Thermoplastic Composites

MATERIALS SCIENCE AND TECHNOLOGIES
THERMOPLASTIC COMPOSITES
EMERGING TECHNOLOGY,
USES AND PROSPECTS
No part of this digital document may be reproduced, stored in a retrieval system or transmitted in any form or
by any means. The publisher has taken reasonable care in the preparation of this digital document, but makes no
expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No
liability is assumed for incidental or consequential damages in connection with or arising out of information
contained herein. This digital document is sold with the clear understanding that the publisher is not engaged in
rendering legal, medical or any other professional services.
MATERIALS SCIENCE
AND TECHNOLOGIES
Additional books in this series can be found on Nova’s website

under the Series tab.
Additional e-books in this series can be found on Nova’s website

under the eBooks tab.
MATERIALS SCIENCE AND TECHNOLOGIES
EMERGING TECHNOLOGY,
USES AND PROSPECTS
ELIZABETH RITTER
EDITOR
New York
Copyright © 2017 by Nova Science Publishers, Inc.
All rights reserved. No part of this book may be reproduced, stored in a retrieval system or transmitted
in any form or by any means: electronic, electrostatic, magnetic, tape, mechanical photocopying,
recording or otherwise without the written permission of the Publisher.
We have partnered with Copyright Clearance Center to make it easy for you to obtain permissions to
reuse content from this publication. Simply navigate to this publication’s page on Nova’s website and
locate the “Get Permission” button below the title description. This button is linked directly to the
title’s permission page on copyright.com. Alternatively, you can visit copyright.com and search by
title, ISBN, or ISSN.
For further questions about using the service on copyright.com, please contact:
Copyright Clearance Center
Phone: +1-(978) 750-8400 Fax: +1-(978) 750-4470 E-mail: [email protected].
NOTICE TO THE READER

The Publisher has taken reasonable care in the preparation of this book, but makes no expressed or
implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is
assumed for incidental or consequential damages in connection with or arising out of information
contained in this book. The Publisher shall not be liable for any special, consequential, or exemplary
damages resulting, in whole or in part, from the readers’ use of, or reliance upon, this material. Any
parts of this book based on government reports are so indicated and copyright is claimed for those parts
to the extent applicable to compilations of such works.
Independent verification should be sought for any data, advice or recommendations contained in this
book. In addition, no responsibility is assumed by the publisher for any injury and/or damage to
persons or property arising from any methods, products, instructions, ideas or otherwise contained in
this publication.
This publication is designed to provide accurate and authoritative information with regard to the subject
matter covered herein. It is sold with the clear understanding that the Publisher is not engaged in
rendering legal or any other professional services. If legal or any other expert assistance is required, the
services of a competent person should be sought. FROM A DECLARATION OF PARTICIPANTS
JOINTLY ADOPTED BY A COMMITTEE OF THE AMERICAN BAR ASSOCIATION AND A
COMMITTEE OF PUBLISHERS.
Additional color graphics may be available in the e-book version of this book.
Library of Congress Cataloging-in-Publication Data

ISBN: (eBook)
Published by Nova Science Publishers, Inc. † New York

CONTENTS
Preface vii
Chapter 1 Thermoplastic Polyurethane Nanocomposites 1
Marija V. Pergal, Milica Balaban,
Biljana Dojčinović and Dragan Manojlović
Chapter 2 Innovative Thermoplastic PET Powder Coating
in Telecommunications Field 61
Yukitoshi Takeshita, Takashi Miwa, Azusa Ishii
and Takashi Sawada
Chapter 3 Mechanical Recycling of Thermoplastic Composites 95
Nathalie Benoit, Rubén González-Núñez
and Denis Rodrigue
Chapter 4 A Suggestion for Standardizing a Traceable Process
for the Determination of the Mechanical Properties
of Concrete Containing Thermoplastic Polymers as
Aggregates 143
Stamatia Gavela, Georgios Papadakos
and Vasilia Kasselouri – Rigopoulou
Index 165
PREFACE
Thermoplastics composites are widely used nowadays. Their matrices are

characterized by their aptitude to be repeatedly melted by heating and
solidified by cooling, and this in a reversible way. The reinforcement add
rigidity and strength to the matrix, thus making a material combining the
properties of both constituents. Thermoplastic composites have become an
area of increased interest due to their relatively low cost, wide range of
mechanical properties, toughness, problem free storage, resistance to chemical
attack, ease of processability, and above all their better recyclability. This
book reviews emerging technologies, uses and prospects of thermoplastic
composites. Chapter One summarizes the preparation, structure-property
relationships and potential applications of thermoplastic polyurethane
nanocomposites. Chapter Two discusses innovative thermoplastic PET powder
coating in the telecommunications field. Chapter Three focuses on mechanical
recycling of thermoplastic composites. Chapter Four provides a suggestion for
standardizing a traceable process for the determination of the mechanical
properties of concrete containing thermoplastic polymers as aggregates.
Chapter 1 – Thermoplastic polyurethanes (TPUs) are an important class of
synthetic polymers with many industrial applications, whose properties can be
custom-made by simply adjusting the composition to meet the highly
diversified requirements of modern technology. TPUs are multiblock
copolymers composed of alternating soft and hard segments. Significant
changes of thermal, mechanical, surface, flame retardant and barrier properties
of TPUs have been achieved by their combination with different nanoparticles.
A large interfacial surface between TPUs and nanoparticles has led to the
possibility for new properties to be created. These properties are often
demonstrated as being necessary in order to meet all requirements for specific
viii Elizabeth Ritter
applications. Properties of TPU-based nanocomposites strongly depend on the

preparation method, the type and content of hard and soft segments in the
TPUs, the type and content of nanoparticles, and intensity of interfacial
interactions between the polymer matrix and nanoparticles. In this chapter,
recent development on nanocomposites based on TPUs and different types of
nanoparticles is presented. The current understanding of nanocomposites
based on TPUs is reviewed in this chapter within the following topics: (a)
conditions for nanoparticle preparation in order to maximize their good
dispersion in the TPU matrix, (b) different methods for preparation of TPU-
based nanocomposites and (c) properties of TPU-based nanocomposites
depending on the type, size and shape of the incorporated nanoparticles. The
research into TPU-based nanocomposites, including fundamental structure-
property relationships and potential applications of TPU-based
nanocomposites is discussed. In this chapter, future prospects for TPU-based
nanocomposites are also presented.
Chapter 2 – Polymeric materials such as those used for coatings offer
highly effective protection against corrosion of outdoor steel structures in the
industrial telecommunications fields. Conventionally, tar epoxy resin has been
used as a coating for steel telephone poles for a long time in Japan. However,
degradation of the resin and steel itself become noticeable 20 to 30 years after
construction. We need to change the material to meet requirements from both
the eco-friendliness and durability standpoints. Thermoplastic polyethylene
terephthalate (PET) provides a good coating with high toughness, good
flexibility, and a large thickness without fine pinholes. In addition,
thermoplastic PET is an eco-friendly material because PET recycling
technology has been well established. The authors here describe the challenges
they faced in applying thermoplastic PET powder coating to steel structures in
the telecommunications field, and present research results on treatment in
polymer crystallinity and composites for higher durability against alkalinity as
well as against any harsh environment. PET is a semi-crystalline polymer in
which the crystallinity is governed by crystallization kinetics represented by,
for instance, the Vogel Fulcher and Abrami’s equation. It is therefore possible
to optimize coating properties by controlling the crystallinity, which can be
further controlled by the air-cooling time in the industrial fluidized powder
vessel process. Thermoplastic PET polymer is generally vulnerable to
hydrolysis, especially under alkaline conditions, because hydroxyl ions attack
the ester group in PET, causing it to decompose. To enhance the durability of
thermoplastic PET powder coating even in harsh outdoor environments such
as alkalinity, thermoplastic PET is mixed with a secondary resin in which
Preface ix
primary PET is on the bottom layer and the secondary resin is on the surface
layer so as to protect the primary PET. Polyvinyl butyral (PVB) material has
been selected as a secondary resin. The resulting PET/PVB composite coating
presents a relatively good mechanical stress-strain property as a coating and an
enhanced alkaline resistance, with PET presenting a good mechanical property
and PVB suppressing the hydrolysis of PET. The composite also exhibits good
performance in terms of sea salt water resistance, UV resistance, and heat
cycle resistance in the environment. Innovated thermoplastic PET, with the aid
of a secondary material, is essential to the function and integrity of steel
structures in industrial telecommunications fields in terms of both eco-
friendliness and long-term durability.
Chapter 3 – Thermoplastic composites have found several applications
over the years, but their main advantage when compared with thermoset
composites is their possibility to be reshaped after processing through further
melting and remolding processes. This allows the materials to be recovered
and reprocessed if they fail quality control (post-industrial) or after their end of
life (post-consumer). This is very attractive as recycling is one of the main
principles of sustainable development. This is even more appropriate as
thermoplastic composites are more and more produced and consumed in a
wide range of applications such as packaging, automotive and aeronautics,
furniture, building and construction, as well as sport and leisure goods. Thus
increasing demands leads to a high number of parts being produced and
discarded. With this high volume of composites production and use, the
emergence of constantly new and stronger policies and laws towards
sustainable development and recycling were developed. This is why it
becomes important to study and understand the conditions of the materials at
their end of life and develop applications like recycling to reintroduce these
high amounts of materials into production lines. To limit both the amount of
material consumed and landfilled, thermoplastic composite waste can be
considered as a source of raw material for the manufacture of products through
recycling and reprocessing. However, to obtain good final properties, it is
essential to understand the degradation processes and the materials behavior
during their life cycle and their recycling. In this chapter, a review of the
different possibility to recycle thermoplastics composites is presented, with a
focus on the mechanical recycling techniques available. Some figures about
composites production and recycling are presented and discussed, followed by
a brief presentation of the different recycling methods available for
thermoplastic composites with their advantages and limitations. Then, an
overview of mechanical recycling is made considering both organic and
x Elizabeth Ritter
inorganic fillers. For all cases, the material behavior towards recycling and the
resulting properties are considered, as well as the solutions developed to
improve the performance and quality of the recycled materials.
Chapter 4 – A lot of research has been performed on the utilization of
waste products in concrete. Due to the fact that concrete is widely used and
has a long service life, the wastes used in it are removed from the waste stream
for a long period. One of the waste types studied has been thermoplastic waste.
Polymers become increasingly a preferable material by product designers
because they provide useful properties as design flexibility and significant
weight savings. For this reason the worldwide production has grown, with
packaging being the main application for plastics. Because disposal of post-
consumer plastics is increasingly being constrained by legislation, there is
considerable demand for alternatives leading to effective life cycle
management. As the amount of aggregates required in concrete production is
large, the environmental benefits of the replacement of natural aggregates by
thermoplastic waste is not only related to its safe disposal, but also to the
mitigation of environmental impacts arising from the extraction of aggregates,
i.e., the visual intrusion and the loss of countryside. On the other hand,
designers need to feel confident about the mechanical properties of the product
in order to be encouraged to add the maximum practicable amount of
thermoplastics in concrete or mortars. Studies have been conducted towards
the assessment of mechanical properties of concrete or mortar containing
thermoplastic polymers as aggregates. Due to (a) the variability of
thermoplastic polymers which could be added to concrete, (b) their significant
differences from natural aggregates and (c) the variability of parameters
affecting the properties of concrete itself, the data obtained from tests on
mechanical properties for concrete specimens containing thermoplastic
polymers are based on a multifactorial process. When these results come from
totally different laboratory processes, their metrological traceability is of great
significance in order to achieve inter-comparability. A standardized protocol
for the experimental design and referencing of all essential relative data is
needed in order to facilitate any attempt to compose the results from studies
when thermoplastic polymers of different origin and characteristics are used
and the percentage of aggregate replacement varies. This study clarifies that
different reporting on the total experimental procedure arises difficulties in
composing results coming from different laboratories and the statistical
inference on the effect of natural aggregates replacement by thermoplastics. A
standardized protocol based on (a) metrological traceability as determined by
ISO and (b) quantitative review principles and procedures, is proposed to
Preface xi
facilitate comparability and compatibility of future results. This protocol aims

at a quantitative review providing a statistical basis for the estimation of the
Compressive Strength Reduction (CSR) for a concrete or mortar mix where
conventional aggregates have been, partially or totally, replaced by
thermoplastic polymers. A reference mix was assessed necessesary to be
properly incorporated in the experimental design of any such study. This
quantitative review was succesfully initiated but needs further extention by the
execution of more experiments that should, in any case, follow the proposed
protocol leading to an accurate model aiming at the maximization of this
thermoplastics use.
In: Thermoplastic Composites ISBN: 978-1-53610-707-4
Editor: Elizabeth Ritter © 2017 Nova Science Publishers, Inc.
Chapter 1
THERMOPLASTIC POLYURETHANE
NANOCOMPOSITES
Marija V. Pergal1,*, Milica Balaban2, Biljana Dojčinović1

and Dragan Manojlović3
1
Institute of Chemistry, Technology and Metallurgy (ICTM)-Center of
Chemistry, University of Belgrade, Belgrade, Serbia
2
Faculty of Sciences, University of Banja Luka,
Banja Luka, Bosnia and Herzegovina
3
Faculty of Chemistry, University of Belgrade, Belgrade, Serbia
ABSTRACT
Thermoplastic polyurethanes (TPUs) are an important class of
synthetic polymers with many industrial applications, whose properties
can be custom-made by simply adjusting the composition to meet the
highly diversified requirements of modern technology. TPUs are
multiblock copolymers composed of alternating soft and hard segments.
Significant changes of thermal, mechanical, surface, flame retardant and
barrier properties of TPUs have been achieved by their combination with
different nanoparticles. A large interfacial surface between TPUs and
nanoparticles has led to the possibility for new properties to be created.
These properties are often demonstrated as being necessary in order to
meet all requirements for specific applications. Properties of TPU-based
*
Corresponding Author address. Email: [email protected]
2 Marija V. Pergal, Milica Balaban, Biljana Dojčinović et al.
nanocomposites strongly depend on the preparation method, the type and

content of hard and soft segments in the TPUs, the type and content of
nanoparticles, and intensity of interfacial interactions between the
polymer matrix and nanoparticles. In this chapter, recent development on
nanocomposites based on TPUs and different types of nanoparticles is
presented. The current understanding of nanocomposites based on TPUs
is reviewed in this chapter within the following topics: (a) conditions for
nanoparticle preparation in order to maximize their good dispersion in the
TPU matrix, (b) different methods for preparation of TPU-based
nanocomposites and (c) properties of TPU-based nanocomposites
depending on the type, size and shape of the incorporated nanoparticles.
The research into TPU-based nanocomposites, including fundamental
structure-property relationships and potential applications of TPU-based
nanocomposites is discussed. In this chapter, future prospects for TPU-
based nanocomposites are also presented.
Keywords: polyurethanes, nanocomposites, thermoplastics, nanoparticles,

structure-property relationships
INTRODUCTION
This chapter summarizes the preparation, structure-property relationships
and potential applications of thermoplastic polyurethane nanocomposites.
Thermoplastic polyurethanes (TPUs) and their nanocomposites (NCs) belong
to the class of thermoplastic elastomers which have been prepared and
investigated for some time with the goal of achieving practical applications,
such as in implants, tissue engineering, scaffolds, drug delivery, electronics or
as coatings or packaging materials [1-6].
Polyurethanes can be synthesized by a polyaddition reaction of three basic
components: diisocyanate, a short-chain diol and a long-chain macrodiol [1, 2,
5, 7]. The resultant thermoplastic is a block copolymer that has a linear chain
structure (Figure 1). Most commercially used polyurethanes are multi-block
copolymers, such as AvcothaneTM 51 (Arrow International), PellethaneTM 2363
(Dow Chemical Co.), Elast-EonTM (Elastomedics Pty Ltd.), CorethaneTM (The
Polymer Technology Group), Angioflex® (Applied Biomedical Corp.),
VialonTM (Becton Dickinson Polymer Research), BiomerTM (Ethicon Corp.)
and Actifit® (Orteq® Sports Medicine) [8]. Segmented TPU elastomers are
multiblock copolymers composed of alternating soft and hard segments [2, 5,
8]. The hard segments are composed of diisocyanate (aliphatic or aromatic)
and chain extender (low molecular weight diol), while the soft segments
Thermoplastic Polyurethane Nanocomposites 3
consist of long-chain macrodiol with a molecular weight of 1000-4000 g/mol.

Classical macrodiols based on polyester and polyether have been used in a
number of TPUs [9-17], but nowdays, poly(dimethylsiloxane)- and
polycarbonate-based TPUs are more commonly prepared and investigated [18-
33]. As a result of thermodynamic incompatibility of the hard and soft
segments with their different natures, microphase separation, mostly on the
sub-nanometer and nanometer scale, occurs in TPUs. The phase consisting of
hard, rigid segments performs as nodes of the physical network, imparting
elastomeric behavior to the soft phase, while soft segments provide these
materials with high-elastic properties.
In TPU production, two basic procedures are used: one-stage and two-
stage (i.e., prepolymer method) in the solution or melt [7]. From the
technological point of view, it is important that various TPU materials can be
prepared from the same reaction mixture just by changing the reaction
conditions, such as temperature, time, molar ratio of isocyanate to hydroxyl
groups, catalyst and solvent. The functional, e.g., the mechanical, thermal,
surface and morphological properties of TPU elastomers depend on a number
of factors, such as the chemical nature of the initial components (macrodiol,
diisocyanate, chain extender), the contents and lengths of the soft and hard
segments, the degree of the phase separation/mixing, the strength of hydrogen
bonds, the mobility and orientation (packing) of the building units within the
hard segments, reaction conditions during TPU preparation and thermal
history of the TPU. Adding different sizes or shapes of nanoparticles (clay,
silica, nanotube, graphene, metal, cellulose nanocrystal, etc.) at low
concentrations is another method to modify the functional properties of TPUs
[34-44]. The use of TPU as a nanocomposite matrix presents interesting
challenges in understanding the nanoscale and microscale morphology, due to
the pre-existing nanoscale morphology of the segmented TPU domain.
Recently, development of environmentally friendly waterborne TPUs
(WTPUs) has been intensifying, especially in the coating industry field where
reducing the evolution of volatile organic compounds during the drying
process is critical [45-48]. WTPUs have many advantages, such as low
viscosity at high molecular weight, non-toxicity and good applicability, which
cannot be obtained with conventional solvent-borne systems. A few WTPU-
based NCs have also been developed. WTPUs could also be utilized in
electronic devices as antistatic coatings or electromagnetic shielding if they
could be modified for improved conductivity. This can be achieved by adding
conductive nanoparticles with an extremely low loading.
Figure 1. The chemical structure of the thermoplastic polyurethanes based on poly(ε-

caprolactone)-b-poly(dimethylsiloxane)-b-poly(ε-caprolactone) (PCL-PDMS-PCL) as
the soft segment and methylenediphenyl diisocyanate - 1,4-butanediol (MDI-BD) as
the hard segment [21].
TPUs combine the mechanical properties of conventional thermoplastics

(elasticity and flexibility) with the characteristics of chemically cross-linked
elastomers (high strength and modulus) [2, 7]. They have an important
advantage over conventional elastomers, as they can be processed as
thermoplastic materials above the glass transition and melting temperatures of
the hard segment [49]. This phenomenon is very important for TPU recycling
processes. TPUs can be processed by the usual melt-processing methods, such
as extrusion, injection molding, and calendering. Since some of the
compositions can be dissolved in solvents, they can also be processed by
solution fabrication.
TPUs are very important materials in spite of the fact that TPU production
accounts for just one-eighth of polyurethane market demand [50]. They
currently comprise approximately 15% of the volume of thermoplastic
elastomers sold annually. TPUs can be produced with properties ranging from
soft to hard elastomers with excellent physical properties, including high
tensile strength, abrasion, durability and tear-resistance. Tremendous
versatility in terms of chemical structure originates from a wide range of initial
reactants, affording tailor-made properties with well-designed combinations of
these reactants. Therefore, TPUs can be easily manipulated to satisfy the
highly diverse requirements of modern technologies. However, TPUs have
some disadvantages, i.e., poor thermal stability and poor gas-barrier properties
[51], although these disadvantages can be partially overcome by controlling
TPU microstructure or by introducing nanoparticles [52].
TPU-based NCs are already established as materials with remarkably
improved properties compared to pure TPUs and conventional
microcomposites. The main goal of nanoparticle-containing TPU matrices is to
develop new and improved nano-sized materials, which are impossible to
obtain using conventional micro-size fillers. Introducing various types of
nanoparticles into TPUs can improve the thermal, mechanical, flame retardant,
gas barrier, electrical and/or biological properties of the final NCs. Favorable
polymer-nanoparticle interactions in TPU-based NCs provide improvements
that are reflected through keeping the inherent properties of the nanoparticles
and also increasing stability, as provided by the TPU matrix. Furthermore,
TPU-based NCs can also provide some new properties which cannot be
observed in the individual components, owing to local arrangements or
organization of the nanoparticles in the polymer matrix. The following chapter
reviews some recent developments being made in these fields.
THERMOPLASTIC POLYURETHANE NANOCOMPOSITES:

GENERAL CONSIDERATIONS
TPU-based NCs are multiphase materials, where one of the phases has
nanoscale particles. They constitute polymer matrix filled with a low amount
of inorganic/organic nanoparticles, which have an important influence on the
macroscopic properties of the polymer matrix and are capable of interacting, at
the molecular level, with the polymer matrix [53-55]. If a low amount of
nanoparticles is successfully dispersed in a polymer matrix, the interface
between the polymer matrix and the nanoparticles increases, allowing the
nanoparticles to fully contribute to improving the final product’s properties.
The nanoparticles can be grouped according to their shape aspect ratio (length-
to-thickness ratio), and one can distinguish zero- (e.g., fullerenes), one- (e.g.,
carbon nanotubes), two- (e.g., layered silicates) and three-dimensional (e.g.,
graphite and spherical particles) versions [49]. In general, nanoparticles in
NCs were previously considered to be 1-100 nm in size [56, 57]. However, it
has been established that particles above 50 nm do not tend to improve the NC
properties as expected. Nanoparticles have significantly different physical and
chemical properties from their bulk materials and micro- and macro-additives,
due to their high surface area-to-volume ratios, i.e., high aspect ratios [58].
TPU-based NCs are expected to show unusual properties that arise from the
combination of each component.
A very important issue during the preparation of TPU-based NCs is
preventing aggregation of nanoparticles and formation of agglomerates, which
can be caused by the high interfacial reactivity and high surface tension energy
of nanoparticles [59]. Nanoparticles can aggregate during the polymerization
procedure or in the initial reaction medium (e.g., solvent). A high extent of
nanoparticle dispersion within the TPU matrix, which is very hard to achieve
due to the incompatibility of hydrophobic polymers with nanopaticles, as their
surface is usually hydrophilic, plays a key role in improvement of properties
[40]. According to the literature, in order to improve dispersion of

nanoparticles within the TPU matrix and compatibility of nanoparticles with
the polymer matrix, the following technique are used: ultrasonic treatment,
high-energy mechanical milling, surface modification of nanoparticles using
suitable modifiers (chemical method), treatment of nanoparticles with
surfactants (physical method) and encapsulation of nanoparticles with
polymers which act as stabilizing agents [59, 60].
Generally, polymer-based NCs can be prepared by ex situ methods (direct
compounding) [55, 61-64], or in situ methods [55, 65-69]. In the ex situ
methods, polymer and nanoparticles are prepared separately and mixed in a
solution, in melt or by using mechanical forces. Dispersion of nanoparticles
within the polymer matrix depends on the processing conditions, such as
configuration of the reactor, temperature, time and shear force, and can be
improved by functionalization of nanoparticles or the polymer matrix. In situ
methods can be devided into two main groups, according to the starting
materials and proccesing techniques:
a) Nanoparticle precursors are preloaded into polymer matrix and then

the precursors are exposed to conditions necessary for the in situ
synthesis of the nanoparticles.
b) Nanoparticles are firstly dispersed into the monomers or macrodiols
and then the mixture is polymerized under desirable conditions.
The advantages of in situ methods are better dispersion control of

nanoparticles within the polymer matrix and and the ability to produce clearly-
defined polymer-NCs with exfoliated structures.
The properties of the NCs are affected by the properties of the
components, type of polymer, type, shape, size and concentration of the
nanoparticles, the morphology of the system, the nature of the interphase that
sometimes develops at the interface of the two components and
nanoparticle/polymer matrix interaction [54]. The extent of property
enhancement depends on many parameters including the aspect ratio
(length/diameter) of the nanoparticle, its degree of dispersion and orientation
in the matrix, and the adhesion at the nanoparticle-matrix interface. The choice
of the polymer is affected mainly by their mechanical and thermal behavior,
hydrophobic/hydrophilic balance, chemical stability, biocompatibility, optical
and/or electronic properties [70]. The outstanding mechanical, thermal, barrier
and electro-physical properties achieved at low nanoparticle content, are the
result of physical and chemical interphase interactions in TPU-based NCs
[57]. The introduction of nanoparticles can enhance or reduce thermal stability

and change the degree of crystallinity of TPU matrix [57]. Moreover, silver
(Ag) and gold (Au) nanoparticles can improve the antimicrobial properties of a
polymer matrix [71, 72], enhance fibroblast attachment and endothelial cell
response, as well as reduce monocyte and platelet activation on the surface of
TPU NCs, relative to pure TPU [73]. The presence of nanoparticles can
change the glass transition temperature, Tg, of a polymer matrix [57]. Changes
in Tg are correlated with changes in polymer chain dynamics and chain
relaxation behavior. Tg value is affected by the nature of the polymer, the
content of the nanoparticles, thermal history of the NC and on the interfacial
interaction between the polymer matrix and nanoparticles. Both increases and
decreases in the Tg have been reported, dependant upon the interaction
between the TPU matrix and the nanoparticle. The strong interactions of TPU
with clay in TPU/clay NCs can lead to the dispersion of organic and inorganic
phases at the nm level, resulting in enhanced tensile properties, increased
strength and heat resistance, high moduli, increased thermal stability,
decreased gas permeability and flammability and increased biodegradability,
which may not occur in their conventionally filled micro-counterparts [6, 74-
77].
THERMOPLASTIC POLYURETHANE/
CLAY NANOCOMPOSITES
TPU/clay NCs as an important familiy of mineral-reinforced thermoplastic
materials have been intensively studied in the last two decades as promising
nano-engineered materials that achieve outstanding properties compared to
neat TPUs. The first examples of TPU/clay NCs were reported in the literature
in 1998 by Wang and Pinnavaia [78]. Layered alumosilicate clay and
especially montrillomonite are mosty used in preparation of TPU-based NCs
because clays are environmentally friendly, easily available and low cost
materials [57, 76]. Montrillomonite consists of layers made up of two
tetrahedral structures of silica dioxide, while the center layer is composed of a
metal oxide octahedral layer [57]. The layered silicate is 1 nm thick, its lateral
dimensions range from 30 nm to several microns, it has a high aspect ratio
(10-1000) and is organized in stacks with regular gaps between them, called
interlayers or galleries. The clay layers are negatively charged, with this
charge being counterbalanced by alkali or alkaline earth cations that reside in
the galleries. In order to uniformly disperse clay within a polymer matrix, clay
layers must be modified with some hydrophobic surfactants by exchanging the
interlayer inorganic cations with alkylammonium or alkylphosphonium ions
[6, 76]. Replacing the inorganic cations in the native clay galleries with
organic ions leads to better compatibility between the inorganic clay and the
hydrophobic polymer matrix. Clay modification also lowers the surface energy
of the clay layers and increases the polymer-clay interfacial interactions which
lead to an increase in the distance between clay layers, allowing the polymer
chain to more easily approach [6]. A range of commercial organoclays is
available for the preparation of TPU/clay NCs (different Cloisite® clays
produced by Southern Clay Products, Inc. USA; clays produced by Nanocor
Inc. USA, clays produced by CO-OP Chemical Co., Ltd. Japan; etc).
In TPU/layered silicate NCs, the aim is to obtain full, nanometer-level
dispersion of clay layers (exfoliation) in order to achieve the best properties
[79-81]. It was reported that intercalation of polymer chains into the interlayer
space resulted in a well-ordered multilayer morphology with some clay layers
in dispersed stacks; this essentially improved the polymer properties [40, 82].
Moreover, the particular structure that is obtained for polymer-based NCs
is affected by the favorable polymer-clay interactions and the preparation
method. Based on the interaction of clay with polymer, three different types of
polymer/clay composites can be formed [6], as presented in Figure 2. The
affinity between the polymer matrix and the organoclay is determined, to some
extent, by the polarity of the polymer and the type of organic modifier used to
form the organoclay [83]. If there is very poor affinity to the polymer, clay
layers will not expand and thus, phase separated microcomposite will be
obtained. In intercalated structures, moderate affinity between clay and
polymer exists, so the polymer chains become sandwiched between the silicate
layers, indicating limited dispersion. In exfoliated structures, silicate layers are
well separated into individual platelets and, due to the high affinity to the
polymer, silicate layers are uniformly dispersed within the continuous polymer
matrix. The uniform dispersion of the anisotropic nano-sized particles can lead
to a large interfacial area between the constituents at extremely low loadings
of the nanoparticles. Moreover, improved or new thermal, mechanical, surface,
optical, and electrical properties can be developed, which may not occur in the
macroscopic counterparts. It has been established that exfoliated structures
generally show better properties than intercalated structures of polymer-based
NCs [79-81,84]. In many cases, partial intercalation and partial exfoliation of
clay layers occur in the TPU/clay NCs. It was reported that the size of the
layered silicate can affect the resulting nanocomposite morphology [85]. The
number of polymer-nanoparticle interactions increases with decreasing the

nanoparticle size, due to the smaller particles having a higher surface area and
a smaller interparticle distance. Also, tensile strength can be significantly
improved with smaller organoclay nanoparticles, without changing Young’s
modulus.
Figure 2. Schematic illustration of different types of polymer/clay composites.
MODIFICATION OF CLAY LAYERS

The aim of modifying the clay layers is to make the layered silicates
compatible with the TPU matrix, by lowering the surface energy, increasing
the interlayer spacing of clay and improving the wetting characteristics of the
polymer matrix [86]. The hydrophilic surface of silicate layers should be
converted to an organophilic one by exchanging the cations present inside the
clay galleries with suitable cationic surfactants such as quaternary alkyl
ammonium and alkyl phosphonium ions (ion-exchange reactions), allowing
the intercalation of polymer. The surface properties of clay can be tailored by
changing the type and length of alkyl chain. Moreover, the alkylammonium or
alkylphosphonium cations can provide functional groups which interact with
polymer chains and, therefore, increase the interfacial interactions. Attractive
interaction between the surfactant and the polymer greatly enhances the
possibility of dispersion of the clay within the polymer matrix. According to
Xiong et al. [87], there are two important factors that should be considered in
obtaining polymer/clay NCs. Firstly, the organic modifier molecules must
enter the interlayer galleries and increase the interlayer space in the clay.
Secondly, this results in the easier interaction of polymer molecules or
monomers and clay layers. The conversion of clays into organophilic
compounds can be also performed by exchanging the cation present inside the
clay galleries using organosilane grafted to the edge of hydroxyl groups on the
clay or interspersed between the clay layers on the clay surface, resulting in
mostly intercalated or partially exfoliated TPU-based NC morphologies [88].
Chavarria and Paul [83] reported the effect of organoclay structure on the
dispersion of TPU NCs prepared by melt mixing. It was shown that the
ammonium ion having one alkyl tail rather than two, the presence of
hydroxyethyl groups rather than methyl groups on the nitrogen and a longer
alkyl tail along with a hydroxyethyl group rather than shorter alkyl chain along
with a hydroxyethyl group lead to better clay dispersion. They found that TPU
with more hard segments, i.e., with more polar urethane segments, favors
better polymer-clay interaction [83]. Good dispersion of clay particles within
the TPU matrix can be achieved taking into account the following factors: the
affinity of macrodiol to swell the modified clays, which indicates the
compatibility of macrodiol and clay; a sufficient concentration of surfactant
used to exchange the clay, and; a longer alkyl chain (carbon number ≥ 12)
leading to the adequate swelling of the clay by macrodiol [78].
PREPARATION METHODS FOR

TPU/CLAY NANOCOMPOSITES
TPU/clay NCs can be prepared by in situ polymerization [78], solvent
casting [89], the melt intercalation method [90] and reactive extrusion [91].
Recently, numerous efforts have been focused on finding new procedures for
obtaining TPU-based NCs.
In situ polymerization is a good choice to produce well-dispersed silicate
layers within the polymer matrix. In in situ polymerization, the clay layers are
dispersed within the polymer matrix by polymerization, mixing the clay with
the monomer in the presence of a catalyst [78]. The advantage of this method
is reflected in the ease with which small monomer molecules penetrate into the
intergallery spaces compared with much larger polymer molecules. The
polymerization reaction then occurs between the clay layers, which is a
precondition for exfoliation structure formation.
In solvent casting, clay is first dispersed into selected solvent, often in the
presence high shear homogenization or ultrasonic energy, and then the
polymer and clay solutions are mixed [89]. Due to the lower viscosity,
macromolecules easily reach the surface of the platelets, and replace adsorbed
solvent. Finally, the solvent is allowed to evaporate, and the intercalated
polymer/clay NCs are formed in thin films. This method can lead to poor
dispersion of clay within the polymer matrix, as dispersion mainly depends on

the solvent (which must be able to completely dissolve the polymer and
disperse the nanoclay). However, the method is a high-cost one due to the
large amounts of solvent needed for appropriate clay dispersion, and there are
technical phase separation problems [75].
In the melt intercalative procedure, polymer and clay are mixed together
in the molten state using manual mixing, internal mixers, or single- or twin-
screw extruders. During the formation of NCs using this method, the polymer
chains are melted to increase their mobility and then the clay is dispersed into
the melt to allow intercalation of the clays [90]. The dispersion of the clay into
the polymer is achieved using shear and extrusion. Thermal degradation of
TPU must be avoided if this method of preparing TPU/clay NCs is used [92,
93].
Figure 3. Transmission electron microscopy micrographs of cross-sectional views of

PTMO based TPU with 39 wt.% hard (MDI-BD) segment and containing (a) 1 wt.% of
montmorillonite modified with 1OH, (b) 1 wt.% montmorillonite modified with 2OH,
and (c) 1 wt.% montmorillonite modified with 3OH: (I) indicated intercalated
structure; (II) indicated exfoliated structure [101].
Nowdays, commercial TPUs are usually manufactured by a reactive

extrusion method, comprising in situ polymerization of diisocyanate,
macrodiol and chain-extender in a twin-screw extruder. Clays can be added as
dry powders or pre-dispersed into the macrodiol [91]. This method is applied
to chemical modification of existing polymers and has some advantages over
solution casting, such as the absence of solvents (so this method is less
environmentally harmful), high-viscosity polymers can be used, and
processing conditions, such as mixing time and temperature are flexible and
thermal degradation of polymer can be more easily avoided [94].
The commercially available Cloisite® series of organoclays have been
used in a large number of studies concerning preparation and structure-
property relationships of TPU NCs. Also, in situ polymerization and solvent
casting for preparation of TPU-based NCs have been used in a large number of
TPU NCs studies. Many studies used solvent casting to avoid complications
due to thermal degradation of the alkylammonium modifiers [95]. Wang and
Pinnavaia [78] prepared intercalated TPU/clay NCs by in situ polymerization
of methylenediphenyl diisocyanate and a mixture of macrodiols and
alkylammonium-exchanged montmorillonites. They found that the tensile
strength, modulus, and strain at break increased by more than 100% at a
loading of 10 wt.% organoclay. Finnigan et al. [62, 96] studied TPU/clay NCs
prepared by solvent casting and found that using the hydrophilic Cloisite 30B
produced NCs with relatively better dispersion compared to Cloisite 15A, and
increased stiffness, tensile hysteresis, and permanent set was observed.
Interestingly, dispersion with the more hydrophobic Cloisite 15A was not
easier than with Cloisite 30B, although increases in tensile and barrier
properties were reported [96]. It has been shown that Cloisite 25A, in the
middle of the hydrophilicity/hydrophobicity range, also resulted in good
dispersion and improvement of NC properties [97, 98]. Chen et al. [99]
obtained an exfoliation structure of TPU/clay NCs by in situ polymerization
(in which the organoclay was added at the end of the second phase of
polyaddition), as shown by X-ray diffraction analysis and transmission
electron microscopy. In this study [99], clay was modified by 12-aminolauric
acid and benzidine separately. TPU-based NCs with benzidine-modified clay
(containing two terminal -NH2 groups that could participate in the reaction
with unreacted -NCO groups) showed much better mechanical properties than
that of treated clay with 12-aminolauric acid (containing one terminal -NH2
groups that could participate in the reaction with unreacted -NCO groups).
Therefore, linear branch chains can be formed by adding 12-aminolauric acid
to the TPU, and crosslinking can be formed by adding benzidine to the TPU.
This result was explained by the difference in the interaction between the
swelling agent and polyurethane molecular chain. Furthermore, they found a
two-fold increase in the tensile strength and a three-fold increase in the
elongation for 1% TPU/benzidine-modified clay as compared to that of pure
TPU, which can be attributed to the dispersion of exfoliated silicate layers
between the soft and the hard segments and the crosslinking effect [99]. In the
studies of Tien and Wei [100, 101], the authors prepared TPU/clay NCs by the
solvent method and found that with the increase of numbers of hydroxyl
groups (which acted as pseudochain extenders for polyurethane prepolymer
because of the hydroxyl groups in the clay modifier which reacted with
isocyanate-terminated polyurethane prepolymer) in quaternary ammonium
ions, the dispersion of layered silicates in TPU matrix transformed from an
intercalated to an exfoliated structure (Figure 3). Therefore, the presence of 3
hydroxyl groups per modifier favored formation of TPU NCs with an
exfoliation structure. Also, the tethering of polymer chains to the organoclay
surface has been used to obtain a strong interfacial bond [101].
Very few research studies concerning TPU-based NCs prepared by bulk
and melt processing have been reported in spite of the obvious advantages of
this approach (which offers advantages over solvent casting by way of
eliminating solvent and enhancing the filler dispersion). Since the melt
intercalation method was first demonstrated by Vaia and Giannelis [102], this
has become a primary method for preparing intercalated polymer/organically-
modified layered silicate NCs. Finnigan et al. [62] first reported preparation of
TPU/clay NCs by melt processing, and these displayed large increases in
stiffness, higher hysteresis, and permanent set while tensile strength and
elongation were not improved on addition of organoclay. They found that
organoclay was well dispersed in the TPU matrix via their melt processing
procedure where molecular diffusion and shear stress provided a good driving
force for intercalation between the polymer and organoclay. Moreover, Meng
et al. [79, 80] prepared polyether-based TPU/organically modified
montmorillonite NCs by the melt intercalation mathod. They showed that
optimized exfoliated structure can be obtained if there is adequate shear stress
and appropriate molecular diffusion to accommodate organoclay-TPU
interaction. Pattanayak and Jana [67] prepared TPU-based NCs with reactive
Cloisite 30B by in situ bulk polymerization. Two different methods were used
in this synthesis of TPU/clay NCs: in the first method the chain extender, 1,4-
butanediol, was added after the addition of clay; in the second method, the
chain extension reaction was carried out before the addition of clay and in the
second method chain-extended polymer molecules with residual -NCO groups
were involved in tethering reactions with clay during mixing. They found that
the addition of clay before the chain extension reaction in the first method led
to poor dispersion of clay particles. In the second method, due to the high
shear stress of mixing and good reaction, the clay particles were well
exfoliated to the scale of individual clay layers and the best improvement in
tensile properties was observed. They concluded that both the shear stress of
mixing and the reaction between polymer and clay are the main forces that
determine the exfoliation of clay, as shown by FTIR and reological data [67].
STRUCTURE-PROPERTY RELATIONSHIPS OF
TPU/CLAY NANOCOMPOSITES
Mechanical Properties
The strong industrial and academic interest in TPU/clay NCs has

prompted ongoing investigations into their structure-property correlations. The
relationships between the morphology of TPU nanocomposites and their
mechanical properties are extensively reported in the literature [62, 79-81].
The mechanical properties of TPU/clay NCs are affected by their morphology,
which is determined by the presence of clay particles. In TPU/clay NCs, the
clay particles can interact with the both the hard and soft segment rich phases,
affecting chain mobility and, therefore, mechanical behavior [19, 40,103].
Well-dispersed nanofiller provides greater contact surface and stronger
interfacial interaction between the nanofiller and the polymer matrix, thus
avoiding local stress concentrations and delocalization of stress [103]. These
are reflected in the TPU/clay NCs’ mechanical and physical properties.
Decreased tensile strength and elongation is observed for TPU/clay NCs with
intercalated structure, which could be explained by existence of voids in the
structure of NCs; such voids would result in weaker clay-polymer interaction.
Indeed, it was established that TPU/clay NCs with exfoliated structure had
better mechanical properties [103].
TenasiTech Pty Ltd. is a startup company formed in 2000 to
commercialize the production of TPU/clay NCs via reactive extrusion, and
outstanding tensile properties of TPU-based NCs has been obtained [6, 91]. It
should be noted that Prof. Darren Martin and his group have evaluated TPU-
based NCs from TenasiTech Pty Ltd. Osman et al. [37] studied
poly(dimethylsiloxane)/poly(hexamethylene oxide (PDMS/PHMO) mixed soft
segment based TPU-based NCs (prepared by solvent casting) containing two

different types of octadecyltrimethylammonium-modified nanosilicates: low
aspect ratio synthetic hectorite (Lucentite SWN (LuO) from Kobo Products)
and high aspect ratio synthetic fluoromica (Somasif™ ME100 (MEO) from
Kobo Products) with a hydrophobic surface modified cation. In this study [37],
mechanical properties of the prepared TPU-based NCs based on
PDMS/PHMO mixed soft segment and pure PDMS/PHMO (i.e., Elast-Eon
E5-325)-based TPU were compared with that of Nusil MED 4860, which is a
biomaterial currently used for insulation in cochlear implants (Figure 4). It was
shown that pure PDMS/PHMO-based TPU had higher mechanical properties
compared with Nusil MED-4860 (silicone rubber), with increases of 167% in
tensile strength, 48% in elongation at break, and 29% in tear strength. Better
dispersion and the best mechanical properties were achieved when 2 wt.%
synthetic hectorite was added, resulting in increases of 15% in tensile strength,
19% in elongation at break, 51% in toughness, and 32% increase in tear
strength, as compared to TPU-NCs containing synthetic fluoromica. The
improvements were due to greater polymer-nanofiller interactions. However,
they found that in the case of higher filler loadings (4 wt.%), the more
hydrophobic, low aspect ratio nanofillers induced morphological changes in
the TPU microdomain texture, causing adverse effects on the mechanical
properties of the TPU. These preliminary results show that PDMS/PHMO-
based TPU NCs are good candidate materials for biomedical applications [37].
Furthermore, Osman et al. [37] revealed two main peaks from dynamic
mechanical thermal analysis (DMTA) data (Figure 5a). The first peak
corresponds to a low temperature process (Tα1), which relates to segmental
motion in the PDMS phase and is observed in the range from -84 to -86°C for
TPU-based NCs. The second peak is assigned to the α,ω-PDMS endgroup
(Tα2), and Tα2 is observed in the range from 3 to 18°C for TPU-based NCs. The
Tα2 of the TPU increased with the addition of the organosilicates as compared
to pure TPU, and this shift was more pronounced in the LuO-based NCs as
compared to MEO-based NCs, probably due to higher phase mixing between
the hard and soft segments in case of LuO organosilicates. These results are in
agreement with results obtained by Runt et al. [30], who showed that the
position of the Tα2 can be influenced by the inclusion of single MDI and short
MDI-BDO sequences into the soft phase. From DMTA data, Osman et al. [37]
also observed increases of storage modulus with the addition of 2 or 4 wt.%
organosilicates (Figure 5b), indicating reinforcement has been provided to the
TPU matrix, which is in agremeement with the mechanical test results. LuO
was most effective in increasing the modulus of Elast-Eon E5-325 at room
temperature, especially for NCs containing 4 wt.% of LuO. They reported, on

the basis of DMTA, that the addition of 2 wt.% LuO in the TPU resulted in a
“partial compatibilizing effect” while adding the organosilicate at 4 wt.%
induced almost full compatibilization between the hard and soft segments [37].
Runt et al. [89, 104] reported the preparation and properties of solvent-cast
NCs derived from the biomedical TPU Biospan™ (chemically similar to
poly(ether-urethane)) and Cloisite® 15A. The prepared NCs showed an
improvement in tensile strength, modulus, and elongation at break when the
nanoclay content increased from 0 to 20 wt.%. Pattanayak and Jana [67]
observed the best improvement in mechanical properties when clay particles (5
wt.%) were fully exfoliated, measuring a 110% increase in tensile modulus,
170% increase in tensile strength, 110% increase in tear strength, 120%
increase in fracture toughness, and 40% increase in abrasion resistance over
pure TPU.
Figure 4. Stress-strain curves of pure PDMS/PHMO-based TPU (i.e., Elast-Eon E5-

