [4.6]. The force
( ) requi red to mai ntai n i s experi mental l y determi ned duri ng
testi ng. Resul ts may be presented i n terms of the tensi l e growth functi on
defi ned by combi ni ng Eq. [4.5] and [4.10]:
[4.12]
The stress growth functi on becomes equal tothe tensi l e vi scosi ty as ti me
goes to i nfi ni ty and a constant val ue of the stress ( ) i s obtai ned:
[4.13]
where and . Bagl ey and Chri sti anson (1988) have
noted that thi s l i mi t may not be observed over experi mental l y practi cal
ti me peri ods.
Mei ssner (1972)devi sed a method of achi evi ng a constant strai n rate
i n pol ymer mel ts usi ng a devi ce (Fi g. 4.1)consi sti ng of twosets of counter
rotati ng gears movi ng at constant speeds. The strai n rate i s rel ated to
the constant sampl e vel oci ti es, i n opposi te di recti ons ( and ), l ocated
a constant di stance ( ) apart:
[4.14]
Constant vel oci ti es can be used because the sampl e l ength i s constant.
The tensi l e force, requi red to mai ntai n a constant strai n rate, i s
measured duri ng testi ng and used to cal cul ate the stretchi ng stress.
Extensi onal vi scosi ty i s determi ned from Eq. [4.12]. I n the ori gi nal
devi ce (Mei ssner, 1972), sampl es were supported i n a si l i cone oi l bath.
Rheometri cs, I nc. (Pi scataway, NJ), i ntroduced a commerci al versi on of
thi s i nstrument, i ntended for pol ymer mel ts, whi ch uses an i nert gas to
support the sampl e.
F(t )

ho
E
+
f (
ho
, t )

E
ho
FL
A
o
L
o
ho
F exp(
ho
t )
A
o
ho
E
lim
t
E
+
E
+
f (
ho
, t )
E
f (
ho
)
u
a
u
b
L
o
ho

u
a
+ u
b
L
o
Figure 4.1. Counter rotating clamps, based on the design of Meissner (1972), to
achieve a constant strain rate in uniaxial extension.
4.3. Biaxial Extension
Bi axi al extensi on can be achi eved i n a l ubri cated squeezi ng fl ow
between paral l el pl ates. Fi g. 4.2 i l l ustrates the normal case where the
l ower pl ate i s fi xed and the upper pl ate moves verti cal l y downward.
Duri ng squeezi ng, the cyl i ndri cal shape i s mai ntai ned whi l e the area
i n contact wi th the pl ate i ncreases as the hei ght of the sampl e decreases.
Test sampl es are l ubri cated wi th a l ower vi scosi ty l i qui d.
The i mportance of proper l ubri cati on cannot be underesti mated.
Unl ubri cated sampl es requi re more force to achi eve deformati on than
that requi red for l ubri cated sampl es (Chri sti anson et al ., 1985). Thi s
i s even true i n food materi al s, such as butter (Fi g. 4.3), that have a hi gh
fat content and may appear to be sel f-l ubri cati ng. The extra force i s
needed to overcome the fri cti on i ntroduced by the shear deformati on
that occurs i n unl ubri cated testi ng. Perfectl y l ubri cated sampl es
experi ence onl y extensi onal deformati on.
L
o
sample
4.3 Biaxial Extension 259
Figure 4.2. Flow between lubricated parallel plates to create sample deformation
in biaxial extension.
Figure 4.3. Force versus compression for lubricated and unlubricated squeezing
flow of butter at 15 C: = 15 mm, = 14mm, = 100 mm/min. (Data from
Rohm, 1993))
t = 0 t > 0
h
h(t)
o
Lubricated
Surface
r
o
r(t)
0 2 4 6 8 10
0
20
40
60
80
100
Compression, mm
F
o
r
c
e
,

N
lubricated plates unlubricated plates
Butter
F
compression
h
o
R
o
u
z
The vel oci ty di stri buti on i n bi axi al extensi on can be expressed i n
terms of Hencky strai n as
[4.15]
[4.16]
[4.17]
where . These equati ons are a sl i ghtl y modi fi ed versi on of Eq.
[1.67], [1.68] and [1.69]. Compressi ve strai n i n the verti cal di recti on i s
[4.18]
where , the hei ght separati ng the pl ates. The strai n rate i s
[4.19]
To produce a constant strai n rate duri ng experi mentati on, the
vel oci ty of the movi ng pl ate must decrease as the pl ates approach each
other. Fol l owi ng the same procedure gi ven for uni axi al extensi on, the
strai n rate equati on descri bi ng verti cal moti on (Eq. [4.19]) can be
i ntegrated to gi ve
[4.20]
or
[4.21]
where i s the i ni ti al sampl e thi ckness. Eq. [4.21] shows that the
di stance separati ng the pl ates must exponenti al l y decrease over ti me.
Assumi ng the materi al i s i ncompressi bl e and the vol ume i s constant
yi el ds:
[4.22]
where i s the i ni ti al area of the sampl e i n contact wi th one pl ate.
Bi axi al strai n i s i nstantaneousl y i ncreased to a constant val ue ( )
i n an i deal experi ment:
[4.23]
whi ch determi nes the pl ate spaci ng requi rements speci fi ed by Eq. [4.21].
Bi axi al stress (radi al net stretchi ng stress) i s determi ned from the
squeezi ng force ( ) whi ch i s measured duri ng testi ng:
u
z
2
B
z

h
z
u
r

B
r
h
r
2
u
h
/2
d
h

dh
h
h f (t )
h

d
h
dt

1
h
_
,
dh
dt
0
t
ho
dt

h
o
h
dh
h
h f (t ) h
o
exp(

ho
t )
h
o
Ah A
o
h
o
A
o
Bo
Bo
for t 0
F f (t )
4.3 Biaxial Extension 261
[4.24]
A bi axi al extensi onal growth functi on i s defi ned by combi ni ng Eq. [4.21]
and Eq. [4.24]
[4.25]
The stress growth functi on becomes equal to the bi axi al extensi onal
vi scosi ty as ti me goes to i nfi ni ty and a constant val ue of the net
stretchi ng stress ( ) i s obtai ned:
[4.26]
where and . Obtai ni ng steady-state condi ti ons
duri ng experi mentati on may be di ffi cul t. Compari ng growth functi ons
(Eq. [4.25]) al one may be suffi ci ent to di sti ngui sh food sampl es.
I f the l ubri cated squeezi ng fl ow experi ment i s operated so the
downward vel oci ty of the upper pl ate i s constant and the bottom pl ate
i s fi xed (the common mode of operati on i n testi ng equi pment such as
the I nstron Uni versal Testi ng Machi ne), then sampl e hei ght decreases
l i nearl y:
[4.27]
The bi axi al extensi onal strai n rate (al so cal l ed the radi al extensi on rate)
i s equal to one-hal f the verti cal Hencky strai n rate:
[4.28]
Extensi onal vi scosi ty i s cal cul ated from the net stretchi ng stress and
the strai n rate:
[4.29]
i s obtai ned from one of the fol l owi ng equati ons dependi ng on the
degree of fi l l between the pl ates duri ng testi ng:
[4.30]
or
B

rr

zz

F
A

Fh
A
o
h
o
B
+
f (
Bo
, t )

B
Bo
2F exp(
ho
t )
A
o
ho
F exp(2
Bo
t )
A
o
Bo
B
lim
t
B
+
B
+
f (
Bo
, t )
B
f (
Bo
)
h f (t ) h
o
u
z
t
1
2
_
1
2h
_
,
dh
dt

u
z
2(h
o
u
z
t )
B
f (t )

B

B
2(h
o
u
z
t )
u
z
B

F
r
2
(partially full gap where r R)
[4.31]
where i s the radi us of the sampl e, i s the radi us of the pl ate, and
i s the experi mental l y determi ned force pushi ng down on the upper pl ate.
When the gap i s fi l l i ng wi th an i ncompressi bl e materi al , the sampl e
vol ume i s constant maki ng . Thi s expressi on can be sol ved
for al l owi ng Eq. [4.30] to be cal cul ated i n terms of :
[4.32]
where i s the i ni ti al radi us of the sampl e.
Hi gh l evel s of strai n can resul t i n l oss of l ubri cati on between pl ates,
and subsequent i ntroducti on of shear fl ow i nto the experi ment. The
maxi mum strai n can be cal cul ated as
[4.33]
Accordi ng to Macosko (1994), l oss of l ubri cati on i s typi cal when the
maxi mum strai n i s near 1.0. Macosko(1994)al sonotes that the decrease
i n l ubri cant thi ckness ( ) i s approxi matel y equal to the square of root
of the gap:
[4.34]
where i s the i ni ti al thi ckness of the l ubri cant. Furthermore, he
recommends (based on Secor, 1988) the fol l owi ng cri teri on for l ubri ca-
ti on:
[4.35]
where and are the Newtoni an shear vi scosi ti es of the l ubri cant and
test sampl e, respecti vel y. These resul ts gi ve us val uabl e i nsi ght i nto
the l ubri cati on probl em. Unfortunatel y, they are not di rectl y appl i cabl e
to the bi axi al extensi on of non-Newtoni an foods. Experi mental condi -
ti ons used i n l ubri cated squeezi ng fl ow of some food materi al s are
summari zed i n Tabl e 4.1.
B

F
R
2
(full gap)
r R F
R
o
2
h
o
r
2
h
r
2
h
B

Fh
R
o
2
h
o
for r R (partially full gap)
R
o
(
B
)
max

1
2
ln
h
h
o
_
h
h
o
o
2
h
<
L
R
2
2
< 20
L

4.4 Flow Through a Converging Die 263
Typi cal data anal ysi s of bi axi al extensi onal fl ow, for the case where
the bottom pl ate i s fi xed and the upper pl ate moves downward at a
constant vel oci ty, i s di scussed i n Exampl e Probl ems 4.9.1. Resul ts show
a typi cal change (sharp i ncrease fol l owed by a gradual decrease) i n
extensi onal vi scosi ty wi th strai n rate. Exampl e Probl em 4.9.2 i l l us-
trates how resul ts from i ndi vi dual tests, such as the one gi ven i n
Exampl e 4.9.1, can be combi ned to exami ne the tensi on-thi nni ng (or
tensi on-thi ckeni ng) characteri sti cs of a sampl e.
Table 4.1. Product, Initial Sample Size, Lubricant, and Test Conditions used to
Determine Biaxial Extensional Viscosity in Lubricated Squeezing Flow
Product lubricant Reference
(cm) (cm) (cm/min)
butter 1.5 2.80 paraffin oil
1
0.1,1.0,10 Rohm, 1993
butter 1.0 3.67 cooking oil
2
0.05 Shukla et al.,
1995
starch 1.2-4.0 3.85 paraffin oil
3
0.5 Christianson
gels et al., 1985
cheese 2.0-4.0 2.85 paraffin oil
3
0.5 - 5.0 Casiraghi et
al., 1985
cheese 3.18 silicone oil
4
1.0 Campanella et
al., 1987
wheat 2.0 3.1 paraffin oil
3
0.2 - 5.0 Bagley and
dough Christianson,
1986
wheat 0.54-1.92 5.72-8.0 Super Lube
5
constant Huang and
dough stress Kokini, 1993
1
Merck 7174; 70 mPa s at 20 C;
2
PAM, American Home Foods;
3
USP/TCC white,
light. Fisher Scientific Co.;
4
Dow Corning 500;
5
Super Lube, Inc., Bohemia, NY
4.4. Flow Through a Converging Die
Fl ow i nto a convergence i nvol ves an energy l oss due to shear and an
addi ti onal l oss due to the extensi on (stretchi ng) of fl ui d streaml i nes.
Convergi ng fl ow i s someti mes cal l ed uncontrol l ed fl ow because fl ui d
streaml i nes are a functi on of fl ui d properti es. I t i s al so not pure
extensi onal fl ow because i t i nvol ves a combi nati on of both shear and
extensi onal deformati on. Anal yses descri bed here are based on sepa-
rati ng the entrance pressure drop i ntotwocomponents: one due toshear
and the other due to extensi on. An al ternati ve method, cal l ed the si nk
fl ow anal ysi s based on the work of Metzner and Metzner (1970), assumes
h
o
R
o
u
z

pure (no shear) extensi onal fl ow. Thi s method requi res knowl edge of
an angl e of convergence typi cal l y found from fl ow vi sual i zati on. Another
techni que, i nvol vi ng energy pri nci pl es, has been devel oped by Bi ndi ng
(1988).
Figure 4.4. Shear (showing velocity profile) and extensional flow components
causing pressure drop in a convergence.
4.4.1. Cogswells Equations
Cogswel l (1972) assumed the entry pressure drop over an area of
convergi ng fl ow, from a ci rcul ar barrel i nto a capi l l ary di e, was made
up of two components, one rel ated to shear fl ow and one rel ated to
extensi onal fl ow (Fi g. 4.4):
[4.36]
Thi s fl ow si tuati on coul d be referred to as unl ubri cated, constrai ned
convergence (Cogswel l , 1978). I t i s assumed that no sl i p occurs at the
wal l .
The pressure drop due to shear can be deri ved by consi deri ng the
di fferenti al pressure drop (based on the power l aw form of the fl ow rate
equati on, Eq. [2.31]) over the l ength at radi us (Fi g. 4.5):
Shear Flow Extensional Flow
P
en
P
en, S
+ P
en, E
dl r
4.4.1 Cogswells Equations 265
Figure 4.5. Geometry for converging flow from a barrel of radius into a capil-
lary die of radius .
[4.37]
Si nce , . Substi tuti ng thi s equati on i nto Eq. [4.37]
yi el ds
[4.38]
whi ch must be i ntegrated between the di e and the barrel :
[4.39]
Eval uati ng the i ntegral and si mpl i fyi ng the resul t yi el ds the component
of the pressure drop i n the convergence due to shear fl ow:
[4.40]
where , the apparent wal l shear rate i n the di e.
The vol umetri c average vel oci ty at any parti cul ar crossecti on of the
di e i s
R
b
r
dl
l
R
R
b
R
d(P
en, S
)
Q
n
r
3n + 1
3n + 1
n
_
,
n
2Kdl
r/l tan dl dr/ tan
d(P
en, S
)
Q
n
r
3n + 1
3n + 1
n
_
,
n
2K
tan
dr
0
P
en, S
d(P
en, S
)

R
R
b
Q
n
r
3n + 1
3n + 1
n
_
,
n
2K
tan
dr
P
en, S

n
3n + 1
4n
_
,
n
2K
3n tan
R
R
b
_
,
3n
_
,
4Q/(R
3
)
[4.41]
Eq. [4.41] may be di fferenti ated to gi ve the average extensi onal strai n
rate at each crossecti on:
[4.42]
Unl i ke the shear rate, thi s expressi on does not depend on the form of
the vel oci ty profi l e. The di fferenti al pressure drop, due tothe di ssi pati on
of extensi onal energy, may be wri tten i n terms of an average extensi onal
stress acti ng on an annul us:
[4.43]
or, recogni zi ng that i s negl i gi bl e, Eq. [4.43] i s si mpl y
[4.44]
Assumi ng a power l aw rel ati onshi p between the average stress and
the average strai n rate,
[4.45]
al l ows Eq. [4.44] to be rewri tten as
[4.46]
Substi tuti ng Eq. [4.42], for the average strai n rate, i ntoEq. [4.46] yi el ds
[4.47]
Eq. [4.47] must be i ntegrated between the radi us of the capi l l ary ( )
and the radi us of the barrel ( ) to determi ne the component of the
entrance pressure l oss due to extensi onal fl ow:
[4.48]
Eval uati ng the i ntegral s and si mpl i fyi ng the resul t gi ves
[4.49]
u
Q
r
2

Q
l
2
tan
2
d u
dl
_
,

(2Q)
l
3
tan
2

2Q tan
r
3
(dP
en, E
) r
2

E
((r + dr)
2
r
2
)
(dr)
2
d(P
en, E
) 2
E
dr
r
E
K
E
(
E
)
m
d(P
en, E
) 2K
E
(
E
)
m
dr
r
d(P
en, E
) 2K
E
2Q tan
r
3
_
,
m
dr
r
R
R
b
0
P
en, E
d(P
en, E
)

R
R
b
2K
E
2Q tan
r
3
_
,
m
dr
r
P
en, E

m
2K
E
3m
_
tan
2
_
,
m
R
R
b
_
,
3m
_
,
4.4.1 Cogswells Equations 267
Taki ng the l ogari thm of thi s equati on yi el ds
[4.50]
whi ch can be used i n regressi on anal ysi s to cal cul ate from the sl ope
of the l i ne, and from the i ntercept. Val ues of and may be used
to cal cul ate the average extensi onal vi scosi ty:
[4.51]
where i s the average extensi onal strai n rate at the di e (where )
computed from Eq. [4.42] as
[4.52]
Usi ng the above equati ons, the fol l owi ng procedure can be uti l i zed
to cal cul ate extensi onal vi scosi ty from di e entry pressure data:
1. Determi ne the total entrance pressure l oss ( ) usi ng the Bagl ey
procedure di scussed i n Sec. 2.5. Al so, determi ne the shear fl ow
rheol ogi cal parameters ( ) usi ng standard methods i n capi l l ary
vi scometry presented i n Chapter 2. Al though i t may be conveni ent
to fi nd shear properti es from capi l l ary data, any standard rheo-
l ogi cal techni que coul d be used to determi ne and .
2. Usi ng Eq. [4.40], fi nd the pressure drop i n the convergence due to
shear fl ow, .
3. Subtract thi s from , found i n Step 1, to yi el d the component of
the pressure drop i n the entrance due to extensi onal fl ow, .
4. Use Eq. [4.50] i n a regressi on anal ysi s procedure to eval uate and
. Cal cul ate the strai n rate and the extensi onal vi scosi ty from
Eq. [4.52] and [4.51], respecti vel y.
The precedi ng equati ons descri be behavi or acceptabl y up to di e
angl es of approxi matel y 45 degrees (Gi bson, 1988). When , shear
fl ow i s domi nant; when , materi al s may form thei r own conver-
gence pattern (Cogswel l , 1981) resul ti ng i n an unknown entry angl e.
Formati on of a product convergence pattern wi l l be a functi on of the
ln(P
en, E
) m ln+ ln
2K
E
3m
_
tan
2
_
,
m
R
R
b
_
,
3m
_
,
1
1
]
m
K
E
K
E
m
E
K
E
(
E, R
)
m 1
E, R
r R
E, R

tan
2
P
en
K, n
K n
P
en, S
P
en
P
en, E
m
K
E
< 10
> 45
rheol ogi cal properti es. I t may be a parti cul arl y si gni fi cant probl em wi th
foods havi ng a hi gh yi el d stress. Cogswel l s equati ons are used to
i nvesti gate soy dough i n Exampl e Probl em 4.9.3.
Cogswel l (1972) formul ated expressi ons for the net average exten-
si onal stress and net average extensi onal strai n for fl at entry di es (Fi g.
4.5 wi th ) as
[4.53]
and
[4.54]
where i s the apparent shear vi scosi ty based on a power l aw rel a-
ti onshi p: . The average extensi onal vi scosi ty i s easi l y cal cu-
l ated from Eq. [4.53] and [4.54]:
[4.55]
Eq. [4.55] i s very conveni ent for maki ng a rapi d compari son between
fl ui ds.
An al ternati ve anal ysi s for convergi ng fl ow has been proposed by
Bi ndi ng (1988) and used to eval uate corn meal dough (Padmanabhan
and Bhattacharya, 1993) and pol ymer mel ts (Padmanabhan and
Bhattacharya, 1994). The techni que i nvol ves axi s-symmetri c fl ow but
emphasi zes pl anar contracti on whi ch i nvol ves the convergence from a
rectangul ar channel i nto a rectangul ar ori fi ce. Al so, a refi ned versi on
of Cogswel l s method has been proposed, and successful l y used for
mol ten pol ymers, by Bersted (1993). Thi s anal ysi s does not requi re a
constant extensi onal vi scosi ty i n the convergence, but an i terati ve
procedure i s needed to fi nd a sol uti on to the pressure drop equati ons.
4.4.2. Gibsons Equations
Usi ng spheri cal coordi nates, Gi bson (1988) devel oped a method of
determi ni ng extensi onal vi scosi ty for the ful l range of di e angl es up to
90 degrees. The anal ysi s i ncl uded a power l aw model (Eq. [4.45])rel ati ng
average extensi onal stress and strai n. Components of the pressure drop
due to shear fl ow and extensi onal fl ow were found to be
90
E

3
8
(n + 1) P
en
E

4
2
3(n + 1) P
en
K()
n 1
E

E
9(n + 1)
2
(P
en
)
2
32
2
4.4.2 Gibsons Equations 269
[4.56]
where i s expressed i n radi ans and
[4.57]
respecti vel y. The di e exi t effect i ntegral ( ) was gi ven as a functi on of
and the angl e of convergence:
[4.58]
can be i ntegrated di rectl y at but a numeri cal sol uti on i s requi red
at other val ues of . Sol uti ons to Eq. [4.58] coveri ng most practi cal
si tuati ons are gi ven i n Tabl e 4.2 and Fi g. 4.6. I nterpol ati on between
di fferent val ues of and di e entry angl es (Tabl e 4.2) does not i ntroduce
a si gni fi cant l evel of error. Equati ons gi ven i n Fi g. 4.6 were generated
from l i near regressi on anal yses of the i nformati on provi ded i n Tabl e
4.2.
The average di e exi t el ongati onal strai n rate i s
[4.59]
Procedures for fi ndi ng , , and are the same as those outl i ned,
i n the precedi ng secti on, for Cogswel l s method. Val ues of and are
determi ned from raw data of versus . Taki ng the l ogari thm of
Eq. [4.57] yi el ds
[4.60]
showi ng that i s the sl ope of the resul ti ng l i ne. The numeri cal val ue
of i s determi ned from the i ntercept. Extensi onal vi scosi ty i s
cal cul ated from Eq. [4.51] wi th defi ned by Eq. [4.59]. The l ast term
i n Eq. [4.60] (equal to ) i s smal l and often negl i gi bl e for di e angl es
l ess than 45 degrees. Thi s anal ysi s, l i ke that of Cogswel l , may be
compl i cated by the tendency of foods wi th a hi gh yi el d stress to form a
natural angl e of convergence. Gi bsons equati ons are used toi nvesti gate
soy dough i n Exampl e Probl em 4.9.4.
P
en, S

2K(sin
3n
)
3n
1 + 3n
1 + 3n
4n
_
,
n
R
R
b
_
,
3n
_
P
en, E
K
E
2
3m
(sin) (1 + cos )
4
_
,
m
R
R
b
_
,
3m
+

4
m
_
,
1
1
]
m
f (m, )

(sin
m + 1
) (1 + cos )
m 1
d
m 1
m
m
E, R

(sin) (1 + cos )
4
P
en, E
K n
K
E
m
P
en, E

ln(P
en, E
) m ln() + ln
K
E
2
3m
(sin) (1 + cos )
4
_
,
m
R
R
b
_
,
3m
+

4
m
_
,
1
1
]
1
1
]
m
K
E
E, R
/4
m
Table 4.2. Numerical Solutions of the Die Exit Effect Integral (fromGibson, 1988):
0.1 0.0660 0.1558 0.2882 0.4677 0.7003
0.2 0.0629 0.1535 0.2892 0.4730 0.7080
0.3 0.0601 0.1515 0.2904 0.4787 0.7162
0.4 0.0575 0.1497 0.2920 0.4848 0.7249
0.5 0.0550 0.1481 0.2939 0.4913 0.7339
0.6 0.0528 0.1467 0.2961 0.4983 0.7434
0.7 0.0507 0.1455 0.2985 0.5055 0.7533
0.8 0.0488 0.1444 0.3011 0.5132 0.7636
0.9 0.0470 0.1435 0.3040 0.5212 0.7743
1.0 0.0453 0.1427 0.3071 0.5295 0.7854
1.2 0.0422 0.1414 0.3139 0.5472 0.8087
1.4 0.0394 0.1406 0.3215 0.5662 0.8334
1.6 0.0370 0.1401 0.3299 0.5865 0.8597
1.8 0.0348 0.1399 0.3391 0.6082 0.8874
2.0 0.0328 0.1399 0.3490 0.6313 0.9167
Figure 4.6. Numerical solutions of the die exit effect integral for different values
of the extensional flow behavior index ( ) and die entry angles.
f (m, )

(sin
m + 1
) (1 + cos )
m 1
d
m 30 45 60 75 90
0 0.5 1 1.5 2
0
0.2
0.4
0.6
0.8
1
Extensional Flow Behavior Index
D
i
e

E
x
i
t

E
f
f
e
c
t

I
n
t
e
g
r
a
l
30 degrees
90 degrees
75 degrees
60 degrees
45 degrees
= .6787 + .1133 m
= .4498 + .0856 m
= .2789 + .0319 m
= .1529 - .0081 m = .0643 - .0172 m
m
4.4.3 Empirical Method 271
4.4.3. Empirical Method
Extensi onal vi scosi ty may be esti mated usi ng a standard materi al .
Assumi ng maki ng (from Eq. [4.51]) , and that the shear
contri buti on to the pressure l oss i s smal l , then Eq. [4.49] shows that an
average extensi onal vi scosi ty i s proporti onal to the entrance pressure
drop di vi ded by the apparent wal l shear rate i n the di e:
[4.61]
where i s a di mensi onl ess constant assumed to be a functi on of the
system geometry, not strai n rate or the rheol ogi cal properti es of the
sampl e. The numeri cal val ue of coul d be esti mated usi ng a standard
Newtoni an materi al wi th a known val ue of .
Eq. [4.61] can al so be used wi th a zero l ength di e where i t i s assumed
the enti re pressure drop i s the entrance l oss. Thi s i dea i s i l l ustrated i n
Fi g. 4.7 where a pl unger, movi ng downward at a constant vel oci ty ( ),
i s forci ng materi al through an ori fi ce wi th a 90 degree entry angl e. The
entrance pressure l oss i s cal cul ated from the force on the pl unger and
the crossecti onal area of the barrel : . Assumi ng the test
materi al i s i ncompressi bl e, the vol umetri c fl ow through the ori fi ce i s a
functi on of the pl unger vel oci ty ( ). Gi ven Eq. [4.61] and the
above defi ni ti ons, extensi onal vi scosi ty may be esti mated as
[4.62]
Eq. [4.62] can be used, for both tapered or fl at entry di es, as the basi s
of a qual i ty control test i f maki ng a rel ati ve compari son between the
extensi onal vi scosi ti es of si mi l ar foods. A commerci al l y manufactured
on-l i ne system (Rheometri cs, I nc., Pi scatawa, NJ), based on thi s pri n-
ci pl e, i s avai l abl e to the food i ndustry. The uni t has been used
successful l y wi th cooki e and cracker dough. I t shoul d be noted that
coul d be a functi on of the strai n rate and the rheol ogi cal properti es of
the test substance. Thi s may be parti cul arl y i mportant for hi gh yi el d
stress materi al s tested i n i nstruments constructed wi th l arge entry
angl es. Menji var et al . (1992) used data from a zero l ength di e to
cal cul ate Trouton rati os whi ch correl ated wel l wi th the extrudate swel l
behavi or of wheat fl our doughs.
m 1
E
K
E
E
C
P
en
C
C
E
u
z
P
en
F/(R
b
2
)
Q u
z
R
b
2
E
C
P
en
C
4
_
FR
3
u
z
R
b
4
_
,
C
Figure 4.7. Flow through an orifice (zero-length die) controlled by the constant
downward velocity of the plunger.
4.5. Opposing J ets
I n the method of opposi ng jets, opposi ng nozzl es are compl etel y
i mmersed i n a test fl ui d (Fi g. 4.8). Usi ng a vacuum, fl ui d i s sucked i nto
the nozzl es creati ng uni axi al extensi onal fl ow i n the regi on between
them. Thi s fl ow causes a tensi l e stress whi ch, i f unrestrai ned, woul d
cause the nozzl es toapproach each other. I n experi mentati on, one nozzl e
i s fi xed and the other nozzl e i s restrai ned from movement but al l owed
totransmi t a resul tant force toan appropri ate transducer. A momentum
bal ance i ndi cates that pressure and momentum forces cancel , so the
force ( )measured at the transducer represents the tensi l e stress (Ful l er
et al ., 1987). An opposi ng jet devi ce coul d al so be used to create uni axi al
compressi ve fl ow i f fl ui d was forced to move out of the nozzl es i nstead
of bei ng sucked i nto them.
R
R
b
F
Side View
Bottom View
bottom plate
barrel
orifice
plunger
u = constant
z
F
4.5 Opposing Jets 273
Figure 4.8. Opposing jets configuration to achieve uniaxial extensional flow by
sucking fluid into the nozzles.
The average extensi onal strai n rate may be taken as the mean
vel oci ty at the nozzl e di vi ded by one-hal f the di stance separati ng the
nozzl es:
[4.63]
where i s the di stance separati ng the nozzl es, i s the total vol umetri c
fl ow rate i nto both nozzl es, and i s the cross-secti onal fl ow area of the
nozzl e. Extensi onal vi scosi ty i s cal cul ated by di vi di ng the mean tensi l e
stress ( ) by the strai n rate:
[4.64]
Si nce the strai n rate i s an average val ue and the resi dence ti me of fl ui d
el ements enteri ng the ori fi ce i s non-uni form, the val ue cal cul ated usi ng
Eq. [4.64] shoul d be consi dered an average extensi onal vi scosi ty.
Usi ng the opposi ng jet method, Ful l er et al . (1987)were abl e toobtai n
good resul ts (approxi matel y correct Trouton rati os)for Newtoni an fl ui ds
consi sti ng of gl yceri n-water mi xtures. Onl y qual i tati ve di fferences,
however, were observed for non-Newtoni an fl ui ds made from pol ymers
(Xanthan gum and pol yacryl ami de) di ssol ved i n mi xtures of gl yceri n
and water. I n the Ful l er et al . (1987) experi ments, a nozzl e di ameter
of 1mm was used and, i n most testi ng, the separati on di stance between
b
Q
A
_
1
b/2
_
,

2Q
Ab
b Q
A
F/A
E

11

22
F/A
2Q/(Ab)

Fb
2Q
nozzl es was 1mm. A commerci al i nstrument based on the opposi ng jet
pri nci pl e i s avai l abl e (Rheometri cs, I nc., Pi scatawa, NJ). Thi s method
of measurement may yi el d val uabl e resul ts for numerous fl ui d food
systems such as pancake syrup where stri ngi ness (formati on of thi n
threads) i s an i mportant factor i n eval uati ng qual i ty. Si mi l ar consi d-
erati ons may be associ ated wi th thi ckened dri nks and oral drugs usi ng
a fl ui d carri er (cough syrup, pai n drugs, and anti bi oti cs).
4.6. Spinning
Figure 4.9. Spinning apparatus to evaluate extensional viscosity.
Spi nni ng i s a method of eval uati ng extensi onal vi scosi ty of moderate
vi scosi ty fl ui ds by subjecti ng the sampl e to a uni axi al el ongati on.
Extrudate drawi ng, mel t spi nni ng, and fi ber spi nni ng are synonyms for
the procedure. The test i s conducted by extrudi ng a sampl e from a smal l
di ameter tube on to a rotati ng drum or wheel (Fi g. 4.9). Extensi on rates
experi enced by the materi al may be vari ed by changi ng the wi nd-up
speed of the drum. Stress on the sampl e i s determi ned from the force
measured on the drum.
L
1
2
F
4.6 Spinning 275
Experi mental data from a spi nni ng test i s ordi nari l y eval uated usi ng
a si mpl i fi ed anal ysi s produci ng average val ues. Assumi ng the extensi on
rate i s constant over the l ength of the sampl e, the average rate of
extensi on may be cal cul ated as
[4.65]
where and are the mean vel oci ti es at poi nts 1 and 2 separated by
a di stance equal to (Fi g. 4.9). Taki ng as the vol umetri c fl ow rate
through the extrusi on tube, Eq. [4.65] may be wri tten as
[4.66]
where and are the radi i at poi nts 1 and 2, respecti vel y. These radi i
are usual l y determi ned by photographi c methods. Strai n i mposed on
the sampl e duri ng testi ng i s cal cul ated as
[4.67]
An average extensi onal stress i s determi ned from the tensi l e force
on the sampl e:
[4.68]
where i s taken as the average radi us over the l ength of the sampl e:
. Usi ng Eq. [4.66] and [4.68], an average extensi onal
vi scosi ty may be cal cul ated:
[4.69]
Spi nni ng tests have commonl y been conducted on pol ymer mel ts and
sol uti ons. The method has al so been used as a means of eval uati ng the
stretchabi l i ty of Mozzarel l a cheese (Cavel l a et al ., 1992). Thi s research
showed that cheese reached a maxi mum stretchabi l i ty over a wel l
defi ned temperature range. Experi mentati on al so al l owed an accurate
eval uati on of maxi mum el ongati on and sampl e strength. I nformati on
of thi s type may be very useful i n compari ng and screeni ng di fferent
Mozzarel l a cheeses for use wi th pi zza and pasta products.
E

u
2
u
1
L
u
1
u
2
L Q
E

1
L
Q
R
2
2
Q
R
1
2
_
,

Q
L
1
R
2
2
1
R
1
2
_
,
R
1
R
2
E
ln
u
2
u
1
_
,
ln
R
1
2
R
2
2
_
E

F
R
2
R
R (R
1
+ R
2
)/2
E

E
FL
R
2
Q
R
2
2
1
R
2
R
1
_
,
2
_
,
4.7. Tubeless Siphon (Fano Flow)
Figure 4.10. Apparatus to evaluate extensional viscosity using the tubeless
siphon phenomenon.
Data col l ected empl oyi ng the tubel ess si phon phenomenon (al so
cal l ed open si phon or Fano fl ow) are useful i n eval uati ng extensi onal
behavi or. Testi ng i s carri ed out usi ng a l ow pressure reservoi r wi th a
protrudi ng capi l l ary tube (Fi g. 4.10). The upper contai ner i s l owered
unti l the tube touches the fl ui d contai ned i n the bottom vessel . Then,
the vacuum i s appl i ed and the tube i s sl owl y rai sed al l owi ng a free
standi ng col umn of fl ui d to be formed. An advantage of thi s method,
over the spi nni ng techni que, i s that the sampl e recei ves l ess severe
treatment pri or to testi ng. Strai n rates and extensi onal vi scosi ti es can
be cal cul ated usi ng the same consi derati ons di scussed for spi nni ng.
Comparati ve data can be easi l y col l ected merel y by determi ni ng the
maxi mum col umn hei ght that can be obtai ned wi th a constant vacuum.
4.8. Steady Shear Properties from Squeezing Flow Data
Bi axi al extensi onal vi scosi ty can be determi ned from squeezi ng fl ow
data (Sec. 4.3). Al ternati ve i nterpretati ons of these data are presented
i n thi s secti on. Power l aw fl ui d properti es, for exampl e, can al so be
vacuum
4.8.1 Lubricated Squeezing Flow 277
esti mated from squeezi ng fl ow data. I n addi ti on, paral l el pl ate equi p-
ment can be used to produce a combi ned shear and extensi onal fl ow
when the i nterface between the pl ate and the fl ui d i s not l ubri cated so
the sampl e adheres fi rml y to the pl ate. I n thi s case, Newtoni an and
power l aw fl ui d properti es as wel l as yi el d stresses may be cal cul ated.
Squeezi ng fl ow data may al so provi de empi ri cal i nformati on that can
form the basi s of a useful qual i ty control test.
4.8.1. Lubricated Squeezing Flow
Squeezi ng fl ow between paral l el pl ates can be achi eved i n many food
rheol ogy l aboratori es. When thi s deformati on i s executed between
l ubri cated pl ates, bi axi al extensi onal fl ow i s achi eved and an exten-
si onal vi scosi ty can be cal cul ated. Data from l ubri cated squeezi ng fl ow
may al so be eval uated i n a manner that produces steady shear fl ui d
parameters. Thi s i s advantageous i n some cases. Food materi al s, for
exampl e, that have a hi gh fat content (raw meat emul si ons, butter,
margari ne, soft cheese, etc.) may vi ol ate the no sl i p boundary condi ti on
requi red i n a tradi ti onal vi scometer. True sl i p, however, i s a necessary
condi ti on i n l ubri cated squeezi ng fl ow. Campanel l a and Pel eg (1987c)
have taken advantage of thi s i dea i n eval uati ng the power l aw fl ow
behavi or of peanut butter. Squeezi ng fl ow can al so be useful for very
thi ck fl ui ds whi ch are di ffi cul t to l oad i nto conventi onal vi scometers.
Figure 4.11. Lubricated squeezing flow with sample completely filling the gap
between two parallel plates.
R
F(t)
h(t)
I n l ubri cated squeezi ng fl ow of a power l aw fl ui d wi th a system
havi ng a fi xed bottom pl ate, and constant sampl e area i n contact wi th
the pl ates (meani ng the gap between the pl ates i s compl etel y ful l duri ng
testi ng, Fi g. 4.11), the force ( ) requi red to mai ntai n a constant down-
ward vel oci ty ( ) of the upper pl ate i s (Campanel l a and Pel eg, 1987c),
for a power l aw fl ui d,
[4.70]
where . Taki ng the l ogari thm of each si de yi el ds
[4.71]
where:
[4.72]
Power l aw fl ui d parameters ( ) may be determi ned from l i near
regressi on of experi mental data usi ng Eq. [4.71]. Earl y data poi nts
shoul d be negl ected: Onl y the l i near porti on of the curve, where steady
fl ow i s present, shoul d be used i n the anal ysi s. Thi s techni que i s
demonstrated to eval uate the behavi or of peanut butter i n Exampl e
Probl em 4.9.5.
The squeezi ng fl ow sol uti on for a power l aw fl ui d i n the case of fl ow
under a constant force, produci ng a constant stress when the gap i s
compl etel y ful l (Fi g. 4.11) duri ng testi ng, i s (Campanel l a and Pel eg,
1987c)
[4.73]
where:
[4.74]
and i s a constant force appl i ed to the sampl e through the upper pl ate.
Si nce contai ns both unknown properti es, mul ti pl e experi ments wi th
at l east two di fferent l oads are requi red to determi ne the numeri cal
val ues of and .
F
u
z
F
3
(n + 1)/2
R
2
Ku
z
n
h
n
h f (t ) h
o
u
z
t
ln(F) C
1
+ n ln
1
h
_
,
C
1
ln(3
(n + 1)/2
R
2
Ku
z
n
)
K, n
ln
h
h
o
_
,
C
2
t
C
2

