Sie sind auf Seite 1von 241

Disposal and Recycling of Organic and Polymeric Construction

Materials

Disposal and Recycling of Organic and


Polymeric Construction Materials
Proceedings of the International RILEM Workshop
Tokyo
2628 March 1995

EDITED BY

Y.Ohama
Department of Architecture,
College of Engineering,
Nihon University, Koriyama,
Japan

E & FN SPON
An Imprint of Chapman & Hall
London Glasgow Weinheim New York Tokyo Melbourne Madras

Published by E & FN Spon, an imprint of Chapman & Hall,


26 Boundary Row, London SE1 8HN, UK
This edition published in the Taylor & Francis e-Library, 2005.
To purchase your own copy of this or any of Taylor & Francis or Routledges collection of thousands of eBooks please go to
www.eBookstore.tandf.co.uk.
Chapman & Hall, 26 Boundary Row, London SE1 8HN, UK
Blackie Academic & Professional, Wester Cleddens Road, Bishopbriggs,
Glasgow G64 2NZ, UK
Chapman & Hall GmbH, Pappelallee 3, 69469 Weinheim, Germany
Chapman & Hall USA, One Penn Plaza, 41st Floor, New York NY10119,
USA
Chapman & Hall Japan, ITP-Japan, Kyowa Building, 3F,
221 Hirakawacho, Chiyoda-ku, Tokyo 102, Japan
Chapman & Hall Australia, Thomas Nelson Australia, 102 Dodds Street,
South Melbourne, Victoria 3205, Australia
Chapman & Hall India, R.Seshadri, 32 Second Main Road, CIT East,
Madras 600035, India
First edition 1995
1995 RILEM
ISBN 0-203-62728-8 Master e-book ISBN

ISBN 0-203-63113-7 (Adobe eReader Format)


ISBN 0-419-20550-0 (Print Edition)
Apart from any fair dealing for the purposes of research or private study,
or criticism or review, as permitted under the UK Copyright Designs and
Patents Act, 1988, this publication may not be reproduced, stored, or
transmitted, in any form or by any means, without the prior permission in
writing of the publishers, or in the case of reprographic reproduction only
in accordance with the terms of the licences issued by the Copyright
Licensing Agency in the UK, or in accordance with the terms of licences
issued by the appropriate Reproduction Rights Organization outside the
UK. Enquiries concerning reproduction outside the terms stated here
should be sent to the publishers at the London address printed on this age.
The publisher makes no representation, express or implied, with regard
to the accuracy of the information contained in this book and cannot
accept any legal responsibility or liability for any errors or omissions that
may be made.
A catalogue record for this book is available from the British Library
Publisher's Note
This book has been produced from camera ready copy provided by the
individual contributors in order to make the book available for the
Workshop.

Contents

Workshop organization
Preface
PART ONE

PLASTICS-BASED MATERIALS

vi
viii
1

Mechanical properties of polymer mortar made from recycled PET-based


unsaturated polyester resin
Y.S.SOH, H.S.PARK and D.S.LEE

Properties of plain and reinforced polyester concretes made with recycled PET
K.S.REBEIZ and D.W.FOWLER

A new kind of hybrid recycled polymer mortar


Y.BAO, D.P.WHITNEY and D.W.FOWLER

22

Utilization of waste plastics as aggregate in asphalt mixture


M.YAMADA

33

The behavior of Portland cement concrete with the incorporation of waste plastic
fillers
D.SANDER, D.W.FOWLER and R.L.CARRASQUILLO

43

Polymer granulates for masonry mortars and outdoor plaster


H.R.SASSE, O.LEHMKMPER and R.KWASNY-ECHTERHAGEN

54

Polymer modified lightweight cement mortar using plastics waste


Y.HAYASHI, R.NANIWA, H.IIBACHI, K.HADA and T.YAMAZAKI

62

PART TWO

FRP-BASED MATERIALS

71

Updating recycling technologies for thermoset composites in Japan


T.KITAMURA

72

Environment-conscious materials design of lightweight precast concrete


components with recyclable FRP rebars
T.FUKUSHIMA, K.YANAGI and T.MAEDA

81

10

Recycling of plastics wastes from electronic parts production processes


M.IJI and S.YOKOYAMA

91

11

Preparation and properties of lightweight high-strength mortars containing FRP


fine powder as aggregate
A.KOJIMA and S.FURUKAWA

99

12

Properties of autoclaved cement paste containing scrap FRP powder


M.WAKASUGI and A.SUGIURA

110

13

Recycling of FRP as a cementitious composite


K.YAMADA and H.MIHASHI

116

14

Properties of artificial woods using FRP powder


K.DEMURA, Y.OHAMA and T.SATOH

127

PART THREE RUBBER-BASED MATERIALS

136

15

Noise abatement by panels of recycled vehicle tires


M.MOTAVALLI, M.FARSHAD and P.FLELER

137

16

Construction materials using powdered rubber made of vehicle tires


R.YAMAMOTO

143

17

Punching resistance of mats made of recycled PVC and rubber in underground


construction
P.FLELER, M.FARSHAD and A.ROLLER

150

PART FOUR WOOD-BASED MATERIALS

156

18

Newly developed wood-chip concrete with recycled timbers


Y.KASAI, M.KAWAMURA, J.D.ZHOU and K.MACHIDA

157

19

Particleboards made from recycled wood


S.SUZUKI

168

20

Development of formwork material made of scrap lumber


T.ONO, S.ONO, T.KUMANO, T.SANO and Y.MUKAWA

177

PART FIVE

PAPER-BASED MATERIALS

185

21

Ecology boards using recycled paper resources for concrete forms


K.KURIHARA, S.TAKATA, Y.TOMIMURA and S.HOSOYA

186

22

Recycling of used paper as a building material


H.MIHASHI, K.KIRIKOSHI, S.ARIKAWA, T.YAMAMOTO and T.NARITA

194

23

Use of paper sludge ash in concrete products


Y.S.SOH, S.Y.SOH and D.S.LEE

203

PART SIX

OTHERS

209

24

Investigation on deterioration of recycled hot-mixed asphalt concrete pavement and


a trial re-recycling of asphalt concrete
T.YOSHIKANE

210

25

Reuse of carpet industrial waste for concrete reinforcement


Y.WANG

222

Author index

230

Subject index

231

Workshop organization

Sponsoring Organization
International Union of Testing and Research Laboratories for Materials and Structures (RILEM)
Architectural Institute of Japan (AIJ)
Japan Technology Transfer Association (JTTAS)
International Advisory Committee
Professor. K.Kamimura
Professor A.M.Brandt
Professor D.W.Fowler
Dr. H.W.Fritz
Dr. T.Kawano
Professor K.Kishitani
Professor W.Koyanagi
Professor S.Nagataki
Professor K.Okada
Dr. S.Okamoto
Dr. A.M.Paillere
Professor H.W.Reinhardt
Professor F.Sandrolini
Professor H.R.Sasse
Professor R.N.Swamy

Utsunomiya University, Japan (Chairman)


Polish Academy of Sciences, Poland
University of Texas at Austin, U.S.A.
Eidgenssische Materialprfungs-und Forschungsanstalt, Switzerland
Maeta Concrete Industry Ltd., Japan
Nihon University, Japan
Gifu University, Japan
Tokyo Institute of Technology, Japan
Fukuyama University, Japan
Building Research Institute, Japan
Laboratoire Central des Ponts et Chaussees, France
Universitt Stuttgart, Germany
Universita di Bologna, Italy
Rheinisch-Westflische Technische Hochschule Aachen, Germany
University of Sheffield, United Kingdom

Organizing Committee
Professor Y.Ohama
Professor H.Mihashi
Dr. K.Demura
Professor T.Arima
Dr. T.Fukushima
Professor Y.Kasai
Mr. A.Kawamura
Professor A.Kojima
Mr. S.Kurihara
Professor A.Moriyoshi
Mr. N.Nishiyama
Professor F.Oishi
Dr. M.Sawaide
Dr. A.Shirai
Mr. M.Wakasugi

Nihon University (Chairman)


Tohoku University (Secretary)
Nihon University (Secretary)
University of Tokyo
Building Research Institute
Nihon University
Kumagaigumi Co., Ltd
Gunma College of Technology
Japan Reinforced Plastics Society
Hokkaido University
Nishimatsu Construction Co., Ltd
Kanagawa University
Shimizu Corporation
Tokyo Kasei Gakuin University
Sumitomo Osaka Cement Co., Ltd

vii

Professor M.Yamada

Osaka City University

RILEM Subcommittee of Architectural Institute of Japan (AIJ)


Professor Y.Ohama
Professor H.Mihashi
Professor T.Arima
Dr. K.Demura
Dr. T.Fukushima
Mr. T.Kaminosono
Professor Y.Kasai
Professor T.Kuwahara
Professor R.Naniwa
Professor T.Soshiroda
Dr. H.Tamura
Mr. K.Tobinai
Professor F.Tomosawa
Mr. K.Yanagi

Nihon University (Chairman)


Tohoku University (Secretary)
University of Tokyo
Nihon University
Building Research Institute
Building Research Institute
Nihon University
Hokkaigakuen University
Kogakuin University
Shibaura Institute of Technology
General Building Research Corporation of Japan
Mitsubishi Materials Corporation
University of Tokyo
Japan Testing Center for Construction Materials

Preface

In recent years, various polymers have been widely used as construction materials, and the disposal and recycling of
organic (polymeric) construction materials has become a serious problem in the construction industry. The
construction industry is one of the major consumers of the polymers, and is considered to be a significant potential
customer for recycled polymers. Accordingly, there is a pressing need for the construction industry to develop
ecologically safe disposal systems and effective recycling systems for the organic (polymeric) construction
materials.
Against such a background, the RILEM Workshop on Disposal and Recycling of Organic (Polymeric)
Construction Materials is to be held in Tokyo, Japan on 2628 March 1995, and is co-sponsored by the International
Union of Testing and Research Laboratories for Materials and Structures (RILEM), the Architectural Institute of
Japan (AIJ) and Japan Technology Transfer Association (JTTAS) under the auspices of the American Concrete
Institute (ACI), the American Society for Testing and Materials (ASTM), the Gypsum Board Association of Japan,
the Japan Cement Association, the Japan Concrete Institute, the Japan Fiberboard and Particleboard Manufacturers
Association, the Japan Housing and Wood Technology Center, the Materials Research Society (USA), the Materials
Research Society of Japan, the Plastic Waste Management Institute (Japan), the Slate Association of Japan, the Japan
Reinforced Plastics Society, the Japan Society of Civil Engineers, the Japan Society of Waste Management Experts,
the Japan Wood Research Society, the Society of Gypsum & Lime (Japan), the Society of Materials Science, Japan,
the Society of Polymer Science, Japan, and the Society of Rubber Industry, Japan. Financial support was provided by
the TOSTEM Foundation for Construction Materials Industry Promotion, AIJ and JTTAS.
The main objectives of this Workshop are to collect recent information about the disposal and recycling of organic
(polymeric) construction materials, and develop new ideas to further improve the ecologically safe disposal systems
and effective recycling systems for organic (polymeric) construction materials.
The main topics of the Workshop are as follows:
1. Disposal and recycling of organic (polymeric) construction materials using plastics, rubber, asphalt, wood and
paper.
2. Development of construction materials using waste organic (polymeric) materials from the other industries.
This Proceedings volume brings together the papers which will be presented at the Workshop. I believe that the
volume will be of interest for the manufacturers, users and researchers of organic (polymeric) construction materials.
On behalf of the Organizing Committee, I would like to thank all the authors of the papers included here for their
co-operation. I wish to acknowledge the national and international organizations or institutions which supported the
Workshop. I would also like to express my sincere appreciation to the members of the International Advisory
Committee for their useful advice and suggestions.
Yoshihiko Ohama
Koriyama, Japan
January 1995

PART ONE
PLASTICS-BASED MATERIALS

1
MECHANICAL PROPERTIES OF POLYMER MORTAR MADE
FROM RECYCLED PET-BASED UNSATURATED POLYESTER
RESIN
Y.S.SOH and H.S.PARK
Department of Architectural Engineering, College of Engineering, Chonbuk National University,
Chonju, Korea
D.S.LEE
Department of Chemical Technology, College of Engineering, Chonbuk National University,
Chonju, Korea

Abstract
Unsaturated polyester (UPE) resins made from recycled poly ethylene terephthalate (PET) were prepared and
the properties of the polymer mortar prepared with PET-modified UPE resin as well as those of the cured resin
itself were investigated. It was found that the degree of unsaturation of the resins was the most important
property affecting the thermal and mechanical properties. The cured unsaturated polyester resin or polymer
mortar made with resins of higher degree of unsaturation showed higher glass transition temperatures,
compressive strength, or higher flexural modulus and lower flexural strength, compared with those made with
resins of lower degree of unsaturation. Such properties were interpreted in terms of unsaturation, crosslink
density, and chain flexibility of the resin molecules.
Keywords: mechanical properties, polymer mortar, unsaturated polyester resin, recycled PET, degree of
unsaturation, glass transition temperature.
1
Introduction
Polymer concretes show excellent mechanical properties and chemical resistance compared with conventional
cement concretes. Polymer concretes can be cured quickly by the use of curing agents. Thus, the applications of
polymer concretes are being increased. One of the popular polymers for polymer concretes is unsaturated polyester
(UPE) resin. The properties of UPE resin can be modified by changing its molecular features. For the synthesis of the
resin, phthalic anhydride or isophthalic acid as well as maleic anhydride can be employed to modify the mechanical
properties or hydrothermal resistance. Terephthalic acid which is also used for the synthesis of poly ethylene
terephthalate (PET) enhances the thermal resistance of the cured UPE resin. However, the synthesis of unsaturated
polyester resin from terephthalic acid is difficult. One method to synthesize unsaturated polyester from terephthalic
acid is the use of recycled PET.
PET is useful polymer used for fiber, film, and plastic containers such as carbonated beverage bottles. Recently,
the recycling of polymers such as PET after use is attracting the attention of many researchers aware of
environmental problems and wishing to find ways to save earth resources. Previous studies showed that unsaturated
polyester resins can be economically prepared from recycled PET and the resins may be useful for resin concretes
(14). However, there is little information on the molecular features of the UPE resins. Thus, we synthesized various
UPE resins from PET. Especially, the PET content, chain flexibility and degree of unsaturation of the resins were
systematically varied and the effects of those variables on the mechanical properties of the cured resins and polymer
mortars made therefrom were studied.
Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

MECHANICAL PROPERTIES OF POLYMER MORTAR

2
Experimental program
Carbonated beverage bottles made of PET, high density polyethylene (HDPE) base cup, and labels were collected,
washed, and crushed into small fragments by using a crusher. The fragments of crushed PET bottles include PET,
HDPE, and various labels from which PET fragments can be easily separated by density difference in water. The
PET collected was dried in vacuum oven. Glycolyses of the PET were carried out using propylene glycol (PG) or
dipropylene glycol (DPG) at 200C for 8 hours. For the glycolyses of the PET, zinc acetate(0.05% by weight of the
PET) was added as a catalyst. Unsaturated polyester resins were prepared by condensation polymerization at 200C
using the products of glycolyses and dibasic acids such as maleic anhydride and adipic acid. The resins were then
diluted with styrene to make 44% (by weight) styrene solution after the polymerizations and hydroquinone(0.5% by
weight of the resin) was added as an inhibitor. Variables in the syntheses of the UPE were the type of glycol, the PET
content, the molecular weight of the resin, and the degree of unsaturation of the UPE. In Table 1, recipes of UPE
resin from recycled PET are given. Hydroxyl values of the resin were measured to check the number average molecular
weight of the resin.
In order to cure the resin, methyl ethyl ketone peroxide(MEKPO), 1 wt.% of the resin, was added as an initiator
and cobalt acetate, 0.5 wt.% of the resin, was also added as an accelerator of the cure. Glass transition temperatures
(Tgs) of the cured resins were measured by employing differential scanning calorimeter (DSC: DuPont Thermal
analyzer 2000). About 10 mgs of samples were heated in DSC in nitrogene gas environment at 10C/min to measure
Tgs of the samples. Mechanical properties of the cured resin were measured by employing universal testing machine
at room temperature (25C). Polymer mortars (PM) were prepared with the UPE resins, calcium carbonate, and
aggregates. Mix formulations of the polymer mortar is given Table 2. Mechanical properties of the polymer mortar
were also measured by employing universal testing machine.
Table 1. Recipes (by molar ratios) for preparation of unsaturated polyester resin from recycled PET
Sample Code

Recycled-PET

Propylene glycol

Dipropylene glycol

Maleic anhydride

Adipic acid

A-1
A-2
A-3
A-4
A-5
B-1
B-2
B-3
C-1
C-2
C-3
D-1
D-2
D-3
B-31
B-32
B-33
B-34

1.4
1.2
1.0
0.8
0.6
1.2
1.0
0.8
1.2
1.0
0.8
1.2
1.0
0.8
0.8
0.8
0.8
0.8

1.4
1.4
1.4
1.4
1.4
1.2
1.2
1.2
1.1
1.1
1.1
1.2
1.2
1.2
1.2

1.2
1.2
1.2
-

1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
0.95
0.90
0.85
0.80

0.05
0.10
0.15
0.20

Table 2. Mix formulation for polymer mortar


Material

Weight ratios

Unsaturated polyester resin


Calcium carbonate
Aggregate

15
15
70

SOH, PARK AND LEE

Fig. 1. Tgs of the cured UPE resins depending on the PET content in the resin:
D-series.

A-series;

B-series;

C-series; ( )

3
Results and Discussion
In Table 3, the hydroxyl values and the number average molecular weights of the resins are summarized. Number
average molecular weights of the resins were found to be dependent on the stoichiometry of the reactants as we
expected to be.
In Fig.1, Tgs of the cured resins depending on the PET content of the reins are shown. It is observed that Tgs of
the resins decrease as the PET contents of the resins are increased. Even though the introduction of PET is expected
to increase chain rigidity of the resin molecules, it results in decrease of unsaturation and crosslink density of the
resin. Thus, the decrease of the Tg as PET content is increased in the resin is attributable to the decreased crosslink
density. It is of interest to note that Tg of the resin made from glycolyses products using PG is higher than that of the
resin made from glycolyses products using DPG. It is also observed that the higher are the molecular weights of the
resin, the higher, Tgs of the cured resin. It seems that Tgs of the cure resins are determined mainly by unsaturation
of the resin.
Table 3. Hydroxyl values and number average molecular weights of the UPE resins prepared
Sample Code

Hydroxyl value (mg KOH/g)

Number average molecular weight

A-1
A-2
A-3
A-4
A-5
B-1
B-2
B-3
C-1
C-2
C-3
D-1
D-2
D-3
B-31
B-32
B-33
B-34

95.02
97.05
110.10
122.51
146.64
76.08
75.58
75.43
55.09
50.05
72.07
64.21
63.65
66.59
80.75
87.96
81.39
77.65

1178.76
1154.10
1017.26
914.20
763.8
1472.13
1481.87
1484.82
2033.04
2237.76
1553.90
1744.28
1759.60
1582.80
1387.08
1273.30
1376.05
1442.37

In Fig. 2 and Fig. 3, flexural strength and flexural modulus of the cured resin depending on the PET content are
shown. It is observed that flexural modulus of the cured resin made from glycolyses product using PG is higher than

MECHANICAL PROPERTIES OF POLYMER MORTAR

Fig. 2. Flexural modulus (FM) of the cured UPE resins depending on the PET content in the resin:

Fig. 3. Flexural strength (FS) of the cured UPE resins depending on the PET content in the resin:

B-series;

B-series;

D-series.

D-series.

that of the cured resin made from glycolyses products using DPG, while flexural strength of the cured resin showed
opposite trend. It is speculated that the cured resins made from glycolyses products using DPG have more flexible
chain and flexural modulus are low compared with the cured resin made from glycolyses products using PG.
However, ultimate strength of the cured resin made from glycolyses products using DPG is superior to the cured
resin made from glycolyses products using PG due to possibly higher ultimate strain. It is of interest to note that
flexural strength of the cured resin made from glycolyses products using PG decrease as PET content in the resin is
increased. As we noted in Fig. 1, higher PET content implies lower unsaturation of the resin and lower crosslink
density of the cured resin. Thus, the decrease of flexural strength with PET content may be attributable to lower
unsaturation and lower crosslink density.
In Fig. 4 and Fig. 5, flexural modulus and flexural strength of the cured resin depending on the adipic acid content
are shown. It is found that flexural modulus of the cured resin decreased as the adipic acid content in the resin was
increased. Introduction of adipic acid instead of maleic anhydride implies decrease of unsaturation of the resin. It
seems that the decrease of flexural modulus resulted from decreased crosslink density due to lower unsaturation. It is
of interest to note that flexural strength, on the contrary, tends to increase as the adipic acid content is increased. It is
speculated that the cured resins of higher adipic acid content have more flexible chain and the flexural modulus are
low compared with that of the cured resin of lower adipic acid content relatively. But, ultimate strength of the cured
resin of higher adipic acid content is superior to the cured resin of lower adipic acid content due to possibly higher
ultimate strain.
In Fig. 6 and 7, changes of compressive strength and flexural strength of PM depending on styrene monomer(SM)
of the various resins are given. It is observed that the mechanical properties of the PMs were not affected by the SM
contents so much. However, it is of interest to note that the compressive strength of PM prepared with the UPE made
from recycled PET using DPG is lower than that of PM prepared with the UPE resin made from recycled PET using
PG while the flexural strength of PM prepared with the UPE made from recycled PET using DPG is higher than that
of PM prepared with the UPE resin made from recycled PET using PG. It is speculated that the UPE resin molecule
made from PET using DPG is more flexible than the UPE resin molecule made from PET and PG and the

SOH, PARK AND LEE

Fig. 4. Flexural modulus(FM) of the cured UPE resins depending on the adipic acid (AA) content in the resin.

Fig. 5. Flexural strength(FS) of the cured UPE resins depending on the adipic acid (AA) content in the resin.

Fig. 6. Compressive strength (CS) of the PM from different UPE depending on the styrene monomer (SM) content in the resin:
B-1;
B-2;
B-3;
D-1; ( ) D-2;
D-3.

phenomena observed in Fig.6 and 7 are due to the different molecular flexibilities and crosslink densities of the
resins as in Fig. 5.
In Fig. 8, changes of compressive strength of the PM depending on the adipic acid content in the resin are given. It
is observed that the compressive strength of the PM decreased as the adipic acid content is increased. The increase of
adipic acid in the UPE resin results in the decrease of the degree of unsaturation of the resin. The decrease of
compressive strength of PM as the adipic acid content is increased is attributable to the decrease of crosslink density
of the resin because of the decreased degree of unsaturation. In Fig. 9, changes of flexural strength of the PM
depending on the adipic acid content in the resin are given. It is observed that the flexural strength of the PM
increased as the adipic acid content is increased. The increase of flexural strength of PM as the adipic acid content is
increased is also attributable to the decrease of crosslink density of the resin, increased molecular flexibility, and
possibly increased ultimate strain.

MECHANICAL PROPERTIES OF POLYMER MORTAR

Fig. 7. Flexural strength (FS) of the PM from different UPE depending on the styrene monomer (SM) content in the resin:
B-1;
B-2;
B-3;
D-1; ( ) D-2;
D-3.

Fig. 8. Compressive strength (CS) of the PM depending on the adipic acid (AA) content in the resin of different SM content (wt.%):
40;
44;
48.

Fig. 9. Flexural strength (FS) of the PM depending on the adipic acid (AA) content in the resin of different SM content (wt.%):
40;
44;
48.

4
Conclusion
Various UPE resins based on recycled poly ethylene terephthalate(PET) were prepared and the properties of the
polymer mortar made with the UPE as well as those of the cured resin itself were investigated. It was found that
degree of unsaturation of the resins was the most important property affecting thermal and mechanical properties of
the cured resin and PM. The cured unsaturated polyester resin or polymer mortar made from resins of higher
unsaturation, i.e., less PET content or adipic acid content in the resin, showed higher glass transition temperatures,

SOH, PARK AND LEE

compressive strength, or higher flexural modulus and lower flexural strength. Such properties could be interpreted in
terms of unsaturation, crosslink density, and chain flexibility of the resin molecules.
5
References
1.
2.
3.
4.

Pearson, W., Emerging Technologies in Plastics Recycling, Andrews, G.D. and Subramanian, P.M. (1992), Edt., Chapter
1, ACS Symposium Series 513, ACS, Washington D.C.
Rebeiz, K.S., Iyer, V.S., Fowler, D.W. and Paul, D.R. (1990), Proceedings of 48th Annual Technical Conference
(ANTEC90).
Schneider, J.B., Ehrig, R.J., Brownell,G.L. and Kosmack, D.A. (1990), Proceedings of 48th Annual Technical Conference
(ANTEC90).
Rebeiz, K.S., Fowler, D.W. and Paul, D.R., (1992), Emerging Technologies in Plastics Recycling, Andrews,G.D. and
Subramanian, P.M. Edt., Chapter 1, ACS Symposium Series 513, ACS, Washington D.C.

2
PROPERTIES OF PLAIN AND REINFORCED POLYESTER
CONCRETES MADE WITH RECYCLED PET
K.S.REBEIZ
Department of Civil and Environmental Engineering, Lafayette College, Easton, Pennsylvania,
USA
D.W.FOWLER
Department of Civil Engineering, The University of Texas at Austin, Austin, USA

Abstract
Recycled poly (ethylene terephthalate), PET, plastic waste can be used to produce unsaturated polyester resins.
The PET waste is typically found in used beverage bottles that are collected after use in many localities.
Research at the University of Texas investigated the use of suitable unsaturated polyester resins based on recycled
PET for the production of polymer concrete (PC) materials. The properties and structural behavior of
unreinforced and steel-reinforced PC materials using resins based on recycled PET were found to be
comparable to those obtained with PC materials using virgin resins. Resins based on recycled PET can also
relatively easily be altered to achieve a wide variety of properties and performances in the PC. An
experimental design also showed that the effect of the level of PET in the resin did not adversely affect the
neat resin and the PC mechanical properties. Resins based on recycled PET help in decreasing the cost of PC
products, saving energy, and alleviating an environmental problem posed by plastics waste.
Keywords: Polymer concrete, polyester, polyethylene, recycled materials
1
Introduction
The high cost of resins used in the production of polymer concrete (PC) makes the material expensive relative to cementbased materials. Not surprisingly, a recent survey ranked lower cost resins as the most important future need for PC
[1]. Recently, some work has been done on the production of unsaturated polyester resins based on recycled poly
(ethylene terephthalate), PET [2]. The PET wastes are typically found in used beverage bottles, and many states have
passed legislation to collect and recycle these bottles. If specially formulated, the unsaturated polyester could be used
in the production of PC [3].
Unsaturated polyesters based on recycled PET might be a potentially lower source cost of resins for producing
useful PC based-products. A main advantage of recycling PET in PC is that the PET materials do not have to be
purified, including removal of colors, to the same extent as other PET recycling applications (such as carpets and
fiberfills), which should facilitate the recycling operation and minimize its cost. The recycling of PET in PC could
also help save energy and allow the long term disposal of the PET waste, an important advantage in recycling
applications. The objective of this paper is therefore to report on investigations of the important properties and behavior
of PC using resins based on recycled PET.

Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

10

REBEIZ AND FOWLER

2
Materials
PET consists of repeating ethylene glycol and terephthalic acid molecules connected together through ester linkages.
For the production of unsaturated polyester, the PET molecules are converted into low molecular weight oligomers
by glycolysis in the presence of a transesterification catalyst. These oligomers are then reacted with unsaturated
dibasic acids or anhydrides to form unsaturated polyester resins [3]. A variety of other chemicals may also be used
during the production process to give the resin some specific properties such as flexibility or rigidity.
Phthalic anhydride or isophthalic acid is typically used in the formulation of conventional unsaturated polyester.
Virgin terephthalic acid is not usually used in the production of unsaturated polyester because it is expensive and
possesses a high melting point, which presents difficulties in synthesis. Conversely, recycled PET is effective in
incorporating terephthalic functionality into the backbone of a polyester resin [4] [5]. Terephthalic-based polyesters
exhibit more linear properties than isophthalic or orthophthalic-based polyesters when the polymerization reaction
with typical glycols and acids occurs because of the location of the carboxyl groups on the benzene ring of phthalic
acid. Terephthalic, isophthalic, and orthophthalic-based polyesters have their carboxyl groups in the para, meta, and
ortho position of the benzene ring, respectively. The degree of linearity, in descending order, results from the para, meta,
and ortho-based structures. Therefore, the more linear molecular structure of terephthalic-based polyester resins, as
compared to isophthalic or orthophthalic based-polyesters, allows the cross-linking reaction to be more accessible
when free radical polymerization takes place upon final curing, thus producing a more uniformed and structured
cross-linking matrix with a higher degree of strength, stiffness, and toughness. Another advantage of using recycled
PET in making unsaturated polyesters, as compared to using virgin materials, is that it takes about 50% shorter
processing time to produce a polyester resin with a certain molecular weight and acid number [3].
The experimental unsaturated polyester resins used in this study were supplied by several chemical companies
since the PET chemical conversion into unsaturated polyester could not be done at the University of Texas
laboratories. These resins were prepolymers with high viscosity. They were therefore diluted with styrene to reduce
their viscosity and allow their further cure to a solid (polymer) upon the addition of suitable free radical initiators and
promoters. The typical styrene content varied between 30 to 40% of the total resin weight, and viscosities were in the
range of 100 to 1000 cps.
3
Mix Design
The PC mix design, optimized for workability, strength, and economy, was 10% resin, 45% pea gravel (3/8-in./10mm), 32% sand (fineness modulus of 3.25), and 13% fly ash (type F). The gravel and sand were oven-dried for a
minimum of 24 hours at 260F (127C) to reduce their moisture content to less than 0.5% by weight, thus ensuring
good adhesion between the polymer matrix and the aggregates. Fly ash was already obtained dry from the supplier
and therefore did not need to be oven-dried. The use of fly ash greatly improved the workability of the fresh mix and
the strength of the hardened material. It also helped produce PC specimens with very smooth surfaces. One percent,
by weight of resin, of methyl ethyl ketone peroxide initiator (9% active oxygen) and 0.1%, by weight of resin, of cobalt
naphthenate promoter (12% solution) were added to the resin immediately prior to its mixing with inorganic
aggregates. The PC mixing procedure followed the Polymer Concrete Test Method 1.0 of the Society of Plastics
Industry [6]. The age at testing of the specimens was three days, unless otherwise specified, although the specimens
could have been tested much sooner.
4
Testing
The compression test used 3-in.6-in. (7.6-cm15.2-cm) cylinders. The specimens were tested in a hydraulic load
machine at a constant loading rate of 10,000 lbs/min (4,500 kg/min). Electrical strain gages were bonded to the
specimens and connected to an automated data acquisition system in a full-bridge configuration. Flexural specimens
used 2-in.2-in.12-in. (5.1-cm5.1-cm30.5-cm) beams. The beams were loaded in third-point loading, at a
uniform rate of 500 lbs/min (230 kg/min).
Bond strength between PC overlays and portland cement concrete substrate was measured using the pull-out test
method [7]. The tensile bond test is illustrated in Fig. 1. Specimens were thin overlays, about 1/2-in. (1.3-cm) thick,
cast directly (without the use of a primer) on sandblasted portland cement concrete slabs. Circular grooves (4-in./10.
8-cm diameter) were cored through the overlays and into the portland cement concrete substrate. Circular steel disks

PLAIN AND REINFORCED POLYESTER CONCRETES

11

were then bonded to the sandblasted overlay at the cored locations using a strong epoxy. The disks were then pulled
out in direct tension to determine the type and magnitude of the bond failure. The mix design used in making the
portland cement concrete substrate was designed to achieve a compressive strength of 5000 psi (34.5 MPa) with the
use of air entraining agents.
The Duponts method was used to measure shrinkage strains [3]. Specimens consisted of 3-in.3-in.12-in. (7.6cm7.6-cm30.5-cm) beams cast in Teflon-lined molds. The molds were wrapped in a plastic sheet to reduce the
effect of ambient temperature changes on the plastic shrinkage readings. Immediately after mixing and placing the
materials in the molds, the shrinkage measuring device was carefully inserted into the fresh PC mix to record
shrinkage strains for different time intervals. The shrinkage device consists of a horizontal rod to which two
removable angles were attached. One angle was fixed while the other was free to move on roller bearings. A direct
current differential transformer, attached to the rod, was used to record the longitudinal displacement induced by
shrinkage. PC peak exotherms were measured by inserting thermocouples inside the shrinkage specimens and
connecting them to a digital temperature indicator.
The thermal expansion test used 3-in.6-in. (7.6-cm15.2-cm) cylinders. Electrical strain gages were
longitudinally bonded to the specimens at mid-height and on opposite sides using a special epoxy system insensitive
to high temperatures (other epoxy systems showed improper behavior beyond 130F (54.4C)). The strain gages
were then connected to a switch and balance unit in a full-bridge configuration. A piece of fused quartz with a known
coefficient of thermal expansion was used as the compensating arm of the full-bridge circuit. The specimens were
subjected to thermal cycles beginning at room temperature. The temperature was increased to 162F (72.2C),
decreased to 10F, and then returned to room temperature. Strain and temperature readings were taken in
increments of 10F (5.6C). For each of the increments, the specimens were left at a constant temperature for a
minimum of eight hours to ensure thermal stabilization before the strain and temperature readings were taken.
Thermocouples, attached to the surface of the specimens and connected to a digital temperature indicator, were used
to monitor specimen temperatures.
Creep specimens consisted of 3-in.6-in. (7.6-cm15.2-cm) cylinders tested in uniaxial compression using a
hydraulic spring-loaded creep frame. During testing, cylinders were aligned on top of each others to ensure uniform
stress in all of them. Electrical gages were bonded to the specimens at mid-height using epoxy and then connected to
an automated data acquisition system in a full-bridge configuration. Electrical strain gages were also attached to
dummy or control specimens that were left unloaded to correct for non-creep related deformations such as shrinkage
of the adhesive used to attach the strain gages, springs decompression in the loading frame, or ambient temperature
changes. Since temperature changes of plus or minus 2F (1.1C) could significantly affect the creep readings, the
specimens were enclosed in an insulating Styrofoam box to ensure that the temperature was maintained at 75F (23.
9C) throughout the testing period.
Rectangular beams, reinforced in tension with longitudinal steel bars, were tested for their flexural behavior. A
typical reinforced flexural beam is shown in Fig. 2. The reinforcing steels were Grade 60 deformed bars conforming
to ASTM Standard A615. The beams were simply supported and two equal concentrated loads were applied at the
third points of the span. The shear span-to-depth ratio was 4.0. The two outer portions of the flexural beams were
over-reinforced with vertical closed stirrups to prevent a shear mode failure. Vertical stirrups were #3 Grade 60
deformed steel bars spaced 3-in. (7.6-cm) apart (high strength wire was used to securely tie the stirrups). Each stirrup
was rectangular in shape and was closed by welding its ends together. Electrical strain gages were bonded to the
reinforcing steel and to the PC at various levels and connected to a data acquisition system. The vertical deflection of
the beams was measured using linear transducers. The beams were loaded to failure with strains and deflections
recorded for each load level.
5
Properties of PC Using Resins Based on Recycled PET
The PC mechanical properties using various unsaturated polyester resins based on recycled PET are shown in
Figs. 3 to 6. Each value in these figures represents the average value of two specimens. The different PC systems
were made with different resins using different formulations. The wide range of properties encountered with PC
using resins based on recycled PET enable these materials to be used in various applications where different
properties are desirable. For example, PC with high strength and modulus can be used in making precast machine
base components. Conversely, PC with low modulus and good bond strength to portland cement concrete can be
used in the repair of pavements and bridges. The properties of PC made with resins using recycled PET are
comparable to those obtained with PC using virgin materials and tested under the same conditions at the University of
Texas.

12

REBEIZ AND FOWLER

The failure of PC in compression was violent. Compression cylinders would shatter violently and the remaining
core of the cylinders had either a cone shape or a near vertical failure surface. Flexural beam specimens also failed in
a violent manner as a tensile crack developed in the zone of maximum moment near mid-depth. The specimens were
broken into almost two identical pieces and the failure surface was near vertical. The tensile bond strength between
the PC overlays and the portland cement concrete substrate was found to be strongly dependent on the type of resin
used. In overlay or repair applications, it is usually desirable to have tensile bond failures occurring in the portland
cement concrete substrate rather than at the interface between the two materials.
A typical stress-strain curve in compression for two PC systems, one using a flexible resin and the other using a
rigid resin, is shown in Fig. 7. PC made with a flexible resin exhibits lower modulus, higher ductility, and more
toughness than the PC made with a rigid resin. In precast components, PC should be based on rigid resins. In overlay
applications, PC should be based on flexible resins capable of stretching when subjected to large thermal or
mechanical movements. The stress-strain behavior and the ultimate compressive strength of PC using resin based on
recycled PET are comparable to those obtained with PC using virgin resins [8]. It can also be observed that the
ultimate compressive strain of PC is much larger than the one corresponding to portland cement concrete.
A typical shrinkage and exotherm curve for PC is shown in Fig. 8. Most of the shrinkage strains in PC took place
within the first eight hours after mixing and stopped after 24 hours. It is also noted in the figure that most of the
shrinkage took place after the occurrence of the peak exotherm. Shrinkage is important in many PC applications. In
precast components, low shrinkage is important because excessive shrinkage strains may significantly affect the
dimension of these structures, thus making their demolding, assembly, or use more difficult. In overlay applications,
relatively low shrinkage is desirable because studies have reported that excessive shrinkage strains may cause
delamination between the overlay and the substrate [9] [10].
A typical thermal expansion curve for PC is shown in Fig. 9. The thermal expansion is important when PC is used
in conjunction with other materials such as steel or portland cement concrete since the coefficient of thermal
expansions of PC is at least twice as high as those corresponding to steel or cement concrete. Hence, changes in
temperature in the composite structure will create shear stresses at the interface between the two materials that may
eventually cause deterioration in the structure.
A common problem encountered with PC systems used in precast components is excessive creep deformation
under service conditions. Creep takes place in PC as the result of molecular movement in the viscoelastic resin
binder. Therefore, variations in time and temperature significantly affect the PC creep behavior. A typical PC creep
compliance (or strain per uniaxial unit sustained stress) and Poissons ratio are plotted versus time in Fig. 10. The
stress intensity ratio (which is the ratio of the applied compressive stress to the ultimate compressive strength) used
in this test was 20%. A stress intensity ratio of 20% was chosen to avoid complications resulting from nonlinear
viscoelastic behavior and because PC is usually designed with a high safety factor. The application of larger load
levels would also have been more difficult because of the high strength of PC [11]. Previous creep studies with PC
determined that excessive creep deformation and catastrophic failure often occur when the creep stress intensity ratio
exceeds about 50% [12]. More than 20% of the final creep for PC took place within one day and more than 90% of
the final creep took place within six days. The specific creep (which is the creep strain divided by the sustained
stress) for PC was 199.0 /ksi (1372 /MPa) after 95 days, which is comparable to what was observed with other
PC systems using virgin resins [13]. The PC creep strain is higher than the one corresponding to portland cement
concrete. It should be noted however that different conditions affect the creep behavior of polymer composites and
cement-based materials. The creep behavior of polymer composites is sensitive to temperature variations while the
creep behavior of cement-based materials is sensitive to humidity changes. It is also noted in Fig. 10 that the
Poissons ratio of PC, measured during the compressive creep stress, increased by about 26% during the three month
test period. This time dependency of the PC Poissons ratio needs to be taken into consideration in the analysis of
structural elements.
6
Effect of Level of PET on Strength Properties
A D-optimal experimental design was developed to determine the effect of recycled PET as a raw component in
unsaturated polyester resins [14]. This study was done on several resins made with the same basic formulation but
with the percent PET and glycol type as the only variables. The three different glycols that were used in the digestion
process of
the recycled PET were ethylene glycol, diethylene glycol, and a combination of ethylene and
diethylene glycol. These three different glycols will be referred to as glycols 1, 2, and 3 respectively. Unsaturated
polyester resins were synthesized with a low PET concentration of 15% by weight and a high PET concentration of

PLAIN AND REINFORCED POLYESTER CONCRETES

13

Fig. 1. Pull-out Bond Test

Fig. 2. Reinforced PC Flexural Beam

40% by weight of the alkyd portion portion of the resin. The D-optimal experimental design consisted of eight
unsaturated polyester resin preparations with two replicates.
The effects of recycled PET on the tensile and flexural strengths of the neat resins (cured resins without the use of
aggregates) are shown in Figs. 11 and 12, respectively. The neat resins were produced and tested according to ASTM
procedures [15]. The tensile strength of the neat resin increases with increasing amount of PET for glycols 1 and 2,
while it remains almost constant for glycol 3. The tensile strength of resins using glycol 3 is the highest, followed by
resins using glycols 1 and 2, respectively. The flexural strength of all neat resins increases with increasing amounts of
PET. The flexural strength of resins using glycol 3 is highest, followed by resins using glycol 1, and then resins using
glycol 2.
The effects of percent PET and glycol type on the PC compressive and flexural strengths are shown in Figs. 13
and 14, respectively. The compressive strength of PC using resins made with glycols 1 and 2 increases significantly
with increasing PET, while it increases very slightly for PC using resins made with glycol 3. The compressive
strength of PC made with resins using glycol 3 is the highest, followed by PC made with resins using glycol 1, and
then PC made with resins using glycol 2. The flexural strength of PC made with resins using glycols 1 and 3
increases with increasing PET, while it decreases for PC made with resins using glycol 2. At a low PET
concentration of 15%, all systems are comparable in flexural strength. At a high PET concentration of 40%, the
flexural strength of PC made with resins using glycol 1 is the highest, followed by PC made with resins using glycol
3, and then PC made with resins using glycol 2.

14

REBEIZ AND FOWLER

Fig. 3. PC Compressive Strength

Fig. 4. PC Flexural Strength

7
Flexural Behavior of Reinforced PC
The effect of the tensile steel reinforcement ratio, , on the load-deflection and moment-curvature responses of PC
beams is shown in Fig. 15. The moment was computed from the applied loads by statics while the corresponding
curvature was calculated from the strain distributions. It is noticed that the load-deflection and moment-curvature
responses are very similar in terms of shape. In the first stage of loading, a linear relationship existed between the
moment (or load) and curvature (or deflection). This proportional limit stage ended with the formation of a major
vertical flexural crack and the resulting change in slope and decrease in stiffness.
The cracking patterns for steel-reinforced flexural PC beams were similar and generally typical to those observed
in steel-reinforced portland cement concrete. However, PC exhibited more cracks that were more uniformly spaced
than what would be observed with portland cement concrete with the same reinforcement ratio, thus indicating a
larger bond strength between PC and steel than between portland cement concrete and steel. The first noticeable
cracks were flexural cracks that originated in the tension zone at mid-span between the two point loads and
propagated vertically toward the compression zone with increasing loads. The second type of cracks observed in the
beams were flexural-shear cracks that originated in the shear span. The initiation of cracks depended on the tensile
steel reinforcement ratio. The lower the steel ratio, the earlier the visible crack occurred at a lower load. As the
reinforcement ratio increased, the number and spacing of cracks decreased.

PLAIN AND REINFORCED POLYESTER CONCRETES

15

Fig. 5. PC Compressive Modulus of Elasticity

Fig. 6. PC Tensile Bond Strength

Variation of strain over the compression region up to failure for the beams was found to be almost perfectly linear.
Failure occurred when the ultimate compressive strain in the PC reached a value of at least 0.005. As the beams
failed, the compressive concrete piece separated as a V-shape, a phenomenon already observed before with other
steel-reinforced PC systems.
The ultimate flexural strength of steel-reinforced PC using resins based on recycled PET was also compared with
other steel-reinforced concrete systems as shown in Fig. 16. The other concrete systems were portland cement
concrete, unsaturated polyester (virgin resin), methyl methacrylate (MMA), polyesteramide resin (PEAR), epoxy,
and vinyl ester. It is observed that the flexural strength of PC using resins based on recycled PET is much higher than
the one corresponding to portland cement concrete and comparable to those obtained with PC systems using virgin
materials. It should be noted that a large portion of the PC internal moment at failure is resisted by the tensile stresses
in the concrete, unlike what would be happening with steel-reinforced portland cement concrete.

16

REBEIZ AND FOWLER

Fig. 7. Typical PC Stress-Strain Curve in Compression

Fig. 8. Typical PC Shrinkage and Exotherm versus Time Curves

8
Conclusions
Resins based on recycled PET can relatively easily be altered to achieve a wide variety of properties and
performances. Resins with high strength and stiffness can be formulated for precast applications, while resins with
low modulus, high elongation at break, and good bond strength to portland cement concrete can be formulated for
overlay applications. The properties of PC materials using resins based on recycled PET are very comparable to
those obtained with PC materials using virgin resins. Potential applications of PC using resins based on recycled PET
can include thin overlays on bridges and floors, repairing concrete bridges and pavements, and the production of
many precast products such as containers for hazardous wastes, floor drains, electric insulators, bases for large metalworking machines, and building panels.
The effect of the level of PET in the resin did not adversely affect the neat resin and the PC properties. Resins
using a maximum amount of recycled PET and impurities are desirable because they did not adversely affect the
materials properties while they helped decrease the cost of PC based-products, thus making them more competitive.
As more localities are instituting recycling programs, the supply of recycled PET is expected to increase and,
consequently, the price of resins based on recycled PET is expected to decrease.
The use of steel bars can be very effective in increasing the strength of PC materials. Compared to steel-reinforced
portland cement concrete, the material is much stronger and more ductile. PC also requires less reinforcement cover
for the tensile reinforcing steel than portland cement concrete because of its inherent high flexural strength, low
permeability, and excellent chemical resistance. The flexural strength of steel-reinforced PC using resins based on
recycled PET was found to be comparable to other PC systems using virgin resins.

PLAIN AND REINFORCED POLYESTER CONCRETES

17

Fig. 9. Typical PC Thermal Expansion Curve

Fig. 10. Typical PC Creep Compliance and Poissons Ratio versus Time Curves

Field applications and continuous monitoring of PC materials using resins based on recycled PET would really
determine the long term behaviors of these materials under field conditions. Special precautions should be taken in
cases involving large sustained loads because the viscoelastic nature of the resin binder can result in unreasonably
high deformations. Special resin formulations, adequate supports, and/or large safety factors would be advisable in
these instances. Trial testing of the materials is also advisable because it would help overcome problems due to
erratic cure or batch to batch inconsistencies of the resins. For applications requiring very high strength, a special
mix using strong aggregates and special reinforcements should be formulated. Future progress in physics and
chemistry should allow the economical chemical conversion of other plastic wastes into resins that can be used
effectively in the production of new PC systems with improved properties that will extend their use in engineering
and structural applications even further.
Acknowledgement
The authors acknowledge the support for this research from the Advanced Research Program of the Texas Higher
Education Coordinating Board.
9

18

REBEIZ AND FOWLER

Fig. 11. Effect of Level of PET on Tensile Strength of Neat Resins

Fig. 12. Effect of Level of PET on Flexural Strength of Neat Resins

References
1.
2.
3.
4.
5.

6.
7.

D.W.Fowler (1989). Future Trends in Polymer Concrete. ACI SP1168, pp 129 143.
U.R.Vaidya and V.M.Nadkarni (1987). Unsaturated Polyester Resins from Poly(ethylene terephthalate) Waste. Industrial &
Engineering Chemistry Research, Vol. 26, No. 2, pp. 194198.
K.S.Rebeiz, D.W.Fowler, and D.R.Paul (1991). Formulating and Evaluating an Unsaturated Polyester Composite made
with Recycled PET. Journal of Materials Education, Vol. 13, No. 5 & 6, pp. 441454.
J.B.Schneider, R.J.Ehrig, G.L.Brownell, and D.A.Kosmack (1990). Unsaturated Polyesters Containing Recycled PET.
Proceedings of the 48th Annual Technical Conference of the Society of Plastics Engineers, pp. 14621465.
A.J.DeMaio (1991). Engineering High Performance Thermoset Resins from Poly(ethylene Terephthalate) Thermoplastics.
Proceedings of the 46th Annual Conference of the Composites Institute of the Society of Plastics Institute,
pp. 18C/118C/5.
Polymer Concrete Test Methods (1987). Composite Institute of the Society of Plastics Industry.
Use of Epoxy Compounds with Concrete (1979). ACI Committee 503 Report.

PLAIN AND REINFORCED POLYESTER CONCRETES

19

Fig. 13. Effect of Level of PET on PC Compressive Strength

Fig. 14. Effect of Level of PET on PC Flexural Strength


8.
9.
10.
11.
12.
13.
14.
15.

C.Vipulanandan, N.Dharmarajan, and E.Ching (1987). Stress-Strain Behaviour of Polymer Concrete Systems. Proceedings
of the Fifth International Congress on Polymers in Concrete, Brighton, England, pp. 165170.
A.Al-Negheimish (1988). Bond Strength, Long Term Performance and Temperature Induced Stresses in Polymer ConcretePortland Cement Concrete Composite Members. Ph.D. Dissertation, The University of Texas.
R.Letsch (1987). Polymer Mortar OverlaysMeasurement of Stresses. Proceedings of the Fifth International Congress on
Polymers in Concrete, Brighton, England, pp. 119123.
K.C.Kyriacou (1991). Accelerated Compression and Flexural Creep Testing of Polymer Concrete. Ph.D. Dissertation, The
University of Texas.
M.Hsu and D.W.Fowler (1985). Creep and Fatigue of Polymer Concrete. ACI SP 89, pp. 323341.
J.Hristova and R.A.Bares (1987). Relation between Creep and Performance of PC. Proceedings of the Fifth International
Congress on Polymers in Concrete, Brighton, England, pp. 99102.
K.S.Rebeiz (1992). Structural Use of Polymer Composites Using Unsaturated Polyester Resins Based on Recycled Poly
(ethylene Terephthalate). Ph.D. Dissertation, The University of Texas.
Annual Book of the American Society for Testing Materials (1988).