325), Nusil MED 4860- and PDMS/PHMO-based NCs containing 2 or 4 wt.%
synthetic hectorite (Lucentite SWN; LuO) or synthetic fluoromica (Somasif™ ME100;
MEO) [37].
Figure 5. DMTA curves as a function of temperature: (a) damping factor (tan δ); (b)
storage modulus (E′) for pure PDMS/PHMO-based TPU (i.e., Elast-Eon E5-325) and
PDMS/PHMO-based NCs containing 2 or 4 wt.% synthetic hectorite (Lucentite SWN;
LuO) or synthetic fluoromica (Somasif™ ME100; MEO). Curves (x) and (y) show that
4 wt.% of LuO has a much higher soft microphase transition temperature (Tα2) and
storage modulus as compared to the pure TPU, which indicates pronounced phase
mixing [37].
Thermal Properties
Generally, including clay in the polymer matrix improves thermal

stability. The observed improvement in thermal stability of polymer/clay NCs
is generally ascribed to the shielding effect of clay layers, which act as a
barrier to the gasses and volatile degradation products [105]. Moreover, the
clay platelets act as heat barriers and obstacles for polymer chain motion,
leading to further stabilization [105]. Rueda et al. [40] studied the structure-
property relationship of TPU/Cloisite 30B NCs based on poly(caprolactone)-b-
polytetrahydrofuran-b-poly(caprolactone) as the soft segment and 1,6-
hexamethylene diisocyanate-1,4-butanediol as the hard segment, prepared by
in situ intercalative polymerization. They observed that the thermal stability of
TPU/clay NCs increased when clay content increased from 1 to 4 wt.%, due in
part to the intercalation/delamination of the clay platelets whose surface was
grafted by polyurethane chains leading to strong interactions between the
organoclay and the TPU matrix. The excellent thermal stability of TPU/clay
NCs might be explained by the barrier effect generated by highly anisotropic
clay platelets, which delayed the degradation in NCs. However, these authors
observed no relationship between thermal stability and morphology.
Pavlićević et al. [106] found that the thermal stability of TPU NCs containing
polycarbonate macrodiol with molecular weight ca. 1000 g/mol (type 5651 or
4671) can be improved by addition of organically modified nanoparticles with
layered structures (bentonite or montmorillonite). Poręba et al. [107]
investigated the thermal stability of TPU/modified organoclay NCs based on
polycarbonate macrodiol with higher molecular weights of 2000 g/mol (types
5652 and 4672), prepared by one-step polymerization, and observed
significant enhancement of the thermal stability in NCs, and observed
significant enhancement of the thermal stability in NCs with higher hard
segment content and well dispersed nano-scale layered bentonite particles,
where hard domains acted like defects and interacted strongly with the well-
dispersed organoclay layered nanoparticles. Salahuddin et al. [108] observed
that temperatures at 5%, 10% and 50% of weight loss obtained during
thermogravimetric analysis (TGA) of TPU/clay NCs, were 40 to 125°C higher
than those of the pure TPU. They concluded that thermal stability of NCs was
much higher than that of pure TPU and increased with the amount of organo-
montmorillonite in TPU. The high thermal stability was attributed to the
interaction between the clay particles and the polymer matrix [108]. On the
other hand, there are reports of worsened thermal stability of polymer-based
NCs containing clay, connected with the degradation of the organic modifiers
[109-111]. It could be assumed that the degradation of the organic ions had an
accelerating effect on the degradation of the polymer. According to the work
done by Tabuani et al. [112], adding Cloisite 30B into the polyether-based
TPU provokes a reduction of about 30°C of the degradation onset temperature,
mainy due to early decomposition of the Cloisite 30B organic modifier and
clay-catalyzed polymer degradation, as also reported by Bellucci et al. [113].
Figure 6. DSC thermograms obtained during: (a) heating scan and (b) cooling scan for
the pure PDMS/PHMO-based TPU (i.e., Elast-Eon E5-325) and PDMS/PHMO-based
NCs containing containing 2 or 4 wt.% synthetic hectorite (Lucentite SWN; LuO) or
synthetic fluoromica (Somasif™ ME100; MEO) [37].
Using differential scanning calorimetry (DSC), Cruz and Viana [103]

found that incorporating nanoclay into the TPU resulted in an increased glass
transition temperature of the hard segment (TgHS) of around 8–10°C and the
formation of thicker hard segment structures with a high degree of
crystallinity. These supramolecular structures, acting as restraining points for
the soft segment domains, reduced the macromolecular mobility, increasing
the initial modulus. Rueda et al. [40] have shown (based on DSC) the presence
of clay particles in both soft and hard segments of the TPU matrix. The
increase in melting temperature of the soft segment of TPU/clay NCs as
compared with pure TPU was ascribed to the nucleating effect of clay [40].
The authors observed decreases in enthalpies of crystallization and melting of
the hard segment, which were explained by the higher viscosity of the samples
when polyurethane chains anchored to the clay, consequently acting as a
deterrent to crystallization of hard segment. Osman et al. [114] observed that
PDMS/PHMO-based TPU NCs containing 2 or 4 wt.% nanofillers, i.e.,
organofluoromica with octadecyltrimethyl ammonium (ODTMA) single
surfactant modification (MEO) and ODTMA/CC-dual surfactant modification
(MEO-C) prepared by melt compounding, ehxibited two endotherms (T1 and
T2) in the DSC thermograms. T1 endotherm appeared in the 53 to 58°C range
and was associated with the disordering of clusters of single (‘lone’) MDI
residues. T2 endotherm was detected in the 99 to 106°C range and was related
to the Tg of the hard domains. The MEO NCs exhibited a slightly higher
temperature endotherm which was associated with the disordering of clusters
of single (‘lone’) MDI residues (T1), as compared to pure PDMS/PHMO-
based TPU (Elast-Eon E5-325) and MEO-C-based NCs. Due to the very
hydrophobic inter-gallery space in this system, single MDI residues and
shorter hard segments are preferentially intercalated and these intercalated
shorter hard segments could contribute to the increase in T1 value [114].
Furthermore, in another study of Osman at el. [37], PDMS/PHMO-based NCs
containing LuO and MEO (2 or 4 wt.%) produced two to four endotherms (T1
to T4) in the DSC thermograms (Figure 6). T1 endotherm appeared at ~ 50°C
and corresponded to the ordering of single (“lone”) MDIs. T2 endotherm was
observed between 99 and 115°C, and was related to the Tg of the hard
domains. PDMS/PHMO-based NCs containing LuO with 2 or 4 wt.% as
compared with NCs containing MEO nanofiller, showed endotherms T3 (for
2LuO) as well as T3 and T4 (for 4LuO) at temperatures of 197 and
194−207°C, respectively. These high temperature melting endotherms for the
NCs were perhaps due to sluggish, but preferential, organization of the longer
hard segments, which were potentially excluded (insoluble) from the soft
segments and interphases due to their thermodynamic incompatibility. Based

on the cooling scan, and compared to NCs containing MEO, PDMS/PHMO-
based TPU NCs containing LuO showed crystallization exotherms at higher
temperatures [37]. However, they also showed lowed enthalpy of
crystallization with a broader and weaker peak, which indicated disruption to
the reorganization of the complete population of hard segments during cooling
from melt. This was perhaps brought about by the hydrophobic LuO
simultaneously acting as both a nucleating and a compatibilizing agent [37].
Barrier Properties
Generally, adding organoclay into TPU matrices improves their barrier

properties. Chang and An [97] found that the gas permeability reduced linearly
with an increasing amount of organoclay in the TPU matrix. Runt et al. [89]
studied preparation of polyether-based TPU containing octadecylammonium-
modified montmorillonite by solvent casting and obtained low permeability
NCs. They found the modulus of NCs was improved, while the water vapor
permeabilities were reduced by up to five times at the highest organoclay
content (20 wt.%). Rafal et al. [19] showed that the presence of 1 wt.% of
bentonite did not practically affect gas transport properties of polycarbonate-
based TPU NCs. In this case, very high values of permeability, solubility and
diffusion coefficient for CO2 were caused by favorable interactions of CO2
with carbonate units in the TPU [19]. Herrera-Alonso et al. [115] observed
significant barrier properties for TPU/clay NCs if clay samples were pre-
processed with sonication instead of plain stirring, suggesting that dispersion
of the clay particles is a critical factor. Moreover, based on WAXS results,
these authors suggested that the TPU achieved better penetration within the
gallery space of the montmorillonite nanoclays when these species were
modified with alkylammonium groups with only one tallow group i.e., for
Cloisite 10A and Cloisite 30B organoclays as compared to Cloisite 20A. This
was due to hydrogen bonding between the two hydroxyethyl groups present in
the organic modifier and the polar section of the polyurethane.
Flame Retardant Properties
The introduction of organoclay into the TPU matrix improves the flame
retardancy as compared to the pure TPU, due to the the formation of a
carbonaceous-silicate char that controls the flammability [116]. It was known

that the flame-retarding mechanism of layered silicate NCs is based on
formation of the char; the char insulates the polymer from heat and acts as
mass and energy transport barrier, reducing the escape of volatile gases from
combustion of the polymer. Visually, a smooth char layer is formed at the
surface of the nanocomposite, finally leaving a black char residue. Tabuani et
al. [112] observed that polyether-based TPU/clay NCs with intercalated
structure, containing both melamine cyanurate and 7 wt.% of Cloisite 30B
clay, showed a significant improvement of fire properties (as demonstrated by
the results of cone calorimeter test and LOI (limiting oxygen index) tests), and
enhanced mechanical properties. Kannan et al. [117] observed that the flame
retardancy of polypropylene (PP)/TPU blands NCs was enhanced when 5
wt.% of Cloisite 10A clay was added. They observed that addition of nanoclay
increased the LOI value and the product took more time to ignite when
compared with pure PP/TPU. Also, nanoclay-added samples had a low
burning rate compared with pure PP/TPU blend, resulting in good flame
retardant properties. The introduction of the nanoclay as the flame retardant to
the PP/TPU improved the mechanical properties. The tensile stress required to
make 200% strain in TPU/PP/MA-g-PP (maleic anhydride-grafted-PP) NC
containing 5 wt.% of nanoclay increased by 120%, and the flexural modulus
increased by 80-90%.
Biocompatible Properties
A few studies have revealed that the release of quaternary alkyl

ammonium from organoclays could induce toxicity of TPU. Rueda et al. [40]
reported that NCs containing 1 and 2 wt.% of Cloisite 30B induced significant
cell (L-929 fibroblast) inhibition and, in consequence, a reduction in the
degree of cell adhesion to the material surface. Styan et al. [118] found that
methyl tallow bis-2-hydroxy ethyl ammonium, which was released from
Cloisite® 30B, inhibited cell growth. Tee et al. [119] found that cytotoxicity
increased as the size of fluoromica decreased. The relationship between size
and toxicity is mainly related to the increased surface area and reactivity
associated with the smaller diameter particles. However, several studies have
shown opposite results. Mishra et al. [120] showed good cell growth and
attachment on the surface of TPU/Cloisite 30B NCs. They found that NCs
were not genotoxic by observing the localization of the enzyme HIPK2 (an
apoptosis activator which means that its presence in a nucleus indicates
genotoxicity stress). Styan et al. [121] observed that cell death was reduced
when montmorillonite modified with amino undecanoic acid was used and cell
growth was not significantly inhibited because this organomodifier was
negatively charged at pH 7.4. The obtained results showed that suitable
material selection is important when incorporating organoclays in NCs. Also,
another important factor is the method of processing TPU. Edwards et al.
[122] showed that some commercial organoclays contained excess surfactants
which could be released and thus produce toxicity.
According to the literature, the addition of a very low percentage (≤10
wt.%) of layered silicates to a TPU matrix can also lead to a significant
improvement in many other properties, such as water vapor permeability
[123], water absorption [124], ionic conductivity [125], electrical properties
[126], heat distortion temperature [127], biodegradability [128], shape
memory behavior [129], and drug delivery properties [130]. These capabilities
make this material a good candidate for different potential applications, such
as in the automotive, military, aerospace, electronics, food packaging,
biotechnology, and biomedical fields.
TPU/CARBON NANOTUBE COMPOSITES

The unique mechanical, thermal, electrical and optical properties of
carbon nanofillers make them good reinforcing agents for high performance
polymer NCs. Carbon nanofillers comprise carbon nanotubes (CNTs), carbon
nanofibers (CNFs) and graphene platelets, and are distinguished by strong
hydrophobic character and good electrical conductivity [131]. Carbon
nanotube-reinforced TPUs have been subject to considerable research, both
academic and applied. CNTs are stretched hollow cylindrical structures [131].
There are two types of CNTs: single-wall carbon nanotubes (SWCNTs) are
graphene sheet rolled to cylinders, and multi-wall carbon nanotubes
(MWCNTs) are such cylinders (walls) inserted into one another. Because of
their very small diameter that forces electrons to move along their length,
CNTs are considered as one-dimensional nanomaterials [131, 132]. They are
specified by their two types, geometry and chirality which is designated by the
angle of orientation of the graphene lattice relative to the tube axis. The
orientation affects electric properties of CNTs, which can behave as either a
metallic conductor or a semiconductor.
A combination of advantageous properties such as large aspect ratios
(300-1000), large surface area (experimental: 150-1587 m2/g, theoretical: 3000
m2/g for SWCNTs), high Young’s modulus of over 1 TPa, high tensile
strength of over 200 GPa, high flexibility, low mass density, thermal
conductivity higher than 3000 W/mK, high corrosion resistance, high
electrical conductivity ranging close to metallic conduction is reason for their
extensive investigation in the literaure. CNTs are prepared using various
methods: laser ablation of graphite, chemical vapor deposition and growth in
the arc discharge from a carbon electrode. Polymer/CNT NCs can be prepared
using the following procedures: in situ polymerization [133-136], solution
blending [64, 133, 137-139] and melt processing [61, 140, 141]. Due to the
strong Van der Waals force, CNTs tend to bundle with each other and thus do
not disperse well in the polymer matrix, leading to phase separation and poor
mechanical properties [142]. To achieve better dispersion of CNTs (especially
in the case of MWCNTs) in polymer matrix and, therefore, to improve the
load transfer across the CNTs-polymer matrix interface, physical (i.e., non-
covalent approach which mainly involves surfactants, biomacromolecules or
polymer wrapping) or chemical surface modification (such as oxidation,
amination, functional groups and other covalent functionalization of CNTs) is
applied [132, 138, 143-147].
TPU/CARBON NANOTUBE COMPOSITES
The literature reveals that very few reasearch works have been published
concerning SWCNT reinforced TPU NCs as compared to MWCNT (and these
are mostly functionalized MWCNT) reinforced TPU NCs. Sen et al. [132]
were first to prepare TPU/SWCNT NCs by the solvent proccesing method and
they observed significantly improved mechanical properties over the pure
TPU. They found that functionalization of SWCNT led to its increased
dispersion in TPU matrix and enhanced interaction between TPU and
SWCNT. Koerner et al. [148] were the first to study TPU/MWCNT NCs and
the main goal of this study was improved the stress-recovery characteristics of
Morthane (a commercially available, shape memory TPU) by adding 1-5
vol.% MWCNT. They found that NCs could store and subsequently release up
to 50% more recovery stress than the unmodified TPU and the products
possesed the ability to recover from stress, above that of the pure TPU, when
exposed to increased temperatures, infrarred radiation and electrical current.

Chen et al. [149] prepared acid-treated MWCNT reinforced TPU NCs by a
melt processing method. Significant enhancement in Young’s modulus (an
eight-fold increase) and tensile strength (2.4 fold increase) was achieved by
incorporating MWCNT into TPU (up to 9.3 wt.%) without sacrificing high
elongation at break compared to pure TPU. These results indicated that strong
interfacial interactions and good MWCNT dispersion within TPU matrix are
responsible for the significant enhancement of mechanical properties of the
composite fibers. Barick and Tripathy [150] prepared TPU NCs with various
MWCNT content (0.5, 2.5 and 5 wt.%) by melt blending and they observed
that TPU/functionalized MWCNT NCs with 0.5 and 2.5 wt.% improved the
tensile properties, tear strength and hardness of the TPU matrix without
sacrificing the elongation at break. Those authors ascribed the changes to
homogeneous dispersion of functionalized MWCNT in TPU matrix and the
presence of strong interfacial adhesion between functionalized MWCNT and
the TPU matrix. They found that the glass transition temperature and storage
modulus determined by DMTA of the NCs were increased with increasing
MWCNT content, which indicated that CNTs were compatible with the TPU
matrix [150]. Raja et al. [151] showed that polymer blend NCs based on TPU,
polylactide (PLA), with functionalized MWCNT prepared by the melt mixing
process, had higher tensile modulus values and improved thermal stability. Tg
of the TPU/PLA blend was slightly decreased by the addition of functionalized
MWCNT, which was attributed to the better dispersion of functionalized
MWCNTs and to improved interaction between functionalized MWCNTs and
polymer matrix [151]. Cruz et al. [103] studied the influence of the type of
filler, such as high-structured carbon black (HSCB), CNF, nanosilica (NS) and
organo-modified nanoclays (OMNC) on the mechanical properties of
composites. All polyester-based TPU NCs show similar behavior, but with
much higher deformation capabilities and stress levels (in some cases by more
than 200%) compared to the pure TPU. Compared to the pure TPU, and with
the exception of TPU/HSCB and TPU/NS composites, the initial elastic
modulus show great improvement. They observed that TPU/CNF composites
had an enhancement of elastic modulus of more than 100% due to good
nanofiller (i.e., CNF) adhesion to TPU matrix and good dispersion of the CNF
into the TPU matrix.
Thermal Properties
The introduction of 2 wt.% of MWCNT amide derivate into

poly(tetramethylene oxide)-based TPU matrix increased the temperature at
maximal degradation rate from 408°C for TPU to 419°C for NCs, as shown by
Xiong et al. [135]. They found that the thermal stability of the composite was
improved by the addition of the functionalized MWCNTs, which can be
attributed to their interactions with the polymer matrix. Kuan et al. [133]
observed that adding functionalized MWCNT (containing COOH and NH2
groups) into waterborne TPU improved the thermal degradation properties by
26°C (from 315 to 341°C) when the CNT content was 2.5 parts per hundred
parts of resin in a covalent bonding system. Barick and Tripathy [61] indicated
that the thermal stability of the NCs was substantially enhanced by the
incorporation of COOH-MWCNT, and attributed that to the excellent
interfacial interactions present between the TPU matrix and MWCNT. The
formation of chemical bonds between the TPU matrix and COOH-MWCNT
significantly reduced the TPU-CNT thermal boundary resistance, which is
responsible for the smooth transfer of heat from the TPU matrix to the CNTs.
This facilitates uniform heat distribution throughout the NCs and as a result,
the thermal decomposition temperature increases. Therefore, the homogeneous
dispersion of the CNTs, large heat transfer due to higher thermal conductivity
of TPU/CNT nanocomposites and formation and stabilization of the CNT
bonded macroradicals were the main factors which enhanced the thermal
stability of TPU matrix [61].
The effect of methylene-bis-ortho-chloroanilline (MOCA)-grafted
MWCNT on the Tg values of TPU/functionalized MWCNTs was investigated
by Xiong et al. [135], using DSC. They observed that the increase of the
functionalized MWCNT content led to a significantly increased Tg value for
the TPU/functionalized MWCNT NCs. The improvements in Tg can be
attributed to the existence of crosslinking CNTs inside the TPU matrix.
Additionally, the glass transition region was widened with increasing
MWCNT content [135]. This shows that MWCNT inserted in hard segments
of TPU chains play the role of the cross-linker in the matrix. Barick and and
Tripathy [61] showed that the Tg, melting temperature and crystallinity slightly
increased with inclusion of MWCNT, which suggested that the functionalized
MWCNT substantially influenced both the crystalline and amorphous
structures of the TPU matrix. Lima et al. [152] carried out DSC analysis of
TPU/MWCNT functionalized with amino and carboxyl groups by spray
coating and buckypaper based techniques, and determined that Tg related to the
hard segment domain of TPU/MWCNT-NH2 was higher, while that for

TPU/MWCNT-COOH was lower than for the pure TPU.
Bourbigot et al. [153] studied the flame retardancy of TPU CNT

composites and potential synergistic effects when CNT was combined with
conventional flame retardant agents (phosphate and phosphinate). They found
that TPU/CNT NCs showed significant reduction of peak heat release rate
(PHRR), but the exfoliated or intercalated structure did not play any
significant role in the flame retardant properties. Incorporating MWCNT in
TPU decreased the PHRR by 50% compared to the pure TPU, but LOI and
UL-94 were not enhanced [153]. They showed that TPU/MWCNT NCs gave
the best flame retardancy results when combined with conventional flame
retardants, which led to synergetic effects.
Electrical Properties
It is important to possess a three-dimensional network of conductive

nanoparticles (i.e., percolation) in order to achieve conductive paths through
the polymer matrix. The filler loading that produces a sharp drop of the NCs’
electrical resistance is considered as the percolation threshold. Furthermore,
the percolation threshold is affected by many factors: the type of polymer, the
route of composite preparation, the size, the particle shape, aspect ratio and
modification of the filler. Lower percolation thresholds are obtained for
nanoparticles with higher aspect ratios. Extensive percolation studies were
carried out in polymer/CNT composites. It was established that percolation
thresholds lower than 1 wt.% or even lower than 0.1 wt.% are usual [138, 154,
155]. Incorporating CNT into TPU significantly increased electrical
conductivity of the TPU/CNT NCs in comparison to the pure TPU. According
to Lima et al. [152], the electrical conductivity was significantly improved
from 10-12 to 10-5 S/cm for the spray-coating composites and to 1.1 and 7.3
S/cm for the buckypaper composites with MWCNT-COOH and MWCNT-
NH2, respectively, as compared to pure TPU. Barick and and Tripathy [150]
observed that dielectric constant, loss tangent and electrical conductivity of the
TPU/MWCNT NCs were much higher than that of the pure TPU matrix. The
conductivity of the TPU/MWCNT NCs increased with increasing MWCNT in
the low frequency range. They found that interconnected physical networks in
the bulk of the TPU matrix were formed due to the presence of MWCNT
clusters, and were responsible for the electrical conductivity of the
TPU/MWCNT NCs [150]. Zhang et al. [141] studied the effect of conductive
network formation in a polymer melt on the conductivity of TPU/MWCNT
composite. A very low percolation threshold of 0.13 wt.% was achieved in
hot-pressed composite film samples, whereas a much higher MWCNT
concentration (3-4 wt.%) was needed to form a conductive network in
extruded composite strands. These results were implied in terms of the
dynamic percolation behaviour of the MWCNT network in the polymer melt.
According to Koerner et al. [138], the addition of small amounts (0.5-10
vol.%) of MWCNTs into commercial TPU Morthane produced NCs with high
electrical conductivity (1–10 S/cm), low electrical percolation (0.005) and
enhanced mechanical properties.
GRAPHENE NANOCOMPOSITES
Graphene is a monolayer of sp2 hybridized carbon atoms arranged in a
two-dimensional lattice, where graphene layers in the graphite lattice consist
of regular hexagons with 0.142 nm sides, and carbon atoms are located in
vertices of the hexagons [131]. Moreover, graphene, a cheap, multifunctional
material, has nanoscale dimensions and unique physical, chemical and
electrical properties. Novoselov et al. [156] reported pioneering studies based
on preparation graphene in free state using a peel-off method. This involved
peeling graphite off a piece of Scotch (adhesive) tape and transferring it to a
stable SiO2 substrate. Recently, graphene has attracted much attention as a
reinforcing agent in the area of polymer NCs due to excellent performances
such as high surface area (2630 m2/g), excellent mechanical properties
(approximately 1000 GPa), excellent electrical conductivity (approximately
6000 S/cm), thermal conductivity (approximately 5000 W/mK) and gas
impermeability [157]. Several methods have been used to prepare graphene,
such as chemical vapor deposition [158], epitaxial growth [159] and raw
graphite exfoliation [160], plus physical and chemical reduction of graphite
oxide (GO) [161, 162].
Pristine graphene can not be dispersed effectively within a polymer
matrix. To obtain homogeneous dispersion, efficient interfacial interactions
and prepare NCs with improved properties, expanded graphite (forcing the
crystal lattice planes apart due to the insertion of some foreign particles),
intercalated (insertion of metal between lattice planes) and surface modified
graphene have been used [131]. Covalent functionalization is an effective
route for producing graphene-based polymer NCs, because the covalent
functionalization dramatically alters the Van der Waals interactions among the
nanofiller aggregates, making them easily dispersed within the polymer matrix
[163-165]. For example, graphene oxide (GO), the precursor of graphene
nanosheets, has abundant functional groups on the surface (hydroxyls,
epoxides, and carboxyls), which can provide the reactive site for covalent
functionalization [166]. Three different approaches can used for preparing
graphene/TPU NCs: in situ polymerization [167-170], solution blending [165,
169-172] and melt intercalation [169, 170]. It has been established that very
small graphene loading into polymer is sufficient to achieve network
formation and percolation in a polymer matrix and to improve stiffness, fire
retardancy, electrical conductivity, abrasion resistance, barrier and
electrooptical properties [169, 170].
TPU/GRAPHENE NANOCOMPOSITES
Fahim et al. [173] prepared TPU/graphene NCs with differing amounts of

graphene powder using two processing methods: mechanical mixing followed
by compression moulding and wet mixing. Dry mixing produced improved
mechanical, viscosity and electrical properties compared to wet mixing, due to
better dispersion of graphene particles in the TPU matrix. In fact, those authors
found that the processing method plays an important role in obtaining the
desired TPU/graphene NCs. TPU/graphene NCs obtained by dry mixing had
the highest compressive strength at 0.1 wt.% of graphene compared with 0.5
and 1wt.% of graphene. Xiang et al. [174] prepared TPU NCs containing
hexadecyl-functionalized low-defect graphene nanoribbons (HD-GNRs) by
solution casting and observed that incorporating HD-GNRs improved the
mechanical properties of NCs. They attributed this to a synergistic effect of
both the incorporated HD-GNRs and the phase separation; higher HD-GNR
concentration led to higher storage modulus. Kim et al. [172] incorporated
graphene oxide (GO), thermally reduced graphene oxide (TRG) and

isocyanate-treated grephene oxide (iGO) into polyester-based TPU elastomer
containing 25 wt.% hard segment via melt intercalation, solvent mixing and in
situ polymerization. Because TRG and GO have surface hydroxyls that can
react with isocyanate groups, they could also serve as pseudo chain extenders
for TPU. TRG dispersed via solvent blending appeared to more effectively
increase the modulus and had good elongation of TPU than TRG dispersed via
melt processing. With only 3 wt.% of iGO or TRG, a 3- to 10-fold
enhancement in the tensile modulus and 80-90% reduction in nitrogen
permeability were attained [172]. Liang et al. [175] prepared three series of
TPU NCs with isocyanate modified graphene, sulfonated graphene and
reduced graphene as nanofillers. They determined significant improvements in
mechanical properties of TPU/sulfonated graphene NCs with 1 wt.% of
sulfonated graphene due to fine dispersion of sulfonated graphene in the TPU
matrix. The tensile strength of TPU/sulfonated graphene (1 wt.%) NCs was
increased by 75% at a strain of 100% and the Young’s modulus was increased
by 120%.
Thermal Properties
Studies on graphene TPU-based NCs have suggested improved thermal

stability with increased levels of exfoliation and interfacial adhesion [105].
Increases in the onset of degradation (in inert atmosphere) of 20-30°C and
higher have been reported with graphene oxide-derived fillers [176]. On the
other hand, the presence of graphene oxide in the TPU matrix was found to
accelerate the decomposition kinetics. Strankowski et al. [177] observed that
the addition of thermally reduced graphene oxide (TRG) into TPU did not
raise the temperature at which degradation starts. Incorporating 0.5-3.0 wt.%
of TRG into TPU matrix led to increased char residues compared to pure TPU
elastomer. Liang et al. [175] showed that the rate of thermal degradation of
TPU/isocyanate modified graphene NCs were significantly higher than those
of the sulfonated graphene and reduced graphene based TPU NCs. Those
authors observed that there are fewer functional groups attached to sulfonated
graphene sheets than to the isocyanate modified graphene.
According to Xiang et al. [174], the Tg of TPU determined by DMTA was
shifted to lower temperatures when HD-GNRs were added, because phase
separation of TPU causes fewer hard segments to be alongside soft segments,
so that the motion of the soft segment becomes easier. Furthermore, the
hindering effect of hard segments plays a more important role than that of HD-
GNRs in determining Tg shifts [178], and this result has also been observed in
TPU/carbon nanotube composites, as reported by Xia et al. [178].
Additionally, Lee et al. [167] showed that the prsence of FGS in waterborne
polyurethane can increase the melting temperature and enthalphy of melting of
the soft segment of waterborne polyurethane-based NCs, as determined by
DSC analysis. The cystallinity of the soft PCL segment was enhanced, while
the crystallinity of the hard segment decreased with increasing FGS content in
NCs. Strankowski et al. [177] observed that the glass transition temperature
increased with increasing content of thermally reduced graphene oxide in TPU
matrix. These results can be explained by the reduced chain mobility of soft
segments due to good dispersion of TRG nanofiller in polyester-based TPU
matrix.
Lee et al. [167] prepared waterborne polyurethane (WTPU) with

functionalized graphene sheets by an in situ method and observed that the
electrical conductivity of the NCs was increased 105 fold in comparison with
pure WTPU. They concluded this was due to the good, homogeneous
dispersion of functionalized graphene sheet particles in the WTPU matrix. The
percolation threshold was obtained at 2 wt.% of functionalized graphene sheet
loading due to agglomeration of FGS during reduction [167]. They found that
the electrical conductivity of WTPU/FGS NCs can be enhanced as effectively
as that of CNTs based TPU NCs. Kim et al. [172] observed that
TPU/thermally reduced graphene oxide (TRG) sheets prepared by solvent
blending became electrically conductive at very low TRG loading (0.5 wt.%).
Li et al. [171] found that the electrical conductivity of TPU/reduced graphene
oxide (RGO)/polyvinyl pyrrolidone NCs prepared by solution mixing were
greatly improved with the addition of RGO, obtaining an electrical percolation
threshold of 0.35 wt.%.
Barrier Properties
Recently, more researchers report the application of graphene and

graphene oxide for gas adsorption, gas storage and gas separation. Kim et al.
[172] found that that in situ polymerized thermally reduced graphene oxide
was not as effective as solution mixed thermally reduced graphene oxide in

reducing gas permeability of TPU. Furthermore, they observed that up to an 80
fold decrease in nitrogen permeation of TPU was observed with only 3 wt.%
phenyl isocyanate-treated graphite oxide. Xiang et al. [174] showed that the
TPU NCs with hexadecyl-functionalized low-defect graphene nanoribbons
(HD-GNRs) had high gas barrier efficiencies at low loadings, a result
attributed to the structure of the low defect HD-GNRs and their uniform
dispersion. They demontrated that nitrogen gas effective diffusivity of TPU
was decreased by 3 orders of magnitude with only 0.5 wt.% of HD-GNRs.
Chen et al. [179] observed that adding iron-graphene (IG) and aluminum
hypophosphite (AHP) into TPU dramatically increased the flame retardant
properties of TPU. They found that with 0.25 wt.% of IG and 9.75 wt.% of
AHP, the product achieved a UL 94 V-0 rating, the LOI value increased by
31.5 vol.%, pHRR value decreased by 90% and peak smoke production rate
(pSPR) value decreased by 72%. According to Xu et al. [180], the presence of
molybdenum trioxide-loaded graphene (MoO3-GNS) hybrids and cuprous
oxide-loaded graphene (Cu2O-GNS) hybrids in TPU resulted in excellent
flame retardant and smoke suppression properties of TPU composites, as
evidenced by the obvious reduction in their heat release rate, total heat release,
and smoke density. They found that the enhancement in the flame retardant
and smoke suppression properties were mainly attributed to the synergistic
effect between the physical barrier effect of graphene sheets and the catalytic
charring effect of MoO3 or Cu2O. Gavgain et al. [181] reported that the LOI
value of intumescent flame retardant polyurethane (IFRPU) composites
containing 2 wt.% GNS increased from 29.0% to 34.0%. Moreover, the
addition of GNS into IFRPU composites produced excellent antidripping
properties as well as UL-94 V0 rating [181]. Hu et al. [182] suggested that
GNS could significantly enhance the char residue and reduce the heat release
rate and total heat release of waterborne polyurethane.
CARBON NANOFIBER COMPOSITES
Carbon nanofibers (CNFs) are cylindrical nanostructures with graphene
layers arranged as stacked cones or as a paper scroll. They have lengths in the
order of micrometers and the diameter varies from tens of nanometers to 200
nm. CNFs are usually prepared by the chemical vapor deposition method.
CNFs possess high aspect ratio, large surface area, good mechanical
properties, good thermal and electrical conductivity, high temperature
resistance and low cost, which make them good candidate materials for
biosensors, Li-ion batteries, energy storage devices, electrochemical
substances, and nanofillers for polymer NCs [183].
CNFs are of particular interest as polymer (including TPU) preparation
modifiers, not only because of their potential for high specific strength but also
for their usefulness in designing functional materials with unique mechanical
properties, electrical properties, dielectric behaviors, thermal properties, shape
memory behaviors, thermal conductivities, and electromagnetic interference
(EMI) shielding effectiveness. TPUs strength and abrasion resistance should
be further improved to deal with more demanding conditions [184]. The
abrasion resistance and other mechanical properties of TPU can be improved
by reinforcing them with short carbon fiber. Recently, TPU materials based on
vapor-grown carbon nanofibers have been the subject of intense investigation.
The reinforcement efficiency of CNFs depends on the content and aspect ratio
of the CNF, on the effect of the intrinsic mechanical properties, and on the
uniform distribution of CNFs and polymer matrix-CNF interfacial adhesion
[185].
TPU/CARBON NANOFIBER COMPOSITES
Mechanical properties of CNF-based polymer composites depend on the

properties of their individual constituents and the interaction degree between
the fibers and polymer matrix. A good fiber-polymer matrix interface provides
an effective load transfer from the polymer to the CNF, which is necessary to
reduce the stress concentrations and enhance the mechanical properties [186].
In addition, the poor interfacial adhesion between CNFs and TPU, caused
intrinsically by the chemically inert surface of CNFs, limits the mechanical
and thermal performance of TPU composites. Also, the mechanical properties
of TPU/CNF composites depend on the fiber length. Zhang et al. [184] used
methylenediphenyl diisocyanate to modified the surface of CNF (content of
MDI-CNFs ranged from 5 to 30 wt%) and found that the mechanical
properties and wettability of TPU composites were significantly improved
over material without MDI. The tensile strength, tensile modulus, elongation
at break, and storage modulus of TPU/MDI-CF were superior to those for
TPU/CNF composites. Namely, the tensile strength of TPU/MDI-CF was
increased by 99%, which was greater than TPU/CNF (53%), and the friction
loss of TPU/MDI-CNF was decreased by 49%. The tensile strength of
TPU/CNF composites improved with increasing fiber length at the same fiber
content, which was one of the reasons why TPU/MDI-CNF composites had
higher tensile strength [184]. The abrasion values of TPU/MDI-CF composites
were lower compared to those of TPU/CNF composites, as well as being lower
than that of TPU when CNF content was more than 5 wt.%. The authors
concluded that the MDI-CNFs had good interfacial adhesion with TPU matrix,
which could be used to manufacture high-performance TPU composites [184].
Jiang et al. [187] developed multiscale graphene oxide/carbon nanofiber
(GO/CNF) reinforcements in which CNFs served as structural reinforcements
and GO distributed uniformly to reinforce the TPU matrix. The multiscale
reinforcement improved the tensile strength of TPU by 46%. Compared with
unmodified TPU/CNF composite, the individual processes of electrophoretic
deposition treatment and mechanical mixing of GO resulted in an
improvement of 17% and 14% of the tensile strength, respectively. The
enhanced reinforcing performance of GO-deposited CNFs was attributed to
the improved interfacial adhesion between TPU matrix and CNF. Barick et al.
[90] found that the presence of vapor-grown CNFs in the TPU matrix (CNF-
based TPU was prepared by melt blending) increased the ultimate tensile
strength and modulus of pure TPU. Both maximum tensile strength and
modulus occurred at 7 wt.% of CNFs, which indicated good fiber dispersion
and interaction between TPU matrix and CNF. They found that at 10 and 15
wt.% of CNF, the formation of aggregates was higher due to a rise of highly
entangled and interconnected network structures, which reduced the interfacial
interaction between TPU and CNF [90]. Therefore, the tensile strength of the
NCs containing higher CNF content was significantly reduced due to the
presence of voids or other defects as well as the formation of bundles of CNFs,
which could act as stress concentration regions leading to weakness in

interactions betweeen polymer matrix and fiber.
THERMAL PROPERTIES
Špirková et al. [188] observed enhanced thermal stability of
polybutadiene-based TPU/CNF composites (prepared by solvent blending) at
very low CNF loading (0.05 wt.%). They observed that TPU NCs with 2.25
wt.% of CNF had lower thermal stability compared to pure polybutadiene-
based TPU matrix. Barick et al. [90] found that thermal stability of TPU
significantly increases with increasing vapor-grown CNF content. Roy et al.
[183] prepared NCs based on a TPU/acrylonitrile butadiene rubber blend using
CNF-layered double hydroxide (LDH) hybrids as nanofillers. In this study, a
new synthetic route was applied to develop CNF-LDH hybrids through a
noncovalent assembly using sodium dodecyl sulfate as bridging linker
between magnesium-aluminum LDH and CNF. DSC and TGA analyses
demonstrated maximum improvements in the melting temperature (5°C),
crystallization temperature (17°C), and thermal stability (14°C) in the 0.50
wt.% surfactant-modified CNF-LDH blend compared to the pure blend. This
enhancement in the thermal properties of the NCs could be attributed to the
homogeneous dispersion, strong interfacial interaction between TPU matrix
and modified CNF, and synergistic effect [183].
Špirková et al. [188] observed no change in Tg value of soft polybutadiene
segment (determined by DSC) between TPU/CNF NCs and pure TPU matrix,
which indicates that the presence of CNF does not significantly influence the
flexibility of polybutadiene-based soft segments in TPU/CNF NCs. On the
other hand, increased values of melting temperature of hard segment and
enthalphy of melting for TPU/CNF NCs compared to pure TPU matrix
confirmed preferable specific interactions between CNF and hard segment
domains in TPU (and also partially in the soft hard segment interphase).
TPU/CNF composite containing 0.75 wt.% of CNF had the best functional
(i.e., mechanical, surface and thermal) properties, and so this material was the
best candidate for applied use, e.g., as elastomeric protective thermoplastic
electroinsulating film. In this study [188], CNFs were used as a reinforcing
filler in low concentrations (up to 2.25 wt.%, i.e., below expected conductivity
threshold concentration of 3.5 wt.%) in order to prepare a non-conductive
composite material. Barick et al. [90] observed that the melting temperature
and the Tg values of soft segments were higher with the inclusion of CNF. The
increase in melting temperature of soft segments with added CNF could be

attributed to the heterogeneous nucleating effect of CNF, which may enhance
the melt crystallization process [90]. The nucleating effect of the CNF can be
explained in terms of the greater chemical affinity of the polar nanofibers
towards the TPU matrix and also due to better interaction between CNF and
the TPU matrix arising from the large surface area of the CNF [90].
Barick et al. [90] reported that the electrical conductivity of TPU NCs
increased with increasing CNF content. Al-Saleh et al. [189] showed that in
the case of polymer/CNF composites, the loading of untreated CNFs above 3.5
wt.% should produce conductive materials. Tibbetts et al. [190] observed that
polymer NCs containing vapor-grown CNF exhibited a percolation threshold
of below 1 wt.% and resistivities of below 0.15 Ω cm were achieved with a
CNF loading near 15 wt.%.
SILVER NANOCOMPOSITES
Introducing inogranic nanoparticles into a polymer matrix can
significantly improve thermal, mechanical, optical, electrical, magnetic and
flammability properties [191]. The most commonly used fillers in polymer
NCs are spherical nanoparticles, although the use of anisotropic nanoparticles
of unusual optical properties can produce interesting properties in the
composites, which arise from interactions between polymers and particles. The
properties of polymer composites depend on the type of nanoparticles, their
size, shape, and concentration and their interaction with polymer matrix.
Silver and gold nanoparticles are of great importance due to their size-
dependent functional properties, and it is significant that they can be
synthesized with narrow size distributions. Moreover, Ag nanoparticles have
attracted considerable attention due to their strong surface plasmon resonance
(SPR) [192]. Different types of Ag nanopaticles such as one-dimensional
nanostructures [193], nanorods [194], nanowires [194], nanoprisms [195],
triangular nanodisks [196], and nanodisks [197] are used to prepare
polymer/Ag NCs. There are various methods for synthesizing Ag
nanoparticles with controlled size and shape. The most common methods are
based on the reduction of Ag salts by strong reducing agents such as sodium
borohydride [198], sodium hydroxide [199] or hydrazine [200] in aqueous or
organic media [201] and stabilization of the synthesized nanoparticles through
functionalization or by surfactants [202]. Another method for synthesizing Ag
nanoparticles is using weak reducing agents such as sodium citrate [203],
glucose [204] or polyols [205].
Ag nanoparticles are used as nanofillers for TPU matrices due to their
antimicrobial activity, low human toxicity, good biocompatibility, oxidation
resistance, good electrical conductivity, and possibility to improve thermal and
other properties of the pure TPU [206]. Application of Ag particles is mostly
related to their physical properties such as size, shape, stability, surface charge
and surface modifications. Ag nanoparticles have found many applications in
different fields such as medicine, nanomedicine, pharmacy, optics and
electronics. Because of its biocompatibility and antimicrobial properties, Ag
has long been used as a nanofiller in the fields of food production, medical
devices, cosmetics, water and air purification systems. Ag has been included
as a coating or filler on or into polymer films through a variety of chemical
and physical methods. Especially, the use of Ag nanoparticles was investigated
as a constituent of coatings on medical devices and implants [207], due to their
strong bactericidal activity against both Gram-positive and Gram-negative
bacteria, including multiresistant strains.
The main problem in polymer nanocomposite technology is the prevention
of agreggation of inorganic particles, which occurs due to their high surface
energy (typically 500-2000 mJ/m2 for inorganic materials) [208]. In addition,
the specific surface area of nanoparticles also plays a crucial role in interfacial
interactions and in the formation of the interface, which consists of a layer of
high density polymer around the particle. This problem can be overcome by
modification of particle surfaces using chemical or physical methods which
can improve the interfacial interaction between the inorganic particles and the
polymer matrix.
Many attemps have been made to disperse Ag nanoparticles in TPU
matrix using various methods. A common ex situ method is the blending
technique to homogenize the previously prepared particles into the polymer.
Other methods of improving Ag nanoparticles’ dispersion employ the
following techniques: in situ chemical reduction of silver nitrate in the
presence of prepared prepolymers [209], modification of Ag nanoparticle
surfaces by iodination [210], and sol-gel mixing of organic polymers into
inorganic solutions [211]. Chemical methods often involve the presence of
stabilizers to prevent agglomeration [212]. Metal nanoparticles must be

stabilized to prevent their uncontrollable aggregation, to control their final
shape and size and to allow particle solubility in various solvents. Therefore, it
is important to explore new methods for preparation of stabilized metal
nanoparticle-polymer composites.
TPU/SILVER NANOCOMPOSITES
According to the literature, adding Ag nanoparticles into TPU matrix has a

great effect on the mechanical properties of TPU. Karak et al. [213] prepared
sunflower-oil based hyperbranched TPU/Ag and linear TPU/Ag NCs by in situ
catalytic reduction of a silver salt. The mechanical properties of both
hyperbranched TPU and linear TPU NCs were improved compared to pure
polymers. The hyperbranched structural units stabilized the Ag nanoparticles
better than the linear TPU and allowed a homogeneous distribution within the
matrix with a narrow size-window even at higher Ag loading. The authors
found the most significant enhancement in elongation at break for
hyperbranched TPU NCs with Ag salt loading up to 5 wt.%, although the
value decreased with increasing Ag loading to 7.5 wt.% Ag salt. The increased
values of elongation at break could be due to the higher molecular weight of
the polymer in hyperbranched TPU NCs, which in turn could be due to the
formation of complexes of the components (prepolymer, macrodiol,
diisocyanate, hyperbranched polyester) with the Ag ions [213]. The decreased
values of elongation at break could have been due to aggregation of Ag
nanoparticles at high loading (7.5 wt.%) as confirmed by TEM results [213].
The use of a natural product, a monoglyceride of sunflower oil, in the
polymeric matrix resulted in further improvement in NC toughness and makes
the materials interesting in the context of the successful use of sustainable raw
materials. Han [214] prepared hyperbranched waterborne TPU (HWPU)/Ag
NCs by incorporating Silver Sol into the HWPU. The DMTA results showed
that the HWPU/Ag NCs had higher glass transition temperature of hard
segments (TgHS) and dynamic storage modulus than did pure HWPU,
especially when incorporated Ag nanoparticles were protected by the third
generation hyperbranched polyester. The increase in TgHS was due to the
interfacial interactions between Ag nanoparticles and the HWPU matrix,

which limits the movement of chain segment [214]. Moreover, the Young’s
modulus and tensile strength of HWPU/Ag NCs increased with increasing
generation number of hyperbranched polyester. This was because the Ag
nanoparticles were better dispersed in the HWPU matrix of the third
generation than the previous generations and so could better restrict the
mobility of the chain segment in this HWPU [214].
Thermal Properties
Adding Ag nanoparticles into the TPU matrix significantly improves the

thermal properties of TPU. Karak et al. [213] observed improved thermal
stability for hyperbranched TPU/Ag and linear TPU/Ag NCs (about 50-75°C
upwards shift of maximum decomposition temperature) compared to the pure
polymers. This was mainly due to the reduced molecular mobility of the
polymer chain, caused by interactions with the Ag nanoparticles [213]. In
addition, metallic Ag is very thermostable, which enhances the thermal
stability of the NCs. Macocinschi et al. [215] demonstrated that TPU and
TPU-extracellular matrix bio-NCs containing in situ generated Ag
nanoparticles had better thermal stability than pure TPU because the initial
decomposition temperature (Tonset) was higher. The well-homogenized Ag
nanoparticles in the TPU matrix restricted the polymer chain mobility and
served as nucleation sites facilitating crystallization, inducing better thermal
stability [215].
Karak et al. [213] observed a slight increase in melting temperature and Tg
with increasing Ag content. However, the crystallinity of hyperbranched TPU
and linear TPU was not significantly affected by the presence of Ag
nanoparticles [213]. A slight increase in crystallinity was observed only for
hyperbranched TPU up to 2.5 wt.% Ag salt which might have been due to the
fact that the nanoparticles can act as a nucleating agent for the crystal growth
of the soft poly(-caprolactone) segment in TPU [213]. Han [214] observed that
the initial decomposition temperatures of the HWPU/Ag NCs increased
slightly with increasing generation number of hyperbranched polyester.
Macocinschi et al. [215] studied thermal behavior of TPU and TPU-
extracellular matrices with and without incorporated Ag nanoparticles. They
demonstrated that the homogeneous dispersion of Ag nanoparticles into the
TPU matrix led to small changes of Tg (up to 1°C) and decreased enthalphy of
melting, because Ag nanoparticles facilitate the crystallization of TPU.
Furthermore, for TPU-extracellular matrix/Ag bio-NCs, small changes (up to

1°C) of Tg and decreases of melting enthalpies were observed [215]. These
results are explained by the presence of both metallic Ag and oxidized state
Ag in the TPU but only oxidized state Ag in the TPU-extracellular matrix.
Antimicrobal and Biocompatible Properties
The use of dispersed Ag nanoparticles for enhancing the biocompatibility

and antimicrobial activity of TPU has been reported [71, 73]. Also, stabilizing
Ag nanoparticles with polymer decreased the nanoparticles’ potential to
aggregate and, thus, increased the antimicrobial activity and biocompatibility
compared to unstabilized Ag nanoparticles [216, 217]. The antimicrobial and
biocompatible properties of TPU/Ag NCs are important properties of these
materials which can be used in medicine to reduce infections, as well as to
prevent bacterial colonization on various medical devices (i.e., catheters,
vascular grafts and prostheses). Hsu et al. [73] prepared polyester-type
waterborne TPU/Ag NCs containing various small amounts (15-75 ppm) of
Ag nanoparticles (size 5 nm) and observed that TPU/Ag NCs showed much
lower bacterial adhesion than the pure TPU for Bacillus subtilis, Escherichia
coli and Ag+-resistant E. coli. TPU/Ag NC containing 30 ppm of Ag
nanoparticles showed superior physicochemical properties, good
biocompatibility and microbiostatic effects [73]. They presumed that Ag
nanoparticles on (or near) the surface of TPU/Ag NCs could directly interact
with bacteria and at the same time release some Ag species to inhibit bacterial
growth [73]. Overloading Ag nanoparticles (50 and 75 ppm) could cause them
to aggregate in the bulk and on the surface, which may interfere with exposure
of the nanoparticles (and release of Ag) to bacteria. Therefore, the authors
concluded that dispersion of Ag nanoparticles was more crucial than the
amount of Ag nanoparticles [73]. TPU NCs prepared with 30 ppm Ag showed
enhanced fibroblast attachment and endothelial cell response, as well as
reduced monocyte and platelet activation, relative to pure TPU or NCs with
the other Ag contents [73]. Subcutaneous implantation in rats confirmed the
better biocompatibility of the NCs [73].
Triebel et al. [218] prepared TPU/Ag NCs using two different methods: in
situ preparation (the Ag nanoparticles were obtained in situ by thermal
reduction of silver acetate in the melt of a TPU during melt mixing) or ex situ
preparation (the Ag nanoparticles were synthesized in an invertible polyester
resulting in much smaller particle sizes and the obtained Ag particles were
surrounded by the polyester cage, which stabilizes them with respect to

agglomeration). Due to the compatibility of the invertible polyester with the
TPU matrix, which was confirmed by DSC, good dispersion of the Ag
particles surrounded by the polyester cage was expected after the incorporation
into the TPU by melt mixing with a twin-screw extruder. At a constant weight
percentage of silver in the polymer, the composite with the ex situ Ag
nanoparticles released Ag ions about two orders of magnitude higher than the
level released from the composite with the in situ Ag nanoparticles. This result
was explained by the smaller particle size the ex situ particles exhibited and,
following from that, by the higher specific surface area of the smaller particles
and a higher coefficient of diffusion for Ag ions [218]. Moreover, the ex situ
preparation method led to a sufficiant release of Ag ions to achieve
antimicrobial characteristics even at low concentrations of less than 0.1 wt.%
Ag [218]. Paladini et al. [219] demontrated that Ag modified TPU catheters
obtained by an innovative Ag deposition technology based on the
photoreduction of silver nitrate had strong antimicrobial effects on
Staphylococcus aureus.
Several hypotheses for the antimicrobial mechanism of Ag nanoparticles
have been proposed in the literature. Kim et al. [220] observed that Ag
nanoparticles generated free radicals and damaged cell membranes. This
implies that Ag nanoparticles could be used as effective growth inhibitors for
various microorganisms, making them applicable in diverse medical devices
and antimicrobial control systems. Shrivastava et al. [221] found that Ag
nanoparticles exerted their antibacterial effect by anchoring to and penetrating
the bacterial wall, and modulating cellular signaling by dephosphorylating
putative key peptide substrates on tyrosine residues.
Good electrical properties can be obtained using anisotropic metal NPs