W
3
(n + 1)/2
R
2
K
_
,
1/n
W
C
2
K n
4.8.2 Nonlubricated Squeezing Flow 279
The potenti al error i ntroduced by the exi stence of a yi el d stress
duri ng l ubri cated squeezi ng fl ow shoul d not be overl ooked. Fortunatel y,
thi s probl em i s not seri ous regardl ess of whether or not a yi el d stress i s
present: I n constant deformati on experi ments fl ow wi l l occur i rre-
specti ve of the yi el d stress, and i n constant stress experi ments, appl i ed
stresses must exceed the yi el d stress for fl ow to occur (Campanel l a and
Pel eg, 1987a).
I t i s i nstructi ve to consi der the theoreti cal rel ati onshi p between the
power l aw fl ui d parameters determi ned by the above procedures and
extensi onal vi scosi ty. Consi der, for exampl e, data obtai ned i n the
constant di spl acement mode. Eq. [4.70], i nvol vi ng steady shear
parameters for power l aw fl ui ds ( ), may be rewri tten as
[4.75]
whi ch, by consi deri ng Eq. [4.28] and [4.31], i s equi val ent to
[4.76]
Si nce , Eq. [4.76] can be expressed as
[4.77]
or, i n terms of bi axi al extensi onal vi scosi ty as,
[4.78]
A comparabl e expressi on for tensi l e extensi onal vi scosi ty was gi ven
earl i er as Eq. [4.51]. For the speci al case of a Newtoni an fl ui d ( and
), Eq. [4.78] shows the extensi onal vi scosi ty tobe si x ti mes the shear
vi scosi ty as predi cted by Eq. [1.79].
4.8.2. Nonlubricated Squeezing Flow
Nonl ubri cated squeezi ng fl ow produces a barrel i ng effect (Fi g. 4.12)
due to shear fl ow caused by adhesi on of the sampl e to the pl ates. The
resul ti ng compl ex fl ow i s not purel y vi scometri cor extensi onal , but some
combi nati on of each. Equati ons rel ati ng the force requi red to move the
pl ates and pl ate separati on di stance can be deri ved by assumi ng speci fi c
consti tuti ve rel ati onshi ps. The devi ce ori gi nal l y made to produce
nonl ubri cated squeezi ng fl ow was cal l ed a paral l el pl ate pl astometer
K(
)
n
F
R
2
3
(n + 1)/2
K
u
z
h
_
,
n
B
3
(n + 1)/2
K(
h
)
n
h
2
B
3
(n + 1)/2
2
n
K(
B
)
n
B

B
B
3
(n + 1)/2
2
n
K(
B
)
n 1
n 1
K
(Di enes and Kl emm, 1946). That termi nol ogy i s sti l l common today, but
the publ i shed l i terature contai ns numerous synonyms for the paral l el
pl ate pl astometer (Bi rd and Lei der, 1974): paral l el pl ate vi scometer,
compressi on pl astometer, transverse fl ow vi scometer, and paral -
l el -pl ate pl asti meter. More recentl y (Covey and Stanmore, 1981), i t has
been cal l ed the squeeze fi l m vi scometer.
Figure 4.12. Flow between lubricated and nonlubricated parallel plates.
I t i s i mportant to remember that the equati ons presented i n thi s
secti on are based on the no-sl i p premi se whi ch may be enhanced when
testi ng wi th metal fi xtures by usi ng vari ous methods such as pi tted
pl ates, an adhesi ve l i ke cyanoacryl ate ester (Supergl ue), emery paper
attached wi th doubl e adhesi ve tape (Rhom and Wei di nger, 1993), or by
coati ng the pl ates wi th spray l acquer and ordi nary sand (Nol an et al .,
1989; Navacki s and Bagl ey, 1983). Unl ess otherwi se noted, the sol uti ons
presented i n thi s secti on are for the case of a ful l y l oaded gap. I ni ti al
sampl e si ze i s i mportant: A suffi ci ent amount of materi al to produce a
t = 0
t = 0
t > 0
t > 0
Lubricated
Plates
Nonlubricated
Plates
lubricant
no slip
barreling
4.8.2 Nonlubricated Squeezing Flow 281
val ue of , where equal s the radi us of the pl ate and i s the
i ni ti al hei ght of the sampl e (al so the i ni ti al di stance between the pl ates),
i s recommended (Di enes and Kl emm, 1946). Devel opment of the fol -
l owi ng equati ons al so i gnores el asti c effects whi ch coul d be si gni fi cant
at hi gh squeezi ng rates.
Newtonian Fluids. Deri vati ons and sol uti ons for Newtoni an fl ui ds
and power l aw fl ui ds are gi ven i n Bi rd et al . (1987) and Lei der and Bi rd
(1974). The nonl ubri cated squeezi ng fl ow of a Newtoni an fl ui d, i n an
experi mental system usi ng an i mmobi l e bottom pl ate wi th a ful l y l oaded
gap, i s (Wi nther et al ., 1991)
[4.79]
where and , the force requi red to mai ntai n a
constant downward vel oci ty of the upper pl ate. Eq. [4.79] i s known as
the Stefan equati on. The vi scosi ty can be cal cul ated as the sl ope of
versus determi ned from regressi on anal ysi s.
The shear rate, eval uated from the vel oci ty profi l e i n the gap, i s
(Wi nther et al ., 1991; Churchi l l , 1988)
[4.80]
where i s the di stance from the center l i ne and i s the verti cal di stance
from the hori zontal mi dpl ane of the sampl e. A maxi mum shear rate i s
found at and :
[4.81]
The average shear rate i s equal to of the maxi mum shear rate.
The case for squeezi ng fl ow of a Newtoni an fl ui d wi th a constant
force, ful l gap, can be determi ned by i ntegrati ng the Stefan equati on.
Fi rst, i t must be recogni zed that , so Eq. [4.79] may be wri tten
as
[4.82]
R > 10h
o
R h
o
F
3R
4
u
z
2h
3
h f (t ) h
o
u
z
t F f (t )
F
1/h
3
f (r, z)
6rzu
z
h
3
r z
r R z h/2
max

3Ru
z
h
2
2/3
u
z
dh/dt
h
3
dh
2W
3 R
4
dt
where i s the constant force appl i ed to the sampl e. Thi s equati on i s
easi l y i ntegrated to gi ve
[4.83]
The constant of i ntegrati on i s eval uated from the i ni ti al condi ti on that
at maki ng . Substi tuti ng thi s i nto Eq. [4.83], and
si mpl i fyi ng the resul t, produces the fi nal sol uti on:
[4.84]
Vi scosi ty can be cal cul ated from the sl ope of versus . Al so, i f the
vi scosi ty was known, coul d be determi ned by sol vi ng for .
Power LawFluids. The nonl ubri cated squeezi ng fl ow of a power l aw
fl ui d, i n an experi mental system usi ng an i mmobi l e bottom pl ate wi th
a ful l gap, i s (Wi nther et al ., 1991)
[4.85]
where . Eq. [4.85] i s known as the Scott equati on and i t
reduces to the Stefan equati on when . Wi nther et al . (1991) suggest
eval uati ng Eq. [4.85] by mul ti pl yi ng both si des by gi vi ng
[4.86]
then, taki ng the l ogari thm of each si de of the equati on, to get
[4.87]
Usi ng Eq. [4.87], may be found from the sl ope of versus .
The val ue of i s determi ned from the i ntercept. Si nce the i nerti a terms
i n the equati ons of moti on are negl ected i n devel opi ng Eq. [4.87], the
earl i est data poi nts may not l i e on the l i ne and shoul d be negl ected. The
average shear rate for a power l aw fl ui d i n nonl ubri cated squeezi ng fl ow
i s (Wi nther et al ., 1991)
W
1
2h
2
2Wt
3 R
4
+ C
C 1/(2h
o
2
) h h
o
t 0
h
o
h
_
,
2
4Wh
o
2
t
3 R
4
+ 1
(h
o
/h)
2
t
h f (t ) h
F

2n + 1
n
_
,
n
2KR
n + 3
n + 3
_
u
z
n
h
2n + 1
_
,
h f (t ) h
o
u
z
t
n 1
h
Fh

2n + 1
n
_
,
n
2KR
n + 3
n + 3
_
u
z
h
2
_
,
n
ln(Fh) ln
2n + 1
n
_
,
n
2KR
n + 3
n + 3
_
,
1
1
]
+ n ln
u
z
h
2
_
,
ln(u
z
/h
2
) n ln(Fh)
K
4.9.1 Biaxial Extension of Processed Cheese Spread 283
[4.88]
I ntegrati ng Eq. [4.85] can produce a sol uti on for descri bi ng the case
where a constant wei ght i s used to deform a power l aw fl ui d. Thi s
probl em has been sol ved by Lei der and Bi rd (1974).
YieldStressEvaluation. Nonl ubri cated squeezi ng fl ow has been used
toeval uate the behavi or of Bi ngham pl asti cand Herschel -Bul kl ey fl ui ds
(Covey and Stanmore, 1981). General sol uti ons are somewhat cum-
bersome but the work veri fi ed a si mpl e procedure to determi ne the yi el d
stress of semi -sol i d materi al s. I n testi ng, a constant force ( ) i s pl aced
on sampl es whi ch compl etel y fi l l the gap between paral l el pl ates. The
yi el d stress i s cal cul ated on the basi s of the asymptoti c or resi dual
thi ckness ( ) of the sampl e:
[4.89]
The techni que works reasonabl y wel l when the materi al s tested have a
hi gh yi el d stress and the rate of deformati on produced by the constant
force i s smal l . Campanel l a and Pel eg (1987a) used thi s method to
eval uate the yi el d stresses of tomato ketchup, mustard, and mayon-
nai se. Al so, val ues of the yi el d stress, for a /treacl e paste, determi ned
wi th Eq. [4.89] compared wel l to those found by extrapol ati on of a
rheogram to zero shear rate (Covey and Stanmore, 1981).
4.9.1. Biaxial Extension of Processed Cheese Spread
Gi ven the data i n Tabl e 4.3, determi ne the bi axi al extensi onal strai n
rate and the extensi onal vi scosi ty of processed cheese spread. Al so
cal cul ate the maxi mum strai n found duri ng testi ng. Data are for the
l ubri cated squeezi ngfl ow between paral l el pl ates where the gap i s fi l l i ng
duri ng experi mentati on (Fi g. 4.12). The l ower pl ate i s fi xed and the
upper pl ate i s movi ng downward wi th a constant vel oci ty. Paraffi n oi l
was used as the l ubri cant.
average

2(2n + 1)
3n
_
u
z
R
h
2
_
,
W
h
a
o

3Wh
a
2R
3
TiO
2
Table 4.3. Data for the Lubricated Squeezing Flow ( = 0.04 m, = 0.057 m,
= 0.0005 m/min) of Processed Cheese Spread at 7 C (Data from Casiraghi et al.,
1985)
compression
(N) (m) (m) (1/s) (1/s) (MPa s)
5.1 0.0014 0.0386 0.000108 0.000216 4.5
13.8 0.0027 0.0373 0.000112 0.000227 11.3
22.9 0.0048 0.0352 0.000118 0.000236 16.7
29.6 0.0067 0.0333 0.000125 0.000250 19.3
39.1 0.0094 0.0306 0.000136 0.000272 21.5
50.0 0.0120 0.0279 0.000149 0.000298 22.9
59.5 0.0141 0.0258 0.000161 0.000322 23.3
69.7 0.0160 0.0240 0.000174 0.000348 23.5
80.8 0.0180 0.0220 0.000189 0.000378 23.0
91.5 0.0200 0.0199 0.000209 0.000418 21.3
106.3 0.0227 0.0173 0.000241 0.000482 18.7
137.1 0.0260 0.0140 0.000298 0.000596 15.8
192.3 0.0294 0.0106 0.000392 0.000784 12.7
Figure 4.13. Raw data of force versus compression for the lubricated squeezing
flow of processed cheese spread at 7 C.
h
o
R
o
u
z
F h

h

B
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035
0
50
100
150
200
250
Compression, m
F
o
r
c
e
,

N
Processed Cheese Spread
F
compression
h
4.9.1 Biaxial Extension of Processed Cheese Spread 285

Pl otti ng the raw data (Fi g. 4.13) cl earl y i l l ustrates the change i n force
duri ng compressi on. Resul ts are summari zed i n Tabl e 4.3 and the
fol l owi ng equati ons show the computati ons needed to generate the fi rst
row of i nformati on:
Bi axi al extensi onal vi scosi ty, pl otted i n terms of the Hencky strai n rate,
i s i l l ustrated i n Fi g. 4.14. The maxi mum strai n found duri ng testi ng i s
cal cul ated from Eq. [4.33]:
Si nce thi s val ue i s l ess than 1.0, i t i s unl i kel y that experi mental errors
associ ated wi th l ubri cant l oss occurred duri ng testi ng.
Figure 4.14. Biaxial extensional viscosity of processed cheese spread at 7 C.
h h
o
- compression 0.0400 0.0014 0.0386 m
B
0.5
u
z
h
_
,
0.5
.0005
(60).0386
_
,
0.000108 s
1
h
2

B
2(0.000108) 0.000216 s
1
B

Fh
R
o
2
h
o
5.14(.0386)
(.057)
2
(.040) (.000108)
4.5(10)
6
Pa s 4.5 MPa s
(
B
)
max

1
2
ln
h
h
o
_
,

1
2
ln
.0106
.04
_
,
0.66
0.0002 0.0004 0.0006 0.0008 0.001
0
5
10
15
20
25
h
.
, 1/s
B
,

M
P
a

s
Processed Cheese Spread

4.9.2. Biaxial Extension of Butter
Bi axi al extensi on data are avai l abl e for butter at 15 C and 20 C (Fi g.
4.15). These tests were conducted usi ng paral l el pl ates wi th a fi xed
l ower pl ate. The upper pl ate was l owered at di fferent downward vel o-
ci ti es i ndi cated by the crosshead speeds. Al l sampl es had an i ni ti al
di ameter of 28 mm and an i ni ti al hei ght of 15 mm. Eval uate the behavi or
of thi s materi al .
Figure 4.15. Biaxial extensional viscosity of butter at 15 C and 20 C (Data from
Rohm, 1993).
Hi gher crosshead speeds produced l arger strai n rates. I n each test,
a sharp i ni ti al i ncrease was fol l owed by a gradual decl i ne, si mi l ar to the
resul ts found for processed cheese spread (Fi g. 4.14) i n the previ ous
exampl e. Pl otti ng a l i ne through the downward sl ope of each data set
(dashed l i ne, Fi g. 4.15) suggest a power l aw rel ati onshi p between vi s-
cosi ty and strai n rate. Fi tti ng the l i ne through representati ve poi nts
yi el ds

0.001 0.002 0.005 0.01 0.02 0.05 0.1 0.2 0.5 1
15 C
20 C
Butter
100
30
10
3
1
0.3
0.1
0.03
0.01
B
,

M
P
a

s
h
, 1/s
.
A
B
C
A : 1 mm / min
B : 10 mm / min
C : 100 mm / min
Crosshead Speed

4.9.3 45 Converging Die, Cogswells Method 287
and
The sl ope terms gi ven above are anal ogous to used to express the
rel ati onshi p (Eq. [1.28]) between apparent vi scosi ty and shear rate i n
a standard power l aw fl ui d. Hence, the resul ts coul d be i nterpreted as
extensi onal -thi nni ng behavi or. Si mi l ar resul ts have been observed for
processed cheese spread. Other materi al s, such as mozzarel l a cheese,
show a l evel i ng off of extensi onal vi scosi ty after the i ni ti al sharp i ncrease
(Casi raghi et al ., 1985) maki ng the i nterpretati on of data much more
di ffi cul t.
4.9.3. 45 Converging Die, Cogswells Method
Taki ng the capi l l ary vi scometer data for soy dough presented i n
Exampl e 2.12.2, esti mate the extensi onal vi scosi ty assumi ng that the
entrance pressure l oss can be di vi ded i nto two separate components:
one rel ated to shear and one rel ated to extensi onal fl ow. Data from
Tabl e 2.8 are summari zed i n Tabl e 4.4. Assume the materi al forms a
natural convergence angl e of i n the di e. Sol ve the
probl em usi ng Cogswel l s equati ons presented i n Sec. 4.4.1.
Anal ysi s of the data conducted i n Exampl e 2.12.2 generated val ues
of the fl ow behavi or i ndex and the consi stency coeffi ci ent: = 0.29, =
23,300 Pa s
n
. Eq. [4.40] must be eval uated to determi ne the component
of the entrance pressure drop due to shear:
Gi ven, =0.00159 m and =0.0075 m, the equati on may be si mpl i fi ed:
Cal cul ati ng for each val ue of shows that the pressure drop,
due to shear fl ow, i s a smal l porti on of the total pressure drop at the
entrance (Tabl e 4.4). The component of the entrance pressure drop due
to extensi onal fl ow i s found from Eq. [4.36]:
B
9.55(10
4
) (
h
)
.08 1
at 15C
B
3.55(10
4
) (
h
)
.10 1
at 20C
n 1
/4 rad 45
n K
P
en, S
()
n

3n + 1
4n
_
,
n
2K
3n tan
R
R
b
_
,
3n
_
,
R R
b
P
en, S
()
.29
3(.29) + 1
4(.29)
_
,
.29
2(23, 300)
3(.29) tan(/4)
.00159
.0075
_
,
3(.29)
_
,
()
.29
(45, 560)
P
en, S

P
en, E
P
en
P
en, S
Resul ts are summari zed i n Tabl e 4.4.
Extensi onal fl ow parameters may be esti mated from regressi on of
Eq. [4.50]:
yi el di ng the sl ope,
and i ntercept,
Substi tuti ng appropri ate val ues i nto the i ntercept equati on gi ves
whi ch can be sol ved for the extensi onal consi stency coeffi ci ent:
Average extensi onal vi scosi ty i s determi ned from Eq. [4.51]:
wi th the average extensi onal strai n rate at the di e eval uated from Eq.
[4.52]:
The Trouton number ( ) i s cal cul ated from Eq. [1.82] where i s
eval uated, as uni axi al extensi on, at a shear rate numeri cal l y equal to
:
Resul ts are summari zed i n Tabl e 4.4.
ln(P
en, E
) m ln+ ln
2K
E
3m
_
tan
2
_
,
m
R
R
b
_
,
3m
_
,
1
1
]
m
d(lnP
en
)
d(ln)
0.159
ln
2K
E
3m
_
tan
2
_
,
m
R
R
b
_
,
3m
_
,
1
1
]
14.479
K
E
2
3(.159)
_
tan(/4)
2
_
,
(.159)
.00159
.0075
_
,
3(.159)
_
,
K
E
(1.96) exp(14.479)
K
E
991, 000 Pa s
m
E
K
E
(
E, R
)
m 1
991, 000(
E, R
)
.159 1
E, R

tan(/4)
2
(.5)
E
/
E, R
K(
E, R
)
n 1
23, 300(
E, R
)
.29 1
4.9.4 45 Converging Die, Gibsons Method 289
Table 4.4. Capillary Viscometer Data for Defatted Soy Flour Dough (34.7%) at
Room Temperature (Data from Morgan, 1979)
(1/s) (MPa) (k Pa) (MPa) (1/s) (kPa s
m
) (kPa s
n
) (-)
47.4 3.58 139.5 3.58 23.7 65.1 1.65 39
47.4 3.58 139.5 3.58 23.7 65.1 1.65 39
47.4 3.58 139.5 3.58 23.7 65.1 1.65 39
94.8 4.63 170.6 4.63 47.4 38.2 1.01 38
94.8 4.63 170.6 4.63 47.4 38.2 1.01 38
94.8 4.63 170.6 4.63 47.4 38.2 1.01 38
190.0 4.38 208.7 4.38 95.0 21.2 0.614 35
190.0 4.38 208.7 4.38 95.0 21.2 0.614 35
190.0 4.38 208.7 4.38 95.0 21.2 0.614 35
948.0 6.17 332.6 6.17 474 5.44 0.196 28
948.0 6.17 332.6 6.17 474 5.44 0.196 28
948.0 6.17 332.6 6.17 474 5.44 0.196 28
4.9.4. 45 Converging Die, Gibsons Method
Reexami ne the soy dough data gi ven i n the previ ous exampl e probl em
usi ng Gi bsons equati ons (Sec. 4.4.2) to separate the pressure drop i nto
the shear and extensi onal components. Assume the natural angl e of
convergence ( ) produced by the materi al i s radi ans or 45 degrees.
Anal ysi s of the data conducted i n Exampl e 2.12.2 gave val ues of the
fl ow behavi or i ndex and the consi stency coeffi ci ent: =0.29, =23,300
Pa s
n
. Eq. [4.56] must be eval uated to determi ne the component of the
entrance pressure drop due to shear:
Gi ven, =0.00159 m and =0.0075 m, the equati on may be si mpl i fi ed:
Entrance pressure l oss due to extensi onal fl ow i s found, from Eq. [4.36],
as
Resul ts of the pressure l oss cal cul ati ons are summari zed i n Tabl e 4.5.
Usi ng these data, extensi onal fl ow parameters are esti mated from
regressi on of Eq. [4.60]:
(P)
en
(P)
en, S
(P)
en, E

E, R

E
N
Tr
/4
n K
P
en, S

2K(sin
3n
)
3n
1 + 3n
1 + 3n
4n
_
,
n
R
R
b
_
,
3n
_
,
R R
b
P
en, S

2(23, 300) (sin
3(.29)
(/4))
3(.29) (/4)
1 + 3(.29)
1 + 3(.29)
4(.29)
_
,
.29
.29
.00159
.0075
_
,
3(.29)
_
,
52, 995( )
.29
P
en, E
P
en
P
en, S
Table 4.5. Capillary Viscometer Data for Defatted Soy Flour Dough (34.7%) at
Room Temperature (Data from Morgan, 1979)
(1/s) (MPa) (k Pa) (MPa) (1/s) (kPa s
m
) (kPa s
n
) (-)
47.4 3.58 162.3 3.42 14.3 92.1 2.38 39
47.4 3.58 162.3 3.42 14.3 92.1 2.38 39
47.4 3.58 162.3 3.42 14.3 92.1 2.38 39
94.8 4.63 198.4 4.43 28.6 51.4 1.46 35
94.8 4.63 198.4 4.43 28.6 51.4 1.46 35
94.8 4.63 198.4 4.43 28.6 51.4 1.46 35
190.0 4.38 242.7 4.14 57.4 28.6 0.889 32
190.0 4.38 242.7 4.14 57.4 28.6 0.889 32
190.0 4.38 242.7 4.14 57.4 28.6 0.889 32
948.0 6.17 386.8 5.78 286 7.40 0.284 26
948.0 6.17 386.8 5.78 286 7.40 0.284 26
948.0 6.17 386.8 5.78 286 7.40 0.284 26
yi el di ng the sl ope,
and i ntercept,
Substi tuti ng known val ues i nto the i ntercept equati on gi ves
whi ch can be sol ved for the extensi onal consi stency coeffi ci ent:
Hence, the average extensi onal vi scosi ty can be expressed as
where the average extensi onal strai n rate at the di e i s determi ned from
Eq. [4.59]:
(P)
en
(P)
en, S
(P)
en, E

E, R

E
N
Tr
ln(P
en, E
) m ln() + ln
K
E
2
3m
(sin) (1 + cos )
4
_
,
m
R
R
b
_
,
3m
+

4
m
_
,
1
1
]
1
1
]
m
d(lnP
en, E
)
d(ln)
0.158
ln
K
E
2
3m
(sin) (1 + cos )
4
_
,
m
R
R
b
_
,
3m
+

4
m
_
,
1
1
]
1
1
]
14.478
K
E
2
3(.156)
(sin(/4)) (1 + cos(/4))
4
_
,
.156
.00159
.0075
_
,
3(.156)
+
.1516
4
.156
_
,
1
1
]
K
E
(2.24) exp(14.478)
K
E
865, 467 Pa s
m
E
K
E
(
E, R
)
m 1
865, 467(
E, R
)
.158 1
4.9.5 Lubricated Squeezing Flow of Peanut Butter 291
The Trouton number ( ) i s cal cul ated from Eq. [1.82] where i s
eval uated at a shear rate of :
Resul ts are gi ven i n Tabl e 4.5.
Compari ng sol uti ons from the Cogswel l (Tabl e 4.4) and Gi bson
(Tabl e 4.5) methods shows practi cal l y i denti cal Trouton Numbers.
Val ues of the extensi onal and steady shear vi scosi ti es are the same order
of magni tude but si gni fi cantl y di fferent. Si nce fl ui d moti on i n a con-
vergence has both shear and extensi onal components, i t i s not consi d-
ered pure fl ow maki ng i t di ffi cul t tosay that one method i s quanti tati vel y
superi or to the other. Both are reasonabl e and provi de a good basi s for
i nvesti gati ng extensi onal fl ow of food i n a convergence.
4.9.5. Lubricated Squeezing Flow of Peanut Butter
Data for the l ubri cated squeezi ng fl ow of peanut butter were col l ected
for the case of constant di spl acement and constant area (Fi g. 4.11, Tabl e
4.6). Assume power l aw fl ui d behavi or, and determi ne the steady shear
rheol ogi cal properti es ( and ) of the materi al . Al so, compute the
bi axi al extensi onal strai n rate and the comparabl e shear rate.
Usi ng the l i near porti on of the curve (Fi g. 4.16: )where
steady fl ow has been achi eved, regressi on of Eq. [4.71],
where, from Eq. [4.72],
yi el ds = 0.79 and = -2.118. Sol vi ng for gi ves
The bi axi al extensi onal strai n rate i s
E, R

(sin) (1 + cos )
4

(sin(/4)) (1 + cos(/4))
4
(.302)
E
/
E, R
K(
E, R
)
n 1
23, 300(
E, R
)
.29 1
K n
1/h(t ) > 200 1/m
ln(F) C
1
+ n ln
1
h
_
,
C
1
ln(3
(n + 1)/2
R
2
Ku
z
n
)
n C
1
K
2.118 ln[3
(.79 + 1)/2
(.0318)
2
K
.05
(100) (60)
_
,
.79
]
K 146 kPa s
n
Figure 4.16. Raw data of force versus reciprocal height for the lubricated squeez-
ing flow of peanut butter at 23 C.
and a comparabl e shear rate, based on our consi derati on of the Trouton
number (Eq. [1.84]), i s
Sampl e cal cul ati ons for the fi rst data poi nt ( = 230 m
-1
) i ncl uded i n
the strai ght l i ne rel ati onshi p between and , are
and
Resul ts are summari zed i n Tabl e 4.6. The approxi mate shear rate range
covered i n determi ni ng and was 0.00332 to 0.00938 s
-1
.
100 200 300 500 700 1,000
1
2
3
5
10
20
30
50
100
1/h(t), 1/m
F
,

N
Peanut Butter
B

u
z
2(h
o
u
z
t )

u
z
2
1
h
_
12

B
1/h
F 1/h
B

u
z
2
1
h
_
,

.05(230)
2(60) (100)
9.58(10)
4
s
1
12

12 (9.58) (10)
4
3.32(10)
3
s
1
K n
4.9.5 Lubricated Squeezing Flow of Peanut Butter 293
Table 4.6. Lubricated Squeezing Flow ( = 7.14 mm, = 0.0318 m, = 0.05
cm/min) Data for Peanut Butter at 23 C(Data fromCampanella and Peleg, 1987c)
F h(t) 1/h(t) ln (F) ln(1/h(t))
N mm 1/m 1/s 1/s
2.25 6.67 150 0.811 5.012 - -
3.16 6.25 160 1.151 5.075 - -
4.50 5.88 170 1.504 5.136 - -
5.75 5.56 180 1.749 5.193 - -
6.90 5.26 190 1.932 5.247 - -
7.50 5.00 200 2.015 5.298 - -
8.50 4.44 230 2.140 5.438 9.58E-4 3.32E-3
9.50 3.85 260 2.251 5.561 1.08E-3 3.75E-3
10.9 3.33 300 2.389 5.704 1.25E-3 4.33E-3
12.3 3.03 330 2.509 5.799 1.38E-3 4.76E-3
14.0 2.50 400 2.639 5.991 1.67E-3 5.77E-3
15.3 2.27 440 2.728 6.087 1.83E-3 6.35E-3
16.7 1.96 510 2.815 6.234 2.13E-3 7.36E-3
18.2 1.67 600 2.901 6.397 2.50E-3 8.66E-3
19.5 1.54 650 2.970 6.477 2.71E-3 9.38E-3
h
o
R u
z
12

B
Chapt e r 5 . Vis c oe las t ic it y
5.1. Introduction
Rheol ogy i s the sci ence of the deformati on and fl ow of matter. There
are three ways to deform a substance: shear, extensi on, and bul k
compressi on. Shear and extensi onal deformati on have been thoroughl y
di scussed i n earl i er chapters. Bul k compressi on, where a uni form
change i n external pressure produces a vol umetri c change i n the
materi al , was bri efl y di scussed i n defi ni ng the bul k modul us (Sec. 1.4).
I t i s possi bl e to conduct tests i n al l three modes of deformati on, under
steady state or dynami c condi ti ons, and compare the resul ti ng modul i
and compl i ances (Ferry, 1980). Thi s chapter wi l l focus on vi scoel asti c
materi al functi ons determi ned from shear and extensi onal deformati on
data. Bul k compressi on testi ng i s not consi dered here because tech-
nol ogy i n the area i s not wel l devel oped. Future research may show the
concept to be val uabl e i n defi ni ng a pressure dependent vi scosi ty
functi on to exami ne hi gh pressure processes such as food extrusi on.
Viscoelastic Material Functions. I n Chapters 2 (Tube Vi scometry)
and 3 (Rotati onal Vi scometry) the emphasi s was on measurement
methods to determi ne fl ow curves for non-Newtoni an materi al s under
steady shear condi ti ons. Al l foods have uni que fl ow curves and thi s
i nformati on i s very useful i n a l arge number of i ndustri al appl i cati ons.
Cl earl y, from an engi neeri ng standpoi nt, the steady fl ow curve i s the
most val uabl e way tocharacteri ze the rheol ogi cal behavi or of fl ui d foods.
Steady shear vi scosi ty i s a property of al l fl ui ds regardl ess of whether
or not they exhi bi t el asti c behavi or; however, many phenomena cannot
be descri bed by the vi scosi ty functi on al one and el asti c behavi or must
be taken i nto consi derati on. Thi s chapter wi l l i nvesti gate experi mental
methods to generate data that refl ect the combi ned vi scous and el asti c
character of materi al s.
I n steady shear, vi scoel asti c fl ui ds exhi bi t normal stresses and
measuri ng them provi des one way of characteri zi ng el asti ci ty. Normal
stress di fferences (Eq. [1.23] and Eq. [1.24]) can be measured on rota-
ti onal rheometers produced by vari ous i nstrument compani es. Com-
putati ons (Eq. [3.57]) requi re an eval uati on of axi al force under steady
shear condi ti ons. Unsteady state shear measurements provi de a
dynami c means of eval uati ng vi scoel asti ci ty. The two major categori es
of unsteady shear testi ng are transi ent and osci l l atory.
5.1 Introduction 295
Table 5.1. Material Functions Determined in Transient Shear Flow Experiments
St art -up flow: Material at rest is suddenly subjected to a constant shear rate.
Shear stress growth function
First normal stress growth function
Second normal stress growth function
Ce s s at ion of s t e ady s he ar flow: Material undergoing steady state shear flow is
suddenly brought to rest.
Shear stress decay function
First normal stress decay function
Second normal stress decay function
St e p s t rain: Material at rest is given a sudden step increase in strain.
Shear stress relaxation function
First normal stress relaxation function
Second normal stress relaxation function
Cre e p: Material at rest is given a sudden step increase in stress.
Shear creep compliance
Steady-state compliance
Re c oil: Stress, in a fluid in steady state shear flow, is suddenly brought to zero.
Material is constrained in one direction and recoil (the degree of retraction) is mea-
sured in the second direction.
Recoil function
Ultimate recoil function
Transi ent shear testi ng (Tabl e 5.1) i s a category that i ncl udes
numerous measurement concepts: start-up fl ow, cessati on of steady
shear fl ow, step strai n, creep, and recoi l . Data generated usi ng these
methods may l ead to numerous materi al functi ons (Tabl e 5.1) such as
the shear stress growth functi on, shear stress decay functi on, shear
stress rel axati on functi on, shear creep compl i ance, and the recoi l
functi on. I n osci l l atory testi ng, a sampl e i s subjected to harmoni cal l y
varyi ng (usual l y si nusoi dal ) smal l ampl i tude deformati ons i n a si mpl e
shear fi el d. Vari ous compani es make i nstruments to accompl i sh
osci l l atory tests whi ch have proved thei r useful ness i n addressi ng
numerous food i ndustry probl ems. Many functi ons (Tabl e 5.2) can be
generated from osci l l atory experi ments. A comprehensi ve l i st of vi s-
coel asti cmateri al functi ons and coeffi ci ents i s avai l abl e i n Deal y (1994).
Extensi onal fl ow was i nvesti gated i n Chapter 4. What has been
di scussed above, rel ati ve to shear fl ow, i s al so true for extensi onal fl ow.
The extensi onal vi scosi ty functi on i s determi ned i n steady-state
extensi onal fl ow, but transi ent experi ments generate data whi ch refl ect
296 Chapter 5. Viscoelasticity
the vi scoel asti c character of the materi al . The fol l owi ng tests, for
exampl e, coul d be conducted i n tensi l e extensi on: tensi l e start-up, ces-
sati on of steady tensi l e extensi on, tensi l e step strai n, tensi l e creep, and
tensi l e recoi l . Thi s type of experi mental testi ng l eads to numerous
vi scoel asti c materi al functi ons (Tabl e 5.3). Si mi l ar functi ons coul d be
generated for experi ments i nvol vi ng bi axi al or pl anar extensi onal fl ow.
Table 5.2. Material Functions Determined in Oscillatory Shear Testing (Har-
monically Varying Simple Shear) Experiments
Complex viscosity
Dynamic viscosity
Out-of-phase component of the complex viscosity
Complex shear modulus
Shear storage modulus
Shear loss modulus
Complex shear compliance
Shear storage compliance
Shear loss compliance
Linear VersusNon-linear Viscoelasticity. I n process engi neeri ng,
data on vi scoel asti ci ty may be very hel pful i n understandi ng vari ous
probl ems. Shear creep data, for exampl e, are useful i n exami ni ng
gravi ty dri ven phenomena such as coati ng and saggi ng. Thi s i nfor-
mati on can al so be an i nval uabl e tool i n product devel opment. Means
of eval uati ng l i near vi scoel asti c behavi or are the pri mary foci of thi s
chapter. When materi al s are tested i n the l i near range, materi al
functi ons do not depend on the magni tude of the stress, the magni tude
of the deformi ng strai n, or the rate of appl i cati on of the strai n. I f l i near,
an appl i ed stress wi l l produce a proporti onal strai n response. Doubl i ng
the stress, for exampl e, wi l l doubl e the strai n response. The l i near range
of testi ng i s determi ned from experi mental data. Testi ng can easi l y
enter the non-l i near range by appl yi ng excessi ve strai n (usual l y greater
than 1%) or hi gh deformati on rates to a sampl e.
The i mportance of l arge deformati on (non-l i near) behavi or i n food
rheol ogy shoul d not be overl ooked. Many processes, such as masti cati on
and swal l owi ng, are onl y accompl i shed wi th very l arge deformati ons.
Col l ecti ng vi scoel asti c data rel evant to thi s type of probl em i nvol ves
5.2 Transient Tests for Viscoelasticity 297
testi ng i n the non-l i near range of behavi or. These data may be useful
i n attacki ng practi cal probl ems; however, from a fundamental stand-
poi nt, they can onl y be used for comparati ve purposes because the
theoreti cal compl exi ty of non-l i near vi scoel asti ci ty makes i t i mpracti cal
for most appl i cati ons. More research i s needed i n thi s area.
Table 5.3. Experimental Tests and Material functions Determined in Transient
Tensile Extension
Te ns ile s t art -up: Material at rest is suddenly subjected to a constant extensional
strain rate.
Tensile stress growth function
Ce s s at ion of s t e ady t e ns ile e xt e ns ion: Material subjected to a steady state
extensional strain rate is suddenly brought to rest.
Tensile stress decay coefficient
Te ns ile s t e p s t rain: Material at rest is given a sudden step increase in strain.
Tensile relaxation modulus
Te ns ile c re e p: Material at rest is given a sudden step increase in tensile stress.
Tensile creep compliance
Tensile creep rate decay function
Te ns ile re c oil: Material subject to steady state stress and strain has the stress
suddenly reduced to zero.
Tensile recoil function
Ultimate tensile recoil function
5.2. Transient Tests for Viscoelasticity
I n thi s secti on, three typi cal transi ent tests of vi scoel asti ci ty are
presented: step strai n (stress rel axati on), creep, and start-up fl ow.
These tests i nvol ve smal l strai ns and can be conducted wi th commer-
ci al l y avai l abl e or easi l y constructed i nstruments. Al though there are
numerous potenti al experi mental methods (Tabl e 5.1 and 5.3) to
el uci date vi scoel asti ci ty, many are di ffi cul t to perform and questi onabl e
for use on food products. Practi cal rheol ogy i s appl i cati on dri ven so the
most appropri ate test for a parti cul ar materi al depends on the probl em
under consi derati on. Tests descri bed i n thi s secti on have al l been
conducted on food, and produced useful resul ts.
5.2.1. Mechanical Analogues
Massl ess mechani cal model s, composed of spri ngs and dashpots, are
useful i n conceptual i zi ng rheol ogi cal behavi or. The spri ng i s consi dered
an i deal sol i d el ement obeyi ng Hookes l aw:
[5.1]
and the dashpot i s consi dered an i deal fl ui d el ement obeyi ng Newtons
l aw:
[5.2]
Spri ngs and dashpots can be connected i n vari ous ways to portray the
behavi or of vi scoel asti c materi al s; however, a parti cul ar combi nati on of
el ements i s not uni que because many di fferent combi nati ons can be used
to model the same set of experi mental data. The most common
mechani cal anal ogs of rheol ogi cal behavi or are the Maxwel l and Kel vi n
(someti mes cal l ed Kel vi n-Voi gt) model s depi cted i n Fi g. 5.1.
Figure 5.1. Maxwell and Kelvin models.
Mechani cal anal ogues provi de a useful means of i nvesti gati ng creep
and step strai n data. These data may al so be presented i n terms of
vari ous compl i ance and modul us di stri buti on functi ons (or spectra) as
wel l as el ectri cal model s (Mohseni n, 1986; Pol akowski and Ri pl i ng,
1966; Sherman, 1970; Barnes et al ., 1989; Whorl ow, 1992; Ferry, 1980).
I n addi ti on, step strai n and creep curves can be normal i zed and pres-
ented i n l i near form (Pel eg, 1980). Thi s techni que can be qui te useful
i n bi ol ogi cal materi al s where i t i s often di ffi cul t to achi eve equi l i bri um
condi ti ons.
G