20

REBEIZ AND FOWLER

Fig. 15. Typical PC Moment-Curvature and Load-Deflection Responses

PLAIN AND REINFORCED POLYESTER CONCRETES

Fig. 16. Ultimate Moment Strength of Various Concrete Systems

21

3
A NEW KIND OF HYBRID RECYCLED POLYMER MORTAR
Y.BAO
East China Hydroelectric Power Investigation & Design Institute, Hangzhou, Zhejiang Province,
China
D.P.WHITNEY and D.W.FOWLER
Department of Civil Engineering, The University of Texas at Austin, Austin, USA

Abstract
Polymer mortar is especially suited to applications requiring quick set time, high strength, and lack of
delamination. Since environmental regulations have become stricter regarding the disposal of the large amount
of waste plastic produced, these waste materials have begun to be incorporated in polymer mortar. In this study,
tests were performed on a new variety of polymer mortar which combines recycled polyester and polyurethane
with traditional aggregates. The resins studied included RPE (recycled polyethylene), HRPE (a hybrid of
RPE), and RPU (recycled polyurethane). Tests included compressive, tensile, and bond strengths versus the
ratio of RPU to RPE in the mortar. The results are presented.
Keywords: Polymer concrete, polyester, polyurethane, recycled materials.
1
Introduction
A new type of polymer mortar was developed by combining two recycled polymers: recycled polyester and recycled
polyurethane. The technology of recycled polyurethane-modified recycled polyester has made possible materials that
combine the flexibility, adhesion, and expansion of polyurethane, and the quick set time, high strength and lower
cost of polyester. The recycled polymer mortar was obtained by mixing these new hybrids, initiator and promoter,
with sand and fly ash. This new kind of recycled polymer mortar has excellent bond and tensile strength,
compressive strength and lower shrinkage. It exhibits no delamination when used for overlays on concrete subjected
to thermal cycling. The short set time and the fast strength development make this recycled polymer mortar a useful
material for fast repairs of concrete.
Research was conducted on the method of recycling polyurethane, formulation of recycled polyurethane-modified
recycled polyester (RPUMRPE) and their mechanical and chemical resistance properties.
2
The Recycling of Rigid Polyurethane Foam [1] [2] [3]
The utilization of plastic products has been a vital development in 20th century technology. Along with more and
more plastic wastes being placed into the environment, the problems of disposal and environmental pollution must be
overcome. Plastic waste products are very light; they occupy about 7% of the weight of garbage, but they represent
about one-fourth of the total volume which requires a large portion of the landfill. Incineration is another disposal
method, but locating sites for new incineration facilities has led to considerable public opposition. Consequently the
ecological disposal of plastic products is presently of great concern.
Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

A NEW HYBRID RECYCLED POLYMER MORTAR

23

Use of polyurethane foam has been increasing dramatically and with this increase there has also developed a
serious problem in disposing of waste foam products. Various approaches have been developed in an effort to reclaim
the waste foam. Flexible polyurethane waste foam generated in production or recovered after use as packaging
materials can be recycled by shredding and mixing with a polyurethane binder to make carpet underlay. The rigid
foams are highly cross-linked materials, and at present the disposal methods are landfill and incineration. Neither
method is acceptable because long-range ecological goals dictate zero pollution as well as conservation of raw
materials. Therefore efficient recycling methods are mandatory in the years to come.
Three methods of recycling of polyurethane foam have been proposed: hydrolysis, pyrolysis and glycolysis.
1. Hydrolysis:

In hydrolysis, a complex mixture of polyol and polyamine is obtained. The method of separation is required, but it is
very difficult and not feasible.
2. Pyrolysis:
This process gives a more complex mixtures of chemical compounds, useful only as gaseous or liquid fuels.
3. Glycolysis:
This process can achieve a recovery of scrap foam into a polyol mixture, which can be used to produce polyurethane
without the need for purification. This method requires low capital investment and is simple enough to cope with
variations in the mix scrap foam. The resulting recycled polyol can be incorporated with polyisocyanate to produce
recycled polyurethane which can be used to modify recycled polyester mortar made with either virgin or recycled resins.
The glycolysis method of recycling polyurethane foam consists of heating scrap foam with glycol or polyol,
increasing the solubilizing agent at a temperature of 180 to 196C. The scrap polyurethane is chopped into particles
of relatively small size to reduce the reaction time. The addition of the scrap can be made over a period of time. The
period of heating will range from about 5 to 8 hours depending upon the nature of the scrap polyurethane and the
glycol or polyol employed. The viscosity of the reactant is mainly controlled by the proportion of scrap polyurethane
to glycol, polyol and increasing-solubilizing agent used. When the reaction is completed, the viscosity and hydroxyl
value will be determined. The procedures are shown in Figures 1 and 2.
The hydroxyl number is very important for the synthesis of polyurethane. It is defined as the number of milligrams
of potassium hydroxide equivalent to the hydroxyl content of one gram of the sample. The principle of the analytical
method is that the hydroxyl group is esterified with a solution of phthalic anhydride in pyridine. The excess reagent
is back-titrated with standard sodium hydroxide solution, and a blank is run on the reagents to determine the amount
of anhydride consumed. The phenolphthalein is an indicator for the titrate. Because the color of recycled polyol is
dark brown, the faint pink endpoint is not clear, so a potentiometric titrate can be applied.
3
Recycled Polymer Mortar from Recycled Polyurethane and Polyster (RPUMRPE) [4]
It is well known that polyurethanes are used for their excellent adhesion, abrasion, toughness, flexibility and ease of
application. Polyester also has many outstanding characteristics: low-cost, quick set time, high strength and good
durability. The combined use of polyurethane and polyester has been studied by many investigators [6]. The high
reactivity of polyurethane with active hydrogen-containing groups is conveniently used for modifying. The approach
provides a material with very useful properties. The characteristics of polyurethane are eminently demonstrated when
used in the rubbery region which is in contrast to the properties of unsaturated polyester resin, which in generally
used in the glassy region. Polymerized polyurethane-modified polyester has a marked increase in toughness that
combines flexibility and rigidity and also exhibits a great improvement in bond strength as well as reduced shrinkage,
which is very helpful in delamination resistance.
In this work, polyurethane was made from recycled polyol obtained from rigid foam and polyisocyanate. The raw
material for the polyester resin was recovered from scrap PET [5], then depolymerized using different amounts of
ethylene, propylene glycol into glycolized monomer and oligomer. These glycolized products were reacted with
maleic or terephthalic acid to obtained recycled polyester. The proportion of recycled polyurethane to recycled
polyester was varied ranging from 1:1, 1:2, 1:2.5, and 0:1. The combined resins were then mixed with initiator,

24

BAO, WHITNEY AND FOWLER

promoter, sand and flyash to yield recycled polymer mortar. The tensile strengths of the mortar made from the
various combination of resin are as follows:
Table 1. Tensile Strength of RPUMRPE Mortara
Specimens

Tensile Strength, psi (MPa)

Ratio of Polyurethane to Polyester Resin


1:1

1:2

1:2.5

0:1

1
2
3
Avg.
a Measured at 7 days

1586 (10.93)
1684 (11.61)
1540 (10.62)
1603 (11.06)

1722 (11.88)
1702 (11.74)
1715 (11.83)
1713 (11.81)

1540 (10.62)
1770 (12.21)
1704 (11.75)
1671 (11.52)

1560 (10.76)
1580 (10.90)
1570 (10.83)

This tensile strength was measured using dog bone specimens with a 1 in.1 in. (25 mm25 mm) crosssection.
The results of Table 1 show that RPUMRPE mortar exhibits an excellent tensile strength of about 1700 psi (11.72
MPa) after 7 days when the ratio (RPU:RPE) ranges from 1:2 to 1:2.5. It can be seen that recycled polyurethane
improves the tensile strength of the recycled polyester.
4
Selection of Initiator and Promoter
Several tests were conducted to identify the best initiator-promoter system. It is well known that expansion occurs
during the polymerization of polyurethane. The slower the curing rate, the larger the expansion and the lower the
strength that will result. Rapid curing can result in greater strength because the fast curing time limits the extent of
expansion so that the mortar is more dense. Table 2 shows the effect of the initiator-promoter on tensile strength and
expansion. Expansion was measured on the increase in length of the specimens which were dog bone shaped.
Table 2. Effects of the initiator and promoter on tensile strength and expansion
Initiator (Cumene hydroperoxide), %
Promoter (Cobalt naphthenate), %
Specimens
1
2
3
Avg.
Expansion, %

0.29
0.58
0.12
0.23
Tensile strength, psi (MPa)
1100 (7.59)
1175 (8.10)
1100 (7.59)
1250 (8.62)
1080 (7.45)
920 (6.34)
1093 (7.54)
1115 (7.69)
35.8
18.4

0.87
0.35

1.16
0.47

1275 (8.79)
1200 (8.28)
1215 (8.38)
1230 (8.48)
9.5

1275 (8.79)
1240 (8.55)
1350 (9.31)
1288 (8.88)
2.5

The amount of initiator and promoter exhibits a significant effect on curing time as well as strength, and the
promoter has a very definite role (Table 3).
The analogous results can also be seen from effect of benzoyl peroxide and cobalt naphthenate in Table 4.
Although the amount of benzoyl peroxide was reduced from 1 to 0.87%, the tensile strength was increased because
the amount of cobalt naphthenate was increased from 0.23 to 0.34%.
The peroxides were used to compare the effect on expansion and tensile strength: methyl ethyl ketone peroxide,
cumene hydroperoxide and benzoyl peroxide. Methyl ethyl ketone peroxide is a popular initiator for low temperature,
particularly in conjunction with cobalt naphthenate, since it results in fast curing, less expansion and high tensile
strength. In contrast, benzoyl peroxide starts to decompose free radicals at about 122F, therefore, at room
temperature and with small quantities of cobalt naphthenate, longer curing time, larger expansion and very low
strength were obtained as shown in Table 5. If more cobalt naphthenate is used, the greater exothermic
reaction raises the temperature, and benzoyl peroxide can also provide fast curing and higher tensile strength.
Cumene hydroperoxide is a good initiator at room temperature which can combine with cobalt naphthenate to yield
higher strength. The results of test are shown as Table 5.
The curves for viscosity vs. time (Figures 1 and 2) show that the cure time can be controlled by initiator and
promoter. The viscosity of polyols changes depending on temperature, particularly below 20C.

A NEW HYBRID RECYCLED POLYMER MORTAR

25

Table 3. Results of The Amount of Initiator and Promoter on Properties


Cumene hydroperoxide, %
Cobalt naphthenate, %
Expansion, %
Tensile strength, psi (MPa)

0.5
0.07
41.3
870 (6.00)

1
0.07
25.9
916 (6.32)

1
0.15
10
1102 (7.60)

1
0.23
3.1
1553 (10.71)

5
Typical Properties of Recycled P Polyurethane Modified Recycled cycled Polyester Mortar
(RPUMRPE)
Typical properties of RPUMRPE mortar in comparison to the hybrid recycled polyester (HRPE) mortar and recycled
polyester (RPE) are presented in Table 6. The HRPE is hybrid polymer of recycled polyester and polyurethane which
comes from F2 hybrid resin part A made by Amoco. The RPE is recycled polyester made by Alpha corporation. The
bond, tensile and compressive strength, modulus of elasticity and shrinkage of these three kinds of polymer mortar were
measured at 4, 8, 12, and 24 hours respectively.
As shown in Table 6, the RPUMRPE possesses excellent properties. Its bond strength is 2 to 4 times higher than
that of RPE because of the polarity of polyurethane molecules. Its shrinkage is about half of that of RPE due to the
expansion during the polymerization of polyurethane. The tensile strength is also about 1.5 to 2 times greater than
that of RPE. This kind of polymer mortar can used for rapid repair materials because the strength cevelopment is
very fast as shown in Table 7 and Figures 10 and 11.
Table 4. The Effect of The Amount of Benzoyl Peroxide and Cobalt Naphthenate
Benzoyl peroxide, %
Cobalt naphthenate, %
Expansion, %
Tensile strength, psi (MPa)

0.58
0.23
12.3
1040 (7.17)

1
0.23
11.8
1326 (9.14)

0.87
0.34
10.6
1416 (9.77)

Table 5. RPUMRPE with Various Catalyst Systemsa


Initiator

Promoter

Expansion, %

Tensile Strength, psi (MPa)

1% Methyl ethyl ketone peroxide


1% Benzoyl peroxide
1% Cumene hydroperoxide
1% Cumene hydroperoxide
1% Cumene hydroperoxide
0.58% Benzoyl peroxide
1% Benzoyl peroxide
0.87% Benzoyl peroxide
1% Benzoyl peroxide
a Measured at 7 days

0.04% Cobalt naphthenate


0.04% Cobalt naphthenate
0.07% Cobalt naphthenate
0.15% Cobalt naphthenate
0.23% Cobalt naphthenate
0.23% Cobalt naphthenate
0.23% Cobalt naphthenate
0.34% Cobalt naphthenate
0.35% Cobalt naphthenate

4
20.7
25.9
10
3.1
12.3
11.8
10.6
2

1476 (10.17)
153 (1.06)
916 (6.32)
1102 (7.60)
1533 (10.57)
1040 (7.17)
1326 (9.14)
1416 (9.77)
1505 (10.38)

6
Thermal Compatibility Test
For the determination of the thermal compatibility, the ASTM C 88478 test method was followed. A layer of
RPUMRPE mortar was applied to a slab of cured and air dried concrete. After the material cured for one week, the
samples were subjected to five cycles of temperature change. In the first cycle, the specimens were placed in the
environmental chamber at -63F (21.1 1.7C) for 24 hours and then removed to room temperature at 731.8F
(231C) for 24 hours. Three beams with 1/2- to 2 inch (12- to 50-mm) overlays and four slab with 1/2-inch (12-mm)
overlays were tested. No delaminations were found in the specimens. The direct shear tests were conducted to
measure the effect of the thermal cycles. The results are shown in Table 8.
Pull-out tests were also conducted to measure the bond strength between the polymer concrete overlay and the
portland cement concrete substrate before and after the thermal compatibility test. The results are shown in Table 9.

26

BAO, WHITNEY AND FOWLER

According to the ASTM C-884-78, delamination of the polymer mortar layer from the concrete test block or the
presence of horizontal cracks in the concrete near the interface shall constitute failure. It is obvious from these results
that the RPUMRPE mortar passes the ASTM C-884-78. After thermal cycling, no delamination was found in the
specimens. The failure in all specimens in the shear and pull out tests occurred within the portland cement concrete
substrate. These results depended on the tensile strength of the concrete substrate.
Table 6. The Comparison of Main Properties of Three Polymer Mortar Within 24 hrs.
PROPERTIES

TIME, hrs

RPUMRPE

HRPE

RPE

Bond Strength, psi (MPa)


8
12
24
Tensile Strength, psi (MPa)
8
12
24
Compressive Strength, psi (MPa)
8
12
24
Modulus of Elasticity, 106 psi (103 MPa)
8
12
24
Shrinkage, in./in.103
8
12
24

4
501 (3.46)
528 (3.64)
535 (3.69)
4
1378 (9.50)
1492 (10.29)
1540 (10.62)
4
8026 (55.35)
8134 (56.10)
8233 (56.78)
4
2.0 (13.8)
2.12 (14.6)
2.37 (16.3)
4
3.15
3.28
3.30

455 (3.14)
162 (1.12)
213 (1.47)
327 (2.26)
1288 (8.88)
247 (1.70)
383 (2.64)
780 (5.38)
7799 (53.79)
1681 (11.59)
2235 (15.41)
3617 (24.94)
1.89 (13.0)
0.086 (0.593)
0.196 (1.35)
0.493 (3.40)
3.00
0.80
1.40
2.25

94 (0.65)
153 (1.06)
182 (1.26)
253 (1.75)
163 (1.12)
636 (4.39)
791 (5.46)
1027 (7.08)
1150 (7.93)
7213 (49.74)
8209 (56.61)
8925 61.55)
0.0634 (0.437)
1.67 (11.5)
1.75 (12.1)
1.83 (12.6)
0.25
5.65
5.75
5.80

113 (0.78)

457 (3.15)

6674 (46.03)

1.58 (10.9)

5.40

Table 7. Strength Gain of RPUMRPE


4 hrs
Compressive
strength, psi
(MPa)
Percentage of
30-day
strength
Tensile
strength, psi
(MPa)
Percentage of
30-day
strength

8 hrs

12 hrs

1 day

7 days

14 days

30 days

7,799 (53.79) 8,026 (55.35)

8,134 (56.10) 8,233 (56.78) 11,153 (76.


92)

11,157 (76.
94)

11,407 (78.
67)

68

70

71.3

72

97

98

100

1288 (8.88)

1378 (9.50)

1492 (10.29)

1540 (10.62)

1625 (11.21)

1633 (11.26)

1650 (11.38)

78

83.5

90.4

93.3

98.5

99

100

The compressive strength has 68% of the 30 day strength in 4 hours and tensile strength has 78% within 4 hours.
Table 8. Comparison of Shear Strength from Thermal Compatibility Test
Mortar Type

Load to Failure, lbs. (N)

Shear Strength, psi (MPa)

Failure Mode

RPUMRPE
Specimen 1
Specimen 2
Specimen 3
Average

(Before Thermal Cycling)


13,000 (58,000)
12,500 (55,600)
11,000 (49,000)
12,166 (54,114)

1035 (7.14)
994 (6.86)
875 (6.03)
968 (6.68)

PCCa
PCC
PCC

A NEW HYBRID RECYCLED POLYMER MORTAR

Mortar Type
Specimen 1
Specimen 2
Specimen 3
Average
HRPE
Specimen 1
Specimen 2
Specimen 3
Average
Specimen 1
Specimen 2
Specimen 3
Specimen 4
Average
aPCC=portland

Load to Failure, lbs. (N)


(After Thermal Cycling)
12,000 (53,000)
8,900 (40,000)
11,950 (53,150)
10,950 (48,710)
(Before Thermal Cycling)
14,700 (65,400)
9,050 (40,250)
12,900 (57,400)
12,216 (54,337)
(After Thermal Cycling)
7,600 (34,000)
14,800 (65,800)
9,900 (44,000)
11,050 (49,150)
10,837 (48,203)

Shear Strength, psi (MPa)

Failure Mode

955 (6.59)
709 (4.89)
951 (6.56)
871 (6.01)

PCC
PCC
PCC

1169 (8.06)
720 (4.97)
1026 (7.08)
972 (6.70)

PCC
PCC
PCC

605 (4.17)
1177 (8.12)
788 (5.43)
879 (6.06)
861 (5.94)

PCC
PCC
PCC
PCC

27

cement concrete

Table 9. Results of The Tensile Bond Strength From Thermal Comparability Test of RPUMRPE
Load to failure, lbs. (N)

Tensile bond strength, psi (MPa)

(Before Thermal Cycling)


3,620 (16,100)
288 (1.99)
3,600 (16,000)
286 (1.97)
3,610 (16,050)
287 (1.98)
(After Thermal Cycling)
Specimen 1
3490 (15,500)
277 (1.91)
Specimen 2
3520 (15,700)
280 (1.93)
Average
3505 (15,600)
278.5 (1.92)
aEpoxy resin was used to bond the circular steel disc to the slab.
bPortland Cement Concrete
Specimen 1
Specimen 2
Average

Failure mode
Epoxya
PCCb

PCC
PCC

7
The Influence of Foam on Permeability
When polyurethane was polymerized in the presence of moisture, some expansion is inevitable. In order to
investigate the influence of foam on permeability a rapid chloride ion permeability test was conducted in accordance
with AASHTO T-277. Three resin systems: RPUMRPE-1, RPUMRPE-2, and HRPE were used. The permeability
results of 12 slices are shown in Table 8. It can be seen that every specimen is impermeable. After 6 hours, the
permeabilities were the same as at the beginning. Although RPUMRPE-2 contains more isocyanate compound and
more expansion than RPUMRPE-1, it still had excellent impermeability which means the pores are closed, and the
water cannot pass through.
8
Overlay
Several overlays from 0.5 to 1.0 inch (12 to 25 mm) in thickness were cast on beams from 16 to 65 inches (406 to
1650 mm) long for each polymer mortar. Three polymers mortars, RPUMRPE, HRPE, and RPE, were tested. They
were demolded after one day, and it was found that RPUMRPE and HRPE overlays showed no delamination, but the
recycled polyester (RPE) developed a 0.51-mm crack at the interface of the overlay and beam. After 3 months, the
RPE overlay separated from the beam. In contrast, the RPUMRPE and HRPE overlays performed well. Two

28

BAO, WHITNEY AND FOWLER

Fig. 1. Viscosity vs. Time for RPUMRPE With Varied Cobalt Naphthenate
Curve 10.23%; Curve 20.15%; Curve 0.1%

RPUMRPE overlays on beams were placed outdoors and subjected to sunlight exposure. After 6 months no change
has been observed.
Other overlays of RPUMRPE from 1 to 2 inches (25 to 50 mm) in thickness were cast on beams. They also
exhibited no delamination, although the shrinkage stresses increased with increased mortar layer thickness. This
mortar shows good compatibility with concrete substrate.
Table 10. Results of Rapid Chloride Permeability Test of RPUMRPE and HRPE
Data on strain gage channel
Specimen

Time

Time

Time

0 hr

3.5 hrs

6 hrs

0.00002
0.00001
0.00001
0.00001
0.00001
0.00001
0.00001
0.00001
0.00002
0.00001
0.00001
0.00001

0.00001
0.00001
0.00000
0.00000
0.00001
0.00001
0.00001
0.00000
0.00001
0.00002
0.00001
0.00001

1
0.00002
2
0.00001
3
0.00001
4
0.00001
5
0.00001
6
0.00001
7
0.00001
8
0.00001
9
0.00002
10
0.00001
11
0.00001
12
0.00001
*14 RPUMRPE-1; 58 RPUMRPE-2; 912 HRPE

9
Conclusion
Laboratory tests on recycled polyurethane-modified recycled polyester mortar have shown that this is an effective
material for making polymer mortar overlays which combines the advantages of both polyurethane and polyester. This
mortar shows higher bond strength, good tensile and compressive strength, lower shrinkage, and impermeability. The
1- to 2-inch (25- to 50-mm) thickness mortar overlay exhibited no delamination. Because of the fast development of
strength, this mortar can be used for rapid repair of concrete structures.
This has particular significance due to the fact that the materials came from recycled plastic wastes. This not only
alleviates pollution of the environment, but also reduces the cost of the polymer concrete. These two kinds of
recycled polymer can complement each other to produce excellent properties.

A NEW HYBRID RECYCLED POLYMER MORTAR

29

Fig. 2: Viscosity vs. Time for RPUMRPE With Varied Amount Cumene Hydroperoxide
Curve 1:1%; Curve 2:0.5%

Fig. 3. Viscosity vs. Temperature for Polyol

Fig. 4. Comparison of Bond Strength of Three Resin Mortars

10
References
1.
2.
3.
4.
5.
6.

Polyether Diols From Urethane Resin Scrap. Organic and Polymer Waste Reclaiming Encyclopedia, p. 341.
Osamu Kinoshita. Process for Decomposition of A Polyurethane Resin. U.S.P., Vol. 3, p. 530, p. 632.
F.F.Furilla, W.A.Odinak. Reclaiming Scrap Polyisocyanate Foam with an Aliphitic Diol and A Dialkanol Amine. U.S.P.,
Vol. 3, p. 440, p. 708.
K.S.Rebeiz, D.W.Fowler, and D.R.Paul (1991). Making Polymer Concrete with Recycled PET. Plastic Engineering, p. 33.
U.R.Vaidya and V.M.Nadkarai (1987). Unsaturated Polyester Resin Foam Poly(ethylene terephthalate) Waste-Part 1:
Synthesis and Characterization. Ind. Eng. Chem. Res., Vol. 26, pp. 194198.
K.H. Hsieh, J.S.Tsai, K.W.Chang (1991). Interpenetrating Polymer Network of Polyurethane and Unsaturated Polyester.
Journal of Materials Science, Vol. 26., pp. 58775882.

30

BAO, WHITNEY AND FOWLER

Fig. 5. Comparison of Tensile Strength of Three Resin Mortars

Fig. 6. Comparison of Compressive Strength of Three Resin Mortars

Fig. 7. Comparison of Modulus of Elasticity of Three Resin Mortars

A NEW HYBRID RECYCLED POLYMER MORTAR

Fig. 8. Comparison of Shrinkage of Three Resin Mortars

Fig. 9. The Initial Expansion and Shrinkage of RPUMRPE

Fig. 10. Compressive Strength Development of RPUMRPE

31

32

BAO, WHITNEY AND FOWLER

Fig. 11. Tensile Strength Development of RPUMRPE

4
UTILIZATION OF WASTE PLASTICS AS AGGREGATE IN
ASPHALT MIXTURE
M.YAMADA
Department of Civil Engineering, Osaka City University, Osaka, Japan

Abstract
This paper proposes a utilization of waste plastics in asphalt pavement. Laboratory tests conducted to examine
effects of substituting crushed plastics for a portion of the aggregate of an asphalt paving mixture.
Six samples of waste plastics were prepared for the tests. Four of them were obtained from a size reduction
facility for industrial plastics recycling. The size of the plastics particles was about 2 to 10mm. The other two
were obtained by being separated from domestic wastes collected in two cities and crushed to produce the size
of 2 to 5 mm particles. These plastics particles were added to an asphalt mixture in quantities from 5 to 10
percent of the aggregate volume.
Marshall properties, dynamic stability. bending strength and strain at failure of the asphalt mixtures with
each of the plastics aggregates were compared with those of a conventional asphalt mixture. Results showed
that dynamic stability of asphalt mixture with a plastics aggregate which softened at the mixing temperature
was higher than the conventional mixture.
Field tests were also carried out to evaluate the constructability and performance of asphalt mixtures with
plastics aggregates.
Keyword: paving material, asphalt mixture, waste plastics, recycling, Marshall test, wheel tracking test.
dynamic stability, bending test.

1
Introduction
Plastic is used widely for various purposes at home and work from its characteristics: inexpensive. moldable. light,
strong, hygienic. colorful. etc. But. most of plastic goods are used only for a short time and then thrown away. We
use and disuse a large quantity of plastics. The amount of plastics supplied in Japan was about ten mililon tons in
1990. The amount of plastics discharged was 3.13 million tons as domestic wastes. 2.44 million tons as industrial
wastes. and 5.57 million tons in total.
A part of the industrial plastics wastes (about 27 percent in 1987) were reutilized for productions of solid fuel, oil,
gas. monomers and so on, A large quantity of the other plastics wastes were disposed of.
Almost all of domestic wastes, which are filled in land after appropriate treatments or incinerated. But. they are
unsuitable for the landfill as being persistency in land. When they are incinerated, the heat of combustion damages an
incinerator and harmful gases may be caused.

Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

34

YAMADA

Recycling is the best alternative. Much of industrial post-consumer plastics can be recycled by some methods
because they are sorted. Meanwhile, the range of plastics collected from households is enlarged substantially. It is
difficult to recycle unsorted plastics.
Therefore, new technologies for reutilization of unsorted plastics wastes must be developed. This paper proposed a
utilization of waste plastics in asphalt pavement [1][2]. Significant amounts of domestic plastic waste collected by a
self-governing community will be able to be utilized for construction of roads in the community, if this utilization
technology is put to practical use.
2
A conception of utilization of waste plastics in asphalt pavement
Crushed waste plastics may be utilized as a base materials in road. But, the utilization is unrecommendable, because
the characters of plastic do not become effective there. It will be desirable that plastics which is organic are used
together with an asphalt which is organic.
Waste plastics will be utilized in an asphalt mixture in two ways as follows.
Dissolution in the asphalt to utilize as a portion of the binder
Substitution for the crushed stones or sands of the same size in the mixture
The former may be more supportable than the latter. If a proper quantity of an appropriate plastic is dissolved in an
asphalt, properties of the asphalt will be modified to become more suitable for road paving. Maybe, there are some
sorts of appropriate plastics like that in the wastes. However. the proportion of the appropriate plastics to the all
waste plastics will is not high. The optimum volume of the plastics dissolved in the asphalt to modify properties of
the asphalt is not high too. Therefore, the former utilization does not spend much waste plastics and it is not a
effective way for the plastics waste reduction.
The latter may spend comparatively much waste plastics become, because they are utilized as aggregate. If the
waste plastics used as aggregate have a good influence upon the properties of asphalt mixture. the latter is an effective
way for plastics waste reduction.
The particles of crushed plastics will present one of the following three states in a hot asphalt mixture, where
asphalt. filler and aggregate are mixed at 150 to 170C.
1. Very soft and soluble in the asphalt
2. Soft and plastic-deformable, but insoluble in the asphalt
3. Almost unchanged and only elastic-deformable, of course insoluble in the asphalt
It is probable that there are the three sorts of plastics in wastes which are discarded from households, and among the
three the most numerous are of the second sort.
Plastics of the first sort are soluble in the asphalt at hot-mixing and have the effect of raising the viscosity of the
asphalt. But. too high viscosity of the asphalt binder may produce bad effects on the workability at construction and
the flexibility of the asphalt mixture in pavement.
Plastics of the second sort are insoluble in the asphalt and have little effect on the properties of the asphalt, but
they may have some effects on the mechanical properties of the asphalt mixture. The particles of the plastics become
soft and plastic at hot-mixing for making the asphalt mixture and deform at compaction, in between the aggregate
particles. They become back to hard and elastic upon cooling. The elastic particles in between the aggregate particles
will restrict the movement of the aggregate particles, and consequently the resistance to permanent deformation of
the asphalt mixture will become higher.
As plastics of the third sort are like aggregate particles, the addition of them will have little influence on the
mechanical properties of the asphalt mixture, if the percentage of the addition is not too high. When the particles of
plastics are too much or large-sized, the strength of the asphalt mixture will drop. because the particles of plastics are
softer than the particles of the aggregates at even low temperature.
The effect of the second sort plastics is worth notice. It is expected that the effect will avail the improvement of
asphalt mixtures and the utilization of plastics wastes.

WASTE PLASTICS FOR ASPHALT MIXTURE

35

Fig.1. Domestic waste plastics

3
Plastics samples and experimental procedures
3.1
Plastics samples tested in this study
Six samples of waste plastics were prepared for this study. Four of them were obtained from a size reduction facility
for industrial plastics recycling. The size of the plastics particles was about 2 to 10 mm. The other two were obtained
by being separated from domestic wastes collected in two cities (see Fig. 1) and crushed to produce the size of 2 to 5
mm particles. The sort and properties of them are them are shown in Table 1.
3. 2
Making of asphalt mixtures with the waste plastics
The type of asphalt mixture used in this study was a densegraded asphalt mixture, which is one of typical asphalt
mixtures in Japan, and the maximum particle size of the aggregates was 13 mm. The asphalt mixture is composed of
asphalt binder, filler and aggregates.
As the asphalt binder, a straight asphalt which complies with the quality requirements specified in JIS K 2207, having
a penetration grade of 60 to 80, was used.
As the filler, a mineral powder produced by crushing a lime stones was used.
As the aggregates, crushed sand stones and a river-mouth sand, and crushed waste plastics were used. The plastics
aggregates were substituted for a portion of the aggregates of the same size as them, in quantities from 0 to 20
percent of the tatal aggeregate volume.
The design asphlt contents of the asphalt mixtures with the waste plastics were determined by the procedure in the
Manual for Asphalt Pavement by Japan Road Association where is applied to the conventional asphalt mixtures.
Mixing of the asphalt mixture materials was conducted for 3 minutes at 152 to 157C in a mixer. Compaction of
the asphalt mixture into a mold was conducted at 140 to 144C with the Marshall test rammer or a roller compactor.
Table 1. Plastic samples tested in this study
Plastic sample No.

Industrial/Domectic

Ind.

Ind.

Ind.

Ind.

Dom.

Dom.

Sort of plastic

PE

PE

PE

PVC

Mix

Mix

16
55
24
4
0.99
120130

4
71
23
2
0.76
130140

4
73
18
5
0.89
150160

0
21
58
17
1.36
over 200

0
9
78
13
0.93
120160

0.89
13016

Weight percent (%)


1013
510
2.55
1.22.5
Specific gravity
Softening tem. (C)

Size (mm)
5
51
41
3

36

YAMADA

Fig. 2. Apparatus for the wheel tracking test

3.3
Test Methods
3.3.1
Method for Marshall test
Marshall test specimens of asphalt mixtures were prepared with the asphalt content varying at 0.5 percent intervals
within the range of 4.5 to 6.5 percent. Density, stability and flow value of the specimens were measured to calculate
the percentage of air void and voids filled with asphalt and to determinate the design asphalt content.
3.3.2
Method for wheel tracking test
Wheel tracking test specimens were prepared by compacting asphalt mixtures into a mold of 300 by 300 mm, and 50
mm in depth, with a roller compactor. The test apparatus is a machine in which a loaded wheel with a solid rubber
tire of 200 mm diameter and 50 mm wide is rolled to and fro on a specimen of an asphalt mixture. A photograph of
the apparatus is given as Fig. 2.
In order to evaluate the resistance to deformation, the testing wheel loaded with 686 N was shuttled across the
center of the specimen 42 rounds per minute at 60 C, and the dynamic stability (DS) (passes/mm) was measured by
using the following equation.
(1)
where d1 is the deformation at 45 minutes (=t1) and d2 is the deformation at 60 minutes (=t2).

WASTE PLASTICS FOR ASPHALT MIXTURE

37

3.3.3
Method for bending test
Bending test specimens were prepared by cutting to the final dimensions of 40 mm wide, 40 mm high and 300 mm
long. The testing equipment used was a four-point bending apparatus made for constant deflection rate tests. The
span length was 270 mm. The rate of the deflection was 32 mm per minute. The specimen temperature during the
test was 10 to 20C. The force and deflection were measured to calculate the stress and strain and by using the formula
deduced from the elastic theory.
4
Experimental results
4.1
Results of Marshall test
4.1.1
State of compaction
Influence of mixing of a plastic on the state of compaction in the asphalt mixture varied with sorts of plastics. Fig. 3
shows the relation air void percent of the compacted asphalt mixture with a quantity of each plastic and asphalt. The
air void percent of compacted mixture decreased as the softening temperature of plastic mixed is higher. This
suggests that careful trial mixing must be conducted to check the quality of the mixture, especially the air void,
before construction.
4. 1. 2
Marshall stability and flow value
The Marshall stability of all asphalt mixtures with plastics was over the lower limit value (=7.35kN) specified in the
Munual for Asphalt Pavement. But. the flow value of mixtures with plastics was larger than the conventional mixture
and that of some of the mixtures was over the upper limit (=40/100mm) specified in the Munual.
4.1.3
Design asphalt contact
The relation between design asphalt contect and volume percent of plastic in the mixtures is shown in Fig. 4. The
design asphalt content was determined in the range of asphalt contents which satisfied the standards specified in
Table 2.
The design asphalt content of mixtures with the plastic sample-1, which has the low softening temperature as
shown in Fig. 1. decreased as the volume percent of plastic was higher. However, it is doubtful whether the low
design asphalt content is the optimum value which must be adopted for construction, because the reduction of
asphalt content may bring deterioration of durability of the asphalt mixture. That must be require further
investigation.
4.2
Results of wheel tracking test
Table 3 shows the dynamic stabilities of asphalt mixtures with six plastics listed in Table 1, where the asphalt
content was 5.5 or 6.0 percent. and the volume percent of plastic was 5 or 10 percent. The all dynamic stabilities,
except of the plastic sample-4, were higher than the value of the mixture with no plastic.
Especially the dynamic stabilities of the mixtures with the plastic sample-1 and 2 were so high as to be over 20
thousands passes per minute in case of 5.5 percent asphalt content. It is assumed that these plastics were of the first
sort mentioned in the paragraph 2. and the dynamic stabilities increased remarkably because a part of plastics dissolved
in the asphalt.

38

YAMADA

Fig. 3. Air void percent of asphalt mixtures

The dynamic stability of the mixture with the plastic sample-4 was nearly equal to that of the mixture with no
plastic. It is assumed that this plastic was of the third sort
Table 2. Standards specified to determine the design asphalt content of a dense-graded asphalt mixture with plastic in this study
Number of blows
Percent of air voids
Voids filled with asphalt
Load at 40 mm flow value

75 blows
36 %
7085 %
7.35 kN

WASTE PLASTICS FOR ASPHALT MIXTURE

39

Fig. 4. Design asphalt content determined by Marshall test


Table 3. Dynamic stabilities of asphalt mixtures measured by wheel tracking test
Sample No.

Pla.

vol.%

10

10

10

10

10

10

D S (passes /mm)
6.0

Asp. w%
5.5
7400

22000
8000

21000
5100

8100
330

480
1100

3700
2200

7400
-

1100
-

5000
300

410

and had little effect on the dynamic stability.


The dynamic stabilities of the mixtures with the plastic sample-5 and 6, from domestic wastes. were moderate
values, which are within the range recommended for heavy traffic road pavements.
4.3
Results of bending test
The relations between bending strength and temperature are shown in Fig. 5 and Fig. 6. The bending strength of the
asphalt mixture with the plastic sample-1 was higher than that of the mixture with no plastic, at the all testing
temperature. The bending strengthes of the mixtures with the other three plastic samples trended to be lower than
that of the mixture with no plastic. The peak of strength. which appeared in the strength-temperature relation of the
mixtures with the sample-1, 3, and 4, and no plastic, disappeared in that with the sample-2.
The strength of the asphalt mixture with plastic will be affected by the hardness and state of softening of plastic
particles, and the state of bonding between plastic and asphalt. It does not necessarily follow that a higher strength of
asphalt mixture is better for asphalt pavements. However, lowering of the strength caused by the bad bonding
between aggregate and asphalt has a harmful influence upon the durability of asphalt mixtures.
Fig. 7 and Fig. 8 show the relations between strain at failure and temperature. The strains at failure of the mixtures
with the plastic sample-1 and 2, which had high dynamic stabilities, were larger at the low temperatures and smaller
at the high temperatures than the mixture with no plastic. This suggests that the temperature susceptibility became
lower because a part of the plastics dissolved in the asphalt.
5
Test constructions in field
The test paving constructions in two places mentioned below were carried out to evaluate the constructability and
performance of the asphalt mixture with plastics. In both places. the surfaces of constructed pavement were enough
flat. but less lustrous them the normal asphalt pavement. A comment of the constructors was that rakers felt a little
heavier and the surface was hair-crackable at compaction by a roller.
1. A paving construction at a roadway in Higashi-Osaka City. Osaka Prefecture (Fig. 9.)

40

YAMADA

Fig. 5. Relations between bendig strength and temperature (asphalt content: 5.5%)

Fig. 6. Relations between bendig strength and temperature (asphalt content: 6.0%)

Fig. 7. Relations between strain at failure and temperature (asphalt content: 5.5%)

An asphalt mixture with the plastic sample-5 was used for paving the surface course of 50 mm in thickness and 443
m2 in area in September, 1993. The total thickness of asphalt mixture layers was 200 mm. The content of plastic was
8.3 volume percent (3.2 weight percent).
As this place is a heavy traffic road, the resistance to rutting, and the performance of the mixture will be evaluated
by the further follow-up survey.
2. A paving construction at a parking zone in Kusatsu City, Shiga Prefecture (Fig.10.)
An asphalt mixture with the plastic sample-6 was used for paving all the asphalt mixture layer of 50 mm in thickness
and 520 m in area in March, 1994. The content of plastic was 10 volume percent (3.6 weight percent).

WASTE PLASTICS FOR ASPHALT MIXTURE

41

Fig. 8. Relations between strain at failure and temperature (asphalt content: 6.0%)

Fig. 9. Construction at a roadway in Higashi-Osaka City

As the asphalt mixture layer is thin. the pavement may be deteriorated early. However, if the utilization like this
case is suitable. many self-governing communities may choose the separate collection of domestic plactis wastes to
use widely for road constructions.
6
Conclusion
From the experimental investigation, it was found that dynamic stability of asphalt mixture with a plastics aggregate
which softens at the mixing temperature is higher than the conventional mixture, and crushed domestic waste plastics
may be utilize as the plastics aggregate.
7
References
1.
2.

Yamada, M. and Inaba, Y. (1994) Utilizationof waste plastics in asphalt mixture, Hoso (Pavement), Vol. 29. No. 7,
pp. 1722 (in Japanese).
Yamada, M. (1994) Use of waste plastics-aggregate in hot-mix asphalt, Journal of Japan Institute of Aggregate
Technology, Vol. 26, No. 102, pp. 7175 (in Japanese).

42

YAMADA

Fig. 10. Construction at a parking zone in Kusatsu City

5
THE BEHAVIOR OF PORTLAND CEMENT CONCRETE WITH
THE INCORPORATION OF WASTE PLASTIC FILLERS
D.SANDER, D.W.FOWLER and R.L.CARRASQUILLO
Department of Civil Engineering, The University of Texas at Austin, Austin, USA

Abstract
There is an abundance of waste plastic which is becoming more difficult to dispose of in landfills because of
stricter environmental regulations. Some of the plastic waste contains heavy metals such as lead. These wastes,
including high and low density polyethylene, PET, polypropylene, PVC, polystyrene, and vent dust (PVC with
inorganics) were used as filler in portland cement concrete. The variables were 1) amount of plastic filler, 2)
water-cement ratio, and 3) type of plaster. Tests included compressive strength, impact, abrasion resistance, deicing, scaling, and chloride content. Hollow concrete masonry units were made and tested in compression. The
results are presented.
Keywords: Portland cement concrete, waste plastic, recycled materials
1
Introduction
In 1976 plastic became the most widely used material in the United States. It represents a $140-billion-per-year
industry8, greater than steel and aluminum combined. Unfortunately the increase in the use of plastic has affected the
problem of waste plastic disposal at about the same rate. The need to recycle plastics to reduce the use of virgin resins
and to reduce the amount of waste materials has become a subject of considerable interest.
In the last 15 years the amount of plastic solid waste has more than doubled. In 1988 it represented about eight
percent by weight of the nations solid waste stream and 20 percent of the volume of waste in the municipal landfills.
By the year 2010 plastic will contribute 10 percent of this weight, according to industry projections8.
There are three ways of disposing solid waste, including plastic. One is in landfills, another is incineration, and the
third is by recycling. In 1988, 73 percent of solid waste was placed in landfills. The other 27 percent was either
incinerated or recycled.
There are four methods to recycle plastics: thermal reprocessing, chemical modification or depolymerization,
chopping and cutting into small pieces for using as fillers, and heat recovery. Thermal reprocessing consists of
heating thermoplastics to the melting point. The plastic is then converted into a new product while it is cooling.
Chemical modification or depolymerization is the method used for recovering the basics chemicals from the plastic
waste. In this case depolymerization occurs when the plastic is heated, along with a catalyst, to remove either water
or methanol. The use of plastics as fillers in other materials such as concrete represents a third method. By shredding
the plastic into smaller pieces and using it as a filler the chemical composition of the plastic is less important in this
case. It is by far the easiest method for recycling plastics whether contaminated or not. The fourth method is
retrieving the heat content of plastic wastes by incineration. Between 10 and 15 percent of all municipal waste is

Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0419 20550 0.

44

SANDER, FOWLER, CARRASQUILLO

being incinerated in the US. Incineration with heat recovery, however, is being opposed by the public because of
potential emissions of furans, dioxins and heavy metals.
Of these four methods, none alone will solve the problem of the disposal of plastics. It is important, rather, to find
the best combination of all four methods that will reduce the amount of waste plastic to the lowest possible
percentage.
Past investigations at the University of Texas at Austin (UT)11 and at the Center for Recycling Research have
demonstrated, however, that plastics added as a filler to concrete tend to reduce the compressive strength of the
concrete but also to increase the resistance to freezing and thawing cycles. The results of the past investigations were
not conclusive and further research was needed for evaluating the influence of waste plastic when added to PCC. The
purpose of this investigation was to identify:
1. Methods to test the influence of waste plastic in different shapes when it is added into the PCC matrix,
2. Types of plastics which could enhance certain PCC properties, and
3. The effect of different levels, shapes and sizes of plastic fillers on PCC properties.
This research investigated the possibility of using the postconsumer waste plastics as an organic filler in PCC mixes.
Eight postconsumer plastics were identified and used to test the effect of different plastic types, shapes, sizes and
levels, on a standard portland cement concrete (PCC) mix. The plastics were: high density polyethylene (HDPE) in
flake and fine form; low density polyethylene (LDPE); polyethylene terephthalate (PET); polypropylene (PP);
polystyrene (PS); polyvinyl choloride (PVC); and vent dust (VD). Vent dust, produced in manufacturing operations,
consists of 10 to 15% PVC, 40 to 45% CaCO3, 20 to 30% TiO2, 15 to 20% A1(OH)3, ~4% Pb stabilizer, and 1 to 2%
Sb2O3. Compressive strength, abrasion resistance, impact resistance and deicer scaling resistance were the properties
which were tested. Four plastics were incorporated in the mix of concrete masonry units (CMUs) and tested under
compressive load.
2
Experimental Program
2.1
Mix Design
Table 1 shows the three mix designs the six different tests performed to the plastic modified concrete. Axial
compression on concrete cylinders was tested in two different mix designs, No. 1 and No. 2. The other properties
were examined only for one of the mix designs.
2.2
Proportioning of Plastic Fillers
Three levels of plastic filler were examined for each of the tests: 0, 1 and 4 percent by weight. Except for the vent
dust and the HDPEfine, the rest of the fillers replaced 1 and 4 percent of the weight of the pea gravel. Vent dust and
the HDPEfine substituted for the sand. Table 2 presents the plastic and the aggregate quantities used in mix No. 1
and No. 2. The cement factors and the water-cement ratios were given in Table 1.
The proportion of materials for the batches made to cast the masonry units is given in Table 3. The levels of
plastics were reduced to 2 and 4 percent of the volume of the sand. The plastic fillers were, in this case, added to the
mix.
2.3
Mixing Procedure
Except for the CMUs which were produced in an industrial block plant, the batches were completed following the
same procedure. Before starting a typical concrete batch, the water, sand and pea gravel were adjusted for moisture
content. The two aggregates and the plastic filler were introduced into the mixer and mixed for approximately 30
seconds. The cement and two-thirds of the total water were added and the drum was rotated for 2 minutes more. The
rest of the water was added within the next 30 seconds. After two minutes the drum was stopped, and the slump test
was performed. The drum was then scraped to loosen any material not previously mixed. It was restarted and rotated

PORTLAND CEMENT WITH WASTE PLASTIC FILLERS

45

for one minute more. The concrete was then scooped in the molds in three layers, consolidated on the vibrating table,
and the top surface was troweled. The specimens were moist cured for 24 hours, after which they were demolded and
placed in a curing chamber for 28 days.
Table 1. Basic Mix Designs for Control Specimens
Mix Number

Proportions

Test

Cement=570 lb/cy
w/c=0.46
No. Plastics=8

Cement=875 1b/cy
w/c=0.40
No. Plastics=6
1:2.75
Cement:Sand by Volume
No. Plastics=4

Axial Compression
Abrasion Resistance
Deicing Scaling Resistance
Chloride Content
Axial Compression
Impact Resistance
Slump
Axial Compression for CMUs

2.4
Test Procedures
2.4.1
Compressive Strength Test
For this test 3-in. by 6 in. (75-mm by 150-mm) concrete cylinders were cast following the method described in
ASTM C 3986. Mix designs No. 1 and No. 2 were evaluated under axial compressive load.
2.4.2
Impact Test
This test method was proposed by Senbetta16. Its original purpose was to evaluate the impact resistance of floor
materials in terms of the capacity to resist surface disintegration due to repetitive low level impact. Instead of floor
materials, this test was employed to measure the influence of plastic fillers on the concrete resistance to repetitive
low impact. The specimens which were used in the test were nine 2-in. (50-mm) concrete cubes made with mix No.
2. The test uses the Los Angeles (LA) abrasion machine, which is traditionally utilized to test aggregates according
to ASTM C 131 and C 535 specifications.
Eighteen 2-in. (50-mm) diameter steel balls were included to provide impact to the concrete cubes in addition to
the free fall that occurs when the drum is rotating. The number of specimens per batch was eighteen 2-in. (50-mm)
cubes, which were cast and cured in a moisture chamber for 28 days according to ASTM C 19290a.
The test started after 9 of the 18 cubes were weighed and placed into the LA machine with the 18 steel balls. The
machine was then turned on to allow the drum to rotate. At the completion of every 500 revolutions the specimens
were taken out, weighed and replaced in the drum until a total of 2000 revolutions were completed. The drum had to
be cleaned from debris prior to beginning of each 500-revolution cycle. This procedure was repeated with the second
set of 9 cubes. The final weight was taken from the average of the two groups of 9 cubes.
Table 2. Plastic Fillers for Mix No. 1 and No. 2
Mix Designation

Mix No. 1

Mix No. 2

Plastic, lb/cy

Coarse, lb/cy

Fine, lb/cy

Plastic, lb/cy

Coarse, lb/cy

Fine, lb/cy

1.
2.

Control-01
HDPEflake-1
HDPEflake-4
LDPE-1
LDPE-4
PET-1
PET-4
PP-1

0
12.7
51.0
12.7
51.0
12.7
51.0
12.7

1274
1243
1141
1240
1129
1246
1154
1240

1845
1845
1845
1845
1845
1845
1845
1845

0
N.A.*

1151
N.A.*

1412
N.A.*

N.A.*

N.A.*

N.A.*

11.5
46.1
11.5

1122
1043
1115

1412
1412
1412

3.
4.
5.

46

SANDER, FOWLER, CARRASQUILLO

Mix Designation
Plastic, lb/cy
6.
7.
8.
9.

Mix No. 1
Coarse, lb/cy
PP-4
PS-1
PS-4
PVC-1
PVC-4
HDPEfine-1
HDPEfine-4
VD-1
VD-4

Fine, lb/cy
51.0
12.7
51.0
12.7
51.0
18.5
51.0
18.5
73.8

Mix No. 2
Plastic, lb/cy
1129
1246
1154
1261
1148
1274
1274
1274
1274

Coarse, lb/cy
1845
1845
1845
1845
1845
1816
1730
1798
1654

*not available

Fine, lb/cy
46.1
1016
11.5
1118
46.1
1042
11.5
1126
46.1
1061
14.1
1151
56.1
1151
14.1
1151
56.1
1151
1 lb/cy=0.593 kg/m3

1412
1412
1412
1412
1412
1371
1263
1383
1326

Table 3. Plastic Fillers for Mix No. 3


Mix Designation

Plastic, % vol.

Cement:Sand, by vol.

Control-0
PET-2
PET-4
PP-2
PP-4
PS-2
PS-4
VD-1
VD-2

0
2
4
2
4
2
4
1
2

1:2.75
1:2.75
1:2.75
1:2.75
1:2.75

2.4.3
Abrasion Resistance Test
The test was conducted in accordance with ASTM C 94490a Standard of Abrasion Resistance of Concrete by the
Rotating-Cutter Method. This method measures the capacity of wear resistance of a concrete surface. The specimens
were 5 in. (127 mm) long, 4 1/2 in. (114 mm) wide and 1 1/2 in. (38 mm) thick and were cut from the tops of the
beams cast for the deicing scaling test. The rotating-cutter was applied to the interior cut surface and not to the toolfinished surface of the sample.
2.4.4
Deicing Scaling Test
The scaling test followed ASTM C 67291a Standard Test for Measuring Scaling Resistance of Concrete Surfaces
Exposed to Deicing Chemicals. Three beams 12 in. (305 mm) long, 6 in. (152 mm) high and 4 1/2 in. (114 mm) wide
were cast for each of the concrete-plastic combinations.
2.4.5
Chloride Content Test
This test followed ASTM C 115290 Standard Test Method for Acid-Soluble Chloride in Mortar and Concrete. The
chloride content was investigated for the samples subjected to the deicing scaling test. Chloride measurements were
taken after 25 freeze-thaw cycles. With an electric rotary-hammer powder samples were taken from concrete at four
different depths: 0 to 12mm, 12 to 25mm, 25 to 38mm, and 38 to 50mm.
2.4.6
Compressive Strength Test for Concrete Masonry Units
For the compressive strength test two concrete masonry units (CMUs) were cast for each plastic combination. The
CMUs were Grade N according to ASTM C 9085. The mixing procedure used to cast the CMUs at an industrial
block plant in San Antonio was different from the procedure normally used at the plant Cement, sand, plastic and

PORTLAND CEMENT WITH WASTE PLASTIC FILLERS

47

Fig. 1. Variation of the Slump when Plastic Content Changes

water were mixed together in a wheelbarrow because the amount of materials needed for each plastic batch was too
small to use the tank which premixed normally sand and cement together. Adding the plastic in the premix container
would have contaminated the material for the subsequent commercial blocks. The water was adjusted according to
the minimum required by the mix to keep the CMUs together after demolding them. The mix was cast in a double-unit
mold and was vibrated for approximately two 5-sec periods. After vibrating the molds, the units were pushed out of
the mold and placed on a shelf. The entire procedure was done automatically by a hydraulically-driven system. The
CMUs were then steam cured for 12 hours and then placed outside. The units were capped with Hydrostone and after
75 days the units were tested in compression in accordance with ASTM C 14091 Standard Method for Sampling
and Testing CMU.
3
Test Results and Discussion
3.1
Workability
Figure 1 presents the slump test data of mix No. 2. The figure indicates that one percent of plastic generally did not
change significantly the workability of fresh concrete. When the plastic content was increased to 4 percent, the slump
decreased in all cases except for the batch with PET.
No noticeable segregation of the mix was found to be produced by the fillers. In fact, the cohesiveness of the mix
was greatly improved with vent dust. During the mixing of the many batches none of the plastic fillers was observed
to flow to the top. An analysis of the concrete samples showed that the mixing procedure had evenly distributed the
plastic chips.
3.2
Compressive Strength
The effect of plastic fillers on the axial compressive strength was measured on two different concretes. The control
specimens of mix No. 1 reached 4150 psi (28.62 MPa) and of mix No. 2 reached 6740 psi (46.48 MPa). The use of a
high strength concrete was for the purpose of determining if a stronger paste would partially nullify the negative
effect of an unbonded filler in the concrete matrix. If true, higher concrete strength would be less affected by the
presence of a plastic filler than a lower concrete strength.
Size and shape of one of the fillers (HDPE) were changed to measure the effect on the strength. It was theorized
that a fine-graded filler like HDPEfine (FM=2.24) with a large unbonded surface area would have a different effect
on the strength than a large-sized filler like HDPEflake (FM=4.89) with less surface area. The tests were designed to
determine this effect.
Figure 2 shows the percentage of compressive strength for mixes No. 1 and No. 2 for 4 percent plastic. It is observed
that the differences in percentage strength between both mixes are not significant

48

SANDER, FOWLER, CARRASQUILLO

Fig. 2. Percent Compressive Strength of Concretes with Four Percent Plastic

From the same figure it is possible to analyze the size effect of HDPEflake and HDPEfine. For mix No. 1 with the
two plastic levels, it is observed that the fine graded filler caused a bigger reduction in the strength than the strength
decrease produced by the HDPEflake.
If vent dust is excluded, one percent of plastic filler changed the strength from 110 percent (HDPEflake-1) to 92
percent (PS-1) of the control value. By incorporating 4 percent of plastic, the concrete strength varied from 99
percent (PET-4) to 82 percent (PVC-4) of the reference strength.
The reason why vent dust reduced compressive strength 25 percent, with only a 1 percent plastic content, deserves
a separate analysis. Two possible reasons for the low strength level are:
1. Vent dust reacted with part of the cement affecting thereby the subsequent hydration. This possibility could exist
because of the presence of calcium carbonate (CaCO3) and aluminum hydroxide (Al(OH)3) as components of
the vent dust.
2. Vent dust particles (<200m) adhered to the surface of the aggregates and interfered with the paste to aggregate
bonding. During the mixing process the vent dust particles were observed to have covered the pea gravel surface
before the addition of the mixing water.
3.3
Impact Test
In Figure 3, the typical types of response to repetitive low impact loading when the content level of plastic was
increased from 0 to 1 and from 1 to 4 percent by weight are shown for three types of plastic. The polyester concrete
(PET-1 and PET-4) and polystyrene concrete (PS-1 and PS-4) behave the same as the control. The plastic content did
not change the impact resistance of these two plastic-modified concretes more than 5 percent from the plain concrete
response.
It was found that increasing amounts of HDPEfine enhanced the impact response of concrete. One percent of
HDPEfine produced a concrete with 8 percent more weight at the end of the 2000 revolutions than the control mix
weight percentage. Meanwhile the HDPEfine-4 concrete finished with 17 percent more weight than the reference
concrete weight percentage.
One percent of polypropylene produced a concrete with lower impact resistance than plain concrete. When PP
content was increased to 4 percent, the impact resistance improved to a level similar to plain concrete. Concrete with
vent dust behaved similar to concrete with polypropylene. However, VD-4 modified concrete did not reach the level
of plain concrete although PP-4 did.
The addition of plastic fillers lowered the material strength, but improved the deformation capacity. A possible
explanation may be that the fillers act as small energy absorbers of the impact load.

PORTLAND CEMENT WITH WASTE PLASTIC FILLERS

49

Fig. 3. Percent of Original Weight after 2000 Revolutions versus Plastic Content for the Impact Test

Fig. 4. Comparison of Seven Different Plastic-Concrete Combinations After Six Minutes of Abrasion

3.4
Abrasion Resistance Test
Figure 4 shows the different responses to abrasion at the end of the tests for filler contents of 0, 1 and 4 percent All
plastic fillers except vent dust reduced abrasion. Vent dust caused an increase in the abrasion when only 1 percent
was incorporated in the concrete, but it reduced the abrasion when the content was increased to 4 percent.
3.5
Deicing Scaling Test
Three types of responses to deicing scaling were observed depending on the content level of plastic in the concrete
(Fig. 5). The first tendency is represented by PET, PP, LDPE and HDPEflake concretes. The fillers were all in flake
form with approximately the same size and fineness modulus. These plastics increased the scaling when their level was
increased in the concrete. The slope of the curves indicate that more than 4 percent of plastic would perhaps increase
the scaling on the concrete surface.
VD-1 concrete suffered less scaling than the reference concrete. But the addition of vent dust beyond 1 percent
increased the scaling to that of plain concrete when the level was 4 percent.