(nanowires, nanorods, nanotubes, etc.), arising from interactions between
polymers and nanoparticles. The conductivity of TPU containing dispersed
conductive metal nanosize particles are controlled by the size, concentration
and shape of the metal nanoparticles and the interactions betweeen the TPU
and the metal particles. Also, the distribution and morphology of the metal
nanoparticles affect electrical percolation in composites. Very few studies
concerning electrical propertis of TPU/Ag NCs have been reported. Ma et al.
[222] demonstrated highly conductive stretchable fibers synthesized by the
scalable wet spinning process using flower-shaped Ag nanoparticles with

nanodisc-shaped petals (Ag nanoflowers) and TPU (Ag-TPU fibers). An
extraordinarily high conductivity (41 245 S/cm) was obtained for Ag
nanoflower-based TPU, which was 2 orders of magnitude greater than that of
fibers synthesized using CNTs, spherical Ag nanoparticles and
poly(vinylidene fluorideco-hexafluoropropylene) (PVDF-HFP) [223]. This
was due to the enhanced surface area and vigorous coalescence of nanodisc-
shaped petals during the curing process [222].
SILICA NANOCOMPOSITES
TPU/silica NCs have received great attention in recent years and these
NCs have been employed in a variety of applications. Fumed silica (SiO2;
nanosilica) is a porous inorganic particle with high specific surface area, high
surface energy, good chemical and physical stability, low toxicity,
biocompatibility, degradability under physiological conditions, and low cost
[224, 225]. Fumed silicas can be used as fillers added into a TPU matrix in
order to improve the thermal, barrier, rheological and mechanical properties as
well as the corrosion resistance of TPU [226-229]. Namely, a nanometer-sized
silica used as filler to prepare TPU NCs increased the tensile strength and
elongation at break beyond the level that was reached by a micrometer-sized
filler [230]. According to the literature, the enhancement in properties was
ascribed to the creation of hydrogen bonds between the hydroxyl groups on the
nanosilica surface and the soft segments of the TPU, favouring the degree of
phase separation [231-234]. Moreover, incorporating nanosilica into the TPU
matrix produced films with enhanced properties, such as increased indentation
hardness, high resistance to whitening and high permeability to water vapor.
TPU/silica NCs are also useful as tough, abrasion-resistant coatings with
increased friction coefficient and reduced tackiness, with no change in gloss.
For the preparation of TPU/silica NCs, three different methods are used:
directly blending polymer with nanosilica, sol-gel processes and in situ
polymerization. Functionalization of the silica is the usual method to generate
in situ a good connectivity between the filler and the polymer matrix [235].
STRUCTURE-PROPERTY RELATIONSHIPS OF TPU/

SILICA NANOCOMPOSITES
According to the literature, the interactions between the silanol groups on

nanosilica and the TPU determine the mechanical properties of these NCs.
Vega-Baudrit et al. [227] observed improvement in tensile strength and
elongation at break of TPU/nanosilica NCs (prepared by solvent blending), by
increasing the silanol content. Also, the TPU/nanosilica NCs showed greater
storage and loss moduli then the pure TPU due to the interactions produced
between the nanosilica and the TPU [227]. Chen et al. [236] reported that
incorporating nanosilica (10 nm) into the polyester-based TPU significantly
increased the tensile strength, but the elongation at break was only slightly
increased. The hardness, glass transition temperature, adhesion, and viscosity
of the resulting TPU NC films were also enhanced. Lee et al. [237] prepared
poly(tetramethylene oxide) (PTMO)-based TPU NCs containing 1-10 wt.%
nanosilica by in situ polymerization. Those authors found NCs containing the
nanosilica at up to 3wt.% exhibited significantly enhanced tensile strength and
elongation at break; these values decreased when the nanosilica content
increased further. The nanosilica particles were well dispersed in the TPU
matrix up to 3 wt.% [237]. The enhanced tensile strength of NC films could be
attributable to the strong interfacial interaction between the polymer chains
and the nanosilica particles [237]. They suggested that large enhancements in
the elongation may be due to the effect of the dispersed nanosilica particles,
which can act as chain extenders. The decreased tensile strength and
elongation at break of the NCs with high contents of nanosilica (above 5 wt.%)
could be due to larger aggregates of the nanosilica particles in the TPU matrix,
as explained for the crystallization behavior of the PTMO segments [237]. For
comparison with TPU NCs containing 3 wt.% silica prepared by in situ
polymerization, in this study [237], TPU without nanosilica was melt blended
with 3 wt.% nanosilica. They found that the nanosilica particles in TPU/silica
NCs prepared by in situ polymerization were covalently linked to the
polyurethane molecules in the interfaces. This was in contrast to TPU NCs
containing the same nanosilica content prepared by melt blending [237]. The
tensile strength and elongation at break of TPU/silica NCs prepared by melt
blending were lower than those of TPU/silica NCs obtained by in situ
polymerization, even though the silica contents were equal in both cases. This
result implies that the nanosilica particles in the melt blended NCs were more
aggregated and less efficiently dispersed in the TPU matrix [237]. Malay et al.
[238] prepared NCs containing polyurethane-urea (PUU) and silica
nanoparticles of various sizes (29, 74 and 215 nm) and filler loadings (1-40
wt.%), by solution blending using isopropyl alcohol. This preparation method
led to even distribution of silica nanoparticles in the PUU matrix. PUU
copolymer was based on a cycloaliphatic diisocyanate and poly(tetramethylene
oxide) soft segments and had urea hard segment content of 20% by weight.
Incorporation of silica nanoparticles of smaller sizes (29 and 74 nm) led to
higher modulus and tensile strength of the NCs, but their elastomeric
properties were retained. Increased filler content of up to about 20 wt.%
resulted in materials with higher elastic moduli and tensile strength.
Improvements in tensile properties of these NCs were attributed to the
interactions between the surface of silica nanoparticles and ether linkages of
the polyether segments of the copolymers [238]. Chen et al. [69] observed that
maximum values of tensile properties, glass transition temperature and
abrasion resistance of TPU/silica NCs were obtained when the particle size of
silica was about 28 nm.
Špirková et al. [50] studied the influence of the size (7 to 40 nm) and
hydrophobic/hydrophilic character of silica particles on the internal particle
arrangement in polycarbonate-based TPU matrix with 18 wt.% hard
(hexamethylene diisocyanate-1,4-butanediol) segment; they also studied the
functional i.e., mechanical, surface and barrier properties of the TPU NCs.
They found that even a small amount of silica nanoparticles (1 wt.%) in the
TPU NCs can significantly influence their functional properties. Mechanical
properties are affected by the size and character of nanoadditives. Nanosilica
particles interact preferably with hard segments, as detected by DMTA. The
authors observed that significant improvement of tensile properties of TPU
NCs compared to pure TPU were achieved only by adding hydrophobic
nanosilica (12 nm particles), resulting in very elastic but simultaneously,
highly tough TPU NC-based coating/film. TEM confirmed that smaller
particles (hydrophobic 12 nm and hydrophilic 7 nm) were dispersed in TPU
matrix in the form of agglomerates of ca 1 µm size, while hydrophilic 40 nm
particles were present as small formations (up to 0.5 µm) [50]. The presence of
nanosilica has only negligible influence on gas transport properties. Recently,
Chiacchiarelli et al. [239] reported the relationship between degree of
nanosilica dispersion and the tensile properties of TPU NCs. They observed
that the final dispersion and the mechanical properties depend on the silica
type and weight concentration, mixing methods and nanosilica surface
properties. They found that using ultrasonic dispersion, high shear mixing or
tip sonication had effects on reducing agglomerate size and the interparticle
distance. Peruzzo et al. [240] investigated how the incorporation route of the
nanosilica dispersion affected nanoparticle distribution and the properties of
waterborne polyurethane/nanosilica composite films. They observed
enhancement of mechanical properties for in situ synthesized composites,
compared with the blend composites, as a result of the polyurethane-nanosilica
interation and distribution of the nanosilica particles within the waterborne
polyurethane matrix.
Thermal Properties
Vega-Baudrit et al. [227] reported that the addition of nanosilicas

increased the thermal stability of the TPU, irrespective of the nanosilica
specific surface area and silanol content. They observed that the thermal
decomposition for TPU/nanosilica NCs starts at higher temperature (384-
390°C) as compared to the decomposition of the pure TPU (380°C). Malay et
al. [238] showed that thermal resistance of polyether based poly(urethane-
urea) NCs was improved in the presence of silica particles. Furthermore, Lai et
al. [241] reported that the initial weight loss temperature of TPU/silica NCs
did not increase with increasing silica content. At a weight loss of 50%, the
silica did not increase the hybrid thermal resistance significantly, which
implied that the silica was not an effective heat barrier for TPU/silica hybrids.
According to Vega-Baudrit et al. [227], the Tg values of the soft segment
in TPU/nanosilica NCs are located at 48°C, being independent of nanosilica
addition and the specific surface area and silanol content of the nanosilica.
They found that adding nanosilica decreased the enthalpy of melting and
crystallization of TPU, which is in agreement with the lower degree of phase
segregation in the TPU without nanosilica. Malay et al. [238] showed that no
significant change in the soft PTMO segment Tg in poly(urethane-urea)
copolymer was observed, regardless of the silica size or the amount
incorporated. This phenomenon, which was also observed by several other
researchers [230, 242-244], indicates no change in the soft segment mobility in
the presence of silica particles dispersed in the polymeric matrix. Petrović et
al. [230] showed that Tg increased with increasing nanosilica particle
concentration (with an average diameter of about 12 nm) in poly(propylene
oxide)-based TPU due to strong interaction between the TPU matrix and the
nanosilica.
Barrier Properties
Khosravi et al. [245] studied the effect of silica nanoparticles on

propane/methane and ethane/methane separation with polyester- and
polyether-based TPU membranes. They demonstrated that by increasing the
amount of silica particles in polyester-based TPU, permeability decreases.
Furthermore, by introducing silica particles in polyether-based TPU, the
permeability of these gases first increases, and then decreases. The increase in
permeability by increasing the silica content up to 2.5 wt.% could be the result
of formation of active sites at the polymer-silica interface that are suitable sites
for adsorption condensable gases, whereas the reduction in permeability by
further increasing the silica content could be the result of the introduction of
tortuous paths in TPU membranes and reduction of gas diffusivity [245]. They
found that the selectivity for propane over methane increases with the
inclusion of silica particles in the polyether-based TPU membranes, while it
decreases in polyester-based TPU membranes. Sadeghi et al. [246] showed
that increasing silica content decreased gas permeability of polyether-based
TPU membranes, but increased CO2/N2 and CO2/CH4 selectivities. Gas
permeability reduction can be attributed to the presence of silica particles in
the soft segments of TPU NCs, which reduces the free volume of soft
segments and reduces the passage areas of gas molecules. In the case of
TPU/silica (20 wt.%) NC membrane, the obtained CO2/N2 selectivity was 1.65
times that of pure TPU, while the CO2 permeability reduction of TPU/silica
membrane was 35.6% in comparison with pure TPU [246]. They found that in
the case of CO2, its lower permeability reduction with increasing silica content
compared to other gases is due to both its small molecular size and also its
more dissolution in the membrane due to the increased polar OH groups in the
polymer with increasing silica content [246].
CONCLUSION AND FUTURE PROSPECTS

Thermoplastic polyurethanes, thermoplastics with the mechanical
performance characteristics of rubber, are described as “bridging the gap
between rubber and plastics”. Considerable efforts have been made in recent
years in order to improve TPU properties, and which have involved chemical
modifications to their molecular structures and addition of nanoparticles into
TPU matrices. The improvements include more attractive mechanical, thermal,
barrier, electrical, flame retardant, biocompatible and antimicrobial properties,
which are usually the result of synergistic effects between TPU matrices and
nanoparticles. One of the key features that determines the enhancement such
properties in NCs is the quality of dispersion of the nano-sized particles in the
polymeric matrix. Dispersion of the nanoparticles depends on polymer matrix-
filler compatibility and surface energy of the particles as well as on the
preparation method. Therefore, the properties of TPU NCs strongly depend on
the synthesis procedure, type of the polymer matrix, the state of the particle
dispersion in polymer matrix and specifications of the used nanofillers, such as
type, size, shape and concentration. The application of TPU NCs in new fields
requires that other properties must be developed to meet the desired
functionalities without affecting the main polymer properties (e.g., mechanical
properties). A prerequisite for useful, functional TPU-based NCs is adequate
interfacial adhesion between the inorganic and the organic materials, which is
critical for producing a high performance composite. Namely, TPU NCs with
tunable properties, which will be capable of numerous future applications, can
be developed by varying the nanoparticle types and characteristics. In order to
understand TPU-nanoparticle property relationships so that the performances
of final materials can be predicted, and for hierarchically ordered NCs to be
fabricated with properties desired, it is necessary to discover methods for
preparing TPU NCs and establish guidelines for process optimization. Better
understanding of the subtle structures and conformations at the TPU-nanofiller
interface can be provided by developing and applying new experimental
techniques. Other areas that need great attention are long-term mechanical,
thermal, and other environmental ageing, tear resistance, and fatigue
performance of TPU NC materials. One important new field in nanocomposite
research is rheology, used to determine the effects of nanocomposite formation
and processing effects on nanocomposite structure and properties. Perspective
future applications of TPU NCs are in biomedical implants (e.g., softer,
tougher, implantable electrode insulation for pacemakers), drive belts, o-rings,
fuel hoses and more durable TPU seals. Further applications of TPU NCs also
likely will include tougher covers on protective wear and thin aliphatic gas
barrier coatings for encapsulating the next generation of thin-film
photovoltaics. Expected future technological development in the TPU
nanocomposite field will lead to new TPU NCs with advanced properties. In
comparison to other thermoplastic elastomers that are currently available,
TPUs are expected to remain leading in the field of nanocomposite
technology.
ACKNOWLEDGMENTS
This work was financially supported by the Ministry of Education,
Science and Technological Development of the Republic of Serbia (research
project numbers: 172062 and 172030).
REFERENCES
[1] Gogolewski, S. Colloid Polym Sci 1989, 267, 757-785.
[2] Cooper, S. L.; West, J. C.; Seymour, R. W. Plastics, Resins, Rubbers,
Fibers. Encyclopedia of Polymer Science and Technology. Volume 1.
New York: John Wiley; 1976. P. 521-543.
[3] Chattopadhyay, D. K.; Raju, K. V. S. N. Prog Polym Sci 2007, 32, 352-
418.
[4] Cherng, J. Y.; Hou, T. Y.; Shih, M. F.; Talsma, H.; Hennink, W. E. Int J
Pharm 2013, 450, 145-162.
[5] Drobny, J. G. Handbook of Thermoplastic Elastomers; William Andrew
Publishing, 2007.
[6] Martin, D. J.; Osman, A. F.; Andriani, Y.; Edwards, G. A. In
Thermoplastic polyurethane (TPU)-based polymer nanocomposites;
Gao F.; Ed.; Advances in Polymer Nanocomposites; Woodhead
Publishing Limited, 2012; pp. 321-350.
[7] Petrovic, Z. S.; Ferguson, J. Prog Polym Sci 1991, 16, 695-836.
[8] Vermette, R.; Griesser, H. J.; Laroche, G.; Guidoin, R. Biomedical
Appliactions of Polyurethanes; Landes Bioscience: Texas, 2001
[9] Rueda-Larraz, L.; d'Arlas, B. F.; Tercjak, A.; Ribes, A.; Mondragon, I.;
Eceiza, A. Eur Polym J 2009, 45, 2096-2109.
[10] Skarja, G. A.; Woodhouse, K. A. J Biomat Sci, Polym Ed 1998, 9, 271-
295.
[11] Skarja, G. A.; Woodhouse, K. A. J Appl Polym Sci 2000, 75, 1522-
1534.
[12] Kultys, A.; Rogulska, M.; Głuchowska, H. Polym Int 2011, 60, 652-659.
[13] Gunatillake, P. A.; Meijs, G. F.; Rizzardo, E.; Chatelier, R. C.;
McCarthy, S. J.; Brandwood, A.; Schindhelm, K. J Appl Polym Sci
1992, 46, 319-328.
[14] Martin, D. J.; Meijs, G. F.; Gunatillake, P. A.; McCarthy, S. J.;
Renwick, G. M. J Appl Polym Sci 1997, 64, 803-817.
[15] Sask, K. N.; Berry, L. R.; Chan, A. K. C.; Brash, J. L. Langmuir 2012,
28, 2099-2106.
[16] Desai, N. P.; Hubbell, J. A. J Biomed Mater Res 1991, 25, 829-843.
[17] Waletzko, R. S.; Korley, L. T. J.; Pate, B. D.; Thomas, E. L.; Hammond,
P. T. Macromolecules 2009, 42, 2041-2053.
[18] Špírková, M.; Poręba, R.; Pavličević, J.; Kobera, L.; Baldrian, J.;
Pekárek, M. J Appl Polym Sci 2012, 126, 1016-1030.
[19] Poręba, R.; Špírková, M.; Brožová, L.; Lazić, N.; Pavličević, J.;
Strachota, A. J Appl Polym Sci 2013, 127, 329-341.
[20] Pergal, M. V.; Antić, V. V.; Govedarica, M. N.; Goäevac, D.; Ostojić,
S.; Djonlagić, J. J Appl Polym Sci 2011, 122, 2715-2730.
[21] Pergal, M. V.; Antic, V. V.; Ostojic, S.; Marinovic-Cincovic, M.;
Djonlagic, J. J Serb Chem Soc 2011, 76, 1703-1723.
[22] Pergal, M. V.; Antic, V. V.; Tovilovic, G.; Nestorov, J.; Vasiljevic-
Radovic, D.; Djonlagic, J. J Biomat Sci, Polym Ed 2012, 23, 1629-1657.
[23] Pergal, M. V.; Nestorov, J.; Tovilovic, G.; Ostojic, S.; Godevac, D.;
Vasiljevic-Radovic, D.; Djonlagic, J. J Biomed Mater Res - Part A
2014, 102, 3951-3964.
[24] Pergal, M. V.; Nestorov, J.; Tovilović-Kovačević, G.; Jovančić, P.;
Pezo, L.; Vasiljević-Radović, D.; Djonlagic, J. Hem Ind 2014, 68, 731-
741.
[25] Pergal, M. V.; Stefanovic, i. S.; Gođevac, D.; Antic, V. V.; Milacic, V.;
Ostojic, S.; Rogan, J.; Djonlagic, J. J Serb Chem Soc 2014, 79, 843-866.
[26] Stefanovic, I. S.; Djonlagic, J.; Tovilovic, G.; Nestorov, J.; Antic, V. V.;
Ostojic, S.; Pergal, M. V. J Biomed Mater Res - Part A 2015, 103, 1459-
1475.
[27] Balaban, M.; Antic, V. V.; Pergal, M. V.; Francolini, I.; Martinelli, A.;
Djonlagic, J. J Serb Chem Soc 2012, 77, 1457-1481.
[28] Balaban, M.; Antić, V.; Pergal, M.; Godjevac, D.; Francolini, I.;
Martinelli, A.; Rogan, J.; Djonlagić, J. Polym Bull 2013, 70, 2493-2518.
[29] Stefanović, I. S.; Špírková, M.; Poręba, R.; Steinhart, M.; Ostojić, S.;
Tešević, V.; Pergal, M. V. Ind Eng Chem Res 2016, 55, 3960-3973.
[30] Hernandez, R.; Weksler, J.; Padsalgikar, A.; Runt, J. Macromolecules
2007, 40, 5441-5449.
[31] Kojio, K.; Nonaka, Y.; Masubuchi, T.; Furukawa, M. J Polym Sci Part
B: Polym Phys 2004, 42, 4448-4458.
[32] Kojio, K.; Nakamura, S.; Furukawa, M. J Polym Sci Part B: Polym
Phys 2008, 46, 2054-2063.
[33] Kojio, K.; Furukawa, M.; Motokucho, S.; Shimada, M.; Sakai, M.
Macromolecules 2009, 42, 8322-8327.
[34] Ma, J.; Liu, C.; Li, R.; Zhu, L.; Wu, H. Polym Bull 2012, 68, 1341-
1355.
[35] Zhang, R.; Deng, H.; Valenca, R.; Jin, J.; Fu, Q.; Bilotti, E.; Peijs, T.
Compos Sci Technol 2013, 74, 1-5.
[36] Prządka, D.; Jęczalik, J.; Andrzejewska, E.; Marciniec, B.; Dutkiewicz,
M.; Szłapka, M. React Funct Polym 2013, 73, 114-121.
[37] Osman, A. F.; Edwards, G. A.; Schiller, T. L.; Andriani, Y.; Jack, K. S.;
Morrow, I. C.; Halley, P. J.; Martin, D. J. Macromolecules 2012, 45,
198-210.
[38] Pei, A.; Malho, J.-M.; Ruokolainen, J.; Zhou, Q.; Berglund, L. A.
[39] Fernández-d’Arlas, B.; Ramos, J. A.; Saralegi, A.; Corcuera, M.;
Mondragon, I.; Eceiza, A. Macromolecules 2012, 45, 3436-3443.
[40] Rueda, L.; Garcia, I.; Palomares, T.; Alonso-Varona, A.; Mondragon, I.;
Corcuera, M.; Eceiza, A. J Biomed Mater Res Part A 2011, 97A, 480-
489.
[41] Rueda, L.; Saralegui, A.; Fernández d’Arlas, B.; Zhou, Q.; Berglund, L.
A.; Corcuera, M. A.; Mondragon, I.; Eceiza, A. Carbohydr Polym 2013,
92, 751-757.
[42] Bistričić, L.; Baranović, G.; Leskovac, M.; Bajsić, E. G. Eur Polym J
2010, 46, 1975-1987.
[43] Yoonessi, M.; Peck, J. A.; Bail, J. L.; Rogers, R. B.; Lerch, B. A.;
Meador, M. A. ACS Appl Mater Interfaces 2011, 3, 2686-2693.
[44] Zia, K. M.; Zuber, M.; Barikani, M.; Hussain, R.; Jamil, T.; Anjum, S.
Int J Biolog Macromol 2011, 49, 1131-1136.
[45] Serkis, M.; Poręba, R.; Hodan, J.; Kredatusová, J.; Špírková, M. J Appl
Polym Sci 2015, 132, doi: 10.1002/app.42672.
[46] Hwang, H.-D.; Kim, H.-J. React Funct Polym 2011, 71, 655-665.
[47] Lee, Y.-H.; Kim, E.-J.; Kim, H.-D. J Appl Polym Sci 2011, 120, 212-
219.
[48] Tseng, H. J.; Lin, J. J.; Ho, T. T.; Tseng, S. M.; Hsu, S. H. J Biomed
Mater Res Part A 2011, 99, 192-202.
[49] Fakirov, S. Handbook of Condensation Thermoplastic Elastomers;
WILEY-VCH Verlag GmbH & Co. KGaA 2005
[50] Špírková, M.; Poręba, R.; Brožová, L. In The influence of the size, shape
and character of nanofillers on functional properties of polyurethane
elastomers Wang X.; Ed.; Nanocomposites; Nova Science Publishers,

Inc.: 2013; Vol. pp 259-279.
[51] Yao, K. J.; Song, M.; Hourston, D. J.; Luo, D. Z. Polymer 2002, 43,
1017-1020.
[52] Zilg, C.; Thomann, R.; Mülhaupt, R.; Finter, J. Adv Mater 1999, 11, 49-
52.
[53] Nicolais, L. Metal Polymer Nanocomposites; John Wiley and Sons:
New Jersey, 2005
[54] Mancini, L. H.; Esposito, C. L. Nanocomposites: Preparation,
Properties and Performance; Nova Science Publishers, Inc.: New York,
2008
[55] Hussain, F.; Hojjati, M.; Okamoto, M.; Gorga, R. E. J Compos Mater
2006, 40, 1511-1575.
[56] Buchachenko, A. L. Russ Chem Rev 2003, 72, 375-391.
[57] Paul, D. R.; Robeson, L. M. Polymer 2008, 49, 3187-3204.
[58] Kamat, P. V. J Phys Chem B 2002, 106, 7729-7744.
[59] Novakov, I. A.; Dang, N. K.; Vaniev, M. A.; Sidorenko, N. V. Russ
Chem Bull 2013, 62, 276-284.
[60] Ma, P.-C.; Siddiqui, N. A.; Marom, G.; Kim, J.-K. Composites Part A
2010, 41, 1345-1367.
[61] Barick, A. K.; Tripathy, D. K. J Appl Polym Sci 2010, 117, 639-654.
[62] Finnigan, B.; Martin, D.; Halley, P.; Truss, R.; Campbell, K. Polymer
2004, 45, 2249-2260.
[63] Mishra, A. K.; Nando, G. B.; Chattopadhyay, S. J Polym Sci Part B:
Polym Phys 2008, 46, 2341-2354.
[64] Buffa, F.; Abraham, G. A.; Grady, B. P.; Resasco, D. J Polym Sci Part
B: Polym Phys 2007, 45, 490-501.
[65] Malkappa, K.; Rao, B. N.; Jana, T. Polymer 2016, 99, 404-416.
[66] Loos, M. R.; Yang, J.; Feke, D. L.; Manas-Zloczower, I.; Unal, S.;
Younes, U. Composites Part B 2013, 44, 740-744.
[67] Pattanayak, A.; Jana, S. C. Polymer 2005, 46, 3394-3406.
[68] Song, M.; Xia, H. S.; Yao, K. J.; Hourston, D. J. Eur Polym J 2005, 41,
259-266.
[69] Chen, Y.; Zhou, S.; Yang, H.; Wu, L. J Appl Polym Sci 2005, 95, 1032-
1039.
[70] Jeon, I.-Y.; Baek, J.-B. Materials 2010, 3, 3654-3674.
[71] Paul, D.; Paul, S.; Roohpour, N.; Wilks, M.; Vadgama, P. J Funct
Biomater 2013, 4, doi: 10.3390/jfb4040358.
[72] Sawant, S. N.; Selvaraj, V.; Prabhawathi, V.; Doble, M. PLOS ONE
2013, 8, 1-9.
[73] Hsu, S.-h.; Tseng, H.-J.; Lin, Y.-C. Biomaterials 2010, 31, 6796-6808.
[74] Pavlidou, S.; Papaspyrides, C. D. Prog Polym Sci 2008, 33, 1119-1198.
[75] Alexandre, M.; Dubois, P. Mater Sci Eng R 2000, 28, 1-63.
[76] Khudyakov, I. V.; Zopf, R. D.; Turro, N. J. Des Monomers Polym 2009,
12, 279-290.
[77] Haider, S.; Khan, Y.; Almasry, W. A.; Haider, A. In Thermoplastic
Nanocomposites and Their Processing Techniques; El-Sonbati A.; Ed.;
Thermoplastic - Composite Materials; InTech, 2012
[78] Wang, Z.; Pinnavaia, T. J. Chem Mater 1998, 10, 3769-3771.
[79] Meng, X.; Du, X.; Wang, Z.; Bi, W.; Tang, T. Compos Sci Technol
2008, 68, 1815-1821.
[80] Meng, X.; Wang, Z.; Yu, H.; Du, X.; Li, S.; Wang, Y.; Jiang, Z.; Wang,
Q.; Tang, T. Polymer 2009, 50, 3997-4006.
[81] Jeong, E. H.; Yang, J.; Hong, J. H.; Kim, T. G.; Kim, J. H.; Youk, J. H.
Eur Polym J 2007, 43, 2286-2291.
[82] Manias, E.; Touny, A.; Wu, L.; Strawhecker, K.; Lu, B.; Chung, T. C.
Chem Mater 2001, 13, 3516-3523.
[83] Chavarria, F.; Paul, D. R. Polymer 2006, 47, 7760-7773.
[84] Xia, H.; Shaw, S. J.; Song, M. Polym Int 2005, 54, 1392-1400.
[85] Finnigan, B.; Jack, K.; Campbell, K.; Halley, P.; Truss, R.; Casey, P.;
Cookson, D.; King, S.; Martin, D. Macromolecules 2005, 38, 7386-
7396.
[86] Tortora, M.; Gorrasi, G.; Vittoria, V.; Galli, G.; Ritrovati, S.; Chiellini,
E. Polymer 2002, 43, 6147-6157.
[87] Xiong, J.; Liu, Y.; Yang, X.; Wang, X. Polym Degrad Stab 2004, 86,
549-555.
[88] Subramani, S.; Lee, J.-Y.; Choi, S.-W.; kim, J. H. J Polym Sci Part B:
Polym Phys 2007, 45, 2747-2761.
[89] Xu, R. J.; Manias, E.; Snyder, A. J.; Runt, J. J Biomed Mater Res Part A
2003, 64A, 114-119.
[90] Barick, A. K.; Tripathy, D. K. Composites Part A 2010, 41, 1471-1482.
[91] Marshall, R.; Martin, D. In Nanocomposite polyurethanes for extreme
applications in oil and gas; High Performance Thermoplastic &
Composites for Oil and Gas Applications 2011; Smithers: Houston,
USA, 2011
[92] Dennis, H. R.; Hunter, D. L.; Chang, D.; Kim, S.; White, J. L.; Cho, J.
W.; Paul, D. R. Polymer 2001, 42, 9513-9522.
[93] Giannelis, E. P. Appl Organomet Chem 1998, 12, 675-680.

[94] Xanthos, M.; Biesenberger, J. Reactive Extrusion: Principles and
Practice; Hanser: Munich, 1992
[95] He, H.; Duchet, J.; Galy, J.; Gérard, J.-F. J Colloid Interface Sci 2006,
295, 202-208.
[96] Finnigan, B.; Martin, D.; Halley, P.; Truss, R.; Campbell, K. J Appl
Polym Sci 2005, 97, 300-309.
[97] Chang, J.-H.; An, Y. U. J Polym Sci Part B: Polym Phys 2002, 40, 670-
677.
[98] Kim, B. K.; Seo, J. W.; Jeong, H. M. Eur Polym J 2003, 39, 85-91.
[99] Chen, T. K.; Tien, Y. I.; Wei, K. H. Polymer 2000, 41, 1345-1353.
[100] Tien, Y. I.; Wei, K. H. J Appl Polym Sci 2002, 86, 1741-1748.
[101] Tien, Y. I.; Wei, K. H. Macromolecules 2001, 34, 9045-9052.
[102] Vaia, R. A.; Giannelis, E. P. Macromolecules 1997, 30, 8000-8009.
[103] Cruz, S. M.; Viana, J. C. Macromol Mater Eng 2015, 300, 1153-1162.
[104] Xu, R.; Manias, E.; Snyder, A. J.; Runt, J. Macromolecules 2001, 34,
337-339.
[105] Leszczyńska, A.; Njuguna, J.; Pielichowski, K.; Banerjee, J. R.
Thermochim Acta 2007, 453, 75-96.
[106] Pavličević, J.; Špirkova, M.; Strachota, A.; Szécsényi, K. M.; Lazić, N.;
Budinski-Simendić, J. Thermochim Acta 2010, 509, 73-80.
[107] Poręba, R.; Špirková, M.; Pavličević, J.; Budinski-Simendić, J.;
Mészáros Szécsényi, K.; Holló, B. Composites Part B 2014, 58, 496-
501.
[108] Salahuddin, N.; Abo-El-Enein, S. A.; Selim, A.; Salah El-Dien, O. Appl
Clay Sci 2010, 47, 242-248.
[109] Chrissafis, K.; Antoniadis, G.; Paraskevopoulos, K. M.; Vassiliou, A.;
Bikiaris, D. N. Compos Sci Technol 2007, 67, 2165-2174.
[110] Chang, J.-H.; An, Y. U.; Sur, G. S. J Polym Sci Part B: Polym Phys
2003, 41, 94-103.
[111] Dlamini, D. S.; Mishra, S. B.; Mishra, A. K.; Mamba, B. B. J Inorg
Organomet Polym Mater 2011, 21, 229-236.
[112] Tabuani, D.; Bellucci, F.; Terenzi, A.; Camino, G. Polym Degrad Stab
2012, 97, 2594-2601.
[113] Bellucci, F.; Camino, G.; Frache, A.; Sarra, A. Polym Degrad Stab
2007, 92, 425-436.
[114] Osman, A. F.; Andriani, Y.; Edwards, G. A.; Schiller, T. L.; Jack, K. S.;
Morrow, I. C.; Halley, P. J.; Martin, D. J. RSC Adv 2012, 2, 9151-9164.
[115] Herrera-Alonso, J. M.; Marand, E.; Little, J. C.; Cox, S. S. J Membr Sci
2009, 337, 208-214.
[116] Kiliaris, P.; Papaspyrides, C. D. Prog Polym Sci 2010, 35, 902-958.
[117] Kannan, M.; Thomas, S.; Joseph, K. J Vinyl Addit Techn 2015, n/a-n/a.
[118] Styan, K. E.; Martin, D. J.; Poole-Warren, L. A. J Biomed Mater Res
Part A 2008, 86A, 571-582.
[119] Tee, N.; Zhu, Y.; Mortimer, G. M.; Martin, D. J.; Minchin, R. F. Int J
Nanomedicine 2015, 10, 2363-2375.
[120] Mishra, A.; Das Purkayastha, B. P.; Roy, J. K.; Aswal, V. K.; Maiti, P.
[121] Styan, K. E. Polyurethane organosilicate nanocomposites for novel use
as biomaterials [Thesis]; 2006.
[122] Edwards, G.; Halley, P.; Kerven, G.; Martin, D. Thermochim Acta 2005,
429, 13-18.
[123] Rahman, M. M.; Kim, H.-D.; Lee, W.-K. J Appl Polym Sci 2008, 110,
3697-3705.
[124] Han, B.; Cheng, A.; Ji, G.; Wu, S.; Shen, J. J Appl Polym Sci 2004, 91,
2536-2542.
[125] Jae-Do, N.; Sung Duk, H.; Hyouk Ryeol, C.; Jong Hoon, L.; Kwang, J.
K.; Seok, H. Smart Mater Struct 2005, 14, 87.
[126] Lee, H.-T.; Lin, L.-H. Macromolecules 2006, 39, 6133-6141.
[127] Liff, S. M.; Kumar, N.; McKinley, G. H. Nat Mater 2007, 6, 76-83.
[128] da Silva, G. R.; da Silva-Cunha Jr, A.; Behar-Cohen, F.; Ayres, E.;
Oréfice, R. L. Polym Degrad Stab 2010, 95, 491-499.
[129] Gunes, I. S.; Cao, F.; Jana, S. C. J Polym Sci Part B: Polym Phys 2008,
46, 1437-1449.
[130] Da Silva, G. R.; Ayres, E.; Orefice, R. L.; Moura, S. A. L.; Cara, D. C.
e. a. J Drug Target 2009, 17, 374-383.
[131] Badamshina, E.; Estrin, Y.; Gafurova, M. J Mater Chem A 2013, 1,
6509-6529.
[132] Sen, R.; Zhao, B.; Perea, D.; Itkis, M. E.; Hu, H.; Love, J.; Bekyarova,
E.; Haddon, R. C. Nano Lett 2004, 4, 459-464.
[133] Kuan, H.-C.; Ma, C.-C. M.; Chang, W.-P.; Yuen, S.-M.; Wu, H.-H.;
Lee, T.-M. Compos Sci Technol 2005, 65, 1703-1710.
[134] Xu, M.; Zhang, T.; Gu, B.; Wu, J.; Chen, Q. Macromolecules 2006, 39,
3540-3545.
[135] Xiong, J.; Zheng, Z.; Qin, X.; Li, M.; Li, H.; Wang, X. Carbon 2006,
44, 2701-2707.
[136] Jung, Y. C.; Sahoo, N. G.; Cho, J. W. Macromol Rapid Commun 2006,
27, 126-131.
[137] Sen, R.; Zhao, B.; Perea, D.; Itkis, M. E.; Hu, H.; Love, J.; Bekyarova,
E.; Haddon, R. C. Nano Lett 2004, 4, 459.
[138] Koerner, H.; Liu, W.; Alexander, M.; Mirau, P.; Dowty, H.; Vaia, R. A.
Polymer 2005, 46, 4405-4420.
[139] Kwon, J.-Y.; Kim, H.-D. J Appl Polym Sci 2005, 96, 595-604.
[140] Xiong, J.; Zheng, Z.; Song, W.; Zhou, D.; Wang, X. Composites Part A
2008, 39, 904-910.
[141] Zhang, R.; Dowden, A.; Deng, H.; Baxendale, M.; Peijs, T. Compos Sci
Technol 2009, 69, 1499-1504.
[142] Tagmatarchis, N.; Prato, M. J Mater Chem 2004, 14, 437-439.
[143] Dieckmann, G. R.; Dalton, A. B.; Johnson, P. A.; Razal, J.; Chen, J.;
Giordano, G. M.; Muñoz, E.; Musselman, I. H.; Baughman, R. H.;
Draper, R. K. J Am Chem Soc 2003, 125, 1770-1777.
[144] Ortiz-Acevedo, A.; Xie, H.; Zorbas, V.; Sampson, W. M.; Dalton, A. B.;
Baughman, R. H.; Draper, R. K.; Musselman, I. H.; Dieckmann, G. R. J
Am Chem Soc 2005, 127, 9512-9517.
[145] Tasis, D.; Tagmatarchis, N.; Bianco, A.; Prato, M. Chem Rev 2006, 106,
1105-1136.
[146] Barisci, J. N.; Tahhan, M.; Wallace, G. G.; Badaire, S.; Vaugien, T.;
Maugey, M.; Poulin, P. Adv Funct Mater 2004, 14, 133-138.
[147] Chen, W.; Tao, X. Macromol Rapid Commun 2005, 26, 1763-1767.
[148] Koerner, H.; Price, G.; Pearce, N. A.; Alexander, M.; Vaia, R. A. Nat
Mater 2004, 3, 115-120.
[149] Chen, W.; Tao, X.; Liu, Y. Compos Scie Technol 2006, 66, 3029-3034.
[150] Barick, A. K.; Tripathy, D. K. Mater Sci Eng B 2011, 176, 1435-1447.
[151] Raja, M.; Ryu, S. H.; Shanmugharaj, A. M. Eur Polym J 2013, 49,
3492-3500.
[152] Lima, A. M. F.; de Castro, V. G.; Borges, R. S.; Silva, G. G. Polímeros
2011, 22, 117-124.
[153] Bourbigot, S.; Samyn, F.; Turf, T.; Duquesne, S. Polym Degrad Stab
2010, 95, 320-326.
[154] Musumeci, A. W.; Silva, G. G.; Liu, J.-W.; Martens, W. N.; Waclawik,
E. R. Polymer 2007, 48, 1667-1678.
[155] Du, F.; Scogna, R. C.; Zhou, W.; Brand, S.; Fischer, J. E.; Winey, K. I.
[156] Novoselov, K. S.; Geim, A. K.; Morozov, S. V.; Jiang, D.; Zhang, Y.;
Dubonos, S. V.; Grigorieva, I. V.; Firsov, A. A. Science 2004, 666-669,
[157] Bunch, J. S.; Verbridge, S. S.; Alden, J. S.; van der Zande, A. M.;
Parpia, J. M.; Craighead, H. G.; McEuen, P. L. Nano Lett 2008, 8, 2458-
2462.
[158] Reina, A.; Jia, X.; Ho, J.; Nezich, D.; Son, H.; Bulovic, V.; Dresselhaus,
M. S.; Kong, J. Nano Lett 2009, 9, 30-35.
[159] Berger, C.; Song, Z.; Li, X.; Wu, X.; Brown, N.; Naud, C.; Mayou, D.;
Li, T.; Hass, J.; Marchenkov, A. N.; Conrad, E. H.; First, P. N.; de Heer,
W. A. Science 2006, 312, 1191-1196.
[160] Xuekun, L.; Minfeng, Y.; Hui, H.; Rodney, S. R. Nanotechnology 1999,
10, 269.
[161] Stankovich, S.; Dikin, D. A.; Piner, R. D.; Kohlhaas, K. A.;
Kleinhammes, A.; Jia, Y.; Wu, Y.; Nguyen, S. T.; Ruoff, R. S. Carbon
2007, 45, 1558-1565.
[162] Wang, G.; Yang, J.; Park, J.; Gou, X.; Wang, B.; Liu, H.; Yao, J. J Phys
Chem C 2008, 112, 8192-8195.
[163] Si, Y.; Samulski, E. T. Nano Lett 2008, 8, 1679-1682.
[164] Yuan, N. Y.; Ma, F. F.; Fan, Y.; Liu, Y. B.; Ding, J. N. Composites Part
A 2012, 43, 2183-2188.
[165] Hu, L.; Jiang, P.; Zhang, P.; Bian, G.; Sheng, S.; Huang, M.; Bao, Y.;
Xia, J. J Mater Sci 2016, 51, 8296-8309.
[166] Stankovich, S.; Piner, R. D.; Nguyen, S. T.; Ruoff, R. S. Carbon 2006,
44, 3342-3347.
[167] Lee, Y. R.; Raghu, A. V.; Jeong, H. M.; Kim, B. K. Macromol Chem
Phys 2009, 210, 1247-1254.
[168] Wang, X.; Hu, Y.; Song, L.; Yang, H.; Xing, W.; Lu, H. J Mater Chem
2011, 21, 4222-4227.
[169] Lee, H.; Jeong, H. M. In Functionalized Graphene Sheet/Polyurethane
Nanocomposites; Mikhailov S.; Ed.; Physics and Applications of
Graphene - Experiments; InTech: Vienna, Austria, 2011.
[170] Kuilla, T.; Bhadra, S.; Yao, D.; Kim, N. H.; Bose, S.; Lee, J. H. Prog
Polym Sci 2010, 35, 1350-1375.
[171] Li, X.; Deng, H.; Li, Z.; Xiu, H.; Qi, X.; Zhang, Q.; Wang, K.; Chen, F.;
Fu, Q. Composites Part A 2015, 68, 264-275.
[172] Kim, H.; Miura, Y.; Macosko, C. W. Chem Mater 2010, 22, 3441-3450.
[173] Fahim, I. S.; Mamdouh, W.; Salem, H. A. G. Am J Nanosci
Nanotechnol 2013, 1, 31-40.
[174] Xiang, C.; Cox, P. J.; Kukovecz, A.; Genorio, B.; Hashim, D. P.; Yan,
Z.; Peng, Z.; Hwang, C. C.; Ruan, G.; Samuel, E. L. G.; Sudeep, P. M.;
Konya, Z.; Vajtai, R.; Ajayan, P. M.; Tour, J. M. ACS Nano 2013, 7,
10380-10386.
[175] Liang, J.; Xu, Y.; Huang, Y.; Zhang, L.; Wang, Y.; Ma, Y.; Li, F.; Guo,
T.; Chen, Y. J Phys Chem C 2009, 113, 9921-9927.
[176] RamanathanT; Abdala, A. A.; StankovichS; Dikin, D. A.; Herrera
Alonso, M.; Piner, R. D.; Adamson, D. H.; Schniepp, H. C.; ChenX;
Ruoff, R. S.; Nguyen, S. T.; Aksay, I. A.; Prud'Homme, R. K.; Brinson,
L. C. Nat Nano 2008, 3, 327-331.
[177] Strankowski, M.; Piszczyk, Ł.; Kosmela, P.; Korzeniewski, P. Pol J
Chem Technol 2015, 17, 88-94.
[178] Xia, H.; Song, M. Soft Matter 2005, 1, 386-394.
[179] Chen, X.; Ma, C.; Jiao, C. Polym Degrad Stab 2016, 129, 275-285.
[180] Xu, W.; Liu, L.; Zhang, B.; Hu, Y.; Xu, B. Ind Eng Chem Res 2016, 55,
4930-4941.
[181] Gavgani, J. N.; Adelnia, H.; Gudarzi, M. M. J Mater Sci 2014, 49, 243-
254.
[182] Hu, J.; Zhang, F. J Therm Anal Calorim 2014, 118, 1561-1568.
[183] Roy, S.; Srivastava, S. K.; Mittal, V. J Appl Polym Sci 2016, 133, doi:
10.1002/app.43470.
[184] Zhang, Y.; Zhang, Y.; Liu, Y.; Wang, X.; Yang, B. Appl Surf Sci 2016,
382, 144-154.
[185] Ma, H.; Zeng, J.; Realff, M. L.; Kumar, S.; Schiraldi, D. A. Compos Sci
Technol 2003, 63, 1617-1628.
[186] Wu, G.; Ma, L.; Liu, L.; Wang, Y.; Xie, F.; Zhong, Z.; Zhao, M.; Jiang,
B.; Huang, Y. Mater Des 2016, 89, 1343-1349.
[187] Jiang, S.; Li, Q.; Wang, J.; He, Z.; Zhao, Y.; Kang, M. Composites Part
A 2016, 87, 1-9.
[188] Špírková, M.; Duszová, A.; Poręba, R.; Kredatusová, J.; Bureš, R.;
Fáberová, M.; Šlouf, M. Composites Part B 2014, 67, 434-440.
[189] Al-Saleh, M. H.; Sundararaj, U. Carbon 2009, 47, 2-22.
[190] Tibbetts, G. G.; Lake, M. L.; Strong, K. L.; Rice, B. P. Compos Sci
Technol 2007, 67, 1709-1718.
[191] Jiyun, C. H. Smart Mater Struct 2003, 12, 264.
[192] Sosa, I. O.; Noguez, C.; Barrera, R. G. J Phys Chem B 2003, 107, 6269-
6275.
[193] Zhang, D.; Qi, L.; Yang, J.; Ma, J.; Cheng, H.; Huang, L. Chem Mater
2004, 16, 872-876.
[194] Jana, N. R.; Gearheart, L.; Murphy, C. J. Chem Commun 2001, 617-618.
[195] Machulek Junior, A.; Paulo Moises de Oliveira, H.; Henrique Gehlen,
M. Photochem Photobiol Sci 2003, 2, 921-925.
[196] Callegari, A.; Tonti, D.; Chergui, M. Nano Lett 2003, 3, 1565-1568.
[197] Chen, S.; Fan, Z.; Carroll, D. L. J Phys Chem B 2002, 106, 10777-
10781.
[198] Angelescu, D. G.; Vasilescu, M.; Somoghi, R.; Donescu, D.;
Teodorescu, V. S. Colloids Surf A 2010, 366, 155-162.
[199] Wang, H.; Qiao, X.; Chen, J.; Ding, S. Colloids Surf A 2005, 256, 111-
115.
[200] Taleb, A.; Petit, C.; Pileni, M. P. Chem Mater 1997, 9, 950-959.
[201] Balan, L.; Malval, J.-P.; Schneider, R.; Burget, D. Mater Chem Phys
2007, 104, 417-421.
[202] Chen, Z.; Gao, L. Mater Res Bull 2007, 42, 1657-1661.
[203] Torres, V.; Popa, M.; Crespo, D.; Calderón Moreno, J. M.
Microelectron Eng 2007, 84, 1665-1668.
[204] Raveendran, P.; Fu, J.; Wallen, S. L. J Am Chem Soc 2003, 125, 13940-
13941.
[205] Andersson, M.; Pedersen, J. S.; Palmqvist, A. E. C. Langmuir 2005, 21,
11387-11396.
[206] Shah, M. S. A. S.; Nag, M.; Kalagara, T.; Singh, S.; Manorama, S. V.
Chem Mater 2008, 20, 2455-2460.
[207] Juan, L.; Zhimin, Z.; Anchun, M.; Lei, L.; Jingchao, Z. Int J
Nanomedicine 2010, 5, 261-267.
[208] Caseri, W. Macromol Rapid Commun 2000, 21, 705-722.
[209] Upadhyay, J.; Kumar, A. Compos Sci Technol 2014, 97, 55-62.
[210] Wu, H.; Chiang, S.; Han, W.; Tang, Y.; Kang, F.; Yang, C. Compos Sci
Technol 2014, 99, 109-116.
[211] Khan, A.; Asiri, A. M.; Rub, M. A.; Azum, N.; Khan, A. A. P.; Khan, S.
B.; Rahman, M. M.; Khan, I. Composites Part B 2013, 45, 1486-1492.
[212] Rozenberg, B. A.; Tenne, R. Prog Polym Sci 2008, 33, 40-112.
[213] Karak, N.; Konwarh, R.; Voit, B. Macromol Mater Eng 2010, 295, 159-
169.
[214] Han, W.-S. Polym Compos 2016, doi: 10.1002/pc.24156.
[215] Macocinschi, D.; Filip, D.; Zaltariov, M. F.; Varganici, C. D. Polym
Degrad Stab 2015, 121, 238-246.
[216] Kim, J. H.; Min, B. R.; Won, J.; Kang, Y. S. J Polym Sci Part B: Polym
Phys 2006, 44, 1168-1178.
[217] Macocinschi, D.; Filip, D.; Vlad, S.; Butnaru, M.; Knieling, L. Int J Biol
Macromol 2013, 52, 32-37.
[218] Triebel, C.; Vasylyev, S.; Damm, C.; Stara, H.; Ozpnar, C.; Hausmann,
S.; Peukert, W.; Munstedt, H. J Mater Chem 2011, 21, 4377-4383.
[219] Paladini, F.; Pollini, M.; Talà, A.; Alifano, P.; Sannino, A. J Mater Sci:
Mater Med 2012, 23, 1983-1990.
[220] Kim, J. S.; Kuk, E.; Yu, K. N.; Kim, J.-H.; Park, S. J.; Lee, H. J.; Kim,
S. H.; Park, Y. K.; Park, Y. H.; Hwang, C.-Y.; Kim, Y.-K.; Lee, Y.-S.;
Jeong, D. H.; Cho, M.-H. Nanomed: Nanotechnol 2007, 3, 95-101.
[221] Siddhartha, S.; Tanmay, B.; Arnab, R.; Gajendra, S.; Ramachandrarao,
P.; Debabrata, D. Nanotechnology 2007, 18, 225103.
[222] Ma, R.; Kang, B.; Cho, S.; Choi, M.; Baik, S. ACS Nano 2015, 9,
10876-10886.
[223] Ma, R.; Lee, J.; Choi, D.; Moon, H.; Baik, S. Nano Lett 2014, 14, 1944-
1951.
[224] Wang, Z.; Zhou, Y.; Yao, Q.; Sun, Y. Appl Surf Sci 2009, 256, 1404-
1408.
[225] Gun'ko, V. M.; Mironyuk, I. F.; Zarko, V. I.; Voronin, E. F.; Turov, V.
V.; Pakhlov, E. M.; Goncharuk, E. V.; Nychiporuk, Y. M.; Vlasova, N.
N.; Gorbik, P. P.; Mishchuk, O. A.; Chuiko, A. A.; Kulik, T. V.;
Palyanytsya, B. B.; Pakhovchishin, S. V.; Skubiszewska-Zięba, J.;
Janusz, W.; Turov, A. V.; Leboda, R. J Colloid Interface Sci 2005, 289,
427-445.
[226] Maciá-Agulló, T. G.; Fernández-García, J. C.; Pastor-sempere, N.;
Orgilés-Barceló, A. C.; Martín-Martínez, J. M. JAdhes 1992, 38, 31-53.
[227] Vega-Baudrit, J.; Sibaja-Ballestero, M.; Vázquez, P.; Torregrosa-Maciá,
R.; Miguel Martín-Martínez, J. Int J Adhes Adhes 2007, 27, 469-479.
[228] Torró-Palau, A. M.; Fernández-Garcı́a, J. C.; César Orgilés-Barceló, A.;
Martıń -Martın
́ ez, J. M. I J Adhes Adhes 2001, 21, 1-9.
[229] Jaúregui-Beloqui, B.; Fernández-Garcıá , J. C.; César Orgilés-Barceló,
A.; Mar Mahiques-Bujanda, M.; Martı́n-Martı́nez, J. M. I J Adhes Adhes
1999, 19, 321-328.
[230] Petrović, Z. S.; Javni, I.; Waddon, A.; Bánhegyi, G. J Appl Polym Sci
2000, 76, 133-151.
[231] Tien, Y. I.; Wei, K. H. Polymer 2001, 42, 3213-3221.
[232] Vega-Baudrit, J.; Navarro-Bañón, V.; Vázquez, P.; Martín-Martínez, J.
M. Int J Adhes Adhes 2006, 26, 378-387.
[233] Nunes, R. C. R.; Fonseca, J. L. C.; Pereira, M. R. Polym Test 2000, 19,
93-103.
[234] Nunes, R. C. R.; Pereira, R. A.; Fonseca, J. L. C.; Pereira, M. R. Polym
Test 2001, 20, 707-712.
[235] Stribeck, N.; Zeinolebadi, A.; Harpen, F.; Luinstra, G.; Eling, B.; Botta,
S. Macromolecules 2013, 46, 4041-4052.
[236] Chen, X.; Wu, L.; Zhou, S.; You, B. Polym Int 2003, 52, 993-998.
[237] Lee, S.-I.; Hahn, Y. B.; Nahm, K. S.; Lee, Y.-S. Polym Adv Technol
2005, 16, 328-331.
[238] Malay, O.; Oguz, O.; Kosak, C.; Yilgor, E.; Yilgor, I.; Menceloglu, Y.
Z. Polymer 2013, 54, 5310-5320.
[239] Chiacchiarelli, L. M.; Puri, I.; Puglia, D.; Kenny, J. M.; Torre, L.
Colloid Polym Sci 2013, 291, 2745-2753.
[240] Peruzzo, P. J.; Anbinder, P. S.; Pardini, F. M.; Pardini, O. R.; Plivelic,
T. S.; Amalvy, J. I. Mater Today Commun 2016, 6, 81-91.
[241] Lai, S. M.; Liu, S. D. Polym Eng Sci 2007, 47, 77-86.
[242] Chen, G.; Zhou, S.; Gu, G.; Yang, H.; Wu, L. J Colloid Interface Sci
2005, 281, 339-350.
[243] Chen, G.; Zhou, S.; Gu, G.; Wu, L. Colloids Surf A 2007, 296, 29-36.
[244] Sadeghi, M.; Semsarzadeh, M. A.; Barikani, M.; Ghalei, B. J Membr Sci
2011, 385–386, 76-85.
[245] Khosravi, A.; Sadeghi, M.; Banadkohi, H. Z.; Talakesh, M. M. Ind Eng
Chem Res 2014, 53, 2011-2021.
[246] Sadeghi, M.; Semsarzadeh, M. A.; Barikani, M.; Pourafshari Chenar, M.
J Membr Sci 2011, 376, 188-195.
Chapter 2
INNOVATIVE THERMOPLASTIC
PET POWDER COATING IN
TELECOMMUNICATIONS FIELD
Yukitoshi Takeshita, Takashi Miwa, Azusa Ishii