Maxwell
G
Kelvin
G
5.2.2 Step Strain (Stress Relaxation) 299
5.2.2. Step Strain (Stress Relaxation)
I n a step strai n test the sampl e i s gi ven an i nstantaneous strai n and
the stress requi red tomai ntai n the deformati on i s observed as a functi on
of ti me. Thi s experi ment i s commonl y known as a "stress rel axati on"
test and i t may be conducted i n shear, uni axi al tensi on, or uni axi al
compressi on. Stress rel axati on data may al so be obtai ned by subjecti ng
a fl ui d to a constant rate of strai n (i n, for exampl e, a concentri c cyl i nder
vi scometer), then suddenl y stoppi ng the deformati on and observi ng the
change i n stress over ti me.
Figure 5.2. Stress relaxation curves.
A wi de range of behavi or may be observed i n stress rel axati on tests
(Fi g. 5.2). No rel axati on woul d be observed i n i deal el asti c materi al s
whi l e i deal vi scous substances woul d rel ax i nstantaneousl y. Vi scoe-
l asti c materi al s woul d rel ax gradual l y wi th the end poi nt dependi ng on
the mol ecul ar structure of the materi al bei ng tested: stress i n
vi scoel asti c sol i ds woul d decay to an equi l i bri um stress ( ), but the
resi dual stress i n vi scoel asti c l i qui ds woul d be zero.
Stress rel axati on data are commonl y presented i n terms of a stress
rel axati on modul us:
o
0
S
t
r
a
i
n
S
t
r
e
s
s
t=0
0
Ideal Elastic Material
Viscoelastic Solid
Viscoelastic Liquid
Ideal Viscous Material
Time
e
e
> 0
[5.3]
I f a materi al i s perfectl y el asti c, the rel axati on modul us i s equal to the
shear modul us defi ned by Eq. [1.10]: . i s a si mi l ar, but
ti me-dependent, quanti ty determi ned from experi mental data. Curves
of the stress rel axati on modul us versus ti me, generated at di fferent
l evel s of strai n, overl ap i f data are col l ected i n the l i near vi scoel asti c
regi on of materi al behavi or. Rel ated functi ons (Ferry, 1980) can be
found i n tensi on ( ) and bul k compressi on ( ).
The Maxwel l model , one whi ch contai ns a Hookean spri ng i n seri es
wi th a Newtoni an dashpot, has frequentl y been used to i nterpret stress
rel axati on data for vi scoel asti c l i qui ds, parti cul arl y pol ymeri c l i qui ds.
The total shear strai n i n a Maxwel l fl ui d el ement (Fi g. 5.1) i s equal to
the sum of the strai n i n the spri ng and the dashpot:
[5.4]
Di fferenti ati ng Eq. [5.4] wi th respect to ti me, and usi ng the defi ni ti ons
provi ded by Eq. [5.1] and [5.2], yi el ds
[5.5]
or
[5.6]
where the rel axati on ti me (al so cal l ed the characteri sti c ti me of a
Maxwel l fl ui d) i s defi ned as
[5.7]
Al though an exact defi ni ti on of i s di ffi cul t, i t can be thought of as the
ti me i t takes a macromol ecul e to be stretched out when deformed.
Rel axati on ti mes for common fl ui ds vary a great deal as shown by the
i nformati on i n Tabl e 5.4. The above equati ons are presented i n terms
of shear deformati on. I f testi ng i s conducted i n uni axi al tensi on or
compressi on, then the rel axati on ti me can be thought of i n terms of an
extensi onal vi scosi ty ( ) and Youngs modul us ( ).
G f (t )

constant
G / G(t )
E(t ) K(t )
()
spring
+ ()
dashpot
d
dt

1
G
d
dt
_
,
+
+
rel
d
dt
_
rel

G
rel
E
E
Table 5.4. Relaxation Time and Limiting Viscosity at Zero Shear for various
Viscoelastic Fluids (Source: Tanner, 1985).
Fluid T
( C) (s) (Pa s)
Water 20 1E-12 0.001
Mineral oil 30 7E-10 0.5
Poly-dimethylsiloxane 30 1E-6 0.3
125 1.7E-4 100
Low-density polyethylene 115 10 2E5
240 0.1 3000
High-density polyethylene 180 0.07 2000
220 0.05 1000
High-impact polystyrene 170 7 2E5
210 3 1E5
0.5% Hydroxyethyl-cellulose 27 0.1 1.3
in water
Glass 27 > 1E5 > 1E18
The Maxwel l model i s useful i n understandi ng stress rel axati on
data. Consi der a step strai n (stress rel axati on) experi ment where there
i s a sudden appl i cati on of a constant shear strai n, . When the strai n
i s constant, the shear rate i s equal to zero ( ) and Eq. [5.6] becomes
[5.8]
Thi s equati on may be i ntegrated usi ng the i ni ti al condi ti on that
at :
[5.9]
or, after eval uati ng the i ntegral ,
[5.10]
Eq. [5.10] descri bes the gradual rel axati on of stress (from to zero)
after the appl i cati on of a sudden strai n. The rel ati onshi p provi des a
means of determi ni ng the rel axati on ti me: i s the ti me i t takes for the
stress to decay to 1/e (approxi matel y 36.8%) of i ts i ni ti al val ue.
rel

o
0.0
+
rel
d
dt
_
,
0

o
t 0
0
t
dt
rel
f (t )
o
exp
rel
_
rel
Experi mental data show that the Maxwel l model does not account
for the stress rel axati on behavi or of many vi scoel asti cmateri al s because
i t does not i ncl ude an equi l i bri um stress ( ). Thi s probl em may be
addressed for numerous foods by constructi ng a model whi ch has a si ngl e
Maxwel l el ement connected i n paral l el wi th a spri ng. The stress
rel axati on equati on descri bed by thi s mechani cal model (Fi g. 5.3a) i s
[5.11]
wi th the free spri ng (where ) accounti ng for the equi l i bri um
stress (Fi g. 5.4). The rel axati on ti me i s defi ned i n terms of the standard
Maxwel l porti on of the model : . Appl i cati on of Eq. [5.11] i s
i l l ustrated i n Exampl e Probl em 5.8.1.
Figure 5.3. Maxwell elements in parallel with a spring: a) one Maxwell element
and a free spring, b) three Maxwell elements and a free spring.
Added compl exi ty can be obtai ned by constructi ng a more general -
i zed Maxwel l model consi sti ng of several Maxwel l el ements i n paral l el
wi th an i ndependent spri ng. I f the system i s subjected to a constant
strai n, the total stress i s the sum of the i ndi vi dual stresses i n each
el ement. I n a four el ement model , a model contai ni ng three Maxwel l
el ements and a spri ng (Fi g. 5.3b), the sol uti on for stress as a functi on
of ti me i s
[5.12]
e
f (t )
e
+ (
o

e
) exp
rel
_
e

o
G
0
rel

1
/G
1
G
0
1
G
1
G
1 2 3
G G
G
0
1 2 3
a
b
f (t ) A
1
exp
t G
1
1
_
,
+ A
2
exp
t G
2
2
_
,
+ A
3
exp
t G
3
3
_
,
+
o
G
0
or
[5.13]
where the subscri pts refer to di fferent mechani cal el ements i n the
system. Each Maxwel l el ement may have a di fferent rel axati on ti me.
Thi s concept can be general i zed to determi ne a rel axati on spectra for a
vi scoel asti c materi al (Ferry, 1980). Eq. [5.10], [5.11], [5.12], and [5.13]
can be expressed i n terms of the rel axati on modul us, defi ned by Eq.
[5.3], by di vi di ng each equati on by the appl i ed strai n.
Figure 5.4. Typical stress relaxation curve modeled with Eq. [5.11] describing a
single Maxwell element in parallel with a spring.
Pel eg and Normand (1983) noted two major probl ems i n col l ecti ng
stress rel axati on data for foods: 1) When subjected to l arge deformati on
they usual l y exhi bi t non-l i near vi scoel asti c behavi or; 2) Natural
i nstabi l i ty or bi ol ogi cal acti vi ty make i t di ffi cul t to determi ne equi l i b-
ri um mechani cal parameters. To overcome these di ffi cul ti es, they
f (t ) A
1
exp
t
(
rel
)
1
_
,
+ A
2
exp
t
(
rel
)
2
_
,
+ A
3
exp
t
(
rel
)
3
_
,
+
e
Time
S
t
r
e
s
s
=
e
+ .368 (
o
-
e
) =
e
+ (
-
e
)
exp (
-t
_
)
e
o
o
rel
rel
suggest stress rel axati on data be cal cul ated as a normal i zed stress (a
normal i zed force term i s al so acceptabl e) and fi t to the fol l owi ng l i near
equati on:
[5.14]
where i s the i ni ti al stress, i s the decreasi ng stress at ti me , and
and are constants. The reci procal of depi cts the i ni ti al decay rate
and i s a hypotheti cal val ue of the asymptoti c normal i zed force. Fi tti ng
experi mental data to Eq. [5.14] i s a qui ck and effecti ve way to handl e
stress rel axati on data for many foods. Typi cal val ues of and are
summari zed i n Tabl e 5.5. Al so, the techni que i s i l l ustrated for appl e
ti ssue i n Exampl e Probl em 5.8.2.
Table 5.5. Stress Relaxation Parameters of Eq. [5.14] for Various Biological
Materials (Source: Peleg and Normand, 1983)
Material
(s or min) -
Cheddar Cheese 3.23 min 1.11
Corn Grains 10.9 min 5.18
Low Methoxyl Pectin 68.2 s 1.21
Gel (No.31)
Pea Beans 2.41 s 2.26
Potato Flesh 4.40 s 1.56
5.2.3. Creep and Recovery
I n a creep test, an i nstantaneous stress i s appl i ed to the sampl e and
the change i n strai n (cal l ed the creep) i s observed over ti me. When the
stress i s rel eased, some recovery may be observed as the materi al
attempts a return to the ori gi nal shape. I ncreased avai l abi l i ty of con-
trol l ed stress rheometers has greatl y i mproved our abi l i ty to conduct
shear creep and recovery experi ments on a wi de vari ety of materi al s.
These tests can be parti cul arl y useful i n studyi ng the behavi or exi sti ng
i n constant stress envi ronments such as those found i n l evel i ng, sedi -
mentati on, and coati ng appl i cati ons where gravi ty i s the dri vi ng force.
o
t
o

k
1
+ k
2
t
o
t k
1
k
2
k
1
k
2
k
1
k
2
k
1
k
2
5.2.3 Creep and Recovery 305
Creep experi ments can al so be conducted i n uni axi al tensi on or
compressi on. The anal yti cal methods presented i n thi s secti on have
been used to study typi cal fl ui d foods l i ke sal ad dressi ng (Paredes et al .,
1989), and very compl ex bodi es such as whol e oranges (Chuma et al .,
1978).
Figure 5.5. Creep and recovery curves.
I deal i zed creep and recovery curves are i l l ustrated i n Fi g. 5.5.
Subjected to a constant stress, strai n i n an i deal el asti c materi al woul d
be constant due to the l ack of fl ow, and the materi al woul d return to the
ori gi nal shape upon removal of stress. An i deal vi scous materi al woul d
show steady fl ow, produci ng a l i near response tostress wi th the i nabi l i ty
to recover any of the i mposed deformati on. Vi scoel asti c materi al s (e.g.,
bread dough) woul d exhi bi t a nonl i near response to strai n and, due to
thei r abi l i ty to recover some structure by stori ng energy, show a per-
manent deformati on l ess than the total deformati on appl i ed to the
sampl e. Thi s strai n recovery, or creep recovery, i s al so cal l ed recoi l and
may be i nvesti gated i n terms of a recoi l functi on (Deal y, 1994).
Creep data may be descri bed i n terms of a creep compl i ance functi on:
0
S
t
r
a
i
n
S
t
r
e
s
s
t=0
0
Time
o
Ideal Elastic Material
Viscoelastic Material
Ideal Viscous Material
Permanent
Deformation
(flow, partial recovery)
(no flow, complete recovery)
(steady flow, no recovery)
[5.15]
Compl i ance curves generated at di fferent stress l evel s overl ap when
data are col l ected i n the range of l i near vi scoel asti c behavi or. Wi th a
perfectl y el asti c sol i d, , the reci procal of the shear modul us;
however, di fferent ti me patterns i n experi mental testi ng mean that
. Eq. [5.15] i s presented i n terms of shear deformati on.
Si mi l ar materi al functi ons (Ferry, 1980) can be determi ned from creep
data generated i n tensi on ( ) and bul k compressi on ( ) studi es.
To devel op a mechani cal anal og descri bi ng creep behavi or, the
starti ng poi nt i s the Kel vi n model (Fi g. 5.1) whi ch contai ns a spri ng
connected i n paral l el wi th a dashpot. When thi s system i s subjected to
shear strai n, the spri ng and dashpot are strai ned equal l y:
[5.16]
The total shear stress ( ) caused by the deformati on i s the sum of the
i ndi vi dual shear stresses whi ch, usi ng Eq. [5.1] and Eq. [5.2], can be
wri tten as
[5.17]
Di fferenti ati ng Eq. [5.17] wi th respect to ti me yi el ds
[5.18]
where the retardati on ti me ( ) i s uni que for any substance. I f a
materi al was a Hookean sol i d, the retardati on ti me woul d be zero and
the maxi mum strai n woul d be obtai ned i mmedi atel y wi th the appl i ca-
ti on of stress: Ti me to achi eve maxi mum strai n i n vi scoel asti c materi al s
i s del ayed (or retarded). The retardati on ti me can be thought of i n terms
of extensi onal vi scosi ty ( ) and Youngs modul us ( ) i f testi ng i s
conducted i n uni axi al tensi on or compressi on.
I n creep, where the materi al i s al l owed to fl ow after bei ng subjected
to a constant shear stress ( ), the change i n stress wi th ti me i s zero
( ) and the sol uti on to Eq. [5.18] i s
[5.19]
J f (t )

constant
J 1/G
J(t ) 1/G(t )
D(t ) B(t )
()
spring
()
dashpot
G +
1
G
d
dt

+ (
ret
)
d
dt
ret
/G
E
E
o
d/dt 0
f (t )

o
G
1 exp
ret
_
,
_
,
showi ng that the i ni ti al strai n i s zero ( at ). Eq. [5.19] predi cts
a strai n that asymptoti cal l y approaches the maxi mum strai n ( )
associ ated wi th the spri ng. i s the ti me taken for the del ayed strai n
to reach approxi matel y 63.2%(1-1/e) of the fi nal val ue. Materi al s wi th
a l arge retardati on ti me reach ful l deformati on sl owl y.
The Kel vi n model (Fi g. 5.1) shows excel l ent el asti c retardati on but
i s not general enough to model creep i n many bi ol ogi cal materi al s. The
sol uti on to thi s probl em i s to use a Burgers model (Fi g. 5.6) whi ch i s a
Kel vi n and a Maxwel l model pl aced i n seri es. Data fol l owi ng thi s
mechani cal anal og show an i ni ti al el asti cresponse due tothe free spri ng,
retarded el asti c behavi or rel ated to the paral l el spri ng-dashpot
combi nati on, and Newtoni an type fl ow after l ong peri ods of ti me due to
the free dashpot (Fi g. 5.7):
[5.20]
where , the retardati on ti me of the Kel vi n porti on of the model .
Figure 5.6. Four element Burgers model.
The Burgers model can al so be expressed i n terms of creep com-
pl i ance by di vi di ng Eq. [5.20] by the constant stress:
[5.21]
Wri ti ng the resul t as a creep compl i ance functi on (Eq. [5.15]) yi el ds
0 t 0
o
/G
ret
f (t )

o
G
0
+
o
G
1
1 exp
ret
_
,
_
,
+
o
t
ret

1
/G
1
G
0
1
G
1
0
o
f (t )
1
G
0
+
1
G
1
1 exp
ret
_
,
_
,
+
t
0
[5.22]
where i s the i nstantaneous compl i ance, i s the retarded compl i ance,
i s the retardati on ti me ( ) of the Kel vi n component, and i s the
Newtoni an vi scosi ty of the free dashpot. The sum of and i s cal l ed
the steady state compl i ance. Usi ng the same procedure, Eq. [5.19] coul d
al sobe expressed i n term of the creep compl i ance functi on. The Burgers
model (Fi g. 5.6), l ess the free spri ng ( ), i s someti mes cal l ed the Jeffreys
model . Eq. [5.22] i s appl i ed to ski m mi l k curd i n Exampl e Probl em 5.8.3.
Parameters of Eq. [5.22], for creamy sal ad dressi ng, are gi ven i n
Appendi x 6.19.
Figure 5.7. Typical creep curve showing where various elements of the Burgers
model (Fig. 5.6 and Eq. [5.20]) describe flow behavior.
When conducti ng creep experi ments, control l ed stress rheometers
al l ow one to measure the strai n recovered when the constant stress i s
removed. The compl ete creep and recovery curve may be expressed
usi ng the Burgers model (Fi g. 5.6). When cal cul ated as compl i ance, the
creep i s gi ven by Eq. [5.22] for where i s the ti me when the
constant stress i s removed. At the begi nni ng of creep, there i s an
J f (t ) J
0
+ J
1
1 exp
ret
_
,
_
,
+
t
0
J
0
J
1
ret

1
/G
1

0
J
o
J
1
G
o
Instantaneous Response
Time
S
t
r
a
i
n
Retarded Elastic Behavior
Long Time Viscous Flow
G
0
1
G
1
0
0 < t < t
1
t
1
i nstantaneous change i n compl i ance ( ) due to the spri ng i n the
Maxwel l porti on of the model (Fi g. 5.8). Then, the Kel vi n component
produces an exponenti al change i n compl i ance rel ated tothe retardati on
ti me. After suffi ci ent ti me has passed, the i ndependent dashpot (Fi g.
5.6) generates a purel y vi scous response. Data from the l i near porti on
of the creep curve (Fi g. 5.8) are rel ated to two parameters: the sl ope i s
equal to ; and the i ntercept, someti mes cal l ed the steady state
compl i ance, i s equal to .
Figure 5.8. Compliance and recovery (or recoil) curves showing compliance
parameters for the Burgers model (Fig. 5.6 and Eq. [5.22]).
At , the stress i s removed ( ) and there i s an i nstantaneous
change i n compl i ance (Fi g. 5.8) equal to . The free dashpot causes
permanent deformati on i n the materi al rel ated to a compl i ance of .
Thi s factor i s di rectl y rel ated to the non-recoverabl e sampl e strai n of
. I f a substance obeyi ng the Burgers model i s tested i n the l i near
J
0
1/
0
J
0
+ J
1
Time
C
o
m
p
l
i
a
n
c
e
Creep Recovery
t
1
0
0
J
1
J
0
J
0
0
= =
0
t
1 0
/
0
Slope = 1/
J
1
t t
1
0
J
0
t
1
/
0
0
t
1
/
0
vi scoel asti c regi on of materi al behavi or, then the val ues of and
determi ned from the creep curve wi l l be equal to the val ues of and
determi ned from the recovery curve.
I f necessary, addi ti onal Kel vi n el ements can be added to the Burgers
model to better represent experi mental data. Mathemati cal l y, thi s i dea
can be descri bed wi th the fol l owi ng equati on:
[5.23]
where i s the total number of Kel vi n el ements i n the model , each havi ng
a uni que retarded compl i ance and retardati on ti me. A system wi th two
Kel vi n el ements ( ) was used by Hal i m and Shoemaker (1990) to
model ski m mi l k curd. Thi s model i s expl ored i n Exampl e Probl em 5.8.3.
The same equati on worked wel l for Purkayastha et al . (1985) i n
eval uati ng the compressi ve creep behavi or of potato fl esh and cheddar
cheese. Bal aban et al . (1988) presented systemati c procedures to
determi ne the constants i nvol ved i n Eq. [5.23]. Some advanced
rheometers provi de computer software to generate appropri ate con-
stants. Asi mpl e l i neari zed model (presented i n Exampl e Probl em 5.8.3),
si mi l ar to Eq. [5.14], has been suggested by Pel eg (1980) to characteri ze
the creep of bi ol ogi cal materi al s.
5.2.4. Start-Up Flow (Stress Overshoot)
Duri ng start-up fl ow, a shear rate i s suddenl y i mposed on a vi s-
coel asti c fl ui d hel d previ ousl y at rest. Shear stress produced by thi s
transi ent deformati on di spl ays an i ni ti al overshoot before reachi ng a
steady-state val ue; hence, the phenomenon i s commonl y referred to as
stress overshoot. Resul ts can be used to produce a shear stress growth
functi on. The general behavi or can be model ed usi ng an empi ri cal
equati on devel oped by Lei der and Bi rd (1974)whi ch i ncl udes rheol ogi cal
properti es rel ated to the fi rst normal stress di fference, and the shear
rate. Thi s equati on, usual l y cal l ed the Bi rd-Lei der equati on, i s
expressed as a functi on of ti me and the appl i ed shear rate:
[5.24]
where:
[5.25]
J
0
J
1
J
0
J
1
J f (t ) J
0
+
i 1
m
J
i
1 exp
t
(
ret
)
i
_
,
_
,
1
1
]
+
t
0
m
m 2
21
+
f (
, t ) K(
)
n
1 + (b
t 1) exp
t
an
_
,
_
K
2K
_
,
1/(n n)
5.2.4 Start-Up Flow (Stress Overshoot) 311
[5.26]
[5.27]
i s the ti me constant whi l e and are adjustabl e parameters. Over
short peri ods of ti me, the equati on model s the el asti c response (sudden
overshoot)of the materi al . Once the peak torque i s reached, exponenti al
decay i s si mul ated. After l ong peri ods of ti me, the Bi rd-Lei der equati on
col l apses to the standard power l aw equati on:
[5.28]
Typi cal curves of Eq. [5.24], model i ng the behavi or of mayonnai se, are
i l l ustrated i n Fi g. 5.9. The i nformati on i s pl otted i n terms of a
di mensi onl ess shear stress:
[5.29]
Figure 5.9. Typical stress overshoot curves at various shear rates for mayonnaise
at 25 C (based on data from Kokini and Dickie, 1981).
21
K(
)
n
11

22
K(
)
n
a b
21
+
f (
, t ) K(
)
n
21
+
21
+
1 + (b
t 1) exp
t
an
_
,
0 2 4 6 8 10
0
0.5
1
1.5
Time, s
100 1/s
10 1/s
1.0 1/s
0.1 1/s
2
1
+
2
1
+
8
Mayonnaise

The Bi rd-Lei der equati on was used by Koki ni and Di cki e (1981) and
Di cki e and Koki ni (1982) to eval uate stress overshoot data for vari ous
foods i ncl udi ng ketchup, mustard, mayonnai se, appl e butter, butter,
margari ne, and canned frosti ng. The authors concl uded that the rel a-
ti onshi p provi ded a moderatel y good predi cti on of peak shear stress and
peak ti mes but gave onl y a crude predi cti on of stress decay. Campanel l a
and Pel eg (1987b) were abl e to get a better fi t of stress growth data for
mayonnai se usi ng a more compl ex model proposed by Larson (1985).
The Bi rd-Lei der equati on, i n stri ctl y empi ri cal form ( not
requi red) has al so been used to i nterpret stress overshoot behavi or i n
cream (Prenti ce, 1992). Stress overshoot data were found useful i n
model i ng the human percepti on of fl ui d thi ckness i n the mouth (Di cki e
and Koki ni , 1983).
5.3. Oscillatory Testing
I n osci l l atory i nstruments, sampl es are subjected to harmoni cal l y
varyi ng stress or strai n. Thi s testi ng procedure i s the most common
dynami c method for studyi ng the vi scoel asti c behavi or of food. Resul ts
are very sensi ti ve to chemi cal composi ti on and physi cal structure so
they are useful i n a vari ety of appl i cati ons i ncl udi ng gel strength
eval uati on, moni tori ng starch gel ati ni zati on, studyi ng the gl ass
transi ti on phenomenon, observi ng protei n coagul ati on or denaturati on,
eval uati ng curd formati on i n dai ry products, cheese mel ti ng, texture
devel opment i n bakery and meat products, shel f-l i fe testi ng, and cor-
rel ati on of rheol ogi cal properti es to human sensory percepti on. Food
sci enti sts have found osci l l atory testi ng i nstruments to be parti cul arl y
val uabl e tool s for product devel opment work.
Osci l l atory testi ng may be conducted i n tensi on, bul k compressi on,
or shear. Typi cal commerci al i nstruments operate i n the shear defor-
mati on mode and thi s i s the predomi nant testi ng method used for food.
Shear strai n may be generated usi ng paral l el pl ate, cone and pl ate, or
concentri c cyl i nder fi xtures. Dynami c testi ng i nstruments may be
di vi ded i nto two general categori es: control l ed rate i nstruments where
the deformati on (strai n) i s fi xed and stress i s measured, and control l ed
stress i nstruments where the stress ampl i tude i s fi xed and the defor-
mati on i s measured. Both produce si mi l ar resul ts. The emphasi s i n
thi s secti on i s on fl ui d and semi -sol i d foods. Dynami c testi ng of sol i d
foods has been revi ewed by Rao and Ski nner (1986).
K, n, K, n
5.3 Oscillatory Testing 313
Application of Stress and Strain. A number of assumpti ons are
made i n devel opi ng the mathemati cal equati ons to descri be osci l l atory
testi ng: strai n i s the same at al l poi nts i n the sampl e, sampl e i nerti a
may be negl ected, and the materi al behaves as a l i near vi scoel asti c
substance. When these assumpti ons are vi ol ated, more compl ex ana-
l yti cal consi derati ons enter the probl em (Whorl ow, 1992).
I n osci l l atory tests, materi al s are subjected to deformati on (i n con-
trol l ed rate i nstruments) or stress (i n control l ed stress i nstruments)
whi ch vari es harmoni cal l y wi th ti me. Si nusoi di al , si mpl e shear i s
typi cal . To i l l ustrate the concept, consi der two rectangul ar pl ates ori -
ented paral l el to each other (Fi g. 5.10). The l ower pl ate i s fi xed and the
upper pl ate i s al l owed to move back and forth i n a hori zontal di recti on.
Assume the sampl e bei ng tested i s l ocated between the pl ates of a
control l ed rate devi ce. Suppose the strai n i n the materi al between the
pl ates i s a functi on of ti me defi ned as
[5.30]
where i s the ampl i tude of the strai n equal to when the moti on of
the upper pl ate i s . i s the frequency expressed i n rad/s whi ch
i s equi val ent to hertz. The peri od of ti me requi red to compl ete
one cycl e i s equal to . I f the two pl ates (Fi g. 5.10) were separated
by a di stance of 1.5 mm and the upper moved 0.3 mm from the center
l i ne, then the maxi mum strai n ampl i tude may be cal cul ated as 0.2 or
20%: . A 10% strai n coul d be achi eved by mai n-
tai ni ng mm and movi ng the pl ate 0.15 mm.
Figure 5.10. Oscillatory strain between rectangular plates.

o
sin(t )
o
L/h
L sin(t )
/(2)
2/
o
L/h 0.3/1.5 0.2
h 1.5
Stationary Plate
L
L sin ( t)
h
Oscillating Plate
SAMPLE
Effects of changi ng the ampl i tude and frequency on the i nput strai n
functi on are i l l ustrated i n Fi g. 5.11. Doubl i ng the ampl i tude from 0.1
(Curve 2) to 0.2 (Curve 1 or Curve 3) doubl es the hei ght of the curve.
Doubl i ng the frequency from 1 rad/s (Curve 1) to 2 rad/s (Curve 2 or
Curve 3) cuts, by one-hal f, the ti me between peaks of the two curves.
Usi ng a si ne wave for strai n i nput resul ts i n a peri odi c shear rate
found by taki ng the deri vati ve of Eq. [5.30]:
[5.31]
whi ch can be eval uated as
[5.32]
Wi th a smal l strai n ampl i tude (so the materi al wi l l behave i n a l i near
vi scoel asti cmanner), the fol l owi ng shear stress i s produced by the strai n
i nput:
[5.33]
where i s the ampl i tude of the shear stress (not to be confused wi th
the yi el d stress symbol i zed by i n earl i er chapters) and i s the phase
l ag or phase shi ft (al so cal l ed the mechani cal l oss angl e) rel ati ve to the
strai n. The ti me peri od associ ated wi th the phase l ag i s equal to .
can be thought of as the peak force per uni t area recei ved by the
stati onary pl ate (Fi g. 5.10). Di vi di ng both si des of Eq. [5.33] by yi el ds
[5.34]
The compl ete resul ts of smal l ampl i tude osci l l atory tests can be
descri bed by pl ots of the ampl i tude rati o ( ) and the phase shi ft ( )
as frequency dependent functi ons. These parameters al one, however,
are not commonl y used to descri be resul ts and other materi al functi ons
(whi ch may al l be wri tten i n terms of and ) have been defi ned.
The shear stress output produced by a si nusoi dal strai n i nput may
be wri tten as
d
dt

d(
o
sin(t ))
dt

o
cos(t )

o
sin(t + )
o

/
o
_
,
sin(t + )
o
/
o

o
/
o

Figure 5.11. Strain input functions showing variations in frequency and strain
amplitude: Curve 1) , ; Curve 2) , ; Curve 3)
, .
[5.35]
(cal l ed the shear storage modul us) and (cal l ed the shear l oss
modul us)are both functi ons of frequency and can be expressed i n terms
of the ampl i tude rati o and the phase shi ft:
[5.36]
and
[5.37]
may be i nterpreted as the component of the stress i n phase wi th
the strai n: may be i nterpreted as the component of the stress 90
out of phase wi th the strai n. Addi ti onal frequency dependent materi al
functi ons (Tabl e 5.2) i ncl ude the compl ex modul us ( ), compl ex vi s-
Time, s
S
t
r
a
i
n

I
n
p
u
t
0.3
0.2
0.1
0
-0.1
-0.2
-0.3
Curve 1
Curve 2
Curve 3
0 2.0 4.0 6.0 8.0 10.0 12.0
o
0.2 1 rad/s
o
0.1 2 rad/s
o
0.2 2 rad/s
G + (G/)
G G
G

o
_
,
cos()
G

o
_
,
sin()
G
o
G
o

G
*
cosi ty ( ), dynami c vi scosi ty ( ), out of phase component of the compl ex
vi scosi ty ( ), compl ex compl i ance ( ), storage compl i ance ( ), and the
l oss compl i ance ( ):
[5.38]
[5.39]
[5.40]
[5.41]
[5.42]
[5.43]
[5.44]
Al though , i t i s i mportant to note that and .
Osci l l atory data for vari ous food products may be found i n Appendi ces
[6.20], [6.21], and [6.22].
Usi ng Eq. [5.40], Eq. [5.35] can be expressed as
[5.45]
whi ch i s an excel l ent equati on to represent materi al behavi or because
i t cl earl y i ndi cates the el asti c ( ) and vi scous ( ) nature of a
substance. Thi s i dea i s expanded i n Sec. 5.5 to expl ai n the behavi or of
"si l l y putty"usi ng the Deborah number concept.
Another popul ar materi al functi on used to descri be vi scoel asti c
behavi or i s the tangent of the phase shi ft or phase angl e (cal l ed tan
del ta) whi ch i s al so a functi on of frequency:
[5.46]
Thi s parameter i s di rectl y rel ated to the energy l ost per cycl e di vi ded
by the energy stored per cycl e. Si nce , can vary from zero
to i nfi ni ty. Observati ons of pol ymer systems gi ve the fol l owi ng
numeri cal ranges for : very hi gh for di l ute sol uti ons, 0.2 to 0.3 for
J
*
J
J
G
*
o
(G)
2
+ (G)
2
G
*
()
2
+ ()
2

G
J
*
1
G
*
J
G
(G)
2
+ (G)
2
J
G
(G)
2
+ (G)
2
J
*
1/G
*
J 1/G J 1/G
G +
tan()
G
G
0 /2 tan
tan
amorphous pol ymers, l ow (near 0.01)for gl assy crystal l i ne pol ymers and
gel s. Val ues of for typi cal food systems (di l ute sol uti on, concen-
trated sol uti on, and gel ) are consi dered i n Sec. 5.4.
To better understand the vi scoel asti c parameters defi ned above, i t
i s hel pful to l ook at behavi or whi ch i s sol el y Hookean or sol el y Newto-
ni an. I f a materi al i s a Hookean sol i d, the stress and strai n are i n phase
and . Hence, and are al soequal to 0 because there i s no vi scous
di ssi pati on of energy. I n thi s case, i s a constant equal to the shear
modul us ( ). I f a materi al behaves as a Newtoni an fl ui d, the stress (Eq.
[5.33])and strai n (Eq. [5.30]) are 90 degrees out of phase ( ); hence,
the shear rate (Eq. [5.32]) i s al so 90 degrees out of phase wi th the shear
stress. I n thi s case, and are zero because the materi al does not
store energy. Then, i s constant and equal to the Newtoni an vi scosi ty
( ). Si mi l ar behavi or i s often observed for non-Newtoni an fl ui ds as the
frequency approaches zero.
Further assessment of the phase l ag concept can hel p cl ari fy the
meani ng of gi ven as part of the ampl i tude rati o i n Eq. [5.34]. As the
phase l ag approaches zero, force i s transmi tted through the sampl e (Fi g.
5.10) qui ckl y and the changes i n stress are observed at nearl y the same
ti me as the appl i ed deformati on produces strai n. I n sol i ds, rapi d force
transmi ssi on i s due to the crystal l i ne nature of the materi al . The
amount of force transmi tted for a gi ven strai n depends on the materi al
modul us. I n a Hookean sol i d the maxi mum force per uni t area trans-
mi tted through the sampl e ( ) i s equal to the shear modul us ti mes the
maxi mum strai n ( ). Vi scous heati ng absorbs some energy i n
vi scoel asti c materi al s resul ti ng i n smal l er val ues of .
The phase l ag approaches the maxi mum val ue of wi th fl ui ds
exhi bi ti ng a hi gh degree of Newtoni an behavi or. The maxi mum force
per uni t area ( ) transmi tted through a Newtoni an fl ui d depends on
the maxi mum shear rate i nduced duri ng deformati on. Consi deri ng Eq.
[5.32], the maxi mum shear rate i s frequency dependent and may be
cal cul ated as . Consequentl y, i s equal to the maxi mum shear rate
ti mes the vi scosi ty ( ) for a Newtoni an fl ui d. i s l arger i n vi s-
coel asti c materi al s that have l ess tendency to fl ow: I n these substances,
greater force i s transmi tted through the sampl e because the vi scous
di ssi pati on of energy i s smal l er.
tan
0 G
G
G
/2
G
o
G
o
o
/2
o

o
o

o
The Maxwel l fl ui d model (Fi g. 5.1) i s often used to i nterpret data
from dynami c testi ng of pol ymeri c l i qui ds. I f the strai n i nput i s har-
moni c, , then . Thi s rel ati onshi p (Eq. [5.32])
can be substi tuted i nto Eq. [5.6] and the resul ti ng di fferenti al equati on
sol ved toproduce a number of frequency dependent rheol ogi cal functi ons
for Maxwel l fl ui ds:
[5.47]
[5.48]
[5.49]
[5.50]
where , the rel axati on ti me of a Maxwel l fl ui d, i s equal to . Looki ng
at experi mental data may al l ow materi al constants of the Maxwel l
model to be eval uated from the asymptotes: as goes to zero, goes to
; and as goes to i nfi ni ty, goes to . These i deas are used i n Sec.
5.5 to exami ne the behavi or of "si l l y putty."
Comparison of Moduli and Compliances. The mathemati cal
rel ati onshi ps presented i n thi s secti on are for shear deformati on.
Anal ogous modul i and compl i ances can be defi ned for the tensi on (or
compressi on) and bul k compressi on modes of deformati on (Tabl e 5.6).
These functi ons are general l y more di ffi cul t to measure than shear
functi ons and few data are avai l abl e i n the publ i shed l i terature.
Anal ogues for the rel axati on modul us and creep compl i ance are al so
presented i n Tabl e 5.6.
Typical OperatingModesof anOscillatory TestingInstrument.
Commerci al l y avai l abl e osci l l atory i nstruments wi l l operate i n
numerous modes. A strai n or stress sweep, conducted by varyi ng the
ampl i tude of the i nput si gnal at a constant frequency (Fi g. 5.12), i s used
to determi ne the l i mi ts of l i near vi scoel asti c behavi or by i denti fyi ng a
cri ti cal val ue of the sweep parameter. I n the l i near regi on (Fi g. 5.13),
rheol ogi cal properti es are not strai n or stress dependent. Storage and

o
sin(t )

o
cos(t )
G
G
2
rel
2
1 +
2
rel
2
G
G
rel
1 +
2
rel
2

1 +
2
rel
2
tan
G
G

1
rel
rel
/G

G G
l oss modul i versus the sweep parameter are pl otted i n Fi g. 5.13. Some
experi menters prefer to pl ot combi ned materi al functi ons such as the
compl ex modul us or the compl ex vi scosi ty.
Table 5.6. Comparison of Moduli and Compliances Determined in Oscillatory
Testing Using Three Modes of Deformation: Shear, Bulk Compression, and Ten-
sion (Ferry, 1980).
Shear Bulk Compression Tension
Complex Modulus
Storage Modulus
Loss Modulus
Complex Compliance
Storage Compliance
Loss Compliance
Relaxation Modulus
Creep Compliance
Figure 5.12. Strain or stress sweep mode in oscillatory testing.
G
*
() K
*
() E
*
()
G() K() E()
G() K() E()
J
*
() B
*
() D
*
()
J() B() D()
J() B() D()
G( t ) K( t ) E( t )
J( t ) B( t ) D( t )
Time
A
m
p
l
i
t
u
d
e
Figure 5.13. Typical response to a strain or stress sweep showing the linear vis-
coelastic region defined by the critical value of the sweep parameter.
Strai n and stress sweeps are conducted on control l ed rate and
control l ed stress i nstruments, respecti vel y. I t has been suggested that
stress sweeps produce superi or resul ts (Hol l and, 1994); however, both
strai n and stress sweeps are known to provi de an excel l ent basi s for
compari ng the vi scoel asti c nature food products. I n addi ti on to esta-
bl i shi ng the l i near vi scoel asti c range of the sweep parameter, strai n and
stress sweeps have been used to di fferenti ate weak and strong gel s:
Strong gel s may remai n i n the l i near vi scoel asti c regi on over greater
strai ns than weak gel s.
The frequency sweep i s probabl y the most common mode of osci l -
l atory testi ng because i t shows how the vi scous and el asti c behavi or of
the materi al changes wi th the rate of appl i cati on of strai n or stress. I n
thi s test the frequency i s i ncreased whi l e the ampl i tude of the i nput
si gnal (stress or strai n) i s hel d constant (Fi g. 5.14). Frequency sweeps
are very useful i n compari ng, someti mes cal l ed "fi nger pri nti ng," di f-
ferent food products or i n compari ng the effects of vari ous i ngredi ents
and processi ng treatments on vi scoel asti ci ty. Materi al s usual l y exhi bi t
more sol i d l i ke characteri sti cs at hi gher frequenci es.
Strain or Stress
M
o
d
u
l
u
s
Storage Modulus
Critical Value
Linear Viscoelastic Region
Loss Modulus
Figure 5.14. Frequency sweep mode in oscillatory testing.
An i sothermal ti me sweep, where frequency and ampl i tude are
constant over ti me, can i ndi cate ti me-dependent structural changes
such as those associ ated wi th fi rmi ng of cheese curd or yogurt. A ti me
sweep (Fi g. 5.15) may be conducted i n conjuncti on wi th a control l ed
change i n temperature (Fi g. 5.16). Thi s type of testi ng i s very useful i n
studyi ng probl ems that i nvol ve temperature i nduced changes i n rheo-
l ogi cal behavi or. Typi cal exampl es associ ated wi th heati ng woul d
i ncl ude the softeni ng of chocol ate or cheese due to the mel ti ng of fat,
gel ati on to form pecti n gel s, thi ckeni ng of sol uti ons from starch gel a-
ti ni zati on, and fi rmi ng of meat or egg products caused by protei n
denaturati on. The study of ti me-dependent fl ow behavi or resul ti ngfrom
chemi cal reacti ons such as these i s cal l ed chemorheol ogy.
Strain in Rotational-TypeFixtures. Paral l el pl ate, cone and pl ate,
and concentri c cyl i nder fi xtures are the preferred geometri es for sub-
jecti ng fl ui d and semi -sol i d foods to an osci l l atory strai n. I n a paral l el
pl ate apparatus, such as that used i n a tradi ti onal paral l el pl ate
vi scometer (Fi g. 5.17), the shear strai n i s a functi on of the radi us. I t
vari es from zero at the center of the sampl e ( ) to a maxi mum at the
outer edge of the pl ate ( ). Maxi mum strai n ( ) i s cal cul ated as the
di stance travel ed at the ri m of the upper pl ate ( )di vi ded by the di stance
between the pl ates ( ):
Time
A
m
p
l
i
t
u
d
e
r 0
r R
o
r
h
Figure 5.15. Time sweep mode in oscillatory testing.
Figure 5.16. Controlled temperature changes in oscillatory testing.
Time
A
m
p
l
i
t
u
d
e
T
e
m
p
e
r
a
t
u
r
e
Time
Ramp Change
Step Change
[5.51]
Assume experi mentati on starts wi th the system at rest, then the sampl e
i s subjected to a si ne wave strai n functi on descri bed by Eq. [5.30]. The
pl ate wi l l move through a posi ti ve sweep angl e ( ), then pass through
the starti ng posi ti on whi l e rotati ng to a negati ve val ue of the sweep
angl e ( ). (i n radi ans) i s the amount of angul ar travel of the upper
pl ate from the starti ng posi ti on to the poi nt of maxi mum rotati on. At
thi s poi nt the maxi mum strai n i s i nduced i n the sampl e. To achi eve,
for exampl e, a 10%maxi mum strai n i n a paral l el pl ate apparatus wi th
50 mm di ameter pl ates separated by a di stance of 2.0 mm wi l l requi re
the fol l owi ng sweep angl e:
[5.52]
A sweep angl e resul ti ng i n a strai n of 10%or l ess i s usual l y requi red to
stay i n the regi on of l i near vi scoel asti c materi al behavi or.
Figure 5.17. Sweep angle in a parallel plate apparatus.
o