50

SANDER, FOWLER, CARRASQUILLO

Fig. 5. Scaling Rating after 25 Cycles in Seven Similar Responses of Fillers

Fig. 6. Variation of Chloride Content with Depth with One and Four Percent of HDPEflake in PCC

Polystyrene (PS) and polyvinyl chloride (PVC) did not produce any significant change in the scaling rating even with
increased levels.
It is clear from the results that of all the plastic fillers only the VD-1 concrete performed better than the control
mix. It could be that the ability to absorb the stresses due to temperature changes was enhanced by the presence of a
small amount of vent dust.
3.6
Chloride Content Test
The chloride profiles for several of the 14 plastic-concrete combinations tested are presented in Figures 6 through 8.
Each figure shows three curves which depict the control concrete, the 1-percent filler concrete and the 4-percent
filler concrete.
As was expected, the chloride content decreased with the increase in depth. The accepted corrosion limit for
chloride content in concrete is 1.5 lb/cy. The highest values corresponded to the specimens containing 4 percent of
HDPEflake (6.27 lb/cy), and 4 percent of PET (6.25 lb/cy). The concretes containing these same fillers in a 1percent proportion also reached high chloride levels, e.g. 5.16 and 5.11 lb/cy of concrete. The control specimens
reached a chloride level of 5.67 lb/cy which is approximately the average between the two values of HDPE and PET
specimens. The rest of the plastic-concrete combinations exhibited a lower chloride level than the control specimens
but always above the corrosion limit of 1.5 lb/cy of concrete.

PORTLAND CEMENT WITH WASTE PLASTIC FILLERS

51

Fig. 7. Variation of Chloride Content with Depth with One and Four Percent of PP in PCC

Fig. 8. Variation of Chloride Content with Depth with One and Four Percent of PVC in PCC

3.7
Compressive Strength of Concrete Masonry Units
Table 4 shows the average compressive strength for the two CMUs cast for each level of plastic in the concrete. The
compressive strength was calculated based on the gross cross-sectional area of each unit. None of the values reached
the ASTM required minimum of 1000 psi. Except for the PS-loaded blocks, higher plastic content lead to lower
compressive strength. Based on previous results, the control specimens should have reached a higher or a similar
strength than the rest of the plastic-loaded CMUs. Due to scheduling difficulties at the plant it was not possible to
cast more CMUs. It is recommended that this test be repeated in order to obtain accurate data from CMU strength testing.
4
Conclusions
4.1
Workability
Slump was most affected by the size and shape of the plastic particles. Typically, small size and high angularity of the
fillers caused a reduction in slump of the concrete batch. There were no noticeable effects on slump when 1 percent
plastic was used regardless of the particle size and shape. A 4-percent plastic level, however, caused a reduction in

52

SANDER, FOWLER, CARRASQUILLO

the slump from 10 to 77 percent, depending on the filler. No noticeable segregation of the mix was found to be
produced by the fillers. In fact, the cohesiveness of the mix was greatly improved with vent dust.
4.2
Concrete Strength
The effect of the fillers on the concrete strength was greatly influenced by the size the shape of the plastic particles.
With 1 percent (wt) of plastic the flake form fillers did not reduce concrete strength more than 8 percent (PS-1) of the
control value. In fact, HDPEflake-1, PET-1 and PVC-1 reached a slightly higher strength value than the control
strength. When the plastic level was increased to 4 percent (wt), the flake form plastics caused a strength reduction
between 1 and 18 percent of the reference value. The results suggest that the more similar the size and the shape of
the filler and of the coarse aggregate are, the smaller the reduction in the axial strength will be.
The small size fillers like HDPEfine and VD caused a strength reduction which can be traced also to their surface
area. Vent dust proved to be very harmful to concrete strength even with just 1 percent of content level. Two factors
may interact to reduce the strength. One is the lack of bonding between the plastic particles and the concrete paste.
The second factor is the increased surface area of the plastic particles.
4.3
Impact Resistance
A proposed impact test using concrete cubes in the Los Angeles abrasion machine proved to be an effective way for
testing the resistance of concrete subjected to repetitive low impact loading. Only HDPEfine improved the impact
resistance of concrete. The impact resistance of concrete was not affected with the incorporation of PET and PS
flakes. Vent dust reduced the impact resistance of the concrete samples for both content levels.
4.4
Abrasion Resistance
The abrasion resistance of concrete was generally enhanced with the fillers. An increase in the plastic level from 1 to
4 percent improved the resistance in all cases.
4.5
Scaling Resistance
Except for PVC-1 and VD-1, one percent of plastic filler lowered the scaling resistance of concrete. Four percent of
plastic increased the scaling of the samples except for the case of PS-4, PVC-4 and VD-4.
4.6
Chloride Penetration
For the 25 freeze-thaw cycles, it was found that plastic fillers did not change the chloride penetration profile of
concrete. More cycles are needed to confirm this result and to explore effects with a larger number of cycles.
Table 4. Compressive Strength of CMUs with Four Plastic Fillers*
Mix Designation

Average Compressive Strength, psi (MPa)

Control
PET-2**
PET-4**
PP-2**
PP-4**
PS-2**
PS-4**
VD-1**
VD-2**

605 (4.17)
940 (6.48)
780 (5.38)
715 (4.93)
520 (3.59)
470 (3.24)
710 (4.90)
760 (5.24)
670 (4.62)

PORTLAND CEMENT WITH WASTE PLASTIC FILLERS

Mix Designation

53

Average Compressive Strength, psi (MPa)

* two replications for each mix


** the numbers represent the percent by volume of plastic added

5
References
1.
2.
3.
4.
5.
6.

7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.

ASME, The Materials Division (1991) Plastics and Plastics Composites: Materials Properties, Part Performance and
Process Simulation, V.J.Stokes GE Corporate Research and Development, Winter Annual Meeting of ASME.
Beaudoin, J. (1990) Handbook of Fiber Reinforced Concrete: Principles, Properties, Developments and Applications,
Noyes Publications.
Bennett, R. Market Research on Plastic Recycling, Center for Plastic Recycling Research, Technical Report #31.
Bilodeau, A., Carette, G.G. (1989) Resistance of Condensed Silica Fume Concrete to the Combined Action of Freezing and
Thawing Cycling and the Deicing Salts, ACI SP 11447 Trondheim Conference, pp. 945970.
Birley, A.W., Heath, R.J., Soctt, M.J. (1988) Plastics Materials: Properties and Applications, Chapman and Hall, 2nd
Edition.
Bordeleau, D., Pigeon, M., Banthia, N. (1992) Comparative Study of Latex-Modified Concretes and Normal Concretes
Subjected to Freezing and Thawing in the Presence of a Deicer Salt Solution, ACI Materials Journal, Nov.Dec. 1992,
pp. 547553.
Chanda, M., Roy, S. (1987) Plastic Technology Handbook, Series: Plastics Engineering, Marcel Dekker, Inc.
Curlee, T.R., Das, S. (1991) Identifying and Assessing Targets of Opportunity for Plastics Recycling, Resources,
Conservation & Recycling, Vol. 5, pp. 343363.
Engelmann, P.V., Monfore, M.D., Dawkins, E.W., McInerney, J.A. (1992) Recycling: Extrusion-Compression of
Commingled Resins Blends, Plastic Engineering, February 1992, pp. 2731.
Foy, C., Pigeon, M., Banthia, N. (1988) Freeze-Thaw Durability and Deicer Salt Scaling Resistance of a 0.25 WaterCement Ratio Concrete, Cement and Concrete Research, Vol.18, pp. 604614.
Hasan, S., Fowler, D.W. (1992) Investigation of Scrap PVC and Nylon in Concrete, Interim Report to Vista Chemical Co.,
Construction Materials Research Group, University of Texas at Austin, September 1992.
Langlois, M., Beapre, D., Pigeon, M., Foy, C. (1989) The Influence of Curing on the Salt Resistance of Concrete with and
without Silica Fume, ACI SP 11448 Trondheim Conference, pp. 971990.
Lin, C.C. (1988) Recycled Plastics as Filler in Polymer Concrete Composites, Center for Plastics Recycling Research,
Technical Report No. 23.
Nasvik, J. Plastic Aggregate, Concrete Construction, May 1991, pp. 399401.
Rebeiz, K., Fowler, D.W., Paul, D. (1991) An Overview on the Properties of Polymer Concrete Using Recycled Plastics,
University of Texas at Austin.
Senbetta, E. (1992) New Test Procedure for Impact Resistance of Industrial Floor Products, ACI Materials Journal, Sept.
Oct. 1992, pp. 495498.
Smith, L.L., Ramer, R.M. (1992) Recycled Plastics for Highway Agencies, 71th Annual Meeting, TRB, Paper No. 920496.
Torricone, P. (1993) Recycled Roads, Civil Engineering, April 1993.
Ulrich, H. Introduction to Industrial Polymers, Hanser Publishers, 1982.
Yang, D., Fowler, D.W., Whitney, D. (1991) Results of Tests on Concrete with PVC and Nylon Scrap, Construction
Materials Research Group, University of Texas at Austin, Report Submitted to Vista Chemical Co., January 1991.

6
POLYMER GRANULATES FOR MASONRY MORTARS AND
OUTDOOR PLASTER
H.R.SASSE
IBAC, Rheinisch-Westflische Technische Hochsclule Aachen, Aachen, Germany
O.LEHMKMPER
PROKU, Kunststoffveredelung Gmbh, Essen, Germany
R.KWASNY-ECHTERHAGEN
SICOWA, Verfahrenstechnik fr Baustoffe Gmbh & Co. KG, Aachen, Germany

Abstract
In Germany, plastic waste materials from households are collected separate from other garbage. An industrial
process has been developed to recycle these plastics in order to get artificial light aggregatesfor mortars. Main
point of the process is a combined heat and pressure treatment of crushed and washed plastic particles (hard
plastics and foils) to form a compact plastic block which can be crushed and grinded to granulate gradings from
0,125 to 4 mm. Theoretical and experimental studies justify the use of these granulates as light aggregate and
additive for numerous mortar and concrete products. In some ranges of application, improvements of technical
properties of conventional mortars and plasters can be achieved.
Keywords: aggregates, filler, mortars, plaster, plastic waste materials, processing methods, recycling.
1
Introduction
Since several years plastic waste materials from households are collected separate from the other garbage in
Germany. This separate collection has become necessary because of the requirement of a new law to recycle a
certain part of plastic waste materials (rising recycling amounts year to year). Recycling could be done as
material recycling (plastic products made by recycled plastics),
raw material recycling, that means e. g. hydrogenation, pyrolysis or also gasification.
Material recycling of old plastic materials requires a great effort for separation the plastic waste in all the different
kinds of plastics (PE, PP, PS, PVC). Without this separation the traditional way of material recycling by melting
and extrusion or injection moulding will lead to very ,,low tech products. On the other hand, technical effort to
reach a high standard of purity involves high charges. Technical quality of plastic products made by recycled plastics
not coming from the same manufacturing process cannot reach the high standard of new plastic material. The big
amount of used plastic materials coming from households obviously is far away from the possibility to find new
markets for products made of recycled plastics.
Thinking of the enormous market of building materials, efforts have been made to find applications for old plastic
materials in the building market and to develop both production methods to use the mixed plastics without any
separation and new products containing plastic granulates as artificial aggregates.
A research and development project has been started in cooperation between SICOWA company and the
University of Technology, Aachen. After first fundamental studies (experimental and theoretical) further work has

Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

POLYMER GRANULATES FOR MORTAR AND PLASTER

55

been sponsored mainly by ,,Rhenisch-Westflische Elektrizittswerke Entsorgung AG (RWE Entsorgung AG) and ,,PR
OKU Kunststoffveredelung GmbH.
2
Main steps of the production process
2.1
Input material
Input material for the production process is dirty, mixed plastic material from private households. One of the main
tasks of the production process is to prepare hard plastics (e. g. HD-PE, PS, PVC-U and PET) as well as foils (LDPE) which are difficult to handle in milling processes. The composition of the waste material differs in a wide range,
main parts are the polyolefines.
2.2
Preparation of (dirty) input material
Simplified, the first main step of the production process consists of the following processing units:
crushing the waste products to small pieces (diameter 10 to 20 cm)
washing the plastic pieces in order to remove dirt and paper
removing impurities (e. g. iron andin casealuminium).
Crushed and purified plastics have a very low bulk density (depending on the foil portion below 100 kg/m3)
2.3
Mechanical compacting and heat/pressure treatment
In order to reduce the volume stream and to optimize the efficiency of the following processing units the plastics
have to be compacted mechanically before heat/pressure treatment starts. Bales of compacted wet plastics reach bulk
densities of about 450 kg/ m3, they are put into moulds which will be transported into autoclaves. During the
autoclave process plastics are heated and plastified under saturated steam conditions. Autoclave process is within a wide
range independent from changing compositions of the plastics. The process temperature (about 170C) does not
provoke decomposition of plastics, used for household packaging materials. Hydrothermal autoclave conditions lead
to further compaction and a stable product (bulk density up to 700 kg/m3) of the mixed different kinds of plastics.
Fig. 1 shows a compacted plastic block after finishing the autoclave process.
2.4
Crushing and grinding
The third main step of the production process leads to plastic granulates which can be used in the building industry
as light aggregate. After the autoclave process the compacted plastic blocks have to be cooled before the primary
crushing process. This crushing process is realised by cutting mills, the result are coarse grains which can be dried
much easier than the compact plastic block. Finally, the coarse grains are put into a special mill which produces fine
grains in the size of 0,125 to 4 mm. Fig. 2 shows fine grains with a diameter up to about 0,5 mm.
3
Properties of polymer granulate aggregates
3.1
Geometric properties
The particle shape should be as compact as possible, the edge angles as obtuse as possible. These parameters can be
influenced by selecting the type of the melting process, the crushing equipment, the temperature during the crushing
process, and by applying a second crushing procedure.

56

SASSE, LEHMKMPER AND KWASNY-ECHTERHAGEN

Fig. 1. Compacted block of mixed different kinds of plastic after finishing the autoclave process (laboratory autoclave in the
background)

The grain surface is very smooth when the material is melted to a high degree. This is advantageous for the
workability (consistency, compactability). The relatively low adhesion strength between the cement matrix and the
smooth aggregate surface is disadvantageous only if very high mortar strengths are aspired to.
The melting process must be controlled in a way that the structure of the crushed aggregate particles will not be
leafy, foamy, or irregular. This does not mean, however, that the raw product before the crushing process must be
homogenous and pore-free.
The rules for the grading curve are the same as for other lightweight aggregates. Figure 3 gives examples for 3
different crushing methods.
Since the densities of polymer granulate and traditional aggregates differ considerably, the grading curves for
mixed aggregates have to be transferred into vol.-%.
Table 1 gives information on general properties of lightweight aggregates in comparison to polymer granulate
aggregates.
3.2
Mechanical properties
The strength of polymer granulate aggregates (acc. to the SICOWA technique) is in the upper range, the E-modulus
in the lower range of the corresponding properties of traditional lightweight aggregates (see table 2). This implies that
all commonly used mortar strengths can be achieved, depending on the water-cement-ratio. High strength mortars
can certainly not be made using high volumes of polymer aggregate. Table 1 gives some reference data.

POLYMER GRANULATES FOR MORTAR AND PLASTER

57

Fig. 2. Microscope photograph of fine grains after grinding, diameter of grains up to about 0,5 mm

Fig. 3. Grading curves for 3 different crushing methods

3.3
Chemical composition
The chemical composition of recycled polymer waste material (in Germany mainly from packing materials) can be
described only for random samples, as the portions of different polymers, additives, and contaminations may differ
considerably, depending on
regional dissemination of polymer products
time dependance of origin (day of the week, season)
method and uniformity of a separation process.

58

SASSE, LEHMKMPER AND KWASNY-ECHTERHAGEN

Table 1. Properties of lightweight aggregates


Type

Grain shape

Grain surface

Grain density Density


kg/m3
kg/m3

natural
pumice

round,
spherical

400 to 1100

2200 to 2500

expanded
perlite

round to
irregular 6
mm
spherical to
angular

dense to
microporous, rough
foamy, glassy
to microporous
dense glassy
to microporous
rough, dense
skin
smooth to
rough,
sometimes
cracked
dense,
smooth

100 to 400

expanded clay

poly-styrene
foam
sintered fly
ash pellets

granulate acc.
to SICOWA
technique

spherical
round,
spherical

angular to
splintered

Water
absorption
M-%

low

10

2100 to 2500 -

very low

500 to 2100

2500 to 2800 5 to 20

low to high

55

10 to 40

about 1000

very low

1300 to 2100 2600 to 3000 10 to 20

medium to
high

25

900 to 1100

medium

35

900 to 1100

50

Grain strength Maximum


mortar
strength
N/mm2

<1

Nevertheless, it can be stated that at the time in Germany the chief constituents are polyethylene (PE), polypropylene
(PP), and polystyrene (PS). Many other polymers may be found in small, mostly neglectible amounts, see table 3.
The relative amount of PE/PP is rising, that of PVC is decreasing. Most of the ,,other material will be potyethyleneterephthalat (PET).
Table 2. Properties of mortar components

quarzite
limestone
natural pumice
expanded clay
hardened cement matrix
granulate acc. to SICOWA technique

Compressive strength
N/mm2

E-modulus
kN/mm2

Thermal expansion coefficient


106/K

125430
80240
25
510
1045
3080

6085
2383
825
1, 53

11, 012, 5
3, 57, 5
911
100150

Table 3. Polymeric packing materials in household waste [3].


Amounts in 1 000t
foils
bags
bottles
cups
stopper, breeches
other packages
sum
rel. amount in %

PE/PP

PVC

PS

other

sum

216
120
104
38
28
10
516
69

10
23
24
1
19
77
10

88
5
13
106
14

48
2
1
51
7

274
120
129
150
35
42
750
100

The selection of polymers from other household waste materials (paper, metals etc.) is made visually and by hand
[1]. Automatically controlled plant units which allow the separation of different polymer types are still in their test
phase [2].

POLYMER GRANULATES FOR MORTAR AND PLASTER

59

With the exception of plasticizers for plasticized (soft) PVC all additives to the polymers concerned are present
only in extreme small amounts. Plasticized PVC appears in German pre-selected household wastes only due to
singular mistakes, as it is not used for packages. Building materials (like floorings, sealing materials, profiles, tubes)
are not allowed to give into the household wastes in Germany.
The amounts of polymers, which may saponify (e.g. esters) are so small in the household waste materials that no
technical relevant influences are possible on the alcalic mortars.
Most contaminants like rests of food, salts, paper, and tensides will be separated by the washing procedures.
An automatic quality control system (spectroscopy) has to be installed in the production line to make sure that the
separation and cleaning processes are working regularly.
3.4
Chemical properties
All relevant polymers (PE, PP, PS, PET, PVC, POM) are not reactive with components of the mortars (cements,
lime, gypsum). Pigments (unsoluble metal salts or oxides) will also not react chemically in the mortar.
Aluminium, mainly in fine grained form, may react after mixing of the mortar with the cement and develop
hydrogen gas [4]. This may result in an increase of air bubbles, with negative effects on the strength. Particles of iron
may corrode on the surface of carbonated mortars and then affect the esthetics. The production line has to consider
these effects by separating the metals or by homogenization in case of the aluminium.
4
Mortars with polymer granulate aggregates
4.1
Mix proportions
The workability of the fresh mortars and the properties of the hardened mortars are determined by the mix
composition (binder, water, traditional aggregate, polymeric granulate, additives) in the same way like in normal
lightweight mortar technology. The dense structure of the grains is advantageous as it makes it unnecessary to take
into account the water suction of the aggregate.
There is no indication that additives from the plastic processing or residual contaminants affect the workability or
the stiffening in an undesired way.
4.2
Properties of hardened mortars
Compared to quartz aggregate the strength of mortars with 50 vol-% of polymer granulate is lower, as to to be
expected, see fig. 4. The differences will be lower if the polymer portion is reduced. As the modulus of elasticity of
the unsorted waste materials varies due to changing composition (see chapter 3), there will be a certain scatter in
mortar strength, resulting from this.
Table 4. Example for an outdoor plaster with polymer granulate aggregate
Property

Dimension

Lightweight plaster

bulk density strength, 28 days


- bending
- compression
water suction thermal expansion
20C+40C

kg/dm3

0,851,05
1,02,0
2,55,0
<0,5
1834

N/mm2
N/mm2
kg/m2h0,5
106/k

Requirement acc. to DIN 188 550 T4


0,6; 1,3
2,5; 5,0
0,5 (water repellent)
-

Table 4 gives an example for a lightweight outdoor plaster mortar made with polymer waste granulate.
Up till now there is only few safe knowledge about stress-induced deformations (E-modulus, creep) and stressindependent deformations (shrinkage, swelling, temperature-expansion). But due to long experiences with other
mortars containing polymers (e.g. lightweight mortars with expanded polystyrene as aggregate, polymer modified
cement mortars PCC), basic problems are not to be expected.

60

SASSE, LEHMKMPER AND KWASNY-ECHTERHAGEN

Fig. 4. Influence of scatter in granulate properties on compressive strength of cement mortars

4.3
Behaviour under fire
In case of fire there are to be expected the typical phenomena for thermoplastic materials:
softening of the polymers (above about 100C)
melting of the polymers (above about 150C)
decomposition and burning of the polymers (above about 350C).
Possible resulting consequences are:
deterioration of technical properties
formation of environmentally dangerous gases
formation of corrosive gases.
The classification of the behaviour of the mortars under fire acc. to DIN 4102 [5] depends mainly on the content and
the grain sizes of the granulates. The following classes are attainable with polymer containing mortars:
normal inflammable (class B2)
self extinguishing (class B1)
not inflammable (class A2).
The formation of detrimental gases depends to a high extent on parameters like oxygen supply, grain size, type of
ignition, and temperature gradient. If enough oxygen is present and only pure hydrocarbons are in the granulate, only
carbon dioxide and water will be generated. Disadvantages concerning detrimental emissions are
low oxygen supply
certain elements in the polymer material, e.g. chlorine
additives, e.g. aluminium.
These emissions from polymer containing mortars are quantitatively insignificant compared with these from
coatings, floor coverings, electrical equipment, household furniture and utensils etc., which in case of fire will
produce detrimental gases in large quantities.

POLYMER GRANULATES FOR MORTAR AND PLASTER

61

4.4
Release of components aggressive to groundwater
In Germany there is a catalogue of materials (chemicals) concerning their detrimental action when contacting
groundwater [6], which defines four hazard classes. Basis is the toxicity against mammals, bacteria, and fishes as
well as the wash-out behaviour. Considering this catalogue and the composition of the mortars and plasters
containing polymer granulates from household wastes and industrial sources, a risk for the groundwater can be
excluded.
5
Conclusive remarks
The use of granulated polymeric waste materials as aggregate and filler in masonry mortar and outdoor plasters is
not only a method of recycling but also a technique which results in advantages for the mortars and plasters, e.g.:
lower thermal conductivity
low bulk density (improvement of efficiency of the workmen)
lower wear of mixing and pumping equipment (due to very low hardness of the polymers compared to mineral
aggregates)
low E-modulus in relation to strength (which reduces the danger of cracking of plasters).
Conditions for a successful application of polymer granulates from household wastes in mortars are:
adequate collecting organization for the waste material
adequate preparation process
adequate economical, legal, and ecological milieu in the respective state.
6
References
1.
2.
3.
4.
5.
6.

Information from the German Waste Material Collecting Organization: Duales System Deutschland, Rochusstr. 26, D
53123 Bonn
Kffner, G.: Schallschutz durch alte Joghurtbecher. Spektralbilder helfen, Kunststoffe zu trennen. FAZ 31.08.1993
Eder, G.: Wohin mit dem Kunststoffmll? Werner Verlag, Dsseldorf; 1992
Reul, H.: Handbuch der Bauchemie, Einfhrung in die Grundlagen, Rohstoffe, Rezepturen. Augsburg: Verlag f. chem.
Industrie, H.Ziolowsky KG, 1991
DIN 4102 Teil 1 05.81. Brandverhalten von Baustoffen und Bauteilen. Baustoffe; Begriffe, Anforderungen und Prfungen
Roth: Loseblattsammlung Wassergefhrdende Stoffe 16. Erg. Lieferg. 9/91

7
POLYMER MODIFIED LIGHTWEIGHT CEMENT MORTAR
USING PLASTICS WASTE
Y.HAYASHI
Technical Research Institute, Obayashi Corporation, Tokyo, Japan
R.NANIWA, H.IIBACHI and K.HADA
Department of Architecture, Faculty of Engineering, Kogakuin University, Tokyo, Japan
T.YAMAZAKI
Acolite Construction (S) Pte Ltd, Singapore

Abstract
Polymer modified lightweight cement mortar (PLCM) using plastics waste as lightweight aggregate, is at
present widely used for plastering work instead of ordinary cement mortar with normal aggregate. The suitable
application of this mortar to substrate layer for exterior ceramic tile installation, has been discussed and studied
in Japan. Then it is important to clarify the advantage of PLCM through certain studies and analyses.
Keywords: Differential movement, fall-off of exterior ceramic tile, honeycomb shaped woven net, polymer
modified lightweight cement mortar, two dimensional finite element method, woven fiber net
1
Introduction
Polymer modified lightweight cement mortar (PLCM) with plastics waste as aggregate is widely used as substrate
mortar layer on the reinforced concrete wall for external ceramic wall tile installations instead of the ordinary cement
mortar conventionally used. The reasons that this mortar is popularly used in building site are its own lightweight
and excellent workability for troweling.
Plastics wastes used as aggregate of the mortar are expanded polystyrene (PS) and expanded ethylene-vinyl
acetate copolymer (EVAc). Expanded PS is obtained from packing materials wastes and so on. Now, about 9,200
tons of expanded PS have been crushed and used as lightweight aggregate for plastering materials every year in
Japan. EVAc chip is found in cushion materials for shoes etc. These plastics wastes are crushed and graded for the
purpose. This mortar was developed about 20 years ago. The suitable application of this mortar has been discussed
and studied until now.
Its quality, code of practice and specification of PLCM are not yet authorized in Japan.
2
Analysis of differential movement in exterior wall tile installation using PLCM as substrate layer
Differential movement caused by cold to heat and wet to dry repetition, is the major cause of delamination, disbond
or fall-off failures of tile covering finish which occur at the interface between layers made up of different materials in
the exterior concrete walls. The authors executed numerical analysis with simplified models for simulation to
examine what internal stress distribution is generated in the exterior wall and how the characteristics of each material
affect the stress generated in the wall under cold to heat and wet to dry cycles.

Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

LIGHTWEIGHT CEMENT MORTAR USING PLASTIC WASTE

63

Fig. 1. Analysis model

2.1
Numerical analysis models
For this study, the authors employed a 1/2 scale analysis model for 60108mm sized ceramic tile piece which is two
dimensionally analogical to the actual size for numerical analysis by means of two-dimensional finite element
method. Fig. 1. shows analysis models, that is, Type A, B and C. Type A is criterion model having a 19mm thick
mortar layer consisting of a 15mm thick mortar substrate and a 4mm thick bonding mortar which are both ordinary
plain cement mortar. Type B and C are for comparison of the effect of mortar characteristics with Type A. Type B
has a 15mm thick substrate made of PLCM only with bonding ordinary plain cement mortar.
Type C has a substrate with a 5mm thick PLCM and a 10mm thick ordinary plain cement mortar. Ordinary plain
cement mortar is hereinafter referred to as mortar(P) and PLCM as mortar(L) in Fig. 1, Fig. 2 and Table 1.
2.2
Numerical analysis conditions
Physical property values of each material employed are given in Table 1. In case of PLCM whose physical property
varies with the sort of lightweight aggregate used in it, mean values were adopted.
As shown in Fig. 1, the model was restrained at the whole bottom in the Y-axis direction, and at the lower point of
the left end in the X-axis direction. Cold to heat condition is changing all the time. As every moment of cold to heat
cycle cannot be followed and simulated, the authors assumed that it was in steady state.
Table 1. Fundamental properties of materials employed
Item

Unit

Tile

Concrete

Mortar(P)

Mortar(L)

Elastic modulus

GPa

59.0

24.5

20.0

5.0

64

HAYASHI, NANIWA, IIBACHI, HADA AND YAMAZAKI

Fig. 2. Cross section of external wall

Fig. 3. Result of stress by cold to heat condition


Item

Unit

Tile

Concrete

Mortar(P)

Mortar(L)

Poissons ratio
Unitary mass density
Thermal expansivity
Thermal conductivity
Specific heat

t/m2
106/C
kj/mhC
kj/kgC

0.19
2.30
7.5
5.40
0.75

0.17
2.30
12.0
4.60
0.90

0.17
2.0
11.0
4.60
1.10

0.17
1.30
11.0
1.70
1.20

The authors gave temperatures of 50C on the tile surface and 20C on the concrete back face, an initial
temperature was 15C to the whole model. Regarding wet to dry condition, it was assumed that concrete was fixed
and free from its own movement, because of the rigidity with its thick cross section and that only mortar received the
restraint and shrank; namely at the drying shrinkage of 100106 for ordinary plain cement mortar and 300106 for
PLCM.
2.3
Results and considerations
The authors examined the effect of PLCM by paying attention to the maximum values of xy , x and y generated
around the boundary surface between different materials. The effect by cold to heat condition is given in Fig. 3, and
the effect by wet to dry condition in Fig. 4.
In case of applying the PLCM, the stress occurred both in a-a and d-d cross sections shown in Fig. 2, was
reduced. Stress reduction is more effective in d-d cross section than in a-a cross section. Furthermore, the thicker
the PLCM layer is, the more effective the stress reduction. This is because PLCM has a smaller elastic modulus than
ordinary plain cement mortar, and also has a smaller thermal conductivity, which is effective to insulate the external
heat through the tile at the d-d cross section. The authors gave PLCM a drying shrinkage 3 times as large as the
ordinary plain cement mortar. The reduction of stress was not observed with PLCM as under cold to heat condition.
On the contrary, there was a stress increase in d-d cross section. The thicker the PLCM layer is, the larger the stress
increase.

LIGHTWEIGHT CEMENT MORTAR USING PLASTIC WASTE

65

Fig. 4. Result of stress by wet to dry condition

Thicker PLCM produces high shear stress xy in b-b and c-c cross sections as shown in Fig. 4.
Physical characteristics of the ideal mortar are as follows: smaller elastic modulus, smaller unitary mass density,
smaller thermal expansivity, smaller thermal conductivity and larger specific heat.
3
Fundamental tests for existing PLCM
3.1
Materials employed and their mix proportions
Three types of lightweight aggregate chosen in the market were used. Type A is composed of crushed expanded
polystyrene (3mm under in size), EVAc chips (3mm under in size) and methyl cellulose etc. This mortar using type
A is mainly applied to internal wall with the thickness of 2~10mm.
Type B is composed of EVAc chips (3mm under in size), Calcium carbonate chips (3mm under in size) and methyl
cellulose etc. This mortar using type B is mainly applied to external wall with the thickness of 2~10mm.
Type C is composed of EVAc chips (1.5mm under in size), EVAc powder (0.5mm under in size) and methyl
cellulose. This mortar using type C is applied to external wall with the thickness of 2~5mm. Mix proportions of these
mortars are shown in Table 2.
Table 2. Mix proportions
Polymer modified lightweight cement mortar

Cement
(kg)

Water
(kg)

PLCM A
PLCM B
PLCM C
Polymer dispersion: EVAc, solid content: 55 %

5.000
5.000
5.000

1.903
1.842
1.689

Polymer dispersion
(kg)

Lightweight aggregate (kg)

0.100
0.100
0.100

0.373

0.395

0.497

3.2
Workability for troweling of fresh mortar
Workability of ordinary plain cement mortar is generally estimated with the flow test by Japanese Industrial Standard
(JIS). But this test method can not estimate sufficiently the workability of PLCM with visco-elastic behavior.
The mechanical troweling apparatus for preparation of adhesion test specimen of plastering materials to concrete
substrate has been proposed by R.NANIWA. Workability of PLCM were estimated by 3 rheological characteristics;
horizontal reaction force Ftr, vertical reaction force Fnr and reaction moment Mr at the grip (supporting point) were
measured. Fig. 5 shows the schema of troweling motion and the apparatus in this test method. Some examples of test

66

HAYASHI, NANIWA, IIBACHI, HADA AND YAMAZAKI

Fig. 5. Oblique load to the plastering materials

results of workability of PLCM are shown in Table 3. As shown in Table 3, these polymer modified lightweight
cement mortars showed high reaction force and reaction moment compared with ordinary plain cement mortar (OCM)
and ordinary polymer cement mortar (PCM).
Fig. 6 and 7 show relationships between Fr,
and Flow value by JIS standard test of various mortars tested. Here,
if there are some correlations between those characteristics, we could draw a correlation curve on each figure. We
can not find, however, a correlation between Fnr and Flow value, but find certain correlation between Ftr and Flow
value. It means that Ftr is almost proportional to Flow value and Flow test is suitable only for estimation of the
viscosity.
Table 3. Results of 3 rheological characteristics
Mortar

Polymer

Unit weight
(kg/1)

Ftr
(N)

Fnr
(N)

Mr
(Ncm)

PLCM A
PLCM B
PLCM C
PLCM A
PCM
OCM

EVA
EVA
EVA
non
SBR
non

1.29
1.68
1.61
1.26
1.71
1.88

24.3
18.7
18.0
10.9
11.4
7.7

50.2
41.5
44.4
29.2
38.8
31.8

551
564
479
388
366
348

3.3
Shrinkage
Free shrinkage of test specimen of 4040160mm prism was measured by contact gauge method. On the other hand,
restrained shrinkage of test specimens of PLCM troweled with 6mm in thickness to concrete substrate was measured
by the same method. Test results of shrinkage are shown in Table 4 and Table 5. PLCM shows high free shrinkage
and restrained shrinkage compared with ordinary plain cement mortar.

LIGHTWEIGHT CEMENT MORTAR USING PLASTIC WASTE

67

Fig. 6. Relationships between FlowFr and Flow


Table 4. Shrinkage (free) ( %)
Polymer modified lightweight cement mortar

Age (days)

14

21

28

PLCM A
PLCM B
PLCM C

0.122
0.095
0.085

0.225
0.199
0.178

0.244
0.230
0.206

0.253
0.258
0.250

Table 5. Shrinkage (restrained) ( %)


Polymer modified lightweight cement mortar

Age (days)

14

21

28

PLCM A
PLCM B
PLCM C

0.007
0.013
0.004

0.015
0.025
0.011

0.018
0.036
0.017

0.024
0.040
0.018

3.4
Water absorption and swelling
Water absorption and swelling of PLCM were measured at the age of 28 days of test specimens (4040160mm
prism) which were cured in a controlled atmosphere (203C, 655 % R.H.). Test results are shown in Table 6 and

68

HAYASHI, NANIWA, IIBACHI, HADA AND YAMAZAKI

Fig. 7. An example of explanatory notes of data

Table 7. Water absorption and swelling of PLCM depended upon specific gravity of PLCM. Swelling of PLCM is
very remarkable. It is very important to reduce swelling as well as shrinkage.
Table 6. Water absorption ( %)
Polymer modified lightweight cement mortar

Age (hours)

24

PLCM A
PLCM B
PLCM C

4.22
3.14
2.14

5.24
3.93
2.61

5.95
4.50
2.98

7.29
5.52
3.68

Table 7. Swelling ( %)
Polymer modified lightweight cement mortar

Age (hours)

24

PLCM A
PLCM B
PLCM C

0.026
0.018
0.016

0.058
0.035
0.029

0.089
0.054
0.044

0.136
0.115
0.091

3.5
Mechanical properties
Bending strength ( b), elastic modulus (E) and compressive strength ( c) were obtained. Table 8 shows test results.
Mechanical properties of PLCM depend upon specific gravity. PLCM is characterized by high bending strength, high
compressive strength and low elastic modulus. Deformability of mortar can be expressed as b/E. An important
advantage of PLCM is its high deformability.
Table 8. Mechanical Properties (Age of 28 days)
Polymer modified lightweight cement mortar

Specific gravity Bending strength


(MPa)

Compressive strength
(MPa)

modulus Elastic
(GPa)

PLCM A
PLCM B
PLCM C

1.21
1.52
1.48

15.1
20.9
21.8

7.55
9.24
10.3

4.83
5.22
6.45

LIGHTWEIGHT CEMENT MORTAR USING PLASTIC WASTE

69

Fig. 8 Ceramic tile installation by PLCM connecting with honeycomb shaped woven net.

3.6
Pull-out adhesion strength
Pull-out adhesion strength between PLCM and concrete substrate was measured using the pull-out adhesion test
method. Test result is shown in Table 9. Pull-out adhesion strength depended upon the different types of PLCM.
PLCM which is troweled with 6mm thickness on the surface of concrete of 300mm square with 60mm in thickness,
is generally cement rich, strongly bonded to concrete substrate. As cohesion of PLCM A and PLCM B was not so
strong, and there were internal failures of PLCM layer. Pull-out adhesion strength of PLCM is big enough.
Table 9. Result of pull-out adhesion test
Polymer modified lightweight cement mortar

Pull-out adhesion strength


(MPa)

Failure type

PLCM A
PLCM B
PLCM C

1.21
1.51
1.68

Cohesion failure in PLCM


Cohesion failure in PLCM
Cohesion failure in concrete substrate

4
Preventive measures for fall-off exterior ceramic tile installation using PLCM as the substrate and
adopting fibrous net
4.1
Improvement of PLCM
Serious problems in case of applying PLCM to exterior ceramic tile installation, are caused by its poor dimensional
stability. Moisture movement such as shrinkage and swelling by changing atmosphere of existing PLCM is too big
compared with ordinary polymer cement mortar conventionally used. It is necessary to improve dimensional stability
of PLCM by applying adequate polymer and blending inorganic aggregate with high dimensional stability.
4.2
Utilization of fiber materials
Sandwiching of fiber material at the interface between concrete substrate and mortar layer as a substrate of ceramic
tile finish was developed and its use is recommended in order to prevent the fall-off of exterior ceramic tile
installation from reinforced concrete wall. This material is honeycomb shaped woven net as shown in Fig. 9.
Continuous woven fiber net of x and y axes as shown in Fig. 9, is effective to reinforce and link the exterior ceramic
tile installation in the in-plane direction. Chopped fiber fixed to continuous woven fiber net at the Z direction,
anchors the exterior ceramic wall tile installation. If delamination and cracking are partially generated in the exterior
ceramic tile installation, honeycomb shaped woven net can prevent the fall-off of tile. Fig. 8 shows typical example
of section of exterior ceramic tile installation using PLCM and honeycomb shaped woven net.

70

HAYASHI, NANIWA, IIBACHI, HADA AND YAMAZAKI

Fig. 9 Honeycomb shaped woven net made of polypropylene

5
Conclusion
Through the foregoing studies and analysis, the authors have come to the following conclusion. Here is summary of
the result of deliberations.
(1) Under cold to heat condition, the stress is reduced when the elastic modulus of mortar is lowered, especially at
the interface between substrate mortar and concrete.
Under wet to dry condition, when the drying shrinkage of both bonding and substrate mortar become (e.g. PLCM)
higher, the stress increased at the interface between substrate mortar and concrete. This tendency means that the
repetition of drying after the infiltration of water (e.g. rain water) into the PLCM layer through cracks etc., will
create extreme stress and can result in the fall-off of tile installation from the concrete wall.
(2) The mechanical troweling apparatus for preparation of adhesion specimen of plastering material to concrete
substrate can estimate satisfactorily the workability of PLCM with visco-elastic behavior. Workability of existing
PLCM is not always sufficient to apply. It is necessary to improve the characteristics of PLCM.
(3) Dimensional stability of existing PLCM is poorer, compared with that of ordinary polymer cement mortar. It is
necessary to improve the dimensional stability by using inorganic aggregates and suitable polymer dispersion.
(4) Existing PLCM is characterized by high flexural, high compressive strength and low elastic modulus. An
important advantage of existing PLCM is its high deformability.
(5) Pull-out adhesion strength depended upon the various types of existing PLCM.
(6) Adoption of honeycomb shaped woven net is surely effective to prevent the fall-off of the exterior ceramic tile
installation using PLCM as a substrate mortar.
Acknowledgments
On this study, the authors are very grateful for the close collaboration of Professor M. KAWAKAMI, Department of
Civil Engineering, Akita University and Mr. E. TAKADA, TOTO Corporation, Japan.
References
1.
2.
3.
4.
5.

Naniwa, R.(1991) Fall-off of Exterior Wall tile and Mortar Covering in Concrete Building and its Preventive measures,
Concrete Journal, Japan, Vol. 29, No. 9, Japan Concrete Institute, PP. 518.
Takada, E et al. (1992) Study on the Differential Movement of External Wall in Building, Part 2: Stress Distribution at the
Interface between Ceramic Tile Joint Mortar. Annual Convention of Architectural Institute of Japan, PP. 211212.
Hayashi, Y et al. (1992) Technique of Application of Fiber Materials to Exterior Ceramic Wall Tiling. Durability of
building Materials and Components 6, Japan, Vol. 1, E&FN Spon, PP1322.
Naniwa, R et al. (1992) Analysis of Differential Movement for Durability Design of Exterior Building Wall. Durability of
Building Materials and Components 6, Japan, Vol. 2, E&FN Spon, PP. 13631371.
Hada, K et al. (1994) Mechanical Troweling Apparatus for preparation of Adhesion Test Specimen of Plastering Materials.
Annual Convention of Architectural Institute of Japan, PP. 13451346.

PART TWO
FRP-BASED MATERIALS

8
UPDATING RECYCLING TECHNOLOGIES FOR THERMOSET
COMPOSITES IN JAPAN
T.KITAMURA
The RTC (RP Waste Recycling and Treatment Council) of The Japan Reinforced Plastics Society
(JRPS), Tokyo, Japan

Abstract
Recycling technologies of scrap RP are classified into major three methods, a material recycle centering around
pulverization and a chemical recycling by pyrolysis or chemical decomposition and an energy recovery to be
represented by incineration. Pulverization is viable for cured scraps from factories, material composition and
molding technic of which are known, but not effective for all the waste from RP products after consumption.
The remaining two methods both yield solid residue. In order to reduce the cost of treatment, the development
of a value-added reutilization method for the solid residue is indispensable.
In 1990, the JRPS organized a tertiary committee, the RP Recycling and Treatment Council (RTC) for the
solution of technological and social problems regarding recycling thermoset composites waste.
Keywords; mixing into asphalt, mixing into cement mortar, solid residue, synthesis of zeolite, pulverization,
pyrolysis, thermoset composites.
1
Introduction
The result of discussion for development of Recycling Thermoset Composites by the RTC and the other
organizations concerned in Japan from 1991 to 1993 lead us to the following policies to take.
No matter what treatment method (pulverization or pyrolysis, etc) is adopted, we should promote market
development action to the other than RP industry (open market policy). This open market development means to
discover users who belong to the outside of RP industry and to promote their assessment for reuse potentiality of the
reclaimed product through joint research with them.
At the same time, we should promote technical development which enables us to get such reclaimed product as
could meet requirement of the outside industries.
We should establish quality, specification or technical standards for the reclaimed product.
We should examine logistics for RP waste which leads us to establish system of recovery, assessment and
warranty of stable supply.

Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

RECYCLING TECHNOLOGIES FOR THERMOSET COMPOSITES

73

Fig. 2.1 From RP Waste to Reuse

2
Subject
2.1
Updated results of development for Recycling Technologies
According to the above mentioned policy, we reassessed all the results of technical progresses which had been
accomplished until now. The largest viewpoint of reassessment was whether or not the developed technic could add
value to the reclaimed product.
First, from the viewpoint of treatment process of RP waste, process itself is distinguished either by handling scrap
from factories or by processing RP product-waste after consumption.
This means that the career of waste restricts both process and its end-use. Relations between processing and enduse are shown in Fig. 2.1 arranging inclusive of technical developments on the way.
2.2
Review of Recycling Technologies
2.2.1
Material RecyclingParticle/Powder Reclamation
The results of reassessments are summarized in the following regarding technical requisites which have been
developed until now. The order of description is a) for result of reassessment; b) for problems remaining; c) for
example of actual reuse, if available.
(1) Mixing into SMC as filler material replacing CaCO3
a) If a quantity is filled less than a certain mixing ratio (for general type SMC, by 20wt%, for class A SMC, by
10wt%) of a certain grain of milled powder (200 mesh pass) pulverized from cured scrap from factory, there is seen
no bad influences on material properties to the molded SMC product. The physical, mechanical properties of
recycled SMCStandard type is shown in Tab. 2.1.1 andClass A in Tab. 2.1.2.[1]
b) It is difficult to recover cost for treatment.
c) An example of molding SMC automotive part for Roof-inner Rack for Toyota Sprinter Carib model by
replacing CaCO3 by 20wt% of milled SMC scrap. Also, in the first half of 1993, Toyota adopted two types of SMC

74

KITAMURA

automotive parts reusing recycled SMC fine powder, one of which is a spoiler for Toyota Supra which used light
weight class A recycled SMC and the other is a cylinder head cover for diesel engine of Toyota Land Cruiser, Prado,
which used light weight heat-resistant recycled SMC.
Tab 2.1.1 Physical/Mechanical Properties of Recycled SMCStandard Type
Formulation
Materials

Resin
CaCO3
Reground
GF (1 inch) (%)
Properties
Shrinkage (%)
Specific Gravity
Flexural Strength (MPa)
Flexural Modulus (GPa)
Tensile Strength (MPa)
Tensile Modulus (GPa)
Izod Impact (MJ/mm2)*
* edgewise, un-notched

100
125
0
30

100
78
32
30

100
36
60
30

0.06
1.73
200
12.3
78
11.8
82.0

0.07
1.64
213
11.8
77
11.5
87.4

0.08
1.59
186
11.4
79
11.2
86.0

Tab 2.1.2 Physical/Mechanical Properties of Recycled SMCClass A SMC


Formulation
Materials

Resin
CaCO3
Reground
GF (1 inch) (%)
Properties
Shrinkage (%)
Specific Gravity
Water Absorption (%)
Surface Quality
Average Deviation ()a)
Flexural Strength (MPa)
Flexural Modulus (GPa)
Tensile Strength (MPa)
Tensile Modulus (GPa)
Izod Impact (MJ/mm2)*
* edgewise, un-notched a) Takeda Method.

100
180
0
28

100
123
38
28

100
76
70
28

0.00
1.85
0.44

0.00
1.77
0.39

0.00
1.68
0.30

3.6
190
11.5
85
11.0
76.5

3.8
178
10.9
82
10.7
81.5

5.6
167
10.4
68
10.2
78.5

(2) Mixing into BMC as filler material replacing CaCO3


a) If a quantity is filled less than a certain mixing ratio (for general type of BMC, by 30wt%) of a certain grain size
of milled powder (25 m) pulverized from cured SMC scrap from factory, there is seen no bad influences on
material properties to the molded product.
b) The same as SMC (1).
c) An example of recycling used head lump housings into BMC and molding the same head lump housing.[2]
(3) Mixing into pulp cement plate as filler material replacing asbestos
a) If a quantity is filled by 5% of a certain grain size of milled powder (16 to 30 mesh pass), there is seen no bad
influences on material properties to the molded product (plate).[3]
b) Test was conducted by milling a virgin RP laminate.

RECYCLING TECHNOLOGIES FOR THERMOSET COMPOSITES

75

Fig 2.1.1 (1) Specific Gravity by RP Addition


Fig 2.1.1 (2) Strength by RP Addition

c) No trial for actual scrap.


(4) Mixing into (polyester resin concrete) as filler material replacing CaCO3 or aggregates (sands, small or
middle)
a) Flexural strength increases as filling ratio increases. The 60 to 80% of mixing ratio makes flexural strength the
largest, which is also enhanced much more by adding silane finishing. This test suggests the possibility of reuse for a
resin concrete manway.[4]
b) No information whether or not to recover cost for treatment.
(5) Mixing into gypsum as reinforcement
a) If a quantity is filled less than a certain mixing ratio by 10% of a certain grain size of milled powder (the
smallest size; 0.125 mm), there is seen no bad influences on material properties to the molded product. If a binder
like PVA is added, strength is increased. When added 10% of powdered RP with minimum diameter of 0.125 mm (=
120 mesh pass), added 0.2% of PVA, maximum 35 MPa of flexural strength was obtained.[5]
b) The effect of a positive improvement like flexural strength will be unexpected.
(6) Mixing into cement mortar as filler material replacing aggregate (silica)
a) If a quantity is filled less than a certain mixing ratio (by 5 to 10 vol%) of a certain grain size of milled powder
(25 m), there is seen no bad influences on material properties to the molded concrete. Also, it became clear that
alkaline fusion to RP powder was hard to be caused.[6][7]
The data from a cement supplier of Japan which cooperated with the RTC are shown from Fig. 2.1.1 to Fig. 2.1.3.
[8] The Fig. 2.1.1; (1) shows specific gravity by RP powder addition under the condition of constant flow value, (2)
shows strength by RP addition under the same condition. The Fig. 2.1.2; (1) shows specific gravity by RP powder
addition under the condition of constant w/c ratio, (2) shows strength by RP addition under the same condition. The
Fig. 2.1.3 shows the same result but the difference from the Fig. 2.1.2 is grain size; the former is for 15 to 20 m, the
latter is under 10 m, respectively.
b) A remarkable effect such as improvement of flexural strength will be unexpected. In the meantime, according to
the recent result of research by Dr. Akira Kojima, Gunma College of Technology, when making a compounded
mortar from portland cement+ fine powder of waste from glass-reinforced thermoset+silas balloon and curing
by steam under the condition of 200C, 3 hours, the cured concrete showed properties of a super light weight
concrete, a specific gravity of 1.0, a flexural strength of 8.0 MPa, a compressive strength of 35.0 MPa. A further
development is very concerned.[9]
c) Until now, representative three test examples as aggregate for mortar cement were conducted.
(7) Mixing into straight asphalt and starting to evaluate its performances
a) The Public Works Research Institute of the Ministry of Construction started a series of studies in 1993 for
designing recyclable materials intending material recycling and the RTC cooperated with this project. To begin with,
a ceratin grain of milled powder (5 to 20 m) was mixed into straight asphalt and evaluated good or bad of
dispersion, improvement effect of toughness and tenacity. Dispersion in mixing was good. Toughness in mixing
increased but tenacity has not almost changed or in some cases a deterioration was observed. The results of analysis
regarding the influence of mixing RP recycled product are shown from Fig. 2.1.4 to 2.1.9. Fig. 2.1.8 and 2.1.9 show
result of measurement of toughness and tenacity when fine powder of RP waste were mixed into straight asphalt.[10]

76

KITAMURA

Fig 2.1.2 (1) Specific Gravity by RP Addition


Fig 2.1.2 (2) Strength by RP Addition

Fig 2.1.3 (1) Specific Gravity by RP Addition


Fig 2.1.3 (2) Strength by RP Addition

Fig 2.1.4 Penetration of RP Powder

b) This assessment has only just begun in 1993 and we will have to wait a little while to accumulate much data
evaluating its real performance by asphalt pavement.
2.2.2
Reuse of Solid Residue after Pyrolysis
It is hard for recovered RP waste after consumption to reuse by pulverizing method, as shapes are not uniform,
material constituents are diverse, other materials than RP (metal, wood, other plastics, paint pigment) and

RECYCLING TECHNOLOGIES FOR THERMOSET COMPOSITES

77

Fig 2.1.5 Ductility of Asphalt

Fig 2.1.6 Viscosity of Asphalt

Fig 2.1.7 Softening Point of Asphalt

contamination during use are seen. Accordingly first after separating RP waste into organic and inorganic
constituent, then method of reuse of each constituent should be considered. At present, the most effective method of
this separation is thought to be pyrolysis method which convert organic constituent into gas and oil and inorganic
into solid residue. This means that after pyrolysis, method of adding value to solid residue should be developed.
According to this process, fundamentally the following two constituents of solid residue are obtained;
a. fiberglass+residue of catalyst, paint pigment
b. fiberglass+filler (CaCO3)+residue of catalyst, paint pigment
As one of the reuse markets which might add value to the recovered solid residue, the RTC tried to synthesize
zeolite.
(1) Zeolite
Zeolite is a crystalline aluminosilicate, chemical constituent of which is SiO2, Al2O3, H2O, Na2O, K2O, CaO.
Micro pore which involves crystal water adsorbs various gaseous molecule. Zeolite well adsorbs polar molecule but
even if non-polar molecule, it can also performs adsorption capability depending upon shape or size of molecule.

78

KITAMURA

Fig 2.1.8 Toughness of Asphalt

Fig 2.1.9 Tenacity of Asphalt

Many other zeolites than the natural are synthesized. Synthesis of zeolite is made under the hydro thermal
condition from room temperature to 673 K. The phase of crystallization varies with sort of raw material, ratio of
constituent of raw material, mold material and synthetic condition. These synthetic zeolites are used for ion
exchanger, molecular sieve, industrial catalyst and carrier of catalyst, builder for detergent and soil conditioner, etc.
(2) A preliminary investigation
Following up the experiment of synthesis of zeolite which was conducted by the Government Industrial Research
Inst., Shikoku (GIRIS)/MITI, the RTC synthesized type A, zeolite using solid residue after pyrolysis by wet reaction
process and investigated the performances of its adsorption.
Type A zeolite is synthesized for the first time by Union Carbide as an industrial material, the character of which
is a low ratio of SiO2/Al2O3 and the largest application is builder for surfactant.
This is due to the character of a high selective capability to Ca2+ ion and its high CEC (cation-exchange capacity)
enables for hard water to change into soft which promoted much to be used as builder of detergent.
(3) Result of the first trial of synthesis
Resulting from synthesizing two kinds of samples, the CEC of the synthesized product showed a 190 meq/100g
and a 300 m2/g of the specific surface area. If obtained 500 to 600 m2/g of the specific surface area, it will be
possible to use as a high grade of industrial adsorbent and the RTC is going to research the optimum condition for
removing CaCO3, mixing ratio of reagents and reaction time, etc. The quantitative constituents of two kinds of RP
wastes which were used for the preliminary synthesis test are shown in Tab. 2.2.1[11]and result of analysis for
character of adsorption in Tab. 2.2.2.[12]
Tab 2.2.1 Result of Quantitative Analysis of Chemical Composition
Constituent

Content (wt%)

Sample A

Sample B

SiO2
TiO2
Al2O3

56.09
0.26
13.19

21.76
0.07
6.05

RECYCLING TECHNOLOGIES FOR THERMOSET COMPOSITES

Constituent

Content (wt%)

Sample A

Sample B

Fe2O3
B2O3
CaO
MgO
MnO
ZnO
Na2O
K2O
Co
Pb
lg.loss*
Total
* Ignition loss

0.34
8.18
19.99
0.48
<0.01
0.02
0.30
0.20
0.05
<0.01
0.96
100.06

79

0.10
2.31
40.80
1.18
<0.01
0.18
0.08
0.10
<0.01
<0.01
26.93
99.56

Tab 2.2.2 Result of Evaluation of Synthesized Zeolite


Sample

Trial No.