and Takashi Sawada
NTT Device Innovation Center, NTT Corporation,
Atsugi-shi, Kanagawa, Japan
ABSTRACT
Polymeric materials such as those used for coatings offer highly
effective protection against corrosion of outdoor steel structures in the
industrial telecommunications fields. Conventionally, tar epoxy resin has
been used as a coating for steel telephone poles for a long time in Japan.
However, degradation of the resin and steel itself become noticeable 20 to
30 years after construction. We need to change the material to meet
requirements from both the eco-friendliness and durability standpoints.
Thermoplastic polyethylene terephthalate (PET) provides a good coating
with high toughness, good flexibility, and a large thickness without fine
pinholes. In addition, thermoplastic PET is an eco-friendly material
because PET recycling technology has been well established. We here
describe the challenges we faced in applying thermoplastic PET powder
coating to steel structures in the telecommunications field, and present

Corresponding Author: [email protected].
62 Yukitoshi Takeshita, Takashi Miwa, Azusa Ishii et al.
research results on treatment in polymer crystallinity and composites for

higher durability against alkalinity as well as against any harsh
environment. PET is a semi-crystalline polymer in which the crystallinity
is governed by crystallization kinetics represented by, for instance, the
Vogel Fulcher and Abrami’s equation. It is therefore possible to optimize
coating properties by controlling the crystallinity, which can be further
controlled by the air-cooling time in the industrial fluidized powder
vessel process. Thermoplastic PET polymer is generally vulnerable to
hydrolysis, especially under alkaline conditions, because hydroxyl ions
attack the ester group in PET, causing it to decompose. To enhance the
durability of thermoplastic PET powder coating even in harsh outdoor
environments such as alkalinity, thermoplastic PET is mixed with a
secondary resin in which primary PET is on the bottom layer and the
secondary resin is on the surface layer so as to protect the primary PET.
Polyvinyl butyral (PVB) material has been selected as a secondary resin.
The resulting PET/PVB composite coating presents a relatively good
mechanical stress-strain property as a coating and an enhanced alkaline
resistance, with PET presenting a good mechanical property and PVB
suppressing the hydrolysis of PET. The composite also exhibits good
performance in terms of sea salt water resistance, UV resistance, and heat
cycle resistance in the environment. Innovated thermoplastic PET, with
the aid of a secondary material, is essential to the function and integrity of
steel structures in industrial telecommunications fields in terms of both
eco-friendliness and long-term durability.
Keywords: thermoplastic, PET, powder coating, semi-crystalline,

crystallinity, alkalinity, sea salt, heat cycle, UV-light, resistance, color
difference, specular gloss
INTRODUCTION
We provide telecommunications services throughout Japan, and this
means that we have an enormous number of telecommunication facilities, for
instance, approximately twenty thousand wireless steel towers, five million
steel telephone poles, and two million kilometres of cables [1]. And they are
exposed to a wide range of environments: UV light, a range of temperatures,
and humidity, sea salt particles, as well as alkalinity and acidity. In this
situation, polymeric materials such as those used for coatings offer highly
effective protection and have been developed and introduced in the industrial
telecommunications fields.
Innovative Thermoplastic PET Powder Coating … 63
Solvent–based coatings have been applied to wireless steel towers and

cable pipes. Conventional organic coatings usually include a wide variety of
volatile organic compounds (VOCs), such as hydrocarbons (hexane, toluene,
and xylene), ketones (acetone, methyl ethyl ketone, and methyl isobutyl
ketone), alcohols (methanol, ethanol, and cyclohexanol), and esters (ethyl
acetate, butylacetate, and isobutyl acetate) [2]. However, since VOCs are
environmentally damaging and pose human health risks, recent increased
environmental awareness has led to demands to reduce their use [3].
One solution is to reduce the proportion of organic solvent in the coating
ultimately to zero. Therefore, low-solvent and water-based coatings have been
in high demand, and they have been studied with a view to protecting human
health and the environment. Another approach involves using powder
coatings, which contain no solvent whatsoever. Such coatings represent the
final destination along the road to VOC reduction.
As for steel telephone poles, tar epoxy resin has been used as a coating for
a long time in Japan. However, degradation of the resin and the steel itself
becomes noticeable 20 to 30 years after construction. Therefore, we need to
change the material to meet requirements from both the eco-friendliness and
durability standpoints.
Among the polymeric materials, thermoplastic polyethylene terephthalate
(PET) provides us with a relative high toughness, flexibility, and a large
thickness without fine pinholes [4, 5]. Besides, a recycling technology has
been well established for thermoplastic PET [6]. Further, if the powder coating
is used, there are no VOCs, ideally. This provides many benefits in industrial
fields. For these reasons, thermoplastic PET powder coating has been selected
as a substitute for tar epoxy resin.
This PET is generally a semi-crystalline polymer, which is partly
crystallized during thermal treatment. The presence of the crystalline has a
significant impact on the properties of a polymer in the field. The thermal
condition during the coating process significantly affects the degree of
crystallinity. Therefore, the relationship between the heating condition and
degree of crystallinity and thus film performance needs to be understood.
Further, the bottom part of a steel pole is generally exposed to more sever
conditions than the rest of the pole. This is because the bottom part comes into
direct contact with the ground, mortar, and concrete and therefore has the
potential to be adversely affected by alkaline conditions. For this reason,
treatments against alkaline environments are also required.
In this chapter, we present the challenges we faced in applying this
thermoplastic PET powder coating to steel structures and share our research on
treatments in polymer crystallinity and composites for higher durability

against alkalinity as well as against any harsh environmental factors.
THEORY
Crystallization Kinetics
The overall crystallization rate is determined by two main factors: one is

the rate of formation of nuclei and the other is the rate of growth of such nuclei
to the final crystalline aggregate. Polymer crystallization behaviour is shown
in Figure 1 [7, 8, 9].
Crystal growth rate
Mobility ΔT(T-Tm)
Tg Temperature Tm
Figure 1. Crystal growth rate as a function of temperature.
Most polymers crystallize at a temperature between Tg and Tm,

specifically, from about 30°C above Tg to about 10°C below Tm [8]. During
Figure
the1.
cooling process, the crystal growth rate increases and then decreases with
decreasing temperature. In principal, the driving force of crystallization is ΔT,
the difference from Tm. Hence, the decrease in temperature provides high ΔT
and thus a high growth rate (black arrow in Figure 1). However, after the
growth rate reaches its maximum point, macromolecule mobility decreases
with decreasing temperature, which causes a low crystal growth rate (white
arrow in Figure 1); that is, the equation for the crystal growth rate consists of
the two parts: the macromolecule movement and the crystallization
mechanism. The macromolecule mobility is expressed in the following Vogel

Fulcher equation:
𝑇
  exp (− 𝑇−𝑇
𝐴
) (1)
𝑉
where 𝑇𝐴 is activation temperature, 𝑇𝑉 is the Voglel temperature. Crystal

growth rate G is expressed as
𝐺 𝐾

 exp (− 𝑇∆𝑇) (2)
where K is the crystal growth rate constant, which contains nucleation and
growth parameters. Abrami’s equation also describes the crystal growth rate in
the field, in which temperature and stress change:
𝑋
𝑋∞
= 1 − exp{−(𝐾𝑡)𝑛 } (3)
where 𝑋 is crystallinity, 𝑋∞ is the maximum crystallinity reached after elapse

of infinite time. The 𝑛 is the Abrami index number, an integer, whose value
depends on the mechanism of nucleation and on the form of crystal growth.
The numerical value of 𝐾 is directly connected with the overall rate of
−1
crystallization 𝑡1/2 with reciprocal of time (s-1). The theoretical values of K
and n are determined depending on the type of nucleation, such as
predetermined or spontaneous, and the form of growth, such as spheres (n =
3), platelets (n = 3), or rodlets (n = 1) [8]. Time dependences of crystal growth
rate in terms of various 𝐾 and 𝑛 are shown in Figure 2.
When K is constant, if n increases, the relative degree of crystallinity
increases as shown in Figure 2 (a). Whereas, when n is constant, if K
increases, it increases as shown in Figure 2 (b). In Equation 3 holds if we
ignore the restriction that new crystal cannot be generated in the area where
the crystal has already grown. Therefore, taking into account the system in
which crystal growth rate always changes such as during the cooling process,
Abrami’s equation is extended to the following:
𝑋 𝑡 𝑛
𝑋∞
= 1 − exp [− {∫0 𝐾(𝜏)𝑑𝜏} ] (4)
where 𝐾(𝜏) represents a crystal growth rate constant that varies with the time.
Relative degree of crystallinity
Relative degree of crystallinity

(a) Time (s) (b) Time (s)
Figure 2. Abrami’s equation showing crystallization behavior: (a) Abrami’s index

number (n)2.dependency; (b) crystal growth rate constant (K) dependency.
Figure
EXPERIMENTAL
Coating Material
Common film-former systems for powder coating are classified into

thermoplastic and thermoset ones. The thermoplastic systems are for instance
polyethylene (LDPE, LLDPE, HDPE), polyamide, polyester and hybrid
epoxy/polyester types. The thermoset systems involve, for instance, the
combination of resin and hardener of epoxy resin and phenolic hardeners,
epoxy resin and modified dicyandiamide, COOH polyester resin and epoxy
resin, and OH polyester resin and isocyanate adduct [10]. Of these materials,
thermoplastic polyester coating film provides a relatively large thickness
without pinholes and with good weathering performance, a good appearance,
superior mechanical properties, and low cost [4,5]. Further, the recycling
technologies of thermoplastic polyesters are already established [6], which
provide many benefits in the industrial fields that meet the needs of today’s
society, as described in the introduction. This is why the thermoplastic
polyester (PET) was selected as a primary powder coating.
Thermoplastic PET powder coating (SAPOE 5000, NTT-AT, Tokyo,
Japan) was used in this study. It has a molecular weight of around 50000, a
melting temperature (Tm) of 230-240°C, a glass transition temperature (Tg) of
approximately 72°C, and an average particle size of around 90 µm.
To improve the alkaline resistance, a primary coating resin was mixed
with a secondary resin that has a higher alkaline resistance than the primary
PET resin. Furthermore, to form a dual phase structure by utilizing the
difference in Tm in addition to the different compatibility during the coating
process in a fluidized powder vessel, a secondary resin was chosen that had
both higher alkaline resistance and lower Tm than the primary PET resin. Due
to film formability, we finally decided to use polyvinyl butyral (PVB) as a
secondary resin.
PVB is a polyvinyl butyral powder resin (Mowital B30H, Kuraray, Tokyo,
Japan), which has a molecular weight of around 33000, a Tm of 140-170°C, a
Tg of approximately 68°C, and an average particle size of around 80 µm. The
samples were mixed manually for 5 min using a poly bag (Unipack J-4, 340 x
240 x 0.04 mm). Some of combinations of PET/PVB were prepared. We refer
to the mixed powder for instance as PET/PVB10 in which a hundred parts of
PET and a ten parts of PVB powder coatings were mixed.
Powder Coating Process
The base substrate was an SS400 steel plate (50 mm × 100 mm × 3.2 mm)
that was galvanized and sweep blasted. The coating process is shown
schematically in Figure 3. We first heated the test pieces at around 3200C (A
in Figure 3) and then dipped them into a fluidized polyester powder vessel for
5 s (B in Figure 3). The test pieces were subsequently removed and kept in
static air (200C) for 2–30 min in a vertical position to ensure that the powder
coating materials on the surface had completely melted (C in Figure 3).
Finally, they were immersed in water to cool (D in Figure 3), and test
specimens were obtained. Here, we refer to this cooling time in air as the “air-
cooling time” (C in Figure 3). Obviously, the heat capacity could affect the
formation of the film. However, since we used SS400 steel with a uniform
thickness in this study, we ignored the effect of heat capacity. The surface
temperature of the test specimens was measured with a thermometer
(TEMPERATURE HiTESTER 3442, HIOKI, Nagano, Japan and IR-TA,
CHINO, Tokyo, Japan).
Characterization
Crystallinity
Semi-crystalline polymers can be characterized by various techniques
such as X‐ray diffraction (XRD), optical microscopy, and differential scanning
calorimetry (DSC). DSC remains the preferred method since it facilitates the
complete characterization of the various thermal transitions in these materials.
Therefore, differential scanning calorimeters (DSC 3100, MAC Science,

Tokyo, Japan and DSC 6200, Seiko Instruments, Chiba, Japan) were used here
to determine the crystallinity. The samples were cooled by immersing them in
a liquid N2 bath, and the coating film was peeled from the test specimen with a
cutter. As shown in Figure 4, an exothermic peak caused by cold
crystallization and an endothermic peak caused by melting were typically
observed in our measurement. Figure 4 shows a typical chart in which the air-
cooing time is 6 min. Todoki suggested that the crystallinity is obtained with
the following equation [11]:
Δ𝐻𝑚 −Δ𝐻𝑐
𝑋𝑐 = ( ) × 100 (5)
𝐴
where Xc is the crystallinity (%), ΔHm is the melting heat (J/g), ΔHc is the
crystallization heat (J/g), and A is the melting heat (118 J/g) of 100%
crystallized PET [12]. For each air-cooling time, the crystallinity was
determined by Eq. (5).
A B
Thermoplastic
polyester powder
SS400 Steel Plate 320℃ Air

＋
Hot dip galvanizing, Fluidized powder vessel
then blasted.
Air cooling Water bath Completion

2 − 30 min
C D
Figure
Figure 3. 3.Schematic diagram of powder coating process.
Identification
Micro FTIR analysis is available for identification of domains in film.
Pieces of coating film were peeled from the steel plates, which had been
coated with wax to facilitate peeling. They were cut to form a surface area and
a cross-sectional area. Each piece was observed with a digital optical
microscope (VHX-900, Keyence, Tokyo, Japan), and the exact structure was
analyzed with a micro FTIR (Continuum Nexus 470, Thermo Fisher Scientific,
MA). This micro FTIR allowed us to obtain point analyses and profiles with a
spatial resolution as high as 10 µm. Here, the differently colored parts
observed with the microscope were found to be around 100–300 µm in size, so
we performed the analysis with a spatial resolution of 50–80 µm. Pretreatment
was used to form a KBr tablet for the cross-sectional area. The cross-sectional
area was analyzed in the transmission mode and the surface area was analyzed
in the reflection mode. There were 32 and 96 scans in the transmission and
reflection modes, respectively.
0
Exotherm
-1 Glass transition Cold crystallization

Heat flow (mW)
-2
Endotherm
-3
Melting
-4
-5
50 100 150 200 250 300
Temperature (oC)
Figure 4. Typical DSC trace of thermoplastic PET powder coating film. Air-cooling
time: 6 min. Heating rate: 10°C/min.
Figure 4.
Measurement
Specular Gloss and Color Difference

The color difference was measured with a color tester (SM-7-IS-2B, Suga
Test Instruments, Tokyo, Japan) and 60° specular gloss was measured with a
glossmeter (UGV-5, Suga Test Instruments, Tokyo, Japan).
Surface Roughness
For the roughness measurements, we employed a digital laser microscope
(Controller: VK-8500, Microscope: VK-8510, KEYENCE, Tokyo, Japan) in
accordance with Japan Industrial Standard (JIS) B 0601 (1994) [13]. With the
microscope, we detected the top and bottom of the surface by focusing on
them and measured the height by 0.10-µm pitches. We then averaged the
results for an area of 1479 µm × 1109 µm. We performed at least three
measurements, and obtained the average roughness.
Adhesive Strength
An adhesion test was performed based on a JIS K 5600-5-7 (ISO4624)
[14] by using an Elcometer Pull off Adhesion Tester (F106-3, Elcometer Inc.,
Michigan, USA). This adhesion tester was attached to a dolly, which was
adhered to the sample with epoxy adhesive (Araldite Rapid, Showa Denko,
Tokyo, Japan). A lifting force was applied to the dolly by manually turning a
control knob. As the applied pressure increased, a gauge on the side of the
tester displayed the current pressure in N/mm2 (MPa); the gauge on this
particular device ranged from 0 to 7 N/mm2. The dolly used in conjunction
with this device has a 20-mm diameter (314 mm2), and it is produced by
Elcometer along with the device. At the moment the dolly becomes separated
from the testing surface, the gauge is read and the number recorded. After the
number is recorded, the failure plane is observed, and the percentage of the
area of each layer or interface at which the sample broke is recorded. We call
this recorded number the adhesive strength.
Impact Resistance
With reference to JIS K 5600-5-3 [15], we used a Du Pont-type falling
impact tester (Yasuda, Tokyo, Japan) for coated plates to measure the impact
resistance. A hemispherical 300-g steel weight 1 in. in diameter was dropped
onto the specimen surface. The dropping height was increased in 10-cm steps
to the height at which the base metal was exposed by the destruction of the
coating layer.
Artificial Accelerated Test
Salt Water Spray Test

Salt water spray resistance was investigated in accordance with JISK5600-
7-1 [16]. A salt water spray instrument (STP-90V, Suga Test Instruments,
Tokyo, Japan) was used in this study. Two test pieces were prepared under
identical conditions, one with a cross-cut (n = 2). In this salt water spay
testing, each test piece was observed to determine whether or not blisters,
cracks, or corrosion occurred. The test period was 3500 h.
Alkaline Resistance Test

To determine the alkaline resistance, the test pieces were immersed in
10 wt% NaOH aqueous solution at 40°C. We observed the film surface
appearance and measured the mass over seven days. Finally, we measured the
adhesive strength as described above.
Accelerated UV Test
The UV resistance was examined in accordance with JISK5600-7-7 [17].
An xenon arc instrument (Ci4000 Xenon Weather-Ometer, Atlas, Toyo Seiki,
Tokyo, Japan) was employed. The test period was 3000 h. The pseudo solar
radiation energy was 0.55 mW/mm2, the black panel temperature was
(63 ± 2)°C, the chamber temperature was (50 ± 10)°C, the chamber relative
humidity was 50%, and the water spraying time was 18 min/2 h. There were
two test pieces for each test period (n = 2). After each test period, the color
difference and the specular gloss of the coating film were measured as
described above.
Heat Cycle Test

Heat cycle resistance was tested using a constant temperature and
humidity cryogenic unit (PL-3KPH, Espec, Osaka, Japan) to provide the
accelerating environment. The temperature range was ‒30 to +70°C, the
relative humidity was 0–90%, and one cycle took 12 h. There were a total of
200 cycles. As with the salt water spray, after testing, the sample was checked
for blisters, cracks, or corrosion.
Outdoor Exposure Test
The resistance to actual sunshine and an outdoor atmosphere was

investigated by outdoor exposure according to JISK5600-7- 6 [18]. There were
three test pieces for each condition (n = 3). The color difference and specular
gloss were mainly tested for approximately 5.5 years in the metropolitan area
of Shinkiba in Tokyo to determine the influence of sunshine. The corrosion
resistance was examined for around five years in a coastal region of Miyake
Island, south of Tokyo, to determine the resistance against seawater particles.
CRYSTALLINITY OF THERMOPLASTIC PET

Cooling Time and Crystallinity
First, to determine the relationship between the cooling time during the
coating process and the obtained crystallinity, the crystallinity was measured
twice for each air-cooling time based on Eq. (5) [19]. The effect of the air-
cooling time on the degree of crystallinity is shown in Figure 5. When the air-
cooling times were 2, 3, 4, 5, 6, and 30 min, the average crystallinities were
0.7, 1.7, 3,7 5.6, 7.0, and 7.6%, respectively. Figure 5 shows that crystallinity
increased almost linearly within 2 to 6 min. When the air-cooling time was 30
min, crystallinity was not very different from that at 6 min.
9
8
7
Crystallinity (%)
6
5
4
3
2
1
0
0 2 4 6 8 28
10 30 14
12 32
Air-cooling time (min)
Figure 5.
Figure 5. Effect of air-cooling time on the degree of crystallinity [19].
Figure 6 shows a typical curve of the surface temperature of a test

specimen as it was being cooled with 5 min air-cooling time. The curve
indicates that coating times of 1, 2, 3, 4, and 5 min exhibit temperatures of
217, 188, 161, 136, and 1200C, respectively. In addition, Tm and Tg were
230–240 and approximately 720C, respectively (Tm and Tg in Figure 6).
These two results suggest that the specimen surface temperature lies between
Tm and Tg at 1–5 min. We already discussed crystallization in the temperature
region between Tm and Tg in the section on theory. Therefore, this condition

appears to allow the micro-Brownian movement of the macromolecular chains
and might eventually induce crystallization within 5 min in the case of 5-min
air-cooling time. It is obvious the temperature after 6 min is below Tg in
Figure 6, where the crystallization ceases, which might show that there is no
great difference between the results at 6 and 30 min. These results are
governed by crystallization kinetics described in the theory section. In the
engineering aspect, this finding suggests that it is possible to manipulate the
degree of crystallization by controlling the air-cooling time.
The coating weight and the heat transfer from the substrate may be
important parameters that could affect the total crystallization behaviour.
However, the coating weights were almost the same for all samples, so we
assumed that these conditions would be the same in this study.
400
350
In fluidized powder vessel
300
Temperature (⁰C)
250
In air-cooling
200
150
Tm (230 to 240°C)
In water bath
100
Tg (72°C)
50
Cooling time
0
0 1 2 3 4 5 6 7
Time (min)
FigureFigure
6. 6. Typical curve showing surface temperature of a test specimen as it is being
cooled with 5 min air-cooling time [19].
Crystallinity and Specular Gloss
As we found that the degree of crystallinity has an almost linear

relationship to the air-cooling time, we next examined the effect of
crystallinity on 60° specular gloss [19]. The relationship between crystallinity
and 60° specular gloss is shown on the left of Figure 7. The specular gloss
decreased slightly until the crystallinity reached around 3%. Once the
crystallinity was around 4%, it decreased greatly to around 80% and then
maintained an almost constant value. We speculate that when the crystallinity
exceeds 4%, the crystalline part has an influence on the reflection of light. As
the specular gloss changed depending on the crystallinity, we focused on the
surface structure of the film. We observed the surface roughness in the way
already described above.
Crystallinity Images Roughness

(air-cooling time)
100
0.7% 0.04 µm
80 (2 min) A
60o Specular gloss (%)
A B
C
1.7%
60 D E (3 min) 0.14 µm
B
40 3.7%
(4 min) 0.21 µm
20 C
5.6%
(5 min) 0.17 µm
0
D
0 2 4 6 8 10
Crystallinity (%) 7.0%
(6 min) 0.90 µm
E
Figure 7. Relationship between crystallinity and specular gloss, along with the surface
images and the roughness [19].
FigureThe
7. right side of Figure 7 shows the crystallinity, microscope images of
the surface, and the roughness. Although it is hard to see, the microscope
images appeared progressively darker as the crystallinity increased. When the
crystallinity increases, the roughness also increases: 0.04 → 0.14 → 0.21 →
0.17 → 0.90 µm. It is considered that the changes in surface roughness affect
the specular gloss. In the industrial aspect, this finding indicates that it is
possible to form a smooth and uniform coating film surface by suppressing
crystallization.
Crystallinity and Adhesive Strength
What we should know next is the mechanical property of the coating. We

examined the adhesive strength in adhesion test in relation to the degree of
crystallinity [19]. Figure 8 shows the failure plane and the crystallinity of the
film, where (a) is an image showing the broken part in the failure plane and (b)
shows the corresponding breaking location (%) vs. crystallinity together with
the air-cooling time. In Figure 8(a), we certainly observe that the failure occurs
in some cases where the interfaces are between the film and Zn substrate, Fe
and Zn phases, and film and dolly. The numerical values in the squares are the
adhesive strengths in Figure 8(b). Figure 8(b) shows that when the crystallinity
was 0.7%, around 21% of the breakage locations were at the interface between
the steel (Fe) and the zinc plate (Zn), around 55% between the zinc plate (Zn)
and the coating film (Film), and around 20% between the coating film (Film)
and the dolly (Dolly). Moreover, when the crystallinity exceeded 1%, the ratio
of breakages at the interface between the zinc (Zn) and the coating film (Film)
increased to around 80%. Overall, the proportion of interface breakage
(Film/Zn) increased as a function of crystallinity until the crystallinity was
3.7%, which simply results in a reduction in adhesive strength to 5.1 MPa.
Zhang et al. also suggested that the crystalline part exhibited greater hardness
than the amorphous part and thus weak adherence to the substrate [20]. We
consider that as the crystallinity increases, the film becomes brittle and fragile
and the adhesive strength is reduced. One other possibility is that the increased
internal stress caused by high crystallinity [20] reduces the adhesion at the
interface between the film and substrate.
100 Fe/Zn
Adhesive strength [MPa]
Film/Zn Film/Zn
90
Film Cohesive
Breakage location(%)
Fe/Zn 80 Film/Dolly
70
Film/Dolly 60
6.2 6.0
50 5.1 4.8
40 4.6
Film/Dolly
30
20
10
0
Fe/Zn Film/Zn 2 min 3 min 4 min 5 min 6 min
(a) (b) 0.7% 1.7% 3.7% 5.6% 7.0%
Figure 8. The failure plane and crystallinity of the film: (a) Image showing the broken
part in the failure plane; (b) the corresponding breaking location (%) vs. crystallinity
Figure 8. with the air-cooling time [19].
together
Crystallinity and Impact Resistance
Finally, we checked the impact resistance [19]. Figure 9 shows the

relationship between the crystallinity and impact resistance. When the
crystallinity was 0.7%, the impact resistance was 20 kg cm. However, when it
was 1.7%, the impact resistance decreased to 13 kg cm. It is known that the
impact resistance decreases as the crystallization increases [21]. In addition,
Thames et al. suggested that a highly crystalline powder coating has decreased
impact resistance [22]. Therefore, we consider that the behavior might be
similar in this case.
These findings suggest that suppressing crystallization is a more effective
way of forming a coating film with relatively good mechanical performance in
a cooling time of less than 2 min, which thus minimizes both the energy
consumption and the cost in this fluidized powder vessel process in the field.
25
Impact resistance (kg cm)
20
15
10
0
0 2 4 6 8
Crystallinity (%)
Figure 9. Relationship between the crystallinity and impact resistance [19].
Figure 9.
COMPOSITE OF THERMOPLASTIC PET
Selection of Secondary Material and Film Formability
of Composite Coating
The next challenge was to form a composite for higher durability [23].
The literature includes many reports on ways to easily hydrolyze polymeric
material in the presence of alkalinity. Hare has reported the paint film
degradation mechanism and its control [24]. He explained that esters are
particularly vulnerable to hydrolysis, especially under alkaline conditions,
where they are attacked by hydroxyl ions. Subsequently, the reaction produces
a polyol (e.g., ethylene glycol) and the salt of a fatty acid. This salt is very
much like soap, and it destroys any adhesion that the coating has achieved
with the substrate. Thus, we have attempted to improve the performance even
under alkaline conditions; that is, we have tried to prepare a new composite
powder coating, in which one material protects against alkalinity without
sacrificing the original weathering resistance, and to develop a technique for
preparing a composite coating film. Among the materials of PVB (polyvinyl
butyral) and PA (polyamide) resins, PVB has been selected as a secondary
protective layer owing to its relatively good film formability without pinholes
and good adhesive strength to the primarily PET film.
First, the film mass and thickness and the specular gloss property were
investigated for a new composite PET/PVB powder coating. Figure 10(a) and
(b) respectively show film mass and thickness (a) and specular gloss and
roughness (b) as a function of temperature. The heating time was constant at
20 min. For Figure 10(a), both mass and thickness increased with heating
temperature. This might be because, when the heating temperature increases,
the amounts of PET and PVB powders that melt also increase. In Figure 10(b),
the specular gloss increases with the heating temperature. It is considered that
this corresponds to the realization of a uniform and smooth film. In relation to
this specular gloss, the surface roughness is shown on the right axis. The
surface roughness decreases with temperature. Therefore, it was found that
when the heating temperature increases, the degree of melting increases, which
results in a smoother surface.
Next, we investigated the effect of PVB content in the mixing and coating
process. Figure 11(a) and (b) respectively show film mass and thickness (a)
and specular gloss and roughness (b) as a function of PVB content. The
heating temperature was constant at 3700C. Both the mass and thickness
increased with PVB content in Figure 11(a). In Figure 11(b), the specular gloss
decreases greatly with PVB content from near 100% to around 40%. The
surface also becomes rough. Correspondingly, the right axis shows that the
surface roughness increases almost linearly to 1.6 μm with PVB content. Thus,
when PVB content increases, more secondary polymer components precipitate
in the surface area, which makes the originally smooth surface rougher. Of the
factors involved in the coating process studied in this research, higher
temperature treatment (3700C) was the best way to realize a thick and smooth
film. For practical applicability and film formability, a higher secondary

polymer concentration and good fluidity in the powder vessel are essential.
7 450 100 3.5
Surface roughness (µm)

400
Thickness of film (µm)

6 Specular gloss
3.0
60o specular gloss (%)

Surface roughness
Film mass (g)
350 80
5 2.5
300
4 250 60 2.0
Mass
3 Thickness 200 1.5
40
150
2 1.0
100 20
1 50 0.5
0 0 0 0.0
300 320 340 360 380 300 320 340 360 380
Temperature (oC) Temperature (oC)
(a) (b)
Figure 10. (a) Film mass and thickness and (b) specular gloss and roughness as a
function of temperature [23].
Figure 10.
7 500 100 2.0
Surface roughness (µm)

6
Film thickness (µm)
60o specular gloss (%)
400 Specular gloss

80
5 Surface roughness 1.5
Film mass (g)
4 300 60
Mass
1.0
3 Thickness 200 40
2
100 0.5
1 20
0 0 0 0.0
0 5 10 15 20 0 5 10 15 20
PVB content PVB content
(a) (b)
Figure 11. (a) Film mass and thickness and (b) specular gloss and roughness as a
function of PVB content [23].
Figure 11.
Structural Analysis and Basic Stress-Strain Property
To determine whether the desired dual phase structure was formed, a

structural analysis was performed. Figure 12 shows microscope photographs
of cross-sectional areas of PET/PVB20, PET/PVB25, and PET/PVB30 [25].
We can recognize two regions: a lighter region and a darker one. To make this
easier to see, a sketch is provided in (d). The thickness of the dark phase in the
surface layer was around 100–150 μm. The dark spot in the middle layer was
around 100–300 μm [A in Figure 12(d)]. Here, we performed micro FTIR on
the dark spot and light region, namely, regions A and B in the sketch. The
FTIR spectra we obtained are shown in Figures 13 and 14. Figure 13 shows
FTIR for A in the sketch. All of the spectra have a similar shape. Some
characteristic absorbance was observed. We therefore found that these spectra
identify the PVB phase [26]. We concluded that the dark region in the cross-
sectional area is only the PVB phase, not mixed with the PET phase. Figure 14
shows FTIR spectra for B in the sketch. We found that these spectra identify
the PET phase. Therefore, we concluded that the broad light part of the cross-
sectional area was only the PET phase.
500 µm
Film A B
Steel
(a)
Film
Steel
Film
(d)
Steel
(b)
Film
Steel
(c)
Figure
Figure12.12. SEM images showing cross-sectional areas of composite coating: (a)
PET/PVB20, (b) PET/PVB25, and (c) PET/PVB30. (d) Sketch for ease of
understanding [25].
PET/PVB20
PET/PVB25
PET/PVB30
Transmittance (%)
PVB
ν(–C=O)
ν(O–H) 1740cm-1 ν(C–O)
3200cm-1 ν(C–H) δ(C–H) 970-1170cm-1
2880-2950cm-1 1340-1420cm-1
3000 2000 1500 1000
Wavenumbers (cm-1)
Figure
Figure 13.FTIR spectra of dark part A in Figure 12(d) [25].
13.
PET/PVB20
PET/PVB25
Transmittance (%)
PET/PVB30
PET
ν(–C=O)
1720cm-1 ν(=C–O)
1285cm-1 ν(C–O)
1120cm-1
3000 2000 1500 1000
Wavenumbers (cm-1) γ(C–H)
720cm-1
Figure 14. FTIR spectra of light part B in Figure 12(d) [25].