R

r
h
2R
2
_
1
h
_
,

R
h
+

(
o
)h
R

.1(2)
50/2
0.008 rad or 0.458 degrees
R
h
r
r
+
-
Shear strai n i n a cone and pl ate fi xture (Fi g. 3.3) i s uni form
throughout the gap and equal to the sweep angl e di vi ded by the tangent
of the cone angl e:
[5.53]
Recal l that for the preferred cone angl es, rad (4 degrees).
Shear strai n i n a concentri c cyl i nder fi xture (Fi g. 3.1) i s a functi on of
the radi us. I n the case where the bob rotates and the cup i s fi xed, the
maxi mum strai n occurs at the bob:
[5.54]
where, recal l , . When the bob i s fi xed and the cup rotates, the
maxi mum strai n occurs at the cup:
[5.55]
Shear stresses for paral l el pl ate, cone and pl ate, and concentri c cyl i nder
fi xtures are cal cul ated usi ng the appropri ate expressi ons devel oped i n
Chapter 3: Eq. [3.71] wi th , Eq. [3.55], and Eq. [3.3], respecti vel y.
5.4. Typical Oscillatory Data
SingleFrequencyTests. Typi cal i nput and output for a di l ute sol uti on
showi ng Newtoni an type behavi or i s i l l ustrated i n Fi g. 5.18 wi th
accompanyi ng data i n Tabl e 5.7. I n thi s case, an i nput strai n ampl i tude
of 10% ( ) wi th a constant frequency of 10 rad/s gi ves an i nput
strai n functi on of . The output stress functi on i s shi fted by
1.48 rad (85 ) and shows an ampl i tude rati o of 0.1 ( ):
[5.56]
Typi cal output data for a gel showi ng very el asti c (sol i d l i ke) behavi or
and a concentrated sol uti on showi ng vi scoel asti c behavi or are seen i n
Fi g. 5.19 wi th compani on data i n Tabl e 5.7. The i nput strai n functi on
i s the same as that shown i n Fi g. 5.18: .
o

tan
tan 0.07
o

bob
2
2
1
_
,
R
c
/R
b
o

cup
2
2
1
_
,
n 1
o
0.1
0.1sin(10t )
G
*

o
/
o
0.1
G
*
o
sin(t + ) .01sin(10t + 1.48)
0.1sin(10t )
5.4 Typical Oscillatory Data 325
Figure 5.18. Typical input and output for a dilute solution showing Newtonian
like behavior at a constant frequency of 10 rad/s.
Table 5.7. Typical Output for aDilute Solution(NewtonianBehavior), Gel (Elastic
Behavior) and a Concentrated Solution(Viscoelastic Behavior) froma StrainInput
Function having a Frequency of 10 rad/s and a Strain Amplitude of 10 Percent:
Dilute Solution Gel Concentrated
Solution
, rad (degrees) 1.48 (85 ) 0.0698 (4 ) 0.873 (50 )
, Pa 0.100 5200 200
, Pa 0.00907 5187 129
, Pa 0.0996 363 153
, Pa s 0.00996 36.3 0.140
, Pa s 0.000907 518.7 12.9
, Pa s 0.010 520.0 20.0
10.9 0.0699 1.19
0 0.2 0.4 0.6 0.8 1 1.2
Time, s
S
t
r
a
i
n

I
n
p
u
t

(
-
)

o
r

S
t
r
e
s
s

O
u
t
p
u
t

(
P
a
)
0.15
0.05
0
-0.05
-0.10
-0.15
0.10
G
*
= .1 Pa ; phase shift = 85 degrees
Strain Input: 10 % strain ; 10 rad/s
Output:
0.1sin(10t )

o
/
o
G
*
G
G
*
cos
G
G
*
sin
*
G
*
/
tan G
/G

Figure 5.19. Typical output (Input: ) for a gel showing elastic
type behavior and a concentrated solution showing viscoelastic behavior.
Output stress for the gel (Fi g. 5.19)has a smal l phase shi ft,
rad (4 ), but a l arge ampl i tude rati o ( Pa) produci ng the
fol l owi ng shear stress functi on:
[5.57]
The output stress for the concentrated sol uti on has a greater phase shi ft,
rad (50 ), and an ampl i tude rati o between the di l ute sol uti on
and the gel ( Pa):
[5.58]
Normalized Strain and Stress. I nput and output at a parti cul ar
frequency may al so be vi sual i zed i n terms of a normal i zed strai n,
[5.59]
and a normal i zed stress,
[5.60]
0 0.2 0.4 0.6 0.8 1 1.2
Time, s
S
t
r
e
s
s

O
u
t
p
u
t
,

P
a
G
*
= 5200 Pa : phase shift = 4 degrees
G
*
= 200 Pa : phase shift = 50 degrees
600
400
200
0
-200
-400
-600
Gel
Concentrated Solution
0.1sin(10t )
0.0698
G
*

o
/
o
5213
G
*
o
sin(t + ) 521.3sin(10t + 0.0698)
0.873
G
*
200
G
*
o
sin(t + ) 20sin(10t + 0.873)
o
Val ues of and range from -1.0 to 1.0.
The normal i zed strai n i nput functi on i s the same for the three
materi al s currentl y under consi derati on:
[5.61]
Normal i zed stress output functi ons are
[5.62]
for the di l ute sol uti on,
[5.63]
for the concentrated sol uti on, and
[5.64]
for the gel . Eq. [5.61] through [5.64] are pl otted i n Fi g. 5.20. Phase
shi fts for the di l ute sol uti on (1.48 rad) and the concentrated sol uti on
(0.873 rad)are i ndi cated on the fi gure. The phase shi ft for the gel (0.0698
rad) i s smal l but sti l l vi si bl e. Advanced rheometers may provi de a
conti nuousl y updated di spl ay of these (or equi val ent) waveforms duri ng
osci l l atory testi ng. Thi s i nformati on can be very useful i n determi ni ng
whether or not the materi al bei ng tested i s di spl ayi ng vi scoel asti c
behavi or. I t can al so be used to detect di storted waveforms that may
produce mi sl eadi ng resul ts.
Further consi derati on of Fi g. 5.20 can provi de added physi cal
meani ng to the phase l ag phenomenon. Frequency of the strai n i nput
i s 10 rad/s sothe frequency of the output si gnal s for each materi al (di l ute
sol uti on, concentrated sol uti on, and gel ) i s al so 10 rad/s. The peri od of
ti me requi red to compl ete a si ne wave strai n cycl e i s . I n thi s
experi ment, the peri od i s equal to or 0.628 s. The maxi mum phase
l ag ( ) whi ch can occur i s meani ng the stress si gnal , correspondi ng
to a parti cul ar strai n i nput, coul d be observed at a maxi mum of
s after the strai n i s appl i ed. Stress i s observed
1.48/10 = 0.148 s after the strai n i n the di l ute sol uti on because i s 1.48
rad. Thi s i s cl ose to maxi mum phase l ag expected for a Newtoni an fl ui d
at the 10 rad/s test frequency. A val ue of 0.0698 rad (the mi ni mum
possi bl e val ue of i s zero)for the gel means the stress si gnal i s 0.0698/10
= 0.00698 s behi nd the strai n i nput. Thi s Hookean-l i ke behavi or i s not
o
0.1sin(10 t )/0.1
o
0.01sin(10 t + 1.148)/0.01
o
20sin(10 t + 0.873)/20
o
521.3sin(10 t + 0.0698)/521.3
2/
2/10
/2
/(2) /(2) (10) 0.157

Figure 5.20. Normalized curves for input strain ( ) and output stresses
( ) showing the phase shift for a dilute solution (1.48 rad or 85 ) and a con-
centrated solution (0.873 rad or 50 ).
unusual due to the hi gh l evel of mol ecul ar structure present i n a gel .
The l ag peri od of 0.0873 s found for the concentrated sol uti on represents
a si gnal transmi ssi on ti me between the di l ute sol uti on and the gel .
Cl earl y, the concentrated sol uti on has more structure than the di l ute
sol uti on but si gni fi cantl y l ess than the gel .
Frequency Sweeps. Fi g. 5.18, 5.19, 5.20 i l l ustrate rheol ogi cal
behavi or at a si ngl e frequency of 10 rad/s. To effecti vel y characteri ze
the rheol ogi cal behavi or of these substances, materi al functi ons shoul d
be determi ned over a wi de range of frequenci es usi ng the frequency
sweep testi ng concept i l l ustrated i n Fi g. 5.14. Data may be pl otted wi th
the frequency gi ven as radi ans per second or hertz by recogni zi ng that
1 hz =1 cycl e/s =2 rad/s. The choi ce i s essenti al l y a matter of i ndi vi dual
preference.
0 0.2 0.4 0.6 0.8 1 1.2
Time, s
N
o
r
m
a
l
i
z
e
d

S
t
r
a
i
n

o
r

S
t
r
e
s
s

(
-
)
1.5
1.0
0.5
0
-0.5
-1.0
-1.5
input
strain
dilute
solution
concentrated
solution
gel
1.48 rad 0.873 rad
/
o
/
o


A di l ute sol uti on, a concentrated sol uti on, and a gel show di sti ncti ve
behavi or (Fi g. 5.21, 5.22 and 5.23)when subjected to a frequency sweep.
Wi th a di l ute sol uti on (Fi g. 5.21), i s l arger than over the enti re
frequency range but approach each other at hi gher frequenci es. and
curves i ntersect at the mi ddl e of the frequency range for the con-
centrated sol uti on (Fi g. 5.22) showi ng a cl ear tendency for more sol i d-
l i ke behavi or at hi gher frequenci es. The crossover frequency i s
someti mes a useful cri teri on for product eval uati on. I t occurs when
, the poi nt where the phase l ag ( ) equal s . i s si gni fi cantl y
hi gher than throughout the frequency range for the gel (Fi g. 5.23).
I t i s meani ngful toobserve that modul i are a strong functi on of frequency
i n the di l ute and concentrated sol uti ons, but practi cal l y constant wi th
the gel . Al ternate ways of pl otti ng osci l l atory data are di scussed i n
Figure 5.21. Mechanical spectra for a dilute solution made from 5% dextrin
(Data from Ross-Murphy, 1988).
G G
G
G
G G /4 G
G
0.1 0.2 0.5 1 2 5 10 20 50 100
0.001
0.002
0.005
0.01
0.02
0.05
0.1
0.2
0.5
1
G

(
P
a
)

o
r

G

(
P
a
)
G
Dilute Solution
G
(rad/s)
Figure 5.22. Mechanical spectra for a concentrated solution made from 5%
lambda carrageenan (Data from Ross-Murphy, 1988).
Figure 5.23. Mechanical spectra for a gel made from 1% agar (Data from Ross-
Murphy, 1988).
0.01 0.03 0.1 0.3 1 3 10 30 100
0.1
0.3
1
3
10
30
100
300
1,000
G

(
P
a
)

o
r

G

(
P
a
)
(rad/s)
G
G
0.01 0.03 0.1 0.3 1 3 10 30 100
10
30
100
300
1,000
3,000
10,000
30,000
100,000
G

(
P
a
)

o
r

G

(
P
a
)
(rad/s)
Gel
G
G
Analysis of the Phase Lag ( ). The tendency of the di l ute sol uti on
and the concentrated sol uti on toexhi bi t more fl ui d or sol i d-l i ke behavi or
wi th i ncreasi ng frequency can be exami ned i n more detai l by consi deri ng
the frequency dependence of the phase l ag. Fi rst, the data shown i n
Fi g. 5.21, 5.22, and 5.23 are summari zed i n the form of power l aw
equati ons (Tabl e 5.8). Usi ng thi s i nformati on the phase l ag may be
cal cul ated from the l oss and storage modul i : . Resul ts
are i l l ustrated i n Fi g. 5.24. The maxi mum phase l ag whi ch can be
observed i s found wi th a Newtoni an fl ui d: The mi ni mum val ue i s
zero found wi th a Hookean sol i d. Hi gh val ues of at l ow frequenci es
i ndi cate a tendency toward more fl ui d-l i ke behavi or for both the di l ute
and concentrated sol uti ons at l ow deformati on rates. More sol i d-l i ke
behavi or i s observed for these sol uti ons at the hi gh deformati on rates
associ ated wi th hi gh frequenci es. The phase l ag for the gel (Fi g. 5.24)
i s practi cal l y constant i ndi cati ng consi stent sol i d-l i ke behavi or over the
enti re frequency range.
Table 5.8. Values of the storage modulus ( ) and the loss
modulus ( ) for typical materials: dilute solution (Fig. 5.21),
concentrated solution (Fig. 5.22), and gel (Fig. 5.23).
Material
(Pa s
b
) ( - ) (Pa s
d
) ( - ) (rad/s)
Dilute Solution .00028 1.66 .01186 .934 .1 - 100
Concentrated Solution 16.26 .840 27.78 .520 .01 - 100
Gel 5626 .0371 344.7 .0145 .01 - 100
Timeto Complete an Oscillatory Test. Osci l l atory data for a par-
ti cul ar frequency must be col l ected over one compl ete si ne wave cycl e,
and two or three cycl es may be needed to obtai n equi l i bri um val ues. I t
i s i mportant to consi der the ti me requi red to compl ete l ow frequency
tests when sel ecti ng a frequency range for the experi mentati on. I f, for
exampl e, a test was conducted at a l ow frequency of = 0.01 rad s
-1
,
then the ti me requi red to compl ete one cycl e can be cal cul ated
( ) as 10.5 mi nutes. Therefore, i t woul d take at l east 31.5
mi nutes to obtai n a si ngl e pai r of and val ues i f three cycl es were
needed! I t i s usual l y unnecessary to col l ect dynami c data at frequenci es
arctan(G/G)
/2
G a
b
G c
d
a b c d
2/(.01) (60)
G G
Figure 5.24. Variation of the phase lag ( ) with frequency ( ) for typical mate-
rials. The upper limit is represented by a Newtonian fluid ( ) and the
lower limit by a Hooke solid ( ).
l ess than 0.01 rad s
-1
and a l ower l i mi t of =0.1 rad s
-1
i s often adequate.
I n contrast, the ti me requi red to compl ete one cycl e at a frequency of
100 rad s
-1
, a sui tabl e upper l i mi t for many experi ments, i s 0.063 s
meani ng that a three cycl e test coul d be compl eted i n l ess than 0.19 s.
5.5. Deborah Number
Marcus Rei ner (Rei ner, 1964) proposed the Deborah number as a
means of di sti ngui shi ng between sol i ds and l i qui ds. The number, named
after the prophetess Deborah, i s based on hi s i nterpretati on of a bi bl i cal
passage found i n Judges 5:5. He stated that Deborahs song, gi ven after
the vi ctory over the Phi l i sti nes, coul d be transl ated as "The mountai ns
fl owed before the Lord." Hi s poi nt was that mountai ns coul d fl ow before
God, not before man, because the ti me of observati on for God i s i nfi ni te.
(A si mi l ar argument coul d be made from Psal m 97:5: "The mountai ns
mel t l i ke wax before the Lord.") Thus, Prof. Rei ner recogni zed the need
to consi der the ti me of the process as a factor when characteri zi ng a
0.1 0.2 0.5 1 2 5 10 20 50 100
0
0.5
1
1.5
Newtonian Fluid
Dilute Solution
Hooke Solid
Gel
(
r
a
d
)
(rad/s)

/2
0
5.5 Deborah Number 333

substance as a sol i d or a l i qui d. Thi s i dea has not been extensi vel y
appl i ed to food products but sti l l provi des a very useful means for
understandi ng the behavi or of vi scoel asti c materi al s.
The Deborah number i s defi ned as
[5.65]
where i s the characteri sti c ti me of the materi al and i s the
characteri sti c ti me of the process. for a Maxwel l fl ui d and
for a Kel vi n sol i d (Fi g. 5.1). Gi vi ng a preci se defi ni ti on of
i s more di ffi cul t. I t i s rel ated to the ti me scal e of the deformati on and
i s roughl y equal to the l ength scal e i n the fl ow di recti on di vi ded by the
mean vel oci ty. I t mi ght be the reci procal of frequency for an osci l l ati ng
surface, or the ti me for a parti cl e to pass through a convergi ng di e and
experi ence the subsequent i ncrease i n vel oci ty. Any process must be
i nvesti gated i ndi vi dual l y but Deborah numbers tend to be hi gh for
processes such as fi ber spi nni ng and pl asti c moul di ng but l ow for
extrusi on.
Al l the vi scometri c fl ow si tuati ons descri bed i n Chapters 2, 3 and 4
i nvol ve l ow val ues of the Deborah Number. Creep and stress rel axati on
di scussed i n thi s chapter al so i nvol ve l ow Deborah Number val ues. The
other mai n topi c of thi s chapter, osci l l atory testi ng, i nvol ves both l ow
and moderate val ues of the Deborah Number. Changes i n the phase
l ag wi th i ncreasi ng frequency for the di l ute and concentrated sol uti ons
i l l ustrated i n Fi g. 5.24 demonstrate the concept. Al though the char-
acteri sti c ti me of the materi al i s unaffected by osci l l atory testi ng, the
characteri sti c ti me of the process i s i nversel y proporti onal to the
frequency; therefore, as the frequency i ncreases, the Deborah Number
i ncreases causi ng the sol uti ons to exhi bi t the more sol i d-l i ke charac-
teri sti cs associ ated wi th l ower val ues of the phase l ag (Fi g. 5.24).
I t i s i mportant to remember that al l materi al s are vi scoel asti c, but
the vi scous or the el asti c character may domi nate i n certai n si tuati ons.
Pi pki n (1986), suggests that the characteri sti c ti me of a materi al may
be provi si onal l y consi dered as an order-of-magni tude esti mate for how
l ong i t takes the substance to compl ete a stress rel axati on process. I f
a materi al i s an i deal el asti c sol i d, , and no rel axati on occurs.
Wi th an i deal vi scous materi al , , meani ng rel axati on i s i mme-
N
De

t
material
t
process
t
material
t
process
rel
t
material
ret
t
material
t
process
t
material

t
material
0
di ate. The characteri sti c ti me of water i s on the order of 10
-12
s and the
characteri sti c ti me for wi ndow gl ass (a super cool ed l i qui d at room
temperature) i s greater than 10
5
s. Hence, when water i s deformed,
el asti c effects are di ffi cul t to observe because the materi al rel axes so
qui ckl y. Conversel y, the vi scous behavi or of gl ass i s not easi l y observed
because the response to the deformi ng force (gravi ty) on a verti cal pane
i s very sl ow; consequentl y, decades are needed for observabl e changes
to occur. The characteri sti c ti me of the materi al and the process must
be exami ned together to determi ne i f vi scous or el asti c type behavi or
wi l l domi nate a parti cul ar fl ow probl em. Materi al s exhi bi t pronounced
vi scoel asti c behavi or when these ti mes are si mi l ar i n magni tude.
The previ ous di scussi on provi des i nsi ght concerni ng how the
numeri cal val ue of can be used as a measure of the degree of vi s-
coel asti ci ty. I f , the stress i s proporti onal to the vi scosi ty ti mes
the shear rate and the materi al behaves as a vi scous l i qui d. On the
other hand, i f , then stress i s proporti onal to the modul us of
ri gi di ty ti mes the strai n and the materi al behaves l i ke an el asti c sol i d.
I f the i s on the order of one, materi al s wi l l show vi scoel asti cbehavi or.
When i s very hi gh, the model for a Hookean sol i d may best descri be
materi al behavi or. The Newtoni an model may be the best choi ce when
very smal l are present.
Tofurther expand the concept of the Deborah number, consi der data
from a common si l i cone pol ymer known as "bounci ng putty" or "si l l y
putty". Thi s substance, sol d as a toy for chi l dren, i s i nteresti ng because
i t has a rel axati on ti me wel l wi thi n the l i mi ts of human percepti on. I f
the materi al i s assumed to behave as a Maxwel l fl ui d, constants may
be determi ned from the asymptotes found i n Fi g. 5.25. Taki ng the l i mi t
(as goes to i nfi ni ty) of Eq. [5.47] shows that Pa and
taki ng the l i mi t (as goes to zero) of Eq. [5.49] i ndi cates that
Pa s. Usi ng these val ues, the rel axati on ti me can be
cal cul ated: s.
N
De
N
De
1
N
De
1
N
De
N
De
N
De
G G 260, 000
80, 000
rel
/G 0.31
5.5 Deborah Number 335
Figure 5.25. Oscillatory shear measurements on a silicone polymer (Data from
Denn, 1980).
Conti nued anal ysi s of si l l y putty shows that Pa and
Pa s at rad/s. Expressi ng thi s i n terms of the shear stress
gi ven by Eq. [5.45] yi el ds
[5.66]
Cl earl y, the el asti c porti on of the equati on ( ) domi nates the
mathemati cal rel ati onshi p. Assumi ng , the Deborah number for
thi s case can be i s approxi mated as
[5.67]
whi ch i s a numeri cal val ue of suffi ci ent magni tude to i ndi cate that
si gni fi cant el asti c effects woul d be present. The characteri sti c ti me of
the process woul d be much l ower, causi ng the Deborah number to be
much hi gher, i n si tuati ons where si l l y putty bounces when dropped and
shows bri ttl e fracture when pul l ed qui ckl y.
0.1 0.3 1 3 10 30 100 300 1,000
100
300
1,000
3,000
10,000
30,000
100,000
300,000
1,000,000
dynamic viscosity storage modulus
(
P
a

s
)

o
r

G

(
P
a
)
(rad/s)
Silicone Polymer
G 260, 000
100 100
G +
260, 000 + 100

260, 000
o
0.1
N
De

rel
t
process

rel
(
o
)
1
0.31(.1) (100) 3.1
Vi ewi ng the probl em at a l ow frequency ( rad/s), Pa
and Pa s, gi ves a shear stress equati on (Eq. [5.45]) of
[5.68]
Testi ng under these condi ti ons resul t i n a l ow val ue of the Deborah
number:
[5.69]
At l ow frequency (l ong characteri sti c ti me of the process), the Deborah
number i s very smal l and the vi scous component ( ) domi nates
materi al behavi or expl ai ni ng why the substance fl ows, as a l i qui d, when
pul l ed sl owl y. Gi ven the above di scussi on, one can concl ude that
adverti si ng si l l y putty as the "real sol i d-l i qui d"i s techni cal l y acceptabl e.
5.6. Experimental Difficulties in Oscillatory Testing of Food
There are vari ous sources of error whi ch may be parti cul arl y prob-
l emati c when testi ng food products usi ng osci l l atory methods. Many
foods may not exhi bi t a wel l defi ned strai n or stress range where the
pri nci pl e of l i near vi scoel asti ci ty appl i es. Thi s may be due to non-
homogeneous sampl es, the presence of wal l sl i p, or ti me-dependent
materi al behavi or. Presheari ng wi th steady rotati on, before conducti ng
osci l l atory tests, can be useful i n control l i ng (or el i mi nati ng) thi xotropy
i n some materi al s. Serrated or roughened surfaces may be effecti ve i n
deal i ng wi th wal l sl i p. Speci al surfaces have been used, for exampl e,
i n testi ng cracker and cooki e doughs (Menji var, 1994), and cheddar
cheese (Rosenberg et al ., 1995).
The presence of a yi el d stress may al so cause seri ous di ffi cul ti es i n
osci l l atory testi ng of food. Thi s probl em was observed many years ago
by El l i ot and Ganz (1971, 1977) i n studi es i nvol vi ng mayonnai se, sal ad
dressi ng, and margari ne. I f the appl i ed strai n causes the resul ti ng
stress to exceed the yi el d stress, then the output si n wave wi l l be
truncated due to parti al fl ow of the materi al . Thi s devi ant behavi or i s
i l l ustrated i n Fi g. 5.26 where the peak val ues of the normal i zed stress
are equal to 0.7: . Si nce the yi el d stress i s equal to 70% of
the maxi mum shear stress ( ), cannot exceed 0.7.
0.1 G 200
80, 000
G +
200 + 80, 000

N
De

rel
t
process

rel
(
o
)
1
0.31(.1) (.1) 0.003
80, 000

*
/
o
0.7
o
5.6 Experimental Difficulties in Oscillatory Testing of Food 337
The transi ti on from sol i d-l i ke to fl ui d-l i ke behavi or may be thought
of i n terms of a cri ti cal strai n or stress, bel ow whi ch the materi al behaves
as an el asti c sol i d. Appl i ed stresses whi ch exceed the yi el d stress cause
a devi ati on from sol i d-l i ke behavi or resul ti ng i n a non-si nusoi dal output
wave; consequentl y, measured val ues of the storage and l oss modul i do
not fal l wi thi n the scope of l i near vi scoel asti ci ty. These modul i , al though
qui te useful i n compari ng di fferent food products, wi l l be a functi on of
both the strai n (or stress) ampl i tude and frequency (Yoshi mura and
Purdhomme, 1987). The probl em can be further compl i cated by the fact
that the yi el d stress i tsel f may be a ti me-dependent parameter for many
food products (Steffe, 1992).
Figure 5.26. Stress output when the applied strain takes the material outside the
regime where the material exhibits solid-like behavior.
The above expl anati on of the yi el d stress i ssue i n osci l l atory testi ng
i l l ustrates the di l emma, but may oversi mpl i fy the probl em. An al ter-
nati ve approach i s to assume that the yi el d stress does not exi st and
everythi ng fl ows. Thi s means that even materi al s appeari ng to have a
very sol i d-l i ke nature woul d, gi ven suffi ci ent ti me, fl ow. Barnes et al .
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Time
N
o
r
m
a
l
i
z
e
d

S
t
r
a
i
n

o
r

S
t
r
e
s
s
1.5
1.0
0.5
0
-0.5
-1.0
-1.5
Strain
Input
Stress
Output
yield stress exceeded
(1987) menti on growi ng evi dence showi ng that concentrated systems
(e.g., gel s, margari ne, i ce cream, and sti ff pastes) fl ow at very l ow
stresses. They expl ai n that the hypothesi s of a yi el d stress, and the
subsequent observati on that a materi al does not fl ow, coul d be i nter-
preted to mean that the materi al has a very hi gh zero shear vi scosi ty.
Normal wi ndow gl ass, for exampl e, has a zero shear vi scosi ty greater
than 10
18
Pa s. Thi s materi al appears to be sol i d but gi ven suffi ci ent
ti me (decades) shows observabl e changes due to fl ow. Latter work
(Barnes, 1992) has acknowl edged sci enti fi c evi dence suggesti ng the
presence of a yi el d stress i n non-i nteracti ng and fl occul ated suspensi ons.
Regardl ess of the expl anati on for devi ant behavi or, osci l l atory i nfor-
mati on i s of l i mi ted useful ness i f data are not col l ected wi thi n the l i near
vi scoel asti c range of the materi al behavi or.
5.7. Viscometric and Linear Viscoelastic Functions
Normal stress functi ons ari se from the vi scoel asti c nature of
materi al s; therefore, i t i s reasonabl e to expect l i near vi scoel asti c
materi al functi ons determi ned from osci l l atory testi ng to be rel ated to
steady shear behavi or. Exact rel ati onshi ps can be determi ned i n the
l ower l i mi ts of shear rate and frequency (Wal ters, 1975):
[5.70]
and
[5.71]
where, recal l (Eq. [1.23]), . Al though thei r use-
ful l ness i s l i mi ted because they onl y appl y at l ow frequenci es and shear
rates, the above equati ons have proven tobe val i d for numerous pol ymer
mel ts and sol uti ons.
Cox and Merz (1958) observed that the compl ex vi scosi ty i s nearl y
equal to the steady shear vi scosi ty when the shear rate and frequency
are equal :
[5.72]
Thi s empi ri cal rel ati onshi p, now referred to as the "Cox-Merz rul e,"may
be useful for materi al s that are more easi l y tested under osci l l atory
i nstead of steady shear condi ti ons. The rul e may, for exampl e, appl y to

0

0
G
2
1
1
1 0

1
2
1
1
1
0
1
f (
) (
11

22
)/(
)
2
5.7 Viscometric and Linear Viscoelastic Functions 339

pol ymeri c fl ui ds that have a l arge normal stress di fference, l eadi ng to
a Wei ssenberg effect (rod cl i mbi ng), that compl i cates steady shear
testi ng.
The Cox-Merz rul e seems to work wel l wi th many syntheti c and
bi opol ymer di spersi ons (da Si l va and Rao, 1992). Rao and Cool ey (1992)
found that Eq. [5.72] coul d be appl i ed to tomato paste by i ntroduci ng a
si mpl e shi ft factor i nto the computati on: Compl ex vi scosi ty was eval u-
ated at a frequency of where , the shi ft factor, was equal to
approxi matel y 0.0074. Bi stany and Koki ni (1983) found the Cox-Merz
rul e to be i nappl i cabl e to vari ous foods. I n the same work, however, the
fol l owi ng rel ati onshi ps were abl e to effecti vel y rel ate steady shear and
dynami c rheol ogi cal properti es:
[5.73]
and
[5.74]
where , , , and are empi ri cal constants determi ned from
experi mental data. Val ues of these constants for vari ous foods are gi ven
i n Tabl e 5.9. The correspondi ng normal stress di fferences and power
l aw fl ui d properti es (needed tocal cul ate apparent vi scosi ty)for the foods
l i sted i n Tabl e 5.9 are gi ven i n Appendi x [6.6].
Dorai swamy et al . (1991) proposed an extended Cox-Merz rul e
rel ati ng the compl ex vi scosi ty to the steady shear vi scosi ty. They found
that a pl ot of the compl ex vi scosi ty versus was equi val ent to a pl ot
of the apparent vi scosi ty versus shear rate. was defi ned as the
"effecti ve shear rate"where i s the strai n ampl i tude gi ven i n Eq. [5.30].
The equi val ence of compl ex and steady shear vi scosi ti es was found to
be val i d over a wi de range of parameters i nvol vi ng concentrated sus-
pensi ons havi ng yi el d stresses.

G
2
C(
1
)
*
C()
C C
o
Table 5.9. Empirical Constants
*
for Eq. [5.73] and [5.74] (Source: Kokini, 1992)
Food
Whipped cream cheese 93.21 0.750 13.87 1.146
Cool Whip 50.13 1.400 6.16 1.098
Stick Butter 49.64 0.986 0.79 1.204
Whipped butter 43.26 0.948 33.42 1.255
Stick margarine 35.48 0.934 1.28 1.140
Ketchup 13.97 0.940 14.45 1.069
Peanut butter 13.18 1.266 1.66 1.124
Squeeze margarine 11.12 1.084 52.48 1.022
Canned frosting 4.40 1.208 4.89 1.098
Marshmallow fluff 3.53 0.988 1.26 0.810
* Data collected at room temperature over a shear rate and frequency range of 0.1
to 100 s
-1
. Units of and used in Eq. [5.73] and [5.74] are rad s
-1
, s
-1
,
Pa, Pa s
2
, Pa s and Pa s, respectively.
Table 5.10. Stress Relaxation Data from Compression Testing of Apple Tissue at
Room Temperature (Cylindrical Sample: Length = 2 cm, Diameter = 2 cm)
(s) (Pa) (s)
0.0 754
0.6 601 3.0
1.2 572 5.0
1.8 562 7.1
2.4 549 8.8
3.0 536 10.4
3.6 526 11.9
6.0 504 18.1
9.0 484 25.1
12.0 468 31.6
15.0 461 38.6
18.0 452 44.9
24.0 435 56.7
30.0 426 69.0
42.0 409 91.8
54.0 406 117.0
C C
,
, G,
1
,
*
t
o
t
o

5.8.1 Generalized Maxwell Model of Stress Relaxation 341
5.8.1. Generalized Maxwell Model of Stress Relaxation
Usi ng the general i zed Maxwel l model (Eq. [5.11]), esti mate the rel ax-
ati on ti me ( ) of the appl e ti ssue descri bed by the data presented i n
Tabl e 5.10.
Recal l Eq. [5.11], wi th referri ng to the stress i n compressi on:
Assume the materi al has reached equi l i bri um at the compl eti on of the
test so = 406 Pa. Then, regressi on of the data (Fi g. 5.27) usi ng the
l ogari thmi c transformati on of Eq. [5.11],
yi el ds s
-1
or s, and Pa maki ng
=606.75 Pa. Note that the val ue of , determi ned from the regressi on
parameter, i s l ower than the experi mental val ue of 754 Pa gi ven for the
i ni ti al stress i n Tabl e 5.10.
Figure 5.27. Stress relaxation data for apple tissue modeled using Eq. [5.11].
rel
f (t )
e
+ (
o

e
) exp
rel
_
e
ln( 406) ln(
o

e
) t /
rel
1/
rel
0.0906
rel
11.04
o
406 200.75
o

o
0 10 20 30 40 50 60
0
100
200
300
400
Time, s
e
,

P
a
e
Apple Tissue
= 406 Pa
5.8.2. Linearized Stress Relaxation
Stress rel axati on data for appl e ti ssue are gi ven i n Tabl e 5.10 wi th the
raw data pl otted i n Fi g. 5.28. Determi ne the coeffi ci ents for the nor-
mal i zed stress equati on proposed by Pel eg and Normand (1983), Eq.
[5.14]:
where the val ue of i s taken from Tabl e 5.10 as 754 Pa.
Li near regressi on of thi s equati on (an excel l ent fi t) yi el ds the fol -
l owi ng: = 4.52 s, = 2.12. Li neari zed data are pl otted i n Fi g. 5.29.
Resul ts are comparabl e to those reported for potato fl esh (Tabl e 5.5:
= 4.40 s, = 1.56).
Figure 5.28. Stress relaxation data for apple tissue at room temperature.
o
t
o

k
1
+ k
2
t
o
k
1
k
2
k
1
k
2
0 10 20 30 40 50 60
300
400
500
600
700
800
Time, s
S
h
e
a
r

S
t
r
e
s
s
,

P
a
Apple Tissue
5.8.3 Analysis of Creep Compliance Data 343
Figure 5.29. Normalized stress relaxation data for apple tissue.
5.8.3. Analysis of Creep Compliance Data
Esti mate the parameters of the four parameter Burgers model (Fi g. 5.6)
to represent the shear creep compl i ance data gi ven i n Tabl e 5.11.
Compare the resul ti ng curve to the 6 parameter model (Fi g. 5.30) for-
mul ated by addi ng an addi ti onal Kel vi n el ement to the Burgers model .
Al so, determi ne the parameters for the si mpl e l i neari zed creep
compl i ance model suggested by Pel eg (1980).
The Burgers model , expressed i n terms of the creep compl i ance
functi on, i s descri bed by Eq. [5.22]:
Model parameters may be esti mated usi ng the fol l owi ng procedure:
1. The i nstantaneous compl i ance, , i s determi ned from the ori gi nal
data (Curve 1 on Fi g. 5.31 or Tabl e 5.11) at as 0.00220
cm
2
/dyne.
0 10 20 30 40 50 60
0
20
40
60
80
100
120
140
Time, s
Apple Tissue
,

s
o
t
o
-
J f (t ) J
0
+ J
1
1 exp
ret
_
,
_
,
+
t
0
J
0
t 0
Table 5.11. Shear Creep Compliance of Skim Milk Curd Fortified with -casein
(Data from Halim and Shoemaker, 1990)
Data Time J
a
J
b
J
c
t/J
d
Point (s) (cm
2
/dyne) (cm
2
/dyne) (cm
2
/dyne) (s dyne/cm
2
)
1 0 0.00220 0.00220 0.00220 0
2 0.89 0.00249 0.00237 0.00241 357
3 2.01 0.00264 0.00257 0.00262 761
4 3.79 0.00293 0.00284 0.00287 1,293
5 8.93 0.00331 0.00337 0.00328 2,698
6 14.06 0.00355 0.00369 0.00352 3,961
7 18.97 0.00372 0.00389 0.00369 5,099
8 23.44 0.00386 0.00401 0.00382 6,073
9 28.71 0.00398 0.00411 0.00395 7,241
10 34.02 0.00408 0.00419 0.00406 8,338
11 39.06 0.00418 0.00426 0.00416 9,344
12 44.20 0.00426 0.00432 0.00424 10,376
13 48.88 0.00434 0.00437 0.00432 11,263
14 58.48 0.00447 0.00447 0.00445 13,083
15 68.97 0.00458 0.00458 0.00458 15,059
16 78.57 0.00469 0.00468 0.00469 16,752
17 93.75 0.00485 0.00484 0.00485 19,330
18 109.25 0.00499 0.00500 0.00502 21,894
19 118.97 0.00509 0.00509 0.00512 23,373
a
rawdata;
b
predictedusing 4parameter Burgers model;
c
predictedusing 6parameter
model;
d
calculated from raw data
2. Subtract from Curve 1 to generate Curve 2 (Fi g. 5.31). Usi ng
the strai ght porti on of the curve (the l ast si x data poi nts, ),
l i near regressi on anal ysi s yi el ds and from the i ntercept and
sl ope of
[5.75]
as = 0.00168 cm
2
/dyne and = 97,847 P. Note that refl ects
the ful l y extended (equl i brated) Kel vi n el ement maki ng the
exponenti al term i n the ori gi nal equati on equal to zero.
J
0
t 58.48
J
1