Zeolite A

H-Sodalite

A
B

3
6

Diffraction Strength of X-rays

Specific SurfaceArea
(m2/g)

Cation Exchange Capacity (CEC)


(meq/100g)

1874
1277

324
201

190
190

0
0

3
The Present ConclusionToward Development of New Technologies
3.1
Research for the entrance
It should be said that chemical decomposition or pyrolysis including dry distillation has just begun. Now, proposal of
such technical development as separates and decomposes RP waste directly into monomer and solid residue is going
to be advocated. Studies for technic to utilize recovery method of raw material by reactive media of super critical
water or non-heating utilization of electro magnetic wave (micro wave) to promote or control chemical reaction by
irradiating are going to start. Study for chemical decomposition of high polymer waste by super critical water will be
done by the initiative of the Governmental Research Institute and non-heat utilization of micro wave by the hand of
the RTC/JRPS.
3.2
Development of the exit
Composite materials originally are made up of the combination of more than two different materials and it is quite
natural to think that recycling begins from separating or treating each component.
Besides, first thinking of separating organic and inorganic component as the intermediate process of recycling,
there will be no problem to find users for reuse of organic component. The core of problems is still reuse method of
solid residue, in particular inorganic part.
Until now, reuse for cement, construction materials (e.g. gypsum), civil engineering material (e.g. asphalt filler)
was mainly pursued and more efforts will be projected toward these streams. The problem is how far we could
recover the treatment cost.
However, from the viewpoint of recovery of process cost, there remains fairly questionable points which drive us
to enhance one more step added value of the reclaimed product. The problem is whether or not the direction of this
technological development will have a good chance of realization.

80

KITAMURA

If such recycled product and technic as would delight users have developed, all the problems would be settled.
Otherwise, we have to bear the cost by ourselves so as to close this recycling loop. But we do not yet think that we
are already in the stage to conclude there is no way.
Development works for reuse technology have just begun.
Acknowledgement
The author wishes to thank researchers of Government Industrial Research Institute, Shikoku (GIRIS)/MITI,
Research Centers cooperated with the GIRIS and of the Ship Research Institute/the Ministry of Transport (MOT) and
of the Public Works Research Institute of the Ministry of Construction (MOC) for their participating enthusiastically
in the discussion to seek new method of reuse technology and with pleasure for their offering data for this paper and
also would like to thank the members of the RP Waste Recycling & Treatment Council (RTC) of the JRPS who
cooperated to make great progress for the solution of this task.
Biography
Tatsundo Kitamura: He is Research and Analysis/Manager of the Japan Reinforced Plastics Society (JRPS) which
handles every affairs to promote development of RP/ Composites Industry in Japan. Received a BA degree from
Waseda University in 1960. He started to serve for the JRPS since 1977 and consistently engaged in Research
&Development works for the Industry. Also preside FRP Forum, Inc./a consulting office for R&D on RP/
Composites.
References
1.
2.
3.

4.
5.
6.
7.
8.
9.
10.
11.
12.

Takashi Shibata and the other three, Takeda Chemical Ind., Ltd., Tokyo Research Lab., Recycling of Molded SMC
Materials, Proceedings of the JRPS/The 36th FRP Annual Conference, p185189; October, 1991.
An experimental presentation at the 37th FRP Annual Conference; CON-EX92 TOC, Gotanda, Tokyo by Koito Mfg.
Co., Ltd. which took place October 20 to 22, 1992.
Kazuhide Hamada and the other three, Prefectural Research Center, Kohchi, Reuse Technology for Composite Materials,
Special Research Report for Important Regional Technology Development of Government Industrial Research Inst.,
Shikoku(GIRIS)/MITI, p9091; March 1993.
Hiroshi Kamishima and the other two, GIRIS. Ibid.[3], p7071 and p77.
Takao Kitamura and the other four, GIRIS. Ibid.[3], p8084.
Akira Kojima, Gunma College of Technology, Recycling RP Waste and A New Development for Building Materials,
Polymer Digest, p210; October, 1990.
Kazuhide Hamada and the other three, Prefectural Research Center, Kohchi. Ibid. [3] p8689.
Result of a joint research with a cement supplier by the RTC, May 1993.
Akira Kojima, Gunma College of Technology, Recycling RP WasteLater Pulverization and Development, at the Particle
Reclamation Sub-committee of the RTC, August, 1993.
Kiyoshi Katawaki, Itaru Nishizaki, Report for the Study of Establishing Evaluation and Design Technology for EcoMaterialization of Materials, The Public Works Research Institute of the Ministry of Construction; July 1994.
The RTC/JRPS, Report of the Result of Synthesis of Zeolite, p4, April 1994.
Ibid. [10] p7.

9
ENVIRONMENT-CONSCIOUS MATERIALS DESIGN OF
LIGHTWEIGHT PRECAST CONCRETE COMPONENTS WITH
RECYCLABLE FRP REBARS
T.FUKUSHIMA
Materials Department, Building Research Institute, Ministry of Construction, Tsukuba, Ibarakiken, Japan
K.YANAGI
Inorganic Materials Testing Division, Japan Testing Center for Construction Materials, Soka,
Saitama-ken, Japan
T.MAEDA
General Institute, Mitsubishi Chemical Corporation, Yokohama, Japan
Abstract

In order to try to reduce the environmental burdens of a mass of waste building materials that partly cause the
dominant global environmental problems, concept of environment-conscious materials (ecomaterials) design
of composite building materials, components, and/or elements is proposed, taking in advance into account of
the easiness of demolition and recycling. As part of this concept, hybrid structural systems composed of steel
reinforced concrete (RC) columns and new continuous fiber reinforced plastics (FRP) reinforced concrete
(FRPRC) beams connected with steel fasteners, recyclable precast reinforced concrete and various effective
demolition methods are considered. Lightweight high strength concrete components reinforced with recyclable
continuous fiber reinforced rebars using thermally, photochemically and biochemically resolvable resin, not
thermosetting epoxy and vinyl ester resin are recommended as ecomaterial design satisfying both the long
service lives and the recyclability of building materials, components and elements. As a trial to obtain effective
data for establishment of this materials design, some experimental research results by bending test of this
lightweight high strength precast concrete component specimens reinforced with both new types of shearing
carbon continuous fiber reinforced plastic (CFRP) rebars hardened by thermally resolvable resin and existing
type of CFRP main deformed rebars by thermosetting epoxy resin are reported.
Keywords: pitch-based high performance-type carbon continuous fiber reinforced plastic, chemically
resolvable resin, environment-conscious materials (ecomaterials) design, deformed rebar, hybrid structure,
lighweight precast concrete, recycle, service life,
1
Introduction
When human beings appeared for the first time on the earth and built houses as shelters from the severe external
environment, avoiding the fierceness of mighty nature, building materials were such natural materials as wood,
bamboo, stone and soil, and since then such natural materials have been often used continuously even until now as
building materials originally gentle to the environment. From the historical viewpoint, many kinds of new artificial
materials supplied by modern mass productive methods have caused various kinds of technological innovation time
by time not only in various industrial fields, but also in building field. Such new materials used as building materials
have become the key technology which promoted new functional building forms, emerging from old-fashioned
building ones.
It may be said that such kinds of modern artificial materials as steel, cement, concrete, glass, plastic and aluminum
have been used massively from the era of industrial revolution to the period of high-speed economical growth as
building materials in building field as it were a right material in a right place in correspondence with the regional
Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

82

FUKUSHIMA, YANAGI AND MAEDA

situations and the phases of the times. Such new advanced materials now developed at high pitch as advanced
composites, new metals, new polymers and fine ceramics will become used as building materials in the near future
after the settlement of technology of utilization in building field.
Various kinds of materials, whether organic or inorganic or metallic, have been used as building materials to
satisfy each required performance for given components and elements composing buildings under the given
environment. Development of new building materials have done so far from the viewpoint of pursuit of frontier
characteristics of lightweight, high strength and high durability in case of such structural materials as steel, concrete
and wood, and amenity characteristics of high function and flexible design in case of such non-structural materials as
finishing polymer and glass. In proportion as the importance of the global environment problem threatening the
existence of human beings itself in the coming 21st century becomes more and more appreciated day by day,
however, the characteristics of harmony with the global environment are becoming more and more required also for
building materials and components as well as characteristics of high performance. It is because the carbon dioxide
generation and consumption of energy at the time of production of building materials and a mass of various waste
building materials at the time of demolition, to not a few extent, cause the burden to the environment. We should in
the future make life cycle design of building materials, considering prediction of service life, life cycle cost analysis
and evaluation of the environmental load for the total life cycle of production, in service, disposal and recycling. The
impotance of this design was emphasized in the RILEM Workshop on Environment and Building Materials held in
Finland last year[1].
Recently the concept of environment-conscious materials (ecomaterials) has been newly proposed to try to make
an effective approach from material side for the preservation of the global environment[23]. We should treat with
the global environmental problem from building side hereafter, considering the correlation between building
materials and the environment in correspondence with this concept. In other words, we should establish the materials
design taking in advance into account of the compatibility of long service lives and recyclability of building
materials, paying specially attention to the harmony of building technology with the environment based upon the
easiness of demolition and recyclability of building materials and components.
This paper deals with environment-conscious materials (ecomaterials) design for lightweight high strength precast
concrete combined with continuous fiber reinforced plastic reinforcement using chemically resolvable resin. As a
trial to obtain effective data for establishment of this material design, some experimental research results by bending
test of this lightweight high strength precast concrete component specimens reinforced with both new types of
shearing carbon continuous fiber reinforced plastic (CFRP) rebars hardened by thermally resolvable resin and
existing type of CFRP main deformed rebars by thermosetting epoxy resin are reported.
2
Research methods
2.1
Methodology of ecomaterials design
There is an increasing requirement for building materials and components to be ecomaterials gentle to the global
environment. We should make ecomaterials design of building materials and components, taking into the
compatibility of their long service lives and harmony with the environment based upon the easiness of demolition
and recyclability.
In order to consider the correlation with building materials and environment, however, the following three aspects
are important.
1. influence of building materials on the global environment
2. influence of building material on the human dwelling environment
3. influence of the global environment on service lives of building materials
This paper mainly deals with the first aspects, paying special attenntion to ecomaterials design satisfying the
compatibility of their long service lives and harmony with the environment of composite building materials and
components said so far to be difficult in demolition and recycling, but the next two aspects should never fail to be
considered from the viewpoint of the compatibility of design as a whole.
As to the second aspect, we should make design not to use such building materials harmful to the human dwelling
as asbestos and radon-generating materials that may cause lung cancer.

ENVIRONMENT-CONSCIOUS MATERIALS DESIGN

83

Fig. 1 Outline of test specimens

As to the third aspect, we should establish new durability design based upon effective new predictive methods of
service lives of building materials and components and deterioration-protective ones, corresponding the increase with
time of deterioration agents with the progress of the global environmental problem. I have already carried out
research on new predictive methods of the progress of carbonation of concrete considering the increases in the
concentration of atmospheric carbon dioxide which is one of the main features of the global environmental problem
and reported in some papers[45]. Although I would like to emphasize the importance of this aspect from viewpoint
of durability design to assure long service lives of building materials and components, these research results are not
included in this paper for conveniences sake of the uniformity of the content of this paper.
2.2
Experimental method of tentative elemental technology for ecomaterials design
2.2.1
Test specimens
In order to design composite building components having both high performance and recyclability, lightweight high
strength precast concrete component specimens reinforced with carbon continuous fiber reinforced plastic (CFRP)
rebars hardened by chemically resolvable resin are considered. Although thermosetting epoxy resin of CFRP rebars
was tried to be wholly replaced with thermochemically, photochemically and biochemically resin, judging from the
present technological state-of the-art, only that of shearing CFRP rebars are converted into polyacetal thermoplastic
resin.
Test specimens are rectangular-shape beams (100100600mm) of lightweight high strength precast concrete
components reinforced with both new types of shearing pitch-based high performance type carbon continuous fiber
reinforced plastic (CFRP) rebars (tensile strength; 1.67GPa, tensile elasticity; 147GPa, 7mm , pitch; 38.5mm, Pt=1.
0%) hardened by thermally resolvable resin (resin; polyacetal, fiber content; 1%) and four existing type of CFRP
main deformed rebars (tensile strength; 2.25GPa, tensile elasticity; 147GPa, 8mm , tensile side; 2, compressive
side; 2, Pt=Pc=l.3%) by thermosetting epoxy resin(fiber content; 65%) were made as shown in Fig. 1.
Test specimens were cured for 26days in 20C water after placing , and dryed for 2days in a room air-conditioned
at constant temperatuter (20C) and constant humidity (60% R.H.) .
2.2.2
Lightweight high strength concrete
Lightweight high strength concrete was made for designed compressive strengths of 29.4, 39.2, and 58.5MPa.
Ordinary Portland cement was used. As coarse aggregate, artificial lightweight aggregate (ovendry specific gravity;
1.30, percentage of water absorption; 26.5%) was used, and as fine aggregate, river sand (surface-dry secific gravity;
2.62, finess modulus; 2.68)) was used. Further standard type and high range water-reducing agent were also used as
chemical admixtures. In Table 1 are shown mix proportions to obtain the designed compressive strengths, and in
Table 2 are shown the really obtained compressive strengths.
2.2.3
Bending test
Bending test was done for test specimens after 28 day curing by a Universal -type testing machine based upon twopoint concentrated loading for 500mm span simple support as shown in Fig. 2. Displacement and

84

FUKUSHIMA, YANAGI AND MAEDA

Fig. 2 Outline of bending test


Table 1 Mix proportions of lightweight concrete
designed
compressive
strength MPa
water

adimixture slump cm water-cement sandkg/m3


ratio %
aggregates
ratio %
cement

sand

29.4
2.98
18.7
60
44.1
4.63
20.1
45
58.8
9.97
15.8
25
artificial lightwe lightwe eight coarse aggr cregate

unit cotnent kg/m3

gravel

43.5
42.5
39.5

179
160
160

kg/m3

298
356
640

unit weight air content

%
810 671 1958
794 683 1993
644 658 2102

3.1
1.9
2.3

Table 2 Really obtained compressive strength of lightweight concrete (after 28 day age)
designed compresive strength (MPa)

number

real compressive strength MPa

29.4
2
3
mean value
44.1
2
3
mean value
58.8
2
3
mean value

1
38.9
36.2
37.9
1
50.1
46.5
48.3
1
69.1
63.3
66.1

38.7

48.4

65.5

strains were measured by using electric displacement meters and strain gauges. The situations of the ocurrence of
crack in concrete during loading were observed by the eyes.
3
Research results and discussion
3.1
Concept of ecomaterials design
3.1.1
Compatibility of recyclability with long servive life
Building materials and components play the important role in that they compose buildings supporting houses and social
capitals. When given building are requested to have service lives of 30 to 500 years, needless to say, structural
materials and components have to show as long life-span as possible during their designed service lives, which is
different from cases of daily necessities and such durable consumer goods as automobiles and electric appliances
often exchanged in considerably short times. If rational durability design for building materials components not

ENVIRONMENT-CONSCIOUS MATERIALS DESIGN

85

Fig. 7 Concept of repeated partial recycle (cascade recycle) of lightweight concrete

easily to deteriorate are made, it will help to assure buildings having long service lives. and decrease the frequency
of generation of waste materials, resulting in the reduction of burdens to the global environment. Consequently, in
order to reasonably evaluate the environmental loads of building materials, it is necessary for us to calculate the
amounts of use of natural resources, amounts of generation of carbon dioxide, amounts of consumption of energy by
dividing them by service lives. Many building materials and components, however, are liable to become burdensome
residues almost difficult in demolition and recycling. From the viewpoint of the recyclability, we should establish
materials design, considering in advance such gadgets as screws and threads which enable the easy demolition of
machines and textiles when they are pulled out after designed service lives.
Fig. 3 and Fig. 4 show the comparison of concepts of new reinforced concrete with CFRP reinforcement with
ordinary reinforced concrete with steel reinforcement, considering the compatibility of recyclability with long
service lives. As to steel reinforced concrete buildings as one of the main structures for the present in building field,
we freshly remember that the deterioration due to corrosion of reinforcing steel caused by unexpectedly fast
carbonation of concrete in the megalopolitan areas, penetration of chloride ions into concrete in the seashore areas,
and mixing of chloride ions into concrete as sea sand and admixtures became a serious problem as concrete crisis.
In order to solve this problem, new approach has been made to durability design for making longer service lives of
steel reinforced concrete components, and further a new active movement has rapidly gained ground to use various
types of new continuous fiber reinforced plastic (FRP) reinforcements as replacement of steel reinforcement liable to
corrode. Such FRP reinforcements have excellent characteristics of light weight, high strength and high corrosionresitance. When they are to be disposed after designed service life, however, it is very dificult to recycle and reuse
them, becuse they are hardened by such thermosetting resins as epoxy and vinyl ester, and they give to a great deal
of load to the global environment. If thermosetting resin is replaced with thermochemically, photochemically, and
biochemically resolvable one, these FRP reinforcement become recyclable. This is one of the elemental material
technology having the compatibility of recyclabilty with long service lives that should be established in the near
future. Fig. 3 and Fig.4 show these situations.
3.1.2
Ecomaterial design of lightweight precast concrete combined with chemically recyclable FRP
reinforcement
Fig. 5, Fig. 6 and Fig. 7 show the fundamental concepts of ecomaterials design of lightweight precast concrete
combined with chemically recyclable FRP reinforcement, considering the recyclability of both FRP reinforcement
and concrete. In Fig. 5 are shown the process of materials design for selection and evaluation of wanted FRP
reinforced concrete (FRPRC) as building materials and components in given buildings under the given environment,
judging from the comparison of required performances of FRPRC. In order to convert this materials design into
ecomaterials design, we should evaluate another important performance of the harmony with the environment. Fig. 6
shows especially the concept of ecomaterials design of lightweight precast concrete reinforced with FRP
reinforcement, taking into account of total life cycle design based upon prediction of service life, life cycle cost
analysis, and life cycle assessment (evaluation of total load to the environment). In Fig. 7 is shown the concept of
repeated partial recycle (cascade recycle) of lightweight precast concrete. In this ecomaterials design, hybrid
structural systems composed of steel reinforced concrete (RC) columns and new continuous fiber reinforced plastics

86

FUKUSHIMA, YANAGI AND MAEDA

Fig. 3 Comparison of concepts of new reinforced concepts and CFRP reinforcement with ordinary reinforced concrete with steel
reinforcement.

(FRP) reinforced concrete (FRPRC)beams connected with steel fasteners are considered, from the viewpoint of the
easiness of demolition. Recyclable FRP reinforcement hardened by chemically resolvable resin, not thermosetting
resin are considered, and in case of FRP hardened by thermosetting resin , recycling by reuse of pulverized FRP
powder as fine aggregate of lightweight concrete are also proposed. On the other hand, considering the
correspondence to the expected situations in the future of the exhaustion of natural aggregates, or the prohibition of
their gathering from the viewpoint of the preservation of natural environment, effective use of artificial lightweight
aggregates are recommended.
As methods of easy demolition of FRPRC, we can consider the interface detachment by use of high-frequency
wave heating based upon the good electric conductivity of carbon fiber, or by use of the strong vibration caused by
microwave, and the expansive destruction of concrete itself by use of external stimulation of micro capsules
including expansive agents in advance mixed in concrete. If such elemental techniques are successfully developed, we
can establish the wanted purpose of the compatibility of long service lives with recyclability of building materials
and components, and it will greatly contribute to the reduction of load to the global environment.
3.2
Results of bending test of CFRP reinforced lightweight high strength concrete
Table 3 shows the results of bending test for CFRPRC specimens. In Fig. 8 are shown load-displacement curves for
test specimens having three designed compressive strengths of lightweight concrete. Both main CFRP
reinforcements and shearing CFRP reinforcements show the completely elastic tensile characteristics, but bending
behaviors of CFRPRC specimens are almost the same as ordinary steel RC, except that CFRP reinforcements do not
show such yielding characteristics as steel reinforcement. As the load was increased, occurrence of initial bending
cracks, bending-shearing cracks, cracks along main reinforcements were observed in turn in the tensile side of beam
specimens, and thereafter shearing cracks across cross sections and compressive failure of concrete

ENVIRONMENT-CONSCIOUS MATERIALS DESIGN

87

Fig. 4 Comparison of concepts of prediction of service life between new reinforced concrete with CFRP reinforcement and ordinary
reinforced concrete with steel reinforcement
Table 3 Results of bending test
Test specimen

ratio of
reinforce
ment pt,
pw

label* number

experime calculate cPcr


ntal
d value
value
cPcr
ePcr

cPcr

bPul*4) bPu2*5)

60

2
3
mean
value
45

Pt=0.
013
37.9

Pw=0.
010
48.3

compres
sive
strength
of
concrete
(MPa)

sPu*6)

5,880
5,390
5,145

initial crack load*2)

experim
ental
value
Pmax

maxium load (N)

calculated value

Pmax Pmax

Pmax

4,165

0.7

44,100

44,100
42,728
43,639

129,037

51,548

51,940
43,120
48,873

129,037

5,782

1
0.9
0.9

6.419

2
5,537
6,527 0.9
3
5,537
0.9
mean
5,831
0.9
value
25
1
8,575
2
66
8,330
7,624 1.1
3
7,987
1.1
mean
8,301
1.1
value
*1) labels show water-cement ratios
*2) calculated values based upon loadstrain curves
*3) based upon the equation

1.1
51,450
48,020 129,037
601,564
53,341

bPul

10,509

5,329

bPu2 sPu

0.3
0.3
0.3

0.3

0.4

0.4
0.4
0.4

0.83
0.81
0.82

0.4

0.5

108,329 56,948

0.4
0.3
0.4

0.5
0.4
0.5

0.91
0.76
0.86

117,198 63,171

0.4
0.5
0.4

0.4
0.4
0.5
0.5

0.4
0.76
0.96
0.84

0.83

0.91

0.81

88

FUKUSHIMA, YANAGI AND MAEDA

Fig. 5 Concept of materials design of FRPRC taking into account of the harmony with the environment.
Test specimen

ratio of
reinforce
ment pt,
pw

compres
sive
strength
of
concrete
(MPa)

label* number

experime calculate cPcr


ntal
d value
value
cPcr
ePcr

cPcr
bPul*4) bPu2*5) sPu*6)
*4) bending failure of concrete
*5) compressive failure of concrete
*6) shearing failure

initial crack load*2)

experim
ental
value
Pmax

Pmax Pmax

calculated value

maxium load (N)

bPul

bPu2 sPu

Pmax

in the compressive side were observed step by step. As seen from Table 3, shearing failure dominates the faille of
these CFRPRC specimens, despite of rather densely spaced shearing reinforcement. In order to obtain bending
failure effective to component design, beam length of test specimens is considered to have to be longer. It was found

ENVIRONMENT-CONSCIOUS MATERIALS DESIGN

89

Fig. 6 Processes of environment-conscious material design of precast lightweiht concrete reinforced with continuous fiber
reinforced plastic (FRP) reinforcement

Fig. 8 Load-displacement curves of CFRPRC specimens

from this bending test that shearing CFRP reinforcements hardened by thermoplastic polyacetal resin are applicable
enough to be used as recyclable reinforcements for precast lightweight high strength concrete.
4
Conclusions
Research on the concept of environment-conscious materials (ecomaterials) design of lightweight precast
components reinforced with recyclable fiber reinforced plastics (FRP) rebars were carried out in order to establish
the compatibility of long service lives with recyclability of building materials and components, as part of methods of
effective solution of the global environmental problems. Further, as a trial to obtain effective data for establishment
of this materials design, bending test of specimens of lightweight high strength precast concrete component
reinforced with both new types of shearing carbon continuous fiber reinforced plastic (CFRP) rebars hardened by

90

FUKUSHIMA, YANAGI AND MAEDA

thermally resolvable resin and existing type of CFRP main deformed rebars by thermosetting epoxy resin were done.
The following results are obtained.
1. From the viewpoint of the easiness of demolition and recycling, not ordinary steel reinforced concrete (RC)
cast-in place completely in a body, but hybrid structural systems composed of steel reinforced precast RC columns
and new continuous fiber reinforced plastics (FRP) reinforced lightweight precast concrete (FRPRC) beams
connected with steel fasteners are recommended.
2. As to hardening resin of FRP reinforcement, althogh the compatibility with fire-resistance should be taken in to
account, in order to make FRP reinforcement more recyclable, conversion from thermosetting resin into
thermochemically, photochemically and biochemically resolvable resin is recommended.
2. As to the recycle of lightweight high strength precast concrete, not complete recycle, but repeated partial
recycle (cascade recycle) is recommennded.
3. It was found from bending test of CFRPRC specimens using both new types of shearing carbon continuous
fiber reinforced plastic (CFRP) rebars hardened by hardened by thermoplastic polyacetal resin and existing type of
CFRP main deformed rebars by thermosetting epoxy resin that this shearing CFRP reinforcements are applicable
enough to be used as recyclable reinforcements for precast lightweight high strength concrete.
5
References
1.
2.

3.
4.

5.

Proceedings of RILEM Workshop on Building Materials and Environment, held on November 1819, at VTT, Otaniemi,
Finland.
Report of Fundamental Research Based upon the Budget in 1992 Fiscal Year for Promotion and Adjustment of Science and
Technology Ecomaterials for the Preservation of the Global Environment (1993). Research Development Bureau of
Science and Technology of Japan, Tokyo (in Japanese).
Report of Research on Rare Metals Ecomaterials (1991), Ecomaterials (II) (1992), The Society of Non-traditional
Technology (in Japanese).
Fukushima, T. (1991) Predictive Methods on the Progress of Neutralization (Carbonation) of Concrete by Unsteady State
Dynamic Analysis Considering the Influence of Tendency of Increase in Concentration of Atmospheric Carbon Dioxide,
Proceedings of the 2nd CANMET/ACI International Conference, held in Montreal, Canada, August 49, 1991, edited by
V.M.Malhortra, Supplementary Papers, American Concrete Institute, pp.-545564.
Fukushima, T. (1994) Predictive Method for the Progress of Neutralization (Carbonation) of Concrete by Step Response
Analysis, Proceedings of the 48th Meeting on Cement Technology, Japan Cement Association, Tokyo (in Japanese)
pp. 614619.

10
RECYCLING OF PLASTICS WASTES FROM ELECTRONIC
PARTS PRODUCTION PROCESSES
M.IJI and S.YOKOYAMA
Resource and Environment Protection Research Laboratories, NEC Corporation, Kawasaki,
Japan

Abstract
We have investigated the recycling of the thermosetting plastic waste generated in the production of printed
wiring boards (PWBs) and IC packages, which are the main plastic products in the electronic parts industry. A
process for pulverizing mold residues of PWBs and separating the resulting powder into copper and the
powder consisting of glassfiber and resin (glassfiber-resin powder) was developed. More than 90% of copper
(maximum 94%) was recoverable from the PWB waste powder, which had an average grain size of 100300
m. The recovered glassfiber-resin powder could be used to improve the strength and the thermal expansion
properties of epoxy resin type paints and adhesives, and also proved useful as a filler for polymer mortars. The
mold residues of molding resin for IC package was also succeeded in pulverizing into a powder, which was
found to have a reactive surface nearly as good as that of silica powder and could be used as a general filler and
most promisingly, as a decorating agent which has the added advantage of improving the surface hardness of
resin type construction materials.
Keyword: IC package, molding resin, recycling, printed wiring board
1
Introduction
For electronic parts, thermosetting plastics, such as epoxy resin and phenol resin, are used for the high thermal
stability and insulating properties, etc. The residues of these molded resins (mold residues) generate from the
production process of electronic parts. Up to now, recycling of these wastes was difficult, because these waste are
impossible to remelt for curing and hard to pulverize for inclusion of inorganic filler, such as glassfiber and silica
powder. For these reasons, the wastes are disposed of as industrial wastes in landfill sites. Therefore, the recycling of
such wastes was a serious problem to the electronic industry. Representatives of these wastes are the mold residues
of PWBs and molding resins for IC packages. We have investigated recycling possibility for these plastic wastes.
2
Recycling PWB waste
PWBs consists of cured epoxy resin, glass fiber and copper circuits. The production process of PWBs generates 10
20% of mold residues. A recycling process, consisting of the combustion of the resin and recovery of copper, was
attempted.[1] However, toxic gases such as organic bromides were generated by the flame retardant included in the
resin. On the other hand, recovery of the copper by etching the surface of PWBs and the reuse of pulverized PWBs
as a filler for plastics was tried. [2] However, the process required expensive equipment for waste water treatment
Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

92

RECYCLING OF PLASTICS WASTE

Fig. 1. Pulverizing and separating for PWB waste

Fig. 2. Copper recovery ratio from pulverized PWBs

and was impossible to apply to multilayer PWBs. We have investigated pulverizing PWB waste and separating the
resulting powder into the copper and glassfiber-resin powder, and reusing the recovered glassfiber- resin powder for
construction materials.
2.1
Pulverizing and separating process
We have developed a process for pulverizing PWB waste, and a process for separating the resulting powder into
copper rich powder and glassfiber-resin powder.
Figure 1 presents the flow chart for these pulverizing and separating processes. The pulverizing process was
highly effective and abrasion resistant by combination of the crushing step which uses cutting and shearing forces,
and the fine-pulverizing step which uses compressive and shearing forces. The separating process consists of the
gravity separating step and the static electronics separating step.
As Figure 2 shows, for the recovery of the copper during the gravity separation step, the effective particle size for
the PWB wastes (16 layers) was found to be from 100 to 300 m average particle size. The recovery ratio of the
copper from the PWB was more than 90% (maximum 94%), and the copper content in the copper rich powder was
more than 20wt% (the original copper content in the PWB waste was 7wt%.). Figure 3 shows size distributions for
each components (copper, glassfiber, resin) in a PWB powder with a average particle size of 150 m. Figure 4

IJI AND YOKOYAMA

93

Fig. 3. Size distribution of each components of pulverized PWBs

Fig. 4. pulverized PWBs (SEM)

shows this powder picture by a scanning electron microscope (SEM). These demonstrated that the copper was
pulverized more roughly than the glassfiber and the resin, and was well separated from them. This size difference was
considered to be due to the ductility of copper and bridling of glassfiber and resin. The effective recovery of copper
from the PWB powders of these particle sizes was thought to be due to this size difference, and to the difference in
specific gravity of each component (copper; 8.9, glassfiber; 2.5, resin; 1.2).
The copper rich powder, produced by the gravity separation step, was further purified by the electrostatic
separation step which uses the difference in electronic conductivity of each component. The copper rich powder
having now a 82wt% of copper content could be recovered by this step and was provided to be a valuable copper
resource.
2.2
Glassfiber- resin powder applications
We have investigated the use of the recovered glassfiber-resin powder as a filler for paints, adhesives and polvmer
mortars.

94

RECYCLING OF PLASTICS WASTE

2.2.1
Use for paints and adhesives
We have investigated the recovered glassfiber-resin powder for its properties as a filler for epoxy resin compounds
which are used as paints or adhesives. and compared it to conventional fillers, such as talc and calcium carbonate.
The epoxy resin compound, composed of bisphenol A type epoxy resin (50.0wt%), aliphatic polyamine type hardener
(18.0wt%) and filler (32.0%), was prepared. Strength and thermal expansion properties were measured for the
molded epoxy resin compound cured 23C for 7 days. Viscosity was measured for the epoxy resin compound before
adding the hardener. Adhesive strength was measured by tearing two ferric boards bonded with the epoxy resin
compound which was composed of bisphenol A type epoxy resin (49.2wt%), polyaminoamide type hardener (18.0wt
%), and filler (32.8wt%), and was cured at 23C for 7 days.
As Table 1 shows, the glassfiber-resin powder was found to be better than either talc or calcium carbonate at
improving the strength properties and decreasing the thermal expansion of the molded epoxy resin compound. The
viscosity with the glassfiber-resin powder was nearly equivalent to that produced by the addition of the talc, and was
greater than that produced by the addition of the calcium carbonate. Table 2 presents that the adhesive strength with
the glassfiber-resin powder proved to be satisfactory, almost equal to that achieved with addition of the talc and the
calcium carbonate.
Table 1. Properties of epoxy resin compound containing glassfiber-resin powder
Filler

Average particle size


(m)

Viscosity
(Pas)

Tensile strength
(kgf/cm2)

Flexural strength
(kgf/cm2)

Impact strength
(kgfcm/cm2)

Calcium carbonate
Talc
Glassfiber and resin
44
20
13

2.6
11
55
75.2
66.2
63.2

25.8
69.2
73.0
622
628
589

477
280
505
888
878
965

661
577
812
3.04
3.15
3.60

2.45
2.50
2.95

Table 2. Adhesive strength of epoxy resin compound containing glassfiber-resin powder


Filler

Average particle size


(m)

Tensile strength
(kgf/cm2)

CaCO3
Talc
Glassfiber and resin
44
20
13

2.6
11
55
158
160
152

157
138
159

2.2.2
Use for polymer mortars
The recovered glassfiber-resin powder was tested as a filler for polymer mortars[3]. The mortars were composed of
filler, silica sand and binder, dried for one day, and cured at 70C for 15 hrs. A glassfiber-resin powder and a
calcium carbonate powder for comparison (each with an average size of 52 m) were used as fillers. Bisphenol F
type epoxy resin, poly methacrylic methyl (PMMA), and unsaturated polyester (UP) were used as binders.
As Figure 5 shows, the strength properties with the glassfiber-resin powder were nearly equivalent to that
produced by addition of the calcium carbonate and proved to be satisfactory for the use as a filler for polymer
mortars.
3
Recycling molding resin waste
We have investigated the reuse of ground up molding resin waste as a filler or a decorating agent for construction
materials. Since the surface reactivity of the molding resin waste powder has a great influence on properties of the

IJI AND YOKOYAMA

95

Fig. 5. Strength properties of polymer mortar containing glassfiber-resin

Fig. 6. Chemical structure of molding resin powder

material including the waste powder, we have also determined the surface reactivity by surface analysis and by
measuring the degree of reaction of the molding resin powder and virgin epoxy resin.
3.1
Surface reactivity of ground up molding resin
A conventional molding resin waste (<150 m), composed of silica filler (70wt%), o-cresol novolac type epoxy
resin, phenol resin type hardener, and additives (amine type catalyze, wax, carbon powder type coloring agent, etc.)
was ground up with cutting, shearing and compressive forces. The chemical surface structure of the molding resin
powder was determined by means -of X-ray micro analyzer (XMA), FT-IR and 13C-NMR. As a result, the surface
was found to have reactive functional groups, such as epoxy group which had not reacted with the phenol resin, and
alcohol group generating from the reaction of the epoxy resin and the phenol resin, and silanol group on the silica
surface.

96

RECYCLING OF PLASTICS WASTE

Fig. 7. Analysis of reactivity between molding resin powder and epoxy resin by 13C-NMR

Fig. 8. Comparison of reactivity between molding resin powder and epoxy resin by gravimetric method

The reactivity of the molding resin powder was determined by measuring the degree of reaction of the powder and ocresol novolac type epoxy resin. The molding resin powder and the epoxy resin were mixed in a 100/5 weight ratio
in acetone, dried for a day and heated at 170C.
13C-NMR measurements performed on this mixture produced the reference signal level for the epoxy functional
group of the epoxy resin before heating, and the reduced epoxy functional group signal level after heating for 5 hrs at
170C, shown in Figure 7. Similarly, we took a mixture before heating and washed away the unreacted epoxy resin
with acetone, and then did the same with a mixture which had been heated as above at 170C for the two times
periods shown in Figure 8. By comparing weights, we were able to determine the degree to which reactions
occurred. Further, we compared these with those produced with silica powder (<150 m)/epoxy resin mixture (also
shown in Figure 8). As may be seen, the molding resin powder has reactivity nearly as good as that of silica powder.
These results show that the molding resin powder added to polar resin matrix, such as epoxy resin, can bind sufficiently
to the resin matrix.

IJI AND YOKOYAMA

97

3.2
Molding Resin Powder Application
3.2.1
Use as a Filler
We investigated the use of the molding resin powder (<150 m) as a filler for construction materials composed of
bisphenol A type epoxy resin and amine type hardener, and compared the material properties with those produced
with a silica powder filler (<150 m). Furthermore, the effect of surface treatment of the molding resin powder on
these properties was examined by using epoxy or amino silane coupling agents, which were added at 1wt% to the
molding resin powder and heated at 100C for 1 hr.
As table 3 shows, the strength properties of the molded construction material with the molding resin were not as
good as that of the silica powder, but were sufficient for use as a construction material, and they were improved by
the surface treatment with the silane coupling agents. The viscosity of the material with the molding resin powder
before adding the hardener was greater than that of the silica powder. This viscosity difference was thought to be
mainly caused by a difference in specific gravity (molding resin; 1.8, silica; 2.2).
Table 3. Use of molding resin powder as filler
Property

Silica
Molding resin
35wt% 50wt% powder
30wt%

40wt% 50wt% Epoxy silane treatment Amino silane treatment

Flexural strength
(kgf/mm2)
Impact strength
(kgf/cm/cm2)
Viscosity (poise)

10.4

10.7

8.6

7.8

8.9

10.2

9.7

1.4

1.5

1.2

1.1

1.2

1.2

1.1

96

240

116

294

1380

1600

1420

3.2.2
Use as a decorating agent
Since the molding resin is black from the original presence of carbon black, we decided to test it as a decorating
agent for a construction material, composed of acrylic resin (35wt%) and aluminum hydroxide powder (65wt%). As
Figure 9 shows, the molded construction material containing 10wt% of molding resin powder (<1mm) had a good
appearance, similar to marble. Also, the material was found to have high stability during cutting, produced by the
strong binding of the molding resin powder and matrix resin. Table 4 shows strength properties, thermal stability and
surface hardness. The strength properties with the molding resin powder were not as good as that of
Table 4. Use of molding resin powder as decorating agent
Property

<Blank>
Acrylic resin
+Al(OH) 65wt%

+Molding resin
<1mm 10 wt%

<1mm 20wt%

<5mm 10wt%

Flexural strength (kgf/mm2)


Impact strength (kgf/cm/cm2)
Tensile strength (kgf/mm2)
Thermal deformantion temperature (C)
Surface hardness (H)

5.5
1.5
2.7
94
6

4.5
1.6
2.9
93
7

4.3
1.7
2.9
93
8

3.6
1.6
2.3
93
7

the original material, but all values tested produced a material which was still sufficiently strong. The thermal
stability with the molding resin powder proved to be satisfactory, equal to that achieved with original material. The
surface hardness with the molding resin powder was better than that produced by original material. This surface
improvement was mainly due to the hard silica filler included in the molding resin.

98

RECYCLING OF PLASTICS WASTE

Fig. 9. Construction material containing molding resin powder

4.
Conclusion
We have investigated the recycling of the thermosetting plastic waste (mold residues) generated in the production of
PWBs and IC packages, which are the main plastic products in the electronic parts industry. We developed a process
for pulverizing mold residues of PWBs, and separating the resulting powder into copper rich powder containing 82wt
% of copper and the glassfiber-resin powder. More than 90% of copper (maximum 94%) was recoverable from the
PWB waste powder, which had an average grain size of 100300 m. The recovered glassfiber-resin powder could
be used to improve the strength and the thermal expansion properties of epoxy resin compounds which are used as
paints and adhesives, and also proved useful as a filler for polymer mortars. The mold residues of molding resin for
IC package was also succeeded in pulverizing into a powder, which was found to have a reactive surface nearly as
good as that of silica powder and could be used as a general filler and most promisingly, as a decorating agent which
has the added advantage of improving the surface hardness of resin type construction materials. We conclude that the
recycling of the thermosetting plastic waste containing glassfiber or silica powder, such as PWBs and molding
resins, is basically possible for the use of the waste powder as a filler or a decorating agent for resin type construction
materials.
5.
Reference
1.
2.
3.

Technical Information Institution. (1991) Plastics Waste Problem and Measure, pp5160.
Rosett, L.K and Rosett, K. (1983) Plastics Compounding, 6 (7), pp4754.
Yokoyama, S., Iji, M., and Ohama, Y. (1993) Proceeding of the 4th Annual Conference of the Japan Society of Waste
Management Expert, pp297300.

11
PREPARATION AND PROPERTIES OF LIGHTWEIGHT HIGHSTRENGTH MORTARS CONTAINING FRP FINE POWDER AS
AGGREGATE
A.KOJIMA
Department of Industrial Chemistry, Gunma College of Technology, Maebashi, Gunma-ken,
Japan
S.FURUKAWA
Department of Civil Engineering, Gunma College of Technology, Maebashi, Gunma-ken, Japan

Abstract
Mortar was prepared with aggregate of FRP fine powder as well as sirasu baloon. The mortar containing FRP
as part of the aggregate became more lightweight by heating at 200C, when its flexural strength increased up
to 2.5 times and its compressive strength upto 3.5 times. Improving of the flexural strength and deflection of the
FRP mortar was tested by reinforcing with fibers (fly ash, asbestos, Vinylon, polypropylene fibers and carbon
fibers). Among the reinforcing fibers, asbestos was the most effective. Partial replacement (10~30%) of
ordinaly portland cement by silica fume also incresed the compressive strength. Finally it resulted in the
preparation of lightweight high-strength mortar with a bulk density of 1.0g/cm3, aflexural strength of 7.1MPa,
and a compressive strength of 41MPa.
Analysis of FRP fine powder and FRP-containing mortar suggest that the improvement in strength of the FRPcontainig mortar by heating is due to the structural change of FRP fine powder.
Keywords: FRP fine powder, recyclization, lightweight mortar, strengthening
1.
Introduction
FRP (glass fiber reinforced plastics) is widely used for auto parts, small-sized boats, bathtubs, etc. as a material with
lightness, high strength, and durability. Its use is increasing yera by year, reaching nealy 500,000 tons per year at
present.
On the other hand, the quantity of scrapped FRP is also increasing, But its disposal and recycling have constituted
a great social problem, as there is no effective reusing system for it.
In Germany, the ERCOM shredder system was developped to recycle SMC automobile parts for reuse in existing
SMC production lines.
After shredding the SMC parts, the material is further processed by grinding the sheredded, material into powder/
fiber mixture, drying the material, air separating and sieving into various sizes and as a results, depending on the type
of composite, it is reported that the fiber with powder randing in the particle sizes of 0.25~3.0mm, and 3~15mm are
obtained.
The authos1) have endeavored for the preparation of light weight mortar with an aggregate of fine-powdered
(10~20m in diameter) FRP waste to utilize the scrapped FRP.
Minute examinations of the preparing conditions showed that a mortar with an aggregate of fine-powdered FRP
(FRP morta) can be lightened and strengthned by high temperature curing at about 200C.

Disposal and Recycling of Organic and Polymeric Construction Materials, Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

100

KOJIMA AND FURUKAWA

But its mechanism has not been clarified. This study aimed to find a preparing process of FRP mortar with less
weight and higher strength by examining the material combination, curing conditions, and fiber addition at FRP
mortar molding, and to clarify the mechanism of lightening and strengthening by high temperature curing.
2.
Preparation of FRP micro fine powder
FRP fine powder was prepared on the FRP pulverizer (ASAOKA Co. Ltd.) as shown in Photo 1. The FRP pulverizer
was devoted consisting of (1) automatic feeder, (2) micropowdering machine, (3) unground material separatoy, (4)
electrostatic remover, and (5) micropowder collector (Fig.1).
The FRP pulverizer features a new grinding technology using diamond wheel cutter capable of pulverizing waste
SMC, BMC, hand lay up, synthetic marble, and CFRP materials into micropowder instanteneously in one process.
The standard diamond wheel cutter (250mm in diameter, 300mm in width) rotates at 800rpm. The cut of size (under
50250600mm) waste materials are fed into the automatic feeder against the diamond wheel cutter to be micropowdered, drown to the micropowder collector by passing througha separator where the unground material is
separated and an electrostatic remover is incorporated to prevent spontaneous explosion. The minimum cutting
capacity of the micropowdering machine is 50kg/hr depending on the cuting size of the waste materials.
In observing the micropowder processed by the diamond wheel cutter with a scanning electron microscope. As can
be seen from Photo 2, there are, in part, comparatively large powder around 20 m but the majority is less than that
and around 4~5 m. By incresing the magnification further, as shown in Photo 3, many powder around 2 m can be
recognized. Also, as glass fiber in its fiber from was recognized, the cut section was observed by the microscope as
per Photo 4. The cross section shows that the glass fiber was cut instantaneously with a very sharp edge. Also, very
minute powder less than 1 m attacched the glass fiber can be seen.
Nevertheless, the grain diameter of the micropowder is within the range from 0.4 m to 100 m but the grains
around 16 m is the greatest and the average grain diameter is 13.5 m.
3.
Experimental
3.1.
Material used
Normal Portland cement was used, as obtained commercially. Aggreates used for the preparation of FRP mortar
were Sirasu balloon (Sankilite B03) and fine powder of silicafume. FRP fine powder was made by pulverizing a
plastic bathtub produced by hand lay up method, using an FRP pulverizer (Asaoka Co. Ltd.). Its average particle size
was 15m. Fig. 2 shows the particle distribusion of FRP fine powder.
Five kinds of fiber were used to reinforce the mortar: fly ash fiber (Itochu Co. Ltd., 10~20mm in fiber length, 12.
6m in diameter, 1246MPa in tensile strength, and 63GPa in elastic modulus), asbestos (LAB Chrysotile Inc., 4T
300, 8~12mm in fiber length), Vinylon (Unichika Co. Ltd., AB 1800, 6mm in fiber length) , polypropylene (5mm in
fiber length), and carbon fiber (pitch-based chopped strand, 5mm in fiber length).
3.2.
Prepation of FRP mortar
The preparation procedure of FRP mortar was shown in Fig.3. Mixture of fine-powdered FRP and Sirasu balloon
with a certain compounding ratio was added to normal Portland cement, and stirred.
The compounds obtained had an FRP/cement ratio (FRP/C) of 0.47 and a Sirasu ballon/cement ratio (S/C) of 0.28.
Water was added so that the water/cement ratio (W/C) would be 0.75, which was stirred in a mortar-mixer.
It was poured into a mold (JIS: 4cm4cm16cm), and before its fluidity was lost, it was pressurized through a
pressing board with an oil press (3.9MPa, 15min.). After one days standing it was unmolded and cured in water for a
week. Then it was heated in a drier kept at 200C for three hours for high-temperature curing in order to give higher
strength.
Moreover, lightneing and strengthening of FRP morta was tested by replacing some part (10~30%) of cement with
silica fume. FRP mortars without high temperature curing, and those without pressurization were also prepared for
comparison.

LIGHTWEIGHT HIGH-STRENGTH MORTARS

101

3.3.
Preparation of fiber-reinforced FRP mortar
Normal Portland cement was mixed with a certain quantity of water to make cement paste. Each kind of the fibers
was mixed into this cement paste by 3% weight of the cement.
Premixuture of FRP fine powder and Sirasu balloon was added to it, poured into a mold (4cm4cm16cm) after
stirring, and pressurized (3.9MPa, 15min.) before losing fluidity.
After one days standing it was unmolded, cured in water (one week), and allowed high temperature curing (200
C, 3hr) to prepare fiber-reinforced FRP mortar.
3.4.
Measurement
Bending strength and compressive strength of the FRP mortar prepared were measured by JIS R 5201. At the
bending test, deflection at the central part of the specimen was measured under the maximum load.
To investigate the structural change of fine-powdered FRP when heated at 200C in mortar, it was kept in a drier
for three hours at certain temperatures (room temperature, 100C, and 200C), out or in sodium hydroxide solution.
With the FRP fine powder after treatment, IR spectra were taken by an infrared analyzer (Nicolay Co. Ltd.,
Magna-IR 750).
For FRP mortars with and without heat curing, their fractured surface were observed by a SEM, while fine
powders of them were analyzed by X-ray diffractometry.
4.
Results and cosideration
4.1.
Strength (FRP mortar without fiber reinforcement)
Table 1 shows the bulk density and mechanical strength of FRP mortars prepared with and without pressurization or
high temperature curing. Figures in parentheses after the mean values show the differences of minimum and
maximum values from the mean, respectively. In the comparison of the properties of two FRP mortars whose only
difference is with and without high temperature curing, FRP mortar with high temperature curing (No. 2) was more
lightend than that unheated (No. 1) and its bending strength and compressive strength were raised by factors of 2.5
and 3.5, respectively.
This shows that FRP mortar is lightened and strengthened by heating. As it was considered that compactification of
FRP mortar is necessary for its strengthening, compactifications by pressurization and by the use of fine aggregate
(silica fume) were examined.
In result, considerably lightweight FRP mortar with high strength (No.3) was obtained only by high temperature
curing and pressurization. When the substitution rate of silica fume for cement was raised from 0.1 to 0.3, bulk
density decresed and compressive strength incresed, while deflection showed a tendency to decrease.
4.2.
Strength (FRP mortar with fiber reinforcement)
Table 2 shows the kind of fiber, combination, bulk density, and mechanical strength of FRP mortars with fiber
reinforcement. Bending strength and deflection of FRP mortar reinforced with asbestos (No. 12) were much greater
than those of FRP mortar without reinforcement. Meanwhile, for FRP mortars reinforced with fly ash fiber,
polypropylene, Vinylon, and carbon fiber (No. 8, 9, 10, and 11), mechanical strength decresed, though defflection
increased.
As to FRP mortars reinforced with polypropylene and with Vinylon, when their fractured sections were observed,
ordinally white to orenge fibers had changed to brown. This suggests that the fibers were degraded by heating during
the high temperature curing to be useless for reinforcement.
In case of carbon fiber, its dispersion was not enough to reveal strength. From these results, it is concluded that the
fiber to be used for the reinforcement of FRP mortar is limited to asbestos out of the five kinds used here.

102

KOJIMA AND FURUKAWA

When strengthening by fine-powdered aggregate (silica fume) for the fiber-reinforced FRP mortar was tried, a
lightweight and high-strength FRP mortar (No. 13) was obtained, which had a bulk density of 1.01g/cm3, abending
strength of 7.1MPa, and a compressive strength of 41.0MPa.
4.3.
Change by heatig
SEM observation was carried out on the fractured sections of FRP mortars heat-treated at certain temperatures from
room temperature to 200C. Noticeable difference was not seed the FRP mortars untreated and heat-treated
(Photo. 5).
X-ray diffraction analysis also gave no difference between them.
So at the heating around 200C, the increase of strength can not be considered attributable to any structural change
in the matrix.
On the other hand, when fine powdered FRP was heated solely in air around 200C, formation of cyclic carbonyl
group was recogniazed IR spectrum of alkai-treated FRP fine powder gave absorption peaks by ester bond around
1740cm1, and by C-H bond around 3050cm1. Of these absorption peaks, when heated at 200C, the one by ester
bond around 1740cm1 was recognaized. From these results, it is considered that strengthning of FRP mortar by
heating is due to the change of FRP fine powder contained in it rather than to the thermal change of cement hydrate.
5.
Application of FRP mortars for construction materialas.
The apprication of FRP mortars is various construction materials, such as the sidewalk concrete flags (Photo 6), the
artificial wood of concrete (Photo 6), the centrifugal concrete pipe with luster surface (Photo 7), and the surface
mortar layer on concrete blocksf or retaing wall and revetment, laying blocks (Photo 8).
6.
Conclusion
(1) Mortar containing fine-powdered FRP was lightened and strengthned by high-temeperature curing. Its mechnaical
strength was greatly improved by pressurization to reduce the structural porosity for its compactification. Moreover,
strengthening of FRP mortar was obtained by addition of asbestos and partial replacement of cement by silica fume.
(2) Heating of unmixed fine-powdered FRP around 200C in air produced the formation of cyclic carbonyl group.
Heating of FRP fine powdwe in alkaki gave no remarkable change to its IR spectrum.
These resuls are presumed due to the structual change of FRP fine powder by heating.
(The author is grateful to Professor Hiroshi Maki in Takushoku University and ex-Professor Hideya Kurihara in
Gunma College of Technology for helpful guidance in this study.
And he thanks Mr. Toshihiko Asada in Asaoka Co. Ltd. for suuplying FRP fine powder.)
References
1.

Furukawa S. et al., Extended Abstracts of the 47th Annual Meeting of Japan Cement Association, 254257. (1993).

Table 1 Mechanical strength of FRP Mortars


No.

Press.
(MPa)

Curing Temp.
(C)

SF
(C+SF)

Bulk density
(g/cm3)

Bending strength
(MPa)

Compressive strength
(MPa)

R. T

200

3.9

200

3.9

200

0.1

3.9

200

0.2

1.15
(0.01~+0.05)
1.08
(0.02~+0.03)
0.97
(0.03~+0.06)
1.12
(0.02~+0.03)
0.99
(0.01~+0.01)

1.4
(0.1~+0.2)
3.4
(0.4~+0.2)
5.3
(0.8~+0.4)
6.5
(1.0~+0.5)
6.6
(0.4~+0.3)

4.0
(0.3~+0.5)
13.7
(0.5~+1.4)
29.5
(0.7~+1.2)
24.5
(1.4~+0.3)
31.2
(2.5~+1.6)

LIGHTWEIGHT HIGH-STRENGTH MORTARS

Fig. 1 FRP pulverizer.


(1) automatic feeder, (2)micropowdering machine,
(3) unground Material separatoy, (4)electrostatic remover,
(5) micropowder collector

Fig. 2 The particle distribusion of FRP fine powder.