Figure 14.
500 µm
(a) PVB
(d)
(b)
Figure 15. (c)
Figure 15. SEM images showing surface area from upper side of composite coating:
(a) PET/PVB20, (b) PET/PVB25, and (c) PET/PVB30. (d) Sketch for ease of
understanding [25].
Similarly, microscope photographs of the surface area taken from above

are shown in Figure 15(a–c). It was relatively clearly observed that there were
dark round spots and light parts. A sketch is shown in Figure 15(d) to make
this easier to understand. The size of the round spot is approximately 100–300
μm. Here, we performed micro FTIR on the spots in the dark and light parts.
Surprisingly, we found that dark and light regions are both the PVB phase, not
the PET phase.
The possible formation process is illustrated in Figure 16. First, the
composite powder coating of PET and PVB was heated above their Tms so
that both components melted (1 in Figure 16). Second, the temperature
decreased during quenching. PET, which has the higher Tm, precipitated
earlier than PVB to form the coat (2 in Figure 16). Consequently, the PVB
remained in position or was pushed toward the upper side and lower side (3 in
Figure 16). Finally, the PVB coagulated to form a continuous layer on the
surface layer (4 in Figure 16).
We found that many PVB domains accumulate and connect to form a
continuous phase in the surface layer. We expect this continuous PVB phase to
protect the film against the outdoor environment.
1 Cooling 2
Zn galvanizing Zn galvanizing
PET and PVB are PET precipitates earlier
melted. than PVB to form coat.
：PET
：PVB
Cooling
4 Cooling 3
Zn galvanizing Zn galvanizing
PVB coagulates to form continuous PVB remains or is pushed
layer on surface. out to upper or lower side.
Figure 16. Possible process to form dual structure [25].
As a basic mechanical property, the stress-strain curve was obtained

through tensile testing. Figure 17 shows tensile strength at the break point as a
Figure 16.
function of PVB content [27]. As seen, tensile strength at the break point is
initially around 46 MPa and then decreases to around 26 MPa as the PVB
content increases. Similarly, an elongation at the break point decreases as the
PVB content increases, although it isn’t shown here. PVB is generally weaker
than PET in terms of tensile strength. It is therefore considered that the
increased PVB, which is weaker than PET, exhibited the smaller tensile
strength of composite coating film. It is also found that these values are within
the range that typical PET resin alone presents and that PET/PVB maintains a
relatively good mechanical stress-strain property as a coating as a whole.
60
Tensile strength at break point (MPa)
50
40
30
20
10
0
0 10 20 30 40
PVB content
Figure 17. Tensile strength at break point as a function of PVB content [27].
Figure 17.
Alkaline Resistance of Composite Coating
As we found the basic structure and basic stress-strain mechanical

property of the composite coating film, we tried to determine the alkaline
resistance, which is the final goal for higher durability [23, 27, 28]. We
immersed a test piece in 10 wt% NaOH aqueous solution at 40oC for seven
days and observed the surface appearance and measured the mass loss by
hydrolysis of the coating. The surface appearance of PET/PVB20 film after
alkaline testing and that of PET alone are shown in Figure 18(a) and (b),
respectively. It was found that there was a noticeable difference between them.
The composite PET/PVB20 film showed no cracking or peeling. It was
therefore also considered that the secondary layer certainly protected the film
from an environment filled with alkaline aqueous solution. The loss of mass
during alkaline testing is shown in Figure 19. It was found that when the
concentration of PVB was high, the degree of mass loss was low. It was
confirmed that the film was protected by the secondary layer of PVB. It was
therefore considered that secondary PVB is polymerized by addition
polymerization, which makes hydrolyzing it difficult even in the presence of
NaOH solution. Finally, the mechanical strength was investigated in both an
accelerated alkaline solution and a mortar environment. Figures 20 and 21
show adhesive strengths before and after alkaline testing for seven days and
testing in the mortar environment for one month, respectively. Figures 20 and
21 indicate that the adhesive strength certainly decreased with time. However,
the decreased strength is not so serious; that is, the numerical value is around 1
MPa or above, which is enough in industrial fields. It is also considered that
the PVB inhibits the hydrolysis of PET, and, consequently, degradation is
small.
(a) (b)
Figure 18. Surface appearance after alkaline testing: (a) PET/PVB20 at seven days; (b)
PET alone at four days [23].
0
-0.1
PET/PVB20
Loss of mass (g)

-0.2
PET/PVB15
-0.3
PET/PVB10
-0.4
PET/PVB5
-0.5
PET/PVB0
-0.6
-0.7
-0.8
0 5 10
Alkaline testing time (day)
Figure 19. Loss of mass during alkaline testing [23].
7
Adhesive strength (MPa)
5
○PET/PVB0
4 □PET/PVB20
Figure 19. 3 △PET/PVB25
2 ◇PET/PVB30
1
0
0 2 4 6 8
Alkaline testing time (day)
Figure 20. Adhesive strength before and after alkaline testing [28].
7
Adhesive strength (MPa)
Figure 20. 5 ○PET/PVB0

4 □PET/PVB20
△PET/PVB25
3
◇PET/PVB30
2
0
初期
0 1ヶ月経過後
1 month
Mortar testing time
Figure 21. Adhesive strength before and after mortar testing [27].
Figure 21.
Salt Water Spray Resistance of Composite Coating
Since we recognized an enhanced alkaline resistance by mixing the PVB,

we continued to check the various resistances against harsh factors in outdoor
environments such as sea salt water, UV light, and heat cycles [23]. Salt water
spray resistance was investigated in accordance with JISK5600-7-1[16].
Figure 22 shows the PET/PVB20, PET/PVB25, and PET/PVB30 test pieces
after 3500 h of salt water spray testing. No blisters, cracks, or peeling was
observed for any of the pieces. Of course, the cross-cut part contained small
particles of white rust, but this did not develop after 2000 h. There was no
blistering or peeling surrounding the cross-cut. Therefore, we found that the
dual-layer structural thermoplastic PET powder coating film offers relatively
good anticorrosion performance even when exposed to salt water spray for
3500 h.
In addition, we also performed outdoor exposure testing at a coastal site
on Miyake Island in Japan, in accordance with JISK5600-7-6 [18]. The site is
merely around 50 meters from the seashore. Figure 23 shows the surface
appearances of pieces at outdoor exposure site. Even after five years of
exposure of PET/PVB25, there was no blistering, cracking, peeling, or white
rust even for cross-cut parts, similar to PET alone after six years of exposure.
From this observation, we found that the salt water particle resistance in both
indoor accelerated and outdoor exposure tests was sufficient. The composite
PET/PVB does not sacrifice the original PET’s sea salt resistance.
PET/PVB20 PET/PVB25 PET/PVB30
2000 h
2500 h
3500 h
Figure 22.
Figure 22. Surface appearance of pieces after salt water spray testing [25].
PET/PVB25 With cross-cut PET
5 years 5 years 6 years
Figure 23. Surface appearance of pieces at outdoor exposure site at coastal area of
Miyake Island in Japan.
2000 h
Figure 23.
2250 h
2500 h
Figure 24. Surface

Figure 24. appearance of pieces after UV test by Xenon arc radiation [25].
UV Resistance of Composite Coating
The UV resistance of this composite coating was investigated in

accordance with JISK5600-7-7 [17]. Figure 24 shows the test pieces
after 2500 h. In Figure 24, PET/PVB20, PET/PVB25, and PET/PVB30 all
have the same appearance. The qualitative evaluations of the color change and
specula gloss are shown in Figures 25(a) and (b), respectively. Figure 25(a)
shows that the color difference, ΔE, increased with the testing time. The plot is
the average of two measurements for each time. When the PVB content
increased, the degree of color change decreased. It is therefore considered that
the secondary PVB layer contributes to protecting the film against color
changes caused by UV light. In contrast, the specular gloss in Figure 25 (b)
decreased with testing time when the PVB content increased. This is probably
because the PVB itself is degraded by the UV light, and this causes high
surface roughness that could degrade the specular gloss. We therefore checked
the surface roughness by employing a surface texture measuring instrument
(Surfcom570A, ACCRETECH, Tokyo, Japan). The measurement revealed
that the surface roughnesses Rmax at 3000 h are 136.7, 142.2, and 167.4 µm for
PET/PVB20, PET/PVB25, and PET/PVB30, respectively. These data were in
agreement with the above consideration. It is therefore considered that the
PVB itself could be sacrificed to maintain a good color difference; however,
the consequent high surface roughness would degrade the specular gloss.
5 100
(a) (b)
60o Specular gloss (%)

PET/PVB20
136.7μm
Color difference (ΔE)
4 PET/PVB25 80
PET/PVB20
PET/PVB30
3 60
PET/PVB25 142.2μm
PET/PVB30
167.4μm
2 40
1 20
0 0
0 1000 2000 3000 4000 0 1000 2000 3000 4000
Time (h) Time (h)
Figure 25. Quantitative evaluation of UV resistance against Xenon arc radiation: (a)
Color difference; (b) 60o Specular gloss [25].
7
5.5 years 5.5 years
100
(a) (b)
6
Figure 25.
Color difference (ΔE)
80 PET
60ﾟSpecular gloss
5
PET/PVB25
4 60
3
40
2 PET
20
1 PET/PVB25
0 0
0 20 40 60 0 20 40 60
Exposure time (month) Exposure time (month)
Figure 26. Quantitative evaluation of UV resistance by outdoor exposure test in Tokyo:

(a) Color difference; (b) 60o Specular gloss.
Lastly we performed outdoor exposure testing at a metropolitan site in

Shinkiba in Tokyo, in accordance with JIS K5600-7-6. [18] Figure 26 shows
Figure
the26.
color difference and specular gloss for 5.5 years exposure. For 5.5 years,
PVB contributed to preventing ΔE degradation. However, in specular gloss,

there was no large difference between them, which agrees with the results of
the indoor accelerated test above.
Heat Cycle Resistance of Composite Coating
Finally, the heat cycle resistance of the composite coating was examined
for a repeated temperature cycle between –30oC to + 70oC. Figure 27 shows
the surface appearance of pieces after 200 cycles. There appears to be no white
rust, blistering, cracking, or peeling in the images. However, one or two small
blisters were observed on the coating. This might be because the air remaining
in the Zn steel was expelled at + 70oC as the temperature rose above the Tg of
the secondary powder coating film (68oC), and the air would cause blistering
relatively easily.
200 cycles
Figure 27. Surface appearance of pieces after 200 cycles of heat cycle testing [25].
Figure 27. CONCLUSION

We described the challenges we faced in applying thermoplastic PET
powder coating to steel structures in the telecommunications field and
presented research results on treatments of polymer crystallinity and
composites for higher durability against alkalinity as well as against any harsh
environment. The main findings are as follows:
(1) Upon controlling the degree of crystallinity, the coating exhibits a

relatively good performance in terms of specular gloss, adhesion, and
impact resistance, which minimizes both the energy consumption and
the cost in the fluidized powder vessel process in the field. The
crystallinity is governed by crystallization kinetics represented by, for
instance, the Vogel Fulcher and Abrami equations. It is therefore
possible to manipulate the degree of crystallization by controlling the

air-cooling time.
(2) To form the dual structure comprising primary PET on the bottom and
secondary PVB on the surface for higher durability, a higher
temperature (370oC) and higher secondary material concentration are
essential. This PET/PVB composite coating presents a relatively good
mechanical stress-strain property as a coating and an enhanced
alkaline resistance, where PET presents a good mechanical property
and PVB suppresses the hydrolysis of PET. The composite also
exhibits good performance in terms of sea salt water resistance, UV
resistance, and heat cycle resistance in the environment. Innovated
thermoplastic PET material, with the aid of a secondary material, is
essential to the function and the integrity of steel structures in
industrial fields in terms of both eco-friendliness and long-term
durability.
ACKNOWLEDGMENT
The authors thank Dr. T. Handa, T. Ueda, R. Nishio and Y. Watanuki of
NTT-AT for their helpful discussions and support in the experiments.
REFERENCES
[1] Y. Takeshita et al., NTT Technical Reviews, 11 (1), 2013.
[2] H. Kawai, Handbook of Industrial Coating, 1st ed., Technosystem.
Tokyo, 2008. p. 256.
[3] K. Hirano, J. Chem. Eng. Jpn. 72, 2008, pp. 78-81.
[4] K. Onishi, Development and prospect of powder coating, CMC, Tokyo,
Japan, 2006, pp. 8-15.
[5] T. Handa, H. Takazawa, Bousei Kanri (Rust prevention & control) 41
(5), 1997, pp. 157-168.
[6] C. Kawamura et al., Coating resins synthesized from recycled PET,
Prog. Org. Coat., vol. 45, 2002, pp. 185-191.
[7] Basic Polymer Science, Tokyo Kagaku Dojin, Kobunshi-Gakkai, 2006,
pp. 170-182.
[8] D. W. Van Krevelen, Properties of polymers, 3rd ed, Elsevier, New

York, 1990, pp. 588-601.
[9] F. Bueche, Physical Properties of Polymers, Interscience publishers, a
division of John & Sons, New York, London, 1962, pp. 305-310.
[10] T. Brock, M. Croteklaes, P. Mischke, European Coatings Handbook, 2nd
ed., Vincentz Network, Hannover, Germany, 2010, pp. 207-210.
[11] M. Todoki, Sen’i Gakkaishi, 65, 2009, pp. 293-297.
[12] G. Groeninckx, H. Reynaers, H. Berghmans, G. Smets, J. Polym. Sci.
Part B: Polym. Phys., 18 (6), 1980, pp. 1311-1324.
[13] JIS (Japanese Industrial Standards) B 0601, Surface Texture –
Definitions and Indication, 1994.
[14] JIS K 5600-5-7 (ISO 4624), Testing Methods for Paint – Part5:
Mechanical Property of Film – Section 7: Adhesion Test (Pull-off
Methods), 1999.
[15] JIS K 5600-5-3, Testing Methods for Paints – Part 5: Mechanical
Property of Film – Section 3: Falling-Weight Test, 2002.
[16] JIS K 5600-7-1, “Testing methods for paints—Part 7: Longperiod
performance of film– Section 1: Resistance to neutral spray”, Tokyo,
Japan, 1999, (ISO 7253:1984).
[17] JIS K 5600-7-7, “Testing methods for paints—Part 7: Long period
performance of film–Section 7: Accelerated weathering (Exposure to
filtered xenon-arc radiation)”, Tokyo, Japan, 1999, (ISO 11341:1994).
[18] JIS K 5600-7-6, “Testing methods for paints—Part 7: Long period
performance of film-Section 6: Natural weathering,” Tokyo, Japan,
2002.
[19] Y. Takeshita, T. Sawada, T. Handa, Y. Watanuki, T. Kudo, Prog. Org.
Coat., 75, 2012, pp. 584-589.
[20] G. Zhang, H. Liao, M. Cherigui, J.P. Davim, C. Coddet, Eur. Polym. J.
43, 2007, pp. 1077-1082.
[21] L.E. Nielsen, Mechanical Property of Polymer, 1st ed., Kagakudojin,
Tokyo, 1968, p. 123.
[22] S. F. Thames, K.G. Panjnani, S. D. Pace, Proceedings of the Water-
Borne High-Solids Powder Coating Symposium, vol. 22, 1995, p. 320.
[23] Y. Takeshita, T. Kamisho, S. Sakata, T. Sawada, Y. Watanuki, R.
Nishio, T. Ueda, J. Coat. Technol. Res., 10, 2013, pp. 503-514.
[24] Clive H. Hare, “Chapter 20: Mechanical-Induced Stress from the
Manufacturing Process.” In: Brady, RF, Jr (ed.) Paint Film Degradation
Mechanisms and Control, pp. 155–156. SSPC, Pittsburgh, PA, 2001
[25] Y. Takeshita, T. Kamisho, S. Sakata, T. Sawada, Y. Watanuki, R.

Nishio, T. Ueda, J. Appl. Polym. Sci., 128, 2012, pp. 1732-1739.
[26] Japan Paint Inspecting Association. Handbook of FT-IR Spectrums of
Coating Resin, Tokyo, Japan, 1975, p 11 and 56.
[27] R. Nishio, Y. Watanuki, T. Ueda, Y. Takeshita, T. Sawada, Bousei
Kanri (Rust prevention & control), 57 (2), 2013, pp. 41-45.
[28] Y. Watanuki, T. Ueda, R. Nishio, Y. Takeshita, T. Sawada, Bousei
Kanri (Rust prevention & control), 56 (1), 2012, pp. 24-28.
BIOGRAPHICAL SKETCH
Name: Yukitoshi Takeshita, PhD
Affiliation: NTT Device Innovation Center, NTT Corporation
Education:
1987.3 B.E., Faculty of Applied Chemistry, Kyushu University, Fukuoka,
Japan
1989.3 M.E., Graduate School of Applied Chemistry, Kyushu University,
Fukuoka, Japan
2001.1 Ph.D. in Engineering, Kyushu University, Fukuoka, Japan
Business Address:
3-1, Morinosato-Wakamiya, Atsugi-shi, Kanagawa 243-0198 Japan
Research and Professional Experience:
 Study of preparation of AlN powders by vapor phase reaction of Al(i-

Bu)3-NH3 system
 Study of preparation of spherical Titania particles from inorganic
precursor by homogeneous precipitation
 R&D on ionic conductive polymer for Li batteries using SBR/NBR
rubber mixed lattices
 R&D on detoxifying harmful substances by supercritical CO2
extraction and H2O decomposition for environmental remediation
 R&D on a thinner and lower friction indoor optical cable for fiber to
the home (FTTH)
 R&D on eco-friendly nano-structured coating system that prevents

corrosion and rust on steel structures, such as wireless steel towers,
steel telephone poles, cable pipes, and bridges that accommodate
cables
Publications Last:
Takashi Miwa, Yukitoshi Takehsita, Azusa Ishii, Takashi Sawada,

Comparison of corrosion behaviors between steel and zinc in accelerated
corrosion test, Rust Prevention & Control Japan, 60 (2016) 7-14.
Takashi Miwa, Yukitoshi Takeshita, Yuichi Akage, Masamitsu Watanabe,
Masaaki Takaya, Takashi Sawada, New Indoor Accelerated Weathering
Condition for Approximating Molecular Weight Distribution of Outdoor-
exposed LDPE, Journal of Chemistry and Chemical Engineering, 10
(2015) 481-493.
Yukitoshi Takeshita, Ethan Becker, Seizo Sakata, Takuya Kamisho, Takashi
Miwa, Takashi Sawada, Water Absorption and Degraded Stress
Relaxation Behavior in Water-Borne Anticorrosive Urethane/Epoxy
Coatings, Journal of Chemistry and Chemical Engineering, 9, 2 (2015)
75-89.
Takashi Miwa, Yukitoshi Takeshita, Yuichi Akage, Masamitsu Watanabe,
Takashi Sawada, Comparison of the Photodegradation Behavior of LDPE
Using Accelerated Weathering Instruments, Corrosion Engineering, 64
(4) (2015) 99-106.
Takashi Miwa, Yukitoshi Takeshita, Seizo Sakata, Takashi Sawada, Study on
effect of dolly shape on adhesion test and correlation with cross-cut test,
Rust Prevention & Control Japan, 58 (2014) 211-215.
Yukitoshi Takeshita, Ethan Becker, Seizo Sakata, Takashi Miwa, Takashi
Sawada, States of Water Absorbed in Water-Borne Urethane/Epoxy
Coatings, Polymer, 55 (2014) 2505-2513.
Takuya Kamisho, Yukitoshi Takeshita, Seizo Sakata, Takashi Sawada, Water
absorption of water-based anticorrosive coatings and its effect on
mechanical property and adhesive performance, Journal of Coatings
Technology and Research, 11, 2 (2014) 199-205.
Takuya Kamisho, Yukitoshi Takeshita, Seizo Sakata, Takashi Sawada, Aging
behavior of water-based and solvent-based coatings in dry and wet
environments, Rust Prevention & Control Japan, 57, 12 (2013) 451-457.
Yukitoshi Takeshita, Takuya Kamisho, Seizo Sakata, Takashi Sawada,
Yoshikazu Watanuki, Ryuichi Nishio, Toshinobu Ueda, Mixed Powder
Coating Film Using Thermoplastic Polyester and Its Alkaline Resistance,

Journal of Coatings Technology and Research, 10, 4(2013) 503-514.
Yukitoshi Takeshita, Takuya Kamisho, Seizo Sakata, Takashi Sawada,
Yoshikazu Watanuki, Ryuichi Nishio, Toshinobu Ueda, Dual Layer
Structural Thermoplastic Polyester Powder Coating Film and Its
Weathering Resistance, Journal of Applied Polymer Science, 128,
3(2013)1732-1739.
Ryuichi Nishio, Yoshikazu Watanuki, Toshinobu Ueda, Yukitoshi Takeshita,
Takashi Sawada, Mechanical property of dual layer structural
thermoplastic coating, Rust Prevention & Control Japan, 57, 2 (2013)
101-105.
Chapter 3
MECHANICAL RECYCLING OF
Nathalie Benoit1, Rubén González-Núñez2

and Denis Rodrigue1,*
1
Department of Chemical Engineering,
Université Laval, Quebec City, Quebec, Canada
2
Departamento de Ingeniería Química,
Universidad de Guadalajara,
Guadalajara, Jalisco, México
ABSTRACT
Thermoplastic composites have found several applications over the
years, but their main advantage when compared with thermoset
composites is their possibility to be reshaped after processing through
further melting and remolding processes. This allows the materials to be
recovered and reprocessed if they fail quality control (post-industrial) or
after their end of life (post-consumer). This is very attractive as recycling
is one of the main principles of sustainable development. This is even
more appropriate as thermoplastic composites are more and more
produced and consumed in a wide range of applications such as
packaging, automotive and aeronautics, furniture, building and
construction, as well as sport and leisure goods. Thus increasing demands
*
Corresponding Author: [email protected].
96 Nathalie Benoit, Rubén González-Núñez and Denis Rodrigue
leads to a high number of parts being produced and discarded. With this
high volume of composites production and use, the emergence of
constantly new and stronger policies and laws towards sustainable
development and recycling were developed. This is why it becomes
important to study and understand the conditions of the materials at their
end of life and develop applications like recycling to reintroduce these
high amounts of materials into production lines. To limit both the amount
of material consumed and landfilled, thermoplastic composite waste can
be considered as a source of raw material for the manufacture of products
through recycling and reprocessing. However, to obtain good final
properties, it is essential to understand the degradation processes and the
materials behavior during their life cycle and their recycling. In this
chapter, a review of the different possibility to recycle thermoplastics
composites is presented, with a focus on the mechanical recycling
techniques available. Some figures about composites production and
recycling are presented and discussed, followed by a brief presentation of
the different recycling methods available for thermoplastic composites
with their advantages and limitations. Then, an overview of mechanical
recycling is made considering both organic and inorganic fillers. For all
cases, the material behavior towards recycling and the resulting properties
are considered, as well as the solutions developed to improve the
performance and quality of the recycled materials.
Keywords: thermoplastics, composites, recycling, foams, properties
INTRODUCTION
Composites materials are composed of two phases, one being the matrix,
and the other one, the reinforcement (fiber or particles). The matrix protects
the reinforcement, provides cohesion to the material and binds the reinforcing
particles together, thus distributing the stresses under load. The reinforcement
increases the rigidity and the performances of the material. The association of
both phases leads to interesting properties, combining the properties of both
components. The performances of the resulting composites depend on the
capacity of the material to transfer the stresses between both constituents, and
thus on the interface quality. These materials have been produced continuously
for a few decades now. The high demand for composites created a new
category of plastic-based waste that has to be taken care of. Nowadays, 8.7
million tons of polymer composites are produced each year, thus generating
high volumes of waste [1]. For example, in France, 30,000 tons of composites
waste are collected every year. As they will reach their end of life soon and the
Mechanical Recycling of Thermoplastic Composites 97
quantities will increase in the future, their recycling should be studied and
understood as to recover the materials and their value, as well as
simultaneously avoid their accumulation in landfill [2]. Recycling of polymer
based materials is environmentally very interesting as it leads to the reduction
of waste simultaneously with saving of virgin resources, and especially
petroleum [2]. However, due to the heterogeneous nature between the matrix
and the reinforcement, and to the lowering of the properties of the recycled
materials, thermoplastic composites are currently rarely recycled at the
industrial scale, as the existing techniques are often expensive [2-4]. Thus,
most composites are currently converted to energy and fuel or fiber recovery.
However, in the latter, the fibers recovered show an important drop of
performance (up to 90% for glass fibers) and suffer from a lack of cost
competitiveness, which usually make them unsuitable for most applications
and markets [5, 6].
Thermoplastics composites are widely used nowadays. Their matrices are
characterized by their aptitude to be repeatedly melted by heating and
solidified by cooling, and this in a reversible way. The reinforcement add
rigidity and strength to the matrix, thus making a material combining the
properties of both constituents. Thermoplastic composites have become an
area of increased interest due to their relatively low cost, wide range of
mechanical properties, toughness, problem free storage, resistance to chemical
attack, ease of processability, and above all their better recyclability. This is
related to their fundamental ability to be reprocessed and reshaped when
reheated leading to much simpler and direct recycling paths [1, 3, 7-9].
Because they have a melting point, they can thus be remelted, reshaped and
reharden multiple times. This is not the case for thermosets having highly
reticulated networks which cannot be melted after their first shaping [10, 11].
Although thermosets are currently the most common composites representing
about 90% of the composite market in 2014, thermoplastics composites are
continuously increasing their market share due to their numerous benefits [1,
3, 9]. Over the last decades, the demand for thermoplastics composites
increased with the development of new applications in several fields such as
decking, outdoors equipment, household, leisure, construction, transportation,
packaging, sports, leisure, domestic applications, energy, automotive and
aerospace [1, 3, 9, 12, 13]. This increased demand on composites produced
and used led to an increase of waste generated, thus increasing the pressure for
the development of efficient and viable recycling methods for such materials
[12, 14, 15]. It was projected that in the next few years, the total global
production of composites will exceed 10 million tons, thus occupying a
volume of over 5 million cubic meters [5]. Among these composites, more
than 90% are glass reinforced (GR) composites, and about 40% of this volume
is associated to thermoplastics composites. With constantly new
environmental directives and legislations being established, both producers
and manufacturers are asked to consider the environmental impact of their
products. This is particularly true for thermoplastics composites, which are
gaining interest over thermosetting composites due to their several advanteges
like design flexibility, easier processing and shorter processing time [10, 11, 13,
14]. Moreover, most of them usually decompose slowly, thus requiring to be
taken care of at the end of their life cycle.
In the last few years, the society has constantly increased its concern about
sustainable development leading the actual governments to consider new laws
and directives towards resource management and waste revalorization [16].
The plastic production continuously increase to fit the increasing demand for
such materials. Around the world, the general plastic production increased by
38% in ten years, reaching 311 million tons in 2014, with an important part of
thermoplastics materials [17]. In Europe, thermoplastics materials represent
more than 72% of the total production [16, 17]. Thus, polymers represent an
important part of the municipal solid waste (MSW) in many countries such as
the United States, Canada and in Europe. For example, they represent up to
20% of the MSW in the United States, and only a few percent were recycled
[14, 18]. Moreover, thermoplastics are generally the main constituents of the
polymer fraction of MSW [19]. In Europe, 7.7 mT of plastic wastes were
recovered in 2015. Among this waste, only 30% of the total volume was
recycled, 39% was energetically revalorized, while the remaining 31% was
simply landfilled [17]. In Canada, in the province of Quebec, plastics waste
constantly increased over the last 20 years. Their revalorization only accounts
for about 5% of the total revalorization for all materials, and represents only
16% of the total plastics waste collected from MSW [20]. These low recycling
rates are essentially due to the lack of techniques available and problems
associated with the collect and sorting processes. This shows that more efforts
must be done in these fields to increase the revalorization rate of such
materials. However, it should be noted that, over the last few years, both
energy recovery and recycling increased, while landfilling rate decreased [17].
For example, in Europe, between 2006 and 2014, for the 25 million tons of the
global plastic waste produced, the landfilled fraction decreased by 38%, while
the energy recovery and recycling shares increased by 46% and 64%
respectively, thus reaching 8 million tons of polymers discarded to landfill,
10.2 million tons of polymers energetically recovered, and 7.7 million tons of
polymer waste recycled [17]. Thus, recycling of thermoplastics and their

derivatives is fundamental in the actual social and political context, especially
since they have a slow decomposition rate, as well as a low density, but
represent very important volumes of materials [18]. Considering these facts,
and especially the important volume generated daily, some efforts still need to
be done to manage resources, as well as to manage, treat and revalorize this
kind of waste. However, if the waste volumes and recycling rate of
thermoplastics are usually known, very few statistics can be obtained about
thermoplastics composites [19]. Although some companies are already using
recycled thermoplastics to produce wood plastics composites (WPC), no use
of recycled composites was reported. In any case, the composition of the waste
stream depends on the geographical location, time of the year (season), current
trends and several other parameters [19].
Recycling consists in the collect, the sorting and the treatment of waste.
Materials sorting is an important issue as the presence of impurities and
contaminants can lead to unsuitable or poor properties of the recycled
materials. Three main approaches can be considered for the recycling of
composite waste [21, 22]. The simplest way is to dispose into landfill.
However, this is also the worse scenario from an environmental and
economical point of view, as it can lead to space problems if the volume of
waste is too important [4, 15]. The second option is the incineration of the
material, with possible energy recovery or heat generated, but this also lead to
air pollution and environmental problems. This option is also known as
thermal recycling and is often considered as a part of the third category, which
is recycling. The last, but most important option, is recycling. It is based on the
partial or total recovery of the constituents of the global material. It is
sustainable, economically viable and appears to be the best environmental and
technical choice [18, 23, 24]. It reduces the energy used and the waste volume
disposed to landfill, while simultaneously reducing the raw material
consumption. However, it induces to some extent some changes in the
mechanical, physical and chemical properties of the final materials, that may
affect the material processing conditions and the quality of the end products
[3, 15, 18, 25]. It often gives, not always justified, a “low quality” image to the
products, thus limiting its attractiveness. Compared to virgin composites,
recycled composites often meet problems to satisfy the quality and economical
requirements, thus limiting the choice of suitable market [4, 18]. Several
works considered the partial addition of virgin materials to counter this
drawback. Coupling agents are also often used to improve the composites
properties, but their sensitivity towards recycling is not well known [2]. It is
thus essential to know how the properties of thermoplastics composites are

modified by recycling, and especially how the reinforcement modify polymer
degradation [24]. This should be done by the study of the recycling effect on
the materials properties, by the development of new, cheaper and more
efficient recycling methods, as well as better separation techniques. But it
should also be done by increasing the recyclability of the produced
composites. Besides thermal recycling, two other categories of recycling
methods exist: chemical recycling and mechanical recycling [2, 3, 16, 24].
Chemical recycling leads to reusing the constituents for the production of new
composites, while mechanical recycling is often performed on the whole
composite.
The recycling of thermoplastics is well known and was the focus of
several authors. The extensive long-term recycling of such polymers was
studied by Oblak et al. [25], Jin et al. [26, 27] and Benoit et al. [28]. However,
the recycling of thermoplastics composites, and especially of their extensive or
long-term recycling, is much more limited. This chapter aims at reviewing the
various works on the mechanical recycling of thermoplastics composites to
present the state of the art in this field. To do so, the main recycling methods
for this type of composites are presented first with their advantages and
drawbacks. Then, the recycling of natural organic fillers reinforced composites
will be considered, followed by the study of the recycling of inorganic fillers
reinforced composites. Finally, the main limitations to the recycling of
thermoplastics composites will be presented, as well as their possible solution.
Finally, this review will summarize and conclude about the actual state and
viability of the mechanical recycling of thermoplastics composites.
RECYCLING METHODS
Due to the lack of knowledge, the complexity of recycling and the
predominance of thermosets until recently, the majority of composites
materials were disposed to landfill [29]. But the same situation is still going
on. However, due to the constantly increasing market of composites materials
and the important volumes of waste generated, as well as the high cost of the
reinforcements and additives used in these materials, other alternatives must
be considered [11, 30]. Nowadays, due to their numerous advantages,
thermoplastics composites are gaining interest and gradually taking the place
of thermosets composites in various applications, making their recycling easier
[30]. They present lower manufacturing time, lower sensitivity to impact
damage, easier manufacturing, and better recyclability due to their aptitude to

be remelted [30]. These advantages allow to consider other ways of managing
composite waste.
Composting can be conducted for biodegradable materials, but in this
case, both matrix and reinforcement should be biodegradable to be sure not to
contaminate the environment with permanent residues [31]. For very specific
cases, reuse of materials after cleaning or basic operations can be considered.
If this method is environmentally respectful, however, in the polymer and
composite fields, reuse is quite limited due to health and safety reasons [11,
16]. Because of their inhomogeneous nature, recycling is a great challenge for
polymer composites. The presence of reinforcement leads to technical
challenges such as equipment wear, incompatibility, restrictions on
temperature and other processing parameters, special attention during
processing or other technical constraints. The waste stream has to be consistent
and sufficient (in quality and quantity), and the price low enough to make the
recycling process viable and profitable [20, 32].
Various recycling methods exist for polymer composite materials. These
methods can be divided in three main categories: thermal recycling, chemical
(or feedstock) recycling and mechanical (or physical) recycling [3, 6, 16, 20,
29-35]. Each one has its own list of advantages and drawbacks [31]. The
choice of the method depends on several parameters including the nature and
quality of materials, their initial state, the final state and quality desired for the
materials, as well as their constituents, the nature of the composite
constituents, the desired post-recycling application combined with economic,
technical and legislative considerations [11]. The recycled composites need to
be of good quality and relatively low price to be competitive with virgin
materials and thus place themselves in the current markets [32]. A complete
overview on recycling issues and technologies can be found in Henshaw et al.
[32] and Goodship [36]. For polymer and composite waste, regardless of the
recycling method, two waste categories can be differentiated: industrial
production waste (from the production rejects and waste which has not been
contaminated or used) and post-consumption waste (subjected to
contamination and degradation during its lifecycle) [16, 18]. The latter
generally requires some additional cleaning and sorting steps as to limit
contamination, separate the metal, paper, adhesive and other foreign parts and,
if applicable, to sort the composites by polymer matrix and fiber types [6, 16].
Thermal Recycling
In thermal recycling, the matrix and the fibers are separated by

decomposition of the polymer at high temperatures (between 300 and
1000°C). This can be done through various thermal processes such as
pyrolysis, gasification, fluidized-bed combustion (FBC), rotary kiln, fluidized
bed and mass burn [32, 34, 35, 37]. All these processes are based on the
decomposition of the product under external heating and its transformation
into energy or liquid fuels [16, 29]. The energy, heat or steam generated during
the process is recovered. Regardless of the reinforcement type, some
degradation takes place to a varying extent during thermal processing [3]. The
heat generated during the process can also be transformed into electricity [34].
In pyrolysis, organic materials are decomposed at high temperature, generally
in the absence of oxygen [32]. In the fluidized-bed combustion, the materials
is decomposed under high temperatures (450-500°C) in a silica sand bed
thermally heated and fluidized by hot air. The organic part of the composite is
volatilized by the hot air and transported through the air stream before
separation. The volatilized polymer can then be burned to produce heat and
energy [34]. In the case of natural fiber composites and other low degradation
temperature fibers, this process cannot be used to recover fibers due to the
high sensitivity of natural fibers towards high temperatures, leading to
important degradation while submitted to high temperatures. Waste plastics
and common reinforcements have high calorific values, thus making them
interesting for energy recovery. Moreover, this type of recycling has also been
shown to significantly reduce the volume occupied by this type of waste, but it
also has been noticed that this recycling method can lead to the emission of
pollutants and noxious emanations, thus limiting its applications and
promoting the recyclers to find other ways of recycling these materials [11].
However, it is efficient and viable when no other method can be applied [31].
It is especially suitable for mixed or contaminated waste, as well as for
composites with insufficient properties after extensive recycling or severe
weathering [3, 6, 18, 24, 32, 34].
Chemical Recycling
Chemical recycling is the chemical de-polymerization and dissolution of

the matrix by organic or inorganic solvents through various advanced
processes such as liquid-gas hydrogenation, gasification, viscosity breaking,
steam treatment, catalytic cracking, pyrolysis, glycolysis, alcoholysis,

solvolysis and hydrolysis [11, 30, 34]. For this type of recycling, both the
fibers and the matrix (as polymers, monomers, fuels or chemicals products)
can often be recovered [30]. The polymer matrix is degraded into its
monomers or basic chemical products (oil/hydrocarbon), while the fibers are
filtered to be recovered [11, 12, 16, 30, 34, 37]. The monomers and chemical
products can then be directly reused as raw materials to create new polymers
or as fuels and composites materials. This method is complex, but suitable for
materials with contaminants or when different materials are mixed to some
extent [6, 11, 31]. In hydrolysis, the depolymerization is conducted with water
and catalysts [32]. This methods achieves high product yield and decrease the
amount of waste [11]. However, this type of recycling is quite controversial as
it uses chemicals, such as solvents, bases, acids and washing fluids, and thus
leads to the production of important amounts of waste chemicals, thus in
contradiction with the benefits of recycling. It is also generally expensive, due
to the cost of the chemicals used and the high energy consumption with low
flexibility [30, 31]. These two points reduce strongly its attractiveness,
especially for thermoplastics [16, 18, 20, 24, 32, 34]. Moreover, before
degradation, the material is often ground to increase its surface area, thus
leading to better diffusion processes, but also to some mechanical degradation
in the fiber and matrix, but at a much lower extent than for thermal and
mechanical recycling [18].
Mechanical Recycling
Mechanical (physical) recycling is the recovery of the composite’s

material through various mechanical processes including grinding/pelletizing,
reprocessing and shaping [3, 6, 11, 12, 16, 18, 20, 24, 29-37]. In the case of
thermoplastics, after the usual cleaning and sorting steps, the materials can
directly be pelletized, remelted and reshaped through extrusion, injection
molding or compression molding. They can be processed in any shape and are
suitable for a wide range of applications [11, 29, 32, 34]. They usually have a
fairly good homogeneity [32]. On the contrary, for thermosets, direct
remelting cannot be done and the ground materials are often used as a new
type of reinforcement [29, 32, 34]. This requires very high pressure and simple
shapes. The resulting materials can then be used for limited applications [32].
However, the high pressures, temperatures, and shear stresses undergone
during these three steps lead to irreversible changes in the polymer structure
and properties, as well as in the fibers [3, 16, 18, 24, 32]. This reduction of
properties and performances of the recycled materials is also reinforced by the
loss of properties undergone by the material during its lifecycle. This loss is
due to their exposure to various long-term threats such as temperature, light
(especially UV), moisture and global use wear, and cannot be compensated
during the reprocessing steps leading to lower properties than virgin materials
[16]. For demanding applications, partial addition of virgin material (polymers
and/or fibers) can be considered [32]. The heterogeneity and diversity of the
matrix is another factor substantially influencing the properties of the materials
mechanically recycled.
Recycling of industrial production waste is called closed-loop recycling,
while recycling of post-consumption products is called open-loop recycling
[18]. The former is obviously better for mechanical recycling due to its low
level of contamination and its homogeneity, but both are suitable for
mechanical recycling. In mechanical recycling, the purity of the waste stream
is very important [11]. In the case of mixed waste recycling, due to the
incompatibility and immiscibility between most polymers, the properties of the
recycled material are generally lower than for separated homogeneous waste
recycling [16, 32]. The more complex and contaminated the composite waste
is, the harder it is to recycle and to obtain interesting properties. Thus, in order
to maximize the properties and performances of the recycled composites, the
waste stream should contain as few types of polymer matrix as possible, and
very few contaminants. Therefore, sorting is a critical step in the mechanical
recycling process. Nowadays, different separation techniques exists for
plastics such as density segregation, selective dissolution/precipitation, manual
sorting, triboelectric, mid infrared (MIR), and near infrared (NIR) techniques
[11]. However, the presence of reinforcement in the composites make them
difficult to apply, thus adding another technical challenge. Some methods
based on pulverization have been developed to limit the negative effect of
polymer mixed waste, but are not applicable to composites, leading to severe
degradation of the reinforcement [16]. Moreover, even for polymers, a
complete separation is rarely conducted, due to its high complexity to
implement and perform [11].
Conclusion
All the methods presented above are summarized in Table 1 for a better
comparison and overview of each processes.
Table 1. Recycling methods for thermoplastics

composites and their characteristics
Type of Based on Examples Advantages Drawbacks Products

recycling obtained
Thermal Decomposition Pyrolysis, Volume Thermal Energy,
recycling at high gasification, reduction, degradation, heat, steam,
temperatures fluidized-bed ideal for gases and liquid fuels
combustion mixed pollutant
fractions emanations
and poor
quality
waste
Chemical De- Liquid-gas High Production of Fibers,
recycling polymerization hydrogenation product waste polymers,
in solvents gasification, yield chemicals, monomers,
viscosity expensive, low fuels,
breaking, steam flexibility, chemicals
treatment, degradation, products
catalytic high energy
cracking, consumption
pyrolysis,
glycolysis,
alcoholysis,
solvolysis,
hydrolysis
Mechanical Reprocessing of Extrusion, Flexible Thermo- Composites
recycling the composites injection (shape and mechanical
molding, process), degradation,
compression suitable for requires initial
molding several materials
applications separation
few steps
Both incineration and chemical recycling lead to the loss of the value
added during polymerization and composite manufacturing. Considering the
current prices of polymers and composites, this appears as an important
drawback [30]. From an industrial point of view, mechanical recycling is very
interesting as it uses conventional processing methods while simultaneously
limits the amount of virgin material and energy used while reducing the
emissions of greenhouse gases. It is also flexible, simple and inexpensive [11,
18, 37]. However, several efforts still need to be done to improve the materials
properties, as well as to develop more efficient recycling and sorting methods
[12, 18]. Some efforts should also be made to integrate the notions of
recycling, dismounting, separation and identification of the composite
materials during the conception/design phase to make these steps easier at
their end-of-life [29, 37]. In various works, it was reported that both
mechanical recycling and chemical recycling are the most widely practiced
methods [12, 16]. However, the mechanical recycling of thermoplastics
composites seems to be the most interesting technique in terms of sustainable
development, as well as from an industrial point of view, due to its low cost,
easiness of application and reliability [12, 16, 31]. The technologies and
infrastructure already exist for the production and recycling of plastics, as well
as for the production of composites. They are widely available and can be
easily adapted for composite recycling with limited efforts [16]. This chapter
will thus focus on the mechanical recycling of thermoplastics composites next.
RECYCLING OF THERMOPLASTIC COMPOSITES

REINFORCED WITH ORGANIC FILLERS
Due to an increasing awareness for sustainable development and
recycling, organic fillers are strongly gaining interest and consideration in the
thermoplastic composite field. New environmental policies are constantly
implemented by more and more countries, thus increasing the pressure on
manufacturers to consider the environmental impact of their products [38].
Therefore, the interest of using natural fibers as a reinforcement in
thermoplastics composites is constantly growing. Moreover, most natural fiber
reinforced composites are produced with thermoplastics matrices. Due to their
numerous advantages compared to more traditional fillers, such as their
renewable origin, biodegradability, non-abrasiveness, low production and
manufacturing cost, low density and high specific properties, they are present
in a wide range of fields [2, 23, 39-42].
Pervaiz and Sain [42] compared the energy consumption of glass and
natural fibers and found that using vegetal fibers instead of glass fibers could
save 60% of energy per ton of product. Natural fibers are indeed often
biodegradable and produce less pollution because of their ability to burn
without noxious gases or solid residues [23, 41]. Compared to glass and
carbon fibers, they also reduce respiratory and physical irritations during
handling [43]. Thus, they are progressively replacing inorganic fillers in some
applications from common non-structural applications to some specific
structural applications [23, 38, 41, 43]. This is especially the case in the
automotive industry where, due to directives towards the reduction of energy
consumption and the requirement of lightweight materials, industrials are
trying to progressively replace traditional materials by others with lower cost
and environmental impact [23, 31, 38, 41]. With this consideration,
mechanical recycling of such composites allows significant waste reduction,
while conserving both materials and energy at the same time [44]. However,
they also present several disadvantages limiting their expansion such as poor
resistance to moisture, sensitivity to high temperatures, tendency to aggregate
and incompatibility with hydrophobic materials [38, 43]. It is well know that,
due to their sensitivity to high temperatures, natural fibers are more prone to
thermal degradation during processing than neat thermoplastics, and that poor
matrix/fiber interface quality leads to low stress transfer resulting in lower
mechanical properties.
Wood
Twite Kabamba et al. studied the mechanical recycling of composites

made of low density polyethylene (LDPE) and yellow birch fibers [39]. In
their work, they performed up to 10 reprocessing cycles by extrusion under
constant conditions to simulate the closed-loop recycling of these composites
with a reinforcement content of 15 wt.%. For each generation, thermal,
molecular, rheological and morphological properties were studied, with the
final aim of determining the effect of wood fiber on polymer degradation and
on the recycling process. The fiber length significantly decreased with the
number of cycles, reaching about the third of its initial value after 10
reprocessing cycles, while fiber diameter was not significantly affected.
However, this decrease was not linear and most fiber degradation took place
during the first generation, while the value leveled off after the 6th generation.
The molecular weight distribution of the polymer showed that the number
average molecular weight Mn of the polymer matrix was more affected than
the mass average molecular weight Mw, with respective decreases of 30% and
4% between the first and the tenth regenerations, indicating that chain scission
mainly occurred in the longest molecules. As the fibers and the polymer chains
broke, the composites crystallinity increased with increasing number of
generations. In the rheological tests, the zero shear viscosity decreased by 30%
between the first and the tenth generations, while in elongational rheology the
strain at break did not change, but the maximum stress decreased by about
46% indicating that modifications in the polymer chains essentially occurred

in chain entanglement rather than in the main backbone. The comparison
between the polymer and the composite rheological behavior towards
recycling indicated that the matrix behavior changed from strain hardening to
strain softening for the composite. The other parameters did not display
significant differences between each generation. Thus, the results showed that
the effect of recycling is mainly affecting the fiber size, the matrix molecular
mass and crystallinity, as well as the shear and elongation viscosities.
However, the results also indicated that the degradation undergone was higher
for the composites than for the same polymer without reinforcement because
the presence of wood fibers in the polymer matrix increased the mechanical
stresses (higher flow resistance) leading to more chain scission and the
production of micro-radicals during processing. The presence of fibers also
limits chains mobility and crystallite growth leading to lower properties after
recycling. However, this difference seems to decrease with regenerations,
probably due to fiber length decrease.
Wang et al. studied the effect of processing parameters on the properties
of wood/polyethylene composites (70/30) with coupling agent [45]. In
comparison with the previous work, they showed than composites processing
with a coupling agent limited the decrease of mechanical properties of the
composites if suitable processing parameters are selected [39, 45]. However,
this work only considered one processing cycle. Thus, it gave some indications
about the recycling behavior of such composites considering that most authors
noted that the first cycle is the most critical one, but the study appears to be
incomplete as some recycling effects could be occurring or be significant for a
higher number of regenerations [38, 39]. It should be extended to take into
consideration the long-term effect of recycling, as well as other types of
materials properties. However, the results of this work gave some directives
and recommendations that should be considered and applied when recycling
wood/thermoplastics composites. The authors indicated that the final
properties of the composites is a combination of the complex interactions
between the constituents’ nature, the equipment characteristics, the processing
conditions and the interfacial properties. On the one hand, high temperatures
or stresses led to a higher level of fiber degradation and to lower properties.
On the other hand, it also increased the composite mixing, homogeneity and
fiber distribution, which led to better properties. Thus, the screw speed and the
extruder temperature profile should not be too high as to limit fiber
degradation, but should be high enough to allow a smooth processing. The
position of the reinforcement feeder appeared to be important to limit
mechanical and thermal degradation of the materials. Thus, an appropriate

choice and combination of parameters, including screw configuration and
rotational speed, throughput rate, barrel temperature profile and feeder
position is essential to limit the mechanical and thermal degradation (color
change) of the wood fibers during processing.
Petchwattana et al. studied the closed-loop recycling of WPC partially
made of waste poly(vinyl chloride) (PVC) [21]. It is the only work considering
the mechanical recycling of partial non virgin composites. The first part of
their work focused on the determination of the best ratio between virgin and
waste PVC for such composites. A content of 30% of industrial waste blended
with 70% of virgin materials was found to be the best formulation, while the
wood flour content was set at 20%. In the second part of the work, the
composites produced were subjected up to seven reprocessing cycles. The
results indicated that these composites could be recycled up to seven times
without any critical loss of mechanical properties. As the two previous studies
[39, 45], the results showed a molecular weight decrease due to chain scission
induced by the shear stresses generated in the material during processing. At
the end of the seven reprocessing cycles, Mn decreased by 35% while Mw
decreased by 30%. Moreover, the flexural strength and the flexural modulus
decreased by 13% and 23% respectively, after the second cycle. Then, no
significant change was observed for the following cycles until the seventh
reprocessing cycle where the properties started to decrease again. The results
also showed no significant effect of recycling on the impact properties. The
authors concluded that, besides the first drop, these composites could be
mechanically recycled without critically affecting their properties and
performances.
The effect of the initial wood particle size and of the addition of coupling
agent on the degradation of mechanically recycled composites was reported by
Soccalingame et al. [46]. The mechanical recycling was simulated by multiple
grinding and injection molding cycles. The polypropylene/spruce wood flour
composites was previously compounded on a twin-screw extruder to ensure
homogeneity. Seven successive reprocessing cycles were then conducted. The
coupling agent used was polypropylene grafted with maleic anhydride
(MAPP) and the wood content was set to 20 wt.%. Two particle sizes were
used. The first one, called G1, was mainly composed of fine particles of
different sizes, while the second one (G2) was made of coarse particles added
with fine ones. Several properties of the composites and their constituents
were considered: morphological, molecular, mechanical, rheological and
thermal. Once again, the results showed that the mechanical properties did not
substantially decreased after the seven reprocessing cycles despite degradation

of both constituents. A very good mechanical stability was observed during the
first five cycles, then a slow but limited loss of properties was observed.
During reprocessing, chain scission and fiber breakup were reported, but they
did not seem to significantly modify the mechanical properties of the recycled
composites. The strength at break remained stable over the seven cycles, while
the elongation at break gradually increased. The authors also observed the
presence of two simultaneous phenomena. During the first five cycles, a slight
decrease of Young’s modulus and strength at break was observed, indicating
chain scission. However, for the last two cycles, both increased suggesting that
a crosslinking mechanism became dominant. Newtonian viscosities also
decreased progressively with recycling due to chain scission and thermo-
oxidation of the matrix induced by the high shear stresses generated during
processing. Similar to other studies, fiber dispersion in the matrix was
improved during the first cycles, while fiber breakup essentially occurred
during the first cycle and for larger wood particles [21, 39, 45]. Moreover, the
fiber length was the most affected parameter by breakup, while fiber diameter
only slightly decreased with recycling. The number average molecular weight
decreased by 34% for the larger fibers and by 16% for the smaller ones
showing an effect of the fiber dimensions on the thermo-mechanical
degradation sustained by the polymer matrix. This was attributed to the
synergistic effect of wood fibers on the shear stresses generated in the polymer
matrix during processing. On the contrary, the weight average molecular
weight remained stable for both fiber sizes for all the generations considered
indicating that chain scission mainly occurred at the extremities of the polymer
chains, rather than randomly.
Beg and Pickering observed the effect of thermo-mechanical degradation
on the physical and mechanical properties of radiata pine/polypropylene
composites during reprocessing [43]. The wood fiber content was set to 40 and
50 wt.% fiber with 4 wt.% of maleated polypropylene as a coupling agent.
Eight reprocessing cycles were conducted. As for Waltz et al., they reported a
slight reduction of impact strength, tensile modulus and strength with
recycling [47]. In agreement with several works, they also reported an increase
of the tensile elongation at break and crystallinity [38, 39, 46]. At the end of
the recycling process, a decrease of tensile strength (25%) was observed for
the composites at 40 wt.% of fiber, while the Young’s modulus decreased by
17% after eight reprocessing cycles. For the composite at 50 wt.% of fiber,
both parameters increased (up to 14% for tensile strength and up to 33% for
Young’s modulus over the first two cycles) due to better fiber mixing and
dispersion during reprocessing, but then decreased after the third cycle due to
chain scission and fiber degradation. During reprocessing, both improvement
of fiber dispersion and fiber degradation occurred simultaneously. For lower
recycling cycles, the dispersion effect was predominant leading to an increase
of the properties, while for higher cycle numbers, fiber degradation prevailed
and the performances decreased. At 40 wt.% of fiber, flexural strength,
flexural modulus and impact strength were found to decrease by 30%, 21%
and 48% respectively, after eight cycles. Interfacial bonding was found to
improve with recycling, while the average fiber length was reduced by 84%
after eight reprocessing cycles. Moreover, the fiber length distribution became
narrower and was shifted towards shorter fiber lengths, confirming that fiber
breakup occurred. Contrary to most previous studies, significant degradation
of the properties was observed in this study [23, 39, 45, 46].
Augier et al. studied the influence of wood fiber addition on the
mechanical and thermal properties of mechanically recycled PVC/wood
composites [48]. Twenty extrusion-milling cycles were performed on
industrial waste and the results were quite different from the trends seen
before. Up to the fifth cycle, the mechanical and thermal properties of the
composites remained constant, but after 10 cycles, a degradation of the
molecular structure was noticed, while the other mechanical properties did not
change significantly. However, the flexural strength increased due to better
fiber dispersion in the matrix. For the PVC alone, an important increase of the
impact strength was observed after 20 cycles, as well as no degradation of the
molecular structure. The different behaviors between the composite and the
neat matrix was explained by the effect of the fibres which accelerated the
matrix degradation, as well as crosslinking reactions. However, the presence
of fibers also increased the composite stability and the mechanical properties
of the composite. A significant decrease in the fibre dimensions and a
widening of the size distribution were also noted between the first and the fifth
cycles, thus confirming fiber degradation during mechanical recycling. After
five cycles, no further degradation was noticed and no change was observed
for the aspect ratio over the whole recycling process. As for the molecular
distribution, a broadening and a clear shift of the peak to higher molecular
weights was observed for the composite after 20 cycles. This is in
contradiction with most studies [21, 38, 39, 46, 49] and was attributed to
dehydrochlorination, formation of unsaturation and crosslinking processes.
The authors concluded that no significant change occurred in the first five
cycles, thus making possible to recycle these composites up to five times, and
even more if the addition of virgin material was considered.
Kurniawana et al. evaluated the recyclability of wood/polypropylene