0
( J J
0
) J
1
+
t
0
J
1

0
J
1
3. Usi ng the exponenti al porti on of the data (the fi rst 8 data poi nts,
s), the retardati on ti me i s determi ned from l i near regres-
si on anal ysi s (i ncl udi ng a l ogari thmi c transformati on of the data)
over of
[5.76]
as = 8.635 s.
Figure 5.30. Six parameter model for creep compliance.
Substi tuti ng the constants found above, the compl ete Burgers model
may be expressed as
t 28.71
J <J
1
+J
0
ln
1
J J
0
J
1
_
,

t
ret
ret
G
1
G
1
0
0
G
2 2
J 0.0022 + 0.00168
1 exp
t
8.635
_
,
_
,
+
t
97847

Figure 5.31. Shear creep compliance of skim milk curd fortified with -casein.
A compari son of the curve predi cted wi th the above equati on and actual
data reveal s (Fi g. 5.31, Tabl e 5.11) that the Burgers model adequatel y
represents the data; however, better accuracy i s obtai ned usi ng a more
compl ex 6 parameter model (Fi g. 5.30) descri bed by Eq. [5.23], wi th
, as
[5.77]
where:
[5.78]
and
[5.79]
0 20 40 60 80 100 120
0
0.001
0.002
0.003
0.004
0.005
0.006
Time, s
C
r
e
e
p

C
o
m
p
l
i
a
n
c
e
,

c
m

/
d
y
n
e
2
6 parameter model
4 parameter Burgers model
Curve 1, raw data
Curve 2, adjusted raw data
Skim Milk Curd
m 2
J f (t ) J
0
+ J
1
1 exp
t
(
ret
)
1
_
,
_
,
+ J
2
1 exp
t
(
ret
)
2
_
,
_
,
+
t
0
(
ret
)
1

1
G
1
(
ret
)
2

2
G
2
The constants i n Eq. [5.77] (for the same -casei n forti fi ed ski n mi l k
curd under consi derati on) were determi ned by Hal i m and Shoemaker
(1990), usi ng nonl i near curve fi tti ng procedures, wi th the fol l owi ng
resul t:
Compari ng thi s predi cti on equati on to the raw data shows that i t does
an excel l ent job model i ng the shear creep compl i ance (Fi g. 5.31, Tabl e
5.11).
Figure 5.32. Linearized shear creep compliance of skim milk curd fortified with
-casein.
Pel eg (1980) suggested that creep data coul d be model ed wi th the
fol l owi ng l i near equati on:
[5.80]
J 0.0022 + 0.00064
1 exp
t
3.14
_
,
_
,
+ 0.001068
1 exp
t
19.5
_
,
_
,
+
t
97707
0 20 40 60 80 100 120 140
0
5,000
10,000
15,000
20,000
25,000
Time, s
t

/

J
,

s

d
y
n
e
/
c
m
2
Skim Milk Curd
t
J
k
1
+ k
2
t
Regressi on anal ysi s of the data, usi ng Eq. [5.80], yi el ded dyne
cm
-2
s and dyne cm
-2
wi th . Raw data and the resul ti ng
curve are pl otted i n Fi g. 5.32. The 6 parameter model i nvol ves compl ex
curve fi tti ng, but gi ves a very accurate representati on of experi mental
data. The si mpl e l i near equati on i s l ess preci se, but often adequate for
i nvesti gati ng practi cal creep probl ems i nvol vi ng bi ol ogi cal materi al s.
5.8.4. Plotting Oscillatory Data
Devel op al ternati ve pl ots showi ng the dynami c behavi or of the con-
centrated sol uti on (5%l ambda carrageenan sol uti on) i l l ustrated i n Fi g.
5.21.
The data gi ven i n Fi g. 5.22 were fi t to a power l aw model resul ti ng
i n the fol l owi ng equati ons (Tabl e 5.8):
and
Usi ng these equati ons other rheol ogi cal parameters can be easi l y cal -
cul ated:
Resul ts are presented i n Fi g. 5.33.
Two very useful methods of representi ng mechani cal spectra are to
pl ot the storage modul us and the l oss modul us, or the storage modul us
and the dynami c vi scosi ty. Crossover poi nts (Fi g. 5.33), such as the one
provi ded by the i ntersecti on of and , may provi de useful bench-
marks for compari ng products or treatments. I n thi s case the crossover
occurs at a frequency of 5.61 rad/s. Si nce at that poi nt, i t i s al so
the frequency where the tangent of the phase angl e ( ) i s equal to 1
k
1
1050.8
r
2
0.99 k
2
196.6
G 16.26()
.84
G 27.78()
.52

G
27.78()
.48

G
16.26()
.16
tan
G
G
1.71()
.32
G
*
(G)
2
+ (G)
2
(16.26()
.84
)
2
+ (27.78()
.52
)
2
*
()
2
+ ()
2
(27.78()
.48
)
2
+ (16.26()
.16
)
2
G G
G G
tan
5.8.4 Plotting Oscillatory Data 349
Figure 5.33. Oscillatory testing results for a 5% lambda carrageenan solution.
because . Observi ng changes i n i s a useful means of eval u-
ati ng the transi ti on from l i qui d-l i ke to sol i d-l i ke behavi or: tan del ta
decreases wi th more sol i d-l i ke behavi or. Combi ned parameters, and
, are al so useful i n l ooki ng at overal l rheol ogi cal behavi or. A popul ar
combi nati on i s to pl ot the vari ati on of , , and wi th frequency. I n
some cases, si ngl e poi nt measurements are useful : , measured at 50
rad/s, correl ates wel l wi th the human percepti on of thi ckness or vi scosi ty
(Hi l l , 1991; Hi l l et al ., 1995).
0.01 0.03 0.1 0.3 1 3 10 30 100
0.1
0.3
1
3
10
30
100
300
1,000
P
a

o
r

P
a

s

o
r

D
i
m
e
n
s
i
o
n
l
e
s
s
(rad/s)
G
tan
G
*
*
G
/4 tan
*
G
*
*
G G
*
6 . Appe ndic e s
6 . 1 . Conve rs ion Fac t ors and SI Pre fixe s
Density
1 g cm
-3
= 1000 kg m
-3
= 62.428 lbm ft
-3
= 0.0361 lbm in
-3
1 lbm ft
-3
= 16.0185 kg m
-3
Force
1 N = 1 kg m s
-2
= 10
5
dyne = 0.22481 lbf = 0.102 kgf
1 lbf = 4.448 N = 0.4536 kgf
1 dyne = 1 g cm s
-2
= 10
-5
N
Length
1 m = 100 cm = 1000 mm = 10
6
m = 3.2808 ft = 39.37 in = 1.0936 yd
1 in = 2.54 cm = 25.40 mm = 0.0254 m = 0.0833 ft = 0.02778 yd
Power
1 hp = 550 ft lbf s
-1
= 745.70 W = 0.7457 kW = 0.7068 Btu s
-1
1 Btu hr
-1
= 0.2931 W = 0.2931 J s
-1
Pressure and Stress
1 bar = 10
5
N m
-2
= 10
5
Pa = 14.5038 lbf in
-2
= 0.987 atm = 10.2 m at 4 C
1 Pa = 1 N m
-2
= 10 dyne cm
-2
= 9.8692 (10
-6
) atm = 0.1020 kgf m
-2
1 lbf in
-2
= 6895 Pa = 6.804 (10
-2
)atm = 6.895 kPa = 2.307 ft at 4 C
1 dyne cm
-2
= 0.10 Pa = 10
-6
bar = 0.987 (10
-6
) atm
1 atm = 1.01325 (10
5
) N m
-2
= 101.325 kPa = 14.696 psi = 1.013 bar = 760 torr
1 atm = 760 mm Hg at 0 C = 33.90 ft at 4 C = 1.013 (10
6
) dyne cm
-2
Revolution and Rotational Speed
1 rev = 1 cycle = 2 rad
1 Hz = 1 cycle/s = 1 rev/s = 2 rad/s = 6.283 rad/s
1 rev/min = 0.1047 rad/s
Temperature
T
Kelvin
= T
Celsius
+ 273.15
T
Kelvin
= (T
Fahrenheit
+ 459.67) / 1.8
T
Fahrenheit
= 1.8 T
Celsius
+ 32
T
Celsius
= (T
Fahrenheit
- 32) / 1.8
Torque, Energy, and Work
1 N m = 100 N cm = 1 J = 1 kg m
2
s
-2
= 10
7
dyne cm = 8.85 in lbf
1 dyne cm = 10
-7
N m = 10
-5
N cm
Viscosity (Absolute or Dynamic, followed by Kinematic)
1 P = 1 dyne s cm
-2
= 0.1 Pa s = 100 cP = 100 mPa s
1 Pa s = 1000 cP = 10 P = 1 kg m
-1
s
-1
= 1 N s m
-2
= 0.67197 lbm ft
-1
s
-1
1 cP = 1 mPa s = 0.001 Pa s = 0.01 P
1 lbm ft
-1
s
-1
= 1.4882 kg m
-1
s
-1
= 1488.2 cP = 2.0885 (10
-2
) lbf s ft
-2
kinematic viscosity (cSt) = absolute viscosity (cP) / density (g cm
-3
)
1 cSt = 0.000001 m
2
s
-1
= 1 mm
2
s
-1
= 5.58001 in
2
hr
-1
= 0.00155 in
2
s
-1
1 St = 100 cSt = 0.0001 m
2
s
-1
1 m
2
s
-1
= 10
-5
cSt = 10.7639 ft
2
s
-1
Volume
1 m
3
= 10
6
cm
3
= 10
3
L (liter) = 264.17 gal (US) = 35.315 ft
3
= 219.97 gal (UK)
1 ft
3
= 0.028317 m
3
= 7.481 gal (US) = 28.317 L = 6.2288 gal(UK)
1 gal (US) = 4 qt = 3.7854 L = 3785.4 cm
3
= 0.8327 gal (UK) = 0.003785 m
3
H
2
O
H
2
O
H
2
O
351
SI Prefixes
Factor Prefix Symbol Factor Prefix Symbol
10
18
exa E 10
-1
deci d
10
15
peta P 10
-2
centi c
10
12
tera T 10
-3
milli m
10
9
giga G 10
-6
micro
10
6
mega M 10
-9
nano n
10
3
kilo k 10
-12
pico p
10
2
hecto h 10
-15
femto f
10
1
deka da 10
-18
atto a
6 . 2 . Gre e k Alphabe t
Alpha Nu
Beta Xi
Gamma Omicron
Delta Pi
Epsilon Rho
Zeta Sigma
Eta Tau
Theta Upsilon
Iota Phi
Kappa Chi
Lambda Psi
Mu Omega

352 Appendices
6 . 3 . Mat he mat ic s : Root s , Powe rs , and Logarit hms
Roots and Powers
*
Logarithms
*
If , then is the logarithm of to the base where is a
finite positive number other than 1. Also, if then . In
this text, values of equal to 10 and are used meaning
and
The following properties of logarithms, expressed in terms of base
, are true for any base:
*
Based on a summary given in Hudson, R.G. 1939. The Engineers Manual (second edition, 25th
printing). John Wiley & Sons, NY.
a
n
= a a a a to n factors
a
n
=
1
a
n
a
m
a
n
= a
m + n
a
m
a
n
= a
m n
(ab)
n
= a
n
b
n
a
b
n
=
a
n
b
n
(a
m
)
n
= (a
n
)
m
= a
mn
(
n
a )
n
= a
a
1
n
=
n
a a
m
n
=
n
a
m
n
ab = (
n
a ) (
n
b ) n
a
b
=
n
a
n
b
log
b
N = x x N b b
b
x
= N log
b
N = x
b e
log
10
N = x or 10
x
= N
log
e
N = lnN = x or e
x
= exp( x) = N
e
ln(MN) = ln(M) + ln(N)
ln
M
N
= ln(M) ln(N)
ln(M
a
) = a ln(M)
ln
a
M
c
=
c
a
ln(M)
353
6 . 4 . Line ar Re gre s s ion Analys is of Two Variable s
Linear regression is a statistical method of fitting two variables
to a linear equation:
The constants, (the intercept) and (the slope), are calculated with
the following formulas:
where:
= number of observations in the sample
= mean of x values
= mean of y values
The method of least squares minimizes the sum of the squared
error between the actual and estimated values: and are found
using a procedure which minimizes
where is the value to be estimated and is the corresponding
estimate of at . Computations are easily done using computer
spread-sheet programs or hand calculators.
A measure of correlation between two variables can be defined in
terms of the sample correlation coefficient ( ) which can take on
values between -1 and 1. A correlation value of zero means there
is no association. A value of -1 means there is a perfect negative
correlation and a value of +1 means there is a perfect positive cor-
relation (Fig. 6.1). Correlation may also be expressed in terms of
the coefficient of determination, . When , all points fall on the
predicted line. If , there is as much variation in the estimate
of as there is in the variable meaning that any observed fluc-
tuations are due to random variations in .
The standard error of the estimate can also be a useful measure
of how well the regression line fits the data. This parameter is
calculated as
y = a + b x
a b
b =
n
i = 1
n
x
i
y
i

i = 1
n
x
i

i = 1
n
y
i
n
i = 1
n
( x
i
)
2

i = 1
n
x
i
2
a = y bx
n
x
y
a b
n
i = 1
n
( y
i
y
x
)
2
y
i
y
x
y x
r
r
2
r
2
= 1
r
2
= 0
y y
y
354 Appendices
where:
= the standard error of the estimate of on
Properties of the standard error of the estimate are analogous to
those of standard deviation. Assuming a sufficiently large sample,
pairs of lines constructed parallel to the regression line of on , at
vertical distances of , 2 and 3 , should include 68%, 95%
and 99.7% (respectively) of the sample data points.
Figure 6.1. Relationship between and for a perfect positive correlation (r =
1), a perfect negative correlation (r = -1), and no correlation (r = 0).
Example Proble m. Determine the power law, Bingham plastic,
Casson, and Herschel-Bulkley model parameters from the data for
the carrageenan gum solution given in Table 6.1.
Linear regression can only be performed on a linear equation of
the form . The Binghamplastic model already exists in this
form; however, the other models may require a transformation of
the data before they can be presented in a linear form. A summary
s
y, x
=
i = 1
n
(y
i
y
x
)
2
n 2
s
y, x
y x
y x
s
y, x
s
y, x
s
y, x
r = -1
r = 1
r = 0
x
y
y x
y = a + bx
355
Table 6.1. Raw and Transformed Data of a 1% aqueous solution of carrageenan
gumat 25 C (Data from Prentice and Huber, 1983)
2.61 9.88 0.96 2.29 1.62 3.14 0.412
2.97 11.4 1.09 2.43 1.72 3.38 0.626
2.81 12.9 1.03 2.56 1.68 3.59 0.536
3.44 14.1 1.24 2.65 1.85 3.75 0.850
3.80 17.6 1.34 2.87 1.95 4.20 0.993
4.85 26.3 1.58 3.27 2.20 5.13 1.322
6.61 42.0 1.89 3.74 2.57 6.48 1.707
6.19 48.6 1.82 3.88 2.49 6.97 1.627
5.89 49.3 1.77 3.90 2.43 7.02 1.567
7.22 55.5 1.98 4.02 2.69 7.45 1.812
8.20 58.8 2.10 4.07 2.86 7.67 1.960
9.08 75.4 2.21 4.32 3.01 8.68 2.076
11.63 104.1 2.45 4.65 3.41 10.20 2.354
10.65 110.4 2.37 4.70 3.26 10.51 2.257
12.75 120.5 2.55 4.79 3.57 10.98 2.455
13.10 136.5 2.57 4.92 3.62 11.68 2.484
14.90 145.8 2.70 4.98 3.86 12.07 2.624
15.85 187.1 2.76 5.23 3.98 13.68 2.691
12.70 210.2 2.54 5.35 3.56 14.50 2.451
20.50 270.0 3.02 5.60 4.53 16.43 2.965
of this transformation, and regression results, are given in Table
6.2. Note that the yield stress for the Herschel-Bulkley model must
be estimated before linear regression can proceed. Estimating ,
, and simultaneously would require a nonlinear estimation
procedure.
Information presented in Table 6.2 can be presented in terms of the
models under investigation:
Bingham, :
Pa, Pa s
Herschel-Bulkley, :
Pa, Pa s
n
,
Casson, :
Pa, Pa
.5
s
.5
Power Law, :
Pa s
n
,
()
.5
(
)
.5

ln ln
ln( 1.1)
o
K n
=
pl
(
) +
o
o
= 3.20
pl
= 0.066
= K(
)
n
+
o
o
= 1.10 K = 0.306 n = 0.74
()
.5
= K
1
(
)
.5
+ (
o
)
.5
o
= 1.16 K
1
= 0.211
= K(
)
n
K = 0.66 n = 0.60
356 Appendices
Coefficients of determination (Table 6.2) indicate that the power law
and Herschel-Bulkley equations provide the best model of the
experimental data.
Table 6.2. Linear regression parameters for different rheological models.
Model
Bingham 0.92 1.44
Power Law 0.98 0.094
Casson 0.96 0.179
*
Herschel- 0.98 0.119
Bulkley
*
Pa was determined from a graphical analysis of the data.
r
2
y x a b s
y, x

o
= 3.20
pl
= 0.066
ln ln
lnK = 0.411 n = 0.60

()
.5
(
)
.5
(
o
)
.5
= 1.078 K
1
= 0.211
ln(
o
) ln
lnK = 1.185 n = 0.74
o
= 1.1
357
6 . 5 . Hooke an Prope rt ie s
Material Material
(N m
-2
) (N m
-2
)
Soft foam rubber 10
2
Carrots (2-4) x 10
7
Rubber 8 x 10
5
Pears (1.2-3) x 10
7
Dry Spaghetti 0.3 x 10
10
Potatoes (0.6-1.4) x 10
7
Lead 1.0 x 10
10
Apples raw (0.6-1.4) x 10
7
Concrete 1.7 x 10
10
Gelatin (gel) 0.02 x 10
7
Glass 7 x 10
10
Peach (0.2-2) x 10
7
Iron 8 x 10
10
Banana (0.08-0.3) x 10
7
Steel 25 x 10
10
Material Material
( - ) ( - )
Cheddar cheese 0.50 Copper 0.33
Potato tissue 0.49 Steel 0.30
Rubber 0.49 Glass 0.24
Apple tissue 0.37 Bread-crumbs 0.00
Apple 0.21-0.34 Cork 0.00
Material Material
(N m
-2
) (N m
-2
)
Dough 1.4 x 10
6
Silver 10
11
Rubber 1.9 x 10
7
Steel 1.6 x 10
11
Granite 3 x 10
10
Glass 3.9 x 10
10
Source: Lewis, 1987.
E E

K K
358 Appendices
6 . 6 . St e ady She ar and Normal St re s s Diffe re nc e
Product
*
( ) (-) ( ) (-)
Apple butter 222.90 0.145 156.03 0.566
Canned frosting 355.84 0.117 816.11 0.244
Honey 15.39 0.989 - -
Ketchup 29.10 0.136 39.47 0.258
Marshmallow cream 563.10 0.379 185.45 0.127
Mayonnaise 100.13 0.131 256.40 -0.048
Mustard 35.05 0.196 65.69 0.136
Peanut butter 501.13 0.065 3785.00 0.175
Stick butter 199.29 0.085 3403.00 0.393
Stick margarine 297.58 0.074 3010.13 0.299
Squeeze margarine 8.68 0.124 15.70 0.168
Tub margarine 106.68 0.077 177.20 0.358
Whipped butter 312.30 0.057 110.76 0.476
Whipped cream cheese 422.30 0.058 363.70 0.418
Whipped desert topping 35.98 0.120 138.00 0.309
*
Data were collected over a shear rate of 0.1 to 100 s
-1
at 30.5 C
Source: Dickie and Kokini, 1983.
K n K n
Pa s
n
Pa s
n
= K(
)
n
N
1
=
11

22
= K(
)
n
359
6 . 7 . Yie ld St re s s of Fluid Foods
Product (Pa) Measurement Method Source
Ketchup 22.8 extrapolation Ofoli et al. (1987)
Mustard 34.0 extrapolation Ofoli et al. (1987)
Miracle Whip 54.3 extrapolation Ofoli et al. (1987)
Apricot Puree 17.4 extrapolation Ofoli et al. (1987)
Milk Chocolate 10.9 extrapolation Ofoli et al. (1987)
Minced Fish Paste 1600-2300 extrapolation Nakayama et
al.(1980)
Mayonnaise 24.8-26.9 stress to initiate flow De Kee et al.(1980)
Ketchup 15.4-16.0 stress to initiate flow De Kee et al.(1980)
Tomato Paste 83.9-84.9 stress to initiate flow De Kee et al.(1980)
Raw Meat Batter 17.9 extrapolation Toledo et al.(1977)
Tomato Puree 23.0 stress decay Charm(1962)
Applesauce 58.6 stress decay Charm(1962)
Tomato Paste 107-135 squeezing flow Campanella and
Peleg(1987)
Ketchup 18-30 squeezing flow Campanella and
Peleg(1987)
Mustard 52-78 squeezing flow Campanella and
Peleg(1987)
Mayonnaise 81-91 squeezing flow Campanella and
Peleg(1987)
Applesauce 45-87 squeezing flow Campanella and
Peleg(1987)
Applesauce 46-82 vane method Qiu and Rao(1988)
Ketchup 26-30 vane method Missaire et al.
(1990)
Spaghetti Sauce 24-28 vane method Missaire et al.
(1990)
Tomato Puree 25-34 vane method Missaire et al.
(1990)
Pumpkin Filling 20 vane method Missaire et al.
(1990)
Applesauce 38-46 vane method Missaire et al.
(1990)
Baby food, pears 49 vane method Missaire et al.
(1990)
Baby food, 25 vane method Missaire et al.
peaches (1990)
Baby food, carrots 71 vane method Missaire et al.
(1990)
o
360 Appendices
Static and Dynamic Yield Stresses of Food Products
Product
*
Static Yield Dynamic Yield
Stress Stress
(Pa) (Pa)
Tomato puree, Brand A 34.4 3.7 28.1 4.2
Tomato puree, Brand B 30.0 4.2 27.6 3.4
Apple sauce, Brand A 77.3 0.0 48.2 4.7
Apple sauce, Brand B 48.2 4.7 38.0 4.7
Tomato ketchup, Brand A 51.3 5.0 40.6 4.5
Tomato ketchup, Brand A 43.2 3.4 39.6 3.4
Spaghetti sauce, Brand A 26.3 4.5 18.3 0.0
Spaghetti sauce, Brand B 24.8 3.4 21.9 0.0
Baby food, pears 31.8 5.0 24.8 4.0
Baby food, carrots 64.0 4.0 35.7 5.0
Baby food, peaches 22.9 3.4 20.1 3.4
Mayonnaise, Brand A 204.4 5.0 112.6 4.0
Mayonnaise, Brand B 163.8 4.2 99.4 4.0
Mustard, Brand A 82.5 5.0 53.1 5.3
Mustard, Brand B 103.8 5.0 53.4 5.0
*
Data were collected at 25 C by using the vane method operating
in the controlled rate mode. Source: Yoo et al., 1995.
Briggs and Steffe (1996) determined the yield stress of frozen ice
cream using the vane method. Results were summarized with the
following equations:
where is the yield stress in Pa, and is the temperature in .
These equations are valid over the range of temperatures typically
maintained for hand-dipped ice cream: -12 to -16 . Yield
stresses, ranging from 2.5 to 8.0 kPa, decreased with reductions in
temperature. Chocolate had sightly higher values than vanilla.

o
= 22.33exp(.37T) for chocolate ice cream
o
= 39.32exp(.33T) for vanilla ice cream
o
T C
C C
361
6 . 8 . Ne wt onian Fluids
Product T
( C) (cP)
Acetic acid 15 1.31
18 1.30
25.2 1.155
30 1.04
41 1.00
59 0.70
70 0.60
100 0.43
Acetone -42.5 0.695
-30.0 0.575
-20.9 0.510
-13.0 0.470
-10.0 0.450
0 0.399
15 0.337
25 0.316
30 0.295
41 0.280
Ethanol -32.01 3.84
-17.59 2.68
-0.30 1.80
0 1.773
10 1.466
20 1.200
30 1.003
40 0.834
50 0.702
60 0.592
70 0.504
Glycol 20 19.9
40 9.13
60 4.95
80 3.02
100 1.99
Glycerin -42 6.71E6
-36 2.05E6
-25 2.62E5
-20 1.34E5
-15.4 6.65E4
-10.8 3.55E4
-4.2 1.49E4
362 Appendices
0 1.21E4
6 6260
15 2330
20 1490
25 954
30 629
Mercury -20 1.855
-10 1.764
0 1.685
10 1.615
20 1.554
30 1.499
40 1.450
50 1.407
60 1.367
70 1.331
80 1.298
90 1.268
100 1.240
150 1.300
200 1.052
250 0.995
300 0.950
340 0.921
Water 0 1.787
5 1.519
10 1.307
15 1.139
20 1.002
25 0.8904
30 0.7975
35 0.7194
40 0.6529
45 0.5960
50 0.5468
55 0.5040
60 0.4665
65 0.4335
70 0.4042
75 0.3781
80 0.3547
85 0.3337
90 0.3147
95 0.2975
100 0.2818
Source: Weast et al., 1984.
363
Vis c os it y Pre dic t ion Equat ions for Various Liquids
Equation 1
Equation 2
Equation 3
where is in units of Pa s and T in degrees Kelvin.
Material Eq. # T range(K)
Acetic Acid 2 1.2106x10
6
-3.6612 270-390
Corn Oil 3 -3.5581 -263.32 183.60 290-340
Corn Syrup: DE=100, 3 -4.4500 -234.80 159.91 280-360
20% dry substance
Corn Syrup: DE=42.9, 3 -4.0137 -129.07 201.23 280-360
20% dry substance
Corn Syrup: DE=35.4, 1 -5.5478 868.35 280-360
20% dry substance
Corn Syrup: DE=35.4, 3 -3.8025 -258.10 186.88 280-360
50% dry substance
Corn Syrup: DE=42.9, 3 -3.9975 -267.36 182.96 280-360
50% dry substance
Corn Syrup: DE=75.4, 1 -5.8508 924.84 201.23 280-360
20% dry substance
Ethanol 3 -5.5972 -846.95 -24.124 210-350
Ethylene Glycol 3 -4.5448 -417.05 146.53 280-420
Groundnut Oil 3 -3.9621 -407.46 151.23 290-340
Rapseed Oil 3 -4.4802 -597.20 1119.99 290-340
Soybean Oil 3 -4.4977 -581.28 115.71 290-340
Sunflower Oil 3 -3.6505 -304.27 168.98 290-340
Water 3 -4.5318 -220.57 149.39 270-380
Source: Viswanath and Natarajan, 1989.
log() = A + (B/T)
= AT
B
log() = A + B/(C T)
A B C
364 Appendices
Vis c os it y (c e nt ipois e ) of Suc ros e Solut ions
% sucrose by weight
T (
o
C) 20 40 60
0 3.818 14.82 -
5 3.166 11.60 -
10 2.662 9.830 113.90
15 2.275 7.496 74.90
20 1.967 6.223 56.70
25 1.710 5.206 44.02
35 1.336 3.776 26.62
40 1.197 3.261 21.30
45 1.074 2.858 17.24
50 0.974 2.506 14.06
55 0.887 2.227 11.71
60 0.811 1.989 9.87
65 0.745 1.785 8.37
70 0.688 1.614 7.18
80 0.592 1.339 5.42
85 0.552 1.226 4.75
90 - 1.127 4.17
95 - 1.041 3.73
Source: International Critical Tables 5:23 (1917).
Vis c os it y (c e nt ipois e ) of Et hanol-Wat e r Mixt ure s
Concentration of ethanol, % by weight
T (
o
C) 10 20 30 40 50 60 70 80 90 100
0 3.311 5.32 6.94 7.14 6.58 5.75 4.76 3.69 2.73 1.77
5 2.58 4.06 5.29 5.59 5.26 4.63 3.91 3.12 2.31 1.62
10 2.18 3.16 4.05 4.39 4.18 3.77 3.27 2.71 2.10 1.47
15 1.79 2.62 3.26 3.53 3.44 3.14 2.77 2.31 1.80 1.33
20 1.54 2.18 2.71 2.91 2.87 2.67 2.37 2.01 1.61 1.20
25 1.32 1.82 2.18 2.35 2.40 2.24 2.04 1.75 1.42 1.10
30 1.16 1.55 1.87 2.02 2.02 1.93 1.77 1.53 1.28 1.00
35 1.01 1.33 1.58 1.72 1.72 1.66 1.53 1.36 1.15 0.91
60 0.91 1.16 1.37 1.48 1.50 1.45 1.34 1.20 1.04 0.83
Source: Matz, 1962.
365
Vis c os it y (c e nt ipois e ) of Corn Syrup
1
DS T ( F)
2
DE=35.4 DE=42.9 DE=53.7 DE=75.4 DE=92.4
85 60 0 0 457000 -
80 7080000 1410000 537000 83200 -
100 1000000 227000 85200 17000 -
120 188000 50100 20000 4270 -
140 44900 13000 6310 1660 -
160 13000 5190 2290 589 -
180 4420 1760 944 275 -
80 60 266000 89100 24000 -
80 126000 59600 17800 4570 -
100 29900 15000 5010 1550 -
120 9810 4840 1800 603 -
140 3350 1860 785 282 -
160 1410 851 367 141 -
180 687 386 196 75.9 -
75 60 39800 18200 7590 6030 -
80 10000 5390 2140 741 501
100 3020 1880 807 331 211
120 1260 817 372 159 106
140 620 389 191 83.2 56.2
160 325 197 103 47.9 32.4
180 180 110 62 28.8 19.5
65 60 1060 560 389 178 -
80 398 237 159 77.6 56.2
100 182 119 83.2 45.7 30.7
120 108 69.2 47.3 26.3 18.6
140 67.9 43.4 29.0 16.2 12.2
160 43.2 26.6 18.6 10.7 8.41
180 29.0 17.8 12.6 7.76 5.96
50 60 54.6 34.0 33.5 18.6 -
80 30.0 19.5 18.4 11.8 9.66
100 18.5 12.0 11.6 7.94 6.31
120 12.9 8.51 7.71 5.50 4.39
140 9.44 6.31 5.43 3.24 3.20
160 6.92 4.49 4.03 3.02 2.50
180 5.27 3.55 3.16 2.19 1.97
35 60 9.16 7.16 7.5 5.13 -
80 6.31 4.75 4.95 3.22 3.35
100 4.52 3.29 3.35 2.74 2.41
120 3.35 2.51 2.51 2.07 1.78
140 2.66 1.99 1.88 1.57 1.41
160 2.20 1.59 1.56 1.27 1.12
180 1.68 1.33 1.26 1.00 0.944
20 60 2.90 2.82 2.63 2.26 -
80 2.24 2.04 1.94 1.70 1.72
100 1.78 1.45 1.45 1.33 1.27
120 1.35 1.14 1.12 1.06 1.00
140 1.12 0.922 0.908 0.832 0.804
160 1.00 0.759 0.794 0.692 0.689
180 0.773 0.673 0.668 0.562 0.576
1
DS = Percent Dry Substance;
2
DE = Dextrose Equivalent
Source: Erickson et al., 1966.
366 Appendices
6 . 9 . Dairy, Fis h and Me at Produc t s
Product T
( C) (-) (Pa s
n
) (Pa) (s
-1
)
Cream, 10% fat 40 1.0 .00148 - -
60 1.0 .00107 - -
80 1.0 .00083 - -
Cream, 20% fat 40 1.0 .00238 - -
60 1.0 .00171 - -
80 1.0 .00129 - -
Cream, 30% fat 40 1.0 .00395 - -
60 1.0 .00289 - -
80 1.0 .00220 - -
Cream, 40% fat 40 1.0 .00690 - -
60 1.0 .00510 - -
80 1.0 .00395 - -
Minced fish paste 3-6 .91 8.55 1600.0 67-238
Raw, Meat Batters
15
1
13
2
68.8
3
15 .156 639.3 1.53 300-500
18.7 12.9 65.9 15 .104 858.0 .28 300-500
22.5 12.1 63.2 15 .209 429.5 0 300-500
30.0 10.4 57.5 15 .341 160.2 27.8 300-500
33.8 9.5 54.5 15 .390 103.3 17.9 300-500
45.0 6.9 45.9 15 .723 14.0 2.3 300-500
45.0 6.9 45.9 15 .685 17.9 27.6 300-500
67.3 28.9 1.8 15 .205 306.8 0 300-500
Milk, homogenized 20 1.0 .002000 - -
30 1.0 .001500 - -
40 1.0 .001100 - -
50 1.0 .000950 - -
60 1.0 .000775 - -
70 1.0 .00070 - -
80 1.0 .00060 - -
Milk, raw 0 1.0 .00344 - -
5 1.0 .00305 - -
10 1.0 .00264 - -
20 1.0 .00199 - -
25 1.0 .00170 - -
30 1.0 .00149 - -
35 1.0 .00134 - -
40 1.0 .00123 - -
1
%Fat;
2
%Protein;
3
%MC
Source: Steffe et al., 1986.
n K
o

367
6 . 1 0 . Oils and Mis c e llane ous Produc t s
Product % Total T
Solids ( C) (-) (Pa s
n
) (Pa) (s
-1
)
Chocolate, melted - 46.1 .574 .57 1.16 -
Honey - -
Buckwheat 18.6 24.8 1.0 3.86 - -
Golden Rod 19.4 24.3 1.0 2.93 - -
Sage 18.6 25.9 1.0 8.88 - -
Sweet Clover 17.0 24.7 1.0 7.20 - -
White Clover 18.2 25.2 1.0 4.80 - -
Mayonnaise - 25 .55 6.4 - 30-1300
- 25 .60 4.2 - 40-1100
Mustard - 25 .39 18.5 - 30-1300
- 25 .34 27.0 - 40-1100
Oils
Castor - 10 1.0 2.42 - -
- 30 1.0 .451 - -
- 40 1.0 .231 - -
- 100 1.0 .0169 - -
Corn - 38 1.0 .0317 - -
- 25 1.0 .0565 - -
Cottonseed - 20 1.0 .0704 - -
- 38 1.0 .0306 - -
Linseed - 50 1.0 .0176 - -
- 90 1.0 .0071 - -
Olive - 10 1.0 .1380 - -
- 40 1.0 .0363 - -
- 70 1.0 .0124 - -
Peanut - 25.5 1.0 .0656 - -
- 38.0 1.0 .0251 - -
- 21.1 1.0 .0647 - .32-64
- 37.8 1.0 .0387 - .32-64
- 54.4 1.0 .0268 - .32-64
Rapeseed - 0.0 1.0 2.530 - -
- 20.0 1.0 .163 - -
- 30.0 1.0 .096 - -
Safflower - 38.0 1.0 .0286 - -
- 25.0 1.0 .0522 - -
Sesame - 38.0 1.0 .0324 - -
Soybean - 30.0 1.0 .0406 - -
- 50.0 1.0 .0206 - -
- 90.0 1.0 .0078 - -
Sunflower - 38.0 1.0 .0311 - -
n K
o

368 Appendices
6 . 1 1 . Fruit and Ve ge t able Produc t s
Product % Total T
Solids ( C) (-) (Pa s
n
) (s
-1
)
Apple
Pulp - 25.0 .084 65.03 -
Sauce 11.6 27 .28 12.7 160-340
11.0 30 .30 11.6 5-50
11.0 82.2 .30 9.0 5-50
Sauce 10.5 26 .45 7.32 .78-1260
9.6 26 .45 5.63 .78-1260
8.5 26 .44 4.18 .78-1260
Apricots
Puree 17.7 26.6 .29 5.4 -
23.4 26.6 .35 11.2 -
41.4 26.6 .35 54.0 -
44.3 26.6 .37 56.0 .5-80
51.4 26.6 .36 108.0 .5-80
55.2 26.6 .34 152.0 .5-80
59.3 26.6 .32 300.0 .5-80
Reliable,
Conc., green 27.0 4.4 .25 170.0 3.3-137
27.0 25 .22 141.0 3.3-137
" ripe 24.1 4.4 .25 67.0 3.3-137
24.1 25 .22 54.0 3.3-137
" ripened 25.6 4.4 .24 85.0 3.3-137
25.6 25 .26 71.0 3.3-137
" overripe 26.0 4.4 .27 90.0 3.3-137
26.0 25 .30 67.0 3.3-137
Banana
Puree A - 23.8 .458 6.5 -
Puree B - 23.8 .333 10.7 -
Puree - 22 .283 107.3 28-200
(17.7 Brix)
Blueberry
Pie Filling - 20 .426 6.08 3.3-530
Carrot
Puree - 25 .228 24.16 -
Green Bean
Puree - 25 .246 16.91 -
Guava
Puree - 23.4 .494 39.98 15-400
(10.3 Brix)
Mango
Puree - 24.2 .334 20.58 15-1000
(9.3 Brix)
n K