No.

Press.
(MPa)

Curing Temp.
(C)

SF
(C+SF)

3.9

200

0.3

Bulk density
(g/cm3)

0.98
(0.05~+0.03)
W/C: 0.75, Sirasu/C: 0.28, FRP/C: 0.47, SF: Silica fume

Bending strength
(MPa)

Compressive strength
(MPa)

5.2
(0.5~+0.2)

35.2
(0.2~+0.9)

Table 2 Mechanical strength of fiber reinforced FRP mortars


No.

Fiber

SF /(C+SF)

Bulk density
(g/cm3)

Bending strength
(MPa)

Compressive strength
(MPa)

Fly ash fiber

0.97
(0.03~+0.06)
1.11

5.3
(0.8~+0.4)
5.2

29.5
(0.7~+1.2)
21.2

103

104

KOJIMA AND FURUKAWA

Fig. 3 Preparation procedure of FRP mortar.

Photo 1 The appearance of FRP pulverizer,


No.

Fiber

SF /(C+SF)

Polypropylen

10

Vinylon

11

Carbon fiber

12

Asbestos

13

Asbestos

0.2

Bulk density
(g/cm3)
(0.05~+0.04)
1.29
(0.02~+0.04)
1.11
(0.10~+0.04)
1.14
(0.02~+0.03)
1.17
(0.03~+0.03)
1.01
(0.03~+0.01)

W/C: 0.75, Sirasu/C: 0.28, FRP/C: 0.47, Fiber/C: 0.03, SF: Silica fume

Bending strength
(MPa)
(0.2~+0.3)
4.0
(0.8~+1.5)
3.5
(1.1~+0.1)
4.5
(0.7~+0.3)
6.3
(0.5~+0.2)
7.1
(0.4~+0.2)

Compressive strength
(MPa)
(1.7~+0.8)
15.3
(1.3~+0.4)
15.1
(2.6~+0.3)
21.7
(1.2~+1.1)
27.2
(1.4~+1.9)
41.0
(2.8~+2.0)

LIGHTWEIGHT HIGH-STRENGTH MORTARS

Photo 2 FRP micro fine powder observed by SEM

Photo 3 FRP micro fine powder observed by SEM.

105

106

KOJIMA AND FURUKAWA

Photo 4 Cross section of glass fiber contained in FRP micro fine powder observed by SEM.

LIGHTWEIGHT HIGH-STRENGTH MORTARS

Photo 5 SEM fractographs of FRP mortars heat-treated. (1) as-received, (2) 200C.

107

108

KOJIMA AND FURUKAWA

Photo 6 The sidewalk concrete flags and the artificial wood of concrete containing FRP fine powder.

Photo 7 The centrifugal concrete pipes with luster surface containing FRP fine powder.

LIGHTWEIGHT HIGH-STRENGTH MORTARS

109

Photo 8 The surface mortar layer on concrete blocks for retaing wall and revetment, laying blocks containing FRP fine powder.

12
PROPERTIES OF AUTOCLAVED CEMENT PASTE
CONTAINING SCRAP FRP POWDER
M.WAKASUGI and A.SUGIURA
Central Research Division, Cement/Concrete Research Laboratory, Sumitomo Osaka Cement Co.
Ltd, Osaka, Japan

Abstract
Strength and bulk specific gravity were tested on the autoclaved (160C, 10hr.) cement paste containing scrap
FRP or glass powder. As a result of the tests it was found that the compressive and flexural strength increase with
increase in bulk specific gravity. The compressive and flexural strengths at 5 and 10% of the scrap FRP
powder addition are smaller than those of plain cement paste. However, those at 15% is larger than those at 5
and 10%. Strength of cement paste containing scrap glass powder generally shows a tendency to increase with
increase in its content. From the above test results, the scrap FRP powder is considered to be able to use as a
filler of the autoclaved cement product.
1
Introduction
One of the effective treatment of recycling method for scrap fiber reinforced plastic (FRP) is pulverization. The
properties of the cement mortar containing the pulverized scrap FRP powder, cured in normal condition are reported
in some papers [1], [2], [3].
In this paper we have investigated the properties of the autoclaved cement paste containing the scrap FRP powder
and silica stone powder. Furthermore, we have investigated the effect of scrap glass power addition on the properties
of the autoclaved cement paste because the scrap FRP powder contains glass fiber. Cement paste specimens were
prepared with various silica stone powder/cement ratios and various percentages of the scrap FRP or glass powder
addition, and tested for flexural and compressive strengths. Their bulk specific gravities were also measured.
2
Experimental
2.1
Materials
Ordinary portland cement as specified in JIS R 5210 (Portland Cement), silica stone powder, scrap FRP powder and
scrap glass powder were used, their properties are shown in Table 1 to Table 4, respectively. Particle size distribution
curve for scrap FRP and glass powder are illustrated in Fig. 1.

Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

AUTOCLAVED CEMENT PASTE CONTAINING FRP POWDER

111

Fig. 1 Particle size distribution curve for scrap FRP and glass powder
Table 1. Properties of ordinary portland cement
Specific gravity

Specific surface area

Chemical analysis (%)

ig.loss

insol.

SiO2

Al2O3

Fe2O3

CaO

MgO

SO3

0.8

0.2

21.5

5.2

2.9

63.8

3.15

3310

cm2/g

1.7

1.9

Table 2. Properties of Silicastone powder


Specific gravity

Specific surface area

Chemical analysis (%)

SiO2

Al2O3

Fe2O3

CaO

MgO

SO3

2.64

4330 cm2/g

94.0

2.1

1.2

0.3

0.3

0.0

Table 3. Properties of scrap FRP powder


Supecific gravity

Supecific surface area

Average Particle size

2.02

1.04m2/cc

13.0 m

Table 4. Properties of scrap glass powder


Supecific gravity

Supecific surface area

Average Particle size

2.51

0.28m2/cc

77.3 m

2.2
Preparation of Specimens
According to JIS R 5201 (Physical testing method for cement), cement pastes were prepared with the mix
proportions in Table 5 (series 1) and Table 6 (series 2). Cement ratios to cement plus silica stone powder (cement
content) were 30, 40 and 50%, and water/cement ratios were constant at 50%. The scrap FRP powder was used as a filler,
and the percentages of addition were 5, 10 and 15% in this study.
Cement paste specimens 4040160mm were molded, and subjected to autoclave cure (160C, 10hr.) after 7 days
in 20C, 85%R.H.
Table 5. Mix proportion of cement paste containing FRP
No

Ordinary portland cement

Silica stone powder

Scrap FRP powde

Water

1
2
3
4
5

50
50
50
50
60

50
50
50
50
40

0
5
10
15
0

50
50
50
50
50

112

WAKASUGI AND SUGIURA

No

Ordinary portland cement

Silica stone powder

Scrap FRP powde

Water

6
7
8
9
10
11
12

60
60
60
70
70
70
70

40
40
40
30
30
30
30

5
10
15
0
5
10
15

50
50
50
50
50
50
50

Table 6. Mix proportion of cement paste containing glass


No

Ordinary portland cement

Silica stone powder

Scrap glass powder

Water

1
2
3
4
5
6
7
8
9
10
11
12

50
50
50
50
60
60
60
60
70
70
70
70

50
50
50
50
40
40
40
40
30
30
30
30

0
10
20
30
0
10
20
30
0
10
20
30

50
50
50
50
50
50
50
50
50
50
50
50

2.3
Flexural and compressive strength
According to JIS R 5201, the autoclaved cement paste specimens were tested for flexural and compressive strengths,
and their bulk specific gravity were also measured.
3
Results and discussion
3.1
Series 1 (effect of scrap FRP powder)
Fig.2 shows the effect of addition of scrap FRP powder on the compressive and flexural strengths. Although both
compressive and flexural strengths at 5 to 10% of scrap FRP powder addition shows a tendency to be smaller than
those of plain paste, those at 15% is larger than those at 5 to 10%. When the cement content is 50%, the largest
compressive and flexural strengths are gained.
Fig. 3 shows the relationship between percentage of scrap FRP powder addition and bulk specific gravity. It shows
a similar tendency to the above-mentioned relationship in Fig. 2. There exists an obvious quantitative relationship
between bulk specific gravity and strengths as shown in Fig. 4.
3.2
Series 2 (effect of scrap glass powder)
Fig. 5 shows the effect of scrap glass powder addition on the compressive and flexural strengths after autoclave
curing. In general, both compressive and flexural strengths shows a tendency to increase with an increase in
percentage of scrap glass powder, except for one case. When the cement content is 50%, flexural strength decreases
with an increase in scrap glass powder addition from 0 to 10%. Because E-glass fiber is easily deteriorated due to
high alkalinity.

AUTOCLAVED CEMENT PASTE CONTAINING FRP POWDER

113

Fig. 2 The effect of scrap FRP powder addition on strengths

Fig. 3 The relationship between scrap FRP powder addition and bulk specific gravity

Fig. 6 shows the relationship between percentage of scrap glass powder addition and bulk specific gravity. There
exists more obvious quantitative relation between them than that of scrap FRP powder.
Furthermore, above-mentioned similar relationship between bulk specific gravity and strengths is seen as shown in
Fig. 7.
4
Conclusion
The conclusion obtained from the test results on the properties of autoclaved cement paste containing scrap FRP or
glass powder are summarized as follows.
Compressive and flexural strengths of the autoclaved cement paste increase with increase in bulk specific gravity.
Strengths at 5 and 10% of scrap FRP powder addition are smaller than those of plain cement paste. However, those
at 15% is larger than those at 5 and 10%. Strength of cement paste containing scrap glass powder generally shows a
tendency to increase with increase in its content. From the above test results, the scrap FRP powder is considered to
be able to use as a filler of the autoclaved cement product.
5
Reference
1

[] Y.Ohama, T.Demura, T.Sato, Water Resistance and Hot Water Resistance of Polymer- Modified Mortars Containing
FRP Powder Summaries of Technical Papers of Annual Meeting Architectural Institute of Japan, Sept. 1994, pp. 137138

114

WAKASUGI AND SUGIURA

Fig. 4 The relationship between bulk specific gravity and strength

Fig. 5 The effect of addition of scrap glass powder on strengths


2
[] S.Furukawa, A,Kojima and T.Asada, Mechanical Properties of Light-Weight Mortar with Powder of Scrap FRP (in
Japan) JCA Proceedings of Cement and Concrete, V. 47, Oct. 1993, pp. 244247
3
[] H.Miyairi, Utilization of Waste FRP as Resources (in Japan) Bulletin of the Chemical Society of Japan, Chemistry and
Chemical Industry, V. 43, No. 11, Nov. 1992, pp. 2432

AUTOCLAVED CEMENT PASTE CONTAINING FRP POWDER

Fig. 6 The relationship between scrap glass powder addition and bulk specific gravity

Fig. 7 The relationship between bulk specific gravity and strength

115

13
RECYCLING OF FRP AS A CEMENTITIOUS COMPOSITE
K.YAMADA
Chemicals Research Laboratories, Nippon Steel Co., Kawasaki, Japan
H.MIHASHI
Department of Architecture, Faculty of Engineering, Tohoku University, Sendai, Japan

Abstract
FRP is now enormously used and recycling as cementitious materials is paid attention. The purpose of this
paper is revealing fundamental flexural properties of cementitious composites including FRP powder.
Three composites including 9.1% (PA0), 13.3% (PA1) and 22.8% (PA3) of FRP powder to all powdery
materials were extruded to rectangular shape (15 mm40 mm) specimens. Furthermore two composites
including 9.1% of A3 silica sand (PA0) and no aggregates (PA) were extruded as the references.
Extrudability, bulk density, flexural strength, modulus of elasticity, absorbed energy of specimens were
examined. After bending tests, fractured surface was observed by SEM and porosity was measured.
Autoclave-cured specimens including FRP powder do not show superiority in flexural strength while watercured specimens including 9.1% of FRP powder indicate excellent strength.
Keywords: asbestos-free mortar, autoclave curing, extrude, flexural strength, FRP powder, porosity, water
curing
1
Introduction
Fine powder of waste FRP is now mainly used for filler material as recycling to the same products [1]. Some
applications such as 5% of recycled filler for rear spoiler of cars [1] are reported. However rapid increasing of waste
FRP (see Table 1) will cause much trouble if recycling of waste FRP for wide applications is not considered.
To consume vast amount of waste FRP, application for building materials is needed.
Table 1 Rapid increase of waste FRP in Japan [4]
Item/Year
Shipment of FRP
Predicted waste:
Total (104KN)
Details

(104

KN)

Construction
Housing
Ships
Cars
Tanks & Containers
Miscellaneous goods

1990

1995

2000

445.3

504.1

584.4

141.5
10.4
70.5
6.8
4.6
10.8
38.4

232.0
22.7
83.2
23.8
15.9
27.8
58.7

370.4
23.9
143.2
41.9
26.2
33.0
102.2

Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

RECYCLING OF FRP AS A CEMENTITIOUS COMPOSITE

117

Fig. 1 DTA and TG of FRP powder

The purpose of this paper is developing compositions of asbestos-free mortar to be used for extruded building
materials that are now common and widely used.
Some studies have been already conducted on low density mortar including FRP fine powder. These studies deal
only with cast mortar of low flexural strength below 10 MPa cured in water or high temperature air [2] or humid air
[3]. No studies have been reported about extruded composites and properties of autoclave-cured specimens.
The study presented in this paper deals with flexural strength of extruded asbestos-free mortar including waste
FRP fine powder.
2
Test program
2.1
Materials
FRP fine powder was made from waste septic tanks, which were formed by Sheet Molding Compound (SMC)
method. The SMC was a composite of resin, inorganic filler and glass fiber (GF). The resin was unsaturatedpolyester resin. The inorganic filler was fine powder of calcium carbonate. These two materials compose 70% of the
weight and the rest of the weight (30%) was GF.
Chemical composition of FRP powder is shown in Table 2. DTA (differential thermal analysis) and TG
(thermogravimetric analysis) versus temperature is shown in Fig. 1, which indicate combustion of resin at 350450
degree Cels.
Powdered FRP was manufactured by a machine that have rotating drums with embedded many diamonds. FRP
tanks were shredded and powdered to become about 30 micrometers of the average diameter.
Table 2 Chemical compositions of FRP powder (wt %)
C

Ca

Si

Al

Mg

Na

Ti

Pb

32.2

36.4

2.44

21.4

5.0

1.53

0.64

0.072

<0.01

<0.01

Table 3 Density of FRP powder


Tapped bulk density
(gr/cc)

Apparent density
(gr/cc)

Average grain size


(micrometer)

0.39

1.93

30.5

Distribution of grain size is shown in Fig. 2 indicating the profile looks mostly like A3 silica sand. Main features of
FRP powder are shown in Table 3. SEM photographs of FRP powder and A3 silica sand are shown in Fig. 3 and
Fig. 4 (see next page). They look similar each other except that FRP powder includes many broken fibers (50 500
micrometers) of GF.

118

YAMADA AND MIHASHI

Fig. 2 Distribution of grain size

2.2
Fabrication of specimens
The five compositions examined are given in Table 4. All of the compositions are determined based on the total
powdery materials weight of cement, silica powder, silica sand and powdered FRP. Weight of pulp, methyl cellulose
(MC) and water of all compositions are 2%, 1.4% and 26.1% to all powdery materials respectively. Weight of cement
of all compositions is 47.3% to all powdery materials. PA composite does not contain any other material. On the
other hand PA0 composite contains 9.1 % of silica sand. PA1, PA2 and PA3 composites contain 9.1%, 13.3% and 22.
8% of FRP powder, respectively.
Table 4 Mix proportion (gr/liter and [%] to all powdery materials)
Powdery materials [100] 1001
Name

Cement

Silica Powder

PA

742
[47.3]
742
[47.3]
728
[47.3]
721
[47.3]
707
[47.3]

827
[52.7]
684
[43.6]
670
[43.6]
600
[39.4]
446
[29.9]

PA0
PA1
PA2
PA3

Silica Sand (A3)

FRP

Pulp

MC

Water

141
[9.1]
203
[13.3]
341
[22.8]

32.1
[2.0]
32.1
[2.0]
31.5
[2.0]
31.2
[2.0]
30.6
[2.0]

21.8
[1.4]
21.8
[1.4]
21.4
[1.4]
21.2
[1.4]
20.8
[1.4]

410
[26.1]
410
[26.1]
402
[26.1]
398
[26.1]
390
[26.1]

143
[9.1]

All materials were mixed and kneaded in kneader (Mikami Machinery Co., MND-10) for 3 minutes of dry mixing
and 8 minutes of wet kneading. After kneaded, fresh mortar was measured its hardness by a rubber hardness tester. Then
the mortar was extruded immediately.
The extruder used to form specimens has two pugging screws, a depressurized room and one auger screw whose
diameter is 75 mm (Honda Machinery Co., HDE-3D). The die used to form specimens has a rectangular hole whose
dimension was 15 mm thick and 40 mm wide (see Fig. 5 and Fig. 6).
During the extrusion of specimens, speed of the extruded specimen, current of the auger screw motor, pressure at
the die and pressure at the depressurized room were measured. After extruded, the hardnesses of formed specimens
were measured.
2.3
Curing
The formed specimens were placed for 24 hours in the curing room whose temperature is 20 degree Cels. and 60%
RH, and then cut into 175 mm length.

RECYCLING OF FRP AS A CEMENTITIOUS COMPOSITE

119

Fig. 3 SEM photographs of FRP powder

In case of autoclave curing, the specimens were cured in the condition of 150 degree Cels. steam for 8 hours. For
getting this condition, it took 5 hours for ascending temperature and 12 hours for cooling. In case of water curing, the
specimens were cured in water of 20 degree Cels. for 28 days.
Before bending tests, all specimens were ground to make the surfaces smooth and to have the same thickness. All
specimens were dried at 105 degree Cels. for 48 hours. Water absorption ratio was measured by the difference of
weight before and after drying only in the case water curing.
2.4
Test setup
The specimens were tested in the condition whose bending span is 160 mm and center loaded at the rate of 0.5 mm/min.
Numbers of specimens were three for every condition. During bending test, load and deflection at the center of
specimens were continuously measured.
After the bending test, some specimens were sliced to investigate fractured surfaces by SEM (scanning electron
microscope).
3
Test results
3.1
Extrudability
All the surfaces of extruded specimens were smooth and had no cracks. It indicates all the composites were
extrudable.
Table 5 shows extrudability of all composites. Base composition (PA) consisting of cement and powdered silica as
the powdery material was most easily extrudable, because it indicated higher extrude speed, lower pressure and
lower load applied to the motor. On the other hand, the composite showed the lowest hardness. PA0 and PA1 showed

120

YAMADA AND MIHASHI

Fig. 4 SEM photographs of A3 silica sand

Fig. 5 Profile of extruder and pressure along material flowExtruding specimen

practically the same extrudability. From PA1 to PA3 the extrudability decreased according with the increase of FRP
powder content. PA3 showed the highest hardness before and after extrude forming.
Table 5 Features of extrudability
Item/Compositions

PA

PA0

PA1

PA2

PA3

extrude speed (cm/min.)


current of auger screw motor (A)
pressure
at die (MPa)
at depressurized room (103 Pa)
hardness
of kneaded compositions (-)
of formed specimens (-)

55
4.0

50
4.2

50
4.2

45
5.3

40
6.2

0.32
2.67

0.35
2.65

0.36
2.66

0.54
2.65

0.72
2.67

15
30

18
40

19
37

25
45

30
52

RECYCLING OF FRP AS A CEMENTITIOUS COMPOSITE

121

Fig. 6 Extruding specimen (15mm40mm)

3.2
Properties of composite
3.2.1
Mechanical properties
In all bending tests, the fracture behavior was brittle.
Absorbed energy was calculated from the area under the load-deflection curves. Flexural modulus of elasticity was
calculated from the load-deflection relation between 0 and the maximum load. Average properties of all composites
are shown in Table 6.
3.2.2
Porosity
After bending tests, the specimens that had medium strength were cut and the porosity was measured by porosity
meter (Micromeritics Co., Autopore 29220). The results were shown in Table 7 and Fig. 7.
Intrusion volume of all autoclave-cured composites was bigger than that of water-cured except PA0. It is
noticeable that the pore diameters of autoclaved composites exceed in diameters of 1 micrometer and 0.01
micrometer while water-cured ones exceed in 0.1 micrometer. Among all cases, water-cured PA1 shows the
minimum intrusion volume, which performed the maximum flexural strength.
Table 6 Average properties of all compositions
Composition

Cure

Bulk density
(gr/cc)

Flexural strength
(MPa)

Absorbed energy
(J)

Flexural modulus
(GPa)

PA

AC150
W28
AC150
W28
AC150
W28
AC150
W28
AC150
W28

1.763
1.811
1.724
1.791
1.722
1.766
1.716
1.699
1.647
1.650

27.92
17.93
25.67
19.49
19.27
22.05
18.73
17.05
17.52
16.04

2.03
1.14
1.51
1.14
1.22
1.58
1.04
0.99
1.02
1.23

16.64
15.28
16.27
14.98
13.35
15.45
12.61
13.59
10.66
10.56

PA0
PA1
PA2
PA3

Water absorption ratio


(%)
13.51
13.87
14.26
15.25
15.65

Average pore diameter of all composites that contain FRP powder increased in accordance with the FRP content.
The ratio of pore surface of autoclave-cured specimens to water-cured ones decreased in accordance with FRP
content ratio, which may indicate emergence of coarse hydration products or generation of bigger pores in case of
autoclave-curing. (See Table 7)

122

YAMADA AND MIHASHI

Fig. 7 Pore distribution of all compositions


Table 7 Porosity of composites
Mix proportion Cure

Intrusion volume Pore surface Area Area ratio


Average pore diameter
(cc/gr)
(m2/gr)
(AC150/ W28) (micrometer)

PA

AC150 0.136
35.795
W28
0.128
21.156
1.692
PA0
AC150 0.147
42.306
W28
0.152
31.039
1.363
PA1
AC150 0.156
20.868
W28
0.117
24.965
0.836
PA2
AC150 0.176
27.988
W28
0.164
33.888
0.826
PA3
AC150 0.168
18.745
W28
0.148
27.574
0.680
0.0214
Note: AC150 means autoclave-cured. W28 means water-cured for 28 days.

0.0151
0.0243
0.0139
0.0193
0.0299
0.0187
0.0252
0.0194
0.0359
2.213

Apparent density
(gr/cc)
2.353
2.382
2.371
2.483
2.308
2.282
2.425
2.357
2.272

3.2.3
SEM observations
Typical SEM photographs of fractured surface after bending tests are shown in Fig. 8 and Fig. 9. Shown in Fig. 8 is
the surface of autoclave-cured PA3 composite. Fig. 9 shows the surface of water-cured PA3 composite. In the case
of water-cured composites, there remained many sound GF compared with autoclave-cured composites that mostly
contain degraded GF-like vague materials.

RECYCLING OF FRP AS A CEMENTITIOUS COMPOSITE

123

Fig. 8 Fractured surface of PA3 (Autoclave cured)

Fig. 10 shows a fractured surface of autoclave-cured PA0 composite and Fig. 11 shows one of autoclave-cured PA1.
There appears smooth surface of A3 silica sand or FRP powder, which may suggest the same effect to multiple cracks
in the early stage of bending.

124

YAMADA AND MIHASHI

Fig. 9 Fractured surface of PA3 (Water-cured)

Fig. 10 Fractured surface of Autoclave cured PA0

Fig. 11 Fractured surface of Autoclave cured PA1

4
Discussion
4.1
Flexural strength
Bulk density (BD) of specimens of all composites is shown in Fig. 12. BD of all composites decreased in accordance
with FRP content. It might be caused because a part of cement was replaced by FRP powder in the mix proportions
used in this study.
The BD of autoclave-cured and water-cured composites is mostly equal and showed the same decreasing feature in
accordance with FRP content.
Flexural strength of autoclave-cured specimens shows a drastic drop by including FRP powder (see Fig. 13). PA
that contains only fine powder as pozzolanic materials shows the highest flexural strength. PA0 that contains silica
sand locates the next.
On the other hand, the order of strength in the case of water-cured specimens differs from that of autoclave-cured
specimens. PA1 shows highest flexural strength. The increase of FRP content means decrease of flexural strength in
this case, too. However, PA and PA0 that contain no FRP powder indicate lower strength.

RECYCLING OF FRP AS A CEMENTITIOUS COMPOSITE

125

Fig. 12 BD vs. FRP content

Fig. 13 Flexural strength vs. BD

Flexural strength of all water-cured specimens showed lower strength than that of autoclave-cuned ones except
PA1. This exception may be caused by the reinforcing effect of GF, only when the content of FRP powder is
comparatively low.
4.2
Merits and demerits of including FRP
4.2.1
Outline
Main chemical component of FRP powder is inorganic oxide such as SiO2 and CaCO3 derived from GF and filler. There
exists no particular difficulties to use FRP powder for the mix proportion in the range shown in Table 4 from the
standpoint of flexural strength.
The grain size of FRP powder is similar to A3 silica sand, which shows no difficulty in extruding. However
extrudability decreases in accordance with FRP content.
4.2.2
Merits and demerits
FRP powder inevitably contains GF. The curing condition that does not degrade GF can bring high strength as
observed water-cured PA1, though durability problem should be considered.
Moreover the smooth surface of FRP powder can bring small dispersed notches into brittle cementitious material
that may produce slightly high strength. However, the effect was not confirmed in this work.
In spite of low tapped bulk density, apparent density of FRP powder is not low compared with silica sand. Therefore
FRP powder itself can not make BD of hydrated composite much lighter.

126

YAMADA AND MIHASHI

FRP powder decreases flexural strength and modulus of elasticity but increases water absorption ratio in all cases.
It may lead to durability problems.
Autoclave-cure did not contribute any significant improvements to the flexural strength because of degradation of
GF and FRP. However, it may lead to least longterm additional degradation.
5
Conclusions
Cementitious composites including FRP powder that was mainly composed of unsaturated-polyester resin, CaCO3
(filler of FRP) and GF (reinforcement of FRP) were investigated. The grain size of FRP powder is similar to A3
silica sand. These suggest FRP powder to use as a substitute of A3 silica sand to extrudable composite.
Flexural strength of extruded specimens of five compositions was examined. The results are as follows:
1 If specimens are cured in the autoclave, the strength of composites including FRP powder drastically dropped to
the level of water-cured specimens.
2 If specimens are cured in water, the composite including 9.1 % of FRP powder showed the maximum strength
among all composites. It was suggested that the result comes from reinforce effect of GF in FRP powder.
3 In all composites investigated in this work, increasing the volume content of FRP powder decreased flexural
strength, flexural modulus of elasticity and increased water absorption ratio.
6
References
1
2
3
4

Fukuda, N. (1994) Utilization technique of powdered FRP: Recycling and handling of FRP waste material, Reinforced
Plastics, Vol. 40, No. 8, pp. 315319.
Furukawa, S., Kojima, A. and Asada, T. (1991) Manufacture of light weight mortar using scrap FRP fine powder, CAJ
Proceedings of Cement & Concrete, No. 45, pp. 310315.
Fujiwara, M., Kohno, T. and Hamada, H. (1994) Utilization of waste FRP for concrete material, Cement & Concrete, No.
563, pp. 6468.
Koyanagi, T. (1994) Prediction of amount of waste FRP based on survey of real state of FRP industry: Recycling and
handling of FRP waste material, Reinforced Plastics, Vol. 40, No. 8, pp. 296302.

14
PROPERTIES OF ARTIFICIAL WOODS USING FRP POWDER
K.DEMURA, Y.OHAMA and T.SATOH
Department of Architecture, College of Engineering, Nihon University, Koriyama, Fukushimaken, Japan

Abstract
Recently, the importance of the countermeasures for waste fiber-reinforced plastics (FRP) products has been
pointed out, because large quantities of waste FRP products have been discarded. For the purpose of recycling
waste FRP products, the present paper deals with the properties of artificial woods (wood-like materials) which
are experimentally manufactured by using the FRP powder made by pulverization of waste FRP products.
Artificial woods using FRP powder are prepared by autoclaving various cementitious compositions consisting
of high-early-strength portland cement, silica fume, FRP powder, shirasu balloon, chemical admixtures, carbon
fibers and water, and tested for bulk specific gravity, hardness, flexural and compressive strengths and nail
withdrawal. As a result, the optimum mix proportions of the artificial woods are proposed. The bulk specific
gravity of an artificial wood with the optimum mix proportions is larger than that of calcium silicate-based
wood-like materials, and their flexural and compressive strengths are higher than those of the calcium silicatebased materials. In addition, they can be nailed and sawed like natural wood.
Keywords: artificial woods, bulk specific gravity, carbon fibers, FRP powder, hardness, nail withdrawal,
strengths
1
Introduction
Fiber-reinforced plastics (FRP) products using unsaturated polyester resin have widely been used in various industrial
applications because of high specific strength, corrosion resistance and excellent moldability. The recent annual
production of FRP products has reached about 450,000 tons in Japan [1]. Although the waste FRP products have
been reclaimed as industrial wastes till now, effective recycling methods for them should be examined from the
viewpoints of environmental protection and resources exploitation in the near future. A few investigations on the
manufacture and properties of cementitious composites with FRP powder made of waste FRP products have already
been reported [2, 3]. In the present paper, artificial woods using the FRP powder made by pulverizing waste FRP
products are prepared by autoclaving cementitious compositions with various carbon fiber, hydroxy propyl methyl
cellulose and shirasu balloon contents, and tested for bulk specific gravity, hardness, flexural and compressive strengths
and wood-processability. From the test results, their physical properties are discussed compared to those of calcium
silicate-based wood-like materials [4].

Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

128

DEMURA, OHAMA AND SATOH

2
Materials
2.1
Cement
High-early-strength portland cement as specified in JIS R 5210 (Portland Cement) was used in all the mortar mixes
for artificial woods.
2.2
Admixtures
Silica fume which is a by-product in the manufacture of ferrosilicon was used as a dispersant for carbon fibers in all
the mixes. The chemical and physical properties of the silica fume is listed in Table 1. Polycarboxylie acid-type
superplasticizer and hydroxy propyl methyl cellulose (HPMC, viscosity: 35005600mPas at 20C as 2% aqueous
solution) were employed as chemical admixtures.
2.3
FRP powder
FRP powder was made by pulverizing the waste trimming of bathtubs. The properties of the FRP powder are listed in
Table 2, and particle size distribution curve for it is illustrated in Figure 1. Figure 2 shows the scanning electron
micrograph of the FRP powder.
2.4
Lightweight aggregate
Shirasu balloon (bulk specific gravity: 0.40, particle size: 30100m) was used for a lightweight aggregate.
2.5
Carbon fibers
Commercially available pitch-based carbon fibers were used for reinforcement, whose physical properties are listed
in Table 3.
Table 1 Chemical compositions and physical properties of silica fume
Chemical compositions (%)
SiO2

Al2O3

82.5
0.19
Physical properties
Specific gravity
(20C)
2.31

Fe2O3

CaO

MgO

Na2O

K2O

ig.loss

7.0

0.67

0.82

1.18

0.24

2.16

3.22

Average particle size


(m)
0.2

Specific surface
(m2/g)
20

Average particle size (m)

Specific gravity (20C)

Chemical compositions (%)

Unsaturated polyester resin

Glass fibers

Ground calcium carbonate

12.4

1.87

30

Table 2 Properties of FRP powder

25

40

ARTIFICIAL WOODS USING FRP POWDER

129

Fig. 1 Particle size distribution curve for FRP powder.

(400)
Fig. 2 Scanning electron micrograph of FRP powder.
Table 3 Physical properties of carbon fibers
Type of carbon fibers

Average diameter
(m)

Average length
(mm)

Specific gravity
(20C)

Elastic modulus in tension


(GPa)

Tensile strength
(MPa)

Pitch-based

18.0

1.65

30

590

3
Testing procedures
3.1
Preparation of specimens
According to JIS R 5201 (Physical Testing Methods for Cement), mortars for artificial woods were prepared with the
mix proportions given in Table 4 by using an ordinary mortar mixer, and their flows were adjusted to be constant at
1705. Specimens 4040160mm were molded, and then precured at 20C and 80%R.H. for 24 hours. Immediately
after precuring, the specimens were cured in autoclave in which a maximum temperature of 180C was maintained
under a pressure of 1MPa for 3 hours.
Table 4 Mix proportions of mortars for artificial woods
Mix no. Cement : silica
fume : FRP
powder
(by mass)
1

Shirasu
balloon
content
(wt%*)

HPMC content Super(wt%*)


plasticizer
content
(wt%*)

Carbon fiber
content
(vol%)

Water-(cement Air content Flow


+ silica fume) (%)
ratio (%)

72.0

5.5

170

130

DEMURA, OHAMA AND SATOH

Mix no. Cement : silica


fume : FRP
powder
(by mass)

Shirasu
balloon
content
(wt%*)

2
0
3
0.4
4
5
14
6
40:10:50
2.0
7
8
0
9
0.8
10
11
14
12
Notes, *: of cement+silica fume
**: unable to mix at flows of 1705

HPMC content Super(wt%*)


plasticizer
content
(wt%*)
4
0
4
0

114.0
92.0
4
0

Carbon fiber
content
(vol%)

Water-(cement Air content Flow


+ silica fume) (%)
ratio (%)

2
92.0
91.0
2
12.4
7.8
2
111.0
110.0
2
4

84.0
11.3
6.1
100.0
166
190**
102.0
12.8
7.7
122.0
128.0

8.8
171
168
8.4

8.3
171
172
9.9
11.3

174

170

168

174
167

3.2
Flexural and compressive strength tests
The cured specimens were tested for flexural strength by center-point loading at a crosshead speed of 0.5mm/min by
using the Instron universal machine. Their midspan deflection was measured by using sensitive linear variable
differential transformer, and their flexural toughness was calculated in accordance with JCI (Japan Concrete
Institute) SF-4 (Method of Tests for Flexural Strength and Flexural Toughness of Fiber Reinforced Concrete). The
portions of the specimens broken in flexural test were tested for compressive strength according to JIS R 5201.
3.3
Measurement of bulk specific gravity
The pieces 404050mm which were cut from the portions of the specimens broken in flexural test were tested for
bulk specific gravity in absolute dry condition according to JIS A 5416[Autoclaved Lightweight Concrete Panels
(ALC Panels)], Section 7.2.
3.4
Hardness test
According to JIS Z 2101(Methods of Test for Wood), the cured specimens were tested for hardness.
3.5
Wood-processability evaluation
The cured specimens were firstly checked up on the possibility of nail driving by using a nail of 2.45mm (in
diameter)45mm (in length). The specimens into which the nail could be driven were tested for nail withdrawal
according to JIS Z 2101. Furthermore, the specimens were evaluated for saw-ability by using a hand saw.
4
Test results and discussion
Figure 3 shows the carbon fiber content vs. water-(cement+silica fume) ratio of fresh artificial woods with flows of
1705. As seen in the figure, the water requirement for the given consistency of the fresh artificial woods increases
with an increase in the carbon fiber content regardless of the HPMC content and shirasu balloon content. The water
requirement increases slightly with raising HPMC content from 0.4 to 0.8wt% irrespective of the carbon fiber content
and shirasu balloon content. The inclusion of 14wt% shirasu balloon in the artificial woods causes an increase in the
water requirement.

ARTIFICIAL WOODS USING FRP POWDER

131

Fig. 3 Carbon fiber content vs. water (cement+silica fume) ratio of artificial woods.

Fig. 4 Carbon fiber content vs. bulk specific gravity of artificial woods.

Figure 4 exhibits the carbon fiber content vs. bulk specific gravity of artificial woods. The bulk specific gravity of
the artificial woods decreases with increasing carbon fiber content, shirasu balloon content and HPMC content. This
tendency corresponds to the relationship between the compressive strength and water(cement+silica fume) ratio as
stated below. The bulk specific gravity of the artificial woods with a shirasu balloon content of 14wt%, an HPMC
content of 0.8% and carbon fiber contents of 2 to 4vol% is under 1.0, which is about twice that of the calcium
silicate-based wood-like materials [4].
Figure 5 represents the flexural load-deflection curves for artificial woods. In general, the maximum flexural load
of the artificial woods tends to increase with an increase in carbon fiber content, and the deflection at the maximum
flexural load increases with increasing shirasu balloon content and HPMC content. A drop in the post-maximum
flexural load shows a more ductile behavior with raising HPMC content from 0.4 to 0.8wt%. From the abovementioned results, the flexural deformation behavior of the artificial woods is markedly improved by using carbon
fibers, HPMC and shirasu balloon.
Figure 6 indicates the carbon fiber content vs. flexural toughness of artificial woods. The flexural toughness of the
artificial woods without shirasu balloon increases with an increase in the carbon fiber content. By contrast, the
flexural toughness of the artificial woods with a shirasu balloon content of 14wt% increases with an increase in the
carbon fiber content, and becomes nearly constant or reaches a maximum at a carbon fiber content of 2vol%,
depending on HPMC content. In general, the flexural toughness of the artificial woods with carbon fiber contents of
2vol% or more attains to 10 or more times that of the artificial woods without carbon fibers. In the case of a carbon
fiber content of 2vol%, the flexural toughness of the artificial woods with a shirasu balloon content of 14wt% is
higher than that of the artificial woods without the shirasu balloon irrespective of the HPMC content
Figure 7 illustrates the carbon fiber content vs. flexural strength of artificial woods. The flexural strength of the
artificial woods without shirasu balloon increases with an increase in the carbon fiber content The flexural strength
of the artificial woods with a shirasu balloon content of 14wt% becomes nearly constant at a carbon fiber content of
2vol% when the carbon fiber content rises from 2 to 4vol%. Regardless of HPMC content, the flexural strength of
the artificial woods with carbon fiber contents of 2 to 4vol% attains to 11.0 MPa or higher, which is higher than that
of the calcium silicate-based wood-like materials [4].

132

DEMURA, OHAMA AND SATOH

Fig. 5 Flexural load-deflection curves for artificial woods.

Fig. 6 Carbon fiber content vs. flexural toughness of artificial woods.

Figure 8 exhibits the carbon fiber content vs. compressive strength of artificial woods. The compressive strength
of the artificial woods decreases with increasing in the carbon fiber content, HPMC content and shirasu balloon
content. Such compressive strength decrease may be explained by increases in both water-(cement+silica fume) ratio
and voids in the artificial woods according to the water-cement ratio theory and voids theory, and is expressed by the
following empirical equation:
(1) Equation by the water-cement ratio theory

where Fc: the compressive strength of the artificial woods


Mw: the mass of water per unit volume of the artificial woods
Mb: the mass of cement plus silica fume per unit volume of the artificial woods
A and B: empirical constants
(2) Equation by the voids theory

ARTIFICIAL WOODS USING FRP POWDER

133

Fig. 7 Carbon fiber content vs. flexural strength of artificial woods.

Fig. 8 Carbon fiber content vs. compressive strength of artificial woods.

Fig. 9 Water- (cement+silica fume) ratio vs. compressive strength of artificial woods.

where Fc: the compressive strength of the artificial woods


Vw: the volume of water per unit volume of the artificial woods
Va: the volume of air per unit volume of the artificial woods
Vb: the volume of cement plus silica fume per unit volume of the artificial woods
and : empirical constants
The relationships between the compressive strength and water-cement ratio or voids of the artificial woods by both
theories are represented in Figures 9 and 10.
Figure 11 illustrates the carbon fiber content vs. hardness of artificial woods. The hardness of the artificial woods
decreases with increases in the carbon fiber content, HPMC content and shirasu balloon content.
In Table 5, the wood-processability of artificial woods is evaluated in comparison with natural wood. Like natural
wood, all the artificial woods can be sawed easily. The artificial woods without carbon fibers are cracked when nails
are driven into them. On the other hand, when the nails are applied in the artificial woods with the carbon fibers,

134

DEMURA, OHAMA AND SATOH

Fig. 10 (Vw+Va)/Vb vs. compressive strength of artificial woods.

Fig. 11 Carbon fiber content vs. hardness of artificial woods.

there are two cases: the nails can be easily driven into them without cracking or bent halfway. This is attributed to
the deformability and energy-absorbing property of the artificial woods which are affected by the carbon fiber
reinforcement and voids. Because the suitable voids absorb the breaking energy while nailing, and the carbon fiber
reinforcement controls to crack the cementitious matrix. The mix proportions of the artificial woods which can be
nailed are recommended as follows:
Cement:silica fume: FRP powder=40:10:50 (by mass)
Shirasu balloon content (wt% of cement+silica fume): 14
HPMC content (wt% of cement+silica fume): 0.8
Superplasticizer content (wt% of cement+silica fume): 2.0
Carbon fiber content (vol%): 2 to 4
The nail withdrawal of the above artificial woods is over 25N/mm, which is about twice higher than that of Japanese
ceder.
Table 5 Wood-processability of artificial woods
Mix no.

Saw-ability

Nail-ability

Nail withdrawal (N/mm)

Japanese Ceder
1
2
3
4
5
6
7
8

Good
Good
Good
Good
Good
Good
Good
Good
Good

Good
Unable to nail by cracks
Unable to nail by the bend of the nail
Unable to nail by the bend of the nail
Unable to nail by cracks
Unable to nail by the bend of the nail
Unable to nail by the bend of the nail
Unable to nail by cracks
Unable to nail by the bend of the nail

12.7*

ARTIFICIAL WOODS USING FRP POWDER

Mix no.

Saw-ability

Nail-ability

Nail withdrawal (N/mm)

9
10
11
12
Note, *: Flat grain .

Good
Good
Good
Good

Unable to nail by the bend of the nail


Unable to nail by cracks
Good
Good

28.8
30.5

135

5
Conclusions
(1) The flexural strength of artificial woods is greatly improved by the addidon of carbon fibers. The flexural
deformation behavior of the artificial woods is markedly improved by using carbon fibers, HPMC and shirasu
balloon. The bulk specific gravity, hardness and compressive strength of the artificial woods decrease with
increasing carbon fiber content, HPMC content and shirasu balloon content.
(2) The mix proportions of artificial woods which can be nailed are recommended in 4. From an economical
viewpoint, the carbon fiber content in the mix proportions can be reduced to 2 vol%. The optimum mix proportions
with a carbon fiber content of 2 vol% is given in Table 7. The bulk specific gravity, flexural and compressive
strengths of an artificial wood with the optimum mix proportions are 0.9, 12.0MPa and 19.0MPa respectively. The
artificial wood also has good wood-processability like natural wood.
(3) The bulk specific gravity of an artificial wood with the above optimum mix proportions is larger than that of
calcium silicate-based wood-like materials, and their flexural and compressive strength are higher than those of the
calcium silicate-based materials.
Table 6 Optimum mix proportions of artificial woods
Mix proportions (kg/m3)
Cement

Silica fume

FRP powder

Shirasu balloon

HPMC

Super-plasticizer

Carbon fibers

Water

266

67

333

47

2.7

13.3

30

406

References
1.
2.
3.

4.

Miyairi, H. (1992) Utilization of Waste FRP as Resources (in Japanese). Chemistry and Chemical Industry, Vol. 43, No.
11. pp. 2432.
Furukawa, S., Kojima, A. and Asada, T. (1993) Mechanical Properties of Light-Weight Mortar with Scrap FRP (in
Japanese). JCA Proceedings of Cement & Concrete, Vol. 47. pp. 244247.
Ohama, Y., Demura, T. and Satoh, T. (1994) Basic Properties of Polymer-Modified Motrars Containing FRP Powder Made
by Pulverization of Waste FRP Products. Journal of the College of Engineering of Nihon University, Series A, Vol. 35.
pp. 3340.
Watanebe, T., Tsukiyama, K. and Kawano, T. (1988) Study on Xonotlite-SBR Latex Composite Material. Proceeding of
the MRS International Meeting on Advanced materials, Vol. 13. pp. 115123.

PART THREE
RUBBER-BASED MATERIALS

15
NOISE ABATEMENT BY PANELS OF RECYCLED VEHICLE
TIRES
M.MOTAVALLI, M.FARSHAD and P.FLELER
Eidgenssische Materialprfungs-und Forschungsanstalt, Dbendorf, Switzerland

Abstract
This contribution contains results of an experimental and theoretical investigation on recycled panels made of
recycled rubber of vehicle tires. These panels are the elements of vertical walls which can be installed along
the highways and are intended to reduce the traffic noise; they were not meant to be used as retaining walls.
The material investigation included tensile, compression, and bending tests on the recycled rubber material. To
simulate the effect of wind action, an inflated rubber membrane balloon with controlled internal pressure was
used; in these tests, the same boundary conditions as expected in practice were simulated. The short-time and
long-time (creep) tests demonstrated sufficient load-bearing capacity and stiffness of the panels. The
experimental results depicted a reasonable correlation with the prediction of the elastic plates. Sound
absorption tests were performed on a system composed of seven panels. The acoustic tests showed that the
panels were effective in abatement of the noise and could be used as traffic noise-reducers. To classify the
material, combustibility tests were also performed on small samples. It was found that the recycled material
had a moderate to low degree of combustibility.
Keywords: Noise abatement, used tire, recycled rubber, rubber walls.
1
Introduction
Recycled polymeric and rubber materials can have various potential applications some of which appear technically
sound and economically attractive. Among the potential products of recycled rubber from vehicle tires, flat and
ribbed panels made of aggregate of these materials have been considered by several manufacturers. It has been
proposed that such panels equipped with waffle plates might be used in the vertical walls which would be installed
along the highways to reduce the traffic noise.
The main apparent technical problems posed in relation with these types of potential noise barriers would be their
noise abatement capability, structural performance, fire resistance, weatherability, chemical activity, and
environmental effects. To assess the quality of new products of such materials, a series of mechanical, noise
abatement, and flammability tests should be performed on a number of samples from the panels made of recycled
rubber products. The primary function of these panels, i.e., their noise abatement capability is among the most
important aspect of their performance.
The motivation for this work and the ensuing request came from a tire-producing industry who intended to reuse
the old tire products; to do so, the plan was to produce ribbed panels made of recycled rubber which could be used in
the noise abatement walls along the highways. The quality of some samples of this product was to be assessed. The
Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

138

MOTAVALLI, FARSHAD AND FLELER

experimental work which was undertaken for this task included material tests of mechanical properties, tests of a
wall module-made of a number of panels- subjected to simulated wind action, noise abatement tests, and
combustibility tests. Some theoretical correlation was also planned.
2
Composition and construction of the panels
A noise-abatement element was composed of a structural part and a non-structural waffle. The structural part was
constituted of a ribbed composite panel. A typical ribbed panel had a dimension of 3000620 mm; in the thickness
direction, the panels were longitudinally subdivided into three rectangular panels the border of which were stiffened
with rubber ribs in intervals of about 900 mm; the thickness of panels between the rubber stiffeners amounted to 40
mm; the thickness of ribs was about 80 mm.
The material constituting the ribbed panels was a recycled product of used tires. The composite section of each
panel was composed of a core made of granular recycled black rubber and two facings at the ribs, but one layer of
facing at the panels between the ribs. The facing(s) were of granular vulcanised EPDM material; they had a green
colour. The particle size of recycled aggregate amounted to about 4.0 millimetre. The thickness of the recycled
rubber core was 40 mm and the thickness of the facing(s) amounted to 8 mm. At the ribbed section, the thickness of
the panel amounted to about 94 mm
In the space between the intermediate panels waffle plates were placed; these nonstructural elements were the
main noise-reducing components of the whole panel. These waffles were constituted of composite sections; their
facing was hollow aluminium sheet of the type d8, Tg12 with 40% hole ratio; the core of the waffles was made of a
50 mm thickness corrugated sheet (85 kg/m2) of mineral fibre. The photograph (1) shows two vertically placed
panels; in the top panel, the stiffening ribs and the plates made of recycled rubber between the stiffeners are shown;
the lower part of the photograph shows the noise-abatement plates which are installed between the stiffening ribs.
Ribbed panels were to be vertically placed and would be longitudinally supported by steel profile members thus
forming a noise-abatement wall system.
In practice, a wall module would consist of two vertical steel columns with wide flange (H) profiles in the web of
which three rubber panels would be placed; additionally, the panels would be connected to each other along their
longitudinal edges via steel profile hollow sections with rectangular cross sections; these profiles would be
embedded in the panel at both sides. The rubber panels would then be fixed in position via a number of point bolts
which when fastened would push against the flange of the wide flange column. A typical wall module is shown in
photograph (1).
3
Properties of recycled rubber
To determine some of the mechanical properties of the recycled materials from the used vehicle tires, a number of
mechanical tests were performed. The mechanical investigation included tensile, compression, three-point bending,
and 4-point bending tests. Table (1) summarises the results of the mechanical tests performed on the recycled rubber
material.
Table 1. Material properties of the recycled rubber
Property

Unit

value

Tensile elastic modulus


Maximum tensile stress
Maximum tensile strain
Compressive elastic modulus
Maximum compressive stress
Maximum compressive strain
Elastic modulus in 3-point bending
Maximum bending strain in 3-point bending

N/mm2

44
1
7.8
27
5
39
36
1.1

N/mm2
%
N/mm2
N/mm2
%
N/mm2
N/mm2

The results presented in table (1) show that the recycled material has a relatively low stretch value in comparison
with homogeneous rubber materials; this is due to the granular structure of the product.

NOISE ABATEMENT BY RECYCLED TIRES

139

Photograph No. 1. A noise-abatement wall module; the noise abatement waffle of the upper panel is not yet installed.

Photograph No. 2. Experimental set-up for testing of a wall module under simulated wind action.

4
Experiments on the panels under simulated wind action
The goal of structural investigation was the laboratory simulation of behaviour of the noise abatement wall system
composed of several 3 meter long rectangular panels under lateral wind loading; these panels were enclosed in a loadbearing frame.
The investigation included a series of tests on the material of the wall, short-time and long-time loading of the
panels by lateral pressurewhich would simulate the wind action, theoretical modelling of the wall behaviour, and
correlation of the experimental and theoretical results.
Figure (1) shows the laboratory set-up of a wall module consisting of two relatively rigid steel profiles with Hsection, three longitudinal steel profile sections, and two rubber panels. For testing, this assembly was placed on a
horizontal floor; to prevent lifting due to under-pressure, the system was fixed to the floor along the wide flange (H)
sections.
To simulate the wind action, an inflated rubber membrane balloon with controlled internal pressure was used;
upon inflating, the balloon would press against the floor, at one side, and against the wall panels, at the other side.

140

MOTAVALLI, FARSHAD AND FLELER

Figure 1. Set-up of laboratory experiments on a wall module

Figure 2. Load-deformation diagram of laterally loaded wall panel

The wall panel system was fixed to the floor and thus was prevented from being lifted due to the pressure from the
balloon. In these tests, the same boundary conditions for the panels as expected in practice were simulated. The
internal pressure was increased to a maximum of 7 kN/m2; this would correspond to about four times the wind load
prescribed by the national standard SIA-160; the pressure change was constantly monitored by manometers.
The lateral deformation of the wall module was monitored and measured by means of five dial gauges. Figure (1)
shows the location of these dial gauges and their designations. Figure (2) presents the load-deformation relation
monitored by the five dial gauges. Testing of the panels under short-time loading was terminated at a maximum
lateral pressure of about 0.05 bar; this would correspond to a wind pressure of 5 kN/m2. Under this pressure, the
lateral deformation of the panels was about 60 mm; the maximum recorded lateral deflection amounted to about 80
mm.
Figure (2) shows the load-deformation diagram in the short-term experiments. The vertical axis of this diagram
represents the controlled pressure changes in the balloon; the horizontal axis represents the lateral deformation which
is recorded by the dial-gauge readings.

NOISE ABATEMENT BY RECYCLED TIRES

141

Figure 3. Deformation-time response of the panel system under sustained lateral pressure of 1.5 kN/m2

To investigate the long-time deformation of the panels under sustained lateral loading, the panels were kept under a
lateral pressure level of 1.5 kN/m2 for a longer time; the long-term experiments lasted up to several months. The
maximum deflection after 4.5 months under a pressure level of 1.5 kN/m2 remained confined to about 28 mm.
Figure (3) shows the deformation-time response of the panel system under sustained lateral pressure of 0.015 bar;
this would correspond to a wind action of 1.5 kN/m2.
Figure (3) shows that the recycled rubber material has an instantaneous elasticity, it also demonstrates that the
panel has some degree of permanent deformation. This lasting deformation might be due to creep behaviour of the
material; it might also be partially due to the relaxation of supporting conditions.
5
Theoretical correlation
For correlation of the experimental results, a simple plate theory was used. As a theoretical model, an elastic plate
model of one of the sub-panels with dimensions 99062096 mm was considered. This plate was assumed to be
supported along its edges elastically and subjected to a uniformly distributed wind action of 1.5 kN/m2. The aim of
this modelling was an indirect determination of the elastic modulus of the recycled rubber material; for this purpose
the measured deflections were correlated with the bending deflection predicted by the theory of thin elastic plates.
For theoretical correlation, the displacement of the sub-panel relative to its edges would be needed. The measured
total deflection of the mid-point of the sub panel would be the sum of two dial gauge readings: one at the edge;
another at the midpoint; this would give
(1)
from which one may determine the following value for the relative deflection:
(2)
On the other hand, the classical plate theory gives the following expression for the mid-point deflection:
(3)
Where q is the uniformly distributed load, a is the side of the square plate, E is its elastic modulus, and h is the plate
thickness. Substituting the measured values in this expression, we obtain E=71 N/mm2; referring to table (1), we
observe that this value is about twice the value obtained by the direct measurements. This difference could be due to
geometrical effects which are present in panel testing, but are not too active in tension tests.
6
Other test.
In addition to mechanical tests, sound absorption test and flammability test were also carried out. The aim of the
sound absorption and the flammability tests was classification of the panels according to the available standards.