composites through three consecutive extrusion/injection molding cycles [49].
Different wood content up to 50 wt.% were tested and 3 wt.% of maleic
anhydride polypropylene was added to the formulation as a coupling agent.
They noticed that recycling did not decrease the properties of the composites,
but increased them instead. Reprocessing appeared to increase the composite
strength until the third cycle, and then remained stable for the composites
without coupling agent, but continued to increase for the composites
containing MAPP. Mechanical recycling also led to an increase of the tensile
modulus and elongation at yield for all composites, but composites with
coupling agent presented higher values. As for Twite Kabamba et al., they
found that the melting point did not change, while the crystallinity increased
with recycling [39]. Those trends were attributed to chain scission and to the
nucleating ability of wood flour. However, the authors expected this trend to
reach a saturation after a number of cycles, as reported by Beg and Pickering
[23]. Finally, this study showed quite unexpected results compared to other
studies [23, 39, 38, 46, 49], but led the authors to the same conclusion that, up
to five cycles, mechanical recycling did not substantially decreased the
performances and properties of the recycled composites.
Maldas et al. studied the mechanical and dimensional stability of
polystyrene/aspen composites after three mechanical recycling under extreme
conditions (exposure to very high and low temperatures, to water and boiling
water [2]. Up to date, this is the only work considering these conditions. They
also considered the influence of coupling agent addition (3% polymethylene
polyphenyl isocyanate or PMPI), as well as the use of fiber coating with a
polymer (10 wt.%) and isocyanate (8 wt.%). They noticed that the mechanical
properties and dimensional stability of the recycled composites did not
substantially change, even after exposure to extreme conditions. Moreover,
they noted that the composites with fiber treatment presented better properties
than the original composites, except after the treatment in boiling and very
cold water. Tensile strength decreased during the first two cycles, and then,
slightly increased. On the contrary, elongation and energy at break increased
during the first two cycles, and then decreased. The modulus did not show any
significant variation with recycling.
Åkesson et al. studied the mechanical recycling of high-density
polyethylene (HDPE)/wood fiber composites [50]. The composites were
subjected to seven consecutive injection molding cycles, and some of them
were also submitted to accelerated thermo-oxidative ageing. The results,
including tensile, flexural and impact properties, thermogravimetric analysis
(TGA), differential scanning calorimetry (DSC), FTIR spectroscopy and

morphological measurements, showed that the performances were not much
affected by the ageing conditions or the repeated injection moulding cycles.
Tensile strength showed a slight continuous decrease with recycling, dropping
by 9% between the first and the seventh cycle, and flexural strength followed a
very similar trend with a total decrease of 7%. No significant change was
observed for the elastic modulus, as well as for flexural modulus. On the
contrary, the elongation at break sharply increased with recycling, from 4.5%
for the original composite up to 7.8% after the seventh reprocessing cycle
(73% increase). This was attributed to lower fiber length which dropped by
64% during the whole recycling process. The results of the aged sample were
very similar to the unaged sample with less than 5% difference for all the
parameters studied. FTIR spectra did not show any significant degradation.
These results indicated that these composites were not much affected by
accelerated thermo-oxidative degradation and could be recycled quite well.
Cellulose
Graupner et al. focused on the behavior of polylactide (PLA) composites

during mechanical recycling by injection molding [51]. The PLA was
reinforced with 30 wt.% of regenerated cellulose fibres (lyocell) of variable
fineness and was subjected to three reprocessing cycles. The first reprocessing
cycle showed some degradation as the tensile strength decreased by 28%. This
degradation continued through the other cycles as the composite lost again
22% of its tensile strength during the second cycle, and 5% more during the
third cycle, thus totalizing a global decrease of 46% after three reprocessing
cycles. The fiber aspect ratio also decreased during reprocessing, leading to
lower properties. It was also shown that fibres with larger diameters were
more sensitive to breakup than finer fibers, thus leading to a lower decrease in
tensile strength for composites reinforced with finer fibres. It was then
concluded that the fiber dimensions, and especially their aspect ratio, strongly
affect the loss of mechanical properties during reprocessing. The properties of
a virgin composite with the same fiber length distribution were also measured
and compared to those of the recycled composites. The results indicated that
the fibers and the matrix were both damaged during reprocessing.
Fonsecca-Valero studied the mechanical recycling of high density
polyethylene/hardwood cellulose composites by injection molding [44].
Various fiber contents were tested between 10 and 48 wt.%, as well as two
different MAPE coupling agents with different molecular weights and maleic
anhydride (MA) contents. Five successive injection molding cycles were
conducted with no addition of virgin materials. The results showed that tensile
strength increased with recycling and, as this parameters appeared to be the
most sensitive to fiber breakup, this effect was higher for lower fiber
concentrations. The coupling agent with the higher Mw and MA contents was
found to be the most stable towards mechanical recycling, while the all others
seemed to lose their compatibilizing effect after a few cycles due to the high
degradation undergone during reprocessing. Recycled composites also
presented a higher relative melt flow index (MFI) compared to virgin ones,
and this effect increased as fiber content increased. ATR-FTIR, SEM and
AFM analyses of both composites showed that important fiber breakup took
place in the composite during recycling, thus explaining the properties
variations and the mechanical behavior observed. Moreover, the carbonyl
index revealed that the fibers underwent more thermo-oxidation than the
matrix. Due to this important degradation, the beneficial effect of increasing
the fiber content on the tensile strength of the composites was much lower
when increasing the number of reprocessing cycle, as this parameter appeared
as the most sensitive to fiber geometry. Elongation at break also increased
substantially with recycling, showing a relative increase of 258%, 128% and
7% for the composites containing 10, 25 and 40 wt.% of cellulose fibers,
respectively. With coupling agents, these relative gains were 216%, 33% and
2% for the low Mw and MA content coupling agent, while being 221%, 59%
and 15% for the higher Mw and MA content coupling agent.
Flax
Moran et al. studied the effect of multiple recycling on the mechanical

properties of polypropylene/flax composites without coupling agent [38]. The
flax concentration was set at 20 wt.% and five reprocessing cycle under
constant conditions were carried out to determine the thermo-mechanical
degradation undergone by the constituents during extrusion cycles, and to
evaluate the feasibility and viability of recycling these composites. The
mechanical properties were found to increase during the first three cycles due
to better mixing and dispersion of the fiber. But decreases were observed for
the next two cycles due to chain scissions and fiber degradation leading also to
darkening. Similarly to Twite Kabamba et al., they also observed that higher
molecular weight chains were more affected by chain scission [39]. They also
noted that most of the fiber degradation occurred during the first reprocessing
cycle. No further significant degradation was seen during the next cycles.
However, contrary to Twite Kabamba et al., they did not observe a
modification of the polymer crystallinity over the whole recycling process.
The recycled composites presented good impact performances and high elastic
modulus in both flexural and tensile tests. The tensile and flexural strengths
were lower than the matrix, but this difference decreased with recycling. The
authors concluded that, even if the recycled material suffered from chain
scission, fiber breakup and global degradation, the mechanical properties were
not significantly affected, and the material could still be used for industrial
applications. For this reason, they concluded that the main parameters
affecting the mechanical properties were the mixing quality and the fiber
modulus, instead of fiber geometry. Arbelaiz et al. also studied the mechanical
properties of polypropylene/flax composites after recycling [52]. The results
revealed that the tensile properties were stable for the first three cycles before
showing a slight decrease after the fourth cycle, with and without coupling
agent. This led the authors to the conclusion that PP/flax composites could be
mechanical recycled up to five cycles with no critical loss of properties.
Le Duigou et al. studied the mechanical recycling of poly(L-lactide)
(PLLA)/flax composites, with 20 and 30 wt.% of fibers [53]. The composites
were subjected to six successive injection molding cycles after an initial
compounding in a single-screw extruder. Young’s modulus appeared to remain
stable until the third cycle without any addition of virgin material. However,
reprocessing by injection molding seemed to lower the polymer molecular
weight and glass transition temperature, as well as the fiber length and aspect
ratio (L/D). It also induced fiber bundle breakup. On the contrary, the degree
of crystallinity showed a slight increase with recycling. These variations also
seemed to improve at higher fiber content, indicating that the latter
significantly influenced the material degradation and the resulting properties
by increasing the strain concentrations and the shear rate during processing.
Fiber concentration also strongly influenced the polymer degradation as the
average molecular weight decreased by 83% and 94% at 20 and 30 wt.% of
fibers respectively over the whole recycling process. The glass transition
temperature Tg decreased by 6°C at 20 wt.% of fiber and by 20°C at 30 wt.%
after seven injection molding cycles. All these changes were attributed to a
higher mobility of the polymer chains due to both chain scission in the
polymer and reduction of fiber dimensions. Once again, fiber breakup
essentially occurred during the first recycling cycles and was attributed to the
shear stresses generated in the processing equipment, which is a function of
temperature, pressure and screw speed. The fibers sustained a decrease in

length of about 20% during the extrusion step, and about 70% after both the
extrusion and injection molding steps. Contrary to most studies, after the third
cycle and regardless of the fiber content, the authors reported that both stress
and strain at break decreased with reprocessing [2, 38, 43, 44, 46, 49, 50]. The
Newtonian viscosity plateau also significantly decreased with increasing cycle
number, indicating an important degradation of the PLLA matrix, thus
confirming that chain scission and fiber breakup occurred.
Sisal and Hemp
In several works, Bourmaud and Baley studied the effect of reprocessing

on the properties of polypropylene/vegetal fiber composites [23, 41]. First,
they observed the effect of reprocessing on the rigidity of polypropylene/sisal
and polypropylene/hemp composites at 30 wt.% [41]. Once again, the results
indicated that the tensile modulus of the composites was not much affected by
mechanical recycling. A slow decrease (10%) of the modulus was observed for
the PP/sisal composites, while the modulus of PP/hemp composites did not
change significantly. This difference was attributed to the different aspect ratio
of both fibers which decrease differently with reprocessing. They finally
concluded that these composites exhibited good recyclability. In their other
work, they compared the thermal and mechanical properties of recycled
polypropylene composites reinforced with hemp, sisal and glass fibers [23].
The results showed that after seven reprocessing cycles, Young’s modulus
decreased by less than 1% and by 10% for PP/hemp and PP/sisal composites
respectively, while decreasing by 40% for the glass fibers. Tensile strength
showed no significant variation over the whole recycling process for PP/hemp
composite, while it decreased by 17% in seven cycles for the PP/sisal and by
52% for PP/glass fiber composites. On the contrary, the elongation at break
increased with recycling, reaching 22% for polypropylene/hemp composites,
28% for polypropylene/sisal composites, and up to 34% for glass fiber
composites. This parameter was strongly linked to interfacial bonding.
Similarly to most studies on this subject, the fiber length was shown to
significantly decrease during the injection molding process and that the natural
fiber bundles were separated by mechanical recycling [38, 39, 53]. However,
the glass fiber only shortened in length, while the natural fibers were affected
in both length and diameter, thus lowering the aspect ratio reduction. The
decrease of fiber length was also higher for natural fiber composites (72% and
77% for hemp and sisal respectively) than for glass fiber (57%). This could
explained the different mechanical behavior between glass fiber- and natural
fiber-reinforced composites, as both fiber length and aspect ratio were
considered as key factors for the rigidity and resistance of a composites.
Moreover, the adhesion between the fibers and the matrix seemed to weaken
with recycling. An important decrease of the interface adhesion was noted
with recycling as fiber debonding and pulling out were observed after the
seven cycles, indicating that the fiber surface treatment was no longer active
after a few reprocessing cycles. The analyses also revealed chain scission
leading to higher ductility, as well as a reduction of viscosity and impact
properties. The results indicated that chain scission preferentially occurred in
the macromolecule center of the highest molecular weight chains. The
Newtonian viscosity plateau also decreased by 57% for the glass fiber
composites and by 65% for the natural fibers composites after six reprocessing
cycles. This drop was related to chain scission lowering the molecular weight,
as well as fiber size reduction. The crystallinity and crystallization temperature
Tc of the natural fibers composites increased with recycling. This was
attributed to the nucleating effect of natural fibers coupled with the higher
mobility of the shortened fibers. The authors finally concluded that PP/vegetal
fibers could be mechanically recycled without any significant properties loss,
while PP/glass fiber composites presented a significant loss of interfacial
adhesion and were not well suited for mechanical recycling without any
preliminary interfacial treatment. The use of a coupling agent was shown to
improvement interfacial adhesion and the mechanical properties were more
stable, but this effect tended to disappear after a few cycles.
Rice Hulls and Kenaf
Srebrenkoska et al. investigated the mechanical recycling of

polypropylene/rice hulls and polypropylene/kenaf composites [54]. The
reinforcement content was set to 30 wt.% and maleic anhydride-grafted PP
was used as a coupling agent to improve the interfacial adhesion in both
composites. The authors reported that the recycling process did not
significantly influenced the mechanical and thermal properties of both
composites, but the composites reinforced with rice hulls were more sensitive
to recycling than the composites reinforced with kenaf fibers: the flexural
strength decreased by 5% for PP/kenaf composites and by 10% for PP/rice
hulls composites. The flexural modulus increased by 20% after recycling for
PP/kenaf composites, while it remained stable for PP/rice hulls composites.

The thermal stability of both composites was only slightly affected by
recycling, and the matrix/fiber interface appeared to be slightly improved.
They noted that the composites properties were very similar to the virgin
composites and concluded that these recycled composites could find suitable
applications in the current market.
Nettle
Steuernagel studied the mechanical recycling of nettle reinforced

polypropylene and compared the results with those obtained with glass fiber
composites [33]. A focus was made on the evolution of stiffness and strength,
as well as fiber dimensions variations. The compounds were produced by a
twin-screw extruder with and without MAPP as a coupling agent. The
resulting composites were then pelletized to be processed by injection molding
up to three times to simulate the mechanical recycling. The results showed a
loss of stiffness with recycling for both glass fiber composites and nettle fiber
composites without coupling agent. The tensile strength of natural fibers
composites decreased slightly, with a maximum drop of 13% for the material
without coupling agent, while it decreased more strongly (about 25%) for the
glass fiber composite. The composite with MAPP showed relatively stable
performances towards recycling. The melt viscosity decreased with recycling
for all composites studied.
Conclusion
Although natural fiber composites production is very well known, the

recycling of these composites is much less investigated [39-54]. The recycling
of this material appears more complex than neat polymer recycling as both the
fiber and matrix can degrade during the recycling process [50]. Moreover,
fiber addition modifies the properties and behavior of the polymer by creating
two new interactions: polymer-filler and filler-filler interactions compared to
polymer-polymer interactions in the neat matrix. In general, fiber addition
decreases the elongation properties while increasing brittleness [44]. However,
with successive recycling cycles, the fibers are shortened and lose their
reinforcing effect, leading to a lower influence on the composite elongation
[50].
Paukszta and Borysiak published a review considering the influence of

processing conditions on the structure and properties of natural fibers
composites [55]. They used the methods available for recycling thermoplastics
composites reinforced with renewable lignocellulosic materials and noted that
mechanical recycling modified the components properties, affected the
coupling agent when used, and changed the interactions between the
components. They also reported that the mechanisms taking place during
composite recycling are similar to those taking place during polymer
recycling. They also reported that, during grinding and reprocessing, the fibers
were often shortened and underwent internal and external fibrillation.
Simultaneously, their dispersion was improved due to a better mixing of the
components after the first few recycling cycles. The interface between the
fibers and the matrix also appeared to degrade with recycling, while the
coupling agent seemed to lose its efficiency after a few cycles.
Soroudi and Jakubowicz presented a review on the mechanical recycling
of bioplastics, their blends and biocomposites [31]. However, due to their
similarity to petroleum-based polyolefins, bio-derived polyolefins were not
considered in their work which focused on the different mechanical recycling
methods, their advantages and drawbacks, as well as the potential of recycled
materials. The authors made the assumption that biocomposites have a lower
recyclability than neat biopolymers due to the higher sensitivity of their
reinforcements to thermo-mechanical degradation, making composting the
best recycling method. However, the different studies on mechanical recycling
of such composites showed that these materials could usually be mechanically
recycled for several cycles without any critical loss of performances. Their
mechanical properties often remained stable during those cycles, and the
fiber/matrix interface was sometimes improved, even if some degradation
occurred to their components such as fiber length and molecular weight.
Table 2 summarizes the main results of the studies reported in this
chapter. These works seem to confirm the conclusions reported by Soroudii et
al., even if some additional observations can be made [31]. First, it was
expected that fiber incorporation increases the material viscosity and the shear
stresses generated in the extruder, leading to higher mechanical degradation, as
well as higher chain scission and fiber breakup, and thus to lower properties.
Then, most of them concluded that fiber length decreases mainly during the
first processing cycle, right after the addition of fibers to the molten matrix,
and to an extent that depends on the processing conditions [39, 40, 56, 57].
Finally, they all agree that, under specific conditions, the recycling of
thermoplastics composites based on natural fibers can be done up to five or ten
cycles without any critical loss of the final properties and performances.
However, the effect of recycling on the mechanical, molecular and thermal
properties of thermoplastics composites based on natural fibers remains quite
uncertain as various results and conclusions, sometime contradictory, were
reported by the different studies. Depending on the types and concentrations of
the components used, the presence of a coupling agent, the processing
parameters or the environmental conditions, several studies reported the
stability of mechanical and thermal properties of their composites during
recycling, while others reported positive or negative variations of some
properties like tensile modulus and strength during reprocessing. In this case,
more work should be done to evaluate more precisely the effect of these
factors on the composite behavior towards recycling.
Table 2. Overview of the investigations published on the

mechanical recycling of natural organic fillers reinforced
composites with their main parameters
Authors Matrix Reinforcement Fiber Coupling Total

content agent number of
(wt.%) cycles
Twite Kabamba et al. [39] LDPE Wood 15 - 10
Wang et al. [45] PE Wood 30 HDPE- 1
AA
Petchwattana et al. [21] PVC Wood 20 - 7
Soccalingame [46] PP Wood 20 MAPP 7
Beg and Pickering [43] PP Wood 40-50 MAPP 8
Augier et al. [48] PVC Wood 40 - 20
Kurniawana et al. [49] PP Wood ≤ 50 MAPP 3
Maldas et al. [2] PS Wood 5-35 PMPI 3
Åkesson et al. [50] HDPE Wood 10 - 7
Graupner et al. [51] PLA Cellulose 30 - 3
Fonsecca-Valero [44] HDPE Cellulose 10-48 MAPE 5
Moran et al. [38] PP Flax 20 - 5
Arbelaiz et al. [52] PP Flax 1-20 MAPP 5
Le Duigou et al. [53] PLLA Flax 20-30 - 6
Bourmaud and Baley [41] PP Sisal, hemp 30 - 7
Bourmaud and Baley [23] PP Sisal, hemp, glass 30 MAPP 7
Srebrenkoska et al. [54] PP Rice hulls, kenaf 30 MAPP 2
Steuernagel [33] PP Nettle N/A MAPP 3
Moreover, even if various studies exist on the subject, several aspects still
remain unknown. Some works investigated the mechanical and morphological
properties of natural fiber composites subjected to five or ten recycling cycles
[21, 23, 38-42, 45, 56-57], but none of them consider more intensive (higher
number of cycles) or open-loop recycling, and not all of them agree on the
behavior of the composites during recycling [21, 23, 38-42, 45, 56, 57].
Furthermore, they do not consider the effect of ageing and contamination
during its lifetime, as well as the cumulative degradation effect of multiple
recycling [50]. Furthermore, very few studies considered high fiber contents
(above 30 wt.%) [50]. A single study considered the effect of recycling under
extreme conditions, subjecting the composite to different temperature and
moisture conditions. None of the works available focused on foamed
composites and some work should be done on these aspects.
RECYCLING OF THERMOPLASTIC COMPOSITES

REINFORCED WITH INORGANIC FILLERS
Due to their high specific properties, carbon and glass fibers are among
the most widely used reinforcement in composites, not only for thermoplastics.
This is associated to their wide range of applications in structural and semi-
structural applications [58]. These composites represent important volumes of
material and thus need to be recycled. In the automotive field, these fibers are
still widely used due to their high specific properties. However, in Europe, the
End-of-Life Vehicles (ELV) directive 2000/53/EC requires that the producers
should reuse and recover a minimum of 95% of a vehicle by weight, and a
minimum of 85% of it should be reused or recycled, thus limiting to 5% the
amount of material that can be discarded [58, 59]. This directive led to
numerous studies on the recycling of such composites, in order to develop
efficient and economical recycling technologies. Moreover, as far as recycling
is concerned, thermoplastics seem to be the best matrix choice due to their
ability to be remelted and reshaped. The resulting composites present good
chemical and impact resistances, as well as design flexibility, good
reprocessability and recyclability [58]. Their recycling methods have thus been
widely studied as they are gaining more and more interest with new
environmental regulations [59]. The fiber length and concentration, interfacial
adhesion and polymer matrix properties are the key factors for the composites
final properties. Thus, mechanical recycling methods should be optimized to

best preserve these properties during reprocessing [58].
Glass
Ville et al. studied the mechanical recycling of glass fiber reinforced

polyamide 12 (PA12) in both laboratory and industrial scale twin-screw
extruders over a range of processing parameters [40]. Contrary to most studies
on this subject, which considered the recycled materials after its reprocessing,
they took samples at various locations along the extruder screw. The results
showed that the fiber length decreased with recycling, leading to a significant
modification of the fiber length distribution. Moreover, an important part of
this breakup occurred at the early stages of mixing, right after the introduction
of the fibers in the extruder, near the feeder. However, some size reductions
also occurred further down the screw. After the first breakup right after the
fiber introduction, the fiber length was also reduced after the first mixing
element, and then continuously decreased until the end of the screw. Finally, a
last sharp fiber size reduction was observed when the composite came through
the die. Then, when the fiber reached a critical length, breakup was more
difficult and the influence of the processing parameters was more limited. The
size distributions shifted towards smaller values and became narrower. The
extent of degradation varied significantly with the processing parameters and
equipment used. It increased with screw speed and residence time, but
decreased with feed rate. Surprisingly, it was higher in the small laboratory
scale extruder than in the industrial one. The authors reported that various
mechanisms took place during fiber breakup such as individual breakage,
bundle breakup into individual fibers by rupture and erosion, by buckling and
by fiber-fiber interactions. They finally concluded that, due to their important
influence on the composites properties, reprocessing conditions should be very
carefully chosen to limit the loss of performances of recycled composites.
Corvaglia et al. proposed a procedure for the secondary mechanical
recycling of polypropylene (PP)-based sandwich panels into short-fiber
composites and studied the influence of processing parameters and fiber
content on the mechanical and physical properties of the recycled material
[59]. The material was made of a polypropylene foam core with
polypropylene/continuous glass fibers laminate skins. This type of recycling
was qualified as “secondary” as the continuous glass fibers were transformed
into short glass fibers during reprocessing. To be recycled, this material was
ground in a mill and then remelted by melt mixing or single-screw extrusion.

The samples for the mechanical and morphological characterizations were
then made by injection molding. The results showed that the tensile strength
was more sensitive to fiber length reduction than tensile modulus. As in Ville
et al., the residence time seemed to strongly influence the thermo-mechanical
degradation undergone by the composite due to the high temperatures and
mechanical stresses generated in the processing equipment [40]. The reduction
of the molecular weight and fiber size induced by this degradation increased
with residence time, depending on the screw speed. However, if the residence
time should be as low as possible to limit degradation, it should also be high
enough to get good fiber dispersion and proper fiber wetting by the matrix.
According to the mechanical results, a maximum screw speed of 40 rpm was
proposed for the processing of such materials. Similarly, the processing
temperature should be set very carefully. It should be low enough as to limit
thermo-degradation of the composite, but high enough to limit the matrix
viscosity which is known to induce high shear stresses and fiber degradation.
This was confirmed by morphological analyses showing a higher amount of
short fibers at lower temperature. The authors also reported that the residence
time should be kept under 15 minutes and processing temperatures should be
set slightly higher than 200°C to avoid excessive degradation for both
components, and preserve the mechanical properties of the recycled material.
They also indicated that, according to their mechanical characterizations, the
internal mixer led to lower matrix degradation compared to the extruder and
should be favored for the reprocessing of these composites.
Cornier-Ríos et al. studied the mechanical recycling of polyethylene
terephthalate (PET)/glass composites [60]. The glass filler content was set to
15 wt.% and six reprocessing cycles were performed for each material and
four recycled content were considered (0, 25, 50, and 100%) in the virgin
material. The thermal properties appeared to remain stable regardless of the
number of generation and recycle ratio, whereas the mechanical properties
slightly decreased (maximum 6%) with recycling and recycle ratio due to fiber
length reduction. The tensile strength showed a reduction of about 3% per
cycle for the totally recycled material, and also decreased with increasing
recycling ratio and recycling generations for samples with mixed generations.
The decrease was however lower for lower recycle ratio as the addition of
virgin materials limited the loss of properties because of its lower degradation
state. On the contrary, elongation at break and elastic modulus did not show
any particular trend with mechanical recycling or recycle ratio. Moreover,
crystallinity remained stable regardless of the number of cycle or the recycled
material content. As the materials were easily melt processed at standard

temperatures, it was concluded that no crosslinking occurred during recycling.
All these results indicated that the recycling of glass fiber reinforced PET
could be done without any significant loss of properties.
Kemmochi et al. focused on the effect of closed-loop recycling on the
properties of fiber reinforced thermoplastics (FRTP), and especially on the
fiber length inside the composites [61]. Different types of glass reinforcement
were considered: powder, short, long and continuous fibers. They simulated
the lifecycle by artificial accelerated ageing coupled with natural weathering.
They reported that the FRTP could be recycled in four main steps, from the
continuous fiber to the powder and then to energy recovery. First, the
continuous fiber reinforced thermoplastics became a long fiber reinforced
thermoplastics after one recycling cycle. Then, due to fiber degradation with
reprocessing, those long FRTP became short fiber reinforced thermoplastics.
After another cycle, they can finally be recycled as glass powder reinforced
thermoplastics. In the end, as the reinforcement effect lose its effect, they can
be energetically recycled or added to virgin materials. They indicated that each
step was equivalent to five to twenty years of normal product use. This work
also revealed that, in the case of short fiber composites, for a defined fiber
length, tensile and bending strengths were higher for recycled materials than
for virgin ones with similar fiber distribution. This was attributed to better
fiber orientation in the thermoplastic matrix. The bending rigidity also showed
no significant influence of the recycling for each category. On the contrary, the
mechanical strengths of these composites were only affected by fiber length
and orientation.
Carbon
A general review was also done on the different ways of mechanically

recycle glass mat reinforced thermoplastic (GMT) [58]. Thermoplastics
composites were shown to be easier to recycle due to their ability to be
remelted and reshaped multiple times. However, due to the necessary milling
step of mechanical recycling, they cannot be used as GMT and present a
certain loss of properties after this type of recycling. Glass mat reinforced
thermoplastic are long glass fibers composites usually made of two layers of
glass mat wetted by a polymer matrix and mostly supplied in sheet form as
semi-finished material. They are shaped by a subsequent molding process and
present high properties. After the milling process, the fibers become shorter
and cannot be used as long fibers anymore. The authors reported that three
ways exist to reuse this type of material. First, small parts can be remelted and
molded into a larger part. However, this method is difficult to implement as it
is hard to find enough parts of the same kind to be recycled. Otherwise, ground
GMT could be used as raw material chips to produce semi-finished sheets of
materials with the same processes again. Finally, they could also be used as
short glass fiber reinforcement in extruded or injection molded thermoplastics
composites.
Colussi et al. studied the mechanical recycling of polyamide 66
(PA66)/short carbon fibers by injection molding [62]. The fiber content was
set to 30 wt.%. As these composites were intended for automotive
applications, the materials were submitted to some artificial aging by
temperature, moisture and ultraviolet (UV) treatments before recycling. These
factors were chosen to simulate common outdoor conditions (moisture, rain
and sunlight) that the material will sustain in combination to mechanical
stresses during its lifecycle, and thus evaluate their effect on the composite
properties. It was found that aging led to lower tensile strength and Young’s
modulus, while the elongation properties slightly increased. These trends were
attributed to the photo- and thermo-oxidative degradations generated in the
polymer matrix by these three factors. Rheological tests revealed that the
complex viscosity decreased by 71% with both aging and recycling, while
unaged samples showed a much lower decrease (12%). This was attributed to
chain scission in the polymer matrix and confirmed the important effect of
ageing on the resulting properties of the recycled composites. The weight
average molecular weight (Mw) of the aged composites increased with
recycling, while the number average molecular weight (Mn) did not
significantly changed. As most of the properties loss occurred after the
artificial ageing, and that the mechanical, morphological and thermal
properties of the unaged composites remained constant with recycling, the
authors concluded that mechanical recycling did not have a significant effect
on the composite properties so it can be used for such composites without any
significant loss of performances.
Sarasua and Pouyet investigated the effect of mechanical recycling on the
mechanical and morphological properties of poly-ether-ether-ketone
(PEEK)/carbon composites [63]. The recycling process was performed by
injection molding and the short carbon fibers content was set at 10% and 30%.
It was found that both fibers and matrix degraded during recycling, while no
significant modification was observed at the fiber-matrix interface. The
morphological analyses revealed that the fiber length distribution was
narrowed and that the average fiber length decreased with recycling.
Moreover, the polymer molecular weight decreased by 28% over the ten
recycling cycles. The results obtained were also in agreement with the trends
reported in most previous studies [40, 59, 60]. Young’s modulus and tensile
strength decreased with recycling, showing a Young’s modulus reduction of
10% and 30% at 10% and 30% fibers respectively, and about 20% for their
tensile strength. Impact strength also decreased with recycling, showing a loss
of 39% in ten cycles for the composite with 10% fibers, and of 24% in ten
cycles at higher fiber content (30%). On the contrary, ductility increased with
reprocessing as elongation at break increased by 225% for the 10% carbon
fiber composites, and by 42% for the 30% carbon fiber reinforced PEEK. The
authors also reported that the failure mode was highly function of the fiber and
that recycling did not influence significantly the high deformation properties
of the composites. They concluded that important fiber damage occurred
during processing, but the properties were well enough conserved during the
recycling process to use these materials for specific applications.
Schinner et al. presented two methods for the mechanical recycling of
thermoplastics/carbon fiber composites [64]. The first one considered the
grinding of the thermoplastic composite waste and its incorporation as
reinforcement into compounds made of virgin materials. The results showed
that the properties evaluated were similar to those of the totally new materials,
thus leading the authors to conclude to the viability of this type of recycling.
The second method was based on the reprocessing and reshaping of the used
thermoplastics/carbon fiber composites. This second approached also revealed
that the properties were not significantly altered, thus confirming the viability
of this second method as a suitable mechanical recycling method for
thermoplastics/carbon fiber composites.
Talc
Wang et al. studied the behaviors of polypropylene/talc composites

subjected to mechanical recycling [65]. This study revealed that the
mechanical properties of these composites depended on the number of
recycling cycle, the reinforcement content and the reprocessing temperature.
Both the tensile modulus and the tensile strength at yield increased with
recycling due to lower fiber dimensions and increase of their aspect ratio. On
the contrary, the compression modulus decreased continuously with increasing
reprocessing. As for previous studies, the authors found that mechanical
recycling led to a reduction of the matrix molecular weights due to chain

scission and a crystallinity increase due to higher mobility of the shortened
fibers [40, 59, 63]. They also reported that the decomposition temperature
remained stable during recycling, allowing the reprocessing at similar
temperatures than for virgin materials.
Conclusion
In their review on mechanical recycling of polyolefin waste, Yin et al.

investigated inorganic filler reinforced composites [16]. They reported that this
type of filler generally improves the mechanical properties of materials, even
at small content and that the final properties of inorganic fillers reinforced
composites depends on the filler geometry and distribution, as well as on the
interfacial adhesion between the fillers and the matrix. Besides the commonly
known effects, they noted that every inorganic filler has its own specific effect
on the composite properties: clay was shown to improve thermal stability and
fire resistance, titanium dioxide had good antimicrobial properties as well as
ultra-violet (UV) protection, talc produced an effective nucleating effect, while
glass fiber led to high specific mechanical properties, low cost, lightweight
and resistance to chemicals. They also confirmed the trends seen in the
previous studies; i.e., an increase of both Young’s modulus and tensile
strength with recycling, fiber breakup, and lower elongation at break. Most of
the studies also concluded that the modifications of the properties during
reprocessing did not significantly changed the final performances of the
recycled materials and these materials could be successfully mechanically
recycled without any important loss of properties. All the works reported in
this chapter are summarized in Table 3.
As for natural organic reinforcement, even if several works were
published on this subject, different aspects still remain unknown. Once again,
mechanical properties have been widely studied, as well as fiber length and
molecular weight distributions [40, 59-63, 65]. Thermal properties were also
frequently studied [59, 60, 65]. However, very few of them focused on the
rheological and physical properties. Moreover, all of them considered less than
ten recycling cycles, as well as closed-loop recycling. Thus, long-term
recycling and open-loop recycling were not studied, as well as material
inhomogeneity, ageing during its life cycle and possible contamination were
also left out of this investigation. The same comment applies to the cumulative
long-term effect of mechanical recycling occurring after a high number of
reprocessing cycles. Once again, recycling under extreme conditions, high

reinforcement content, and recycling of foamed composites were also left out.
In order to completely understand the behavior under recycling, additional
work must be done on these different aspects.
Table 3. Overview of the works considering the mechanical recycling of

inorganic fillers reinforced composites with their main characteristics
Authors Matrix Reinforcement Fiber Number of

content cycles
(wt.%)
Ville et al., [40] PA12 Glass fiber 30 1
Corvaglia et al., [59] PP Glass fiber 40 1
Cornier-Ríos et al., [60] PET Glass fiber 15 5
Kemmochi et al., [61] Nylon 6 Glass - 4
Colussi et al.,[62] PA66 Carbon 30 1
Sarasua and Pouyet [63] PEEK Carbon 10-30 10
Wang et al., [65] PP Talc 10-20 6
LIMITATIONS AND SOLUTIONS

As seen in the previous parts, for both organic and inorganic fillers, some
problems are limiting the use of mechanical recycling for thermoplastic
composites. First, for natural fiber reinforced thermoplastics, the main
limitation to mechanical recycling is the sensitivity of these fibers towards
high temperatures. As seen previously, natural fibers often present degradation
temperature around 150 to 200°C, which is lower than most of the
thermoplastics meting points, and below usual processing temperature [45, 66,
67]. According to Prasanth et al., natural fibers are usually composed of
cellulose, hemicellulose, and lignin, and each of these components presents its
own behavior and limits with respect to thermal degradation, which usually
occurs in two consecutive steps [67]. First, lignin degrades at low temperature
(80-180°C), and then cellulose degrades at much higher temperatures (280-
380°C). In order to produce and recycle such composites, processing
temperatures should be higher than the melting point of the polymer matrix,
and are thus higher than the onset thermal degradation temperature of most
natural fibers. This lead to some additional degradation and darkening of the
fiber during reprocessing (mechanical recycling), leading to properties losses,
as well as to volatile emissions producing undesired foaming [67]. As this
additional thermal degradation is added to the thermo-mechanical degradation

generated by the shear stresses during processing, once the components of the
composites chosen, the only remaining way to limit the properties losses is to
limit the thermo-mechanical degradation induced by the shear stresses. This
can be done by an optimization of the processing parameters. The processing
temperatures and profiles should be high enough to favor good reinforcement
dispersion and to allow smooth processing, but not too high to limit thermal
degradation. Similarly, screw speed should be set high enough to maximize
homogenization and throughput and to minimize residence time, but not too
high as to limit the shear rate undergone by the composite during processing
[67-69]. Winandy at al. also reported that using waste newsprints instead of
natural fibers could lead to lower loss of performance due to their lower
degradation and higher stability under high temperature [70, 71]. The use of
counter-rotating screws also seems to decrease this loss of performance [71].
Prasanth et al. noted that some chemical fiber modifications could improve the
thermal stability of these fibers, and thus inhibit the performances losses
undergone by these composites [67].
Another problem encountered for natural organic fiber reinforced
thermoplastics is the loss of performance induced by moisture absorption. As
most natural organic fibers are hydrophilic, they absorb moisture during their
lifecycle and their reprocessing leading to swelling and to foaming of the
material during reprocessing, if no previous drying treatment was applied [67].
Moisture is also absorbed in the fibers by capillary forces [67]. This modifies
the fiber dimensions and density, affecting the physical, chemical, mechanical
and morphological properties of the recycled composites [71]. But foaming
can be avoid by preliminary fiber drying before recycling, but swelling will
still occur during the use phase. The absorption of moisture by natural fibers is
usually reduced by chemical modification of the fiber surface [67]. The use of
coupling agents, such as polyolefins grafted with maleic anhydride, seems to
counter this effect as the anhydride groups react with the hydroxyl groups of
natural fibers decreasing the number of sites where water molecules can bond
[71].
The second and most important challenge for mechanical recycling of
thermoplastics composites is inhomogeneity of the waste stream. This appears
to limit the final performances of the recycled materials by many aspects.
First, most thermoplastics are immiscible and incompatible, forming different
phases when mixed and leading to lower mechanical properties of the final
recycled composites [14, 19, 72, 73]. This problem requires the use of
stabilizers and compatibilizers at high content to improve the cohesion
between the phases, which is hardly economically feasible, or the use of

mechano-chemical methods leading to substantial degradation of the
composite, and especially of the reinforcement [14, 16]. Mechano-chemical
processes are simple, fast, economic and ecofriendly. However, important
degradation of the composite’s reinforcement limits its used for this type of
materials. Mechano-chemical methods are based on the pulverization of the
polymer materials to obtain reactive blending without using any additives [16,
19, 71, 73, 74]. They use a wide range of reactions based on various
solicitations such as impact, shear, elongation, fracture and compression,
generally generating chains scission in the polymer matrix, but also induce the
formation of free radicals which are from different molecular species and react
with each other to induce coupling and crosslinking [16]. Mechanical milling
is the copulverization of materials with a high-energy ball milling process to
get the materials in a powder form. This increases the blending degree, but
also leads to a major reduction of the material size and, in the case of
composite materials, to major fiber breakup. Some methods also use liquid and
cryogenic gases to make the pulverization step easier [16, 71, 73, 74]. The use
of stabilizers and compatibilizers is possible, but adds extra costs to the
recycling process because these agents have to be renewed regularly as they
seem to lose their efficiencies after a few regenerations [19]. However, the
addition of such agents leads to better properties, and slightly decreases the
need for sorting before reprocessing. Another way of improving the interfacial
adhesion is via irradiation with electrons to generate free radicals in the
materials increasing reactivity and adhesion between the phases [31].
Due a wide variety of composite waste sources, diversity in grade and
composition even after sorting, and inconsistencies of the waste stream lead to
a significant variability of the properties of the final materials [71, 74, 34]. The
best solution would be to implement better sorting methods, but this is not
economically and technically feasible to reach a perfect sorting of each
composite by matrix, fiber and additive types, especially considering that
composites are multi-components materials. Some differences in the
composition will always remain, especially regarding the additives contents in
a materials. Due to this diversity of sources, but also in life conditions, the
composites present different properties and behaviors, depending on their
nature, composition, and degradation extent [19]. The most important problem
is the melting point. If it seem very uncommon that recycling modify the
melting point of the polymer matrix, as no significant difference could be
noticed in most recycling studies, in some specific cases, if the material
contains some impurities and additives, the melting point can be changed [19].
In this case, the blends will have a distribution of melting temperatures instead
of a unique melting point, which is an issue for recycling as the matrices with
low melting point will flow faster than the ones with high melting point,
leading to a heterogeneous material induced by partial melting or thermal
degradation.
Postconsumer materials also usually contain contaminants that could
further decrease the final properties of the recycled composites, or induce
toxicity, unpleasant odors or discoloration, which highly reduce the
attractiveness of the final recycled products [19, 22, 24, 34, 71, 72, 74, 75].
Contamination can be associated to the environment of the product, or to its
assembly with other parts. Common sources of contamination are inserts,
screws, bolts, staples, nails, metals parts, labels, adhesives, paints, varnish and
other finish products, printing products and residues [69, 71, 72]. Once again,
compatibilizers can be used, but only for small volumes of contaminants [22,
24]. The best solution remains to improve the waste collection, separation and
cleaning steps, but adds extra costs and leads to several technical challenges.
Still today, these steps remain quite limited for composites [3, 34, 76].
However, these considerations should become easier in the future as a new
trend toward considering the dismounting, separation and recycling of the
materials progressively settle into the developers’ minds [31, 34, 76].
Deconstruction, separation of the various components and recycling are more
and more considered during the research and development phase of the
composites, as to modify and adjust the composite design to facilitate these
steps at the end of the composite lifecycle.
Another limitation for the use of mechanical recycling for composites is
the degradation undergone during its lifecycle, which is not taken into account
in most studies. This is also expected to decrease the stability and
performances of the final recycled material [16, 31, 72]. Various types of
degradation can occur during a composite lifecycle such as mechanical (shear
and elongation) stresses, oxidation, burning, chemical attacks, ultra-violet
(UV), biological and moisture degradations [16, 31, 67, 70, 72]. Besides
decreasing the final properties and performances, these degradation processes
can lead to partial reticulation of the polymer matrix, thus to higher viscosity
and the creation of stress concentration points making more difficult the
recycling due to the high pressure, energy and cost required, and the lower
final performances [3, 16, 22]. This degradation is unavoidable and is the
focus of most actual studies on thermoplastics composites recycling.
Finally, the lack of markets available for recycled composites also appear
as a limitation to the implementation and the spreading of the mechanical
recycled composites, as well as high costs [3]. If these composites are more
expensive than virgin ones, lower properties could be of interest for less
demanding applications, but the attractiveness of such materials is very limited
[3]. This is why more research should be done to improve these aspects as to
meet the quality requirements at low cost, as well as being able to fit into the
same applications than for virgin composites. All these limitations and their
potential solutions are summarized in Table 4.
Table 4. Main limitations and solutions for the mechanical recycling of

thermoplastic composites
Limitations Possible solutions

Sensitivity of natural fibers to Optimization of the processing parameters and
high temperatures material components
Sensitivity of natural fibers to Fiber drying and surface treatment
humidity
Inhomogeneity of the waste Improved sorting methods, use of compatibilizers,
streams mechano-chemistry
Contamination of the waste Improved sorting methods, design modifications for
better sorting and recycling, use of compatibilizers
Degradation during recycling Optimization of the processing parameters
Lack of market Final properties improvement, lowering of the costs
CONCLUSION
Although thermoplastics composites processing is very well known, the
behavior of these materials towards mechanical recycling is much less
understood. The recycling of these materials appears to be more complex than
neat polymer recycling as both the fiber and the matrix can degrade during
recycling. Furthermore, the presence of reinforcing particles modifies the
polymer behavior and properties (morphology, mechanics and rheology),
while simultaneously creating new interactions between the filler and the
matrix, as well as between the particles of the matrix [50]. However, most
degradation mechanisms are similar to those occurring during the recycling of
polymers [55]. The number average molecular weight Mn decreases with
recycling, while the weight average Mw remains mostly constant [23, 39, 46,
53]. Crystallinity increases with the number of cycles [23, 39, 43, 49, 53, 65]
and chain scission mostly occurs around the mid-point of longer molecules
[23, 38, 39], leading to a narrower molecular weight distribution.
However, most of the materials properties differ from the neat polymer
case: fiber addition leads to more degradation [38, 39, 48]. This degradation
depends on the filler type and size [23, 33, 41, 46, 51], as larger particles seem
to lead to higher degradation [46, 51] and more rigid fibers like glass fibers
seem more affected by recycling [23, 33]. Viscosity increases with filler
addition, but decreases faster due to particle breakup during recycling [23, 39,
46, 53, 62]. Similarly, elongation at break decreases sharply with filler
addition, but was found to increase with recycling as filler dimensions
decreases with successive recycling losing their reinforcing effect [23, 41, 44,
46, 49, 50, 62, 63]. Several works also pointed out that fiber breakup mainly
occurs during the first cycle, and that its extent depends on the processing
conditions [38-40, 53]. Thermo-mechanical degradation is generated in the
material by high temperature, high pressure and high shear stress. This
combination was found to depend on screw speed, residence time and feed rate
[38, 40, 44, 45, 59, 63]. These parameters should be carefully chosen to get the
best compromise between good homogeneity and low degradation. If the
temperature is too high, thermal degradation will be important and
homogeneity will be good due to the low viscosity of the melt, but if the
temperature is too low, poor homogeneity (filler distribution) occurs and
mechanical degradation is important due to high viscosity (stresses on the
particles) [59].
Several aspects related to composite recycling remain unknown. The
presence of crosslinking remains unclear [39, 46, 60]. Furthermore, if all the
authors agree to say that fiber length decreases due to fiber breakup, not all of
them agree about changes in the diameter and aspect ratio [38, 39, 40, 43, 44,
46, 50, 51, 53, 59, 63, 65]. Some works reported that fiber diameter remained
constant leading to a decreasing aspect ratio (L/D) [23, 39, 51, 53], while
others observed that diameter decreases and that the L/D was almost constant
[23, 38, 48]. Mechanical properties changes is also not clear. The majority of
works indicated that the tensile and flexural strengths slightly decreased with
recycling [23, 33, 43, 50, 61, 63], while others reported that these parameters
increased [44, 48, 49, 60-62] or even remained constant [41, 46]. Similar
results were obtained for impact strength [21, 63 77]. Also, the modulus was
found to increase in some works [49, 54, 60], while most studies reported no
significant variation [2, 21, 41, 50, 53, 61]. Nevertheless, all the literature
agrees to say that these variations were not important and that no significant
loss of performance resulted from the mechanical recycling of thermoplastics
composites [2, 21, 23, 38, 41, 45, 50, 52, 54, 62].
The behavior and the performances of thermoplastics composites have