369
Orange Juice
Concentrate
Hamlin, early - 25 .585 4.121 0-500
42-5 Brix - 15 .602 5.973 0-500
- 0 .676 9.157 0-500
- -10 .705 14.25 0-500
Hamlin, late - 25 .725 1.930 0.500
41.1 Brix - 15 .560 8.118 0.500
- 0 .620 1.754 0-500
- -10 .708 13.87 0-500
Pineapple, - 25 .643 2.613 0-500
early
40.3 Brix - 15 .587 5.887 0-500
- 0 .681 8.938 0-500
- -10 .713 12.18 0-500
Pineapple, late - 25 .532 8.564 0-500
41.8 Brix - 15 .538 13.43 0-500
- 0 .636 18.58 0-500
- -10 .629 36.41 0-500
Valencia, early - 25 .583 5.059 0-500
43.0 Brix - 15 .609 6.714 0-500
- -10 .619 27.16 0-500
Valencia, late - 25 .538 8.417 0-500
- 15 .568 11.80 0-500
41.9 Brix - 0 .644 18.75 0-500
- -10 .628 41.41 0-500
Naval
65.1 Brix - -18.5 .71 29.2 -
- -14.1 .76 14.6 -
- -9.3 .74 10.8 -
- -5.0 .72 7.9 -
- -0.7 .71 5.9 -
- 10.1 .73 2.7 -
- 19.9 .72 1.6 -
- 29.5 .74 .9 -
Papaya
Puree - 26.0 .528 9.09 20-450
(7.3 Brix)
Peach
Pie Filling - 20.0 .46 20.22 1.-140
Puree 10.9 26.6 .44 .94 -
17.0 26.6 .55 1.38 -
21.9 26.6 .55 2.11 -
26.0 26.6 .40 13.4 80-1000
29.6 26.6 .40 18.0 80-1000
37.5 26.6 .38 44.0 -
40.1 26.6 .35 58.5 2-300
49.8 26.6 .34 85.5 2-300
58.4 26.6 .34 440.0 -
Puree 11.7 30.0 .28 7.2 5-50
370 Appendices
11.7 82.2 .27 5.8 5-50
10.0 27.0 .34 4.5 160-3200
Pear
Puree 15.2 26.6 .35 4.25 -
24.3 26.6 .39 5.75 -
33.4 26.6 .38 38.5 80-1000
37.6 26.6 .38 49.7 -
39.5 26.6 .38 64.8 2-300
47.6 26.6 .33 120.0 .5-1000
49.3 26.6 .34 170.0 -
51.3 26.6 .34 205.0 -
45.8 32.2 .479 35.5 -
45.8 48.8 .477 26.0 -
45.8 65.5 .484 20.0 -
45.8 82.2 .481 16.0 -
14.0 30.0 .35 5.6 5-50
14.0 82.2 .35 4.6 5-50
Plum
Puree 14.0 30.0 .34 2.2 5-50
14.0 82.2 .34 2.0 5-50
Squash
Puree A - 25 .149 20.65 -
Puree B - 25 .281 11.42 -
Tomato
Juice Conc. 5.8 32.2 .59 .223 500-800
5.8 38.8 .54 .27 500-800
5.8 65.5 .47 .37 500-800
12.8 32.2 .43 2.0 500-800
12.8 48.8 .43 2.28 500-800
12.8 65.5 .34 2.28 500-800
12.8 82.2 .35 2.12 500-800
16.0 32.2 .45 3.16 500-800
16.0 48.8 .45 2.77 500-800
16.0 65.5 .40 3.18 500-800
16.0 82.2 .38 3.27 500-800
25.0 32.2 .41 12.9 500-800
25.0 48.8 .42 10.5 500-800
25.0 65.5 .43 8.0 500-800
25.0 82.2 .43 6.1 500-800
30.0 32.2 .40 18.7 500-800
30.0 48.8 .42 15.1 500-800
30.0 65.5 .43 11.7 500-800
30.0 82.2 .45 7.9 500-800
371
6 . 1 2 . Polyme r Me lt s
Polymer T
( C) (-) (Pa s
n
) (Pa s
n
) (s
-1
)
High impact 170 0.20 7.58 x 10
4
2.1 x 10
5
100-7000
polystyrene 190 0.21 4.57 x 10
4
1.48 x 10
5
100-7000
210 0.19 3.61 x 10
4
1.05 x 10
5
100-7000
Polystyrene 190 0.22 4.47 x 10
4
1.4 x 10
4
100-4500
210 0.25 2.38 x 10
4
9.2 x 10
3
100-4000
225 0.28 1.56 x 10
4
6.6 x 10
3
100-5000
Styrene Acrylonitril 190 0.21 9.0 x 10
4
2.2 x 10
4
100-9000
220 0.27 3.22 x 10
4
9.0 x 10
3
100-8000
250 0.35 1.11 x 10
4
4.2 x 10
3
100-8000
Thermoplastic 200 0.27 2.75 x 10
4
3.6 x 10
4
100-5000
olefin 220 0.30 1.83 x 10
4
2.15 x 10
4
100-4000
240 0.28 1.99 x 10
4
1.35 x 10
4
100-3000
Acrylonitrile buta- 170 0.25 1.19 x 10
5
7.95 x 10
4
100-5500
diene styrene 190 0.25 6.29 x 10
4
4.4 x 10
4
100-6000
210 0.25 3.93 x 10
4
2.6 x 10
4
100-7000
Polypropylene 180 0.37 6.79 x 10
3
4.21 x 10
3
100-4000
190 0.41 4.89 x 10
3
3.02 x 10
3
100-3500
200 0.41 4.35 x 10
3
205 x 10
3
100-4000
Ethylene ethyl 170 0.38 1.21 x 10
4
5.4 x 10
3
100-6000
acrylate 190 0.43 6.91 x 10
3
3.5 x 10
3
100-4000
210 0.48 3.77 x 10
3
2.3 x 10
3
100-6000
High density 180 0.56 6.19 x 10
3
2.1 x 10
3
100-1000
polyethylene 200 0.59 4.68 x 10
3
1.52 x 10
3
100-1000
220 0.61 3.73 x 10
3
1.17 x 10
3
100-1000
Low density 160 0.41 9.36 x 10
3
6.3 x 10
3
100-4000
polyethylene 180 0.46 5.21 x 10
3
3.2 x 10
3
100-6500
200 0.47 4.31 x 10
3
1.7 x 10
3
100-6000
Nylon 220 0.63 2.62 x 10
3
1.6 x 10
3
100-2500
230 0.66 1.95 x 10
3
1.3 x 10
3
100-2000
235 0.66 1.81 x 10
3
1.1 x 10
3
100-2300
Polymethylmetha- 220 0.19 8.83 x 10
4
1.3 x 10
3
100-6000
crylate 240 0.25 4.27 x 10
4
6.0 x 10
3
100-6000
260 0.27 2.62 x 10
4
2.9 x 10
3
100-7000
Polycarbonate 280 0.64 8.39 x 10
3
1.54 x 10
3
100-1000
300 0.67 4.31 x 10
3
8.0 x 10
2
100-1000
320 0.80 1.08 x 10
3
4.2 x 10
2
100-1000
Source: Tadmor, Z. and C.G. Gogos. 1979.
n K
o

372 Appendices
6 . 1 3 . Cos me t ic and Toile t ry Produc t s
Approximate parameters, at 25 C, for the apparent viscosity func-
tion based on the power law form of the model:
(Curve #): Product
( - ) (Pa s ) (1/s)
1: Shine On Lasting Color Nail Enamel
*
0.86 754 3.0 - 99
2: Cover Girl Marathon Mascara
*
0.24 191 0.1 - 75
3: Muppets Toothpaste, Bubble Gum Gel 0.28 117 0.1 - 90
4: Johnsons Baby Sunblock Lotion
*
0.28 75 0.1 - 50
5: Ponds Cold Cream
*
0.45 25 0.3 - 90
6: Oil of Olay Beauty Fluid 0.22 26 0.1 - 90
7: Head & Shoulders Shampoo 1.0 4 0.1 - 6
8: Jergens Lotion Enriched Liquid Soap 1.0 4 0.1 - 6
*
Average values, original data indicate some time-dependent behavior.
0.1 0.2 0.5 1 2 5 10 20 50 100
0.1
1
10
100
1,000
10,000
Shear Rate, 1/s
A
p
p
a
r
e
n
t

V
i
s
c
o
s
i
t
y
,

P
a

s
1
2
3
4
5
6
7
8
Approximate Temperature = 25 C
Data from Laba, 1993a
= K(
)
n 1
n K

373
Approximate parameters, at 25 C, for the apparent viscosity func-
tion based on the power law form of the model:
(Curve #): Product
( - ) (Pa s ) (1/s)
1: Colgate Toothpaste
*
0.22 121.7 3.0 - 700
2: Suave Balsam & Protein Shampoo 0.95 11.1 20.0 - 700
3: Vaseline Intensive Care Lotion 0.34 13.7 0.09 - 750
4: Moisturizing Sunblock Lotion
*
0.32 9.1 0.05 - 700
5: Baby Magic Baby Lotion
*
0.38 6.7 0.02 - 700
6: BAN Antiperspirant/Deodorant Roll-on 0.52 3.3 1.5 - 750
7: Dry Idea Dry Roll-on Deodorant 0.13 9.5 12.0 750
*
Average values, original data indicate some time-dependent behavior.
0.1 0.3 1 3 10 30 100 300 1,000
0.01
0.1
1
10
100
1,000
Shear Rate, 1/s
A
p
p
a
r
e
n
t

V
i
s
c
o
s
i
t
y
,

P
a

s
1
2
3
4
5
6
7
Approximate Temperature = 25 C
Data from Laba, 1993a
= K(
)
n 1
n K

374 Appendices
6 . 1 4 . Ene rgy of Ac t ivat ion for Flow for Fluid Foods
Fluid food Concentration at 50 C
1
(-) (kcal/g (mPa s
n
)
mol)
Depectinized apple 75 Brix 1.0 14.2 150.0
juice
a
50 Brix 1.0 8.4 4.0
30 Brix 1.0 6.3 1.6
15 Brix 1.0 5.3 0.7
Cloudy apple juice
a
40 Brix 1.0 5.8 4.9
30 Brix 1.0 5.1 2.0
Concord grape juice
a
50 Brix 1.0 6.9 15.0
30 Brix 1.0 6.2 1.8
Cloudy apple juice
a
65.5 Brix 0.65 9.1 258.5
50.0 Brix 0.85 6.1 25.0
Apple sauce
a
11.0 Brix 0.30 1.2 730.0
Peach puree
a
11.7 Brix 0.30 1.7 190.0
Pear puree
a
16.0 Brix 0.30 1.9 375.0
Filtered orange juice
b
18.0 Brix 1.0 5.8 1.5
10 Brix 1.0 5.3 0.8
Whole egg
c
75 Wt.% water 1.0 5.9 3.7
Stabilized egg white
c
88 Wt.% water 1.0 5.9 1.9
Plain yolk
c
55 Wt.% water 1.0 6.4 48.2
Salted yolk
c
10 Wt.% salt 1.0 8.3 153.0
Sugared yolk
c
10 Wt.% sugar 1.0 6.9 32.1
1
= apparent viscosity at 100 1/s.
Source:
a
Rao (1986);
b
Saravacos (1970);
c
Scalzo et al. (1970)
n E
a

375
6 . 1 5 . Ext e ns ional Vis c os it ie s of Ne wt onian Fluids
Trouton Ratio
Liquid (Pa s) (s
-1
) (-)
Silicone Oil 102.5 0.5-10 2.4-3.6
Polybutene 24 1.6-5 2.7-3.3
Polybutene 23 2-10 2.6-3.8
Corn Syrup 25 - 4
Oil 750 0.08-0.14 3.1-3.5
Maltose Sryup 104 2-5 5
Glycerol-water 0.357 4000-9000 1.7-3.4
Glycerol-water 0.4-1.7 40-180 2.7-3.3
Glycerol-water 0.12-0.25 200-4000 2.4-3.9
Viscasil 30 1-30 2.4-3.9
Source: Gupta and Sridhar, 1987.
h
376 Appendices
6 . 1 6 . Ext e ns ional Vis c os it ie s of Non-Ne wt onian Fluids
Solute Solvent Conc. Trouton
(%) (s
-1
) ( ) Ratio
(-)
PAA Glycerol 1.5 0.1-10 2-8E3 -
PAA Water 1.0 8-230 2.4-250 -
PAA Glycerol 1.0 1-10 - 0.1-6.0
PAA Glycerol 0.175-0.5 0.17-1.4 - 20-1000
PAA Glycerol 0.5 0.1-0.2 1-5E4 -
PAA Maltose 0.1 0.5-5.0 - 70
PAA Water 0.1 100-750 - 3000
PAA Water 0.1 40-80 - 250-400
PAA Water 0.01-0.5 33-19000 - 500-29000
PAA Corn Syrup 0.05 - - 300
PAA Glycerol 0.01-0.03 4-17 30-100 -
PAA Glycerol 0.005-0.05 50-1400 1-100 -
PAA Water 0.0051 50-800 0.5-25 -
PEO Sucrose 3.0 1-20 20-500 -
PEO Glycerol 3.0 0.4-2.0 0.4-7.0E6 -
PEO Water 0.1 100-750 8-11.4 -
PEO Water 0.1 40-80 - 1500-2400
PIB Decalin 6.4-11.6 2-100 0.5-7 -
PIB Kerosene 3.0-4.0 3-100 - 1.2-40
PIB Polybutene 0.18 0.4-3.0 - 3-30
XG Glycerol 0.03-0.05 20-40 1-10 -
XG Glycerol 0.005-0.01 50-1400 1-8 -
HPC Water 2.0 0.1-1.0 - 2.0-5.0
HPC Acetic Acid 40.0 0.01-10.0 - 10
HC Jet Fuel 0.4-1.0 3-40 - 80-2000
PU DMF 18-30 - 8-30 -
PAA, polyacrylamide; PEO, polyethyleneoxide; PIB, poly-isobutylene; XG,
xanthan gum; HPC, hydroxy-propylcellulose; HC, hydrocarbon; PU,
polyurethane; DMF, dimethylformamide.
Source: Gupta and Sridhar, 1987.
h

E
Pa s
3
7
7
6
.
1
7
.

F
a
n
n
i
n
g

F
r
i
c
t
i
o
n

F
a
c
t
o
r
s
:

B
i
n
g
h
a
m

P
l
a
s
t
i
c
s

100 200 300 500 1,000 2,000 3,000 5,000 10,000
0.001
0.002
0.005
0.01
0.02
0.05
0.1
0.2
0.5
1
F
a
n
n
i
n
g

F
r
i
c
t
i
o
n

F
a
c
t
o
r
100 1,000 5,000 10,000 25,000 50,000 100,000 200,000
N
He
N
Re,B
Bingham Plastics
in
Laminar Flow
378 Appendices
6 . 1 8 . Fanning Fric t ion Fac t ors : Powe r Law Fluids

2
,
0
0
0
4
,
0
0
0
6
,
0
0
0
8
,
0
0
0
1
0
,
0
0
0
0
.
0
0
2
0
.
0
0
4
0
.
0
0
6
0
.
0
0
8
0
.
0
1
0
.
0
1
2
0
.
0
1
4
F a n n i n g F r i c t i o n F a c t o r
N
R
e
,
P
L
n

=

1
.
2
n

=

1
.
0
n

=

0
.
8
n

=

0
.
6
n

=

0
.
4
n

=

0
.
3
n

=

0
.
2
P
o
w
e
r

L
a
w

F
l
u
i
d
s

i
n

T
u
r
b
u
l
e
n
t

F
l
o
w
379
6 . 1 9 . Cre e p (Burge rs Mode l) of Salad Dre s s ing
Creamy-style Salad
Dressing at 2.8 C (cm
2
dyne
-1
) (cm
2
dyne
-1
) (s) (P)
Regular, bottled
1
0.000704 0.00100 16.2 79800
Reduced Calorie, bottled
1
0.00182 0.00192 43.2 148800
Regular, bottled
1
0.000490 0.000481 18.5 587400
Reduced Calorie, bottled
1
0.000870 0.00370 26.1 672000
Regular, dry mix
2
0.00161 0.000980 8.92 19800
Regular, dry mix
2
0.00164 0.00208 2.42 3600
1
Constant Applied Shear Stress: = 55.2 dyne cm
2
2
Constant Applied Shear Stress: = 22.8 dyne cm
2
Source: Paredes et al., 1989
The above constants were determined for the Bugers model (Eq.
[5.22]) written in terms of the shear creep compliance function:
Results are typical of creamy-style salad dressing purchased in a
bottle or made from dry mix. Creep testing may be useful in con-
sidering the stability of salad dressing.
J
o
J
1

ret

o
o
J = f (t ) = J
0
+ J
1
1 exp
ret
+
t
0
380 Appendices
6 . 2 0 . Os c illat ory Dat a for But t e r
Approximate mathematical parameters describing the above curves:
and
T
( C) (Pa s
1+b
) ( - ) (Pa s
d
) ( - )
5 1,073,300 -1.0 4,395,900 0.18
10 350,100 -1.0 1,339,500 0.16
15 140,400 -1.0 561,400 0.16
20 35,000 -1.0 151,700 0.17
0.001 0.01 0.1 1 10 100
Frequency, rad/s
T( C)
5
10
15
20
5
10
15
20
storage modulus
dynamic viscosity
10
9
10
8
10
7
10
6
10
5
10
4
10
3
Butter
Data from Rohm and Weidinger, 1993
S
t
o
r
a
g
e

M
o
d
u
l
u
s

(
P
a
)

o
r

D
y
n
a
m
i
c

V
i
s
c
o
s
i
t
y

(
P
a

s
)
= a()
b
G = c()
d
a b c d
381
6 . 2 1 . Os c illat ory Dat a Iot a-Carrage e nan Ge l
This figure illustrates the effect of temperature in transforming a
gel into a solution. Iota-carrageenan (1%) was prepared by dis-
persing the gumincold water andheating for 15 min. The frequency
sweep was conducted at a strain amplitude of 8% using a coaxial
cylinder geometry ( ). A structured gel is observed at 25 C
where the storage modulus is only slightly dependent on frequency.
The storage modulus and the loss modulus are almost the same at
35 C. Behavior typical of a macromolecular solution (the storage
modulus is less than the loss modulus) is observed at 39 C.
0.01 0.03 0.1 0.3 1 3 10 30 100
0.001
0.01
0.1
1
10
100
Frequency, rad/s
S
t
o
r
a
g
e

o
r

L
o
s
s

M
o
d
u
l
u
s
,

P
a
Iota-Carragenan-Water (1%)
Data from Cuvelier et al., 1990
25 C
Storage Modulus
Loss Modulus
35 C
39 C
R
b
/R
c
= 0.96
382 Appendices
6 . 2 2 . St orage and Los s Moduli of Fluid Foods

Product
( C) (Pa s
b
) ( - ) (Pa s
d
) ( - ) (rad/s)
Mustard
1
fine 25 800.6 0.120 149.5 0.177 0.01 - 100
standard 25 897.5 0.130 185.9 0.171 0.01 - 100
course 25 1160.0 0.135 265.0 0.139 0.01 - 100
Tomato Paste
2
40 8434.0 0.117 2101.0 0.153 5.0 - 55.0
Blueberry Pie 40 278.3 0.17 64.2 0.26 1 - 100
Filling
3
85 237.9 0.13 45.5 0.26 1-100
Cookie Dough
4
RT
*
4.66E7 1.49 1.20E7 1.30 0.01-100
Cracker Dough
4
RT
*
6.50E6 1.24 2.10E6 1.12 0.01-100
Mozzarella
Cheese
5
natural 70 22,700 0.17 10,300 0.19 0.01 - 100
with 1% Ca 70 59,200 0.20 19,800 0.14 0.01 - 100
caseinate
with 2% Ca 70 15,900 0.21 19,800 0.16 0.01 - 100
caseinate
*
RT = RoomTemperature
1
from Aguilar et al., 1991.
2
from Rao and Cooley, 1992.
3
from Steffe et al., 1989.
4
from Menjivar and Faridi, 1994.
5
from Nolan et al., 1989.
G = a
b
G = c
d
T a b c d
383
0.01 0.03 0.1 0.3 1 3 10 30 100
Frequency, rad/s
S
t
o
r
a
g
e

M
o
d
u
l
u
s
,

P
a
Mustard, Standard
25 C
Tomato Paste
40 C
Blueberry Pie Filling
40 C
Mozarella Cheese, Natural
70 C
Cookie Dough
Room Temperature
Cracker Dough
Room Temperature
1E+13
1E+11
1E+09
1E+07
1E+05
1E+03
1E+01
384 Appendices
0.01 0.03 0.1 0.3 1 3 10 30 100
Frequency, rad/s
L
o
s
s

M
o
d
u
l
u
s
,

P
a
Mustard, Standard
25 C
Tomato Paste
40 C
Blueberry Pie Filling
40 C
Mozarella Cheese, Natural
70 C
Cookie Dough
Room Temperature
Cracker Dough
Room Temperature
1E+12
1E+10
1E+08
1E+06
1E+04
1E+02
1E+00
Nome nc lat ure
shi ft factor, di mensi onl ess
area, m
2
i ni ti al sampl e area, m
2
hei ght of mi xer bl ade, m
bul k compressi on creep compl i ance, Pa
-1
bul k compressi on compl ex compl i ance, Pa
-1
bul k compressi on storage compl i ance, Pa
-1
bul k compressi on l oss compl i ance, Pa
-1
, di mensi onl ess
speci fi c heat, J kg
-1
K
-1
mass concentrati on, g/dl or g/100 ml
i mpel l er bl ade or vane di ameter, m
equi val ent di ameter, m
di ameter, m
tensi l e creep compl i ance, Pa
-1
tensi l e compl ex compl i ance, Pa
-1
tensi l e storage compl i ance, Pa
-1
tensi l e l oss compl i ance, Pa
-1
di ameter of extrudate, m
Youngs modul us or modul us of el asti ci ty, Pa
tensi l e rel axati on modul us, Pa
energy of acti vati on for fl ow, cal /g mol e
tensi l e compl ex modul us, Pa
tensi l e storage modul us, Pa
tensi l e l oss modul us, Pa
Fanni ng fri cti on factor, di mensi onl ess
force, N
accel erati on due to gravi ty, 9.81 m s
-2
shear modul us, Pa
shear rel axati on modul us, Pa
shear compl ex modul us, Pa
a
T
A
A
o
b
B
B
*
B
B
c
o
/
w
c
p
C
d
d
e
D
D
D
*
D
D
D
e
E
E
E
a
E
*
E
E
f
F
g
G
G
G
*
386 Nomenclature
shear storage modul us, Pa
shear l oss modul us, Pa
hei ght, m
asympoti c or resi dual thi ckness, m
effecti ve hei ght, m
i ni ti al sampl e hei ght, m
shear creep compl i ance, Pa
-1
i nstantaneous compl i ance, Pa
-1
retarded compl i ance, Pa
-1
shear compl ex compl i ance, Pa
-1
shear storage compl i ance, Pa
-1
shear l oss compl i ance, Pa
-1
fal l i ng bal l vi scometer constant [ ], m
2
s
-2
gl ass capi l l ary vi scometer constant [ ], m
2
s
-2
thermal conducti vi ty, W m
-1
K
-1
fri cti on l oss coeffi ci ent, di mensi onl ess
mi xer vi scometer constant, rad
-1
bul k modul us, Pa
consi stency coeffi ci ent, Pa s
n
bul k compressi on rel axati on modul us, Pa
bul k compressi on compl ex modul us, Pa
bul k compressi on storage modul us, Pa
bul k compressi on l oss modul us, Pa
extensi onal consi stency coeffi ci ent, Pa s
m
temperature dependent constant
temperature and concentrati on dependent constant
shear rate, temperature and conc. dependent constant
vari abl e l ength, m
l ength, m
i ni ti al l ength or l ength of undeformed sampl e, m
extensi onal fl ow behavi or i ndex, di mensi onl ess
torque, N m
G
G
h
h
a
h
o
h
o
J
J
0
J
1
J
*
J
J
2R
2
g/(9L) k
ghR
4
/(8LV) k
k
k
f
k
K
K
K
K
*
K
K
K
E
K
T
K
T, C
K
, T, C
l
L
L
o
m
M
387
torque to overcome yi el d stress, N m
end effect torque on a bob, N m
fl ow behavi or i ndex, di mensi onl ess
average fl ow behavi or i ndex, di mensi onl ess
fi rst normal stress di fference [ ], Pa
second normal stress di fference [ ], Pa
Boussi nesq number, di mensi onl ess
Deborah number, di mensi onl ess
Froude number [ ], di mensi onl ess
Hedstrom number [ ], di mensi onl ess
power number [ ], di mensi onl ess
Newtoni an fl ui d Reynol ds number [ ], di mensi onl ess
Bi ngham fl ui d Reynol ds number [ ], di mensi onl ess
i mpel l er Reynol ds number [ ], di mensi onl ess
Power l aw fl ui d Reynol ds number, di mensi onl ess
Trouton rati o, di mensi onl ess
Weber number [ ], di mensi onl ess
Wei ssenberg number [ ], di mensi onl ess
power i nput to a mi xer [ ], N m s
-1
pressure, Pa
stati c pressure (absol ute) i n the undi sturbed fl ow, Pa
exi t pressure (absol ute) of a sl i t vi scometer, Pa
exi t pressure of a sl i t vi scometer, Pa
l i qui d vapor pressure (absol ute), Pa
vol umetri c fl ow rate i n a pi pe, m
3
s
-1
measured vol umetri c fl ow rate, m
3
s
-1
vol umetri c fl ow rate wi thout sl i p, m
3
s
-1
radi al coordi nate, m
uni versal gas constant, 1.987 cal /(g-mol e K)
radi us, m
M
o
M
e
n
n
N
1

11

22
N
2

22

33
N
Bo
N
De
2
d/g N
Fr
D
2
o
/(
pl
)
2
N
He
P/(d
5
3
) N
Po
N
Re
Du/
N
Re, B
Du/
pl
d
2
/ N
Re, I
N
Re, PL
N
Tr
N
2
d
3
/
st
N
We
N
Wi

1
/
P M
P
P
atm
P
ex
P
ex
P
vap
Q
Q
m
Q
ws
r
R
R
388 Nomenclature
bob radi us, m
radi us of barrel i n convergi ng di e, m
cup radi us, m
radi us of undeformed cyl i nder, m
i ni ti al radi us of sampl e, m
cri ti cal radi us, m
radi us of shaft, m
radi us of truncated porti on of cone, m
ti me, s
temperature, K or C
reference temperature, K or C
vel oci ty, m s
-1
turbul ent vel oci ty [ ], di mensi onl ess
fri cti on vel oci ty [ ], m s
-1
vel oci ty i n the and di recti ons, m s
-1
vel oci ty i n the , and di recti ons, m s
-1
effecti ve sl i p vel oci ty, m s
-1
termi nal vel oci ty, m s
-1
vol umetri c average vel oci ty [ ], m s
-1
vol ume of mi xi ng vessel , m
3
vol ume of a gl ass capi l l ary vi scometer bul b, m
3
wi dth of sl i t or bl ade, m
work output per uni t mass, J kg
-1
constant force, N
Cartesi an coordi nates, m
pi pe l ength requi red for ful l y devel oped fl ow, m
di stance from pi pe wal l i nto fl ui d [ ], m
di stance from the tube wal l [ ], di mensi onl ess
hei ght above a reference pl ane, m
di stance between two poi nts (mi xer vi scometer), m
axi al coordi nate, m
R
b
R
b
R
c
R
o
R
o
R
o
R
s
R
T
t
T
T
r

u
u
+
u/u
*
w
/ = u
f /2 u
*
u
1
, u
2
, u
3
x
1
, x
2
x
3
u
r
, u
z
, u
r z
u
s
u
t
Q/(R
2
) u
V
V
w
W
W
x
1
, x
2
, x
3
X
E
y R r
y
+
u
*
y/
z
z
z
389
, di mensi onl ess
, di mensi onl ess
ki neti c energy correcti on factor, di mensi onl ess
effecti ve sl i p coeffi ci ent [ ], m Pa
-1
s
-1
corrected sl i p coeffi ci ent [ ], m
2
Pa
-1
s
-1
angl e of shear, rad
shear strai n, di mensi onl ess
ampl i tude of strai n functi on [ ], di mensi onl ess
constant shear strai n, di mensi onl ess
strai n at the ri m of a paral l el pl ate, di mensi onl ess
normal i zed shear strai n ( ), di mensi onl ess
shear rate, s
-1
average shear rate, s
-1
shear rate at the bob, s
-1
shear rate at the ri m of a paral l el pl ate, s
-1
shear rate at the wal l , s
-1
apparent wal l shear rate [ ], s
-1
phase shi ft or phase angl e, rad
change i n hei ght, m
l i near di spl acement, Pa
pressure drop, Pa
entrance pressure l oss, Pa
entrance pressure l oss due to extensi onal fl ow, Pa
entrance pressure l oss due to shear fl ow, Pa
radi al di spl acement, Pa
change i n ti me, s
Cauchy strai n [ ], di mensi onl ess
Hencky strai n [ ], di mensi onl ess
Hencky strai n rate, s
-1
constant Hencky strai n rate, s
-1
bi axi al extensi onal strai n rate, s
-1
R
c
/R
b
d/D
u
s
/
w
c
R
o
=
o
sin(t )
*
/
o
w
4Q/(R
3
)
h
L
P
P
en
P
en, E
P
en, S
R
t
c
(L L
o
)/L
o
h
ln(L/L
o
)
ho
B
390 Nomenclature
constant bi axi al extensi onal strai n rate, s
-1
average extensi onal strai n rate i n convergi ng fl ow, s
-1
average extensi onal strai n rate at the di e, s
-1
apparent vi scosi ty, Pa s
bi axi al extensi onal vi scosi ty, Pa s
bi axi al growth functi on, Pa s
equi l i bri um apparent vi scosi ty, Pa s
tensi l e extensi onal vi scosi ty, Pa s
tensi l e growth functi on, Pa s
i nherent vi scosi ty, dl g
-1
i ntri nsi c vi scosi ty, dl g
-1
l i mi ti ng vi scosi ty at zero shear rate, Pa s
pl anar extensi onal vi scosi ty, Pa s
reference apparent vi scosi ty, Pa s
reduced vi scosi ty, dl g
-1
rel ati ve vi scosi ty, di mensi onl ess
speci fi c vi scosi ty, di mensi onl ess
l i mi ti ng vi scosi ty at i nfi ni te shear rate, Pa s
compl ex vi scosi ty, Pa s
dynami c vi scosi ty, Pa s
out of phase component of , Pa s
angl e, rad
apex angl e of cone at bottom of bob, degrees
hal f angl e of convergi ng di e, degrees or rad
ci rcumferenti al coordi nate, rad
structural parameter, di mensi onl ess
ti me constant i n Bi rd-Lei der equati on, di mensi onl ess
equi l i bri um val ue of structural parameter, , di mensi onl ess
rel axati on ti me, s
retardati on ti me, s
Newtoni an vi scosi ty, Pa s
Bo
E, R
B
+
E
+
inh
int
red
rel
sp
rel
ret
391
corrected vi scosi ty, Pa s
pl asti c vi scosi ty of a Bi ngham fl ui d, Pa s
reference Newtoni an vi scosi ty, Pa s
ki nemati c vi scosi ty ( ), m
2
s
-1
Poi ssons rati o, di mensi onl ess
densi ty, kg m
-3
l i qui d densi ty, kg m
-3
sphere densi ty, kg m
-3
shear stress, Pa
average shear stress, Pa
shear stress at bob, Pa
bi axi al (radi al ) stretchi ng stress, Pa
shear stress at cup, Pa
shear stress on ends of vane, Pa
equi l i bri um shear stress, Pa
tensi l e stretchi ng stress, Pa
constant tensi l e stretchi ng stress, Pa
stress on pl ane perpendi cul ar to i i n the di recti on of j, Pa
yi el d stress, Pa
i ni ti al shear stress, Pa
constant shear stress, Pa
ampl i tude of the stress functi on [ ], Pa
shear stress at the ri m of a paral l el pl ate, Pa
surface tensi on, N m
-1
shear stress at the wal l of tube or sl i t, Pa
normal i zed shear stress ( ), di mensi onl ess
shear stress growth functi on, Pa
di e exi t effect i ntegral , di mensi onl ess
sweep angl e, rad
fi rst normal stress coeffi ci ent, Pa s
2
second normal stress coeffi ci ent, Pa s
2
pl
r
/
Eo
ij
o
=
o
sin(t )
st
*
/
o
21
+
2
392 Nomenclature
angul ar vel oci ty at r, rad s
-1
angul ar vel oci ty [2 (rpm)/60], rad s
-1