142

MOTAVALLI, FARSHAD AND FLELER

Noise abatement tests were performed on a system composed of seven panels. To classify the combustability grade of
the material, combustibility tests were performed on small samples of the recycled rubber material; these tests were
performed according to [3].
7
Conclusions
(1) The material properties of the panels made of recycled rubber aggregate characterise a material of low stiffness
(low elasticity modulus) and low strength (low tensile stress).
(2) The behaviour of a wall module composed of steel profiles and the inlaid panels under simulated wind action
proves a strength factor of safety of about 34 against the expected wind load as prescribed in the SIA standard No.
160 [1]. No comment on the allowable long-time deformation of the panels and also the whole system is, however,
presented at this stage.
(3) The sound absorption tests, performed according to the ISO Standard No. 354, showed that the absorption
capacity of the panels would fall in the range of comparable systems.
(4) The combustibility tests on small samples associated a classification of IV (moderate combustible) to the recycled
tire and a value V (low combustible) to the dominantly green EPDM material. The EPDM material embodied some
recycled rubber; due to such inclusions, the combustibility class was reduced.
8
Closing remarks
In this work, the static response of a wall module was experimentally investigated. In reality, noise isolation wall are
built at relatively long distances along the highways; hence, for complete assessment, the dynamic response of the
whole structural system under wind-induced loading should be additionally considered. Moreover, the long-term
loading tests have shown that the panel construction may undergo some creep deformations. Such deformations, if they
exceed a certain amount, can present a non-pleasing appearance; thus the need for stiffening elements or schemes
may arise. No judgement about the presently investigated product in this direction has, however, been passed; to
arrive at such conclusions, the long-term behaviour of the large-scale system under thermal gradientsmust be
investigated in the field tests. Moreover, based on the material combustibility tests, no conclusion may be yet drawn
about the combustibility of the whole system. In addition to the mechanical and physical aspects, there are also some
environmental issues which may of importance in the choice of such systems.
9
References
1.
2.
3.

SIA 160, Int. Standard Organization, (1988), Einwirkungen auf Tragwerke , Zrich.
ISO 354, Int. Standard Organization (1985), Schallabsorptionsgrad (Hallraum-Methode.
Wegleitung fr Feuerpolizeivorschriften, Baustoffe und Bauteile, Teil B: Prfbestimmungen (Brennbarkeitsgrad und
Qualmgrad), Ausgabe 1988, Switzerland.

Acknowledgement
The tested product was produced by the industries Frutiger and Maloya; these industries have provided the test
samples. Such support is thankfully acknowledged.

16
CONSTRUCTION MATERIALS USING POWDERED RUBBER
MADE OF VEHICLE TIRES
R.YAMAMOTO
Muraoka Rubber Reclaiming Co. Ltd, Ichikawa, Chiba-ken, Japan

Abstract
Powdered rubber is generally manufactured from used vehicle tires by mill-grinding and shredder-cutting
methods, or sometimes freeze-grinding method. The powdered rubber is an effective means for the disposal
and recycling of the vehicle tires. In the present paper, the typical manufacturing processes of the powdered
rubbers made of used vehicle tires are outlined, and the process technology, properties and applications of
construction materials such as mat-like paving materials, interlocking rubber blocks and asphalt mixtures using
the powdered rubbers are summarized.
Keywords: Interlocking rubber blocks, mat-like paving materials, mill-grinding, powdered rubber,
recycling, shredder-cutting, vehicle tires
1.
Introduction
Since the 1950s, the automotive industry in Japan has prodigiously been developed, and the amount of used vehicle
tires has outstandingly been increased. The disposal of the used vehicle tires became a big social problem in the
1970s, and various techniques have been developed for the disposal and recycling of the vehicle tires till now. One
of such techniques is the application of the powdered rubber made of the used vehicle tires to the construction
industry.
In this paper, the typical manufacturing processes of the powdered rubbers made of used vehicle tires are outlined,
and the process technology, properties and applications of construction materials such as mat-like paving materials,
interlocking rubber blocks and asphalt mixtures using the powdered rubbers are briefly introduced.
2.
Manufacture of powdered rubbers made of vehicle tires
Powdered rubbers are generally manufactured from used vehicle tires by the processes illustrated in Fig. 1. The used
vehicle tires are cracked with cracking mills, and metallic substances in the tires are separated. Then mill-ground
powdered rubber is produced by grinding the cracked tires with grinding mills and separating nonmetallic substances.
Shredder-cut powdered rubber is also manufactured by cutting the cracked tires with shredders. Powdered rubber
with particle sizes of 300m or finer is produced by applying a freeze-grinding method to the cracked tires. Fig. 2
shows the mill-ground and shredder-cut powdered rubbers.

Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0

144

YAMAMOTO

Fig. 1 Manufacturing Processes for Powdered Rubbers.

Fig. 2 View of powdered rubbers.

3.
Mat-like paving materials for pedestrian areas
Mat-like paving materials for pedestrian areas are produced by mixing the compositions with the formulations as
given in Table 1, placing them in molds, pressing them with hot rollers and curing at ambient temperature.
Table 1 Typical Formulations of mat-like paving materials
Type of material

Formulations (phr*)

Powdered rubber (Mill-ground or shredder-cut)


Urethane prepolymer
Promoter (for cure of urethane prepolymer)
Coloring agent
Note, *: phr, parts per hundred parts of rubber.

100
30
0.51.5
515

Fig. 3 represents an example of the mat-like paving materials. According to the execution method as shown in
Fig. 4, the mat-like paving materials are generally bonded to cement concrete or asphalt concrete substrates by using
adhesives, and finished with acrylic resin topcoatings. Sometimes, fresh compositions with the same formulations are
placed on the primer-treated cement concrete or asphalt concrete substrates, pressed with hot rollers, and finished
with the same topcoating in the field.

CONSTRUCTION MATERIALS USING POWDERED RUBBER

145

Fig. 3 Mat-like paving materials.

Fig. 4 Execution method for mat-like paving materials.

Table 2 shows the properties of mat-like paving material using mill-ground and shredder-cut powdered rubbers.
The mat-like paving materials have good cushioning property, skid resistance, water or vapor permeability, sound
absorption and weatherability.
Table 2 Properties of mat-like paving materials
Property

Type of powdered rubber

Mill-ground

Shredder-cut

Tensile strength (MPa)


Elongation (%)
Hardness (Type A durometer)
Tear strength (N/mm)
Impact resilience (%)

0.48
47
60
3.5
46

0.78
65
63
0.7
44

The applications of precast mat-like paving materials are pavements for play facility areas in parks or
playgrounds, and floorings for entrance halls, verandas, etc. in buildings. The cast-in-place applications of the matlike paving materials are pavements for stadiums, playgrounds, tennis courts, cycling roads, sidewalks, pedestrian

146

YAMAMOTO

Fig. 5 Protective mats under swings.

Fig. 6 Pavement of playground in a primary school.

bridges, promenades, jogging roads, play facility areas and poolsides in parks, etc. Figs. 5 and 6 represent the
examples of the applications of the mat-like paving materials.
4.
Interlocking rubber blocks for pedestrian areas
Interlocking rubber blocks or elastomeric tiles for pedestrian areas are manufactured by mixing the compositions
with the formulations as given in Table 3, and applying a compression molding method to them at 160C under a
gage pressure of 10MPa for 10 minutes. Fig. 7 shows an example of the interlocking rubber blocks. In accordance
with the execution method as illustrated in Fig. 8, the interlocking rubber blocks are bonded to cement concrete or
asphalt concrete substrates by using adhesives in various pedestrian areas.
Table 3 Typical Formulations of interlocking rubber blocks
Type of material

Formulations (phr)

Powdered rubber (Mill-ground)


Urethane prepolymer
Promoter (for cure of urethane prepolymer)
Coloring agent

100
15
0.01
610

Table 4 exhibits the properties of interlocking rubber blocks using mill-ground powdered rubber. The interlocking
rubber blocks have good cushioning property, skid resistance, water permeability, sound absorption and
weatherability like mat-like paving materials. The advantages of the rubber blocks are simple execution, easy
maintenance, free design of patterns and comfortable feeling for walk due to the cushioning property.

CONSTRUCTION MATERIALS USING POWDERED RUBBER

147

Fig. 7 Interlocking rubber blocks.

Fig. 8 Execution method for interlocking rubber blocks.

Fig. 9 Pavement of poolside in a community center.


Table 4 Properties of interlocking rubber blocks
Specific gravity
Tensile strength (MPa)
Elongation (%)
Hardness (Type A durometer)
Tear strength (N/mm)
Impact resilience (%)
Compression set (%)

1.20
3.9
120
66
24
45
28

The applications of interlocking rubber blocks are pavements for amusement parks, playgrounds, tennis courts,
sidewalks, pedestrian bridges, bus stops, promenades, poolsides, etc. Figs. 9 and 10 represent the examples of the
applications of the interlocking rubber blocks.

148

YAMAMOTO

Fig. 10 Pavement of promenade on the riverside.

Fig. 11 Relation between test temperature and strain at failure in bending test of asphalt mixtures with and without powdered
rubber.

5.
Asphalt mixtures with powdered rubbers for roadways
Asphalt mixtures with powdered rubbers for roadways are produced with the materials and mix design as given in
Table 5, and applied by the same process as for ordinary asphalt mixtures.
Figs. 11 to 13 illustrate the temperature dependence of strain at failure (in bending test), change in surface
roughness with additional traffic exposure period, and skid resistance of the asphalt mixtures using powdered
rubbers, respectively. The advantages of the asphalt mixtures are improvements in skid resistance and abrasion
resistance, a reduction in traffic noise and a prevention effect of freezing.
Table 5 Materials and mix design for asphalt mixtures with and without powdered rubber
Type of asphalt mixture

With powdered rubber

Asphalt: Straight asphalt


Filler: Ground limestone
Powdered rubber
Material
Coarse aggregates: Crushed stone Nos. 5, 6
and 7
Fine aggregates: Coarse and Fine sands
Aggregate gradation
Gap-graded
Asphalt content (%)
7.09.0
Powdered rubber content (%) 2.05.0

Without powdered rubber


Except for powdered rubber, the same
materials as the left column

Dense-graded
5.56.0
0.0

CONSTRUCTION MATERIALS USING POWDERED RUBBER

149

Fig. 12 Relation between traffic exposure period and surface roughness of asphalt mixtures with and without powdered rubber.

Fig. 13 Skid resistance of asphalt mixtures with and without powdered rubber in winter.

6.
Concluding remarks
Powdered rubber is an effective means for the disposal and recycling of used vehicle tires. The amount of the used
vehicle tires has markedly been increased year by year. Their consumption as the powdered rubber is still a little at
present. The development of the use of the powdered rubber in construction materials is expected in the near future.
Acknowledgment
The author wishes to thank Mr. Toyoaki Taniguchi of Obayashi Road Construction Co., Ltd. for sending data of his
work, which was cited in Table 5 and Figs. 11 to 13 in this paper.

17
PUNCHING RESISTANCE OF MATS MADE OF RECYCLED
PVC AND RUBBER IN UNDERGROUND CONSTRUCTION*
P.FLELER, M.FARSHAD and A.ROLLER
Eidgenssische Materialprfungs-und Forschungsanstalt, Dbendorf, Switzerland

Abstract
In this contribution, results of a series of punching tests on two types of protection mats are reported. The
purpose of these experiments was investigation of punching resistance of the samples under two different effects;
one of these would simulate the point contact forces and the other would simulate the action of the falling
objects. In the test set-up, some samples of these mats were placed on a fleece while some had no fleece; all
samples were located on a PVC waterproofing membrane. To simulate the punching effect of the heavy weight
tools, samples were subjected to the action of freely falling masses with standard nuzzle geometry; the falling
objects consisted of a 6 kg mass with spherical nuzzle and a 0.560 kg mass with cylindrical punching head; the
falling height of these weight was varied until perforation in the PVC sub-layer occurred. The result was that,
for the samples tested, the requirements of the SIA Norm 280 with specified falling height was fulfilled, i.e., the
water-tightness of the underlying membrane was assured. To simulate the conditions at the construction site,
another series of experiments on these mats, i.e., impact under 1.1 falling mass were performed. It was found
that the protection mat remained effectively intact up to the drop height of over 3 meters and that the
waterproofing membrane remained water-tight. In all tests, some localised plastic deformation in a number of
protective mat samples was, however, observed.
Keywords: PVC mats, rubber mates, recycled mats, protective mats.
1
Introduction
During some fabrication processes, recycled plasticized PVC (with low plasticizer content) and rubber materials
having bonded granular structures are compressed into blocks from which mats of desired thickness, in the order of
about 58 mm, are produced. These mats are employed for protection of waterproofing sheets used in roofing, tunnels,
underground constructions, gardens, channels, and geomembranes. They are normally placed in locations where the
waterproofing layer requires protection against mechanical damage which might jeopardise its water-tightness. To
perform their function as protective layers, the mats must withstand various effects which may cause punching and
perforation of the underlying membrane.
The waterproofing body underneath the mat may be a plastic membrane, a thin aluminium sheet, or a bituminous/
concrete bed. These mats may be loosely placed or may be glued at certain points on the sub-layer. When placed on
the concrete or on the asphalt, point forces may be exerted by the underlying aggregate. In addition, during
construction activities, falling of reinforcements and other objects may induce local damage in the mat and in the

*RILEM Workshop on Disposal and Recycling of Organic (Polymeric) Construction Materials, 1995

Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

PUNCHING RESISTANCE OF MATS OF RECYCLED PVC

151

Diagram 1. Principle of protected multi-layer waterproofing system

waterproofing membrane. Therefore, the main mechanical requirement for the use of these mats would be their
punching and tearing resistance against the localised damages caused by the contact forces and the falling objects.
The goal of this work was an experimental investigation of behaviour of two new types of protective mats made of
two different material groups. In this investigation, two types of external effects were simulated: the first type was
the punching effect of heavy tools; the second was the effect of falling objects on the mats.
2
Test objects
One of the products to be tested was made of recycled plasticized PVC material with granular texture and had an
apparent density of 900 kg/m3; it would be produced and delivered in rolls having a surface area ranging from 12 m2
(for 5 mm mat thickness) to 8 m2 (for 8 mm thickness). The other material consisted of recycled rubber which was
bonded with a Polyurethane Polymer; it had a density of 820 kg/m3. The material of geotexile fleece was Polyester.
The features of waterproofing layers and their protective mats are presented in table 1. Test specimens were taken
from these protected multi-layer waterproofing systems. Diagram 1 depicts the principle of protected multi-layer
waterproofing system.
Table 1. Features of the protected multi-layer waterproofing systems
Characterisation Protective layer
System type 1

Waterproofing membrane
P(1);

Recycled mat type


thickness:
h=8 mm
System type 2
Recycled mat type P(1); thickness:
h=5 mm
System type 3
Recycled mat type R(2); thickness:
h=8 mm
System type 4
Recycled mat type R(2) with
geotextile polymer sheet;
thickness: h=8 mm
System type 5
Recycled mat type P(1); thickness:
h=8 mm
(1) Mats made of recycled PVC material
(2) Mat made of recycled rubber material
(3) PVC waterproofing membranes

M(3)

Bedding

PVC material type


thickness:
d=2 mm
PVC material type M(3); thickness:
d=2 mm
PVC material type M(3); thickness:
d=3 mm
PVC material type M(3); thickness:
d=3 mm

Polymeric geotextile sheet with


weight of 500 g/m2
Polymeric geotextile sheet with
weight of 500 g/m2
Polymeric geotextile sheet with
weight of 300 g/m2
Polymeric geotextile sheet with
weight of 300 g/m2

PVC material type M(3); thickness:


d=3 mm

Polymeric geotextile sheet with


weight of 300 g/m2

3
Sample preparation and testing conditions
Samples to be tested were kept more than 61 hours in the temperature condition of 23C and relative humidity of
50%; this corresponded to the requirements of SN-ISO 291. Sample stripes were taken from the system composed of
protective mat, waterproofing membrane, and the underlying polymeric geotextile polymer bedding; the length of the
strips amounted to 250 mm. The strips were bonded at discrete end points according to the actual conditions of
waterproofing systems.

152

FLELER, FARSHAD AND ROLLER

To simulate the actual conditions corresponding to the falling objects, the samples were loosely placed on the
underlying geotextile polymer bedding. To simulate the effect of roughness of the concrete bedding, a sandpaper (P
60) was placed between the waterproofing system and a steel base plate.
To simulate the conditions at the construction site, i.e., of the falling reinforcement, another series of experiments
on these mats were performed. The aim of these tests was simulation of the punching effect of the falling steel
reinforcement on the mat. In these experiments, a bar with a mass of 1.1 kg was dropped from various heights.
4
Dynamic puncture resistance tests
The purpose of punching tests was an investigation of behaviour of the waterproofing system under punching
effects. To simulate this situation, samples of the waterproofing system were subjected to the effect of falling objects
with standardized mass and geometry. The goal of these tests was to determine the maximum height of falling which
would destroy the water-tightness of the waterproofing system.
The punching tests were basically carried out according to SIA standard number 280 (1983)-test No. 14; some
slight modifications in this standard were, however, implemented. Accordingly, the side dimension of the samples
amounted to 150 mm; a total number of five samples were tested. A mass of 500 g having a spherical tip with radius
12.7 mm. The water-tightness would be controlled with a vacuum bell and soup water; the bell diameter amounted to
30 mm and a vacuum (negative) pressure of 50 kPa would be applied during 30 seconds. The SIA 280 (1983)
requires a maximum falling height of 750 mm at which the waterproofing system remains completely water-tight.
Table (2) summarises the results of punching tests.
Table 2. Results of dynamic puncture resistance tests
Bedding

Waterproofing system

Thickness
[mm]

Fall height
[mm]

Sandpaper on steel plate


type 2
type 3
type 4
type 5
Polymeric geotextile inter-layer on steel plate
type 2
type 3
type 4
type 5

type 1
2
3
3
3
type 1
2
3
3
3

2
900
2000
2000
2000
2
1100
2000
2000
2000

900

1100

Table (2) shows that all systems tested fulfil the requirement of the cited standard. It also shows the strengthening
effect of the underlying geotextile polymer sheet in the case of the waterproofing system in which a thin synthetic
waterproofing membrane has been used. In this case, the increase in the thickness of waterproofing membrane causes
an increase in the punching resistance of the waterproofing system. In the waterproofing system having a 3 mm
membrane, a strength against punching is achieved; for such system the 300 g/m2 geotextile sheet primarily
functions as the levelling medium for the waterproofing system. A common conclusion is that the punching
resistance of the waterproofing system has a strong dependence on the thickness of the waterproofing membrane and
the sub-layer conditions.
5
Punching tests
The aim of these tests was a simulation of the local damage of the waterproofing system by falling objects. For this
purpose, freely falling masses with specified spherical and cylindrical nuzzle were dropped on the specimen which
had been placed on a stiff bedding. The bedding was to simulate a rough concrete substratum; for such simulation, a
sandpaper (P 60) on a heavy metallic base was used. The goal of these tests was determination of the maximum fall
height at which the waterproofing system remained water-tight. Two types of falling objects included (a) a mass of 6.
024 kg with a 10 mm spherical nuzzle; (b) a mass of 0.560 kg with cylindrical head with a diameter of 6 mm. The

PUNCHING RESISTANCE OF MATS OF RECYCLED PVC

153

water-tightness tests were carried out with soup water and vacuum as described in the previous section. Figure (1)
shows the falling object with the spherical nuzzle mounted on the testing device and the specimen which is
constituted of protective mat, the waterproofing membrane, and the geotextile polymeric sheet. Figure (2) shows the
falling object with cylindrical nuzzle mounted on the testing device and the same type of specimen.
Table (3) shows the results of punching tests. This results show that the upper layer of the system type 1 (8 mm
mat) was already damaged at a fall height of 6 m; the failure bound of this system was then estimated to be about 6 m.
In the system type 5, however, the upper layer remained intact at a fall height of 6 m and was not yet damaged.
Table 3. Results of punching tests
Falling object

Protective mat

Steel ball, 6024 gr, 10 mm


type 1
type 2
5
type 3
8
type 4
8
type 5
8
Steel rod, 560 gr, 6 mm
type 1
type 2
5
type 3
8
type 4
8
type 5
8
* Limit of the fall height in the tests

Thickness
[mm]

Thickness of Waterproofing membrane


[mm]

Fall height
[mm]

8
2
3
3
3
8
2
3
3
3

2
300
450
600
850
2
1750
3000
3250
>6000*

650

>6000*

6
Simulation of construction site
To simulate the condition of falling reinforcement bars during construction of reinforced concrete structures, a series
of tests on various mat-membrane systems were carried out. To simulate the underlying concrete, a base consisting
of a sandpaper on a rigid steel plate was used. The mat-membrane system was placed on this base and a mass of
1176 g was dropped from various height on this assembly. The end of the falling object was relatively flat, but had a
sharp edge with diameter of about 12 mm. The water-tightness test were carried out with soup water and vacuum as
described in section 4. Table (4) summarises the results of the falling reinforcement tests.
Table 4. Results of falling reinforcement tests
Falling object

Protective mat Thickness Thickness of Waterproofing membrane


[mm]
[mm]

Fall height
[mm]

Sharp-edge cylinder, 1176 gr, 12 mm


type 2
type 3

type 1
5
8

v: 4500

type 4
8
type 5
8
v: fall of the object normal to the mat
s: Fall of the object slanted to the mat surface

8
2
3
3
3

2
s: 4250
v: 4000
s: <4000
v: 4250
v: 5500

7
Conclusions
In the punching tests, the maximum fall height was achieved by the waterproofing systems type 1 and 5 (2 and 3 mm
PVC membranes and 8 mm protective PVC mat). The protective mat of systems (3), (4) was beyond a certain fall
height protruded; this protrusion remained in the mat after unloading. In the protective mat of systems (1), (2), and
(5), beyond a certain falling height (about 3750 mm) a permanent deformation normal to the direction of fall was
observed; this deformation could have resulted from the plastic response of the PVC material. Under point action of
the underlying concrete aggregate, such permanent deformation might lead to puncture of the waterproofing

154

FLELER, FARSHAD AND ROLLER

Figure 1. The falling object with the spherical nuzzle mounted on the testing device and the specimen which is constituted of
protective mat, the waterproofing membrane, and the geotextile polymeric sheet.

Figure 2. The falling object with cylindrical nuzzle mounted on the testing device and the same type of specimen as in Figure 1.

membrane and potentially jeopardise its water-tightness. Figure (3) shows the composition of the protective mat; it
also shows the mode of damage in one of the samples. Figure (4) shows the photograph of perforation in the
waterproofing PVC membrane.
In the falling reinforcement simulations (1.1 kg falling object), the maximum falling height for punching was
achieved by the system type (5). In these tests also, some permanent deformation in the PVC membrane was
observed. From the results of the punching tests one may conclude that the punching resistance of the systems tested
is dependent on the thickness of the waterproofing membrane and the protective mat. Moreover, it may be concluded
that, in the case of the waterproofing membranes placed on rough surfaces, a higher dynamic puncture resistance of
the waterproofing can be achieved by application of a geotextile sheet.
The perforated protection mats depicted a one-sided lapping as shown in the photograph (3). Because of this
lapping, during placement of fresh concreteand due to loadingthe waterproofing membrane may be damaged.
In this investigation, the short-term behaviour of the protected waterproofing has been studied. It should be
pointed out, however, that, due to the plastic properties of the recycled material, including migration of plasticizers,
modificators, and so on, the long-term behaviour of the system is also of importance and requires due attention.

PUNCHING RESISTANCE OF MATS OF RECYCLED PVC

155

Figure 3. Composition of the protective mat and the mode of damage in one of the samples

Figure 4. Perforation in the waterproofing PVC membrane

Finally, it should be emphasized that, while the concept of recycling proves to be of value, but in evaluation of the
recycled products the environmental aspects and compatibility of the recycled systems with new applications must
also be assessed.
The new concept of recycling of used roofing membranes and rubber products seems to have a promising future; it
appears to be not only an economically sound idea, but it also can potentially open up new horizons for wider
applications.
8
References
1.
2.
3.

ISO 291 (1977), International Standard: Plastics-Standard atmospheres for conditioning and testing, ISO Standards
Handbook 21 Volume 1, Plastics, Second edition (1990), Switzerland.
SIA 280 (1983), Swiss Standard SN 564 280, Edition 1983, Zrich, Switzerland.
ISO 3127 (1980), International Standard: Unplasticized Polyvinyl Chloride (PVC) pipes for the transport of fluidsDetermination and specification of resistance to external blows, ISO Standards Handbook 21, Volume 3, Plastics, Second
edition 1990, Switzerland.

PART FOUR
WOOD-BASED MATERIALS

18
NEWLY DEVELOPED WOOD-CHIP CONCRETE WITH
RECYCLED TIMBERS
Y.KASAI, M.KAWAMURA and J.D.ZHOU
Department of Architecture and Architectural Engineering, College of Industrial Technology,
Nihon University, Narashino, Chiba-ken, Japan
K.MACHIDA
Central Laboratory, Japan Testing Center for Construction Materials, Soka, Saitama-ken, Japan

Abstract
Used timbers are exhausted more than 12 millions cubic meter per year from demolished buildings mainly from
wooden houses in Japan. The authors have carried out a study of recycling used timber as wood-chip concrete.
This special concrete was made under following conditions, water-cement ratios: 0.35, 0.40, 0.50 and 0.60, and
wood-chip packing ratios in volume: 30, 40, 50, 60%. Density of the concrete could be controlled by the woodchip packing ratio, and usually fall into the range from 920 kg/m3 to 1250 kg/m3. Bending strength,
compressive strength, shrinkage, thermal conductivity, coefficient of thermal expansion, working with hand
tools were tested to evaluate the performance of the wood-chip concrete. From these tests, it is found that
wood-chip concrete can be used as a building material.
Keywords: wood-chip concrete, recycling of used timber, special cement paste, injection method of cement
paste, wood-chip packing ratio, strength, specific gravity, thermal conductivity
1
Introduction
It is estimated that the used timber from demolished wooden constructions amounts to about 12 millions cubic meter
per year in Japan. Most of this timber is crushed into chip by a hammer-type crusher at by-product treatment plants.
There are about eighty of this kind of plants in Japan and about 3.6 millions cubic meter of used timber can be
crushed into chip every year. But only a little of these wood-chip are reused as industrial raw materials and the most
of chip are used as fuel. Open up new avenues for the use of the wood-chip is of urgent necessity.
This paper deals with the reuse of used timber as wood-chip concrete which is made by injecting cement paste into
the void of compacted wood-chip. It will be finding that a valuable building material could be made by the method
developed by the authors.

Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

158

KASAI, KAWAMURA, ZHOU AND MACHIDA

Photo. 1 The wood-chip used for industrial raw materials

2
Properties of the wood-chip made of used timber
2.1
Physical properties of wood-chip
The physical properties of wood-chip of used timber crushed by a hammer-type crusher were investigated. The
samples of wood-chip were taken from five by-product treatment plants in Kanto area. Table 1 shows the physical
properties of the wood-chip mentioned above. In the table, the sorts of timber for the industrial raw materials were
conifer trees e.g. Japanese cedar, Japanese cypress, pine and Japanese beech. The wood-chip for fuel were mixed of
conifer trees and broad-leafed trees. Photo. 1 shows the sample of wood-chip used for this experiment.
Table 1. Physical properties of wood-chip
Class by uses

Apparent specific gravity

Ratio of void (%)

Weight of unit vol. Moisture content


(kg/m3)
(% in weight)

under air dried

under oven dried

under oven dried

under air dried

under oven dried

wet condition*

Industrial raw
0.430.45
0.410.43
6673
130150
1222
5865
materials
Fuel
0.440.48
0.420.45
5870
130160
1721
5363
* Wet condition: Wood-chip was dipped into water 30 minutes, thereafter moisture was removed by a centrifugal machine for 1
minute

2.2
Grain size distribution and the ratio of void of wood-chip
The grain size of wood-chip produced in each plant could be investigated by the method of sieving test. The test
method is as follows: the first, the sample of wood-chip is vibrated for three minutes by a sieving machine, and the
weight of remaining wood-chip in each sieve is measured, then, vibrated for 1 minute and measured, until the
difference of weight before and after sieving becomes less than 1% to the weight of befor sieving. The results of
sieving test showed that the grain size distribution can be expressed by a normal frequency distribution, and the grain
size of wood-chip in each plant are almost the same.
The ratio of void in wood-chip was investigated by the sand replacement method using 3,000 ml cylindrical
vessel. The used sand was the standard sand for the physical testing method of cement (JIS R 5201). It is found that
the ratio of void is in proportional to median diameter calculated from sieving curve as shown in Fig. 1.
2.3
Aspect ratio of wood-chip
The shape of wood-chip crushed by a hammer-type crusher is long in fiber direction and narrow in right angle
direction. Table 2 shows the aspect ratios of wood-chip. The aspect ratio of wood-chip for industrial raw materials is
larger than that of fuel.

WOOD-CHIP CONCRETE WITH RECYCLED TIMBERS

159

Fig. 1 Median diameter vs. the ratio of void in wood-chip


Table 2. Aspect ratio of wood-chip
Class by uses
Industrial raw materials
range
Fuel
range

Apparent sieve opening (mm)


2.5

5.0

10.0

15.0

20.0

25.0

mean
7.223.6
mean
6.923.7

11.15
3.519.6
10.07
4.229.2

10.13
4.517.8
10.07
2.027.1

10.09
3.413.2
11.76
2.310.5

8.46
2.99.2
8.09
2.110.9

7.12
5.29.8
8.70
2.411.5

7.18
6.04

3
Properties of the injection cement paste
3.1
Materials used for the cement paste
1. Cement: Ordinary portland cement (specific gravity: 3.16, specific surface area: 327,000 mm2/g)
2. Water: Tap water
3. Chemical admixture: Special admixture for flowing concrete (Polycarboxylate and special water-solulade
polymer)
3.2
Mixing method and primary water-cement ratio
The small quantities of bleeding and 12~15 seconds of the flowing time through a J-funnel were required to the fresh
cement paste for injection. A special mixing method called double mixing was applied to make a flowing fresh
cement paste with minimum segregation. Fig. 2 and Fig. 3 show the results of double mixing. In these figures, when
a primary mixing water (W1) was 0.23 in water-cement ratio, the bleeding and flowing time became the minimum.
So, at the overall water cement ratio of 0.40, 0.50 and 0.60, the appropriate primary mixing water cement ratio (W1/
C) was 0.23.

160

KASAI, KAWAMURA, ZHOU AND MACHIDA

Fig. 2 Primary mixing water cement ratio (W1/C) vs. bleeding

Fig. 3 Primary mixing water cement ratio (W1/C) vs. flowing time

3.3
Effect of chemical admixture
When water cement ratios are 0.40, 0.50, 0.60, an optimum dosage of chemical admixture was 1% of cement in
weight. But when water cement ratio became 0.35, the cement paste could not flow through the J-shape funnel and
the optimum dosage of chemical admixture in this case was 1.5%.
4
Method of making wood-chip concrete
4.1
Materials used
1. Cement, water and chemical admixture were same as to 3.1.
2. Wood-chip: made of used timber crushed by a hammer-type crusher
4.2
Preliminary test on previous treatment of wood-chip
The previous treatment on wood-chip made of used timber is necessary, because some wood release harmful
extractives which delay the hydration of portland cement. Four kinds of previous treatment were tested, e.g. cold tap
water, warm tap water (60~70C), solutions of 0.1% NaOH and 0.5%NaOH. The strength test results of wood-chip
concrete using these treated chip indicated almost the same, because the wood-chip used in this experiment were
conifer so the harmful water-soluble extractives were little. After this result, tap water was used as a previous
treatment liquid.

WOOD-CHIP CONCRETE WITH RECYCLED TIMBERS

161

Photo. 2 (image) The cross-section of specimens with different packing ratio of wood-chip (white part is constituted of cement
paste)

Photo. 3 The outlook of the specimen which was halted during the injection (packing ratio of wood-chip: 50%)

4.3
The method making test pieces
After dipping in cold tap water about 30 minutes, water was removed by a centrifugal machine for 1 minute. Then,
the wood-chip were packed into a steel mold and compressed to achieve the desired packing ratio, and the
compressive stress were 0.5~1.0 MPa. The fresh cement paste holding 2 meter head was injected from the bottom of
mold which was slanted about 15 degree, and it took about 30 seconds to fill the specimen. Photo. 2 (image) shows
the cross-section of specimens with different packing ratio of wood-chip and Photo. 3 shows the outlook of the
specimen which was halted during the injection.

162

KASAI, KAWAMURA, ZHOU AND MACHIDA

The dimension of the wood-chip concrete specimens was 100 100400mm. The three pieces of specimens were
made for each test, and the test result was given by the mean of them.
4.4
Mixing and Curing method
Table 3 shows the test plan of wood-chip concrete. Fresh cement paste for injection was mixed by the double mixing
method. The primary mixing water cement ratio was 0.23 and the chemical admixture was added to the primary
water.
Table 3. Test plan of wood-chip concrete
Water cement ratio

Packing ratio of wood-chip in volume

30%

40%

0.35
0.40
0.50
0.60

50%

60%
o
o
o
o

Curing method: The specimens were demolded at the age of 24 hours, and cured in the room of 20C under sealed
condition in plastic bag until the age of 7 days. After that, it was cured in the room of 20C and 60% relative
humidity until the age of 28 days, 91 days, or 183 days.
4.5
Testing method
Testing methods for wood-chip concrete were shown as below:
1. Bending test: JIS A 1106 (Method of Test for Flexural Strength of Concrete).
2. Compressive test: JIS A 1114 (Method of Test for Compressive Strength of concrete using portions of Beams
Broken in Flexure).
3. Static modulus of elasticity test: with 100100200mm specimen.
4. Thermal conductivity test: JIS A 1412 (Test Method for Thermal Transmission Properties of Thermal
Insulations).
5. Linear thermal expansion test: JIS A 1325 (Test Method for Linear Thermal Expansion of Building Materials).
6. Nail pull out test: The nail pull out test was done by an Autograph Testing Machine.
5
Result and consideration
5.1
The mechanical properties of wood-chip concrete
5.1.1
Bending strength
The relationship between the water cement ratio and the bending strength of wood-chip concrete under curing age is
shown in Fig. 4. Generally speaking, when the water cement ratio decreases, the bending strength increases. On the
other hand, the bending strength increases almost in a linear relation until the age of 28 days, but decreases from the
age of 28 days to 183 days. The reason should be considered by the drying shrinkage of wood-chip. Because the
drying shrinkage at radial and tangential direction of the annual ring wood are considerably larger than that at the
fiber direction, so the bond strength between the wood-chip and cement paste was reduced. Therefore, the bending
strength of wood- chip concrete decreased at the age of three months. From Fig. 4, it is seen that the bending strength
gradually becomes constant when the age exceeds three months.

WOOD-CHIP CONCRETE WITH RECYCLED TIMBERS

163

Fig. 4 Curing age and water cement ratio vs. bending strength of wood-chip concrete

Fig. 5 Packing ratio of wood-chip vs. bending strength of wood-chip concrete

The relationship between the packing ratio of wood-chip and the bending strength of wood-chip concrete is shown
in Fig. 5. At the age of 3 days and 7 days, the bending strength decreases linearly when the packing ratio is
increased.
5.1.2
Compressive strength
The relationships between the curing age, the water cement ratio and the compressive strength of wood-chip concrete
are shown in Fig. 6. The compressive strength increase linearly while the curing age increases from 3 to 91 days. The
compressive strength becomes almost constant from the age of three months to six months except for the case of
water cement ratio of 0.35. Fig. 6 shows that the compressive strength increases when water cement ratio decreases,
but it is almost the same when water cement ratio are 0.40 and 0.35.
Fig. 7 shows the relationship between the packing ratio of wood-chip and compressive strength of wood-chip
concrete. The compressive strength decreases linearly when the packing ratio of wood-chip is increased. From above
mentioned results, it is clear that the compressive strength of wood-chip concrete is influenced by the both of
strength of cement paste and packing ratio of wood-chip. The ratio of bending strength to compressive strength is
relatively larger than that of normal concrete.
5.1.3
Compressive static modulus of elasticity
The compressive static modulus of elasticity is shown in Table 4. The static modulus decreases when water cement
ratio is increased, and the static modulus increases when the packing ratio of wood-chip is decreased.

164

KASAI, KAWAMURA, ZHOU AND MACHIDA

Fig. 6 Curing age and water cement ratio vs. compressive strength of wood-chip concrete

Fig. 7 Packing ratio of wood-chip vs. compressive strength of wood-chip concrete


Table 4. Compressive static modulus of elasticity of wood chip concrete unit: 103MPa
Water cement ratio

Packing ratio of wood-chip (Vol. %)

30

40

0.50
Packing ratio of wood-chip(%)
0.35
50

3.52
2.81
Water cement ratio
0.40
0.50
4.00
3.50

50

60
2.37

2.06

0.60
2.37

1.68

5.2
The physical properties of wood-chip concrete
5.2.1
Density and specific strength
The apparent density of wood-chip concrete is fall into the range of 920~1250 kg/m3 under air drying condition.
Table 5 and Table 6 show the specific strength and density of wood-chip concrete.

WOOD-CHIP CONCRETE WITH RECYCLED TIMBERS

165

Fig. 8 Packing ratio of wood-chip vs. thermal conductivity of wood-chip concrete


Table 5. Specific strength and density of wood-chip concrete
(a). In the case of water-cement ratio: 0.50
Packing ratio of wood-chip (Vol. %)
30
Specific strength (MPa)
5.8~6.2
3
Density (kg/m )
1210~1250
(b). In the case of packing ratio of wood-chip: 50%
Water cement ratio
0.35
Specific strength (MPa)
5.5~6.0
Density (kg/m3)
11901240

40
5.7~6.3
11001190

50
5.5~6.5
10101100

60
5.6~6.2
920990

0.40
5.4~6.1
1170~1200

0.50
5.5~6.5
10101100

0.60
5.7~6.5
9401000

5.2.2
Thermal conductivity
The thermal conductivity of wood-chip concrete was measured with the guarded hot plate method at JTCCM. The
relationship between the packing ratio of wood-chip and the thermal conductivity is shown in Fig. 8. After the
figure, it is known that the thermal conductivity decreases linearly when the packing ratio is increased, and the
moisture content has a great influence on thermal conductivity of wood-chip concrete.
5.2.3
Coefficient of linear thermal expansion
The coefficient of linear thermal expansion of wood-chip concrete was measured by the push rod meter at JTCCM.
The result of measurement is shown in Table 6. Because of the influence of moisture content, the coefficient of
linear thermal expansion are different with the temperature range below and over 40C. Within the range of 8~40C,
when the packing ratio of wood-chip is increased, the coefficient of linear thermal expansion increases.
Table 6. Coefficient of linear thermal expansion of wood-chip concrete* unit: 106(1/K)
Temperature range(C)

Packing ratio of wood-chip (Vol.%)

30

40

50

60

8~40
*Specimens were in air dried conditions.

12~16

16~20

22~26

26~30

166

KASAI, KAWAMURA, ZHOU AND MACHIDA

Fig. 9 Water cement ratio vs. nail pull out load of wood-chip concrete

Fig. 10 The packing ratio of wood-chip vs. nail pull out load of wood-chip concrete

5.3
Workload of wood-chip concrete
5.3.1
Nail pull out load
The nails were knocked into wood-chip concrete by a 500 gram hand hammer. The length of nails were 32, 50, 65
and 90 mm, and two third of the length were knocked into the wood-chip concrete. Fig. 9 shows the relationship
between water cement ratio and the pull out load. The pull out load decreases when water cement ratio is increased,
and the pull out load increases when the nail becomes larger.
The relationship between the packing ratio of wood-chip and the pull out load of nails are shown in Fig. 10. From
the figure, it is clear that the pull out load decreases when the packing ratio of wood-chip is increased.
5.3.2
Sawing
The wood-chip concrete with the packing ratio of 50, 60% and water cement ratio of 0.50, 0.60 can be sawed by a
handy saw.
6
Conclusions
The results obtained after these experiments are as follows:

WOOD-CHIP CONCRETE WITH RECYCLED TIMBERS

167

1. The ratio of void of wood-chip crushed by a hammer-type crusher can be estimated by the median diameter of
wood-chip which can be easily obtained from the sieving curve.
2. The suitable fresh cement paste for injection into the void of wood-chip can be made by double mixing with the
special admixture. The amount of primary mixing water is 0.23 in water-cement ratio.
3. The wood-chip concrete can be made by injecting cement paste into the void of wood-chip.
4. The bending strength of wood-chip concrete is about 4.0~7.0 MPa, the compressive strength is about 5.0~8.0
MPa, and the ratio between bending strength to compressive strength is in the range of 0.5~0.9, it is greater than
that of the normal concrete.
5. The compressive static modulus of elasticity of wood-chip concrete is in the range of 1.7~4.0103MPa.
6. The density of wood-chip concrete is in the range from 920 to 1250 kg/m3.
7. The thermal conductivity of wood-chip concrete is 0.2~0.36 W/mK in the air dried state.
8. The wood-chip concrete have a good work in nailing and sawing.
Acknowledgment
This study was supported by Kishitani Trust for the International Research of Building Materials and Fire Protection
Technology in 1994.
References
1.
2.
3.
4.
5.

Y.Kasai, Y.Fukushima and K.Tominaga, Production and properties of cemented crushed wood-chip board made of used
timber Transaction of the Japan Concrete Institute Vol. 4, 1982, pp. 1724
Y.Kasai, Y.Fukushima and K.Tominaga, Properties of cemented crushed wood-chip board reinforced with excelsior
Transaction of the Japan Concrete Institute Vol. 5, 1983, pp. 18
Sarja A, Structural wood-fiber concrete Concrete International 1989.7
E.Tazawa and T.Kasai, Double mixing effect of fresh cement paste Proceedings of JICE No. 396/V-9, 1988.8
Y.Kasai, Making a Board of portland cement and excelsior with used timber Gypsum & Lime No. 234, 1991, pp.160166

19
PARTICLEBOARDS MADE FROM RECYCLED WOOD
S.SUZUKI
Department of Forest and Wood Engineering, Faculty of Agriculture, Shizuoka University,
Shizuoka, Shizuoka-ken, Japan

Abstract
Particleboard or mat-formed composite board is one of the products that can use recycled wood as a raw
material. According to questionnaires, recycled wood used in the particleboard industry is estimated to be 20%
of the total chips, and the amounts have been increasing. Laboratory scale particleboards were made using
recycled P-chip and S-chip to evaluate chip quality. The results showed that the quality of P-chip was
equal to that of ordinary chip. Hammer-milled particle from low grade chips can be used as a core material of
particleboard. A new type of structural board with three-layer construction was proposed to expand the use of
recycled wood chips.
Keywords: Bending property, composite board, OSB, particleboard, recycled wood, waste wood.
1
Introduction
The total amount of all kinds of wooden waste being generated in Japan was roughly estimated to be about 30
million m3 per year (1). One of the important resources is building construction and demolition because of its
quantity and continuous appearance in the waste stream (2) (3). There are no precise statistics on waste wood. The
estimated value of the waste wood from demolition wood was 5.6 million m3 in 1989. This was calculated from the
total floor area of demolished buildings in 1989, 40 million m2, and the rate of generation per unit area which was
assumed to be 0.14 m3/m2 (1). Another statistics showed that wooden house construction consumed wood at 0.212 m3
per unit floor area and generated 0.032 m3/m2 wood waste. If the collection from generation points is proved
economically possible, the waste wood can become an important wood material.
Chip makers dealing with waste wood produced about 1.8 million tons per year, and about 70 % of chips were for
fuel and about 30% were recycled as a raw material in the paper and board industries (4). The amount of reuse of
waste wood was very small compared with the generation. Particleboard or mat-formed board is a product that can
use recycled wood (5) (6). Thus, the problems of recycled wood for board making was discussed in this paper.
This paper consists of three phases involving the investigation of the problems of recycled wood usage. In the first
phase, conditions covering recycled wood used in the particleboard industry were surveyed and analyzed through
questionnaires submitted to particleboard makers. Chip and particle quality in board making was investigated by an
experimental study conducted in the second phase. Particle processing operations were also discussed. In the third
phase, mechanical properties of three-layer composite board were evaluated and a new type of board named
Japanese OSB was proposed.

Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

PARTICLEBOARDS MADE FROM RECYCLED WOOD

169

Fig. 1. Particleboard production of eleven makers in 1992 and the proportion of recycled wood used.

2
Waste wood used in particleboard industry
2.1
Method of investigation
To ascertain the situation regarding the reuse of waste wood as a raw material in the particleboard (PB) industry in
Japan, a questionnaire was sent out for survey. Replies to the questionnaire were received from eleven PB makers.
As fifteen makers (sixteen mills) now produce particleboard in our country, there was a recovery rate of 73%. Main
questions included in the survey were; amount of waste wood usage and its percentage to the total raw material;
prediction of future trends; kinds of waste wood used; collection of waste wood; quality of waste wood and cleaning;
quality of board produced; and technical problems when using waste wood. The survey was conducted as a part of the
research project supported by the Japan Housing and Wood Technology Center (HOWTEC).
In addition, information on reuse of waste wood was obtained through interviews with technical managers of PB
mills. Six PB makers gave me an opportunity to meet and discuss with them abut technical problems.
2.2
Amount of waste wood chip used
The eleven PB makers, who returned our questionnaire cards, together produced about 683 thousand tons of
particleboard in 1992. This corresponded to about 80% of total amount of the annual production of PB, which was
estimated to be about 768 thousand tons. Since the location of the eleven PB makers was widespread from the north
(Hokkaido) to the south (Kyushu) in Japan, the results derived from the questionnaire survey were thought to be a
fair reflection of the current condition of the particleboard industry.
Figure 1 shows the annual PB production of each maker and the shaded portion indicates the ratio of recycled
wood usage. This was calculated simply; it was assumed that a percentage of PB made from recycled wood was
comparable to the weight percentage of recycled wood chip used in each mill.
It was found that three makers among the eleven did not use any recycled wood and that the amount of recycled
wood used in four makers through Maker-E to -H was very small. Maker-B and -D, which produced more than one
hundred thousand tons of boards, used approximately 20% recycled wood as a raw material. This may be the typical
case in the Japanese particleboard industry.
On the other hand, Maker-A and -C strongly depended on recycled wood resources. The highest percentage of
90% was found in Maker-A. The use of recycled wood by PB maker in Japan differed according to the mill location.
Generation of waste wood varied with the location and the cost of collection also varied with the distance between
mill and generation point.
The sum of the shaded portion shown in Figure 1 reached about 136 thousand tons. Since the total production of
eleven makers was 638 thousand tons, it was concluded that about 20% of PB was produced from the waste wood
resources in 1992. When the survey was conducted ten years ago (7), only a few mills used recycled wood. Changes
in conditions for procuring raw material could have elevated the importance of waste wood in PB production over
the last decade.

170

SUZUKI

Fig. 2. Predicting the amount of recycled wood usage in the future. Shaded portion is an estimate based on production in 1992.

Fig. 3. Kinds of recycled wood used in the particleboard production in Japan.

For estimating the future trend, the amount of recycled wood that will be used both in the next five years and in
the not too distant future needs to be known. Figure 2 shows the predicted amounts of recycled wood use by each
maker. Maker-I and -J did not have predictions for the future. Their values were regarded as zero in the calculation.
Two makers predicted that their PB production in the future would be thoroughly depended on recycled wood. Two
makers replied that it would be more than 60%, and another two makers replied that it would be about from 40% to
50%. Total amount of recycled wood usage of the eleven makers was roughly estimated by weighting the production
scale of each mill in 1992. The calculation shows that the percentage of recycled wood to total raw material in the PB
industry will be 37% in the next five years and 45% in the not too distant future.
Reasons for the importance of recycled wood in PB manufacturing can be summarized as follows. The Japanese
PB industry has used wood residues from plywood production. However, such wood residues have become
insufficient for PB production because of the increasing production in the PB industry and the decreasing trend in the
plywood industry. PB mills have had to use other domestic wood chip or imported wood chips. With such problems
of raw material, some mills in favorable locations for collection have began to use recycled wood.
Many technical managers of PB mills said that they want to or will have to use recycled wood in the future, even
if the recycled wood possesses problems such as collection, cleaning, quality, cost and so on.
2.3
Kinds of recycled wood
Replies on the kinds of recycled wood chips and their ratios were received also from seven makers. Ratios to the
total amount were calculated by weighting the annual production of each mill. Results are shown in Figure 3.
About a half (51%) of the recycled wood used was obtained from demolished buildings. Wood resources
generated from demolition can be a promising raw material for PB production, because the amount of waste wood in
construction and demolition is estimated to be about ten million m3 per year (3).

PARTICLEBOARDS MADE FROM RECYCLED WOOD

171

Fig. 4. Problems on the recycled wood used in particleboard production.

Ratios of wood chip from palette, crating or packing material, concrete-form plywood, and others were 18%, 14%,
13%, and 4%, respectively. Mills that can use this type of recycled wood were limited. They selected the waste wood
resources according to their own locations.
2.4
Problems on recycled wood use
Problems on recycled wood when used as a raw material in PB production were analyzed according to the
questionnaire survey from the eleven makers. Problems can be classified into eight items, and the results of the
importance calculated numerically with some assumptions are shown in Figure 4.
Main concern was the cleaning and removal of foreign materials in wood chips such as metals, sand, plastics,
rubber and so on. Metals and sand can cause serious damage to the knives of the flaker in the PB line, and those
mounted in panel products also can make troubles to end-users in its secondary processing. Metals can be removed
by magnetic separators that are set at many points in PB lines, and nonferrous metals are removed by metal
detectors. Contaminants such as sand, plastics, and rubbers can be cleaned by means of washing and air separating.
However, complete cleaning is difficult and the improvement of chip quality is always the main problem on recycled
wood chips.
The calculated importance of the items was in the following order; cleaning and removal of foreign substance,
chip quality, stable supply, cost, quality of products, facility investment, and others.
3
Properties of particleboard made from demolition wood
There are two types of recycled wood chip obtained from house construction and demolition. These are called P-chip
and S-chip (8). P-chip is defined as high quality knife-cut chips obtained from sound timbers or lumbers with a
relatively large cross section as used in house construction. P-chips are mainly used in the pulp and paper industry. Schip is a general term for crushed chips obtained from demolition wood or recycled wood. S-1 chip is a high grade
chip in for paper or board making. S-2 is for the board industry, and S-3 a low quality chip is for fuel.
In this section, properties of particleboard made from demolition wood were evaluated using P-chip and S-chip in
laboratory scale tests.
3.1
Wood furnish
Four types of particle (a, b, c1, c2) were prepared as wood furnish for the board making. Particle-a was processed
from P-chip using a knife-ring flaker, which is commonly used in PB mills. Quality of this particle was comparable
to or higher than that of ordinary particles used in PB mills in Japan. In general, milling residues in plywood making
or recycled wood chips were further reduced to desired size using this type of flaker.

172

SUZUKI

There is a possibility that recycled wood chips contain foreign materials such as metal, aluminum, sand, plastics,
and so on. A hammer milling operation seems to be preferable for processing recycled wood chip into wood furnish,
because foreign materials can cause damage to knives of the flaker.
Hammer milling was chosen for the preparation of the wood furnish. A laboratory scale hammer milling tool
(Hosokawa Micron Co., H86), which has fifty-six hammer plates on its cylinder and has 15 mm opening screen, was
used. Chips were crushed into particles at 4400 rpm. Particle-b was a hammer milled particle obtained from sound
chip, P-chip. Expecting some decrease in the quality of the demolition wood, P-chip was subjected to one hour
boiling treatment before hammer milling. Two lots of S-chip with some differences in quality were also hammer
milled to prepare Particle-c1 and Particle-c2 after 30-minute steaming at 0.5 MPa.
Mean values in particle dimension, thickness, length and slenderness ratio (length/thickness), used in this test are
shown in Table 1. These mean values were based on two hundred measurements of each furnish type and calculated
using the Weibull distribution function.
Table 1. Results of shape analysis of each type of particle.
Particle type

c1

Thickness (mm)
Length (mm)
Slenderness ratio

0.43
21.4
57

1.98
19.0
14

1.56
13.3
11

3.2
Particleboard manufacturing
Using four types of wood furnish, i.e., Particle-a, -b, -c1, and -c2, particleboards were made in the laboratory. Eight
types of board were made from these particles; six types of homogeneous board and two types of three-layer board
which had a 15 weight percent of each face layer and 70% of core layer. Furnish type and construction of PBs are
shown in Table 2. Specifications for board manufacturing were as follows;
1. Board size: 34032010 (mm),
2. Target density: 0.7 g/cm3, based on air-dry weight and volume,
3. Resin: liquid phenol-formaldehyde resin,
4. Resin content: 10% based on solid/dried wood weight,
5. Pressing: Temperature: 180 C, Pressure: 2.5 MPa, Time: 7.5 minutes,
6. Surface: not sanded.
Table 2. Furnish type and construction of particleboard tested.
Board type

Particle

Construction

1
a
homogeneous
a(100%)
2
b

b(100%)
3
c1

c1(100%)
4
c2

c2(100%)
5
a+c2

a(50%)+c2(50%), mixed
6
b+c1

b(50%)+c1(50%), mixed
7
a/c2/a
3-layer
a(15%; face)/c2(70%; core)/a(15%; face)
8
b/c1/b

b(15%; face)/c1(70%; core)/b(15% ; face)


a: Particle-a made from P-chip using knife-ring flaker.
b: Particle-b made from boiled P-chip using hammer-mill.
c1, c2: Particle-c1 and -c2 made from steam treated S-chip.

3.3
Properties test
After conditioning the boards at 25 C and RH 65% for one month, bending specimens measuring 50 mm in width
and 280 mm in length were cut. The three-point bending test was conducted according to the JIS (Japanese Industrial
Standard) A 5908, and modulus of rupture (MOR) of each type of board were determined.

PARTICLEBOARDS MADE FROM RECYCLED WOOD

173

Fig. 5. MORs of boards made from demolition wood particles. Particle type and construction of each board: Confer Table 2.