been the subject of several studies and are well known. As the mechanical
properties do not vary significantly, it can be concluded that composites
present great aptitudes to short-term mechanical recycling. However, the
influence of recycling on the mechanical, molecular and thermal properties of
thermoplastics composites remains quite uncertain as various results and
conclusions were reported by the different studies. This is why more work
should be done as to evaluate more precisely the effect of these factors on the
composite behavior towards recycling. Moreover, most of these studies
considered a very low number of cycles, generally less than 10. Extensive
long-term recycling of thermoplastics composites has not been studied, and
additional work should be conducted on these aspects to fill this lack of
knowledge. Moreover, very few studies took into consideration extreme
conditions like moisture and temperature [2, 44] or very high filler contents
(above 30 wt.%), and none considered open-loop recycling. Most of them are
not taking into consideration the ageing and contamination of the material
during its lifetime, the inhomogeneity of the waste stream, as well as the
cumulative degradation effect of recycling occurring after a higher number of
reprocessing cycles. Finally, no work considered the mechanical recycling of
foamed thermoplastics composites as these materials are becoming very
popular in view of material savings and weight reduction. These aspects
should be the focus of more research.
Mechanical recycling of thermoplastics composites also reached some
limitations affecting their performances and their viability. The inhomogeneity
of the waste, as well as the possible contamination of the waste stream lead to
lower properties. Compatibilizers and filler surface treatments could be used to
limit the properties losses, but they were shown to lose their performances
with recycling [23, 44]. The sensitivity of natural fibers composites to high
temperatures and humidity also decrease their performances. Furthermore, the
thermo-mechanical degradation undergone by the composites and their
constituents during recycling also limit their performances and their possible
applications, leading to a lack of market for such composites. Partial solutions
exist and should be considered, but more research must be conducted in this
field to overcome these limitations and increase the performances of recycled
composites, as well as to increase the number of possible applications for such
composites.
ACKNOWLEDGMENTS
Financial support from the Natural Sciences and Engineering Research
Council of Canada (NSERC, Canada), the Research Center for High
Performance Polymer and Composite Systems (CREPEC, Quebec), the
Quebec Research Funds – Nature and Technologies (FQRNT, Quebec), the
Quebec Center for Functional Materials (CQMF, Quebec), the Research
Center for Advanced Materials (CERMA, Quebec), and the Renewable
Materials Research Centre (CRMR, Quebec) is acknowledged for this work.
REFERENCES
[1] Ruch, F., Cetim-Cermat. Composites et recyclage : état de l’art et
perspectives. 2016 [04/09/2016]. Available from: http://recyclage-
plastique.com/wp-content/uploads/2016/03/160316-Journ%C3%A9e-
Recyclage.pdf [Ruch, F., Cetim-Cermat. Composites and recycling: state
of the arts and perspectives [online]. 2016 [04/09/2016]. Available from:
http://recyclage-plastique.com/wp-content/uploads/2016/03/1603 16-
Journ%C3%A9e-Recyclage.pdf].
[2] Maldas, D. & Kokta, B.V. (1990). Effect of Recycling on the
Mechanical Properties of Wood Fiber-Polystyrene Composites. Part I:
Chemithermomechanical Pulp as a Reinforcing Filler. Polymer
Composites, 11, 77-83.
[3] Yang, Y.; Boom, R.; Irion, B.; Van Heerden, D.J.; Kuiper, P.; De Wit,
H. (2011). Recycling of composite materials, Chemical Engineering and
Processing, 51, 53-68.
[4] Rybicka, J.; Tiwari, A.; Leeke, G. A. (2016). Technology readiness level
assessment of composites recycling technologies. Journal of Cleaner
Production, 112, 1001-1012.
[5] Thomason, J. L.; Yang, L.; Meier, R. (2014). The properties of glass
fibres after conditioning at composite recycling temperatures.
Composites: Part A, 61, 201-208.
[6] PlasticsEurope. Plastics: Too valuable to be thrown away, Recovery,
recycling and resource conservation [online]. 2009 [04/09/2016].
Available from: http://www.plasticseurope.org/Document/plastics-too-
valuable-to-be-thrown-away.aspx.
[7] Li, J. & Sun, F.F. (2009). The Effect of Maleic Anhydride Graft on the
Interfacial Adhesion of Carbon Fiber Reinforced Thermoplastic
Polystyrene Composite. Journal of Composite Materials, 43, 2717-2725.
[8] Mahfoudh, A. (2013). Etude de la production et de la caractérisation de
composites bois-plastiques. Master Science Thesis, Université Laval,
Québec, Canada. [Mahfoudh, A. (2013). Study of the production and the
characterization of wood-plastic composites. Master Science Thesis,
Laval University, Québec, Canada].
[9] Witten, E; Kraus, T.; Kühne, M. Composites Market Report 2015:
Market developments, trends, outlook and challenges [online]. 2015
[04/09/2016]. Available from: http://www.eucia.eu/userfiles/files/
Composites_Market%20Report_2015.pdf.
[10] Migneault, S.; Koubaa, A.; Erchiqui, F.; Chaala, A.; Englund, K.;
Wolcott, M. P. (2009). Effects of processing method and fiber size on
the structure and properties of wood–plastic composites. Composites:
Part A, 40, 80-85.
[11] Kazemi, Y. (2013) Mechanical and morphological characterization of
wood plastic composites based on municipal plastic waste. Master
Science Thesis, Université Laval, Québec, Canada.
[12] Hamad, K.; Kaseem, M.; Deri, F. (2013). Recycling of waste from
polymer materials: An overview of the recent works. Polymer
Degradation and Stability, 98, 2801-2812.
[13] Buggy, M.; Farragher, L.; Madden, W. (1995). Recycling of composite
materials. Journal of Materials Processing Technology, 55, 448-456.
[14] Lu, N.; Swan, R. H.; Ferguson, I. (2011). Composition, structure, and
mechanical properties of hemp fiber reinforced composite with recycled
high-density polyethylene matrix. Journal of Composite Materials, 46,
1915-1924.
[15] Oliveux, G.; Dandy, L. O.; Leeke, G. A. (2015). Current status of
recycling of fibre reinforced polymers: Review of technologies, reuse
and resulting properties. Progress in Materials Science, 72, 61-99.
[16] Yin, S.; Tuladhar, R.; Shi, F.; Shanks, R. A.; Combe, M.; Collister T.
(2015). Mechanical Reprocessing of Polyolefin Waste: A Review.
Polymer Engineering & Science, 55, 2899-2909.
[17] PlasticsEurope. Plastics - the Facts 2015: An analysis of European
plastics production, demand and waste data [online]. 2009 [04/09/2016].
Available from: http://www.plasticseurope.org/documents/document/
20151216062602-plastics_the_facts_2015_final_30pages_
14122015.pdf.
[18] Verdolotti, L.; Iucolano, F.; Capasso, I.; Lavorgna, M.; Iannace, S.;
Liguori, B. (2014). Recycling and Recovery of PE-PP-PET-based Fiber
Polymeric Wastes as Aggregate Replacement in Lightweight Mortar:
Evaluation of Environmental Friendly Application. Environmental
Progress & Sustainable Energy, 33, 1445-1451.
[19] Kazemi Najafi, S. (2013). Use of recycled plastics in wood plastic
composites - A review. Waste Management, 33, 1898-1905.
[20] Hélène Gervais, Recyc-Québec. : Les plastiques : Fiches informatives
[online]. 2009 [04/09/2016]. Available from: https://www.recyc-
quebec.gouv.qc.ca/sites/default/files/documents/Fiche-info-platique.pdf
[Hélène Gervais, Recyc-Québec : Plastics : Information notes [online].
2009 [04/09/2016]. Available from: https://www.recyc-quebec.gouv.qc.
ca/sites/default/files/documents/Fiche-info-platique.pdf].
[21] Petchwattana, N.; Sirijutaratana Covavisaruch, S.; Sanetuntikul J.
(2012). Recycling of wood–plastic composites prepared from poly(vinyl
chloride) and wood flour. Construction and Building Materials, 28, 557-
560.
[22] Sjöqvist, M. & Boldizar, A. (2011). Molecular Modification and
Compatibilization of Collected Polyethylene. Journal of Polymers and
the Environment, 19, 335-340.
[23] Bourmaud, A. & Baley, C. (2007). Investigations on the recycling of
hemp and sisal fibre reinforced polypropylene composites. Polymer
[24] Strömberg, E. & Karlsson, S. (2009). The design of a test protocol to
model the degradation of polyolefins during recycling and service life.
Journal of Applied Polymer Science, 112, 1835-1844.
[25] Oblak, P.; Gonzalez-Gutierrez, J.; Zupancic, B.; Aulova, A.; Emri, I.
(2015). Processability and mechanical properties of extensively recycled
high density polyethylene. Polymer Degradation and Stability, 114, 133-
145.
[26] Jin, H.; Gonzalez-Gutierrez, J.; Oblak, P.; Zupancic, B.; Emri, I. (2013).
Effect of extensive recycling on flow properties of LDPE. Proceedings
of the 71st Annual Technical Conference & Exhibition, Cincinnati, Ohio,
USA.
[27] Jin, H.; Gonzalez-Gutierrez, J.; Oblak, P.; Zupancic, B.; Emri, I. (2012).
Effect of extensive recycling on flow properties of LDPE. Polymer
[28] Benoit, N.; González-Núñez, R.; Rodrigue, D. (2016). High density

polyethylene degradation followed by closed-loop recycling. Progress in
Rubber, Plastics & Recycling Technology (accepted).
[29] Conroy, A.; Halliwell, S.; Reynolds T. (2006). Composite recycling in
the construction industry. Composites: Part A, 37, 1216-1222.
[30] Papaspyrides, C. D.; Poulakis, J.G.; Arvanitopoulos, C. D. (1995).
Recycling of glass fiber reinforced thermo-plastic composites. I.
Ionomer and low density polyethylene based composites. Resources,
Conservation and Recycling, 14, 91-101.
[31] Soroudi, A. & Jakubowicz, I. (2013). Recycling of bioplastics, their
blends and biocomposites: A review. European Polymer Journal, 49,
2839-2858.
[32] Henshaw, J. M.; Han, W.; Owens, A. D. (1996). An overview of
recycling issues for composite materials. Journal of Thermoplastic
Composite Materials, 9, 4-20.
[33] Steuernagel, L. (2013). Recycling behaviour of natural fiber Reinforced
Thermoplastics. Proceedings of the Polymer Processing Society 29th
Annual Meeting, Nuremberg, Germany.
[34] Asmatulu, E.; Twomey J.; Overcash M. (2014). Recycling of fiber-
reinforced composites and direct structural composite recycling concept.
Journal of Composite Materials, 48, 593-608.
[35] Poulakis, J. G.; Varelidis, P. C.; Papaspyrides, C. D. (1997). Recycling
of Polypropylene-Based Composites. Advances in Polymer Technology,
16, 313-322.
[36] Goodship, V. (2010). Recycling and Reuse of Waste Composites
Management. Cambridge, UK: WP and CRC Press.
[37] Vilaplana, F. & Karlsson, S. (2008). Quality Concepts for the Improved
Use of Recycled Polymeric Materials: A Review. Macromolecular
Materials and Engineering, 293, 274-297.
[38] Moran, J.; Alvarez, V.; Petrucci, R.; Kenny, J.; Vazquez A. (2007).
Mechanical Properties of Polypropylene Composites Based on Natural
Fibers Subjected to Multiple Extrusion Cycles. Journal of Applied
Polymer Science, 37, 228-237.
[39] Twite-Kabamba, E.; Fassi Fehri, Z.; Rodrigue. D. (2011). Properties of
recycled LDPE/Birch Fibre Composites. Progress in Rubber, Plastics
and Recycling Technology, 27, 1-20.
[40] Ville, F.; Inceoglu, N.; Ghamri, J. L.; Pradel, A.; Durin, R.; Valette, B.;
Vergnes, B. (2013). Influence of Extrusion Conditions on Fiber
Breakage along the Screw Profile during Twin Screw Compounding of

Glass fiber-reinforced PA, International Polymer Processing, 28, 49-57.
[41] Bourmaud, A. & Baley, C. (2009). Rigidity analysis of
polypropylene/vegetal fibre composites after recycling. Polymer
[42] Pervaiz M. & Sain MM. (2003). Carbon storage potential in natural fibre
composites. Resources, Conservation and Recycling, 39, 325-40.
[43] Beg, M. D. H. & Pickering, K. L. (2008). Reprocessing of wood fibre
reinforced polypropylene composites. Part I: Effects on physical and
mechanical properties. Composites: Part A, 39, 1091-1100.
[44] Fonseca-Valero, C.; Ochoa-Mendoza, A.; Arranz-Andrés, J.; González-
Sánchez, C. (2015). Mechanical recycling and composition effects on
the properties and structure of hardwood cellulose-reinforced high
density polyethylene eco-composites. Composites: Part A, 69, 94-104.
[45] Wang, Y.; Chan, H. C.; Lai, S. M.; Shen, H. F.; Hsiao, Y. K. (2001).
Effect of Compounding Conditions on the Wood Flour/Polyethylene
Composites in Twin-Screw Extruders. Proceedings of the 59th Annual
Technical Conference, Dallas, Texas, USA.
[46] Soccalingame, L.; Bourmaud, A.; Perrin, D.; Béenézet, J. C.; Bergeret,
A. (2015). Reprocessing of wood flour reinforced polypropylene
composites. Polymer Degradation and Stability, 113, 72-85.
[47] Walz K.; Jacobson R.; Sanadi A. R. (1994). Effect of
reprocessing/recycling on the mechanical properties of kenaf-PP
composites. Internal report, University of Wisconsin-Madison,
Wisconsin, US.
[48] Augier, L.; Sperone, G.; Vaca-Garcia, C.; Borredon, M. E. (2007).
Influence of the wood fibre filler on the internal recycling of poly(vinyl
chloride)-based composites. Polymer Degradation and Stability, 92,
1169-1176.
[49] Kurniawana, D.; Kim, B. S.; Lee, H. Y.; Lim, J. Y. (2013). Effects of
repetitive processing, wood content, and coupling agent on the
mechanical, thermal, and water absorption properties of
wood/polypropylene green composites. Journal of Adhesion Science and
Technology, 27, 1301-1312.
[50] Åkesson, D.; Fuchs, T.; Stöss, M.; Root, A; Stenvall, E; Skrifvars, M.
(2016). Recycling of wood fiber-reinforced HDPE by multiple
reprocessing. Journal of Applied Polymer Science, 133, 43877-43884.
[51] Graupner, N.; Albrecht, K.; Ziegmann, G.; Enzler, H.; Müssig, J. (2016).
Influence of reprocessing on fibre length distribution, tensile strength
and impact strength of injection moulded cellulose fibre-reinforced

polylactide (PLA) composites. EXPRESS Polymer Letters, 10, 647-663.
[52] Arbelaiz, A.; Fernández, B.; Ramos, J. A.; Retegi, A.; Llano-Ponte, R.;
Mondragon, I. (2005). Mechanical properties of short flax fibre
bundle/polypropylene composites: Influence of matrix/fibre
modification, fibre content, water uptake and recycling. Composites
Science and Technology, 65, 1582-1592.
[53] Le Duigou, A.; Pillin, I.; Bourmaud, A.; Davies, P.; Baley, C. (2008).
Effect of recycling on mechanical behaviour of biocompostable
flax/poly(L-lactide) composites. Composites: Part A, 39, 1471-1478.
[54] Srebrenkoska, V.; Gaceva, G. B.; Avella, M.; Errico M. E.; Gentile, G.
(2008). Recycling of polypropylene-based eco-composites. Polymer
International, 57, 1252-1257.
[55] Paukszta, D. & Borysiak, S. (2013). The Influence of Processing and the
Polymorphism of Lignocellulosic Fillers on the Structure and Properties
of Composite Materials - A Review. Materials, 6, 2747-2767.
[56] Von Turkovich, R. & Erwin, L. (1983). Fiber fracture in reinforced
thermoplastic processing. Polymer Engineering & Science, 23, 743-749.
[57] Inceoglu, F.; Ville, J.; Ghamri, N.; Pradel, J. L.; Durin, A.; Valette, R.;
Vergnes, B. (2011). Correlation between processing conditions and fiber
breakage during compounding of glass fiber-reinforced polyamide.
Polymer Composites, 32, 1842-1850.
[58] Anonymous (1991). Recycling: the key to future success. Reinforced
Plastics, 35, 20-23.
[59] Corvaglia, P.; Passaro, A.; Manni, O.; Barone, L.; Maffezzoli, A. (2006).
Recycling of PP-based Sandwich Panels with Continuous Fiber
Composite Skins. Journal of Thermoplastic Composite Materials, 19,
731-745.
[60] Cornier-Ríos, H.; Sundaram, P. A.; Celorie, J. T. (2007). Effect of
Recycling on Material Properties of Glass-filled Polyethylene
Terephthalate. Journal of Polymers and the Environment, 15, 51-56.
[61] Kenmochi, K.; Takayanagi, H.; Nagasawa, C.; Takahashi, J.; Hayashi,
R. (1995). Possibility of Closed Loop Material Recycling for Fiber
Reinforced Thermoplastic Composites. Advanced Performance
Materials, 2, 385-394.
[62] Colucci, G.; Ostrovskaya, O.; Frache, A.; Martorana, B.; Badini, C.
(2015). The effect of mechanical recycling on the microstructure and
properties of PA66 composites reinforced with carbon fibers. Journal of
Applied Polymer Science, 132, 42275-42284.
[63] Sarasua, J. R. & Pouyet, J. (1997). Recycling effects on microstructure

and mechanical behaviour of PEEK short carbon-fibre composites.
Journal of Materials Science, 32, 533-536.
[64] Schinner, G.; Brandt, J.; Richter, H. (1996). Recycling Carbon-Fiber-
Reinforced Thermoplastic Composites. Journal of Thermoplastic
Composite Materials, 9, 239-245.
[65] Wang, K.; Bahlouli, N.; Addiego, F.; Ahzi, S. (2016). Elastic and yield
behaviors of recycled polypropylene-based composites: Experimental
and modeling study. Composites Part B, 99, 132-153.
[66] Xie, B.; Cui, Y. H.; Xu, J.; Wang, X. X.; Zhang, H. H.; Zhang, Z. D.
(2012) Investigation on Microstructure and Properties of Foamed (Wood
Fiber)/(Recycled Polypropylene) Composites. Journal of Vinyl and
Additive Technology, 18, 105-112.
[67] Prasanth, R.; Shankar, R.; Dilfi, A.; Thakur, V. K.; Ahn, J. H. Eco-
Friendly Fiber- Reinforced Natural Rubber Green Composites: A
Perspective on the Future. Thakur, V.K. Green Composites from Natural
Resources. Boca Raton, Florida, USA: CRC Press; 2013; 205-254.
[68] Borsoi, C.; Cavitano, C.; Scienza, L.C.; Zattera, A.J. (2013). Processing
of recycled expanded polystyrene and waste newsprint fibers
composites. Proceedings of the Polymer Processing Society 29th Annual
Meeting, Nuremberg, Germany.
[69] Consuegra Siaotong, B. A. (2006). Effects of fiber content and extrusion
parameters on the properties of flax fiber-polyethylene composites.
Master Science Thesis, University of Saskatchewan, Saskatoon, Canada.
[70] Winandy, J. E.; Stark, N. M.; Clemons, C. M. (2004). Considerations in
recycling of Wood-Plastic Composites. 5th Global Wood and Natural
Fibre Composites Symposium, Kassel, Germany.
[71] Yeh, S. K. & Gupta, R. K. (2008). Improved wood–plastic composites
through better processing. Composites Part A, 39, 1694-1699.
[72] Scelsi, L.; Hodzic, A.; Soutis, C.; Hayes, S. A.; Rajendran, S.;
AlMa'adeed, M. A.; Kahraman, R. (2011). A review on composite
materials based on recycled thermoplastics and glass fibres. Plastics,
Rubber and Composites, 40, 1-10.
[73] Cavalieri, F. & Padella, F. (2002). Development of composite materials
by mechanochemical treatment of post-consumer plastic waste. Waste
Management, 22, 913-916.
[74] Khait, K.; Riddick, E. G.; Torkelson, M. (2001). Recovery of post-
consumer plastic waste via Solid State Mechanochemistry. Proceedings
of the 59th Annual Technical Conference, Charlotte, North Carolina,

USA.
[75] Twite-Kabamba, E. (2006). Etude du recyclage des mousses de PE sur
une ligne d’extrusion. Master Science Thesis, Université Laval, Québec,
Canada. [Twite-Kabamba, E. (2006). Study of the recycling of foamed
PE on an extrusion line. Master Science Thesis, Laval University,
Québec city, Canada].
[76] Verdolotti, L.; Lucolano, F.; Capasso, I.; Lavorgna, M.; Iannace, S.;
Liguori, B. (2014). Recycling and Recovery of PE-PP-PET-based Fiber
Polymeric Wastes as Aggregate Replacement in Lightweight Mortar:
Evaluation of Environmental Friendly Application. Environmental
Progress & Sustainable Energy, 33, 1445-1451.
[77] Otheguy, M. E.; Gibson, A. G.; Findon, E.; Cripps, R. M.; Ochoa
Mendoza, A.; Aguinaco Castro, M. T. (2009). Recycling of end-of-life
thermoplastic composite boats. Plastics, Rubber and Composites, 38,
406-411.
Chapter 4
A SUGGESTION FOR STANDARDIZING

A TRACEABLE PROCESS FOR
THE DETERMINATION OF THE MECHANICAL
PROPERTIES OF CONCRETE CONTAINING
THERMOPLASTIC POLYMERS
AS AGGREGATES
Stamatia Gavela1,*, Georgios Papadakos2

and Vasilia Kasselouri – Rigopoulou
1
Chemical Engineer, MSc, PhD, 2Mechanical Engineer, PhD,
Metrology consultant, 3Professor Emeritus
School of Chemical Engineering,
National Technical University of Athens, Greece
ABSTRACT
A lot of research has been performed on the utilization of waste
products in concrete. Due to the fact that concrete is widely used and has
a long service life, the wastes used in it are removed from the waste
stream for a long period. One of the waste types studied has been
thermoplastic waste. Polymers become increasingly a preferable material
*
Corresponding Author Email: [email protected].
144 S. Gavela, G. Papadakos and V. Kasselouri–Rigopoulou
by product designers because they provide useful properties as design

flexibility and significant weight savings. For this reason the worldwide
production has grown, with packaging being the main application for
plastics. Because disposal of post-consumer plastics is increasingly being
constrained by legislation, there is considerable demand for alternatives
leading to effective life cycle management. As the amount of aggregates
required in concrete production is large, the environmental benefits of the
replacement of natural aggregates by thermoplastic waste is not only
related to its safe disposal, but also to the mitigation of environmental
impacts arising from the extraction of aggregates, i.e., the visual intrusion
and the loss of countryside. On the other hand, designers need to feel
confident about the mechanical properties of the product in order to be
encouraged to add the maximum practicable amount of thermoplastics in
concrete or mortars. Studies have been conducted towards the assessment
of mechanical properties of concrete or mortar containing thermoplastic
polymers as aggregates. Due to (a) the variability of thermoplastic
polymers which could be added to concrete, (b) their significant
differences from natural aggregates and (c) the variability of parameters
affecting the properties of concrete itself, the data obtained from tests on
mechanical properties for concrete specimens containing thermoplastic
polymers are based on a multifactorial process. When these results come
from totally different laboratory processes, their metrological traceability
is of great significance in order to achieve inter-comparability. A
standardized protocol for the experimental design and referencing of all
essential relative data is needed in order to facilitate any attempt to
compose the results from studies when thermoplastic polymers of
different origin and characteristics are used and the percentage of
aggregate replacement varies. This study clarifies that different reporting
on the total experimental procedure arises difficulties in composing
results coming from different laboratories and the statistical inference on
the effect of natural aggregates replacement by thermoplastics. A
standardized protocol based on (a) metrological traceability as determined
by ISO and (b) quantitative review principles and procedures, is proposed
to facilitate comparability and compatibility of future results. This
protocol aims at a quantitative review providing a statistical basis for the
estimation of the Compressive Strength Reduction (CSR) for a concrete
or mortar mix where conventional aggregates have been, partially or
totally, replaced by thermoplastic polymers. A reference mix was
assessed necessesary to be properly incorporated in the experimental
design of any such study. This quantitative review was succesfully
initiated but needs further extention by the execution of more experiments
that should, in any case, follow the proposed protocol leading to an
accurate model aiming at the maximization of this thermoplastics use.
A Suggestion for Standardizing a Traceable Process … 145
Keywords: concrete, mortar, aggregate replacement, thermoplastics,

standardized protocol, quantitative review
INTRODUCTION
Concrete is a widely used construction material and doubtless a lot of
research has been made in the entire world for many years on concrete
strength, concrete additives, fillers, aggregate substitutions, utilization of waste
in concrete. Concrete technology is being taught in universities and
researchers publish a big amount of papers. In this study we will focus only on
aggregate substitution by thermoplastic waste. A lot of research has been made
on the utilization of thermoplastic waste in concrete. Of course in each study
different materials, substitution percentage and procedures are used. Taking
into account the variability of parameters affecting the properties of concrete
itself, any attempt to compose the results seems complicated. This study does
not only consist a typical literature review, but it will propose a standardized
protocol for the experimental design based on (a) metrological traceability as
determined by ISO and (b) quantitative review principles and procedures, to
facilitate comparability and compatibility of future results.
Plastic is a widely used material, especially in packaging. Plastics
increasingly become the material of choice by product designers because they
provide useful properties as design flexibility and significant weight savings.
For this reason the worldwide production has grown up significantly. Due to
the fact that disposal of post-consumer plastics is increasingly being
constrained by legislation there is a demand for alternatives to disposal or
landfilling. In fact, the disposal and landfilling of plastics contributes
significantly to environmental problem as most of them are non-degradable so
they take a long time to decompose. Plastics industry already uses a part of
scrap mixed with virgin material in new production uses. But the volume of
scrap resulting from the startup procedures or from the “out of specification
products” is bigger than the scrap that can be used. The large volume of
materials required for construction is a major area for the reuse of waste
materials. The suggested use of polymers as aggregates in concrete has the
advantage that sorting of different kinds of the polymer is not necessary.
The number of variables that affect concrete properties is large. One
important property, strength of concrete, depends on the hardening of cement
paste, which depends on the hydration process. The hardening of the cement
paste and its bond to aggregates results to concrete strength. Because of this
direct relationship of strength to hydrated cement paste, strength usually gives

an overall picture of the quality of concrete. For this reason and also due to the
simplicity of its measurement and the fact that it is important parameter for the
structural design it is always determined in any study.
Focusing on concrete containing thermoplastic waste as conventional
aggregate substitution, we could enumerate the following factors that it is
known to affect concrete strength: Conventional aggregates’ strength, grading,
maximum size and specific gravity, thermoplastic waste’s grading and specific
gravity, proportions (they include cement content and aggregate to cement
ratio), percentage of replacement, water to cement ratio, cement type,
additives, air content of the mix, size and shape of specimen, age and physical
condition of specimen.
The process for quantifying the strength of concrete or mortar containing
thermoplastics is essential as the benefit of adding such material into the mix
should not end up to a collapse of the mechanical properties. The designer
would not be encouraged to add any or even little quantities of thermoplastics
as aggregates if the risk of mechanical properties collapse would be
significant. A statistical model such the one proposed would increase the
confidence of the designer on his calculations about the strength of the
concrete or mortar mix. It is expected that an increasing accuracy and
reliability of such a statistical model will lead to the maximization of the
amounts of thermoplastics used in these mixes, as the designers will be able to
predict accurately the reduction caused to the compressive strength of the mix.
REVIEW ON DATA SOURCES

Concrete or mortar containing plastic aggregates is typically produced by
replacing conventional coarse or fine aggregates with plastic coarse or fine
aggregates and the following cases of defining the replacing process exist: a)
replacing conventional coarse or fine aggregates with the same volume of
plastic coarse or fine aggregates, which is called by volume replacement, or b)
replacing conventional coarse or fine aggregates with the same mass of plastic
coarse or fine aggregates, which is called by volume replacement. Some
researchers replace only fine aggregates, some only coarse aggregates and
other both. The replacement may be partial or full. Some researchers try to
keep the same gradiation curve after the replacement, or some other ratio. In
this study, with the phrase “by volume replacement of fine aggregates, it is
meant that only fine aggregates is being replaced by the same volume of
plastic aggregates, and the coarse aggregates (if they exist in the mix) are in
the same mix proportion with the reference mix.
Table 1. Types of replacement of conventional aggregates by plastic in

mortar or concrete and properties of the plastic used
D: density/SG: specific
origin and type of

gravity/ABD: apparent
mortar/concrete
particle size of
bulk density/UW: Unit
replacement
Refe-rence
weight/BDC: bulk
criterion
density compacted/SD:
plastic
plastic
Specific density of
plastic
[2] mortar No reference PET bottle max size: 4mm SG: 1.27g/cm3
mix
[3] concrete 0, 10, 30, 50% shredded car max size: 13mm UW: 0,485g/cm3
by volume of bumpers
coarse
[4] mortar No reference PET bottles max size: 4mm SG: 1.27g/cm3
mix granules
[5] concrete 0, 5, 10, 15% No specific max size: 5mm --
by mass of type
coarse
[6] concrete 5% by mass of PET bottles max size: 5mm SG: 1.32, BDC:
fine 660kg/m3
[7] concrete 0, 10, 20% by PET bottles two types: --
volume of fine type A: average size:
2,6mm
type B: max size:
11,4mm
[8] mortar 0, 2, 5, 10, 15, PET bottles three types: type A: ABD: 326kg/m3
20, 30, 50, 70, type A: max size: 5mm type B: ABD: 345kg/m3
100% by type B: max size: 2mm type C: ABD: 408kg/m3
volume of fine type C: max size: 1mm
[9] concrete 0, 20, 40% by HDPE planar dimension: SG: 1.0
volume of fine 10mm
thickness: 1.9mm
[10] mortar 0, 3, 10, 20, two types: type A: max size: 10 type A: BD: 547kg/m3,
50% by type A: PET mm SG: 1.358
volume of fine type B: PC type B: max size: 5mm type B: BD: 612kg/m3,
SG: 1.240
[11] concrete 0, 10, 15, 20% plastic varying length of 0.15- D: 386.7kg/m3
by mass of containers 12mm and width of
fine consisting 0.15-4mm
approx. 80%
PE and 20%
PS
Table 1. (Continued)
D: density/SG: specific
origin and type of

gravity/ABD: apparent
mortar/concrete
particle size of
bulk density/UW: Unit
replacement
Refe-rence
weight/BDC: bulk
criterion
density compacted/SD:
plastic
plastic
Specific density of
plastic
[13], concrete 0, 7, 12, 20% two types: type A: max size 6mm type A: SG: 1020kg/m3
[14] by volume of type A: PP type B: max size type B: SG 910kg/m3
fine type B: HDPE 4,5mm
[16] concrete 0, 0.5, 1, 2.5, two types: two types: type A: D: 1320kg/m3
5, 7.5% by type A: PET type A: max size type B: D 860kg/m3
mass of fine type B: LDPE 12.5mm
type B: max size
4.8mm
[17] concrete 0, 3, 6, 9, 12 -- 0.15-4.75mm --
by mass of
fine
[18] concrete 0, 5, 10, 15% PET bottles max size 7mm UW: 464kg/m3
by volume of SG: 1.11g/cm3
fine
[19] concrete 0, 5, 10, 15% PET bottles max size 4,75mm SG: 1200kg/m3
by mass of
fine
When the densities of plastic and conventional aggregates used in the

studies are being refered, then one can calculate the by volume replacement
from the by mass replacement and the opposite. But if the is no such reference,
then the calculation cannot be made.
For the studies that did not include a reference mix, the replacement
percentage (by mass or by volume) is not defined. In this case we consider that
the reference mix would be a mix that has a conventional aggregate proportion
same as the sum of plastic and conventional aggregate of a mix of the study.
The plastic aggregates used in studies were prepared from plastic waste
obtained from different ex uses and of different types. Table 1 lists the types,
origin and properties of plastic aggregates used in concrete or mortar used in
various studies. In general plastic was shredded in mills, after washing [2, 16]
or unwashed [6]. In some cases the obtained particles were sieved to obtain a
granulometric curve similar to the one of the conventional aggregates that they
would replace. Of course the bulk density and specific gravity of plastic
aggregates is much lower than the ones of conventional aggregates. Bulk
density is reported to be depended on the recycling method [1].
In case where the researchers made the replacement but also altered the
mix proportions in order to keep a constant workability (water to cement ratio)
[8] or for any other reason [15] the results do not serve the purpose of this
study, because a) the compressive strength of specimens containing plastic
waste cannot be compared to the so called reference specimens because the
mix proportions is a significant factor for the value of compressive strength
and b) even doing so these results could be used if the altered values of mix
proportions would be refered briefly.
Plastic aggregates do not achieve chemical bonds with cementitious
materials. The result is the lower bond between plastic aggregates and cement
paste in comparison to that of conventional aggregates and cement paste. To
this lower bond charecteristics of plastics it is attributed the lower strength of
concrete containing plastic aggregates. Although a pretreatment on plastic
aggregates can be made to enhance the bonding to the cement paste, most
researchers avoid it so as not to introduce another process and extra cost.
Chemical treatments that have been used are bleach and alkaline bleach [12].
DATA SELECTION
Two datasets were formed and used for the statistical analysis described:
(a) Data derived from studies shown in Table 1, which refer to

experiments on the compressive strength of concrete or mortar when
plastic material was used to replace part of the conventional
aggregates, fine or coarse, compared to corresponding reference mix
proportion specimens. In order to achieve that, only studies which
used a reference mix proportion specimen and also specimens
containing thermoplastics with the same mix proportions with the
reference (appart from the replaced aggregate quantity and the fine or
coarse aggregate quantity that was replaced) were included [3, 5-7, 9-
11, 13-14, 16-19]. Descriptive statistics for this dataset is provided in
Table 3. The aim of using this dataset is to investigate the statistical
relation of Compressive Strength Reduction in specimens containing
thermoplastics, to variables that describe how the replacement was
realized.
(b) Date derived from the group of reference specimens of the above
dataset, enriched by other studies [20-29] providing data on the results
of Reference Specimen Compressive Strength experiments.
Descriptive statistics for this dataset is provided in Table 3. The aim

of using this dataset is to investigate aspects of the statistical analysis
for the abovementioned dataset that could not exhibit adequate
statistical significance due to the very little population of reference
specimens within the corresponding replacement experiments.
As shown in Table 2 and Table 3, the used datasets cover a wide variety of
parameters and measurement results, adequate to represent a globally
sustainable statistical analysis. Many of these variables and their values are
directly refered by the researchers, such as the Compressive Strength of the
Specimen (SCS) for given Curing Days (CD), which in the case of a reference
specimen is refered herein as RSCS (Reference SCS), the Maximum Polymer
Size (MPS), the Specific Gravity of Polymer (SGoP) or the Shape of the
Specimen (SS). If these data are not refered by the researcher it is impossible
to deduce their value by combining other sources of information.
In other cases, some of the variables and their values are either refered by
the researchers or they can be deduced by combining other sources of
information or variables values. Such cases are:
(a) the proportion of replacement of fine (RoF) or coarse (RoC)

conventional aggregates, in the case where this is conducted by the
criterion of equal masses, needing only the mix proportions of the
specimen containing polymer and of the corresponding reference
specimen. Of course, this calculation is not plausible, not only in the
case where the mix proportions are not refered, but also in the case
where the mix proportions are not kept the same between the polymer
containing specimen and the reference, for the substances apart from
aggregates. The calculation used in this occasion is described in Eq. 1
and in Eq. 2, using the variables of Fine Aggregates Content and the
Coarse Aggregates Content in the Refernece specimen (RFAC,
RCAC) and in the tested replacement specimen (FAC, CAC):
RoF = (RFAC – FAC) / RFAC Eq. 1

RoC = (RCAC – CAC) / RCAC Eq. 2
where the measurement units for RFAC, FAC, RCAC and CAC are
kg/m3 (kilograms of aggregate per cubic meter of the fresh concrete or
mortar).
(b) the Water to Binder (WTB) and the Aggregates to Binder (ATB)
ratios, which were calculated as in Eq. 3 and in Eq. 4, by using the
variables of Water Content (WC) and Binder Content (BC, taking into
account all types of binder like cement, fly ash etc):
WTB = WC / BC Eq. 3
ATB = (FAC + CAC + PC) / BC Eq. 4
For the statistical analysis some of the below mentioned variables were
completely calculated by other sources of information. Such variables were:
(a) the Polymer to total Aggregates ratio calculated by masses or by

volumes of the corresponding materials, PTAM and PTAV
respectively. As it is shown in the statistical analysis, by using these
variables, the weaknesses of the variables RoF and RoC are being
dealt adequately: the RoF and RoC variables do not depict the
decision criterion upon replacement (by mass or by volume) and they
do not take into account the mix proportion of the complementary
aggregete material (the fine aggregates when the coarse aggregates are
replaced and vice versa) so they should be both incorporated into any
statistical model aiming to a quantitative review. At the contrary,
PTAM and PTAV illustrate the exact quantitave relation of polymers
to the conventional aggregates into the tested specimen.
These variables are calculated as:
PTAM = PC / (WC + BC + FAC + CAC + PC) Eq. 5

PTAV=(PC/SGoP)/(FAC/SGoFA+CAC/SGoCA+PC/SGoP) Eq. 6
(b) the Compressive Strength Reduction (CSR) and the Proportional

Compressive Strength Reduction (PCSR) for the specimens
containing thermoplastics as aggregates, compared to the
corresponding reference specimens. These variables were calculated
as:
CSR = RSCS – SCS Eq. 7
PCSR = 100 * CSR / RSCS = 100 * (RSCS-SCS) / RSCS Eq. 8
Using these two variables as estimators of the investigated result

gives the statistical analysis a global sense. In case that this should be
performed only by using the croude result of Specimen Compressive
Strength (SCS), the exact mix proportions of the specimens would the
prevailing factor affecting the result, leaving no chance for any other
factor to emerge as a significant parameter into a statistical model
aiming at a quantitave review of widely different aggregate
replacement experiments. CSR is expected to be more adequate
estimator in the case of systematic effects, but PCSR is expected to be
adequate even when some parameters cause proportional reduction to
the Compressive Strength of the concrete or mortar. Even more,
PCSR incorporates the level of the reference specimen’s Compressive
Strength, so it should be considered as a more global estimator.
A weakness of the dataset that should be dealt with in future research, is

that there is a lack of information on the Compressive Strength of tested
specimens for curing days’ time points apart from that of 28 days.
The combination of the results from the above two datasets is expected to
provide an estimation of the Compressive Strength of concrete or mortar in the
case of conventional aggregates replacement by thermoplastic polymers. By
solving Eq. 8, it comes that the Compressive Strength after replacement could
be estimated as:
SCS = RSCS – CSR = [(100 – PCSR)/100] * RSCS Eq. 9
In both realizations of this equation, the variable in question, SCS, could

be estimated by using only the mix proportion characteristics of the
corresponding reference mix (the RSCS, which corresponds to a mix
proportion where the used plastic polymer as aggregate is assumed to be an
equivalent typical aggregate) and a reduction score, CSR or PCSR, which
could be given as a function of only the characteristics of the presence of
plastic polymer in the mix (a theoretical replacement assumed, even if there
are no reference specimens).
On the other hand, even when the performance of a concrete or mortar
which contains plastic polymer as aggregates substitute is estimated by
laboratory tests on adequate specimens, the above equations could be used in
order to verify the laboratory test results. This should incorporate the
uncertainty of the estimator provided by Eq. 9. The corresponding sensitivity
coefficients and the equation derived by the application of the error
propagation method (see ISO GUM) is described in a following paragraph.
Table 3. Descriptive statistics for the dataset used to investigate the

replacement results on the Compressive Strength of concrete or mortar
specimens
Variable N Mean Min Max

Name Description Type Units value value
SCS Specimen Compressive Strength Continuous MPa 238 29.2 3.6 77.7
(the result of the corresponding
measurement)
RSCS Reference Specimen Compressive Continuous MPa 238 37.4 14.8 78.0
Strength (the result of the
measurement on the reference
specimen, corresponding to each
SCS measurment)
CSR Compressive Strength Reduction Continuous MPa 238 6.9 -6.2 47.2
(in specimen with replacement, as
compared to the corresponding
reference specimen)
PCSR Proportional Compressive Strength Continuous % 238 19.8 -22.5 73.2
Reduction (CSR devided by
Compressive Strength measured
on the corresponding reference
specimen)
PTAM Polymer to total aggregates ratio Continuous - 164 0.041 0.002 0.339
calculated by masses
PTAV Polymer to total aggregates ratio Continuous - 188 0.109 0.004 0.500
calculated by volumes
WTB Water to binder ratio calculated by Continuous - 236 0.51 0.28 0.93
masses
ATB Aggregate to binder calculated by Continuous - 200 4.3 1.4 7.0
masses
RoF Replacement proportion of Fine Continuous % 236 7.9 0.0 50.0
aggregates
FAC Fine Aggregates Content (mix Continuous kg/m3 164 699.2 343.9 1493.0
proportion) of conventional
aggregates
RoC Replacement proportion of Coarse Continuous % 236 2.7 0.0 50.0
aggregates
CAC Coarse Aggregates Content (mix Continuous kg/m3 164 910.6 0.0 1600.0
proportion) of conventional
aggregates
PC Polymer Content (mix proportion) Continuous kg/m3 164 60.2 3.8 394.3
of any kind of polymer
DoR Decision on Replacement (0 = by Binomial - 236 0.54 - -
volume, 1 = by masses)
MPS Maximum Polymer Size Continuous mm 182 6.0 1.9 13.0
SGoP Specific Gravity of Polymer Continuous g/cm3 196 1.10 0.39 1.36
CD Curing Days Continuous d 238 31 1 365
SS Specimen Shape (0 = rectangular, Binomial - 204 0.34 - -
1 = cylindrical)
STATISTICAL ANALYSIS AND DISCUSSION

A crude estimation on the correlation of Specimen’s Compressive
Strength (SCS) with various parameters is conducted by using the second
dataset (shown in Table 3), in order not to include the confounding effect of
the replacement procedure. The corresponding correlation coefficients are
shown in Table 4.
Table 4. Descriptive statistics for the dataset made by reference specimens
Variable N Mean Min Max

Name Description Type Units value value
RSCS Reference Specimen Compressive Continuous MPa 143 33.6 8.1 78.0
Strength (the result of the
corresponding measurement)
WC Water Content (mix proportion) Continuous kg/m3 126 206.8 135.0 436.6
BC Binder Content (mix proportion) Continuous kg/m3 114 345.2 175.0 592.2
RFAC Fine Aggregates Content (mix Continuous kg/m3 114 720.3 420.0 1539.0
proportion) in Reference
Specimen
RCAC Coarse Aggregates Content (mix Continuous kg/m3 114 1100.5 0.0 1856.0
proportion) in Reference
Specimen
WTB Water to binder ratio calculated Continuous - 142 0.61 0.28 1.19
by masses
ATB Aggregate to binder calculated by Continuous - 121 5.6 1.4 11.1
masses
MCAS Maximum Conventional Continuous mm 125 17.4 4.0 38.0
Aggregate Size
SS Specimen Shape (0 = rectangular, Binomial - 136 0.19 - -
1 = cylindrical)
CD Curing Days Continuous d 143 64.8 1 365
Table 5. Correlation coefficients between RSCS and various parameters
Parameter Correlation coefficient used value significance

WTB Pearson r -0.555 < 0.01
Spearman rho -0.515 < 0.01
ATB Pearson r -0.469 < 0.01
Spearman rho -0.496 < 0.01
MCAS Pearson r -0.117 0.19
Spearman rho 0.004 0.97
CD Pearson r 0.216 < 0.01
Spearman rho 0.443 < 0.01
SS Kendal’s tau -0.028 0.70
Figure 1. Correlation between RSCS and WTB.
According to the above correlations, the parameters of Curing Days (CD)

and Water to Binder ratio (WTB) should be included in any statistical model
that is intended to predict a Specimen’s Compressive Strength (SCS). ATB is
also strongly correlated to SCS, but using WTB into any model is adequate to
represent both the ATB parameter, as these are strongly connected to each
other (Pearson r = 0.793). As also seen in Figure 1, as WTB increases, SCS
decreases. The correlation of SCS with the CD parameter seems to have a
sigmoidal shape, as of the model proposed by Pedersen [30]. By combining
the effect of WTB and CD into Eq. 10, the result explains the dispersion of
SCS values derived from the various studies been reviewed (see Figure 2).
Figure 2. Correlation between RSCS, WTB and CD.

The resulting equation shown in Figure 2 is:
RSCS = (RSCS∞-a*WTB) * exp (-(b/CD)c) Eq. 10
After a regression process on the used dataset the values of the above
parameters were estimated as: RSCS∞ = 100 ± 26 MPa, a = 76 ± 23 MPa,
b = 3.1 ± 1.8 d and c = 0.43 ± 0.23 (significance of the regression result on Eq.
10: r = 0.793). The values of these parameters are strongly correlated to each
other, so any attempt to estimate the corresponding uncertainty for an RSCS
estimate should incorporate the covariances between the equation parameters.
In order to assess the effect of aggregates replacement by polymer
material, the CSR and the PCSR variables are assessed. Table 5 and Table 6
show the correlation of these variables with various parameters related to the
replacement process, by using the first dataset shown in Table 2.
Table 6. Correlation coefficients between CSR and PCSR

with various parameters
Parameter Correlation Correlation with CSR Correlation with PCSR

coefficient used value significance value significance
PTAV Pearson r 0.581 < 0.01 0.531 < 0.01
Spearman rho 0.795 < 0.01 0.783 < 0.01
PTAM Pearson r 0.415 < 0.01 0.512 < 0.01
Spearman rho 0.792 < 0.01 0.815 < 0.01
CD Pearson r -0.034 0.60 -0.095 0.15
Spearman rho 0.072 0.27 -0.039 0.55
RoF Pearson r 0.083 0.20 0.339 < 0.01
Spearman rho 0.120 0.07 0.259 < 0.01
RoC Pearson r 0.625 < 0.01 0.340 < 0.01
Spearman rho 0.268 < 0.01 0.186 < 0.01
DoR Kendal’s tau 0.049 0.36 0.032 0.55
ATB Pearson r -0.292 < 0.01 -0.222 < 0.01
Spearman rho -0.310 < 0.01 -0.233 < 0.01
MPS Pearson r 0.418 < 0.01 0.375 < 0.01
Spearman rho 0.203 < 0.01 0.196 < 0.01
SGoP Pearson r -0.356 < 0.01 -0.090 0.21
Spearman rho -0.012 0.86 0.133 0.06
WTB Pearson r 0.158 0.01 -0.054 0.41
Spearman rho -0.052 0.42 0.062 0.34
RSCS Pearson r 0.317 < 0.01 -0.020 0.76
Spearman rho 0.186 < 0.01 -0.055 0.40
SS Kendal’s tau 0.015 0.79 0.164 < 0.01
According to the above correlations, it is obvious that either PTAV or

PTAM is a more adequate index representing the replacement process (see
Figure 3) rather than the RoF or RoC, indexes. Moreover, the PTAV and
PTAM indexes are independent from any replacement assumption, which is
very critical in cases where there is no reference mix propotion used. Also,
between these two alternative indexes, PTAV seems to provide more
sensitivity, when PTAM provides small value variations for the same mix
proportions. At the contrary, RoF and RoC are significantly correlated to the
Decision on Replacement (DoR, Spearman rho equal to -0.272 and 0.222
respectively), so any attempt to incorporate them into an equation for the
estimation of either CSR or PCSR should simultaneously incorporate also the
DoR variable, rising up the complexicity of the regression model.
The result of no statistical correlation between the PCSR estimator and the
WTB or RSCS variables adds value to the reliability of the model described in
Eq. 9, as it is deduced that a separate regression for the two product parameters
in it is statistically valid. Otherwise, a common regression should be
performed for the total function described in this equation, and so rising the
complexity and the need for more reviewed studies.
What is mostly interesting is that both CSR and PCSR variables are totally
independent from the variable of CD (Curing Days). This leads to the useful
finding that when using a CSR or PCSR estimator in combination with a
RSCS estimator, as described in Eq. 9, these two estimators are not interacting
depending on the variable of Curing Days.
The above provided a best regression between PCSR and the parameters
of PTAV, MPS and SS, as shown in Eq. 11 and Figure 4.
PCSR = (d0 + d1 * MPS + d2 * MPS2) * PTAVn + m * SS Eq. 11
The values provided by regression for the above equation using the first
dataset of this study are: d0 = 86.4 ± 35.6 %, d1 = -11.7 ± 10.9 % per mm, d2 =
1.02 ± 0.70 (% per mm)-1, n = 0.39 ± 0.11, m = -7.7 ± 6.7 with an adequate
regression result (r = 0.699).
The abovementioned results is a proof that a procedure for optimizing a
statistical model that aims at the prediction of Compressive Strength for
concrete and mortar mixes containing thermoplastics is justified. Once a
dataset which was not designed in the objective of a quantitative review gives
statistically significant results, the optimization is rather encouragable.
Figure 3. Correlation between CSR, PCSR and PTAM, PTAV.
Figure 4. Correlation between PCSR, PTAV and MPS.