Bibliography
Agui l ar, C., S.S.H. Ri zva, J.F. Rami rez and A. I nda. 1991. Rheol ogi cal
behavi or of processed mustard. I : effect of mi l l i ng treatment. J.
Texture Stud. 22: 59-84.
Ak, M.M. and S. Gunasekaran. 1995. Measuri ngel ongati onal properti es
of Mozzarel l a cheese. J. Texture Stud. 26: 147-160.
Al tomare, R.E., M. Anel i ch and R. Rakos. 1992. An experi mental
i nvesti gati on of the rheol ogy of ri ce fl our dough wi th an extruder-
coupl ed sl i t di e rheometer. I n: Koki ni , J.L., C-T. Ho and M.V. Karwe
(edi tors). Food Extrusi on Sci ence and Technol ogy. Marcel Dekker,
I nc., New York. pg. 233-254.
Bagl ey, E.B. 1957. End correcti ons i n the capi l l ary fl ow of pol yethyl ene.
J. Appl . Phys. 28: 624-627.
Bagl ey, E.B. and D.D. Chri sti anson. 1986. Response of chemi cal l y
l eavened dough to uni axi al compressi on. I n: Fari di , H. and J.M.
Faubi on (edi tors). Fundamental s of Dough Rheol ogy. Amer. Assoc.
of Cereal Chemi sts, St. Paul , MN.
Bagl ey, E.B. and D.D. Chri sti anson. 1988. Behavi or of foods i n l arge
deformati ons. I n: Bl anshard, J.M.V. and J.R. Mi tchel (edi tors). Food
Structure - I ts Creati on and Eval uati on. Butterworths, London. pg
401-416.
Bagl ey, E.B., D.D. Chri sti anson and W.J. Wol f. 1985. Fri cti onal effects
i n compressi onal deformati on of gel ati n and starch gel s and com-
pari son of materi al response i n si mpl e shear, torsi on, and l ubri cated
compressi on. J. Rheol . 29: 103-108.
Bal aban, M., A.R. Carri l l o and J.L. Koki ni . 1988. A computeri zed
method to anal yze the creep behavi or of vi scoel asti c foods. J. Texture
Stud. 19: 171-183.
Banerjee, T.K., D. Manas and S.K. Das. 1994. Non-Newtoni an l i qui d
fl ow through gl obe and gate val ves. Can. J. Chem. Engn. 72: 207-211.
Barbosa Canovas, G.V. and M. Pel eg. 1983. Fl ow parameters of sel ected
commerci al semi -l i qui d food products. J. Texture Stud. 14: 213-234.
Barnes, H.A. 1992. The yi el d stress myth? revi si ted. I n: Mol denaers,
P. and R. Keuni ngs (edi tors). Theoreti cal and Appl i ed Rheol ogy,
Vol ume 2. Proc. XI th I nt. Congr. on Rheol ogy, Brussel s, Bel gi um,
Aug. 17-21, 1992. El sevi er Sci ence Publ i shers, New York. pg.
576-578.
Barnes, H.A., J.F. Hutton and K. Wal ters. 1989. An I ntroducti on to
Rheol ogy. El sevi er Sci ence Publ i shi ng Company I nc., New York.
Barnes, H.A. and K. Wal ters. 1985. The yi el d stress myth. Rheol . Acta
24: 323-326.
Bersted, B.H. 1993. Refi nement of the convergi ng fl ow method of
measuri ng extensi onal vi scosi ty i n pol ymers. Pol ymer Engn. Sci .
33(16): 1079-1083.
Bhattacharya, M. 1993. Sl i t rheometer studi es of wheat fl our dough. J.
Texture Stud. 24:391-409.
394 Bibliography
Bhattacharya, M., and M. Padmanabhan. 1992. On-l i ne rheol ogi cal
measurements of food dough duri ng extrusi on cooki ng. I n: Koki ni ,
J.L., C-T. Ho and M.V. Karwe (edi tors). Food Extrusi on Sci ence and
Technol ogy. Marcel Dekker, I nc., New York. pg. 213-232.
Bhattacharya, M., M. Padmanabhan and K. Seethamraju. 1994. Uni -
axi al extensi onal vi scosi ty duri ng extrusi on cooki ng from entrance
pressure drop method. J. Food Sci . 59: 221-226, 230.
Bi ndi ng, D.M. 1988. A approxi mate anal ysi s for contracti on and con-
vergi ng fl ows. J. Non-Newtoni an Fl ui d Mech. 27: 173-189.
Bi rd, R.B., R.C. Armstrong and O. Hassager. 1987. Dynami cs of Pol y-
meri c Li qui ds, Vol ume 1. John Wi l ey and Sons, New York.
Bi rd, R.B., W.E. Stewart and E.N. Li ghtfoot. 1960. Transport Phe-
nomena. John Wi l ey and Sons, I nc., New York.
Bi stany, K.L. and J.L. Koki ni . 1983. Compari son of steady shear rhe-
ol ogi cal properti es and smal l ampl i tude dynami c vi scoel asti c prop-
erti es of fl ui d food materi al s. J. Texture Stud. 14: 113-124.
Bl ock, H. and J.P. Kel l y. 1986. El ectro-rheol ogy. J. Phys. D: Appl . Phys.
21: 1661-1677.
Boger, D.V. 1982. Ci rcul ar entry fl ows of i nel asti c and vi scoel asti cfl ui ds.
I n: Mujumdar, A.S. and R.A. Mashel kar (edi tors). Advances i n
Transport Processes, Vol . I I . Hal sted Press. John Wi l ey and Sons,
New York. pg. 43-104.
Boger, D.V. and K. Wal ters. 1993. Rheol ogi cal Phenomena i n Focus.
El sevi er Sci ence Publ i shers, New York.
Bogue, D.C. and J.L. Whi te. 1970. Engi neeri ng Anal ysi s of Non-
Newtoni an Fl ui ds. AGARDograph No. 144. North Atl anti c Treaty
Organi zati on. Advi sor Group for Aerospace Research &
Devel opment. Etat Major Forces Ari ennes, Caserne Pri nce Bau-
doui n, Pl ace Dai l l y, Bruxel l es, 3, Bel gi um.
Bourne, M.C. 1968. Texture profi l e of ri peni ng pears. J. Food Sci . 33:
223-226.
Bourne, M.C. 1974. Texture changes i n ri peni ng peaches. J. Can I nst.
Food Sci . Technol . 7: 11-15.
Bourne, M.C. 1978. Texture profi l e anal ysi s. Food Technol . 32(7): 62-66,
72.
Bourne, M.C. 1982. Food Texture and Vi scosi ty: Concept and Mea-
surement. Academi c Press, New York.
Bourne, M.C., J.F. Kenny and J. Barnard. 1978. Computer-assi sted
readout of data from texture profi l e anal ysi s curves. J. Texture
Studi es 9: 481-494.
Bowen, R.L. 1986. Unravel i ng the mysteri es of shear-sensi ti ve mi xi ng
systems. Chem. Engn. June 9: 55-63.
Brennan, J.G. 1980. Food texture measurement. I n: Ki ng, R.D. (edi tor).
Devel opments i n Food Anal ysi s Techni ques - 2. Appl i ed Sci . Pub.
Ltd., London. pg. 1-78.
Bri ggs, J.L. 1995. Methods to Characteri ze Power Fl ui ds Usi ng a
Brookfi el d Vi scometer. MS Thesi s. Dept. of Agri cul tural Engi -
neeri ng. Mi chi gan State Uni versi ty.
395
Bri ggs, J.L. and J.F. Steffe. 1995. Ki neti c energy correcti on factor of a
Herschel -Bul kl ey fl ui d. J. Food Proc. Engn. 18: 115-118.
Bri ggs, J.L. and J.F. Steffe. 1996. Vane method to eval uate the yi el d
stress of frozen i ce cream. J. Dai ry Sci . 79: 527-531.
Butl er, F. and P. McNul ty. 1995. Ti me-dependent rheol ogi cal charac-
teri zati on of buttermi l k at 5 C. J. Food Proc. Engn. 25: 569-580.
Brodkey, R.S. and H.C. Hershey. 1988. Transport Phenomena.
McGraw-Hi l l Book Company. New York.
Cal l i ster, W.D. 1991. Materi al Sci ence and Engi neeri ng (Second Edi -
ti on). John Wi l ey & Sons, I nc., New York.
Campanel l a, O.H. and M. Pel eg. 1987a. Determi nati on of the yi el d stress
of semi -l i qui d foods from squeezi ng fl ow data. J. Food Sci . 52: 214-217.
Campanel l a, O.H. and M. Pel eg. 1987b. Anal ysi s of transi ent fl ow of
mayonnai se i n a coaxi al cyl i nder vi scometer. J. Rheol . 31: 439-452.
Campanel l a, O.H. and M. Pel eg. 1987c. Squeezi ng fl ow vi scosi metry of
peanut butter. J. Food Sci . 52: 180-184.
Campanel l a, O.H., L.M. Poppl ewel l , J.R. Rosenau and M. Pel eg. 1987.
El ongati onal vi scosi ty measurements of mel ti ng Ameri can process
cheese. J. Food Sci . 52: 1249-1251.
Campos, D.T., J.F. Steffe and R.Y. Ofol i . 1994. Stati sti cal method to
eval uate the cri ti cal Reynol ds number for pseudopl asti c fl ui ds i n
tubes. J. Food Engn. 23: 21-32.
Carreau, P.J. 1968. Ph.D. Di ssertati on. Dept. of Chemi cal Engi neeri ng,
Uni versi ty of Wi sconsi n, Madi son.
Casi raghi , E.M., E.B. Bagl ey and D.D. Chri sti anson. 1985. Behavi or of
Mozzarel l a, Cheddar and process cheese spread i n l ubri cated and
bonded uni axi al compressi on. J. Texture Stud. 16: 281-301.
Casson, N. 1959. A fl ow equati on for pi gmented-oi l suspensi on of the
pri nti ng i nk type. I n: Mi l l , C.C. (edi tor). Rheol ogy of Di spersed
Systems. Pergamon Press, New York. pg. 84-104.
Castal do, D., L. Pal mi eri , A. Lo Voi and P. Costabi l e. 1990. Fl ow
properti es of Babaco (Cari ca Pentagona) purees and concentrates. J.
Castel l -Perez, M.E., R.G. Morei ra and J.F. Steffe. 1993. Mtodo si m-
pl i fi cado para caracteri zar fl ui dos que si guen l a l ey de potenci a con
un vi scos metrode mezcl a ti pobandera. Revi sta Espaol a de Ci enci a
y Tecnol og a de Al i mentos 33: 529-547.
Castel l -Perez, M.E. and J.F. Steffe. 1990. Eval uati ng shear rates for
power l aw fl ui ds i n mi xer vi scometry. J. Texture Stud. 21: 439-453.
Castel l -Perez, E.M. and J.F. Steffe. 1992. Usi ng mi xi ng to eval uate
rheol ogi cal properti es. I n: Rao, A.M. and J.F. Steffe (edi tors). Vi s-
coel asti cProperti es of Food. El sevi er Appl i ed Sci . Pub., Ltd., Barki ng,
Engl and. pg. 247-284.
Castel l -Perez, M.E., J.F. Steffe and R.G. Morei ra. 1991. Si mpl e
determi nati on of power l aw fl ow curves usi ng a paddl e type mi xer
vi scometer. J. Texture Stud. 22: 303-316.
Cavel l a, S., S. Chemi n and P. Masi . 1992. Objecti ve measurement of
the stretchabi l i ty of Mozzarel l a cheese. J. Texture Stud. 23: 185-194.
396 Bibliography
Charm, S.E. 1962. Determi nati on of shear stress shear rate behavi or i n
foods i n the presence of a yi el d stress. J. Food Sci . 28: 107-113.
Chatraei , SH., C.W. Macosko. 1981. Lubri cated squeezi ng fl ow: a new
bi axi al extensi onal rheometer. J. Rheol . 25: 433-443.
Cheng, D.C-H. 1970. A desi gn procedure for pi pel i ne fl ow of non-
Newtoni an di spersed systems. Proceedi ngs of the Fi rst I nternati onal
Conference i n the Hydraul i c Transport of Sol i ds i n Pi pes
(Hydrotransport I ), Sept. 1-4. Bri ti sh Hydromechani cs Research
Associ ati on, Cranfi el d, Bedford, Engl and. pg. J5-77 to J5-95.
Cheng, D.C-H. 1973. A di fferenti al form of consti tuti ve rel ati on for
thi xotropy. Rheol . Acta 12: 228-233.
Cheng, D.C-H. 1986. Yi el d stress: a ti me-dependent property and how
to measure i t. Rheol . Acta 25: 542-554.
Cheng, D.C-H. and B.R. Parker. 1976. The determi nati on of wal l -sl i p
vel oci ty i n the coaxi al cyl i nder vi scometer. Proceedi ngs of the
Seventh I nternati onal Congress on Rheol ogy. Gotenberg, Sweden.
pg. 518-519.
Chhabra, R.P. 1992. Bubbl es, Drops and Parti cl es i n Non-Newtoni an
Fl ui ds. CRC Press, Boca Raton, FL.
Chri stensen, C.M. 1987. Percepti on of sol uti on vi scosi ty. I n: Moskowi tz,
H.R. (edi tor). Food Texture. Marcel Dekker, New York.
Chri sti anson, D.D., Casi raghi , E.M. and E.B. Bagl ey. 1985. Uni axi al
compressi on of bonded and l ubri cated gel s. J. Rheol . 29: 671-684.
Chuma, Y., T. Shi ga and M. I wamoto. 1978. Mechani cal properti es of
Satsuma orange as rel ated to the desi gn of a contai ner for bul k
shi pment. J. Texture Stud. 9: 461-479.
Churchi l l , S.W. 1988. Vi scous Fl ows: The Practi cal Use of Theory.
Butterworths, Boston.
Cl app, R.M. 1961. Turbul ent heat transfer i n pseudopl asti c non-
Newtoni an fl ui ds. I nternati onal Devel opment i n Heat transfer.
ASME, Part I I I , Sec. A., pg. 652-661.
Cogswel l , F.N. 1972. Convergi ng fl ow of pol ymer mel ts i n extrusi on
di es. Pol y. Engn. and Sci . 12(1): 64-73.
Cogswel l , F.N. 1978. Convergi ng fl ow and stretchi ng fl ow: a compi l a-
ti on. J. Non-Newtoni an Fl ui d Mech. 4: 23-38.
Cogswel l , F.N. 1981. Pol ymer Mel t Rheol ogy. Hal sted Press, New York.
Col l i ns, M. and W.R. Schowal ter. 1963. Behavi or of non-Newtoni an
fl ui ds i n the entry regi on of a pi pe. AI ChE J. 9: 804-809.
Covey, G.H. and B.R. Stanmore. 1981. Use of the paral l el pl ate pl as-
tometer for the characteri zati on of vi scous fl ui ds wi th a yi el d stress.
J. Non-Newtoni an Fl ui d Mech. 8: 249-260.
Cox, W.P. and E.H. Merz. 1958. Correl ati on of dynami c and steady fl ow
vi scosi ti es. J. Pol ymer Sci . 28: 619-622.
Crane Co. 1982. Fl ow of fl ui ds through val ves, fi tti ngs and pi pe.
Techni cal Paper No. 410M, twenty-fi rst pri nti ng. Crane Co. 300 Park
Ave., New York.
Cri stescu, N. 1989. Rock Rheol ogy. Kl uwer Academi c Pub., Boston.
397
Cross, M.M. 1965. Rheol ogy of non-Newtoni an fl ow: equati on for
pseudopl asti c systems. J. Col l oi d Sci . 20: 417-437.
Cuvel i er, G., C. Pei gney-Noury and B. Lunay. 1990. Vi scoel asti c prop-
erti es of physi cal gel s: cri ti cal behavi or at the gel poi nt. I n: Phi l l i ps,
G.O., P.A. Wi l l i ams and D.J. Wedl ock (edi tors). Gums and Stabi l i zers
for the Food I ndustry 5, I RL Press, New York.
Dai l , R.V. and J.F. Steffe. 1990a. Rheol ogi cal characteri zati on of
crossl i nked waxy mai ze starch sol uti ons under l ow aci d asepti c
processi ng condi ti ons usi ng tube vi scometry techni ques. J. Food Sci .
53: 1660-1665.
Dai l , R.V. and J.F. Steffe. 1990b. Di l atancy i n starch sol uti ons under
l ow aci d asepti c processi ng condi ti ons. J. Food Sci . 53: 1764-1765.
Dal l y, J.W. and W.F. Ri l ey. 1965. Experi mental Stress Anal ysi s.
McGraw-Hi l l Book Company, New York.
DAppol oni a, B.L. and W.H. Kunerth (edi tors). 1984. The Fari nograph
Handbook, Thi rd Edi ti on. The Ameri can Associ ati on of Cereal Che-
mi sts, St. Paul , MN.
Darby, R. 1976. Vi scoel asti c Fl ui ds: An I ntroducti on to Properti es and
Behavi or. Marcel Dekker, New York.
Darby, R. 1985. Couette vi scometer data reducti on for materi al s wi th
a yi el d stress. J. Rheol . 29: 369-378.
da Si l va, J.A.L., M.P. Gonal ves and M.A. Rao. 1994. I nfl uence of
temperature on the dynami c and steady-shear rheol ogy of pecti n
di spersi ons. Carbohydrate Pol ymers 23: 77-87.
da Si l va, J.A.L. and M.A. Rao. 1992. Vi scoel asti c properti es of food
hydrocol l oi d di spersi ons. I n: Rao, A.M. and J.F. Steffe (edi tors).
Vi scoel asti c Properti es of Food. El sevi er Appl i ed Sci . Pub., Ltd.,
Barki ng, Engl and. pg. 247-284.
Daubert, C.R. and J.F. Steffe. 1995. El ectrorheol ogy of mi l k chocol ate.
J. Texture Stud. 27: 93-108.
Deal y, J.M. 1982. Rheometers for Mol ten Pl asti cs. Van Nostrand
Rei nhol d Company. New York.
Deal y, J.M. 1994. Offi ci al nomencl ature for materi al functi ons
descri bi ng the response to a vi scoel asti c fl ui d to vari ous sheari ng and
extensi onal deformati ons. J. Rheol . 38: 179-191.
Deal y, J.D. and A.J. Gi acomi n. 1988. Sl i di ng pl ate and sl i di ng cyl i nder
rheometers. I n: Col l yer, A.A. and D.W. Cl egg (edi tors). Rheol ogi cal
Measurements. El sevi er Appl i ed Sci ence, New York. pg. 383-404.
Deal y, J.M. and K.F. Wi ssburn. 1990. Mel t Rheol ogy and i ts Rol e i n
Pl asti cs Processi ng. Van Nostrand Rei nhol d, New York.
de Brui jne, D.W., J. de Looff and A. van Eul em. 1990. The rheol ogi cal
properti es of bread dough and thei r rel ati on to baki ng. I n: Carter,
R.E. (edi tor). Rheol ogy of Food, Pharmaceuti cal and Bi ol ogi cal
Materi al s wi th General Rheol ogy. El sevi er Appl i ed Sci ence, New
York.
De Kee, D., G. Turcotte and K. Fi l dey. 1980. New method for the
determi nati on of yi el d stress. J. Texture Stud. 10: 281-288.
398 Bibliography
Denn, M.M. 1980. Process Fl ui d Mechani cs. Prenti ce-Hal l , I nc.,
Engl ewood Cl i ffs, NJ.
Dervi sogl u, M. and J.L. Koki ni . 1986. Steady shear rheol ogy and fl ui d
mechani cs of four semi -sol i d foods. J. Food Sci . 51: 541-545, 625.
Di cki e, A.M. and J.L. Koki ni . 1982. Use of the Bi rd-Lei der equati on i n
food rheol ogy. J. Food Proc. Engn. 5: 157-174.
Di cki e, A.M. and J.L Koki ni . 1983. An i mproved model for food thi ckness
from non-Newtoni an fl ui d mechani cs i n the mouth. J. Food Sci . 48:
57-61, 65.
Di enes, G.J. and H.F. Kl emm. 1946. Theory and appl i cati on of paral l el
pl ate pl astometer. J. Appl . Phys. 17: 458-471.
Di ntenfass, L. 1985. Bl ood Vi scosi ty, Hypervi scosi ty & Hypervi sco-
saemi a. MTP Press Li mi ted. Boston, MA.
Dodge, D.W. and A.B. Metzner. 1959. Turbul ent fl ow of non-Newtoni an
systems. AI ChE J. 5: 189-204.
Dol an, K.D. and J.F. Steffe. 1990. Model i ng the rheol ogi cal behavi or of
gel ati ni zi ng starch sol uti ons usi ng mi xer vi scometry data. J. Texture
Stud. 21: 265-294.
Dol an, K.D., J.F. Steffe and R.G. Morgan. 1989. Back extrusi on and
si mul ati on of vi scosi ty devel opment duri ng starch gel ati ni zati on. J.
Food Proc. Engn. 11: 79-101.
Dorai swamy, D., A.N. Mujumdar, I . Tsao, A.N. Beri s, S.C. Danforth and
A.B. Metzner. 1991. The Cox-Mertz rul e extended: A rheol ogi cal
model for concentrated suspensi ons and other materi al s wi th a yi el d
stress. J. Rheol . 35: 647-685.
Drake, B. 1987. Food psychorheol ogy. I n: Moskowi tz, H.R. (edi tor). Food
Texture. Marcel Dekker, New York.
Durguei l , E.J. 1987. Determi nati on of the consi stency of non-Newtoni an
fl ui ds usi ng a Brookfi el d HBTvi scometer. Proc. South Afri can Sugar
Technol . Assoc., June: 32-39.
Edwards, D.A., H. Brenner and D.T. Wasan. 1991. I nterfaci al Transport
Processes and Rheol ogy. Butterworth-Hei nemam, Stonheam, MA.
El l i ott, J.H. and A.J. Ganz. 1971. Modi fi cati on of food characteri sti cs
wi th cel l ul ose hydrocol l oi ds, I . Rheol ogi cal characteri zati on of an
organol epti c property. J. Text. Stud. 2: 220-229.
El l i ott, J.H. and A.J. Ganz. 1977. Sal ad Dressi ngs - prel i mi nary rhe-
ol ogi cal characteri zati on. J. Text. Stud. 8: 359-371.
El l i s, S.B. 1927. Thesi s, Lafayette Col l ege, Pa. Ci ted i n: Whorl ow, R.W.
1980. Rheol ogi cal Techni ques. Hal sted Press (Di v. of John Wi l ey &
Sons), New York.
Eri ckson, E.R., R.A. Berntsen and M.A. El i ason. 1966. Vi scosi ty of corn
syrup. J. of Chem. and Engn. Data 11: 485-488.
Fari di , H. and V.F. Rasper. 1987. The Al veograph Handbook. Amer.
Assoc. of Cereal Chemi sts. St. Paul , MN.
Ferguson, J. and Z. Kembl owski . 1991. Appl i ed Fl ui d Rheol ogy. El sevi er
Appl i ed Sci ence, New York.
Ferry, J.D. 1977. Osci l l ati on vi scometry - effects of shear rate and fre-
quency. Measurement & Control , September - October: 89-91.
399
Ferry, J.D. 1980. Vi scoel asti c Properti es of Pol ymers (Thi rd Edi ti on).
John Wi l ey and Sons, New York.
Ford, E.W. and J.F. Steffe. 1986. Quanti fyi ng thi xotropy i n starch-
thi ckened, strai ned apri cots usi ng mi xer vi scometry techni ques. J.
Fri edman, H.H., J.E. Whi tney and A.S. Szczesnai k. 1963. The Textu-
rometer - a new i nstrument for texture measurement. J. Food Sci .
28: 390-396.
Ful l er, G.G., C.A. Cathey, B. Hubbard and B.E. Zebrowski . 1987.
Extensi onal vi scosi ty measurement for l ow-vi scosi ty fl ui ds. J. Rheol .
31: 235-249.
Garci a, E.J. and J.F. Steffe. 1986. Opti mum economi c pi pe di ameter
for pumpi ng Herschel -Bul kl ey fl ui ds i n l ami nar fl ow. J. Food Proc.
Engn. 8: 117-136.
Garci a, E.J. and J.F. Steffe. 1987. Compari son of fri cti on factor equa-
ti ons for non-Newtoni an fl ui ds i n tube fl ow. J. Food Proc. Engn. 9:
93-120.
Gi bson, A.G. 1988. Convergi ng di es. I n: Col l yer, A.A. and D.W. Cl egg
(edi tors). Rheol ogi cal Measurements. El sevi er Appl i ed Sci ence, New
York. pg. 49-92.
Gi esekus, H. and G. Langer. 1977. Di e Besti mmung der wahren Fl i e-
kurven ni cht-newtonscher Fl ssi gkei ten und pl asti scher Stoffe mi t
der Methode der reprsentati ven Vi skosi tt. Rheol . Acta 16: 1-22.
Gogos, C.G. and S. Bhakuni . 1992. An on-l i ne sl i t rheometer for mea-
surement of the rheol ogi cal properti es of doughs. I n: Koki ni , J.L.,
C-T. Ho and M.V. Karwe (edi tors). Food Extrusi on Sci ence and
Technol ogy. Marcel Dekker, I nc., New York. pg. 255-262.
Goodri ch, K., A. Yoshi mura and R.K. Prudhomme. 1989. Measurement
of the modul us and yi el d strength of soft gel s: experi ments and
numeri cal si mul ati on. J. Rheol . 33: 317-327.
Gri skey, R.G. and R.G. Green. 1971. Fl ow of di l atant (shear-thi ckeni ng)
fl ui ds. AI ChE J. 17: 725-728.
Grovi er, G.W. and K. Azi z. 1972. The Fl ow of Compl ex Mi xtures i n Pi pes.
R.E. Kri eger Publ i shi ng Co., Mal abar, Fl ori da.
Grul ke, E.A. 1994. Pol ymer Process Engi neeri ng. PTR Prenti ce Hal l ,
Engl ewood Cl i ffs, NJ.
Gupta, B.P. 1984. Di mensi onal anal ysi s. I n: Hi l l , H.E. and J.W. Prane
(edi tors). Appl i ed Stati sti cal Techni ques for Sel ected I ndustri es.
John Wi l ey and Sons, New York. pg. 551-581.
Gupta, R.K. and T. Sri dhar. 1988. El ongati onal rheometers. I n: Col l yer,
A.A. and D.W. Cl egg (edi tors). Rheol ogi cal Measurements. El sevi er
Appl i ed Sci ence, New York. pg. 211-246.
Haghi ghi , K., A.K. Sri vastava and J.F. Steffe. 1987. The rheol ogi cal
approach to soi l model i ng. Trans. ASAE 30: 1661-1672.
Hai ghton, A.J. 1959. The measurement of hardness of margari ne and
fats wi th cone penetrometers. J. Amer. Oi l Chem. Soc. 36: 345-348.
400 Bibliography
Hal i m, H.K. and C.F. Shoemaker. 1990. Effect of addi ti on of -, -, and
-casei n, and Na-casei nate on vi scoel asti c properti es of ski m mi l k
curd. J. Texture Stud. 21: 323-337.
Hal l i day, P.J. and A.C. Smi th. 1995. Esti mati on of the wal l sl i p vel oci ty
i n the capi l l ary fl ow of potato granul e pastes. J. Rheol . 39: 139-149.
Hal mos, A.L. and C. Ti u. 1981. Li qui d foodstuffs exhi bi ti ng yi el d stress
and shear-degradabi l i ty. J. Texture Stud. 12: 39-46.
Hamann, D.D. 1983. Structural fai l ure i n sol i d foods. I n: Pel eg, M. and
E.B. Bagl ey (edi tors). Physi cal Properti es of Foods. AVI Publ i shi ng
Co., Westport, CT. pg. 351-384.
Hamann, D.D. and T.C. Lani er. 1987. I nstrumental methods for pre-
di cti ng seafood sensory texture qual i ty. I n: Kramer, D.E. and J.
Li ston (edi tors). Seafood Qual i ty Determi nati on. El sevi er Appl i ed
Sci . Publ i shers, Barki ng, Essex, Engl and. pg. 123-136.
Han, C.D. 1988. Sl i t rheometry. I n: Col l yer, A.A. and D.W. Cl egg (edi -
tors). Rheol ogi cal Measurements. El sevi er Appl i ed Sci ence, New
York. pg. 25-48.
Hanks, R.W. 1963. Lami nar-turbul ent transi ti on of fl ui ds wi th a yi el d
stress. AI ChE J. 9: 306-309.
Harnby, N., M.F. Edwards and A.W. Ni enow (edi tors). 1985. Mi xi ng i n
the Process I ndustri es. Butterworth, Boston.
Harper, J.C. and A.F. El Sahri gi . 1965. Vi scometri c behavi or of tomato
concentrates. J. Food Sci . 30: 470-476.
Hartnett, J.P. and R.Y.Z. Hu. 1989. The yi el d stress-an engi neeri ng
real i ty. J. Rheol . 33: 671-679.
Hayashi , N., I . Hayakawa and Y. Fuji o. 1991. Entrance effect correcti on
on the fl ow of moi sturi zed soy protei n i sol ate mel t i n an extrusi on
rheometer. I nt. J. Food Sci . Technol . 26: 567-574.
Heywood, N. 1991a. Pi pel i ne desi gn for non-settl i ng sl urri es. I n: Brown,
N.P. and N.I . Heywood (edi tors). Sl urry Handl i ng: Desi gn of Sol i d
Li qui d Systems. El sevi er Appl i ed Sci ence, New York. pg. 125-165.
Heywood, N.I . 1991b. Rheol ogi cal characteri zati on of non-settl i ng
sl urri es. I n: Brown, N.P. and N.I . Heywood (edi tors). Sl urry Han-
dl i ng: Desi gn of Sol i d-Li qui d Systems. El sevi er Appl i ed Sci ence, New
York.
Hi l l , M.A. 1991. Dynami c rheol ogy: appl i cati on to the formul ati on of a
real food product. Food Sci . Technol . Today 5(1): 29-31.
Hi l l , M.A., J.R. Mi tchel and P.A. Sherman. 1995. The rel ati onshi p
between the rheol ogi cal and sensory properti es of l emon pi e fi l l i ng.
Hol dsworth, S.D. 1993. Rheol ogi cal model s used for the predi cti on of
the fl ow properti es of food products: a l i terature revi ew. Trans. I .
Chem. E. 71 (Part C): 139-179.
Hol l and, D. 1994. Letter to the edi tor: whi ch i s more cri ti cal - the stress
or the strai n. J. Rheol . 38: 1941-1943.
Hol l and, F.A. and F.S. Chapman. 1966. Li qui d Mi xi ng and Processi ng
i n Sti rred Tanks. Rei nhol t Publ i shi ng Corp., New York.