Internal bond (IB) strength was also determined using a 5050 mm specimen according to the JIS A 5908. An
accelerated aging treatment test was conducted for the evaluation of durability performance. The aging treatment
consisted of vacuum soaking for 30 minutes, a pressure soaking for 24 hours at 0.3 MPa and drying in an air
circulating oven for 24 hours at 60 C. This treatment was repeated, and the IB strength and thickness changes were
determined after 1, 3, 5, and 10 cycles.
3.4
Mechanical properties
MORs of the eight kinds of boards made from demolition wood are shown in Figure 5. Board-1 composed of
Particle-a showed MOR value higher than 30 MPa. This proved the sufficient quality of both the P-chip and the ringflaked particle for board making.
Bending strength of particleboard generally depends on particle shapes, especially the length and the slenderness
ratio of particle (length/thickness) (9). Thin and long particle called flake can give good bending properties for
particleboard. As shown in Table 1, the slenderness ratio of Particle-a was significantly higher than those of hammer
milled particle (Particle-b and -c). Table 1 revealed the superiority in shape of flaked particle.
The MOR value of board-2, i.e., 18.7 MPa, seemed to be fairly good, even if the quality of particle shape
processed by hammer-milling was lower than that obtained by the flaking operation. This showed the potential for
use of hammer-milling for recycled wood processing.
Bending properties of boards made from Particle-c were lower than board-1 and board-2 as shown in Figure 5.
This may have been caused by the low quality of particle shape and by the steam treatment of chips before hammer
milling.
The steam treatment for S-chip and the boiling treatment for P-chip before hammer milling was conducted
considering deterioration and decrease in chip quality. The quality of demolition wood chips in actual use could be
lower than that of chips in this experiment if demolition wood was collected from many types of housing and from
different regions. Thus, the chips were treated before processing. The chips used here were thought to be the lowest
grade of demolition wood chip.
Homogeneous board composed of a mixture of Particle-a and -c2 showed a higher MOR than that of board with
Particle-c2. Also the addition of Particle-b to Particle-c1 increased the bending strength as shown in Figure 5.
However, the effect of mixing combination was not so large in this case.
On the other hand, the MORs of Board-7 and Board-8 were higher than those of Particle-c (c1 and c2) board, and
were higher than those of the mixed homogeneous boards (board-5 and board-6). This indicated that the three-layer
construction gave a significant enhancement in bending strength.
Internal bond (IB) strength is usually evaluated as a index of the bond quality in particleboard. Although, the IB
strength of demolition wood particleboard was also affected by the particle shape and the board construction, each IB
value was higher than 0.5 MPa. This means that all board made here satisfied the JISs requirement of 0.3 MPa.
Another important property of the board was its durability performance, because its manufacture involved hotpressing with resin. Durability performance was evaluated by the aging treatment. There were no differences
between the thickness swellings of each board type after repeated cycles of the aging treatment. It was concluded

174

SUZUKI

from the aging tests that the durability performance of demolition wood particleboard is comparable with that of
board made from ordinary particles.
3.5
Conclusion on recycled wood particleboard
Based from the results of tests, the following points were derived:
Quality of P-chip from demolition wood was equal to that of ordinary chip.
Hammer-milling operation was preferable for processing demolition waste wood chip to wood furnish for board
making.
Demolition wood chips with some contamination or deterioration can be used as core material of particleboard.
4
Japanese OSB using recycled wood
There are two problems concerning wooden resources in Japan. One is the use of waste wood material mentioned
above, and the other is how to use the Japanese conifers planted widely in our country. Especially, finding an
effective use for the thinnings or small diameter logs has always been a difficult problems to solve.
In this section, a new type of structural panel manufactured with a combination of these two wood resources is
discussed. Waste wood chips were processed to particle and used as a core material. Small diameter logs were cut by
a disk flaker to produce wafer or strand, which was used in Wafer-board or OSB production in North America.
These strands were used as face layers of three-layer structural board in this experiment. Strength properties of the
composite board made from Sugi-strand and recycled wood particle were evaluated.
4.1
Experiment
4.1.1
Furnish
Strands were prepared from Sugi (Japanese ceder; Criptomeria japonica D.Don) wood with an oven-dry specific
gravity of 0.39. Each small diameter Sugi bolt was cut into blocks with dimensions of 50 mm-width in longitudinal
direction and 20 mm-thickness in the cross-grain direction. Strands (50 mm by 20 mm by 0.6 mm) were prepared by
a disk flaker from these blocks.
Recycled wood particles were obtained from a PB mill manufacturing boards from recycled wood resources. The
particle was processed from demolition wood in the PB mill.
4.1.2
Board manufacturing
Specifications for board manufacturing were the same as mentioned in section 3.2. Construction of the board is shown
in Table 3. Board-1 and board-5 were homogeneous board, and board-2, -3, and -4 were three-layer board. Board-1
was recycled wood particleboard and the board-5 was strand-board without orientation.
For the evaluation of the fundamental properties of composite board named Japanese OSB, no orientation was
given in the face layers in this experiment.
4.1.3
Determination of mechanical properties
Bending test was conducted as shown in section 3.3.

PARTICLEBOARDS MADE FROM RECYCLED WOOD

175

Fig. 6. Effect of face layer thickness on MOR of three-layer composite made from Sugistrand and recycled wood particle.
Table 3. Furnish type and construction of composite board tested.
Board type

Face layer ratio (%)

Weight percent of each layer (%)

Strand* (Face)

Particle** (Core)

Strand (Face)

1
0
2
20
3
34
4
50
5
100
* Strand: Sugi-strand.
** Particle: Recycled wood particle.

0
10
17
25
50

100
80
66
50
0

0
10
17
25
50

4.2
Results and discussion
MORs of the composite board made from Sugi-strand and recycled wood particle are shown in Figure 6. The mean value
of the MOR of the board-1 made only from the recycled wood particle was 17.4 MPa. Bending strength of composite
board increased linearly with an increasing face layer ratio up to 50%, and then leveled off as shown in Figure 6. The
MOR of the strand board (board-5), of which face layer ratio was 100% in the Figure, seemed to be equal to or
slightly smaller than that of board-4. The MOR of board-3 and board-4 was 34.8 MPa and 41.0 MPa, respectively.
Bending strength of three layer-board having Sugi-strand faces were quite excellent compared with the commercial
particleboard used in Japan.
Figure 7 shows the relationship between the face layer ratio and the modulus of elasticity (MOE) in bending of the
composite boards. It was obvious that the MOE increased with an increasing face layer ratio of the composite board,
because the elastic modulus of the strand layer was higher than that of the core layer with recycled wood particle. If a
simple combination of different elastic bodies was introduced, the MOE of three-layer board can be predicted easily
by a cubic equation of face layer ratio. It was thought that the MOE of the composite board shown in Figure 7 was
predictable from the MOE data of recycled wood particleboard and strand-board.
The MOE of three-layer board was also higher than that of commercial particleboard. This showed the possibility
of manufacturing high performance panel from strands of domestic conifer and recycled wood particles.
From the results of the bending tests as shown in Figure 6 and Figure 7, it could be concluded that a preferable
face layer ratio was 30% to 40%. In the case of board-3, which had 17% by weight of strand layer in both faces,
MOR and MOE were 4.04 MPa and 34.8 GPa, respectively.
If the strands of face layer were oriented like OSB, we can obtain higher bending properties along the orientation.
In that case, a decrease in strength in orthogonal direction to the orientation should be considered. In an other
attempt, a cross oriented three-layer board was manufactured in the laboratory. Sugi-strands in the face layers were
oriented, and the core layer particle obtained from recycled wood was cross-oriented to the orientation of the face
layers. This may become one of the Japanese OSB using recycled wood. We can also try another combinations
using two types of furnish for boards, such as a five-layer oriented board, a multi layer-board, and so on.

176

SUZUKI

Fig. 7. Effect of face layer thickness on MOE of composite board.

4.3
Conclusion
Method of manufacturing a high performance board composed of both Sugi-strands and recycled wood particles
were proposed.
The bending test results showed that a preferable face layer thickness in each face was 15% to 20% of the total
thickness.
5
References
1.
2.
3.
4.
5.
6.
7.
8.
9.

Hamada, M. (1991) For recycling of wood resources (Part 2), Wood Industry (Mokuzai Kougyo), Vol. 46, No. 3.
pp. 1106.
Arima, T. (1994) Timber as eco-material, J. Soc. Mat. Sci., Japan (Zai-Ryo), Vol. 43, No. 485. pp. 12736.
Kuwabara, K. (1993) Wastes from demolished building (Heisei kaitai shin-sho), Nippo Co., Japan.
Japan Housing and Wood Technology Center (1992) Report of the survey on recycling of the wooden waste, HOWTEC
Report-1992, Japan.
Suzuki, S., Saito, F. and Arima, T. (1985) Properties of particleboards made from wood element from demolished building,
Wood Industry (Mokuzai Kougyo), Vol. 40, No. 1. pp. 1520.
Rowell, R.M., et. al. (1993) Opportunities for composites from recycled wastewood-based resources; a problem analysis
and research plan, Forest Prod. J., Vol. 43, No. 1, pp. 5563.
Building Contractors Society (1982) Usage of old timber from demolition (Part 4), Wood Industry (Mokuzai Kougyo), Vol.
37, No. 2. PP. 647.
Building Research Institute (1986) Report of research on reuse technology of waste material for construction (in
Japanese), Building Research Institute, Japan.
Kelly, M.W. (1977) Critical literature review of relationships between processing parameters and physical properties of
particleboard, U.S.D.A. FPL-Report FPL-10.

20
DEVELOPMENT OF FORMWORK MATERIAL MADE OF
SCRAP LUMBER
T.ONO, S.ONO and T.KUMANO
Civil Engineering Division, Shimizu Corporation, Tokyo, Japan
T.SANO
Tokyo Branch, Shimizu Corporation, Tokyo, Japan
Y.MUKAWA
Technology Development Division, Shimizu Corporation, Tokyo, Japan

Abstract
The crisis facing the worlds tropical rainforests and the need to preserve them are widely acknowledged
around the globe. To support the movement for preventing destruction of the Earths valuable forests, we have
developed an alternative material to lauan and conifer plywood for concrete forms. This material basically
consists of particle board made of chips of lumber-recycled material.
To provide this form material with the required rigidity, meshed glass fibers are added to the particle board
as reinforcing material. By applying special resin to the surface of the material and cutting the glass fibers as
required, rigid forms like plywood can be obtained.
Raw lauan and conifer lumber has been conventionally used to make plywood forms. However, the
development of this new form material obviates the requirement to use such resources, leading to the
preservation of large tracts of forest.
Key Words: Alternative Material To Plywood, Conifer Plywood, Flexible Reinforcement, Forest
Preservation, Glass Fiber, Lauan Plywood, Particle Board, Scrap Lumber.
1
Introduction
World concern over the destruction of tropical rain forests has become significant. In Japan too, measures to reduce
the amount of tropical timber are being rapidly adopted. In the construction industry, 100 million sheets of lauan
plywood are being used annually to make concrete formwork. In order to save these forests from destruction and
protect the Earths natural environment, a positive effort is being made to deal with the problem and research is being
carried out on alternative materials; new methods using alternatives are being developed and the effective use of
recycled materials is being established.
The authors decided to focus their attention on using recycled material, developing a particle board made out of chips
from scrap lumber. In order to impart the necessary degree of rigidity to this material, glass fiber reinforcement was
developed. Its effective use to make form material (referred to as chip form below) contributes to the protection of
forests because it obviates the need to use materials derived from lauan or conifer.
This report describes a study of the application of recycled particle board in the making of formwork.

Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

178

ONO, ONO, KUMANO, SANO AND MUKAWA

2
Particle Board Survey
Particle board, developed with the aim of making effective use of timber, has been in use in Japan for over thirty five
years. As a construction material it is manufactured to the Japanese Standard JIS A 5908 and the recent domestic
annual production is around 1 million m2 [1].
2.1
Raw Materials and Manufacture
The raw material for the manufacture of particle board is scrap material obtained, for example, from the leftovers
from sawmills, house demolition, forest thinning and felling, and plywood production. These scrap materials are
broken up into small pieces (chips), sprayed with adhesive and then shaped at high temperature and under high
pressure.
In Figure 1 raw material can be seen, and in Figure 2, the chips.
2.2
Classification
Particle board is classified according to its flexural strength and the kind of adhesive used in its manufacture [2]. M
type, mainly melamine resin, and P type, mainly phenolic resin, are the two main types of adhesive generally used.
2.3
Application
Particle board is used in, among other things, the construction industry, furniture manufacture and for electrical
equipment. Among these, the construction industry alone consumed around 30% of the 1992 total consumption [3].
As a result of increased durability and price stability there has been an increase in demand in recent years. It is being
used for such applications as the flooring or walls of prefabricated houses, and in the dry type sound proof flooring
method used in building condominiums or apartments and gymnasiums.
3
Proposal for Using Chip Formwork Made From Scrap
3.1
Characteristics of Particle Board Used in the Construction Industry
The main particle board used in the construction industry is made using phenol resin adhesive (P type), and its
bending strength in both the longitudinal and width directions is more than 180 kgf/cm2 (200 type). It is the strongest
and the most durable.
It was with attention to this construction industry particle board that the authors began development of a new form
material.
Table 1 shows a comparison of the bending strength characteristics of lauan plywood and construction industry
particle board. From the table it can be seen that the flexural strength and Youngs modulus in flexure are less for the
particle board than for the lauan plywood, it is therefore believed to be difficult to use the material for forms in that
condition.
In order to use particle board as the basic material for forms it was believed necessary to increase its flexural
strength and Youngs modulus in flexure.
3.2
Increase in Flexural Strength and Young's Modulus
(1) Selection of Material for Reinforcement
Various reinforcement methods based on the principal of concrete reinforcement were examined. After
consideration of the requirements below, glass fiber was decided upon as the reinforcement material.

FORMWORK MATERIAL MADE OF SCRAP LUMBER

179

Fig. 1 Examples of Scrap Timber

Fig. 2 Wooden Chips

i) It must have a higher tensile strength and Youngs modulus than the parent particle board
ii) It must be light
iii) It must be capable of being sawn
iv) It must be economically viable
Remark: It was decided to insert the glass fibers near the surface, where the tension due to bending during concrete
placing will be highest.
Figure 3 is a schematic drawing of the chip form.
(2) Study of Methods to Improve Characteristics
The following properties are required of the glass fiber:
i) There must be no discontinuity between the surface layer of the particle board and the original particle board
material.

180

ONO, ONO, KUMANO, SANO AND MUKAWA

Fig. 3 Shematic Drawing of Proposed Form

ii) The force must be positively transferred via the adhesive applied to the surface of the chips.
iii) It must be of adequate thickness to resist the tension during bending.
Table 1 Mechanical Properties of Particle Board and Plywood
Flexural strength (kgf/cm2)
Youngs modulus in flexure (104kgf/cm2)
Notes) Based on the sampling data of O manufacturer.

Particle board

Lauan plywood

200~230
3.0~3.5

465
4.5

As a result of the above considerations, a strand that has a Youngs modulus in the region of 7.0105 kgf/cm2, and
tensile strength of around 70kgf, woven into mesh form was chosen
It is only when the force is positively transferred, as described in ii) above, that the ability of the glass fiber to
resist the tensile load is fully exploited. For this reason the fibers were impregnated with an adhesive of the same
phenol group as the parent material and in addition it was processed using a silane additive. In this way a covalent
bonding of the parent particle board adhesive with the adhesive applied to the fiber and the fiber itself is achieved,
resulting in a composite of the three. In this case the fiber is not simply physically adhering to the parent board but is
actually forming a covalent bonding at the molecular level, so the resulting combination of particle board and its
reinforcement (glass fibers) is greatly improved leading to a considerable increase in the bending strength
characteristics of the parent board.
4
Testing of the Chip Form
4.1
Outline of the Tests
In order to confirm achievement of the necessary properties of the chip form and to compare it with lauan plywood
and particle board, the various tests shown in Table 2 were performed. The conditions of the bending test are shown
in Figure 4.
The procedure for performing the nail extraction resistance test and the percentage weight increase with water
absorption are not specified in any JIS (Japanese Industrial Standards). For this test a drivescrew was driven into a
12mm thick board and the maximum load during withdrawal was measured. For the percentage weight increase, the
weight of the test piece was measured at the time of testing for percentage of widthwise expansion due to water
absorption.
The suitability of chip form for sub-flooring and flooring applications was also thought worthy of consideration. In
this case resistance to impact is of prime importance, for this reason test specimens, as shown in Figure 5, were
prepared and impact tests were carried out.
For the impact testing, with reference to the Housing and Urban Development

FORMWORK MATERIAL MADE OF SCRAP LUMBER

181

Fig. 4 Bending Test

Fig. 5 Impact Test Model


Table 2 Outline of Test
Test item

Test method

Density
Water content
Flexural strength
Youngs modulus in flexure
Flexural strength when damp
Peeling resistance
Wooden screw tenacity
Pegs resistance against pulling
Hygroscopic thickness swelling
Water absorption
Formaldehyde emission

JIS A 5908 particleboards3)


5.3 Water content test
5.4 Flexural strength test

5.2 Density test

5.5 Flexural strength when damp test


5.6 Peeling resistance test
5.7 Wooden screw tenacity test
JIS A 5908 Particleboards
JIS A 5908 Particleboards

5.8 Hygroscopic thickness swelling test

5.9 Formaldehyde emission test

Corporation test guidelines, a 30kgf sandbag was dropped from a height of 50cm and 100cm respectively. This
test is shown in Figure 6.
4.2
Results and Considerations
(1) Bending Strength Characteristics
Figure 7 shows the results of the bending tests. The results confirmed that the glass fibers of the chip form
improve the bending strength and Youngs modulus in flexure to about the same level as lauan plywood. In addition,
it was found that on exceeding the maximum load, the failure was not by sudden destruction, the reinforcing material
effectively imparts toughness to the material.
Figure 8 shows the dispersion in the experimental values of flexural strength and Youngs modulus. It was
confirmed that the dispersion in the flexural strength results was less for the chip form than for the lauan plywood,
and in the case of the Youngs modulus in flexure, the dispersion was about the same for the two materials.

182

ONO, ONO, KUMANO, SANO AND MUKAWA

Fig. 6 Impact Test

Fig. 7 Improved Flexural Strength

The raw material for the particle board is scrap timber that has been cut into chips, and its final properties are to a
large extent due to the nature of the adhesive used to bond the chips together, it is believed that for this reason the amount
of dispersion of values is relatively small. In the case of lauan plywood on the other hand, the raw material is natural
wood, so the dispersion in strength is relatively large but the wood fibers give only little dispersion in elongation
results.
(2) Various Properties
Table 3 shows the test results for various properties. Even though the density of the chip form material is three
times that of lauan plywood, the separation strength, warping strength, and nail withdrawal resistance properties
were all found to be about the same as for the lauan plywood. Also the amount of formaldehyde released (the
amount of vaporized formalin, a solvent used in the adhesive), was found to be less than 1/40 of the amount released
by the lauan plywood.
(3) Impact Resistance
The results of the impact tests are shown in Table 4. Chip form of 12mm thickness was seen to have an impact
strength exceeding that of 25mm thickness general construction use particle board.
It is believed that the superior strength of the chip form is derived from the ability of the glass fiber mesh
reinforcement to distribute the load.

FORMWORK MATERIAL MADE OF SCRAP LUMBER

183

Fig. 8 Dispersion Values of Flexural Strength and Youngs Modulus in Flexure


Table 3 Test Results
Thickness 12 mm
Chip form material*1 Lauan plywood *2 Particle board type 200*3 Remarks3)*4

Item
Standard type
(g/cm3)

Density
Water content (%)
Flexural strength (kgf/cm2)
Youngs modulus in
flexure (104 kgf/cm2)
Flexural strength when
damp (kgf/ cm2)
Peeling resistance (kgf/
cm2)
Wooden screw tenacity
(kgf)

Water resistant type


0.75
7.5
450
4.5

0.58
8.5
465
4.5

0.71
8.0
220
3.5

0.74
7.7
235
3.6

280

290

120

130

0.50~0.90
5~13
180 or more
3.0 or more for lateral
direction (reference value)
90 or more

7.0

5.5

5.0

7.8

3.0 or more

75

50

60

70

50 or more (to be applied to


a thickness of 15 mm or
more)

Pegs resistance against


40
35

pulling (kgf)
Hygroscopic thickness
5.0
3.5
10
4.1
swelling (%)
Water absorption (%)
25.0
30.0

Formaldehyde emission
0.2
8.0
0.2
0.1
(mg/ )
Notes) * 1 Based on the flexural test data.
* 2 Based on the sampling data of lauan plywoods of five manufacturers, O, SA, SI, K and I.
* 3 Construction particle board by O was used.
* 4 From JIS A 5908 Particleboards

12 or less

5 or less

Table 4 Impact Resistance (Residual Deflection (mm) and Cracking by Impact Test)
50cm

100 cm

Free at both ends*1

Crackin
g ( **
When
cracked
)

Free at both ends*1

Free at both ends*1

1st

2nd

3rd

1st

2nd

3rd

1st

2nd

3rd

Chip form material 12 mm


No. 1 specimen

0.5

*2

No
crackin
g

184

ONO, ONO, KUMANO, SANO AND MUKAWA

50cm

100 cm

Free at both ends*1

Crackin
g ( **
When
cracked
)

Free at both ends*1

Free at both ends*1

1st

2nd

3rd

1st

2nd

3rd

1st

2nd

3rd

No. 2 specimen

0.05

0.05

0.05

0.15

Particle board *3 20 mm
No. 1 specimen

0.15**

No. 2 specimen

0.05**

Particle board*3 25 mm
No. 1 specimen

0.20**

No. 2 specimen

0.30**

0.15

No
crackin
g
Cracki
ng
Cracki
ng
Cracki
ng
Cracki
ng

Notes) * 1 Free at both ends means both ends of a specimen are free from anything, while fixed at both ends a specimen
whose both ends are fixed by a mans weight.
* 2 indicates that a specimen was visually monitored for cracking, without measurement of deflection.
* 3 Type 200 as stipulated in JIS A 5908 Particle boards was used.

5
Conclusions
This research was carried out to assess the effectiveness of using glass fibers to enhance the flexural strength of
particle board for use as chip form. The results obtained can be summarized as follows:
i) Flexural strength characteristics are about the same as those of lauan plywood.
ii) Density of the chip form material is three times that of lauan plywood.
iii) Separation strength, warping strength, and nail withdrawal resistance are about the same as for lauan plywood.
iv) Chip form with a thickness of 12mm has an impact strength exceeding that of 25mm thickness general
construction use particle board.
v) Chip form demonstrates the qualities required to make it a suitable alternative to lauan plywood for use in both
concrete formwork and as sub-flooring and flooring material for use in building construction.
6
A Word of Gratitude
This research is the result of the joint efforts of four companies, Shimizu, Onahama Gouhan, Yoshimoto Ringyo and
Nitto Boseki, and the authors would like to extend their thanks to all those involved in this work.
7
References
1.
2.
3.

Ministry of International Trade and Industry (MITI). (1992) Building Materials Statistics.
Japanese Industrial Standard JIS A 5908 (1994) Particle boards.
Japan Fiber Board and Particle Board Manufacturers Association. (1992) Timber and Board Manufacturing Data.

PART FIVE
PAPER-BASED MATERIALS

21
ECOLOGY BOARDS USING RECYCLED PAPER RESOURCES
FOR CONCRETE FORMS
K.KURIHARA and S.TAKATA
Technical Research Institute, Nishimatsu Construction Co. Ltd, Yamato, Kanagawa-ken, Japan
Y.TOMIMURA and S.HOSOYA
Forestry and Forest Products Research Institute, Kukizaki-machi, Ibaraki-ken, Japan

Abstract
For the purpose of protecting the global environment, substituting materials for the plywood from the tropical
rainforests as concrete forms have been studied. It was confirmed that recycled paper resources could be
regenerated to Ecology Board which has enough properties to be substituted for the conventional plywood.
Among the recycled paper resources, newsprints and magazine papers which are mainly made from high-yield
pulps were found to be the best materials since those pulps have similar chemical components to original wood
itself.
First of all, recycled papers were defibrated by the dry defibration process using FS mill developed by Kobe
Steel Co., Ltd. and VMC Co., Ltd.. Defibrated fibers were then oxidized by gaseous ozone to improve the
adhesion properties of the fiber. This is the key step to give the enough strength properties to Ecology Board.
Ozonized fibers were mixed with adhesives and then hot-pressed to form Ecology Board.
Thus manufactured Ecology Board has similar properties to the conventional plywood although the strength
properties depend on the density of the board. It was confirmed that the Ecology Board at the density 0.85 g/
cm3 had enough strength properties for the concrete forms. It was also proven that the strength properties of the
board could be kept almost unchanged during four times repeated use as concrete forms.
Keywords: Concrete form, dry process defibrator, lignin, ozone treatment, tropical rainforest, wastepaper.
1
Introduction
It was reported that in tropical regions deforestation rates have been 1020 times greater than reforestation in recent
years [1]. The world forest resources are decreasing at the rate 0.2% a year whereas the tropical forest is lost at the rate
of 1% a year [2]. Especially, in Southeast Asia wood production has increased rapidly in recent years and the timber
resources are declining. Thailands forests have been so depleted that they are importing timbers. Philippines is
heading toward a similar situation. Malaysia is cutting back its logging [3]. Thus, the forest products industry in
Southeast Asia is in a transitional period and the destruction of the tropical rainforests is of great concern to many
people.
Since the construction industry is one of the big consumers of the plywood as concrete forms which are originally
coming from the tropical rainforests, environmental problems concerning tropical rainforests are very important and
urgent for the industry. Considering the substituting materials for the plywood as concrete forms, what came up is
the recycled paper resources which are also forest products.

Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0419 20550 0.

ECOLOGY BOARDS FOR CONCRETE FORMS

187

Fig. 1 Ultraviolet micrograph of newsprints fibers (280nm).

Since paper is made from wood, recycling of the paper is also very important for not wasting forest products and
for protecting the global forest environment. In Japan, recycling of paper is very advanced compared to other
countries. More than fifty percent of the paper and paper board are already recovered and recycled. Japanese pulp
and paper industry has a plan to achieve 55% recycling of the paper and paper boards by the end of this year. But, owing
to the present economical conditions in Japan, it seems very difficult to reach the goal within this year. Right now
there seems to be excess supply of the recovered paper and paper boards for the pulp and paper industry. In order to
achieve the incremental recovery of the paper, we need to find a new utilization of recycled paper other than paper
and paper boards. The objective of this paper is to make use of the recycled paper resources for concrete forms and
contribute to the preservation of tropical rainforests. Another merit of this new approach is to reduce the urban trash
which otherwise goes to incineration or landfilling.
2
Materials and methods
2.1
Wastepaper resources
Among the wastepaper resources preliminarily tested, newsprints and magazine papers showed better results. These
wastepapers are mainly made from so called high-yield pulps and contain a lot of lignin in the fibers. In other words,
the chemical components of high-yield pulps are very similar to wood itself, whereas office wastepapers such as
copy paper are well delignified and do not contain lignin in the fibers. The structures of wood are often compared to
RC structures, where cellulose molecules are steel and lignin is concrete. This is very interesting to see that low
grade lignin containing wastepapers are better materials for the Ecology Board to be used for concrete forms.
Another merit of using lignin containing fibers is that lignin in the fibers could be chemically modified so as to
perform better adhesion. As illustrated in the ultraviolet micrograph of defibrated newsprints (fig. 1), lignin locates
on the surface of fibers as well as in the cell walls. In the micrograph, the black portion shows the ultraviolet light
absorbing substance which is lignin in the fibers. The lignin on the surface of fibers could contribute to the fiber to
fiber bonding when it is chemically modified. Thus, through this study, newsprints wastepapers were selected as raw
materials for Ecology Board.
2.2
Dry process defibration
Newsprints were first cut to the size 123 mm by the use of a commercial shredder and then subjected to FS mill
(Fibsonic mill) which was developed by Kobe steel Co., Ltd. and VMC Co., Ltd.. Going through this procedure,
newsprints wastepapers were well defibrated to single fibers. In the pulp and paper industry, wastepapers are usually
defibrated by the wet process. But, in this study the dry defibrating process was applied so as not to produce any
polluted effluents.

188

KURIHARA, TAKATA, TOMIMURA AND HOSOYA

2.3
Ozonization of fibers
Oxygen gas was used for ozone generation. The ozonizer used was Yamanashi High-Tech SA-100PV. Oxygen flow
was 1.5 1/min, and the ozone concentration was 34%. Amounts of ozone applied onto fibers were 0.252.00% on
fibers by weight. About 1 kg of fibers were put into a plastic bag and an appropriate amount of ozone gas was
introduced into the bag. Ozone reacted with lignin instantaneously.
2.4
Manufacturing of Ecology Board
Ozonized fibers of 100130 g were mixed with powdered phenol resin (PR-311 or PR-11078, Sumitomo Bakelite
Co., , Ltd.) Addition of the resin was 1520% on fibers. The fibers were then smoothly set in a 2521 cm form and prepressed between aluminum plates with Teflon sheets. The pre-pressed mat of fibers was hot-pressed at 200C for 3
minutes with a 2 mm spacer between aluminum plates.
2.5
Determination of board properties
Board properties determined were modulus of rupture (MOR), modulus of elasticity (MOE), swelling and water
absorption after 24-hour water soak. Testing was done according to Japanese Industrial Standard (JIS) 5906 except
that 130 25 mm bending specimens were tested over a 50 mm span.
2.6
Manufacturing of concrete forms
For the concrete placing tests, Ecology Board for the real size concrete forms was manufactured as a pilot plant scale
test by Kobe Steel Co., Ltd.. The size of the board was 90 90 cm with a thickness of 12 mm. The manufacturing
procedures were basically the same as those for 2 mm size Ecology Board described above. But, in this case, liquid
melamine resins were used as adhesives, and more resins were distributed at the surface layers of the boards than
core layers in order to obtain the higher Youngs modulus.
2.7
Durability test
Durability during four times repeated concrete placing was tested with 12 mm thickness Ecology Board for forms.
After each run of concrete placing, a part of the boards was subjected to the determination of board properties. The
other boards were air-dried for a few days and then used for next run of concrete placing.
3
Result and discussion
3.1
Effect of ozone treatment on board properties
The objective of the ozone treatment is to increase the numbers of functional groups in the lignin on the surface of
defibrated fibers. Increased numbers of functional groups are expected to contribute to the better adhesion properties.
It is well-known that the lignin has excellent reactivities toward ozone [4]. The main reactions of lignin with ozone
are illustrated in figure 2. The original aromatic rings are oxidatively cleaved and thus carboxylic acid groups are
introduced to the lignin, which can contribute to the fiber-fiber bonding.
In this experiment, the dosage of ozone on fibers was varied 0.252.00% by weight. Figure 3 and 4 show the
effect of ozone treatment on MOR and MOE respectively. The best effect was observed at the ozone dosage around
0.25 0.50%. Too much ozone treatment caused negative effects. This is probably because of the damages of the
fibers caused by the reaction of cellulose with ozone. Cellulose is much less reactive with ozone than lignin but it
does react with ozone. Accordingly, if too much ozone is charged onto the fiber, cellulose starts reacting with ozone
and causes fiber damages. Thus, more than 0.50% ozone gave negative effects on the fiber strength.

ECOLOGY BOARDS FOR CONCRETE FORMS

189

Fig. 2 Reactions of lignin with ozone

Fig. 3 Relationship between ozone charge and modulus of rupture.

Fig. 4 Relationship between ozone charge and modulus of elasticity.

3.2
Effect of ozone treatment on swelling and water absorption
Ecology Board was soaked in water for 24 hours and the thickness swelling and the water absorption were
determined. These are the important properties for the use of concrete forms. The effects of the ozone treatment on
the thickness swelling and the water absorption are shown in figure 5 and 6 respectively. Again the swelling and the
water absorption were minimum at the ozone dosage around 0.250.50%. The explanation for this results is that the
appropriate ozone charges made the fiber bonding better and thus the swelling and the absorption became

190

KURIHARA, TAKATA, TOMIMURA AND HOSOYA

Fig. 5 Relationship between ozone charge and thickness swelling in 24 hr water soak test.

Fig. 6 Relationship between ozone charge and water absorption in 24 hr water soak test.

minimized. But, excess ozone charges made the fibers too hydrophylic because of the presence of carboxylic acid
groups and then the swelling and the water absorption increased.
3.3
Effect of undefibrated paper particles on board properties
In this study, the defibration is one of the key steps for the good board performance. In order to see the effect of the
ratio of undefibrated paper particles (incomplete defibration) on the board properties, shredded paper particles (123
mm) before going through defibration process were artificially mixed with completely defibrated fibers, and the
boards were manufactured. Figure 7 and 8 show MOR and MOE of the boards respectively.
It is clearly shown in the figures that the less the undefibrated paper particles were, the higher the MOR and MOE
were. The thickness swelling and the water absorption after 24 hours water soak test are shown in figure 9 and 10
respectively. It is again clearly illustrated in the figures that the swelling and the water absorption were smaller when
the boards had less undefibrated paper particles. From the results shown in the figures 710, it can be said that the
complete defibration is one of the key steps for the best boards manufacturing.

ECOLOGY BOARDS FOR CONCRETE FORMS

191

Fig. 7 Effect of undefibrized paper particles on modulus of rupture.

Fig. 8 Effect of undefibrized paper particles on modulus of elasticity.

3.4
Durability during repeated concrete placing
Durability of the Ecology Board during four times repeated use for the concrete placing was tested with 12 mm
thickness boards. The concrete placing test was also done with the commercial plywood forms. The properties of
both Ecology Board forms and the plywood forms were compared with each other before and after each concrete
placing run. Figure 11 and 12 show MOR and MOE of the board together with those of the plywood respectively.
As is shown in figure 11, MOR slightly decreased during the repeated use, but there were not much differences
between the Ecology Boards and the plywood. Figure 12 shows MOE of the Ecology Board and the plywood. MOE
also decreased slightly as the times of recycling increased. The plywood showed higher MOE. From this recycling
test, no serious problems were observed in terms of durability. Some properties of the concrete surface such as
smoothness, hardness and color were also checked. No inferior results were observed for the Ecology Board forms
compared to the plywood forms used this time.
3.5
Further improvement of the board properties
The density of the board used for the concrete placing test this time was 0.95 g/cm3. When the nailing or cutting
properties and the weight of the forms are considered, this density is too high. But, if the density is lowered the
strength properties also become lower, so that this is very important to manufacture light and strong boards. This is

192

KURIHARA, TAKATA, TOMIMURA AND HOSOYA

Fig. 9 Effect of undefibrized paper particles on thickness swelling in 24 hr water soak test

Fig. 10 Effect of undefibrized paper particles on water absorption in 24 hr water soak test.

under way right now. The key processes are the appropriate ozone treatment, the higher adhesives distribution on the
surface layers and complete defibration of the paper particles. We have so far achieved 40 MPa (MOR) and 4.5 GPa
(MOE) at the density 0.85 g/cm3. There are no nailing or cutting problems. We are now planning the concrete
placing test for the real size construction.
4
Conclusion
Recycled paper resources, especially newsprints, can be successfully regenerated as Ecology Board for concrete
forms. The strength properties of Ecology Board so far achieved are more than 40 MPa (MOR) and 4.5 GPa (MOE)
at the density 0.85 g/cm3. This board has no nailing or cutting problems in using for concrete forms. Ecology Board
can be used repeatedly at least four times and there are no notable deteriorations during four times repeated use.
5
References
1.
2.
3.

World Resources. (1987) Basic Books, Inc., New York.


Forestry White Paper. (Fiscal Year 1991) Forestry Agency, Japan.
Elvin T.Choong, Rubini Atmawidjaja and Suminars Achmadi. (1993) The forest products industry Asia: An emphasis on
Indonesia. Forest Products Journal, Vol. 43, No. 5. pp. 4452.

ECOLOGY BOARDS FOR CONCRETE FORMS

193

Fig. 11 Effect of recycling test on MOR

Fig. 12 Effect of recycling test on MOE


4.

Matsumoto, Y. and Hosoya, S., New Technologies of Ozone Utilization. (1993) Section 13. pp. 497536. Sanyu
Publishers, Tokyo.

22
RECYCLING OF USED PAPER AS A BUILDING MATERIAL
H.MIHASHI, K.KIRIKOSHI and S.ARIKAWA
Department of Architecture, Faculty of Engineering, Tohoku University, Sendai, Japan
T.YAMAMOTO
Tohoku Pole Company, Inc., Sendai, Japan
T.NARITA
Tohoku Electric Power Company, Inc., Sendai, Japan

Abstract
The present paper shows an idea to utilize used paper as a building material. A kind of artificial lightweight
aggregates was developed in which a large amount of fly ash from coal fire electric power plants and used
paper produced from office automation machines. Since strength properties of the cementitious composite
material containing the lightweight aggregates are poor, improvements of the undesirable properties were
studied by fiber reinforcements.
Keywords: Used paper, cementitious composite, fiber reinforcement, sandwich plate.
1
Introduction
Although the amount of by-products such as used paper from office automation machines increases very rapidly,
quite few technologies to utilize them have been developed. Especially the huge amount of used paper produced in
banks and other offices is a waste to be burnt out to keep the printed secret matters. Generally the huge amount of
used paper causes a big problem in urban waste disposal facilities. On the other hand, saving forests in many
countries is desired from the view point of environmental problems in the world.
The present study tried to develop cementitious composite materials for buildings with a kind of artificial
lightweight aggregates which was produced by mixing a large amount of fly ash and shredded used paper. Potential
weak points of this new material are poor strength, high water absorption which may cause large shrinkage
deformation, rather poor fire resistance and resistance to frost damage [1].
The aim of this paper is to show an idea to improve the strength properties by fiber reinforcement.
2.
Experiment Program
2.1
Lightweight artificial aggregates including used paper
By mixing a large amount of fly-ash from coal fire electric power plants with a small quantity of Portland cement
and water, Yamamoto et al. developed artificial aggregates (coal-ash sand) [1], [2]. Various types of coal-ash sand
can be made by changing the mix proportion, turning speed of the mixer and also adding some fibers. Lightweight

Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

USED PAPER AS A BUILDING MATERIAL

195

artificial aggregates including used paper is one of the extension of the coal-ash sand. Hereafter it is called pulp
sand.
Pulp sand were made with high-early-strength Portland cement, shredded paper used in copying machines, fly ash
and a small amount of water mixed by means of Irich mixer. After mixing, produced particles were cured in a moisture
room for about 10 hours whose temperature is 80 C. The most desirable point of this sand is that it is very light and
insensitive to the type of paper. The specific gravity on oven-dried condition is 0.73. The most undesirable property
as a building material may be much higher absorption than usual one, which causes a large shrinkage deformation.
The water absorption ratio is 81.5%.
2.2
Production of core plates with pulp sand mortar
Two types of pulp sand mortar were used as the core plate of sandwich specimens. Mortar of Type M was made by
mixing high-early-strength Portland cement, silica fume and pulp sand. Mortar of Type P was made directly with
pulp sand. Mix proportions of each type of mortar are shown in Table 1 and Table 2. Pressurized forming was used
to produce mortar plates of each type, whose pressure was 10MPa. Then specimens were put in a moist room of 20 C
for about one day. After steam curing at 50 C for 24 hours, they were autoclave-cured on condition of 180 C for 3
hours. Finally they were dried out in an electric oven of 110 C for 24 hours. The geometry of the specimen was
5010200 (mm).
Table 1 Mix proportion of pulp sand mortar (Type M) per batch
Series

Water/Binder

Sand/Binder

Water (g)

Cement (g)

Silica Fume (g)

A
1.12
1.00
253
134
27
B
1.52
1.50
257
100
20
C
1.93
2.00
259
79
16
* Oven dried one whose mix proportion was same as Series I in Table 2,

Silica Sand (g)

Pulp* Sand (g)

66
49
39

227
254
268

Table 2. Mix proportion of pulp sand mortar (Type P) (weight ratio)


Series

Fly ash

Cement

Silica sand

Silica fume

Used paper

I
II
III

5
2
4

2
2
2

0.0
2.5
1.5

0.0
0.5
0.5

3
3
2

2.3
Sandwich plates
Sandwich specimens were made with core plates of pulp sand mortar reinforced by glass fiber mesh and aluminum
sheet as shown in Fig. 1. Thickness of the aluminum sheet was 15 micrometers. Tensile strength
Table 3. Mechanical properties of Type M pulp sand mortar plates
Series

Specific gravity

Flexural strength (MPa)

Shear strength (MPa)

Youngs modulus (GPa)

A
B
C

1.26
1.19
1.13

8.35
6.20
5.63

4.06
3.51
3.18

15.77
8.58
3.51

Table 4. Mechanical properties of Type P pulp sand mortar plates


Series

Specific gravity

Flexural strength (MPa)

Shear strength (MPa)

Youngs modulus (GPa)

I
II
III

0.81
0.97
1.01

0.62
3.21
3.08

1.84
3.80
4.52

0.45
1.47
2.20

196

MIHASHI, KIRIKOSHI, ARIKAWA, YAMAMOTO AND NARITA

Fig. 1. Composition of sandwich plate.

Fig. 2. Test set up for bend test

Fig. 3. Double shearing test

of the glass fiber mesh (Japan Electric Glass Co., AR2500H-200) was about 14.5 MPa. These reinforcements were
glued with the core plate by an adhesive agent of phenol resin (Ohshika Resin D-17). These sandwich plates were
hot-pressed for 2 minutes on a condition of 145 C and 1 MPa after pressing for 12 hours by a weight at room
temperature.
2.4
Test methods
Flexural strength of core plates and sandwich plates were determined by three point bend test as shown in Fig. 2. The
span was 150 mm. Deflection of the center of the span was measured with a laser displacement censor. Vertical
movements of the plate on the supporting points were also measured with displacement transducers which were used
to delete the rigid body displacement from the measured deflection. The applied load was measured with a load cell.
Shear strength of core plates were determined by double shearing test as shown in Fig. 3.
3.
Experimental Results and Discussion
3.1
Mechanical properties of core plates
Test results of flexural and shear strength of core plates are shown in Table 3 (Type M mortar) and Table 4 (Type P
mortar). All of the results are the mean values obtained with three specimens.
As the content of pulp sand is increased in case of Type M mortar, the specific gravity is reduced but the strength
is decreased. Especially the flexural strength is more sensitive than the shear strength. As it is clearly shown in the
relative strength per unit specific gravity, effectiveness to increase the pulp sand content is poor from the view point
of strength properties.

USED PAPER AS A BUILDING MATERIAL

197

Fig. 4. Load vs. deflection curves of core plates (Type M mortar)

Although the specific density of Type P mortar is lower than that of Type M mortar, the strength is much lower
especially in the flexural one. Series I mix proportion lead to an extremely lower strength in comparison with other
two series. It may mean that silica sand and silica fume are essential for this curing condition. Typical load-deflection
curves of each mix proportion are shown in Figs. 4 and 5. Descending portion of Type M mortar was generally very
steep and the deformability was very poor. On the other hand, the behavior of Type P mortar was apparently ductile
though the strength was much lower than that of Type M.
3.2
Mechanical properties of sandwich plates
Experimental results of sandwich plates are shown in Table 5. Series B, C, II and III performed the flexural strength
more than 8 MPa. Some correlations were expected between the strength of core plates and that of sandwich plates.
However, sandwich plates of Series A whose core plate was the strongest in the flexural strength were weaker than
those of Series B and C. The strongest one was Series III, though the Youngs modulus was much lower than those
of Series A, B and C. Toughness (in this case, total energy absorption capacity) was measured from the area under
the load-deflection curve divided by the section area.
The load-deflection curves of sandwich plates are shown in Figs. 6 to 11. In case of Series A, B and C, a small
plateau at about 2025 kgf was observed and the behavior was scattered. On the other hand,
Table 5. Mechanical properties of sandwich plates
Series

Specific gravity

Flexural strength
(MPa)

Youngs modulus
(GPa)

Toughness
(Nm)

A
B
C
I
II
III

1.27
1.19
1.14
0.93
0.97
1.05

7.55
8.41
8.20
3.87
8.22
9.39

6.52
4.93
4.46
1.21
2.26
2.81

0.16
0.29
0.37
0.11
0.57
0.64

load-deflection curves of Series II and III were less scattered and the behavior was more ductile than those of Series
A, B and C. As the result, the toughness of Series II and III was much larger than that of Series A, B and C. Only the
curves of Series I were very different from those of other series.
In Fig. 12, relation between the flexural strength of core plates and that of the corresponding sandwich plates. The
flexural strength of sandwich plates are absolutely independent of that of the core plates. Fig. 13 shows the relation
between the shear strength of core plates and the flexural strength of sandwich plates. In this figure, a strong-

198

MIHASHI, KIRIKOSHI, ARIKAWA, YAMAMOTO AND NARITA

Fig. 5. Load vs. deflection curves of core plates (Type P mortar)

Fig. 6. Load vs. deflection curves of sandwich plates (Type M mortar, Series A).

correlation is observed. It means that increasing the shear strength of core plates may be the key point to produce
light weight but rather high strength sandwich composites.
As the failure mode is concerned, only the weakest Series I was broken by the shear failure of core plates but
others were broken by the tensile failure of glass fibers. Therefore a certain higher flexural strength of sandwich
plates is expected if the reinforcing fiber is stronger.
4
Conclusions
Artificial light weight aggregates were developed with used paper. As an example to overcome the weak strength
property, sandwich plates reinforced with glass fiber mesh was demonstrated. Since the shear strength of core plates
is one of the key factor, it is expected to produce stronger plates with some additional reinforcements of chopped
fiber in the core plate and stronger meshes.
5

USED PAPER AS A BUILDING MATERIAL

199

Fig. 7. Load vs. deflection curves of sandwich plates (Type M mortar, Series B).

Fig. 8. Load vs. deflection curves of sandwich plates (Type M mortar, Series C).

References
1.
2.

Yamamoto, T., Mihashi, H. and Hirai, K. (1991) Advanced utilization of fly-ash as artificial aggregates, Waste Materials in
Concrete, (ed. J.J.J.M.Goumans, H.A.van der Sloot and Th.G. Aalbers), Elsevier, pp. 621622.
T.Narita, H.Mihashi, K.Hirai and T.Yamamoto (1991) Experimental studies on frost damage properties of fly-ash sand
mortar reinforced with fiber, Concrete Research and Technology, Japan Concrete Institute, vol. 2, No. 1, pp. 6775 (in
Japanese).

200

MIHASHI, KIRIKOSHI, ARIKAWA, YAMAMOTO AND NARITA

Fig. 9. Load vs. deflection curves of sandwich plates (Type P mortar, Series I).

Fig. 10. Load vs. deflection curves of sandwich plates (Type P mortar, Series II).

USED PAPER AS A BUILDING MATERIAL

Fig. 11. Load vs. deflection curves of sandwich plates (Type P mortar, Series III).

Fig. 12. Relation between flexural strength of sandwich plates and that of core plates.

201

202

MIHASHI, KIRIKOSHI, ARIKAWA, YAMAMOTO AND NARITA

Fig. 13. Relation between flexural strength of sandwich plates and shear strength of core plates.

23
USE OF PAPER SLUDGE ASH IN CONCRETE PRODUCTS
Y.S.SOH and S.Y.SOH
Department of Architectural Engineering, College of Engineering, Chonbuk National University,
Chonju, Korea
D.S.LEE
Department of Chemical Technology, College of Engineering, Chonbuk National University,
Chonju, Korea

Abstract
Ashes from incinerated paper industries sludges were studied for their potential use in cement concrete
interlocking block. Addition of ashes induced an increase of the water/cement ratios and moisture absorption
and a decrease of the flexural strength of cement mortar or concrete. However, such disadvantages could be
overcome by adding polymer latex and the interlocking blocks containing ashes met the Korean Standard.
Keywords: Paper sludge ash, interlocking block, polymer latex, W/C ratio, moisture absorption, flexural
strength.
1
Introduction
The production of sludges from various waste water increases and the cost of their treatment constitutes more than
half of the cost of the general treatment of waste water. One cost effective treatment of sludges is land-fill. But, areas
for land-fill are limited in Korea and underground water pollution if induces is worrisome. Thus, the incineration of
sludges is being increased. Proper incinerators generally reduce the volume of the sludges by 90%. However, sludge
incineration generates a large amount of ashes. In our province in Korea, the amount of sludges generated annually
was 373,000 metric tons in 1990. More than 2/3 of those were generated by paper industries. The most of part of the
paper industries in Chonlabuckdo are incinerating sludges. Thus, The use of the ashes from the paper sludges is
necessary and attracting the attentions of researchers.
Ash is a chemically stable inorganic material. Thus, there is a potential for their application as a construction
material. Possible applications of ashes reported in the literature include asphalt, cement concrete block, road
construction, and melt slag (13). Recently, we carried out studies on the possible application of ashes from
incinerated paper sludges for interlocking blocks. Acrylic latex was employed to improve the properties of cement
mortar or concrete modified with ashes. Our results are partly reported in this paper.
2
Experimental program
Material: Ordinary Portland cement was used in the experiment. Chemical composition and physical properties of
the cement are given in Table 1 and Table 2 respectively. Aggregates used in the experiment was standard sand
satisfying Korean Standard L 5100. Ash from incineration of paper sludge was investigated in this study
especially. Chemical composition of the ash is given in Table 3. From the composition of the ash, it can be

Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

204

SOH, SOH AND LEE

classified into ash from sludges with polymer. In order to improve the properties of the cement mortar or
concrete, acrylic latex was used in the experiment. Properties of the acrylic latex are given in Table 4.
Table 1. Chemical composition of the Portland cement used in the experiment
SiO2

Al2O3

CaO

Fe2O3

MgO

SO3

K2O

lg.loss

C3S

C2S

C3A

C4AF

21.0

6.0

62.1

2.8

3.4

2.0

1.2

1.7

43.1

27.9

11.2

8.5

Table 2. Physical properties of the Portland cement used in the experiment


Specific gravity

Specific surface area (cm2/g)

Stability

Setting

Initiation (sec)

Termination (hr)

3day

7day

28day

3.15

2,900

Good

230

6:10

Compressive strength
160

242

340

Table 3. Chemical composition of the ash used in the experiment


Comp
ositio
n

SiO2

Cont
ent
(%)

49.
96

Al2O

TiO2

Fe2O

CaO

MgO

Na2O

K2O

MnO

P2O5

Ig.los
s

5.45

5.37

2.03

0.73

0.47

0.87

1.87

29.
82

2.31

1.60

Table 4. Properties of the acrylic latex used in the experiment


Appearance

Color

Solid content

pH

Viscosity (cps)

Specific gravity

Milky liquid

White

47.1%

10

34.5

1.095

Preparation and Test Methods: In the preparation of cement mortar or concrete, content of ash was varied with or
without polymer latex. In order to compare the properties of the cement mortar or concrete at a same flow
property, flow values were maintained 1205 mm. Mix formulations studied are given in Table 5, Table 6, and
Table 7. Test methods of the specimens prepared are given in Table 8.
Table 5. Mix formulation of the cement mortar without polymer latex
Ash content (vol. % by composite)

C:S+Ash
(wt.ratio)

Weight proportion (kg/m3)

Paper sludge ash

Sand

Cement

1:3
110

0
30
70
1124
150
180

0
3
7
11
15
18

1234
1204
1164
76.5
1084
1054

411
120

W/C (%)

Flow (mm)

52.0
62.6
67.8

118
123
123

78.2
80.3

120
124

Table 6. Mix formulation of the cement mortar with polymer latex


Ash content (vol.% by composite

C:S+Ash
(wt.ratio)

P/C

Weight Proportion (kg/m3)

paper sludge ash

Sand

Cement

Polymer

10
1025

0
30
70
53.3
150
180

0
3
7
11
15
18

1:3
110

1135
1105
1065
120
985
955

378

80.4

W/C (%)

Flow (mm)

43.2
48.8
51.2

117
123
120

57.4
61.3

120
120

PAPER SLUDGE ASH IN CONCRETE PRODUCTS

205

Fig. 1. W/C ratios depending on ash content of cement mortar with and without polymer latex.
Table 7. Mix formulation of the cement concrete for interlocking block
(kg/m3)

(a) Without polymer latex


Ash content (wt. % by Aggregate)

Fine aggregate

Ash

sand

0
2
5
8
10
12

0
34
85
137
171
205

854
837
812
786
769
752

Coarse aggregate

Cement

W/C (%)

854
837
812
786
769
752

430
430
465
500
535
570

30.0
37.8
40.1
42.7
54.2
66.1
(kg/m3)

(b) With polymer latex


Ash content (wt.% by Aggregate)

P/C (%)

Ash

sand

2
5
8
10

5
137

Fine aggregate

Coarse aggregate

Cement

Polymer

W/C (%)

34
85
786
171

837
812
500
769

430
465
53.2
535

45.7
49.5
28.7
56.9

26.7
28.0

837
812
786
769

29.4

Table 8. Test methods employed in the experiments


Test item

Test method

Preparation of cement mortar


Flow test of cement mortar
Moisture absorption of cement mortar
Flexural strength of cement mortar

KS F 2403, 2425
KS F 2428
KS F 2503
KS F 2405, 2407

3
Results and Discussion
In Fig. 1, water/cement ratios of the cement mortar into which ash were incorporated with and without polymer latex
are shown. It is observed that water/cement ratios of the cement mortar increased as the ash contents were increased.
It is speculated that the increase of the water/cement ratios resulted from surface property of the ash. It is worthwhile
to note that the increase of water/cement ratios for cement mortar with polymer latex was less than that for cement
mortar without polymer latex. Water/cement ratio of the cement mortar was decreased about 20% by adding polymer
latex into the cement mortar.
In Fig. 2, moisture absorptions (%) of the cement mortar into which ashes were incorporated with and without
polymer latex are shown. In case of cement mortar without polymer latex, moisture absorption increased as the ash

206

SOH, SOH AND LEE

Fig. 2. Moisture absorption depending on ash content of cement mortar with and without polymer latex.

Fig. 3. Flexural strength depending on ash content of cement mortar with and without polymer latex.

Fig. 4. W/C ratios depending on ash content of cement concrete with and without polymer latex.

content is increased. The cement mortar with polymer latex showed less moisture absorption compared with the
cement mortar without polymer latex. Furthermore, moisture absorption increased and levelled off as the ash content
was increased in the cement mortar with polymer latex. It seems that decreased water absorption of the cement
mortar with polymer latex is due to film formation of the latex polymer in the capillary and voids of the cured
cement mortar (45).
In Fig. 3, flexural strength of the cement mortar into which ash were incorporated with and without polymer latex
are shown. It is observed that flexural strength of the cement mortar decreased as the ash content were increased whether
polymer latex was added or not. However, the flexural strength of the cement mortar with polymer latex was
superior to that of cement mortar without polymer latex. Furthermore, the decrease of the flexural strength of the
cement mortar with polymer latex was less than that of the cement mortar without polymer latex as the ash content was
increased.

PAPER SLUDGE ASH IN CONCRETE PRODUCTS

207

Fig. 5. Moisture absorption depending on ash content of cement concrete with and without polymer latex.

Fig. 6. Flexural strength depending on ash content of cement concrete with and without polymer latex.

In Fig. 4, water/cement ratios of the cement concrete for interlocking block into which ashes were incorporated
with and without polymer latex are shown. It is observed that water/cement ratios of the cement concrete increased
as the ash contents were increased. It is worthwhile to note that the increase of water/cement ratios for cement
concrete with polymer latex was less than that for cement mortar without polymer latex and almost constant.
In Fig. 5, moisture absorptions (%) of the cement concrete for interlocking block into which ashes were
incorporated with and without polymer latex are shown. In case of cement concrete without polymer latex, moisture
absorption increased as the ash content is increased. The cement concrete with polymer latex showed less moisture
absorption compared with the cement mortar without polymer latex. Furthermore, moisture absorption decreased as
the ash content was increased in the cement concrete with polymer latex. It is believed that the decreased moisture
absorption of the cement concrete with polymer latex are due to film formation of the latex polymer in the capillary
and voids of the cured cement concrete. According to KS F 4419, moisture absorption of the interlocking block
should be less than 7%. It seems that addition of polymer latex at least 5% of cement to the cement concrete with
ashes is necessary to meet the KS F 4419.
In Fig. 6, flexural strength of the cement concrete into which ash were incorporated with and without polymer
latex are shown. It is observed that flexural strength of the cement concrete decreased as the ash content were
increased whether polymer latex was added or not. It was also found that the flexural strength of the cement concrete
with polymer latex was superior to that of cement concrete without polymer latex. In case of flexural strength for the
interlocking block, it is required to be higher than 51 kg/cm2 generally in Korea. Thus, only the flexural strength of
the cement concrete without polymer latex the ash contents of which were equal to or less than 10% was acceptable
for the interlocking block in Korea.
4
Conclusion
Effects of incorporation of ashes into cement mortar or concrete are as follows;

208

SOH, SOH AND LEE

1. Water/cement ratios increase as the ash contents in the cement mortars or cement concretes are increased.
2. Moisture absorptions of the cement mortars or cement concretes also increase as the ash contents are increased.
3. Flexural strength of the cement mortars or cement concretes decrease as the ash contents are increased.
All the disadvantages mentioned above could be overcome by adding 5% of acrylic polymer latex of cement into
cement concrete to make acceptable interlocking block. Interlocking block made from cement concrete into which
ash was added up to 10% without polymer latex could meet only the required flexural strength.
5
References
1.
2.
3.
4.
5.

Ohkubo, M. and Urakawa, Y. (1993), Test of mortar or concrete mixed with waste incinerated ash on the compressive
strength and other characteristics, Journal of Struct. Constr. Engng, AIJ, No. 443, Jan., pp. 19
Hiraoka, M., Fukui,T., Yashiki, D., Murakami, T., Shimizu, K. and Oshima, H. (1993), Sewage Sludge Melting and
Recycling, Jornal of PPM, Vol 24, No. 2, pp, 3546.
Soh, Y.S. (1992), Development of Lightweight Polymer Cement Concrete, Korea-Japan Symposium on Development and
Application of Concrete-Polymer composits, Chonbuk National Univ., pp. 172159.
Ohama, Y. (1987), Principle of Latex Modification and Some Typical Properties of Latex-Modified Mortars and
Concretes, ACI Materials Jounal, NovemberDecember, pp. 511518.
Manson, J.A. (1981), Applications of Polymer Concrete, Publication SP-9, ACI, Detroit.

PART SIX
OTHERS

24
INVESTIGATION ON DETERIORATION OF RECYCLED HOTMIXED ASPHALT CONCRETE PAVEMENT AND A TRIAL RERECYCLING OF ASPHALT CONCRETE
T.YOSHIKANE
Taiyu Kensetsu Co. Ltd, Nagoya, Japan

Abstract
Plant-recycled hot-mixed asphalt pavement in Japan began in 1975 and has entered on the development period
rapidly since 1985. When we think about the life of asphalt pavement, the development of the technology
regarding re-recycling will become a future theme. From such background, on the basis of the follow-up
survey during the past 15 years for the recycled pavement that was carried out in 1978, we have measured a
change with passage of time between new pavement and recycled pavement. Both the new and recycled
pavements are in good condition even to date, and serve as a source of valuable information about pavement
durability. We will also report on the results of our investigation of recycled and re-recycled hot-mixed asphalt
pavement using asphalt concrete lumps collected at a study site.
Keywords: Asphalt concrete pavement, durability, pavement recycling, recycled hot-mixed asphalt concrete,
re-recycling.
1
Introduction
The use of plant-recycled hot-mixed asphalt (HMA) for public road pavement in Japan first began in 1975. The
Japan Road Association promoted the technology for plant-recycled HMA pavement in Japan with its 1984
publication of The Plant-Recycled HMA Pavement Technical Guideline, and since 1985, positive efforts have
been made to reuse more reclaimed asphalt pavement (RAP).
It has thus been about 20 years since pavement recycling first started in Japan, and about 10 years since this
country began carrying out policies to promote an increase in the application of the guideline. Now, amidst a
growing stock of public road pavement and an increased awareness of issues such as the need to conserve our natural
resources, pavement materials are required to have longer life cycles; to be recycled if possible; and even re-recycled,
if possible. To allow for further development of pavement recycling, we must investigate from the long-range
outlook, the durability of recycled pavement and its ability to withstand long-term performance.
This report contains the results of one study of long-term performance of recycled HMA pavement. The study in
question is a 15 year follow-up survey of one of Japans earlier (1978) full-scale applications of recycled pavement
using recycled HMA to pave municipal roads in Nagoya City. In this pavement test, new HMA was applied at the
same time so as to compare it with to the recycled pavement. Both the new and recycled pavements are in good
condition even to date, and serve as a source of valuable information about pavement durability. We will also report
on the results of our investigation of recycled and re-recycled HMA using asphalt concrete lumps collected at a study
site.
Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

DETERIORATION OF HOT-MIXED ASPHALT CONCRETE

211

Fig. 1 Total amount of HMA, recycled HMA and proportion of recycled HMA in Japan

Fig. 2 Change volume for use of RAP and average RAP mixing ratio

2
Progress of plant-recycled HMA pavement in Japan
Production figures for both HMA and recycled HMA can be seen in Fig. 1. As this figure shows, the total production
of HMA (including recycled HMA) has not changed much over the past 10 years. However, the production of
recycled HMA has grown at a rapid pace. Also, with a total annual production of 1,560 tons in 1993, recycled HMA
now occupies nearly 20% for all HMA production.
The amount of RAP used and the average RAP mixing ratios for recycled HMA are shown in Fig. 2. The amount
of RAP used has increased steadily over the years, reaching 6.24 millions tons in 1993. (This figure was calculated
from the total HMA production for that year and the average RAP mixing ratio of 40% can be seen in Fig. 2). On the
other hand, while the average RAP mixing ratio reached a level of better than 60% in the first and second surveys
(conducted in 1985 and 1987, respectively), the ratio dropped sharply to only 42.5% in the 1992 survey. This is due
to the reduced availability of asphalt concrete lumps that has accompanied the spread of recycled HMA pavement
technology out of metropolitan areas and into the rural areas, resulting in an increased demand for HMA with lower
RAP mixing ratios. (The spread of HMA recycling technology into rural areas was made possible by the fact that

212

YOSHIKANE

Fig. 3 Average RAP mixing ratio with the whole country and regions

Fig. 4 Proportion to the amount of HMA on classification by RAP mixing ratio

Fig. 5 Total amount of recycled HMA by each kind

HMA recycling plants need only simple equipment to produce recycled HMA with RAP mixing ratios of less than
30%.)
Figure 3 shows the changes in the RAP mixing ratio for various regions. In this figure, we see that the average
RAP mixing ratio has dropped drastically in the Kanto region, which includes Tokyo. Large decreases can also be
seen in the Hokuriku, Chugoku and Shikoku regions; however, the figures for the Chubu and Kinki regions have
changed little. Of course, the RAP mixing ratios are not fixed within a region: a wide range of RAP mixing ratios can
be found in each region. Nonetheless, high RAP mixing ratios are not common in the rural areas, and more mixtures
tend to have lower RAP mixing ratios as pavement recycling becomes more common. Figure 4 shows changes in the
distribution of RAP mixing ratios. This figure shows well the trend toward lower RAP mixing ratios.
Looking at the shares occupied by different types of recycled HMA in Fig. 5, we find that these proportions have
not changed much since the first survey. Table 1 shows the changes in the penetration of old bitumen included in RAP.
The penetration of old bitumen is an important factor in the recycled HMA mix design. The average penetration has
dropped, if only slightly, between the second and third surveys, and the rate of this change is on the increase.
To summarize the trends described above: (1) Production of recycled HMA has been increasing rapidly over the
past several years in Japan. Recycled HMA now is 20% for all HMA production. (2) The average RAP mixing ratio
has started to drop sharply as plant-recycling HMA spreads into the rural areas. (3) The average penetration of old
bitumen has shown a tendency to drop slightly over the past several years, and the rate of this change is showing
signs of increasing.

DETERIORATION OF HOT-MIXED ASPHALT CONCRETE

213

3
Follow-up survey of test pavement
3.1
Outline of construction work
Test paving was carried out on a municipal road having Class C traffic volume within Nagoya City limits. New and
recycled HMA were used for three construction areas, as seen in Table 2.
Table 1 Change of penetration of old bitumen from RAP
Year of investigation
Number of plant
Maximum
Minimum
Mean of mean in each plant
Mean of standard deviation in each plant

1987

1992

101
62
17
33.3
5.7

128
57
16
31.9
5.1

5
1
1.4
0.6

Table 2 Summary of test pavement project


Place

City road in Nagoya

Year

March 1978

Condition of design

Classification of traffic volume: C traffic TA=28.35cm


Design CBR: 6 H=47cm

RAP mixing ratio

Dense grade: 50% Corase grade: 60%


Bituminous stabilization: 70%

Layer component
No. 1 area
Surface
Binder course
Base course
Subbase course

Kind of layers
No. 2 area
Dense grade
Coarse grade
Bituminous stabilization
Crusher-run
(M-25)

Type of mixture
No. 3 area
5
5
12
25

cm

Combination (only HMA layers)

New
New
New

Recycled
Recycled
Recycled

New
Recycled
Recycled

3.2
Mixture composition and characteristics
The RAP used was adjusted to a gradation from 13 to 0 mm. The physical characteristics of RAP are shown in
Table 3. Table 4 shows the composition of the recycled HMA. Recovered bitumen is old bitumen whose penetration
has been adjusted to a desired level by the addition of new bitumen and, in some cases, a rejuvenator. For the present
test, the penetration of the recovered bitumen was adjusted to 55, 50, and 40and the RAP mixing ratios were set to
50, 60, and 70% for dense-graded concrete, coarse-graded concrete and bituminous stabilization, respectively. The
recovery of recycled HMA penetration was accomplished using new bitumen (pen. 80100) only. In addition, since
the bitumen content was set so that the dynamic stability of the recycled HMA (as measured by the wheel-tracking
test) would be similar to that of new HMA, the designed recovered bitumen content of the recycled HMA was
between 0.2 and 0.4% greater than that of new HMA.

214

YOSHIKANE

Table 3 Properties of RAP


Gradation of extracted aggregates(Weight % of fraction passing a sieve) % Extracte
d
bitumen
content
(%)

Number
Mean
Standard
deviation
Maximu
m
Minimu
m

Properties of old bitumen

20mm

13

2.5

0.6

0.3

0.15

0.074

Penetrati Softenin
on
g point
(1/10mm) (C)

Penetrati
on index

12
100

12
96.0
1.1

12
65.2
2.9

12
47.8
2.9

12
29.6
2.3

12
18.0
1.5

12
11.9
1.1

12
9.0
1.0

12
4.98
0.18

12
35
2.2

12
55.0
1.2

12
0.85
0.10

98.4

70.0

53.3

33.2

21.0

13.7

10.5

5.28

38

56.2

0.74

94.9

59.2

43.2

25.8

15.6

9.6

7.7

4.33

32

53.7

0.92

Table 4 Mix design


Items

Type of Mixture

Bituminous stabilization

Coarse-graded HMA

New

Recycled

New

Recycled

New

Recycled

Mixing ratio (%)


S-20
S-13
S-5
Screenings
Fine sand
Filer
Bitumen content (%)
Old bitumen
Total

RAP
20.0
85%

19.4
10.0

4.00
3.40
4.50

70.0
20.0
33.8
18.0
12.5
12.5
3.8
1.10

5.00

10.1
15.0
5.0

5.00
2.88
5.20

60.0
20.0
31.2
15.0
20.7
20.7
2.3
2.32

5.40

(Coarse sand)
15%
New bitumen

4.00

Dense-graded HMA

50.0

7.0
5.0
8.5
8.0
1.5
5.40
2.38
5.60

3.22

Table 5 Results of quality control test


Items

Type of mixture

Bituminous stabilization Coarse-graded HMA Dense-graded HMA

New

Recycling

New

Recycling

New

Recycling

3.93
2.514
2.398
5.5
10.5

4.50
2.538
2.349
5.2
11.5

4.91
2.478
2.408
4.4
12.1

5.18
2.519
2.357
4.0
12.4

5.24
2.454
2.405
3.3
13.1

65.6
1,259
34
37
86.9
65

68.9
927
26
36
84.3
41

73.3
1,162
36
32
88.5
66

75.6
1,022
25
41
86.0
49

79.9
1,307
33
40
91.8
69

53.4
0.80

49.3
0.71

52.5
0.64
3,150
0

48.8
0.73
3,100
0

52.1
0.51
1,016
0

Extracted bitumen content (%)


Marshall stability test
Maximum density (g/cm3)
3
Apparent density (g/cm )
2.307
Air voids (%)
8.3
Percentage of bitumen by
9.0
volume (%)
Degree of saturation (%)
52.2
Stability (kgf)
672
Flow value (1/100cm)
24
S/F (100kgf/cm)
29
Residual stability (%)
81.7
Properties of recovered
Penetration (1/10mm)
bitumen
Softening point (C)
49.5
Penetration index
0.70
Wheel tracking test
Dynamic stability
Immersion stripping (%)
0

5.60
2.487

54

926

DETERIORATION OF HOT-MIXED ASPHALT CONCRETE

215

Fig. 6 Change of cracking ratio

Fig. 7 Change of rutting depth

Table 5 shows the results of quality control testing for the HMA mixture varieties produced by an asphalt mixing
plant. The recycled HMA had a smaller percentage of air voids and a greater Marshall stability than the new HMA.
This is because recycled HMA has a tendency to be denser than the new HMA, due to a slightly greater filler
content, a slightly lower penetration and a slightly higher design recovered bitumen content. Also, given the dynamic
stability (as measured by the wheel-tracking test) and the immersion stripping percentage, we can expect that the
flow resistance and stripping resistance were of more or less comparable levels.
3.3
Results of a follow-up survey
3.3.1
Changes in surface conditions
The present survey was conducted in June of 1993, 15 years and 3 months after the test pavement was laid. It was the
7th follow-up survey regarding this pavement test.
No cracks were found in any of the construction areas after ten years. The present survey found that a few cracks
had appeared in both the new and recycled HMA areas. The new HMA area had slightly more cracks at 6% or more,
than did the recycled HMA areas, which had a ratio of only 1% (Fig. 6).
Rutting had increased gradually in both the new and recycled HMA areas. The total rutting after 15 years was only
slight, however, at over 10 mm. No great difference could be observed between the new and recycled HMA
construction areas (Fig. 7).

216

YOSHIKANE

Fig. 8 Change of surface roughness

Fig. 9 Change of deflection

Fig. 10 Change of skid resistance coefficient

The results for surface smoothness, deformation and skid resistance (Figs. 810) were also good for both the new
and recycled HMA areas, and no difference was seen between these areas. The recycled HMA areas were not
inferior to the new HMA area.
3.3.2
Changes in air void percentage
Figure 11 shows the changes in air void percentage for the new and recycled HMA areas. We have already
mentioned that the air void percentages of the recycled HMA areas were generally lower than that of the new HMA
area from the time immediately following application. However, the proportions of subsequent changes were smaller
for the recycled HMA areas, which had the lower original percentages. Most of the attrition of the air void
percentage occurred over several years following paving. Any changes thereafter were comparatively small.
Although the air void percentages for the recycled HMA areas were low overall, this does not seem to have increased
rutting in these areas. In fact, it may have had a beneficial effect on the pavements durability. We interpret this as a
result of using the wheel-tracking test to set the bitumen content.

DETERIORATION OF HOT-MIXED ASPHALT CONCRETE

217

Fig. 11 Change of air void

Fig. 12 Change of penetration

3.3.3
Changes in characteristics of extracted bitumen
Figure 12 shows the changes in the penetration of bitumen extracted from core samples of the bituminous pavement
in the test areas. As the figure shows, the penetration for both the new HMA and recycled HMA dropped rapidly
between mixing and application, whereas changes after application were small. Although penetration on the surface
course tended to become progressively worse after use, the binder course and the bituminous stabilization course
showed only a slight decrease. Also, since the change in the penetration in the new HMA was proportionally greater
than that in the recycled HMA, the difference observed between the two during the mix design and construction
stages tended to shrink after use.
Figure 13 shows the results of the spectrum analysis performed on the bitumen extracted from the surface course
RAP of the new HMA and recycled HMA areas after 15 years of use. It also shows data for the new bitumen at the
time of application, and for extracted old bitumen with extremely deteriorated penetration. (The extraced old bitumen
data, presented for reference, is from another RAP source unrelated to the pavement in the present test.) It is known
that infrared absorption spectrum analysis will show changes reflecting bitumen deterioration. In particular, spectrum
changes in wavelengths near 1,700 cm1 are attributed to oxidation. As the results of the analysis show, the spectrum
sharply depresses around 1,700 cm1 for the deteriorated bitumen. However, the bitumen extracted from both the new
and recycled HMA test areas showed only slight spectrum changes. Even when inspected by infrared absorption
spectrum analysis, the deterioration of bitumen extracted from recycled HMA whose penetration has been adjusted
using only the new bitumen does not appear to be greatly different from the deterioration of bitumen extracted from
new HMA.

218

YOSHIKANE

Fig. 13 Infrared adsorption spectrum

3.3.4
Changes in dynamic stability (DS)
Table 6 shows the DS values measured by the wheel-tracking test for core samples collected during the present
survey, compared with those for core samples collected at
Table 6 Results of wheel tracking test
condition of
load
(kgf)

No. 1 work area (by new HMA) No. 2 work area (by recycled
HMA)

Dense-grade

Coarse-grade

Dense-grade Coarse-grade

Dynamic
stability (DS)
Plant mix
sample
(Number/
mm)
70

At the time of
construction
60

Mix design

60

724

2,000

1,016

3,150

926

3,100

Cutting core at this time

60

4,219

1,804

3,120

1,775

1,248

1,213

712

2,625

1,933

2,378

the time of construction. The preliminary test conditions for the present survey differed somewhat from those at the
time of construction. Nonetheless, the table does indicate that while the DS values changed greatly for the surface
courses of both the new and recycled HMA areas, the changes in the values for the binder course were rather rare.
This shows that bitumen deterioration is more severe in the surface. However, because the surface course of the
recycled HMA areas were given bitumen contents slightly greater than that of the new HMA area, there were fewer
changes in the DS for these areas. Also, given the balance between the DS values for the surface and binder, the higher
bitumen content appears to provide a recycled HMA with advantages in terms of cracking resistance and durability.
4
Mix Design for re-recycling HMA
Asphalt concrete lumps were collected from the new and recycled HMA areas of the present survey only, and
pulverized for use as aggregates in trial mix designs for recycled and re-recycled HMA, respectively, with gradations

DETERIORATION OF HOT-MIXED ASPHALT CONCRETE

219

similar to those used in the original test pavement. A mix design using the new HMA was also prepared for a
comparison of the results. In the following sections we have reported on the test results for these mixtures, including
a comparison of the results of composition testing of the original pavement at the time of application.
4.1
Characteristics of materials used
The characteristics of the new aggregates and RAP used in the mix designs can be seen in Table 7. The new
aggregates were taken from the same source location (as nearly as possible) as those use in the original test, and the
RAP gradation was adjusted to the same sizes from 13 to 0 mm. The gradation of RAP used in this test resembles
that of RAP used in the original test pavement (Table 3). Also, the material used deteriorated less with a higher
penetration than the old bitumen. New HMA with new bitumen (pen. 80100) was used to prepare the recycled and
re-recycled HMAs, and new HMA with new bitumen (pen. 6080) was used as a comparison.
Table 7 Properties of use materials
Items

Use material

New aggregate

S-20

S-13

S-5

Weight of
26.5 (mm)
passing sieves
(by weight%)
19
97.4
13.2
49.0
4.75
0
2.36
0.6
0.3
0.1
575 m
Apparent specific gravity
Maximum specific gravity
Coefficient of water absorption (%)
Old bitumen content (%)
Properties of
Penetration (1/
recovered
10mm)
bitumen
Softening point

(C)
Penetration index

Mixture

Screenings Sand

Filer

No. 1 Work area No. 2 work area


(by HMA)
(by recycled
HMA)

100.0

100.0
6.1
0

100.0
97.0
85.0
2.661

1.78

100.0
94.2
65.4
49.0
27.2
17.9
12.6
9.5
2.710

100.0
95.3
61.1
43.4
25.6
17.1
12.1
9.3

2.677

4.70
48

50.4

52.1

1.19

1.18

100.0
95.1
32.2
8.7
7.0
5.7
4.6
2.710 2.709

0.51
0.69

100.0
100.0 98.7
98.0
74.2
51.2
31.2
37.9
12.2
27.5
5.2
19.5
1.8
2.691 2.676

0.93
1.70

2.721

4.87
40

4.2
Mixture composition and Marshall stability test results
Since RAP had physical properties similar to those of RAP in the original test pavement, the same RAP mixing ratio
of 50% and the same proportions of new aggregates were used. Also, new HMAs were also adjusted to gradations
similar to those used in the original test pavement (Tables 4 and 8).
The Marshall test results showed that the recycled and re-recycled HMAs a had slightly lower bitumen content,
and slightly greater stability than the new HMA. The dynamic stability of the recycled and re-recycled HMAs (as
measured by the wheel-tracking test) was also somewhat higher. As Fig. 14 shows, however, it is possible to achieve
a dynamic stability equal to that of new HMA by increasing the bitumen content of the recycled and re-recycled
HMAs to about 0.2%. This finding shows a similar trend to that seen in the new and recycled HMAs of the original
test pavement.
These results support the idea that re-recycling is possible if (1) a RAP mixing ratio of 50% is used, (2) the
penetration of old bitumen is adjusted using only for new HMA, and (3) the bitumen content is set using a reiterative

220

YOSHIKANE

Fig. 14 Relationship between bitumen content and dynamic stability

testing method such as the wheel-tracking test. Further research will be needed to determine if re-recycling is
possible when the RAP mixing ratio greatly exceeds 50%, or when rejuvenators, a modified bitumen or other elements
are introduced.
Table 8 Properties and mixing ratio of mixture
Items

Type of mixture

Dense-graded HMA

New

Recycled

Re-recycled

Aggregate proportion (%)


S-20
S-13
S-5
Screenings
Sand
Filer
Bitumen content (%)
Old Bitumen
Total
Penetration of recovered bitumen (1/10mm)
Marshall stability test
Apparent density (g/cm3)
Air voids (%)
Percentage of asphalt by volume (%)
Degree of saturation (%)
Stability (kgf)
Flow value (1/100cm)
S/F (100kgf/cm)
Residual stability (%)
Dynamic stability (number/mm)

RAP
24.0
19.3
13.0
20.7
20.7
2.3
New Bitumen

5.4

20.0
7.0
5.0
8.5
8.0
1.5
5.4
2.3
5.3
70
2.475
2.392
3.4
12.3
78.3
955
34
28
88.0
1,212

Maximum density (g/cm3)


2.389
3.5
12.5
78.0
919
31
30
92.7

50.0
20.0
7.0
5.0
8.5
8.0
1.5
3.0
2.4
5.2
57
2.475
2.412
3.4
12.2
78.2
1,138
36
32
94.1
1,575

50.0

2.8

52
2.496

1,500

DETERIORATION OF HOT-MIXED ASPHALT CONCRETE

221

5
Conclusion
This study of a test pavement using recycled HMA, in which the penetration of the old bitumen had been adjusted
using new HMA, confirmed the durability of such a pavement and showed that it could maintain a comparatively
good level of utility even after 15 years and 3 months of use. Also, with respect to the physical properties, we have
found that the bitumen deterioration of layers of the HMA subject for public use was comparatively small and that
the repeated reuse of pavement is indeed possible. Such a high evaluation of pavement recycling from the viewpoint
of both utility and physical characteristics should broaden the possibilities for expansion into re-recycling.
In this report, we also have examined the mix design for re-recycled HMA. As the pavement repair cycle is
growing shorter, it may be that some recycled asphalt concrete lumps are already being mixed in with asphalt
concrete lumps, and therefore, being used as RAP. Thus, partial re-recycling is already happening, in combination
with the recycling process. Also, the annual average penetration of old bitumen in asphalt concrete lumps being
conveyed to recycling plants is tending to drop, if only slightly. This may be due to the increases in the influx of
HMA with advanced deterioration, recycled HMA and modified HMA used in recent years as a flow resistance
measure. To ensure even more use of recycled HMA, such problems must be studied, and dealt with responded to,
from the standpoint of a long-range perspective.
6
References
1.
2.
3.
4.
5.

Yoshikane, T., Kondo, K. and Funahasi, H. (1979) Properties of recycled hot-mixed asphalt mixture, Proceeding of 13th
Japan road conference.
Yoshikane, T. (1979) Design and construction of pavement for saving energy, Proceeding of 13th Japan road conference.
Takagi, T., Funahasi, H. and Miyata, T. (1987) Surface properties of recycled hot-mixed asphalt pavement by long-term
performance, Proceeding of 17th Japan road conference.
Kuriyama, M., Yoshikane, T., Kaneda, M. and Ookouchi, T. (1994) Durability and re-recyclability of hot-mixed asphalt
pavement, Road Construction.
Japan Asphalt Mixture Association, Report of research on the actual condition of recycled hot-mixed asphalt mixture,
Mixture product annual report.

25
REUSE OF CARPET INDUSTRIAL WASTE FOR CONCRETE
REINFORCEMENT
Y.WANG
School of Textile & Fiber Engineering, Georgia Institute of Technology, Atlanta, USA

Abstract
About 2 million tons of industrial and post-consumer carpet waste is disposed of each year in the United
States. The waste could offer very attractive uses in the construction industry as an additive for fiber reinforced
concrete (FRC). An experimental program demonstrated that recycled fibers from carpet industrial waste
provide very effective reinforcement for concrete. Performance enhancement has been observed in
compressive, flexural toughness, and drying shrinkage tests. FRC using recycled carpet waste fibers has also
been successfully used in the construction project of a 11,000 m2 building, demonstrating that low-cost carpet
waste fiber can be used in full scale construction projects. This paper reports on the experimental programs and
compares the effectiveness of such waste fibers with that of conventional fibers for concrete reinforcement. It
also discusses issues that need to be addressed for the application of such FRC in large scale construction
projects and the benefits of using such FRC as a construction material, which include improved durability,
reliability, and earthquake resistance.
Keywords: Carpet industrial waste, fiber reinforced concrete, field study.
1
Introduction
Concrete is the most heavily used construction material in the world. It is durable, inexpensive, readily molded into
complicated shapes, and has adequate compressive strength and stiffness. However, concrete has low tensile strength,
low ductility, and low energy absorption. Due to its lack of tensile strength, concrete is often reinforced with steel
reinforcing bars (rebars) in structural applications. However, under adverse loading conditions such as during an
earthquake, a reinforced concrete structure could collapse due to cracking and spalling of concrete.
An intrinsic cause of the poor tensile behavior of concrete is its low toughness and the presence of defects.
Therefore improving concrete toughness and reducing the size and amount of defects in concrete would lead to better
concrete performance. An effective way to improve the toughness of concrete is by adding a small fraction (usually
0.52% by volume) of short fibers to the concrete mix during mixing. In the fracture process of fiber reinforced
concrete (FRC), fibers bridging the cracks in the matrix can provide resistance to crack propagation and crack
opening before being pulled out or stressed to rupture, as illustrated schematically in Figure 1. After extensive
studies it is now beyond doubt that such fiber reinforcement can significantly improve the tensile properties of
concrete. Orders of magnitude increases in toughness (energy absorption) over plain concrete is commonly observed
[14].
Another important advantage of fiber reinforcement is the reduction of the shrinkage and shrinkage cracking of
concrete associated with hardening and curing. Although the exact mechanism is not fully understood, it is generally
Disposal and Recycling of Organic and Polymeric Construction Materials. Edited by Y.Ohama. RILEM.
Published by E & FN Spon, 26 Boundary Row, London SE1 8HN, UK. ISBN 0 419 20550 0.

CARPET WASTE FOR CONCRETE REINFORCEMENT

223

Fig 1. Schematic illustration of fiber bridging action in FRC

believed that fibers help retain water in concrete and thus reduce concrete plastic shrinkage. This water retention
ability could also extend the cement hydration process, leading to a stronger concrete. Reduced shrinkage cracking
has been observed even with fiber volume fractions as low as 0.1% of polypropylene fibers [5].
Other benefit of fiber reinforced concrete include improved fatigue strength, wear resistance and durability. By
using FRC instead of conventional concrete, section thickness can be reduced and cracking can be effectively
controlled, resulting in lighter structures with longer expected life. Fiber reinforced concrete is currently being used
in many applications including buildings, highway overlays, bridges, and airport runways [13]. In load bearing
applications it is generally used along with traditional steel reinforcement [6]. In building constructions it has
become a more common practice to use low-dosage synthetic fiber reinforcement for floor slabs.
Fibers commonly used for concrete reinforcement include polypropylene. steel and alkali-resistant glass. The
fibers are often of short length (<30 mm) and are randomly mixed in concrete.
Although FRC has become widely available from concrete suppliers, its use has been limited primarily due to cost
considerations. Effort to reduce the cost includes limiting the fiber dosage and developing inexpensive fibers. A very
attractive alternative, however, would be to use suitable recycled fibers from industrial waste which otherwise would
be discarded, the carpet industrial waste being an example.
This paper summarizes the results of an experimental program to evaluate the effectiveness of using recycled
fibers from carpet waste for concrete reinforcement. It also discusses issues that need to be addressed for the
application of such FRC in large scale construction projects. It then reports on a building construction project using
carpet waste fiber reinforced concrete. The results suggest that using carpet industrial waste fibers in construction
would not only improve the reliability and life of the concrete structure, it but also could reduce the landfill spaces
needed to dispose the waste material.
2
The carpet industrial waste
The U.S. carpet industry produces about 1 billion m2 of carpet and consumes about 1 million tons of synthetic fibers
per year. About 70% of the carpet produced is for replacement of used carpet, which translates into about 2 million
tons of used carpet for disposal. In the Dalton, Georgia area where many carpet manufacturers are located, over 40,
000 tons of carpet waste has to be disposed of each year. Significant amount of carpet trim waste is also being
disposed of by other industries such as the automobile manufacturers. Because of the high cost of developing and
managing landfills, waste disposal in landfills has become increasingly difficult.
A carpet typically consists of two layers of backing (usually fabrics from polypropylene tape yarns), joined by
CaCO3 filled styrene-butadiene latex rubber (SBR), and face fibers (majority being nylon 6 and nylon 66 textured
yarns) tufted into the primary backing, as illustrated in Figure 2. Most of the carpet industrial waste is selvage trim,
seams, and lint. Other materials polyester, wool, jute, acrylic, etc., form a relatively small portion of the
waste. The SBR adhesive is a thermoset material, which cannot be remelted or reshaped. Some waste is generated
before the application of SBR. Such waste is termed soft waste, and most of it is reused as a filling material or as
nonwoven mats. The waste containing the SBR (termed hard waste) has not found suitable uses and it forms the
major part of the waste going into the landfills.
Previous studies reported in the literature have indicated that nylon and polypropylene fibers can provide effective
reinforcement for concrete [15]. The configurations of the fibers in carpet, nylon face yarn being highly crimped
and polypropylene fibers being of flat tape shape, help improve the fiber-matrix interlocking which is essential for
effective reinforcement. All the major components of the carpet waste are shown to have good durability in Portland
cement [1, 4].

224

WANG

Fig. 2. Illustration of tufted carpet structure

The carpet industrial waste generated each year and that accumulated in landfills represent an abundance of useful
resources, as they can provide effective reinforcement for concrete. As to be discussed in the following sections,
concrete reinforced with recycled fibers from hard carpet waste is indeed a suitable material for construction. It
suggests that using carpet waste in construction could be a very cost-effective way to improve the durability and
performance of the concrete structure, and to reduce the needs for landfill spaces.
3
Laboratory evaluation
The laboratory study on concrete reinforcement with carpet waste fibers was carried out at Georgia Tech. Details of
the study has been reported elsewhere [7]. The concrete matrix consisted of Type I Portland cement, river sand,
crushed granite aggregate (maximum size=10 mm), and water. The weight ratios were 1.0 (cement)/0.35 (water)/0.85
(sand)/0.61 (aggregate). A small amount of superplasticizer was also added. Recycled carpet waste fiber used was
from hard carpet waste, disassembled mechanically by the Crown America, Inc. of Dalton, Ga. After disassembling,
surface yarns (nylon) and some backing fibers (polypropylene) were collected. This collection is referred to as Type
I waste fiber and had a typical length range between 12 to 25 mm. The disassembled waste after removal of Type I
fiber is referred to as Type II, which contained backing fibers, SBR coated face yarns, and SBR particles. An
analysis of Type II fiber indicated that the fiber length was about 3 to 25 mm and that about one third by weight was
actually fiber, the other two thirds being CaCO3 filled SBR. Fiber volume fractions for the waste fibers were 1% and
2%. Only the actual fiber portion was included for calculating fiber volume fractions for Type II waste fiber
reinforced concrete. FiberMesh, a virgin polypropylene fiber, at 0.5% and 1% volume fractions was also included
in this study for comparison purposes. The fiber length was 19 mm.
3.1
Compressive and flexural properties
Four point flexural test and cylinder compressive test were conducted on a hydraulic testing machine. The age of
specimens at testing was 28 days except the 1 day compressive test.
The results for compression and flexural tests of various mixes are given in Table 1. Six or seven specimens were
tested for each setup. In the one day compressive test, similar strength values were observed for plain concrete and
various FRCs. It appeared that the 28 day compressive strengths of carpet waste FRCs were lower than that of plain
concrete. The plain concrete specimens failed in a brittle manner and shattered into pieces. In contrast, all the FRC
samples after reaching the peak load could still remain as an integral
Table 1. Compressive and Flexural Test Results
Fiber in FRC Mix

Vf

Compressive Strength

Flexural Strength

Flexural Toughness Indices

1 day
MPa

28 days
MPa

MPa

I5

I20

Concrete control
Fiber Mesh PP
Fiber Mesh PP
Type I waste fiber
Type II waste fiber

0
0.5
1.0
1.0
1.0

20.9
24.2
19.9
22.4
20.1

52.6
52.2
51.5
59.4
61.8

4.65
4.58
4.99
5.24
4.09

1.0
2.6
3.3
1.8
2.1

1.0
6.9
12.5
4.4
5.4

CARPET WASTE FOR CONCRETE REINFORCEMENT

225

Fig. 3. Typical flexural test curves


Fiber in FRC Mix

Vf

Compressive Strength

Flexural Strength

Flexural Toughness Indices

1 day
MPa

28 days
MPa

MPa

I5

I20

Type I waste fiber


Type II waste fiber

2.0
2.0

20.9
18.6

39.7
40.7

4.69
4.35

3.3
3.5

7.5
9.8

piece, with fibers holding the concrete matrices tightly together.


The flexural strengths of all mixes tested were essentially the same and the standard deviations were low. Figure 3
shows the typical load versus displacement curves for the flexural test. The plain concrete samples broke into two
pieces once the peak load was reached, with very little energy absorption. The FRC specimens, on the other hand,
exhibited a pseudo ductile behavior and fibers bridging the beam crack can be seen. Because of the fiber bridging
mechanism, the energy absorption during flexural failure was significantly higher than that for plain concrete. One
method of characterizing the energy absorbing ability, or toughness, of FRC is to calculate the toughness indices, as
proposed in ASTM C1018. These indices indicate the energy absorption of FRC compared with a brittle material
such as concrete (toughness index=1.0) for the deformation range specified. It can be seen from Table 1 that on an
equal volume fraction basis, the toughness of carpet waste FRC is lower than that of FiberMesh FRC. But at a higher
fiber volume fraction (2%), the carpet waste FRC shows very impressive toughness which is high than that of FRC with
0.5% FiberMesh fibers.
3.2
Drying Shrinkage
The drying shrinkage of concrete was compared with that reinforced with carpet waste fiber and FiberMesh
polypropylene. The test was carried out by measuring the length changes of 50.850.8635 mm prism specimens at
19.5 C and 50% RH over a period up to 21 days. The reduction of drying shrinkage due to waste fiber
reinforcement at a 2% fiber volume fraction was about 20%.
4
Use of carpet waste FRC in full scale construction projects
The laboratory study described above demonstrated that recycled fibers from carpet industrial waste can provide
effective reinforcement for concrete. However, concerns do exist that must be addressed before such FRC can be
widely accepted for construction.
First, the carpet waste must be disassembled into short fibers of appropriate length. By optimizing the fiber length
and fiber/matrix bond properties, improved toughening effect can be achieved [8, 9]. This generally means that the

226

WANG

fiber should be long enough to carry significant load after matrix cracking, and yet the load is still lower than the
fiber strength so that the fiber will be pulled out rather than ruptured. Since long fibers tend to entangle during
mixing causing uneven fiber distribution, the fiber length should be limited to, say, 25 mm. In a full scale
construction project, large quantities of carpet waste fibers may be needed, and the shredding and cutting operation
should be equipped to handle the large volume and at the same time to maintain the product quality.
Next, the material must be handled by conventional construction equipment following conventional procedures,
including mixing, pouring, and finishing. The fibers must be metered and added to the concrete mix according to
proportion specifications. An automated fiber dosing system is preferred. Uniform fiber dispersion in concrete is
essential. The workability of fresh concrete is reduced when fibers are added. Superplasticizer is often added to
maintain an acceptable level of workability.
A most important requirement for using carpet waste FRC is, of course, that its properties must meet or exceed
specifications. As for conventional concrete, field and laboratory testing for quality control must be carried out.
Cost is another concern associated with the use of recycled carpet waste fibers in concrete. Currently a vast amount
of carpet waste is disposed of each year and therefore the cost of raw material for the recycled fibers is negligible.
The disassemble process to convert the waste into fibers suitable for concrete reinforcement requires only simple,
inexpensive shredding operation. Therefore the cost of the recycled fibers will remain very competitive with virgin
fibers for FRC.
The major factor limiting the use of carpet waste FRC in construction is the lack of experience with the material.
To overcome this barrier, Shaw Industries, Inc., the worlds largest carpet manufacturer, in 1993 decided to use
carpet waste reinforced concrete in the construction of its new Research & Development Center, a decision driven by
Shaws commitment to end landfill disposal of carpet production waste by the end of 1996. The R&D building, with
an approximate area of 11,000 m2, is located in Shaws headquarters in Dalton, Georgia. It consists of laboratories,
offices, and a pilot production plant. The construction was completed in May, 1994 and it represents the first full
scale project using a large amount of carpet industrial waste for concrete reinforcement [10, 11]. Shaw Industries,
Georgia Tech, and North Georgia Ready Mix (concrete supplier) worked together on the project along with
architects, engineers, testing firms and a waste recycling company. Initial trials were carried out to evaluate the
characteristics of carpet waste reinforced concrete under field conditions. The success of the initial tests led to the
specification of such concrete in this construction project. The dosage used was 5.95 kg of waste fiber per m3 of
concrete (10 lbs/yard3), which was much higher than the industry standard (0.89 kg of polypropylene fiber per m3, or
1.5 lbs/yard3). Such FRC was used for floor slabs, driveways, and walls of the building. A total of about 20 tons of waste
from Shaws production plant was used in the project, which would otherwise go to a landfill. The waste was
shredded into a typical fiber length of 12 mm, bagged, and added to the concrete mixing truck. On site testing of
workability, air entrainment and temperature of the fresh concrete was carried out and specimens for compressive
and flexural tests were prepared. It was observed that mixing, pouring and finishing with such FRC went smoothly.
The compressive and flexural strengths exceeded specifications, and improved ductility and shatter resistance of the
material were evident. This project has clearly demonstrated that low-cost carpet waste fiber can be used in a full
scale construction project.
Besides for building constructions, the carpet waste FRC can also be used for highway construction as pavements,
columns, bridge decks and barriers, and for airport construction as runways and taxiways. Recycled fibers from
carpet waste could become a standard additive for concrete.
5
Summary
A large amount of carpet waste is disposed of in landfills each year. This not only poses economical and
environmental problems to the industry, it also represents a severe waste of resources because the waste material can
prove to be valuable for construction applications. This study focused on the use of carpet waste fibers in fiber reinforced
concrete and demonstrated that such reinforcement can effectively improve the shatter resistance, toughness, and
ductility of concrete. Performance enhancement has also been observed in the drying shrinkage test. Such
improvements in concrete performance is especially beneficial for concrete structures in seismic zones as the
increased toughness could improve the reliability and shock resistance of the structures.
FRC using recycled carpet waste fibers has also been used in the construction project of a 11,000 m2 building.
Conventional equipment and procedures were used in the construction. The project consumed about 20 tons of
carpet waste as concrete additives. The waste fibers were directly added into concrete mixing trucks. The mixing,
pouring and finishing went smoothly; the compressive and flexural strengths exceeded specifications; and improved
ductility and shatter resistance of the material were evident. This project has demonstrated that low-cost carpet waste

CARPET WASTE FOR CONCRETE REINFORCEMENT

227

fiber can be used in full scale construction projects. Other uses of the material may include highway, bridge, and
airport construction.
Acknowledgment
This work has been partially supported by the National Textile Center which is funded by the U.S. Department of
Commerce. The author wishes to acknowledge the contributions by the Shaw Industries, Inc. and the North Georgia
Ready Mix Co.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.

ACI Committee 544 (1986) State-of-the-art report on fiber reinforced concrete. 544.1R-82 (reapproved 1986), American
Concrete Institute, Detroit.
Keer, J.G. (1984) Fiber Reinforced Concrete, in New Reinforced Concretes: Concrete Technology and Design, Vol. 2 (ed.
R.N.Swamy), Surrey University Press, pp. 52105.
Bentur, A. and Mindess, S. (1990) Fiber Reinforced Cementitious Composites, Elsevier, London.
Wang, Y., Backer, S. and Li, V.C. (1987) An experimental study of synthetic fiber reinforced cementitious composites.
Journal of Materials Science, Vol. 22, pp. 42814291.
Fibermesh Company Product Literature, Fibermesh Co., Chattanooga, TN.
ACI Committee 544 (1988) Design consideration for steel fiber reinforced concrete. ACI 544.4R-88, American Concrete
Institute, Detroit.
Wang, Y., Zureick, A.H., Cho, B.S., and Scott, D.E. (1994) Properties of fiber reinforced concrete using recycled fibers
from carpet industrial waste. Journal of Materials Science, Vol. 29, No. 16, pp. 41914199.
Wang, Y., Li, V.C., and Backer, S. (1991) Tensile failure mechanisms in synthetic fiber reinforced mortar. Journal of
Materials Science, Vol. 26, No. 24, pp. 65656575.
Li, V.C., Wang, Y., and Backer, S. (1991) A statistical-micromechanical model of tension-softening behavior of short fiber
reinforced brittle matrix composites, Journal of Mechanics and Physics of Solids, Vol. 39, No. 5, 1991, 607625.
Long, V.F. (1994) A new life for waste carpet at Shaw, Business Section, January 23, Citizen-News, Dalton, GA.
Castle, R. (1994) Recycling in the carpet industry. The 75th World Conference of the Textile Institute, September 2528,
Atlanta, GA.

RILEM
RILEM, The International Union of Testing and Research Laboratories for Materials and Structures, is an
international, non-governmental technical association whose vocation is to contribute to progress in the construction
sciences, techniques and industries, essentially by means of the communication it fosters between research and
practice. RILEM activity therefore aims at developing the knowledge of properties of materials and performance of
structures, at defining the means for their assessment in laboratory and service conditions and at unifying
measurement and testing methods used with this objective.
RILEM was founded in 1947, and has a membership of over 900 in some 80 countries. It forms an institutional
framework for cooperation by experts to:
optimise and harmonise test methods for measuring properties and performance of building and civil engineering
materials and structures under laboratory and service environments;
prepare technical recommendations for testing methods;
prepare state-of-the-art reports to identify further research needs.
RILEM members include the leading building research and testing laboratories around the world, industrial research,
manufacturing and contracting interests as well as a significant number of individual members, from industry and
universities. RILEMs focus is on construction materials amd their use in buildings and civil engineering structures,
covering all phases of the building process from manufacture to use and recycling of materials.
RILEM meets these objectives though the work of its technical committees. Symposia, workshops and seminars
are organised to facilitate the exchange of information and dissemination of knowledge. RILEMs primary output is
technical recommendations. RILEM also publishes the journal Materials and Structures which provides a further
avenue for reporting the work of its committees. Details are given below. Many other publications, in the form of
reports, monographs, symposia and workshop proceedings, are produced.
Details of RILEM membership may be obtained from RILEM, cole Normale Suprieure, Pavillon du Crous, 61,
avenue du Pdt Wilson, 94235 Cachan Cedex, France.
RILEM Reports, Proceedings and other publications are listed below. Full details may be obtained from E & FN
Spon, 26 Boundary Row, London SE1 8HN, Tel: (0) 71865 0066, Fax: (0) 71522 9623.
Materials and Structures
RILEMs journal, Materials and Structures, is published by E & FN Spon on behalf of RILEM. The journal was
founded in 1968, and is a leading journal of record for current research in the properties and performance of building
materials and structures, standardization of test methods, and the application of research results to the structural use
of materials in building and civil engineering applications.
The papers are selected by an international Editorial Committee to conform with the highest research standards.
As well as submitted papers from research and industry, the Journal publishes Reports and Recommendations
prepared by RILEM Technical Committees, together with news of other RILEM activities.
Materials and Structures is published ten times a year (ISSN 00255432) and sample copy requests and
subscription enquiries should be sent to: E & FN Spon, 26 Boundary Row, London SE1 8HN, Tel: (0) 71865 0066,
Fax: (0) 71522 9623; or Journals Promotion Department, Chapman & Hall Inc., One Penn Plaza, 41st Floor, New
York, NY 10119, USA, Tel: (212) 564 1060, Fax: (212) 564 1505.
RILEM Reports
1
2
3
4
5
6
7
8
9
10
11
12
13

Soiling and Cleaning of Building Facades


Corrosion of Steel in Concrete
Fracture Mechanics of Concrete StructuresFrom Theory to Applications
GeomembranesIdentification and Performance Testing
Fracture Mechanics Test Methods for Concrete
Recycling of Demolished Concrete and Masonry
Fly Ash in ConcreteProperties and Performance
Creep in Timber Structures
Disaster Planning, Structural Assessment, Demolition and Recycling
Application of Admixtures in Concrete
Interfacial Transition Zone in Concrete
Performance Criteria for Concrete Durability
Ice and Construction

229

RILEM Proceedings
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30

Adhesion between Polymers and Concrete


From Materials Science to Construction Materials Engineering
Durability of Geotextiles
Demolition and Reuse of Concrete and Masonry
Admixtures for ConcreteImprovement of Properties
Analysis of Concrete Structures by Fracture Mechanics
Vegetable Plants and their Fibres as Building Materials
Mechanical Tests for Bituminous Mixes
Test Quality for Construction, Materials and Structures
Properties of Fresh Concrete
Testing during Concrete Construction
Testing of Metals for Structures
Fracture Processes in Concrete, Rock and Ceramics
Quality Control of Concrete Structures
High Performance Fiber Reinforced Cement Composites
Hydration and Setting of Cements
Fibre Reinforced Cement and Concrete
Interfaces in Cementitious Composites
Concrete in Hot Climates
Reflective Cracking, in Pavements
Conservation of Stone and other Materials
Creep and Shrinkage of Concrete
Demolition and Reuse of Concrete and Masonry
Special ConcretesWorkability and Mixing
Thermal Cracking in Concrete at Early Ages
Concrete Technology: New Trends, Industrial Applications
Disposal and Recycling of Organic and Polymeric Construction Materials
Durability of Building Sealants
Non-Metallic (FRP) Reinforcement for Concrete Structures
Freeze/Thaw Durability of Concrete

RILEM Recommendations and Recommended Practice


RILEM Technical Recommendations for the Testing and Use of Construction Materials Autoclaved Aerated
ConcreteProperties, Testing and Design

Author index

Arikawa, S. 261
Bao, Y. 31

Ohama, Y. 169
Ono, S. 235
Ono, T. 235

Carrasquillo, R.L. 61

Park, H.S. 3

Demura, K. 169

Rebeiz, K.S. 13
Roller, A. 199

Farshad, M. 181, 199


Fleler, P. 181, 199
Fowler, D.W. 13, 31, 61
Fukushima, T. 111
Furukawa, S. 137
Hada, K. 87
Hayashi, Y. 87
Hosoya, S. 247

Sander, D. 61
Sano, T. 235
Sasse, H.R. 75
Satoh, T. 169
Soh, S.Y. 271
Soh, Y.S. 3, 271
Sugiura, A. 151
Suzuki, S. 223

Iibachi, H. 87
Iji, M. 126

Takata, S. 247
Tomimura, Y. 247

Kamamura, M. 209
Kasai, Y. 209
Kirikoshi, K. 261
Kitamura, T. 101
Kojima, A. 137
Kumano, T. 235
Kurihara, K. 247
Kwasny-Echterhagen, R. 75

Wakasugi, M. 151
Wang, Y. 297
Whitney, D.P. 31
Yamada, K. 157
Yamada, M. 47
Yamamoto, R. 189
Yamamoto, T. 261
Yamazaki, T. 87
Yanagi, K. 111
Yokoyama, S. 126
Yoshikane, T. 281

Lee, D.S. 3, 271


Lehmkmper, O. 75
Machida, K. 209
Maeda, T. 111
Mihashi, H. 157, 261
Motavalli, M. 181
Mukawa, Y. 235

Zhou, J.D. 209

Naniwa, R. 87
Narita, T. 261

230

Subject index

This index has been compiled from the keywords assigned to the papers, edited and extended as appropriate. The
page references are to the first page of the relevant paper.
Aggregates 75
Artificial woods 169
Asphalt
concrete pavement 281
concrete, recycled hot-mixed 281
mixing into 101
mixture 47
Autoclave curing 157

Fibre reinforcement 261


Field study 297
Filler 75
Flexible reinforcement 235
Flexural strength 157, 271
Forest preservation 235
FRP powder 137, 157, 169
scrap 151

Bending property 223


Bending test 47
Bulk specific gravity 169

Glass fibre 235


Glass powder, scrap 151
Glass transition temperature 3

Carbon continuous fibre reinforced plastic


pitch-based high-performance type 111
Carbon fibres 169
Carpet industrial waste 297
Cement mortar, mixing into 101
Cement paste 209
autoclaved 151
injection, method of 209
Cementitious composite 261
Ceramic tile, exterior fall-off 87
Chemically resolvable resin 111
Composite board 223
Concrete
fibre reinforced 297
form 248
Conifer plywood 235

Hardness 169
Honeycomb-shaped woven net 87
Hybrid structure 111
IC package 126
Interlocking block 271
Lauan plywood 235
Lightweight precast concrete 111
Lignin 248
Lumber, scrap 235
Marshall test 47
Mechanical properties 3
Mill-grinding 189
Moisture absorption 271
Molding resin 126
Mortar 75
asbestos-free 157
lightweight 137

Deformed rebar 111


Degree of unsaturation 3
Differential movement 87
Dry process defibrator 248
Durability 281
Dynamic stability 47

Nail withdrawal 169


Noise abatement 181

Environment-conscious materials (ecomaterials) design 111


Extrude 157

OSB 223
Ozone treatment 248

231

232

SUBJECT INDEX

Paper sludge ash 271


Particle board 223, 235
Pavement recycling 281
Paving material 47, 189
Plaster 75
Plywood, alternative material to 235
Polyester 13, 31
Polyethylene 13
Polymer
concrete 13, 31
latex 271
mortar 3, 87
Polyurethane 31
Porosity 157
Portland cement concrete 61
Printing wiring board 126
Processing methods 75
Protective mats 199
Pulverization 101
PVC mats 199
Pyrolysis 101
Recycled materials 13, 31, 61
mats 199
PET 3
Recycling 47, 75, 111, 126, 137, 189
Re-recycling 281
Rubber
blocks, interlocking 189
mats 199
powdered 189
recycled 181
walls 181
Sandwich plate 261
Service life 111
Shredder-cutting 189
Silica stone powder 151
Solid residue 101
Specific gravity 209
Strengthening 137
Strengths 169, 209
Thermal conductivity 209
Thermoset composites 101
Timber, recycling of used 209
Tropical rainforest 248
Two-dimensional finite element method 87
Tyres 189
used 181
Unsaturated polyester resin
Waste paper 248, 261
plastics 47, 61, 75
wood 223
Water curing 157
W/C ratio 271
Wheel tracking test 47
Wood, recycled 223

Wood chip concrete 209


packing ratio 209
Woven fibre net 87
Zeolite, synthesis of 101

Das könnte Ihnen auch gefallen