401
Houska, M., J. Sesrk, J. Jeschke, M. Adam and J. Pri dal . 1988. I n: H.
Gi esekus (edi tor). Progress and Trends i n Rheol ogy I I . Proceedi ngs
of the Second Conference of European Rheol ogi sts, Prague, June
17-20, 1986. Spri nger-Verl ag, New York. pg 460-463.
Huang, H. and J.L. Koki ni . 1993. Measurement of bi axi al extensi onal
vi scosi ty of wheat fl our doughs. J. Rheol . 37: 879-891.
James, A.E., D.J.A. Wi l l i ams and P.R. Wi l l i ams. 1987. Di rect mea-
surement of stati c yi el d properti es of cohesi ve suspensi ons. Rheol .
Acta 26: 437-446.
James, D.F. and K. Wal ters. 1993. A cri ti cal apprai sal of avai l abl e
methods for the measurement of extensi onal properti es of mobi l e
systems. I n: Col l yer, A.A. (edi tor). Techni ques i n Rheol ogi cal Mea-
surement. Chapman & Hal l , New York.
Jastrzebski , Z.D. 1967. Entrance effects and wal l effects i n an extrusi on
rheometer duri ng the fl ow of concentrated suspensi ons. I nd. Engn.
Chem. Fund. 6: 445-454.
Jones, D.M., K. Wal ters and P.R. Wi l l i ams. 1987. On the extensi onal
vi scosi ty of mobi l e pol ymer systems. Rheol . Acta 26: 20-30.
Kal etunc-Gencer, G. and M. Pel eg. 1984. Di gi ti zer ai ded determi nati on
of yi el d stress i n semi -l i qui d foods. J. Food Sci . 49: 1620-1621.
Kawanari , M., D.D. Hamann, K.R. Swartzel and A.P. Hansen. 1981.
Rheol ogi cal and texture studi es of butter. J. Texture Stud. 12:
483-505.
Kawata, M.K. Kurase, A. Nagashi ma and K.Yoshi da. 1991. Capi l l ary
vi scometers. I n: Wakeham, W.A., A. Nagashi ma and J.V. Sengers
(edi tors). Measurement of the Transport Properti es of Li qui ds, Vol .
I I I . Bl ackwel l Sci enti fi c, London.
Keentok, H. 1982. The measurement of yi el d stress of l i qui ds. Rheol .
Acta 21: 325-332.
Ki l janski , T. 1989. A method for the correcti on of wal l -sl i p effect i n a
Couette rheometer. Rheol . Acta 28: 61-64.
Ki ttredge, C.P. and D.S. Rowl ey. 1957. Resi stance coeffi ci ents for
l ami nar and turbul ent fl ow through one-hal f i nch val ves and fi tti ngs.
Trans. ASME 79: 1759-1766.
Koki ni , J.L. 1992. Rheol ogi cal properti es of food. I n: Hel dman, D.R. and
D.B. Lund (edi tors). Handbook of Food Engi neeri ng. Marcel Dekker,
New York. pg. 1-38.
Koki ni , J.L. and E.L. Cussl er. 1987. The psychophysi cs of fl ui d texture.
I n: Moskowi tz, H.R. (edi tor). Food Texture. Marcel Dekker, New
York.
Koki ni , J.L. and A. Di cki e. 1981. An attempt to i denti fy and model
transi ent vi scoel asti c fl ow i n foods. J. Texture Stud. 12: 539-557.
Koki ni , J.L. and G.L. Pl utchok. 1987. Vi scoel asti cproperti es of semi sol i d
foods and thei r bi opol ymeri c components. Food Technol . 41: 89-95.
Koutsakos, E. and A.W. Ni enow. 1990. Effects of rheol ogi cal properti es
of si mul ated fermentati on broths on fl ow i n sti rred bi oreactors: a l aser
402 Bibliography
anemometry study. I n: R.E. Carter (edi tor). Rheol ogy of Food,
Pharmaceuti cal and Bi ol ogi cal Materi al s wi th General Rheol ogy.
El sevi er Appl i ed Sci ence, New York. pg. 284-303.
Krayni k, A.M., J.H. Aubert, R.N. Chapman and D.C. Gyure. 1984. The
hel i cal screw rheometer: a new concept i n rotati onal rheometry. Proc.
Ann. Tech. Conf. Soc. Pl asti cs Engrs. pg. 405-408.
Kri eger, I .M. 1968. Shear rate i n couette vi scometer. Trans. Soc. Rheol .
12: 5-11.
Kri eger, I .M. and H. El rod. 1953. Di rect determi nati on of the fl ow curves
of non-Newtoni an Fl ui ds. I I . Shear rate i n the concentri c cyl i nder
vi scometer. J. Appl . Physi cs. 24: 134-136.
Kri eger, I .M. and S.H. Maron. 1952. Di rect determi nati on of the fl ow
curves of non-Newtoni an fl ui ds. J. Appl . Physi cs 23: 147-149.
Kri eger, I .M. and S.H. Maron. 1954. Di rect determi nati on of the fl ow
curves of non-Newtoni an Fl ui ds. I I I . Standardi zed treatment of
vi scometri c data. J. Appl . Physi cs. 25: 72-75.
Laba, D. 1993a. Rheol ogi cal profi l es. I n: Laba, D. (edi tor). 1993. Rhe-
ol ogi cal Properti es of Cosmeti cs and Toi l etri es. Marcel Dekker, I nc.
New York. pg. 403-420
Laba, D. (edi tor). 1993b. Rheol ogi cal Properti es of Cosmeti cs and Toi -
l etri es. Marcel Dekker, I nc. New York.
Lang, E.R. and C. Rha. 1981. Determi nati on of yi el d stresses of
hydrocol l oi d di spersi ons. J. Texture Stud. 12: 47-62.
Langhaar, H.L. 1980. Di mensi onal Anal ysi s and Theory of Model s.
Robert E. Kri eger Publ i shi ng Company, Hunti ngton, New York.
Larson, R.G. 1985. Consti tuti ve rel ati onshi ps for pol ymeri c materi al s
wi th power-l aw of rel axati on ti mes. Rheol . Acta 24: 327-334.
Larson, R.G. 1992. I nstabi l i ti es i n vi scoel asti c fl ows. Rheol . Acta 31:
213-263.
Launay, B. and J. Bur. 1977. Use of the Chopi n Al veographe as a
rheol ogi cal tool . I I . Dough properti es i n bi axi al extensi on. Cereal
Chem. 54: 1152-1158.
Lei der, P.L. and R.B. Bi rd. 1974. Squeezi ng fl ow between paral l el
pl ates. I . Theoreti cal anal ysi s. I nd. Engn. Chem. Fundam. 13:
336-341.
Lei ghton, A., A. Levi ton and O.E. Wi l l i ams. 1934. The apparent vi s-
cosi ty of i ce cream. J. Dai ry Sci . 17: 639-650.
Lewi cki , P.P. and K. Ski erkowski . 1988. Fl ow of frui t and vegetabl e
purees through pi pel i nes. I n: H. Gi esekus (edi tor). Progress and
Trends i n Rheol ogy I I . Proceedi ngs of the Second Conference of
European Rheol ogi sts, Prague, June 17-20, 1986. Spri nger-Verl ag,
New York. pg 443-445.
Lewi s, M.J. 1987. Physi cal Properti es of Foods and Food Processi ng
Systems. El l i s Horwood. Chi chester, Engl and.
Lodge, A.S. 1988. Normal stress di fferences from hol e pressure mea-
surements. I n: Col l yer, A.A. and D.W. Cl egg (edi tors). Rheol ogi cal
Measurements. El sevi er Appl i ed Sci ence, New York. pg. 345-382.
403
Lord, D.L., B.W. Hul sey and L.L. Mel ton. 1967. General turbul ent pi pe
fl ow scal e-up correl ati on for rheol ogi cal l y compl ex fl ui ds. Soc. Petrol .
Engrs. J. 7: 252-258.
Mackey, K.L., R.Y. Ofol i , R.G. Morgan and J.F. Steffe. 1989. Rheol ogi cal
model i ng of potatofl our duri ng extrusi on cooki ng. J. Food Proc. Engn.
12: 1-11.
Macosko, C. W. 1994. Rheol ogy: Pri nci pl es, Measurements and
Appl i cati ons. VCH Publ i shers. New York.
Matz, S.A. 1962. Food Texture. AVI Publ i shi ng Co., I nc., Westport,
Connecti cut.
McCarthy, K.L. and J.D. Seymour. 1994. Gravi ty current anal ysi s of
the Bostwi ck Consi stometer for power l aw fl ui ds. J. Text. Stud. 25:
207-220.
Mei ssner, J. 1972. Devel opment of a uni versal extensi onal rheometer
for the uni axi al extensi on of pol ymer mel ts. Trans. Soc. Rheol . 16:
405-420.
Menji var, J.A., C.N. Chang and L. Macei l . 1992. Extrudate swel l
behavi or of wheat fl our doughs. I n: P. Mol denaers and R. Keuni ngs
(edi tors). Theoreti cal and Appl i ed Rheol ogy, Vol ume 2. Proceedi ngs
of the XI th I nternati onal Congress on Rheol ogy. Brussel s, Bel gi um,
Aug. 17-21, 1992. El sevi er Sci ence Publ i shers, New York. pg.
714-716.
Menji var, J.A. and H. Fari di . 1994. Rheol ogi cal properti es of cooki e and
cracker doughs. I n: Fari di , H. (edi tor). The Sci ence of Cooki e and
Cracker Producti on. Chapman & Hal l , New York. pg. 283-322.
Metzner, A.B. 1956. Non-Newtoni an technol ogy: fl ui d mechani cs,
mi xi ng, heat transfer. I n: Drew, T.B. and J.W. Hoopes (edi tors).
Advances i n Chemi cal Engi neeri ng, Vol . 1. Academi c Press, I nc., New
York. pg. 77-153.
Metzner, A.B. 1961. Fl ow of non-Newtoni an fl ui ds. I n: Handbook of
Fl ui d Dynami cs, Secti on 7, McGraw-Hi l l Book Co., I nc. pg. 7-1 to7-30.
Metzner, A.B. and A.P. Metzner. 1970. Stress l evel s i n rapi d extensi onal
fl ows of pol ymeri c fl ui ds. Rheol . Acta 9: 174-181.
Mewi s, J. 1979. Thi xotropy - a general revi ew. J. Non-Newtoni an Fl ui d
Mech. 6: 1-20.
Mi chi yosi , L., K. Mi zuno and Y. Hoshi ai . 1966. Studi es on the fl ow of
sl urri es through pi pe. I . Entrance Regi on of l ami nar fl ow. I nt. Chem.
Engn. 6: 373-381
Mi ddl eman, S. 1968. The Fl ow of Hi gh Pol ymers. I ntersci ence Pub-
l i shers, New York.
Mi ddl eman, S. 1977. Fundamental s of Pol ymer Processi ng. McGraw-
Hi l l Book Co., New York.
Mi ssai re, F., C-G. Qi u and M.A. Rao. 1990. Yi el d stress of structured
and unstructured food suspensi ons. J. Texture Stud. 21: 479-490.
Mi tschka, P. 1982. Si mpl e conversi on of Brookfi el d R.V.T. readi ngs i nto
vi scosi ty functi ons. Rheol . Acta 21: 207-209.
Mi zrahi , S. and Z. Berk. 1972. Fl ow behavi or of concentrated orange
jui ce: mathemati cal treatment. J. Texture Stud. 3: 69-79.
404 Bibliography
Mohseni n, H. 1986. Physi cal Properti es of Pl ant and Ani mal Materi al s,
Vol . 1 (Second Edi ti on). Gordon and Breach Sci ence Publ i shers, New
York.
Montejano, J.G., D.D. Hamann and T.C. Lani er. 1985. Compari son of
two i nstrumental methods wi th sensory texture of protei n gel s. J.
Mooney, M. 1931. Expl i ci t formul as for sl i p and fl ui di ty. J. Rheol . 30:
337-357.
Morgan, R.G. 1979. Model i ng the effects of temperature-ti me hi story,
temperature, shear rate and moi sture on the vi scosi ty of defatted soy
fl our. Ph.D. Di ssertati on. Texas A&M Uni versi ty, Col l ege Stati on,
TX.
Morgan, R.G., J.F. Steffe and R.Y. Ofol i . 1989. A general i zed rheol ogi cal
model for extrusi on of protei n doughs. J. Food Proc. Engn. 11: 55-78.
Morton-Jones, D.H. 1989. Pol ymer Processi ng. Chapman and Hal l . New
York.
Muoz, A.M., A.S. Szczesnai k, M.A. Ei nstei n and N.O. Schwartz, 1992.
The texture profi l e. I n: Hootman, R.C. (edi tor). Manual on
Descri pti ve Anal ysi s Testi ng for Sensory Eval uati on. ASTM Manual
Seri es: MNL 13. Amer. Soc. for Testi ng and Materi al s, 1916 Race
Street, Phi l adel phi a, PA.
Murata, T. and S. Oka. 1968. Determi nati on of parameters of a materi al
obeyi ng Cassons equati on by some rheometers. Bi orheol ogy 5:
95-102.
Murphy, C.E. 1950. Si mi l i tude i n Engi neeri ng. The Ronal d Press
Company, New York.
Nagata, S. 1975. Mi xi ng: Pri nci pl es and Appl i cati ons. John Wi l ey and
Sons, New York.
Nagata, S., N. Masabuma, T. Hi sayuki , H. Hi rabayashi and S. Gotoh.
1970. Power consumpti on of mi xi ng i mpel l ers i n Bi ngham pl asti c
l i qui ds. J. Chem. Engn. of Japan 3: 237-243.
Nakayama, T., E. Ni wa and I . Hamada. 1980. Pi pe transportati on of
mi nced fi sh paste. J. Food Sci . 45: 844-847.
Navi cki s, L.L. and E.B. Bagl ey. 1983. Yi el d stresses i n concentrated
di spersi ons of cl osel y packed, deformabl e gel parti cl es. J. Rheol . 27:
519-536.
Ni el sen, L.E. and R.F. Landel . 1994. Mechani cal Properti es of Pol ymers
and Composi tes (Second Edi ti on). Marcel Dekker, I nc. New York.
Nguyen, Q.D. and D.V. Boger. 1985. Di rect yi el d stress measurement
wi th the vane method. J. Rheol . 29: 335-347.
Nguyen, Q.D. and D.V. Boger. 1987. Characteri zati on of yi el d stress
wi th concentri c cyl i nder vi scometers. Rheol . Acta 26: 508-515.
Nol an, E.J., V.H. Hol si nger and J.J. Shi eh. 1989. Dynami c rheol ogi cal
properti es of natural and i mi tati on mozzarel l a cheese. J. Texture
Stud. 20: 179-189.
Ofol i , R.Y., R.G. Morgan and J.F. Steffe. 1987. A general i zed rheol ogi cal
model for i nel asti c fl ui d foods. J. Texture Stud. 18: 213-230.
405
Ofol i , R.Y. and J.F. Steffe. 1992. Some observati ons on the use of sl i t
rheometry for characteri zi ng the pri mary normal stress di fference of
extrudates. J. Food Engn. 18: 145-157.
Ol droyd, J.G. 1956. Non-Newtoni an fl ow of sol i ds and l i qui ds. I n: Ei ri ch,
F.R. (edi tor). Rheol ogy: Theory and Appl i cati ons, Vol . 1. Academi c
Press I nc., New York. pg 653-682.
Ol dshue, J.Y. 1983 Fl ui d Mi xi ng Technol ogy. Chemi cal Engi neeri ng,
McGraw-Hi l l Publ i cati ons Co., New York
Osori o, F.A. and J.F. Steffe. 1984. Ki neti c energy cal cul ati ons for non-
Newtoni an fl ui ds i n ci rcul ar tubes. J. Food Sci . 49: 1295-1296, 1315.
Osori o, F.A. and J.F. Steffe. 1986. Back extrusi on of power l aw fl ui ds.
J. Texture Stud. 18: 43-63
Padden, F.J. and T.W. Dewi tt. 1954. Some rheol ogi cal properti es of
concentrated pol yi sobutyl ene sol uti ons. J. Appl . Physi cs 28:
1086-1091.
Padmanabhan, M. and M. Bhattacharya. 1993. Pl anar extensi onal
vi scosi ty of corn meal dough. J. Food Engn. 18: 389-411.
Padmanabhan, M. and M. Bhattacharya. 1994. I n-l i ne measurements
of rheol ogi cal properti es of pol ymer mel ts. Rheol . Acta 33: 71-87.
Paredes, M.D.C., M.A. Rao and M.C. Bourne. 1989. Rheol ogi cal char-
acteri sti cs of sal ad dressi ngs 2: effect of storage. J. Texture Stud. 20:
235-250.
Parzonka, W. and J. Vocadl o. 1968. Mthode de l a caractri sti que du
comportement rhol ogi que des substances vi scopl asti ques daprs l es
mesures au vi scosi mtre de Couette (modl e nouveau troi s
paramters). Rheol . Acta. 7: 260-265.
Patton, T.C. 1964. Pai nt Fl ow and Pi gment Di spersi on. I ntersci ence
Publ i shers, New York.
Pel eg, M. 1980. Li neari zati on of rel axati on and creep curves of sol i d
bi ol ogi cal materi al s. J. Rheol . 24: 451-463.
Pel eg, M. and M.D. Normand. 1983. Compari son of two methods for
stress rel axati on data presentati on of sol i d foods. Rheol . Acta 22:
108-113.
Pel eg, M. and M.D. Normand. 1995. Sti ffness assessment from jagged
force-deformati on rel ati onshi ps. J. Texture Stud. 26: 353-370.
Petri e, C.J.S. 1979. Extensi onal Fl ows. Pi ttman, London.
Phi l i ppoff, W. 1935. Kol l oi d-Zei t. 71:1-16. Ci ted i n: Brodkey, R.S. 1967.
The Phenomena of Fl ui d Moti ons. Addi son-Wesl ey Publ i shi ng Co.,
Menl o Park, Cal i forni a.
Pi pki n, A.C. 1986. Lectures on Vi scoel asti ci ty Theory (Second Edi ti on).
Spri nger-Verl ag, New York.
Pokorny, J., J. Davi dek, S. Dobi asova, P. Stern and J. Cmol i k. 1984.
Compari son of sensory anal ysi s and rotati onal rheometry i n the
texture eval uati on of margari nes. J. Texture Stud. 15: 395-406.
Pol akowski , N.H. and E.J. Ri pl i ng. 1966. Strength and Structure of
Engi neeri ng Materi al s. Prenti ce-Hal l , I nc., Engl ewood Cl i ffs, NJ
Powel l , R.E. and H.J. Eyri ng. 1944. Mechani sms for the rel axati on
theory of vi scosi ty. Nature 154: 427-428.
406 Bibliography
Powel l , R.L. 1988. Rotati onal vi scometry. I n: Col l yer, A.A. and D.W.
Cl egg (edi tors). Rheol ogi cal Measurements. El sevi er Appl i ed Sci ence,
New York. pg. 247-296.
Prenti ce, J.H. 1992. Dai ry Rheol ogy: A Conci se Gui de. VCH Publ i shers,
I nc., New York.
Prenti ce, J.H. and D. Huber. 1983. Resul ts of a col l aborati ve study on
measuri ng rheol ogi cal properti es of foodstuffs. I n: Jowi tt R., F.
Escher, B. Hal l strm, H.F.Th. Meffert, W.E.L. Spi es and G. Vos
(edi tors). Physi cal Properti es of Foods. Appl i ed Sci ence Publ i shers.
pg. 123-184.
Pri ncen, H.M. 1986. A novel desi gn to el i mi nate end effects i n a con-
centri c cyl i nder vi scometer. J. Rheol . 30: 271-283.
Progel hof, R.C. and J.L. Throne. 1993. Pol ymer Engi neeri ng Pri nci pl es.
Hansen Publ i shers, New York.
Purkayastha, S., M. Pel eg, E.A. Johnson and M.D. Normand. 1985. A
computer ai ded characteri zati on of the compressi ve creep behavi or
of potato and cheddar cheese. J. Food Sci . 50: 45-50, 55.
Qi u, C-G. and M.A. Rao. 1988. Rol e of pul p content and parti cl e si ze i n
yi el d stress of appl e sauce. J. Food Sci . 53: 1165-1170.
Qi u, C-G. and M.A. Rao. 1989. Effect of di spersed phase on the sl i p
coeffi ci ent of appl esauce i n a concentri c cyl i nder vi scometer. J. Tex-
ture Stud. 20: 57-70.
Rao, M.A. 1975. Measurement of fl ow of a food suspensi on wi th a mi xer.
Rao, M.A. 1986. Rheol ogi cal properti es of fl ui d foods. I n: Rao, M.A. and
S.S.H. Ri zvi (edi tors). Engi neeri ng Properti es of Foods. Marcel
Dekker, I nc., New York. pg. 1-48.
Rao, M.A. and H.J. Cool ey. 1984. Determi nati on of effecti ve shear rates
i n rotati onal vi scometers wi th compl ex geometri es. J. Texture Stud.
15: 327-335.
Rao, M.A. and H.J. Cool ey. 1992. Rheol ogi cal behavi or of tomato pastes
i n steady and dynami c shear. J. Texture Stud. 23: 415-425.
Rao, M.A. and J.F. Steffe (edi tors). 1992. Vi scoel asti c Properti es of
Foods. El sevi er Appl i ed Sci ence, New York.
Rao, V.N.M. and G.E. Ski nner. 1986. Rheol ogi cal properti es of sol i d
foods. I n: Rao, M.A. and S.S.H. Ri zvi (edi tors). Engi neeri ng Properti es
of Foods. Marcel Dekker, I nc., New York. pg. 215-254.
Rasper, V.F. and K.R. Preston (edi tors). 1991. The Extensi graph
Handbook. The Ameri can Associ ati on of Cereal Chemi sts, St. Paul ,
MN.
Rei ner, M. 1964. The Deborah Number. Physi cs Today. January: 62.
Rha, C. and P. Pradi pasena. 1986. Vi scosi ty of protei ns. I n: Mi tchel l ,
J.R. and D.A. Leward (edi tors). Functi onal Properti es of Food
Macromol ecul es. El sevi er Sci ence Publ i shi ng Co., I nc., New York.
Rodri guez, F. 1982. Pri nci pl es of Pol ymer Systems (Second Edi ti on).
Hemi sphere Publ i shi ng Corp., New York.
Rohm, H. 1993. Rheol ogi cal behavi our of butter at l arge deformati ons.
407
Rohm, H. and K.-H. Wei di nger. 1993. Rheol ogi cal behavi our of butter
at smal l deformati ons. J. Texture Stud. 24: 157-172.
Roos, Y.H. 1992. Phase transi ti ons i n food systems. I n: Hel dman, D.R.
and D.B. Lund (edi tors). Handbook of Food Engi neeri ng. Marcel
Dekker, I nc., New York.
Roos, Y.H. 1995. Gl ass transi ti on - rel ated chemi cal changes i n foods.
Food Technol . 49: 97-102.
Rosenberg, M., Z. Wang, S.L. Chuang and C.F. Shoemaker. 1995.
Vi scoel asti c property changes i n Cheddar cheese duri ng ri peni ng. J.
Food Sci . 60: 640-644.
Ross-Murphy, S.B. 1988. Smal l deformati on measurements. I n: Bl an-
shard, J.M.V. and J.R. Mi tchel (edi tors). Food Structure - I ts Creati on
and Eval uati on. Butterworths, London. pg 387-400.
Ryan, N.W. and M.M. Johnson. 1959. Transi ti on from l ami nar to tur-
bul ent fl ow i n pi pes. AI ChE J. 5: 433-435.
Saki adi s, B.C. 1984. Fl ui d and parti cl e mechani cs. I n: Perry, R.H., D.W.
Green and J.O. Mal oney (edi tors). Perrys Chemi cal Engi neers
Handbook (Si xth Edi ti on), Secti on 5. McGraw-Hi l l Book Company,
New York. pg. 5-4 to 5-68.
Saravacos, G.D. 1970. Effect of temperature on vi scosi ty of frui t jui ces
and purees. J. Food Sci . 35: 122-125.
Saunders, S.R., D.D. Hamann and D.R. Li neback. 1992. A systems
approach to food materi al adhesi on. Lebensm.-Wi ss. u.-Technol . 25:
309-315.
Scal zo, A.M., R.W. Di ckerson, J.T. Peel er and R.B. Read. 1970. The
vi scosi ty of egg products. Food Tech. 24 (Nov.): 113-119.
Secor, R.B. 1988. Operabi l i ty of Extensi onal Rheometry by Stagnati on,
Squeezi ng, and Fi ber Drawi ng Fl ows: Computer-Ai ded-Anal ysi s,
Vi scoel asti c Characteri zati on, and Experi mental Anal ysi s. Ph.D.
Di ssertati on. Uni v. of Mi nnesota.
Seethamraju, K. and M. Bhattacharya. 1994. Effect of i ngredi ents on
the rheol ogi cal properti es of extruded corn meal . J. Rheol . 38:
1029-1044.
Senouci , A. and A.C. Smi th. 1988a. An experi mental study of food mel t
rheol ogy I . Shear vi scosi ty usi ng a sl i t di e vi scometer and a capi l l ary
rheometer. Rheol . Acta 27: 546-554.
Senouci , A. and A.C. Smi th. 1988b. An experi mental study of food mel t
rheol ogy I I . End pressure effects. Rheol . Acta 27: 649-655.
Severs. E.T. 1962. Rheol ogy of Pol ymers. Rei nhol d Publ i shi ngCompany,
New York.
Sharma, S.K., A.R. Hi l l , H.D. Goff and R. Yada. 1989. Measurement of
coagul ati on ti me and curd fi rmness of renneted mi l k usi ng a Nameter
vi scometer. Mi l chwi ssenschaft 44: 682-685.
Sherman, P. 1970. I ndustri al Rheol ogy. Academi c Press. New York.
Sherman, P. 1988. The sensory - rheol ogy i nterface. I n: Bl anshard,
J.M.V. and J.R. Mi tchel (edi tors). Food Structure - I ts Creati on and
Eval uati on. Butterworths, London. pg 417-432.
408 Bibliography
Shuey, W.C. and K.H. Ti ppl es (edi tors). 1980. The Amyl ograph
Handbook. The Ameri can Associ ati on of Cereal Chemi sts, St. Paul ,
MN.
Shukl a, A. and S.S.H. Ri zvi . 1995. Measurements of fl owabi l i ty of butter
by capi l l ary rheometry. J Texture Stud. 26: 299-311.
Shukl a, A., S.S.H. Ri zvi and J.A. Bartsch. 1995. Rheol ogi cal charac-
teri zati on of butter usi ng l ubri cated squeezi ng fl ow. J Texture Stud.
26: 313-323.
Si l vester, R.S. 1985. Mi xi ng of Non-Newtoni an Medi a, A Techni cal
Revi ew. (Bri ti sh Hyrdomechani cs Research Associ ati on. Fl ui d
Engi neeri ng. Fl ui d Engi neeri ng Seri es; Vol . 12.) BHRA, Cranfi el d,
Bedford, Engl and
Skal ak, R. and S. Chi en (edi tors). 1987. Handbook of Bi oengi neeri ng.
McGraw-Hi l l Book Company, New York.
Skel l and, A.P.H. 1967. Non-Newtoni an Fl ow and Heat Transfer. John
Wi l ey and Sons, I nc. New York.
Skel l and, A.P.H. 1983. Mi xi ng and agi tati on of non-Newtoni an fl ui ds.
I n: Cheremi si noff, N.P. and R. Gupta (edi tors). Handbook of Fl ui ds
i n Moti on. Ann Arbor Sci ence Publ i shers, Ann Arbor, Mi chi gan. pg.
179-209.
Sone, T. 1972. Consi stency of Foodstuffs. D. Rei del Publ i shi ng Co.,
Dordrecht, Hol l and.
Steffe, J.F. 1984. Probl ems i n usi ng apparent vi scosi ty to sel ect pumps
for pseudopl asti c fl ui ds. Trans. ASAE 27: 629-634.
Steffe, J.F. 1992. Yi el d stress: phenomena and measurement. I n: Si ngh,
R.P. and A. Wi rakartakusumah (edi tors). Advances i n Food Engi -
neeri ng. CRC Press, Boca Raton, FL. pg. 363-376.
Steffe, J.F., E.M. Castel l -Perez, K.J. Rose and M.E. Zabi k. 1989. Rapi d
testi ng method for characteri zi ng the rheol ogi cal behavi or of gel a-
ti ni zi ng corn starch sl urri es. Cereal Chem. 66: 65-68.
Steffe, J.F. and E.W. Ford. 1985. Rheol ogi cal techni ques to eval uate the
shel f-stabi l i ty of starch-thi ckened, strai ned apri cots. J. Texture Stud.
16: 179-192.
Steffe, J.F., I .O. Mohamed and E.W. Ford. 1986. Rheol ogi cal properti es
of fl ui d foods: data compi l ati on. I n: Okos, M.R. (edi tor). Physi cal and
Chemi cal Properti es of Foods. ASAE. St. Joseph, Mi chi gan. pg. 1-13.
Steffe, J.F., I .O. Mohamed and E.W. Ford. 1984. Pressure drop across
val ves and fi tti ngs for pseudopl asti c fl ui ds i n l ami nar fl ow. Trans.
ASAE 27: 616-619.
Steffe, J.F. and R.G. Morgan. 1986. Pi pel i ne desi gn and pump sel ecti on
for non-Newtoni an fl ui d foods. Food Technol . 40: 78-85. [Addendum:
Food Technol . 41: 32].
Steffe, J.F. and F.A. Osori o. 1987. Back extrusi on of non-Newtoni an
fl ui ds. Food Technol . 41: 72-77.
Steffe, J.F., F.A. Osori o and R.Y. Ofol i . 1989. Dynami c and steady
rheol ogi cal properti es of starch based pi e fi l l i ng. I n: Ang, H.G. (edi -
409
tor). Trends i n Food Sci ence. Part Si x, Rheol ogi cal Studi es," pg.
128-130. The Si ngapore I nst. Food Sci . and Technol ., Republ i c of
Si ngapore.
Sweeney, E.T. 1978. An I ntroducti on and Li terature Gui de to Mi xi ng.
- (Bri ti sh Hyrdomechani cs Research Associ ati on. Fl ui d Engi neeri ng.
Fl ui d Engi neeri ngSeri es; Vol . 5.)BHRA, Cranfi el d, Bedford, Engl and
Szczesnai k, A.S. 1995. Texture profi l e anal ysi s-methodol ogy i nter-
pretati on cl ari fi ed. J. Food Sci . (Letters) 60: vi i .
Szczesnai k, A.S., M.A. Brandt and H.H. Fri edman. 1963. Devel opment
of standard rati ng scal es for mechani cal parameters of texture and
correl ati on between the objecti ve and sensory methods of texture
eval uati on.. J. Food Sci . 28: 397-403.
Tadmor, Z. and C.G. Gogos. 1979. Pri nci pl es of Pol ymer Processi ng.
John Wi l ey and Sons, New York.
Tamura, M.S., J.M. Henderson, R.L. Powel l and C.F. Shoemaker. 1989.
Eval uati on of the hel i cal screw rheometer as an on-l i ne vi scometer.
J. Food Sci . 54: 483-484.
Tanaka, M., J.M. De Man and P.W. Voi sey. 1971. Measurement of
textural properti es of foods wi th constant speed cone penetrometer.
Tanner, R.I . 1985. Engi neeri ng Rheol ogy. Oxford Uni versi ty Press,
Oxford.
Tanner, R.I . 1988. Recoverabl e el asti c strai n and swel l i ng rati o. I n:
Col l yer, A.A. and D.W. Cl egg (edi tors). Rheol ogi cal Measurements.
El sevi er Appl i ed Sci ence, New York. pg. 93-118.
Tattersal l , G.H. and P.F.G. Banfi l l . 1983. The Rheol ogy of Fresh
Concrete. Pi tman Publ i shi ng I nc., Boston.
Ti rtaatmadja, V. and T. Sri dhar. 1993. A fi l ament stretchi ng devi ce for
measurement of extensi onal vi scosi ty. J. Rheol . 37: 1081-1102.
Ti u, C. and D.V. Boger. 1974. Compl ete rheol ogi cal characteri zati on of
ti me-dependent food products. J. Texture Stud. 5: 329-338.
Tol edo, R., J. Cabot and D. Brown. 1977. Rel ati onshi p between com-
posi ti on, stabi l i ty and rheol ogi cal properti es of raw commi nuted meat
batters. J. Food Sci . 42: 725-727.
Toms, B.A. 1958. Fundamental techni ques: fl ui ds. I n: Ei ri ch, F.R.
(edi tor). Rheol ogy: Theory and Appl i cati ons, Vol ume I I . Academi c
Press, New York. pg. 475-500.
Trouton, F.T. 1906. On the coeffi ci ent of vi scous tracti on and i ts rel ati on
tothat of vi scosi ty. Proceedi ngs of the Royal Soci ety of London. Seri es
A, Vol . LXXVI I : 426-440.
Uhl , V.W. and J.B. Gray (edi tors). 1986. Mi xi ng, Theory and Practi ce,
Vol . I I I . Academi c Press, I nc., New York.
Ul brecht, J.J. and P. Carreau. 1985. Mi xi ng of vi scous non-Newtoni an
l i qui ds. I n: Ul brecht, J.J. and G.K. Patterson (edi tors). Mi xi ng of
Li qui ds by Mechani cal Agi tati on. Gordon and Breach Sci ence Pub-
l i shers, New York. pg. 93-138.
410 Bibliography
Ul brecht, J.J. and G.K. Patterson (edi tors). 1985. Mi xi ng of Li qui ds by
Mechani cal Agi tati on. Gordon and Breach Sci ence Publ i shers, New
York.
Ul bri cht, D. M.D. Normand and M. Pel eg. 1995. Creati ng typi cal jagged
force-deformati on rel ati onshi ps from the i rregul ar and i rreproduci bl e
compressi on data of crunchy foods. J. Sci . Food Agri c. 67: 453-459.
Ury, J.F. 1966. Resi stance tosteady l ami nar fl ow of pseudopl asti c fl ui ds
through pi pes and val ves. J. Mech. Engn. Sci . 8: 226-233.
Van Wazer, J.R. 1963. Vi scosi ty and Fl ow Measurement. I ntersci ence
Publ i shers, New York.
Vercruysse, M.C.M. and J.F. Steffe. 1989. On-l i ne vi scometry for pureed
baby food: correl ati on of Bostwi ck Consi stometer readi ngs and
apparent vi scosi ty data. J. Food Proc. Engn. 11: 193-202.
Vocadl o, J.J. and M.E. Charl es. 1971. Measurement of yi el d stress of
fl ui d-l i ke vi scopl asti c substances. Canadi an J. Chem. Engn. 49:
576-582.
Vi swanath, D.S. and G. Natarajan. 1989. Data Book on the Vi scosi ty of
Li qui ds. Hemi sphere Publ i shi ng Corp., New York.
Vi tal i , A.A. and M.A. Rao. 1984. Fl ow properti es of l ow pul p concentrated
orange jui ce: effect of temperature and concentrati on. J. Food Sci .
49: 882-888.
Vyal ov, S.S. 1986. Rheol ogi cal Fundamental s of Soi l Mechani cs.
El sevi er Sci ence Publ i shers, New York.
Wal ters, K. 1975. Rheometry. Chapman and Hal l , London.
Warren, R.C. 1988. Vi scous heati ng. I n: Col l yer, A.A. and D.W. Cl egg
(edi tors). Rheol ogi cal Measurements. El sevi er Appl i ed Sci ence, New
York. pg. 119-150.
Weast, R.C., M.J. Astl e and W.H. Beyer. 1985. CRC Handbook of
Chemi stry and Physi cs. CRC Press, I nc.. Boca Raton, FL
Wei pert, D., H.-D. Tscheuschner and E. Wi ndhab (edi tors). 1993.
Rheol ogi e der Lebensmi ttel . B. Behrs Verl ag GmbH &Co., Hamberg,
Germany.
Whorl ow, R.W. 1992. Rheol ogi cal Techni ques (Second Edi ti on). El l i s
Horwood, New York.
Wi kstrm, K., L. Bohl i n and A. El i asson. 1994. Bi axi al extensi on of
wheat fl our doughs. Rheol ogy 94, 4: 192-197.
Wi l l i ams, M.L., R.F. Landel and J.D. Ferry. 1955. The temperature
dependence of rel axati on mechani sms i n amorphous pol ymers and
other gl ass formi ng l i qui ds. J. Am. Chem Soc. 77: 3701-3707.
Wi l l i ams, R.W. 1979. Determi nati on of vi scometri c data from the
Brookfi el d R.V.T. vi scometer. Rheol . Acta 18: 345-359.
Wi nther, G., K. Al mdal and O. Kramer. 1991. Determi nati on of the
pol ymer mel t vi scosi ty by squeezi ng fl ow wi th constant pl ate vi s-
cosi ty. J. Non-Newtoni an Fl ui d Mech. 39: 119-136.
Wood, F.W. and T.C. Goff. 1973. The determi nati on of the effecti ve
shear rate i n the Brabender Vi scograph and i n other systems of
compl ex geometry. Di e Starke 25: 89-91.
411
Yang, T.M.T. and I .M. Kri eger. 1978. Compari son of methods for cal -
cul ati ng shear rates i n coaxi al vi scometers. J. Rheol . 22: 413-421.
Yanovsky, Y.G. 1993. Pol ymer Rheol ogy: Theory and Practi ce. Chap-
man and Hal l , New York.
Yoo, B., M.A. Rao and J.F. Steffe. 1995. Yi el d stress of food di spersi ons
wi th the vane method at control l ed shear rate and shear stress. J.
Yoshi mura, A.S. and R.K. Prudhomme. 1987. Response of and el asti c
Bi ngham fl ui d to osci l l atory shear. Rheol . Acta 26: 428-436.
Yoshi mura, A.S. and R.K. Prudhomme. 1988. Wal l sl i p correcti ons for
couette and paral l el di sk vi scometers. J. Rheol . 32: 53-67.
Yoshi mura, A.S. R.K. Prudhomme, H.M. Pri ncen and A.D. Ki ss. 1987.
A compari son of techni ques for measuri ng yi el d stress. J. Rheol . 31:
699-710.
Zukoski , C.F. 1993. Materi al properti es and the el ectrorheol ogi cal
response. Annu. Rev. Mater. Sci . 23: 45-78.
Index
Activation energy, 33, 85
Adams Consistometer, 67
Adhesiveness, 75
Agar gel, 330
Alveograph, 66
Amplitude ratio, 314
Anelasticity, 8
Angle of shear, 5
Apparent viscosity, 24
Apparent wall shear rate, 100
Apple tissue, 340, 343
Apricots, 240, 250
Arrhenius equation, 33, 85
Average shear rate, 165, 216
Average shear stress, 215
Bagley plot, 112
Biaxial extension, 45
Bingham plastic fluid
definition, 20
in concentric cylinder viscometer,
163
in mixer viscometer, 199
minimum bob speed, 210
tube flow rate, 103
tube velocity profile, 105
Bioyield point, 71
Bird-Leider equation, 310
Bostwick Consistometer, 67
Boussinesq number, 53
Brabender-FMC Consistometer, 70
Brinkman number, 183
Brookfield Viscometer, 69
Bubble growth, 40
Buckingham pi theorem, 186
Buckingham-Reiner equation, 103
Bulk compression, 294
Bulk modulus, 11
Bulk viscosity, 53
Burgers model, 50, 307, 343
Butter, 258
Calendering, 40
Cannon-Fenske viscometer, 3, 94
Cantilever beam, 10
Capillary viscometer, 94
errors in operating, 110
Carrageenan gum solution, 77
Casson equation, 24
with milk chocolate, 82
Casson fluid, 149
Cauchy strain, 4
Cavitation, 182, 236
Cessation of steady shear flow, 48,
295
Cessation of steady tensile extension,
297
Characteristic time
of material, 333
of process, 333
Cheese spread, 283
Chemorheology, 321
Chewiness, 75
CMC solution, 146
Coating, 91
Coefficient of viscosity, 19
Cogswells equations, 264, 287
Cohesiveiness, 75
Complex compliance, 316
Complex modulus, 49, 315
Complex viscosity, 49, 315
Compression-extrusion cell, 70, 71
Compressive stress, 6
Concentric cylinder viscometry, 158
Bingham plastic fluids, 162
cavitation, 236
end correction, 174
Krieger method, 165
Newtonian approximation, 164
Newtonian fluids, 161
413
power law approximation, 165
power law fluids, 162
secondary flow, 182
shear rate calculations, 163
simple shear approximation, 164
viscous heating, 177
wall effects (slip), 181
Cone and plate viscometry, 3, 169,
226, 227
truncated cone, 183
Cone penetrometer, 66
Conservation of momentum equa-
tions, 141
Constitutive equation, 1, 7
Converging die, 263, 287, 289
Corn starch, 79
Corn syrup, 150
Couette system, 158
Cox-Merz rule, 48, 338
Creep, 48, 295, 304
Creep compliance, 48, 305, 343
Critical radius
from bob, 225
tube flow, 105
Cross equation, 24
Darcy friction factor, 129
Data corrections
concentric cylinder viscometer, 174
cone and plate viscometer, 182
parallel plate viscometer, 182
tube viscometer, 110
Deborah number, 12, 332
Dextrin solution, 329
Die exit effect integral, 269
Die swell
see jet expansion
Dilatent behavior, 20
Dimensional analysis, 186
DIN standard, 176, 214
Disk surface viscometer, 54
Dough testing equipment, 65
Dynamic viscosity, 19, 49, 316, 348
Edge failure, 183
Elastic behavior, 49
Elastic solids, 8
Elastoplastic material, 11
Electrorheology, 55
Ellis equation, 24
Elongational viscosity, 39
End correction
Energy of activation, 33, 85
Entrance length in tube, 114
Equation of state, 7
Extensigraph, 65
Extensional-thickening, 45
Extensional-thinning, 45
Extensional flow, 39
biaxial, 43, 258, 283
planar, 43
tensile, 43
uniaxial, 255
Extensional viscosity, 39
biaxial, 45
planar, 46
tensile, 45
Extrudate diameter, 122
Extrudate drawing, 40, 274
Extrudate swell
see jet expansion
Falling ball viscometer, 69, 82
Fanning friction factor, 128, 131
Fano flow, 276
Farinograph, 65
Fiber spinning, 274
414 Index
Finite bob in infinite cup, 168
First normal stress difference
definition, 16
from cone and plate data, 171
in jet expansion, 122
Flexural testing, 9
Fluidity, 19
FMC Consistometer, 70
Fracturability, 75
Frequency sweep, 320
Friction loss coefficient, 133, 135,
136
Froude number, 189
Gibsons equations, 268, 289
Glass capillary viscometer, 2, 94, 125
kinetic energy correction, 127
Glass transition temperature, 34
Gleisselss mirror relation, 48
Gumminess, 75
Hardness, 75
Hedstrom number, 109
Helical screw rheometer, 185
Helipath Stand, 69
Hencky strain, 5
Herschel-Bulkley fluid, 20
tube flow rate, 103
Hoeppler Viscometer, 70
Hole pressure error, 120
Honey, 82
Hookes law, 8, 298
Hookean behavior, 49
Hystersis loop, 29
Impeller Reynolds number, 188
Inelastic fluids, 19
Infinite cup solution, 168, 221, 223,
225
Inherent viscosity, 27
Interfacial rheology, 53
Interfacial viscometer, 54
Intrinsic viscosity, 27
Jet expansion, 16, 47, 121
Kelvin model, 50, 298, 306
Kinematic viscosity, 19, 127
Kinetic energy correction, 110, 127
Kinetic energy correction factor, 131,
133
Kramer Shear Cell, 70, 71
Krieger method, 165
Lambda carrageenan solution, 330
Laminar flow criteria
Launs rule, 48
Leibnitz rule, 99, 173
Limiting viscosity
at infinite shear, 23
at zero shear, 23
Linear elastic material, 11
Linear viscoelastic behavior, 47, 49,
318
Linear viscoelasticity, 296
Loss compliance, 316
Loss modulus, 49, 315, 348
Lubricated squeezing flow, 277
Magnetorheological fluids, 55
Margules equation, 162
Matching viscosity method, 192
Maxwell model, 50, 298, 318, 341
Mechanical analogues, 298
Mechanical energy balance, 128, 152
Melt flow index, 77
Melt Flow Indexer, 76
Melt fracture, 16
Melt spinning, 274
Milk chocolate, 81
Mixer viscometer, 3
Mixer viscometer constant
definition, 190
415
matching viscosity method, 193,
199
slope method, 191
variables influencing, 194
Mixer viscometry, 185
average shear rate, 190
Bingham plastic fluids, 199
matching viscosity method, 192,
194
power law fluids, 190, 195, 237
rheomalaxis, 208
slope method, 191, 194
Mixing
commercial, 185
impellers, 186
power consumption, 187
Mixograph, 65
Modeling rheological behavior, 32
concentration effects, 34
shear effects, 86
temperature effects, 34, 86
Modulus of elasticity
see Youngs modulus
Mooney-Couette bob, 176, 233
Nahme number, 119
Newtonian behavior, 49
Newtonian fluid
definition, 19
162, 213
in glass capillary viscometer, 125
in parallel plate viscometer, 173
in slit viscometer, 122
tube flow rate, 101
turbulent pipe flow, 138, 155
Non-linear elastic material, 11
Non-linear viscoelasticity, 296
Normal stress coefficient, 15
Normal stress difference, 16
On-line viscometer
capillary flow type, 58
concentric cylinder type, 59
falling piston type, 62
off-set rotating sensor, 60
vibrating rod type, 60
vibrating sphere type, 61
Opposing jets, 272
Orange juice concentrate, 86
Orifice viscometer, 68
Oscillatory testing, 48, 336
application of stress and strain, 313
frequency sweep, 321
operating modes, 318
strain sweep, 320
temperature changes, 322
time sweep, 322
typical data, 324
Ostwald viscometer, 94
Parallel plate plastometer, 280
Parallel plate viscometer, 3
Parallel plate viscometry, 172, 229
Peanut butter, 291
Phase shift, 314
Pipe viscometer, 3, 94, 96
Pipeline design calculations, 128
Planar extension, 45
Poiseuille-Hagen equation, 101, 125
Poissons ratio, 9
Powell-Eyring equation, 24
Power law fluid, 20
162, 212, 216
in cone and plate viscometer, 171
in mixer viscometer, 195
in parallel plate viscometer, 173
tube flow rate, 102
turbulent pipe flow, 140, 156
Power number, 188
Pseudoplastic behavior, 20
416 Index
Rabinowitsch-Mooney equation, 97,
100, 145, 147
Rapid Visco Analyser, 68
Recoil, 17, 47, 48, 295, 305
Recoverable shear, 122
Reduced viscosity, 27
Reiner-Philippoff equation, 24
Reiner-Riwlin equation, 163
Relative viscosity, 27
Relaxation time, 301, 341
Representative shear rate, 165, 216
Representative shear stress, 215
Reynolds number
Bingham, 108
Newtonian, 107
power law, 107
Rheodestruction, 28
Rheogoniometer, 18
Rheogram, 19
Rheological equation of state, 1, 7
Rheological instruments, 2
Rheomalaxis, 28
evaluation by mixer viscometry,
208, 250
Rheopectic material, 28
Rising bubble viscometer, 69
Rod climbing, 47
Rolling ball viscometer, 70
Rupture point, 71
Sagging, 40
Salad dressing, 223, 227
Scott equation, 282
Searle system, 158
Second normal stress difference
definition, 16
from parallel plate data, 174
Secondary flow, 182
Shear-thickening, 20
Shear-thinning, 20, 23
Shear modulus, 8, 300
Shear rate
definition, 13
in mixing, 51
in spreading or brushing, 51
tube flow, 52
typical values, 15
Shear stress, 6, 13
Sheet stretching, 40
Silicone polymer, 335
Silly putty, 334
Simple compression, 71
Sink flow analysis, 263
Skim milk curd, 343
Slip
oscillatory testing, 336
Slit viscometry, 122, 150
Slope method, 191
Small amplitude oscillatory testing,
49
Sodium carboxymethylcellulose solu-
tion, 146
Sol-gel transition, 28
Solid behavior
elastoplastic, 12
linear elastic, 12
non-linear elastic, 12
Solution viscosities, 27
Soy dough, 143, 287, 289
Specific viscosity, 27
Spinning, 274
Spreadability, 66
Springiness, 75
Squeeze film viscometer, 280
Squeezing flow, 276
lubricated, 277, 291
nonlubricated, 279
Start-up flow, 48, 295, 310
Steady shear flow, 13
Stefan equation, 281
Step strain, 48, 295, 299
Stokes law, 82
417
Storage compliance, 316
Storage modulus, 49, 315, 348
Strain, 4, 313
axial, 9
Cauchy, 4
concentric cylinders, 324
cone and plate, 324
engineering, 4
Hencky, 5
lateral, 9
rotational, parallel plates, 321
true, 5
volumetric, 11
Strain sweep, 318
Stress, 4, 313
compressive, 7
tensile, 7
Stress overshoot, 310
mayonnaise, 311
Stress relaxation, 48, 299, 342
Stress relaxation modulus, 48, 299
Stress sweep, 318
Stress tensor, 14
Stringiness, 75
Structural parameter, 30
Surface loading, 4
T-bars, 69
Tan delta, 316, 348
Taylor vortices, 182
Tensile creep, 297
Tensile recoil, 297
Tensile start-up, 297
Tensile step strain, 297
Tensile stress, 6
Tension-thickening, 45
Tension-thinning, 45
Texture Profile Analysis, 72, 73
Texture profile curve, 74
Thixotropic material, 28
structure, 37
Three-point bending, 10
Time-dependent functions, 27, 50
Time-dependent thickening, 28
Time-dependent thinning, 28
Time-independent functions, 13, 49
Time sweep, 321
Tomato ketchup, 218, 226, 235
Torsional flow
see parallel plate viscometry
Trouton number, 47, 288, 291
Trouton viscosity, 39
Truncated cone, 183
Tubeless siphon, 16, 47, 276
Turbulent pipe flow, 138
Uniaxial compression, 8
Uniaxial extension, 43
Unlubricated squeezing flow, 279
Unsteady shear testing
oscillatory, 294
transient, 294
U-tube viscometer, 94
Vane method
see yield stress
Velocity profile in tube flow
Bingham plastic fluid, 105
Herschel-Bulkley fluid, 106
Newtonian fluid, 103
power law fluid, 104
turbulent flow, 138
Vibrational viscometer, 4
Visco-Amylograph, 68
Viscoelastic behavior
mechanical analogues, 298
Viscoelastic functions, 338
Viscometric flow, 15
Viscometric functions, 13, 15, 338
Viscosity
apparent, 24
extensional, 39
inherent, 26
intrinsic, 26
418 Index
limiting at infinite shear, 23
limiting at zero shear, 23, 301
reduced, 26
relative, 26
solution, 26
specific, 26
Viscous behavior, 49
Viscous heating
concentric cylinder viscometer, 177,
235
tube flow, 118
Volume loaded viscometer, 4
Wall effects
see slip
Warner-Bratzler Shear, 67
Weber number, 189
Weissenberg number, 189
Weissenberg effect, 16, 47
Williams-Landel-Ferry equation, 34
Wind-up characteristics, 207, 250
Yield number, 38
Yield stress
definition, 35
dynamic, 37, 202
in Bingham plastic fluid, 21
in Herschel-Bulkley fluid, 21
measurement methods, 36
oscillatory testing, 336
static, 37, 202
vane method, 200, 243, 244, 247
Youngs modulus, 9
Zahn viscometer, 68

Rheological Methods in Food Process Engineering (2nd Edition)

Hochgeladen von

Dokumentinformationen

Originalbeschreibung:

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

Rheological Methods in Food Process Engineering (2nd Edition)

Hochgeladen von

Copyright:

Verfügbare Formate

RHEOLOGICAL METHODS IN

FOOD PROCESS ENGINEERING

6 Chapter 1. Introduction to Rheology

10 Chapter 1. Introduction to Rheology

1.5.1 Time-Independent Material Functions 17

1.5.3 Modeling Rheological Behavior of Fluids 33

38 Chapter 1. Introduction to Rheology

50 Chapter 1. Introduction to Rheology

56 Chapter 1. Introduction to Rheology

1.13 Empirical Measurement Methods for Foods 69

0 50 100 150 200

1.14.2 Concentrated Corn Starch Solution 79

80 Chapter 1. Introduction to Rheology

Shear Rate, 1/s

82 Chapter 1. Introduction to Rheology

86 Chapter 1. Introduction to Rheology

1.14.5 Orange Juice Concentrate 87

88 Chapter 1. Introduction to Rheology

90 Chapter 1. Introduction to Rheology

2.2 Rabinowitsch-Mooney Equation 101

112 Chapter 2. Tube Viscometry

2.5 Data Corrections 119

120 Chapter 2. Tube Viscometry

1.960 + 0.815n 1.628n log

1.960 + 0.815(.45) 1.628(.45) log

3.2.1 Derivation of the Basic Equation 159

3.2.1 Derivation of the Basic Equation 161

166 Chapter 3. Rotational Viscometry

3.5 Corrections: Concentric Cylinder 177

3.7.3 Yield Stress Calculation: Vane Method 207

210 Chapter 3. Rotational Viscometry

3.8.3 Newtonian Fluid in a Concentric Cylinder 213

218 Chapter 3. Rotational Viscometry

3.8.6 Concentric Cylinder Data - Tomato Ketchup 219

3.8.10 Infinite Cup - Yield Stress Materials 225

3.8.13 Parallel Plate - Methylcellulose Solution 229

3.8.14 End Effect Calculation for a Cylindrical Bob 231

2(101.35 7.38) (1000)

3.8.21 Vane Rotation in Yield Stress Calculation 249

1 1.5 2 2.5 3 3.5

3.8.22 Rheomalaxis from Mixer Viscometer Data 253

0 100 200 300 400 500 600

Chapt e r 4 . Ext e ns ional Flow

4.9.1 Biaxial Extension of Processed Cheese Spread 285

286 Chapter 4. Extensional Flow

312 Chapter 5. Viscoelasticity

326 Chapter 5. Viscoelasticity

328 Chapter 5. Viscoelasticity

5.4 Typical Oscillatory Data 329

5.5 Deborah Number 333

260, 000 + 100

200 + 80, 000

5.7 Viscometric and Linear Viscoelastic Functions 339

lnK = 0.411 n = 0.60

lnK = 1.185 n = 0.74

Das könnte Ihnen auch gefallen