FUTURE USE
It is provident that there are two cases where the procedure for estimating
or assessing the Compressive Strength of a concrete or mortar mix by using
Eq. 9 could be used:
(a) For given characteristics of a mix proportion and for the concrete or
mortar mix and so calculating a value for the Compressive Strength of
a respective specimen or
(b) for given laboratory test results in order to assess them in comparison
with an interval of expected values.
In case (a), the benefit comes from the fact that this assessment could be
performed before the realization of the concrete or mortar mix. For the so
called reference mix, it will be a statistically reliable process to estimate the
resulting Compressive Strength, taking into account basic features of this mix
and Eq. 10. Therefore, for the corresponding mix containing polymers, a
relative reduction of Compressive Strength will be calculated by using Eq. 11,
taking into account only the features of the presence of polymer into the mix.
Using such a standardized process, a concrete or mortar mix designer has
the ability to estimate the risk of a non appropriate Compressive Strength as a
function of the quantitative features of plastic polymers addition into the
designed product. The positive effect of plastic polymers addition into
concrete or mortar mixes is profound, but this should always be delt on the
basis that the realization does not lead to failed attempts (collapsed mechanical
properties).
A critical aspect of the standardization of this process is the level of
uncertainty that characterizes the estimation provided by Eq. 9. The lowest the
uncertainty of the estimator, the more accurate will be the designing
calculations of the designers leading to greater amounts of plastic polymers
additions into concrete or mortar mixes, without a significant risk of failure of
the corresponding mechanical properties.
In case (b), the same model could be used for the verification of laboratory
measurements, in order to assess wether a result is normal or not. Also in this
case the uncertainty of the calculation provided by the model is of critical
sense for obvious reason.
CONCLUSION
Research on the use of thermoplastic wastes as aggregates in concrete and
mortar mixes has been performed successfully. Many results exist showing
that there is a significant potential for such a utilization of large amounts of
wastes, which would otherwise end up into the environment. The profit gain
from the use of thermoplastic wastes as concrete or mortar constituents is a
matter of assessment against the risk of non-fulfillment for the relative
technical specifications on mechanical properties of the mix. Accurate
quantitative analysis is necessary for the decision upon the extent of
thermoplastic polymer addition into concrete or mortar mixes, in order to
maximize the rate of wastes utilization but in a way that does not compromise
the safety of the constructions.
The model proposed within this study could be used as a tool to assist a
standardized procedure for the estimation of Compressive Strength in concrete
or mortar mixes containing thermoplastic polymers. Its realization based on
the two datasets used in this study provided statistically significant result.
Although, the relatively low population of properly refered studies leads to a
significant level of uncertainty in the corresponding results. Adequate research
could optimize this model, not only for the values of the used parameters, but
also for a possible statistically significant presence of other parameters into the
equation, building up the specificity of the model.
Any attempt to standardize a procedure aiming to the use of the proposed
model within the two above objectives needs a protocol to be kept by all
researchers seeking to it. By this protocol, at least the following should be
noted and refered properly:
(a) The Compressive Strength of Reference Specimens. These

Reference Specimens should not differ from the tested Specimens
containing thermoplastic polymers as a matter of the mix
proportions for all other materials, apart from the replaced
proportion of conventional aggregate.
(b) The mix proportions of both Reference and of the specimens
containing thermoplastic polymer.
(c) The Specific Gravity of all materials contained into these mixes.
(d) The Curing Days that should be identical between the containing
thermoplastic polymers specimens and the corresponding
reference specimens, and which should, also, cover a
represantative span of time (i.e. at least one time point of early
Curing Days, 1 – 7 days, one of mid Curing Days, 14 – 28 days

and one of mature Curing Days, 90 – 365 days).
(e) The Specimen shape (cylindrical or rectangular) and dimensions.
(f) The type of polymer.
(g) The Maximum Polymer Size.
For all the above, measurement units should be also refered properly (i.e. in
SI units) for reasons of intercomparability.
Nevertheless, a future technoeconomical study would offer more
information in this kind of use of thermoplastics. A realistic quantification on
the potential for thermoplastics use into concrete or mortar products demands
a knowledge upon the uses where the expected reduction in mechanical
properties would not rise risks on the safety of constructions. Different kinds
of constructions are in a different way and extent succeptible to variance upon
the estimations on mechanical properties of construction materials. The
identification of concrete constructions that do are not significantly sensitive to
these variations would also encourage the designers to research and optimize
the process of adding amounts of thermoplastics in these materials.
REFERENCES
[1] Gu, L., Ozbakkaloglu, T, 2016. Use of recycled plastics in concrete: A
critical review. Waste Management 51, 19-42.
[2] Akcaozoglu, S., Atis, C.D., Akcaozoglu, K., 2010. An investigation on
the use of shredded waste PET bottles as aggregate in lightweight
concrete. Waste Management 30, 285-290.
[3] Al-Manaseer, A.A., Dalal T.R., 1997. Concrete containing plastic
aggregates. Concrete International 19, 47-52.
[4] Akcaozoglu, S., Atis, C.D., 2011. Effect of granulated blast furnace slag
and fly ash addition on the strength properties of lightweight mortars
containing waste PET aggregates. Construction and Building Materials
25, 4052-4058.
[5] Ghaly, A.M., Gill, M.S., 2004. Compression and deformation
performance of concrete containing postconsumer plastics. Journal of
Materials in Civil Engineering 16, 289-296.
[6] Frigione, M., 2010. Recycling of PET bottles as fine aggregate in
concrete. Waste Management 30, 1101-1106.
[7] Albano, C., Camacho, N., Hernandez, M., Matheus, A., Gutierrez, A.,
2009. Influence of content and particle size of waste pet bottles on
concrete behavior at different w/c ratios. Waste Management 29, 2707-
2716.
[8] Marzouk, O.Y., Dheilly, R.M., Queneudec, M., 2007. Valorization of
post-consumer waste plastic in cementicious concrete composites.
Waste management 27, 310-318.
[9] Soroushian, P., Alhozaimy, A., Eldarwish, A. I., 1993. Recycling of
plasticcs in concrete to enhance toughness characteristics and resistance
to shrinkage cracking. Proceedings of the Symposium on Recovery and
Effective Reuse of Discarded Materials and By-Products for
Construction of Highway Facilities, Colorado, October 19-22, pp. 5-41
to 5-53.
[10] Hannawi, K., Kamali-Bernand, S., Prince, W., 2010. Physical and
mechanical properties of mortars containing PET and PC waste
aggregates. Waste Management 30, 2312-2320.
[11] Ismail, Z., Al-Hashmi, E.A., 2008. Use of waste plastic in concrete
mixture as aggregate replacement. Waste Management 28, 2041-2047.
[12] Naik, T.R., Singh, S.S., Huber, C.O., Brodersen, B.S., 1996. Use of
post-consumer waste plastics in cement-based composites. Cement and
Concrete Research 26, 1489-1492.
[13] Gavela, S., Ntziouni, A., Rakanta, E., Kouloumbi, N., Kasselouri-
Rigopoulou, V., 2013. Corrosion behavior of steel rebars in reinforced
concrete containing thermoplastic wastes as aggregates. Construction
and Building Materials 41, 419-426.
[14] Gavela S., Kolias, S., Kordatos, K., Kasselouri-Rigopoulou, V., 2009. A
new approach to utilization of thermoplastic waste. Fresenius
Environmental Bulletin 18, 224-228.
[15] Evangelista, A.C., Almeida, V.C., Cordeiro, A.C., Alves, L.S.,
Rodrigues, R.A., 2004. Alternative aggregate study for production of
concrete and mortar. Proceedings of 17th ASCE Engineering Mechanics
Conference, Newark, June 13-16.
[16] Galvao, J.C.A., Portella, K.F., Joukoski, A., Mendes, R., Ferreira E.S.,
2011. Use of waste polymers in concrete for repair of dam hydraulic
surfaces. Construction and Building Materials 25, 1049-1055.
[17] Batayneh, M., Marie, I., Asi, I., 2007. Use of selected waste materials in
concrete mixes. Waste Management 27, 1870-1876.
[18] Rahmani, E., Dehestani, M, Beygi, M.H.A., Allahyari, H., Nikbin, I.M.,
2013. On the mechanical properties of concrete containing waste PET
particles. Construction and Building Materials 47, 1302-1308.
[19] Sadrmomtazi, A., Milehsara, S.D., Omran, O.L., Nik, A.S., 2016. The
combined effects of waste Polyethylene Terephthalate (PET) particles
and pozzolanic materials on the properties of self-compacting concrete.
Journal of Cleaner Production 112, 2363-2373.
[20] Zakaria, M., Cabrera, J.G., Performance and durability of concrete made
with demolition waste and artificial fly ash-clay aggregates. Waste
Management 16, 151-158.
[21] Debieb, F., Kenai, S., 2008. The use of coarse and fine crushed bricks as
aggregate in concrete. Construction and Building Materials 22, 886-
893.
[22] Oner, A., Akyuz, S., 2007. An experimental study on optimum usage of
GGBS for the compressive strength of concrete. Cement and Concrete
Composites 29, 505-514.
[23] Elinwa, A.U., Mahmood, Y.A., 2002. Ash from timber waste as cement
replacement material. Cement and Concrete Composites 24, 219-222.
[24] Khatib, J.M., 2005. Properties of concrete incorporating fine recycled
aggregate. Cement and Concrete Research 35, 763-769.
[25] Ganjan, E., Khorami, M., Maghsoudi, A.A., 2009. Scrap-tyre-rubber
replacement for aggregate and filler in concrete. Construction and
Building Materials 23, 1828-1836.
[26] Katz, A., 2003. Properties of concrete made with recycled aggregate
from partially hydrated old concrete. Cement and Concrete Research
33, 703-711.
[27] Siddique, R., 2003. Effect of fine aggregate replacement with Class F
fly ash on the mechanical properties of concrete. Cement and Concrete
Research 33, 539-547.
[28] Eldin, N.N., Senouci, A.B., Rubber-tire particles as concrete aggregate.
Journal of Materials in Civil Engineering, 5, 478-496.
[29] Hwang, K., Noguchi, T., Tomosawa, F., 2004. Prediction model of
compressive strength development of fly-ash concrete. Cement and
Concrete Research 34, 2269-2276.
[30] Freiesleben Hansen, P., Pedersen, J., 1985. Curing of Concrete
Structures. CEB Information Bulletin 166, 42 p.
[31] JCGM/WG1, BIPM, IEC, IFCC, ILAC, ISO, IUPAC, IUPAP, OIML,
JCGM 100:2008 (GUM 1995 with minor corrections): Evaluation of
measurement data – Guide to the expression of uncertainty in
measurement, First edition, 2008.
INDEX
automotive application, 125

A
Accelerated UV Test, 71 B
acid, 12, 23, 25, 77
acrylonitrile, 35 Bacillus subtilis, 40
additives, 5, 100, 130, 145, 146 barrier properties., 21
adhesion, 6, 22, 25, 30, 33, 34, 40, 43, base, 3, 6, 8, 13, 19, 21, 22, 35, 44, 46,
47, 70, 74, 77, 88, 92, 117, 121, 127, 63, 67, 70, 92, 119, 137, 139, 140,
130 141, 142
Adhesive Strength, 70, 71, 74, 77, 83 behaviors, 33, 92, 111, 126, 130, 141
adsorption, 31, 46 beneficial effect, 114
aerospace, 23, 97 benefits, x, 63, 66, 97, 103, 144
AFM, 114 biocompatibility, 6, 37, 40, 42
Ag nanoparticles, 36, 37, 38, 39, 40, 41, Biocompatible Properties, 22, 40
42 biodegradability, 7, 23, 106
aggregate replacement, x, 144, 145, 152, biodegradable materials, 101
162, 163 biomaterial, 15, 54
aggregation, 5, 38 biomedical applications, 15
alcoholysis, 103, 105 biomedical implants, 47
Alkaline Resistance Test, 71 biopolymers, 119
alkalinity, viii, 62, 64, 76, 88 biosensors, 33
amino, 23, 26 biotechnology, 23
ammonium, 9, 13, 20, 22 blends, 119, 131, 138
antimicrobial activity, 37, 40 bonding, 21, 26, 111, 116, 149
apoptosis, 22 bonds, 3, 26, 42, 149
aptitude, vii, 97, 101 breakup, 110, 111, 113, 114, 115, 119,
aspect ratio, 5, 6, 7, 15, 23, 27, 33, 111, 122, 127, 130, 133
113, 115, 116, 126, 133 brittleness, 118
atmosphere, 30, 71 bulk materials, 5
attachment, 7, 22, 40 bulk polymerization, 13
166 Index
burn, 102, 106 Color Difference, 69

butadiene, 35 combined effect, 163
combustion, 22, 102, 105
commercial, 8, 12, 23, 28
C compatibility, xi, 6, 8, 10, 41, 47, 66,
144, 145
cables, 62, 92 compatibilizers, 129, 131, 132
calorimetry, 20, 67, 113 competitiveness, 97
capillary, 129 complex interactions, 108
carbon, 5, 10, 23, 24, 25, 28, 31, 33, 34, complexity, 100, 104, 157
106, 121, 125, 126, 140, 141 composites, v, vii, viii, ix, 8, 9, 23, 24,
carbon atoms, 28 25, 27, 31, 32, 33, 34, 35, 36, 38, 41,
carbon fibers, 106, 125, 140 45, 51, 52, 53, 55, 56, 57, 58, 62, 64,
carbon nanotubes, 5, 23 88, 95, 96, 97, 99, 100, 101, 102, 103,
casting, 10, 12, 13, 15, 21, 29 104, 105, 106, 107, 108, 109, 110,
catalyst, 3, 10 111, 112, 113, 114, 115, 116, 117,
catalytic cracking, 103, 105 118, 119, 120, 121, 122, 123, 124,
cation, 9, 15 125, 126, 127, 128, 129, 130, 131,
cationic surfactants, 9 132, 133, 134, 135, 136, 137, 138,
cell death, 23 139, 140, 141, 162, 163
cell membranes, 41 composition, vii, 1, 99, 130, 139
cellulose, 3, 113, 128, 139, 140 composting, 119
cellulose fibre, 113, 140 compounds, 9, 118, 126
chain mobility, 14, 31, 39 compression, 29, 103, 105, 126, 130
chain scission, 107, 109, 110, 111, 112, compression molding, 103, 105
114, 115, 117, 119, 125, 127, 132 concrete, vii, x, 63, 143, 145, 146, 147,
challenges, viii, 3, 61, 63, 88, 101, 131, 148, 149, 150, 152, 153, 157, 159,
136 160, 161, 162, 163
charring, 32 conditioning, 135
chemical bonds, 26, 149 conductivity, 3, 23, 24, 26, 27, 28, 29,
chemical properties, 5, 99 31, 33, 35, 36, 37, 41
chemical stability, 6 constituents, vii, 8, 33, 96, 97, 98, 99,
chemical vapor deposition, 24, 28, 33 101, 108, 109, 114, 134, 160
chemicals, 103, 105, 127 construction, viii, ix, 61, 63, 95, 97, 138,
Clay modification, 8 145, 161
Cloisite®, 8, 12, 16, 22 consumption, 76, 88, 99, 101, 103, 104,
CMC, 89 105, 107
CO2, 21, 46, 91 contamination, 101, 104, 121, 127, 131,
coastal region, 72 134
coating, vii, viii, 3, 26, 27, 37, 44, 61, COOH, 26, 27, 66
62, 63, 66, 67, 68, 69, 70, 71, 72, 73, cooling, vii, viii, 19, 21, 62, 64, 65, 67,
74, 76, 77, 79, 81, 82, 83, 85, 86, 88, 68, 69, 72, 73, 75, 76, 89, 97
89, 92, 93, 112 copolymers, vii, 1, 2, 44, 45
coatings, viii, 2, 3, 37, 42, 47, 61, 62, 63, correlation, 92, 154, 155, 156, 157
67, 92 correlation coefficient, 154
cochlear implants, 15
Index 167
corrosion, viii, 24, 42, 61, 71, 92 124, 128, 130, 131, 132, 133, 134,
cosmetics, 37 137, 138
cost, vii, 7, 11, 33, 42, 66, 76, 88, 97, degradation mechanism, 77, 132
100, 103, 106, 107, 127, 131, 132, degradation process, ix, 96, 131
149 degradation rate, 26
coupling agent, 108, 109, 110, 112, 114, degree of crystallinity, 7, 20, 63, 65, 72,
117, 118, 119, 120, 129, 139 73, 75, 88, 115
covalent bonding, 26 dehydrochlorination, 111
crosslinking, 12, 26, 110, 111, 124, 130, density, 24, 32, 37, 99, 104, 106, 107,
133 112, 113, 129, 136, 137, 138, 139,
crystal growth, 39, 64, 65, 66 147, 148
crystalline, viii, 26, 62, 63, 64, 67, 74, depolymerization, 103
75, 76 deposition, 24, 28, 33, 34, 41
crystallinity, viii, 7, 20, 26, 31, 39, 62, designers, x, 144, 145, 146, 159, 161
63, 64, 65, 67, 68, 72, 73, 74, 75, 76, dielectric constant, 27
88, 107, 110, 112, 115, 117, 123, 127, differential scanning calorimeter, 68
132 differential scanning calorimetry, 20, 67,
crystallization, viii, 20, 35, 36, 39, 43, 113
45, 62, 64, 65, 66, 68, 72, 73, 74, 76, diffraction, 12, 67
88, 117 diffusion, 13, 21, 41, 103
crystallization kinetics, viii, 62, 73, 88 diffusion process, 103
crystallization temperature, 35, 117 directives, 98, 107, 108
CSR, xi, 144, 151, 152, 153, 156, 157, dispersion, viii, 2, 5, 6, 8, 9, 11, 12, 13,
158 15, 21, 24, 26, 28, 29, 31, 32, 34, 35,
curing process, 42 37, 39, 40, 41, 44, 47, 110, 111, 114,
current prices, 105 119, 123, 129, 155
cycles, 71, 85, 88, 107, 109, 110, 111, distribution, 26, 33, 38, 41, 44, 45, 107,
112, 113, 114, 115, 116, 118, 119, 108, 111, 113, 122, 124, 125, 127,
120, 121, 123, 126, 127, 128, 132, 131, 132, 133, 139
134 diversity, 104, 130
cyclohexanol, 63 DSC, 19, 20, 26, 31, 35, 41, 67, 69, 113
cytotoxicity, 22 ductility, 117, 126
durability, viii, 4, 61, 63, 64, 76, 83, 88,
89, 163
D
debonding, 117 E
decomposition, 19, 26, 30, 39, 45, 91,
99, 102, 127 E. coli, 40
decomposition temperature, 26, 39, 127 Elast-Eon E5-325, 15, 16, 17, 19, 20
deformation, 25, 126, 161 elastomers, 2, 3, 4, 47, 51
degradation, viii, ix, 11, 12, 18, 26, 30, electrical conductivity, 23, 24, 27, 28,
61, 63, 77, 83, 88, 96, 100, 101, 102, 29, 31, 33, 36, 37
103, 104, 105, 107, 108, 109, 111, electrical properties, 8, 23, 27, 28, 29,
113, 114, 115, 119, 121, 122, 123, 31, 33, 36, 41
electrical resistance, 27
168 Index
electricity, 102 fiber content, 34, 110, 113, 115, 121,

electromagnetic, 3, 33 122, 125, 126, 141
electron, 11, 12, 23, 130 fiber diameter, 107, 110, 133
electron microscopy, 11, 12 fiber length, 34, 107, 110, 111, 113, 115,
elongation, 13, 14, 15, 16, 25, 30, 34, 38, 116, 119, 121, 122, 123, 124, 125,
42, 43, 82, 108, 110, 112, 113, 116, 127, 133
118, 123, 125, 126, 127, 130, 131, fibers, 25, 33, 41, 97, 102, 103, 104,
133 106, 107, 109, 110, 111, 113, 114,
encapsulation, 6 115, 116, 117, 118, 119, 121, 122,
endothermic, 68 124, 125, 127, 128, 129, 132, 133,
endotherms, 20 134, 140, 141
energy, 5, 8, 9, 10, 22, 33, 37, 42, 47, 71, fibrillation, 119
76, 88, 97, 98, 99, 102, 103, 105, 106, filler surface, 134
112, 124, 130, 131 fillers, x, 4, 30, 36, 42, 96, 100, 106,
energy consumption, 76, 88, 103, 105, 120, 127, 128, 145
106 film degradation, 77
energy recovery, 98, 99, 102, 124 films, 10, 37, 42, 43, 45
engineering, 2, 73 fire resistance, 127
environment, viii, 62, 63, 71, 81, 83, 88, fire retardancy, 29
89, 101, 131, 160 first generation, 107
environmental awareness, 63 flame, vii, 1, 4, 21, 27, 32, 46
environmental conditions, 120 Flame Retardant Properties, 21, 27, 32
environmental factors, 64 flame retardants, 27
environmental impact, x, 98, 106, 107, flammability, 7, 22, 36
144 flax, 114, 115, 140, 141
environmental regulations, 121 flax fiber, 141
environments, viii, 62, 63, 85, 92 flexibility, viii, x, 4, 24, 35, 61, 63, 98,
epitaxial growth, 28 103, 105, 121, 144, 145
equipment, 97, 101, 108, 115, 122, 123 flour, 109, 112, 137, 139
ethyl acetate, 63 fluidized bed, 102
ethylene, 77 foam, 96, 122, 128, 129
ethylene glycol, 77 food production, 37
execution, xi, 144 force, 6, 13, 24, 64, 70
exfoliated structures, 6, 8 formation, 5, 10, 11, 13, 20, 21, 26, 28,
experimental design, x, 144, 145 29, 34, 37, 38, 46, 47, 64, 67, 81, 111,
exposure, 40, 71, 85, 86, 87, 104, 112 130
extracellular matrix, 39, 40 fracture toughness, 16
extraction, x, 91, 144 France, 96
extrusion, 4, 10, 11, 12, 14, 103, 107, free radicals, 41, 130
111, 112, 114, 116, 123, 141, 142 free volume, 46
friction, 34, 42, 91
FTIR, 14, 69, 79, 80, 81, 113, 114
F fuels, 102, 103, 105
Fumed silica, 42
fiber bundles, 116 functionalization, 6, 24, 29, 37
Index 169
high density polyethylene, 113, 137, 139

G high strength, 4
homogeneity, 103, 104, 108, 109, 133
gasification, 102, 105
human, 37, 63
gel, 37, 42
human health, 63
geometry, 23, 114, 115, 127
humidity, 62, 71, 132, 134
glass, 4, 7, 20, 25, 26, 31, 38, 43, 66, 97,
hydrazine, 37
98, 106, 115, 116, 118, 120, 121, 122,
hydrocarbons, 63
123, 124, 127, 133, 135, 138, 140,
hydrogen, 3, 21, 42
141
hydrogenation, 102, 105
glass transition temperature, 7, 20, 25,
hydrolysis, viii, 62, 77, 83, 89, 103, 105
31, 38, 43, 66, 115
hydrophilicity, 12
glucose, 37
hydrophobic polymer matrix, 8
glycol, 77
hydrophobicity, 12
glycolysis, 103, 105
hydroxide, 35, 37
gold nanoparticles, 36
hydroxyl, viii, 3, 9, 13, 42, 62, 77, 129
governments, 98
hysteresis, 12, 13
grading, 146
granules, 147
graphene platelets, 23 I
graphene sheet, 23, 30, 31, 32
graphite, 5, 24, 28, 29, 32 identification, 69, 106, 161
gravity, 146, 147, 148 images, 74, 75, 79, 81, 88, 99
greenhouse gases, 105 impact, 63, 70, 76, 88, 98, 100, 106, 107,
grinding, 103, 109, 119, 126 109, 110, 111, 112, 115, 117, 121,
growth, 22, 24, 28, 39, 40, 41, 64, 65, 130, 133, 140
66, 108 Impact Resistance, 70, 76
growth rate, 64, 65, 66 impact strength, 110, 111, 133, 140
guidelines, 47 implants, 2, 15, 37, 47
improvements, 4, 15, 26, 30, 35, 46
impurities, 99, 130
H in situ polymerization, 10, 12, 24, 29,
30, 42, 43
hard segments, vii, 1, 2, 3, 10, 13, 20,
incompatibility, 3, 5, 21, 101, 104, 107
26, 30, 38, 44
indentation, 42
hardness, 25, 42, 43, 75
industry, 3, 107, 138, 145
HDPE, 66, 112, 120, 139, 147, 148
infrastructure, 106
heat capacity, 67
inhibition, 22
Heat Cycle Test, 71
inhomogeneity, 127, 129, 134
heat release, 27, 32
initial state, 101
heat transfer, 26, 73
injection molding, 4, 103, 105, 109, 112,
hemicellulose, 128
113, 115, 116, 118, 123, 125
hemp, 116, 120, 136, 137
inorganic fillers, x, 96, 100, 106, 127,
hemp fiber, 136
128
heterogeneity, 104
insulation, 15, 47
hexafluoropropylene (HFP), 42
integrity, ix, 62, 89
170 Index
intercalated structures, 8 low temperatures, 112

interface, 5, 6, 24, 33, 37, 46, 47, 70, 75,
96, 107, 117, 118, 119, 125
interfacial adhesion, 25, 30, 33, 34, 47, M
117, 121, 127, 130
interfacial bonding, 116 macromolecular chains, 73
interfacial reactivity, 5 macromolecules, 10
interference, 33 macroradicals, 26
interphase, 6, 35 magnesium, 35
ion-exchange, 9 magnitude, 32, 41, 42
ions, viii, 8, 9, 13, 19, 38, 41, 62, 77 majority, 100, 133
irradiation, 130 management, x, 98, 144, 162
isocyanate, 3, 13, 30, 32, 66, 112 manufacturing, 100, 105, 106
issues, 101, 138 mass, 22, 24, 71, 77, 78, 83, 84, 102,
107, 146, 147, 148, 151
material degradation, 115
J material surface, 22
materials, viii, ix, 2, 3, 4, 5, 6, 7, 15, 33,
Japan, viii, 8, 61, 62, 63, 66, 67, 68, 69, 36, 37, 38, 40, 44, 47, 61, 62, 63, 66,
70, 71, 85, 86, 87, 89, 90, 91, 92, 93 67, 77, 95, 96, 97, 98, 99, 100, 101,
102, 103, 105, 107, 108, 109, 114,
119, 122, 123, 124, 125, 126, 127,
K 129, 130, 131, 132, 133, 134, 135,
136, 138, 141, 145, 149, 151, 160,
KBr, 69 161, 163
kenaf, 117, 120, 139 matrix, vii, viii, 2, 3, 5, 6, 8, 9, 10, 11,
ketones, 63 13, 14, 15, 18, 20, 21, 23, 24, 26, 27,
kinetics, viii, 30, 62, 73, 88 29, 30, 31, 33, 35, 36, 37, 38, 39, 41,
42, 43, 44, 45, 47, 96, 97, 101, 102,
L 104, 107, 108, 110, 111, 113, 114,
115, 116, 117, 118, 119, 121, 123,
laboratory tests, 152 124, 125, 127, 128, 130, 131,132,
laser ablation, 24 136, 140
Layered alumosilicate clay, 7 measurement, 68, 70, 86, 87, 113, 146,
lead, 7, 8, 10, 23, 47, 99, 102, 103, 105, 150, 153, 154, 159, 161, 164
128, 130, 131, 133, 134, 146, 159 mechanical degradation, 103, 105, 110,
legislation, x, 144, 145 114, 119, 123, 129, 133, 134
leisure, ix, 95, 97 mechanical properties, vii, x, 4, 12, 14,
lending, 24, 29, 34, 44 15, 16, 22, 24, 28, 29, 33, 38, 42, 43,
life cycle, ix, x, 96, 98, 127, 144 44, 47, 66, 97, 107, 108, 109, 110,
lifetime, 121, 134 111, 112, 113, 114, 116, 119, 123,
light, 62, 74, 79, 80, 81, 85, 86, 104 126, 127, 129, 134, 136, 137, 139,
lignin, 128 144, 146, 159, 160, 161, 162, 163
liquid fuels, 102, 105 mechanical stress, ix, 62, 82, 89, 108,
localization, 22 123, 125
low density polyethylene, 107, 138 media, 37
Index 171
medical, 37, 40, 41 monolayer, 28

medicine, 37, 40 monomer molecules, 10
melt, 3, 4, 6, 10, 11, 13, 20, 24, 25, 28, monomers, 6, 9, 103, 105
29, 30, 34, 36, 40, 43, 77, 114, 118, montrillomonite, 7
123, 124, 133 morphology, 3, 6, 8, 14, 18, 41, 132
melt flow index, 114 mortar, x, 63, 83, 84, 144, 145, 146, 147,
melt intercalation method, 10, 13 148, 149, 150, 152, 153, 157, 159,
melting, ix, 4, 20, 26, 31, 35, 39, 45, 66, 160, 161, 162
68, 77, 95, 97, 112, 128, 130 MSW, 98
melting temperature, 4, 20, 26, 31, 35, multi-block, 2
39, 66, 131 multiphase materials, 5
membranes, 41, 46 multi-wall carbon nanotubes, 23
memory, 23, 24, 33 municipal solid waste, 98
metal, 3, 7, 29, 38, 41, 70, 101
metal nanoparticles, 41
methyl groups, 10 N
MFI, 114
Micro FTIR, 69 nanocomposites, vii, viii, 2, 14, 26, 48,
micrometer, 42 54
microorganisms, 41 nanofibers, 23, 33, 36
microphase separation, 3 nanomaterials, 23
microscope, 69, 70, 74, 78, 81 nanomedicine, 37
microscopy, 11, 67 nanometer, 3, 8, 42
microstructure, 4, 140, 141 nanometer scale, 3
Ministry of Education, 48 nanometers, 33
mixing, 3, 10, 11, 12, 14, 15, 17, 25, 29, nanoparticles, vii, 1, 2, 3, 4, 5, 6, 8, 18,
31, 34, 37, 40, 44, 77, 85, 108, 110, 27, 36, 37, 38, 39, 40, 41, 44, 46
114, 119, 122, 123 nanoribbons, 29, 32
modifications, 37, 46, 108, 127, 129, nanorods, 36, 41
132 nanostructures, 33, 36
modulus, 4, 9, 12, 15, 17, 20, 21, 22, 24, nanotube, 3, 23, 31
25, 29, 34, 38, 44, 109, 110, 112, 113, nanowires, 36, 41
115, 116, 117, 120, 123, 125, 126, natural fiber, 102, 106, 116, 118, 119,
127, 133 121, 128, 129, 132, 134, 138
moisture, 104, 107, 121, 125, 129, 131, nettle, 118
134 next generation, 47
molecular mass, 108 NH2, 12, 26, 27
molecular mobility, 39 NIR, 104
molecular structure, 46, 111 nitrogen, 10, 30, 32
molecular weight, 2, 3, 18, 38, 66, 67, non-abrasiveness, 106
107, 109, 110, 111, 114, 115, 117, nucleating, 20, 36, 39, 112, 117, 127
119, 123, 125, 126, 127, 132 nucleating agent, 39
molecules, 9, 10, 13, 43, 46, 107, 129, nucleation, 39, 65
132 nuclei, 64
molybdenum, 32 nucleus, 22
172 Index
polycarbonate, 3, 18, 21, 44

O polyester, 3, 25, 30, 31, 38, 39, 40, 43,
46, 66, 67
optical microscopy, 67
polyether, 3, 13, 19, 21, 22, 44, 45, 46
optical properties, 23, 36
poly-ether-ether-ketone, 125
optimization, 47, 129, 157
polyethylen, 141
organ, 8, 10, 12, 13, 18, 21
polyethylene terephthalate, viii, 61, 63,
organic compounds, 3, 63
123
organic fibers, 129
polylactide, 25, 113, 140
organic polymers, 37
polymer chain, 7, 8, 9, 11, 13, 18, 39,
orientation, 3, 6, 23, 124
43, 107, 110, 115
Outdoor Exposure Test, 71
polymer chains, 8, 9, 11, 13, 43, 107,
oxidation, 24, 37, 110, 114, 131
110, 115
oxygen, 22, 102
polymer composite material, 101
polymer composites, 33, 36, 38, 96, 101
P polymer films, 37
polymer materials, 130, 136
packaging materials, 2 polymer matrix, viii, 2, 5, 6, 8, 9, 10, 11,
panels, 122 14, 18, 24, 25, 26, 27, 28, 33, 36, 37,
paper, 33, 101 42, 47, 101, 103, 104, 107, 110, 121,
pelletizing, 103 124, 125, 128, 130, 131
percolation, 27, 29, 31, 36, 41 polymer matrix-filler compatibility, 47
permeability, 7, 21, 23, 30, 32, 42, 46 polymer molecule, 9, 10, 13
PET, v, vii, viii, 61, 62, 63, 66, 67, 68, polymer nanocomposites, 48
69, 72, 76, 77, 78, 79, 81, 82, 83, 85, polymer properties, 8, 47
86, 88, 89, 123, 128, 137, 142, 147, polymer structure, 103
148, 161, 162, 163 polymeric materials, 62, 63
petroleum, 97, 119 polymerization, 5, 10, 12, 13, 18, 24, 29,
pH, 23 30, 42, 43, 83, 102, 105
phase separated microcomposite, 8 polymer-nanoparticle interactions, 4, 9
phosphate, 27 polyolefins, 119, 129, 137
photographs, 78, 81 polypropylene, 22, 109, 110, 112, 114,
physical and mechanical properties, 110, 116, 117, 118, 122, 126, 137, 139,
139 140, 141
physical properties, 4, 6, 14, 37, 122, polystyrene, 112, 141
127 polyurethane, vii, 2, 4, 13, 18, 20, 21,
physicochemical properties, 40 31, 32, 43, 45, 48, 50
plastics, x, 46, 98, 102, 104, 106, 135, polyurethanes, vii, 1, 2, 4, 46, 52
136, 137, 144, 145, 149, 161, 162 population, 21, 150, 160
platelets, 8, 10, 18, 23, 65 Powder Coating, v, 61, 67, 90, 93
polar, 10, 21, 36, 46 precipitation, 91, 104
poly(dimethylsiloxane), 3, 4, 14 preparation, vii, viii, 2, 3, 5, 7, 8, 12, 13,
poly(vinyl chloride), 109, 137, 139 16, 21, 27, 28, 33, 38, 40, 42, 44, 47,
polyamide, 66, 77, 122, 125, 140 91
polybutadiene, 35 prevention, 37, 89, 91
Index 173
principles, ix, x, 95, 144, 145 Recycling, v, 95, 97, 99, 100, 102, 103,
processing, ix, 6, 12, 13, 23, 24, 25, 29, 104, 105, 106, 121, 135, 136, 137,
47, 95, 98, 99, 101, 102, 105, 107, 138, 139, 140, 141, 142, 161, 162
108, 109, 110, 115, 119, 122, 126, regenerated cellulose, 113
128, 132, 133, 136, 139, 140, 141 regression, 156, 157
product design, x, 144, 145 regression model, 157
propagation, 152 regulations, 121
propane, 46 reinforcement, vii, 15, 33, 34, 96, 97,
propylene, 45 100, 101, 102, 103, 104, 106, 107,
prostheses, 40 108, 117, 121, 124, 125, 126, 127,
protection, viii, 61, 62, 127 129, 130
purification, 37 reliability, 106, 146, 157
PVC, 109, 111, 120 reprocessing, ix, 96, 103, 107, 109, 110,
pyrolysis, 102, 103, 105 113, 114, 115, 116, 119, 120, 122,
123, 124, 126, 127, 128, 129, 130,
134, 139
Q requirements, vii, viii, 1, 4, 61, 63, 99,
107, 132
quality control, ix, 95 researchers, 31, 45, 145, 146, 149, 150,
quantification, 161 160
quantitative review, x, 144, 145, 151, residues, 20, 22, 30, 32, 41, 101, 106,
157 131
quaternary ammonium, 13 resins, 77, 89
resistance, vii, ix, 4, 7, 16, 24, 26, 27,
R 29, 33, 37, 42, 44, 45, 47, 62, 66, 70,
71, 76, 77, 83, 85, 86, 87, 88, 89, 97,
radiation, 25, 71, 86, 87, 90 107, 108, 117, 127, 162
radicals, 41, 108, 130 resource management, 98
raw materials, 38, 103 resources, 97, 99
reactants, 4 response, 7, 40
reaction medium, 5 restrictions, 101
reactions, 9, 14, 111, 130 rheology, 47, 107, 132
reactive extrusion, 10, 12, 14 rice hulls, 117
reactivity, 5, 22, 130 room temperature, 16
recommendations, iv, 108 roughness, 70, 74, 77, 78, 87
recovery, 24, 97, 98, 99, 102, 103, 124 rubber, 15, 35, 46, 91, 163
recycling, vii, viii, ix, 4, 61, 63, 66, 95,
96, 97, 98, 99, 100, 101, 102, 103,
S
104, 105, 106, 107, 108, 109, 110,
111, 112, 113, 114, 115, 116, 117, safety, 101, 160, 161
118, 119, 120, 121, 122, 123, 124, Salt Water Spray Test, 70
125, 126, 127, 128, 129, 130, 131, saturation, 112
132, 133, 134, 135, 136, 137, 138, savings, x, 134, 144, 145
139, 140,141, 142, 148 scaffolds, 2
scanning calorimetry, 20, 67, 113
174 Index
segregation, 45, 104 stabilization, 18, 26, 37

selectivity, 46 stabilizers, 38, 129
SEM, 79, 81, 114 standardization, 159
semiconductor, 23 standardized protocol, x, 144, 145
sensitivity, 99, 100, 102, 107, 119, 128, state, 11, 28, 40, 47, 100, 101, 123, 135
134, 152, 157 statistical inference, x, 144
shape, viii, 2, 5, 6, 23, 24, 27, 33, 36, 37, statistics, 99, 149, 150, 153, 154
38, 41, 47, 50, 79, 92, 103, 105, 146, steam treatment, 103, 105
155, 161 steel, viii, 61, 62, 63, 67, 69, 70, 75, 88,
shear, 6, 10, 11, 13, 45, 103, 107, 109, 89, 92, 162
110, 115, 119, 123, 129, 130, 131, stiffness, 12, 13, 29, 118
133 storage, vii, 15, 17, 25, 29, 31, 33, 34,
showing, 66, 73, 75, 79, 81, 110, 114, 38, 43, 97, 139
115, 123, 126, 160 strength, vii, 3, 4, 7, 9, 12, 13, 14, 15,
silanol groups, 43 16, 24, 25, 29, 33, 34, 39, 42, 43, 70,
silica, 3, 7, 42, 43, 44, 45, 46, 102 71, 74, 77, 82, 83, 84, 97, 109, 110,
silica dioxide, 7 111, 112, 113, 114, 116, 117, 118,
silica particles, 44, 45, 46 120, 123, 125, 126, 127, 133, 139,
silver, 7, 37, 38, 40 145, 146, 149, 161, 163
single-wall carbon nanotubes, 23 stress, ix, 13, 14, 22, 23, 24, 33, 62, 65,
SiO2, 28, 42 75, 82, 83, 89, 107, 116, 131, 133
sisal, 116, 137 Stress-Strain, 78
society, 66, 98 strong interaction, 7, 18, 45
sodium, 35, 37 structure, vii, viii, 2, 4, 8, 10, 11, 12, 13,
sodium dodecyl sulfate, 35 14, 18, 22, 27, 32, 47, 66, 69, 74, 78,
sodium hydroxide, 37 82, 83, 89, 103, 111, 119, 136, 139
soft segments, viii, 2, 15, 21, 30, 35, 42, structure formation, 10
44, 46 structure-property relationships, vii, viii,
sol-gel, 37, 42 2, 12
solid waste, 98 substitution, 145, 146
solubility, 21, 38 substrate, 28, 67, 73, 75, 77
solution, 3, 4, 6, 12, 24, 29, 31, 32, 44, surface area, 5, 9, 22, 23, 28, 33, 36, 37,
63, 71, 83, 100, 130, 131 41, 42, 45, 69, 77, 81, 103
solvent casting, 10, 12, 13, 15, 21 surface energy, 8, 9, 37, 42, 47
solvents, 4, 12, 38, 102, 105 surface layer, ix, 62, 79, 81
solvolysis, 103, 105 surface modification, 6, 24, 37
species, 21, 40, 130 surface properties, 9, 45
specific gravity, 146, 147, 148 Surface Roughness, 70
specific surface, 37, 41, 42, 45 surface structure, 74
specifications, 47, 160 surface tension, 5
spectroscopy, 113 surface treatment, 117, 132, 134
Specular Gloss, 69, 73 surfactants, 6, 8, 9, 20, 23, 24, 35, 37
stability, 4, 5, 7, 18, 25, 26, 30, 35, 37, sustainable development, ix, 95, 98, 106
39, 42, 45, 110, 111, 112, 118, 120, swelling, 10, 13, 129
127, 129, 131 synergistic effect, 27, 29, 32, 35, 47, 110
Index 175
synthesis, 6, 13, 47 thermoplastic elastomers, 2, 4, 47

synthetic polymers, vii, 1 thermoplastic polyurethane
nanocomposites, vii, 2
Thermoplastic polyurethanes, vii, 1, 2,
T 46
thermoplastics, ix, x, 2, 4, 46, 96, 97, 98,
techniques, ix, 6, 26, 37, 47, 67, 96, 97, 100, 103, 105, 106, 107, 108, 119,
98, 100, 104 121, 124, 125, 126, 128, 129, 131,
technologies, vii, 4, 66, 101, 106, 121, 132, 133, 134, 141, 144, 145, 146,
135, 136 149, 151, 157, 161
telecommunications, vii, viii, 61, 62, 88 thermosets, 97, 100, 103
TEM, 38, 44 tissue engineering, 2
temperature, 3, 6, 12, 15, 17, 19, 20, 23, titanium dioxide, 127
26, 30, 31, 33, 35, 39, 43, 45, 64, 65, toluene, 63
66, 67, 71, 72, 73, 77, 78, 81, 88, 89, total product, 98
101, 102, 104, 108, 116, 117, 121, toxicity, 3, 22, 37, 42, 131
123, 125, 126, 128, 133, 134 transformation, 102
TenasiTech Pty Ltd., 14 transition temperature, 7, 17, 20, 25, 31,
tensile strength, 4, 9, 12, 13, 14, 15, 16, 38, 44, 66, 115
24, 25, 30, 34, 39, 42, 43, 82, 110, transmission electron microscopy, 12
113, 114, 118, 123, 125, 126, 127, transport, 21, 22, 44
139 transportation, 97
testing, 70, 71, 82, 83, 84, 85, 86, 87, 88 treatment, viii, 6, 34, 62, 63, 77, 99, 103,
texture, 15, 87 105, 112, 117, 129, 132, 141
TGA, 18, 35, 113
thermal analysis, 15
thermal decomposition, 26, 45 U
thermal degradation, 12, 26, 30, 107,
109, 128, 131, 133 uniform, 8, 26, 32, 33, 67, 74, 77
thermal history, 3, 7 urea, 44, 45
thermal properties, 33, 35, 39, 111, 117, urethane, 10, 16, 45
120, 123, 125, 134 UV light, 62, 85, 86
Thermal Properties, 18, 26, 30, 35, 39,
45
thermal resistance, 45 V
thermal stability, 4, 7, 18, 25, 26, 30, 35,
39, 45, 118, 127, 129 vapor, 21, 23, 24, 28, 33, 34, 35, 36, 42,
thermal treatment, 63 91
thermodynamic incompatibility, 3, 21 variables, 145, 149, 150, 151, 156, 157
thermograms, 19, 20 variations, 114, 115, 118, 120, 133, 157,
thermogravimetric analysis, 18, 112 161
thermoplastic, vii, viii, ix, x, 2, 4, 7, 35, VDF, 42
47, 61, 62, 63, 66, 69, 72, 85, 88, 89, vegetal fibers, 106, 117
93, 95, 97, 106, 124, 126, 128, 132, versatility, 4
140, 142, 143, 145, 146, 152, 160, vinyl chloride, 109, 137, 139
162 vinylidene fluoride, 42
176 Index
viscosity, 3, 10, 12, 20, 29, 43, 102, 105, weight loss, 18, 45
107, 116, 117, 118, 119, 123, 125, weight reduction, 134
131, 133 Wisconsin, 139
viscosity breaking, 102, 105 wood, 99, 107, 108, 109, 110, 111, 112,
volatile organic compounds, 3, 63 136, 137, 139, 141
worldwide, x, 144, 145
W
X
waste, ix, x, 96, 97, 98, 99, 100, 101,
102, 103, 104, 105, 107, 109, 111, xenon, 71, 90
126, 127, 129, 130, 131, 132, 134,
136, 141, 143, 145, 146, 148, 149,
161, 162, 163 Y
water, ix, 21, 23, 37, 42, 62, 63, 67, 70,
71, 85, 89, 92, 103, 112, 129, 139, yellow birch fibers, 107
140, 146, 149 yield, 103, 105, 112, 126, 141
water absorption, 23, 139
water vapor, 21, 23, 42 Z
waterborne TPUs, 3
WAXS, 21 zinc, 75, 92
wear, 47, 101, 104

Thermoplastic Composites

Uploaded by

Copyright:

Available Formats

Thermoplastic Composites

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Thermoplastic Composites

Uploaded by

Copyright:

Available Formats

MATERIALS SCIENCE AND TECHNOLOGIES

Additional books in this series can be found on Nova’s website

Additional e-books in this series can be found on Nova’s website

NOTICE TO THE READER

Library of Congress Cataloging-in-Publication Data

Published by Nova Science Publishers, Inc. † New York

Thermoplastics composites are widely used nowadays. Their matrices are

applications. Properties of TPU-based nanocomposites strongly depend on the

facilitate comparability and compatibility of future results. This protocol aims

Marija V. Pergal1,*, Milica Balaban2, Biljana Dojčinović1

nanocomposites strongly depend on the preparation method, the type and

Keywords: polyurethanes, nanocomposites, thermoplastics, nanoparticles,

consist of long-chain macrodiol with a molecular weight of 1000-4000 g/mol.

Figure 1. The chemical structure of the thermoplastic polyurethanes based on poly(ε-

TPUs combine the mechanical properties of conventional thermoplastics

THERMOPLASTIC POLYURETHANE NANOCOMPOSITES:

[40]. According to the literature, in order to improve dispersion of

a) Nanoparticle precursors are preloaded into polymer matrix and then

The advantages of in situ methods are better dispersion control of

[57]. The introduction of nanoparticles can enhance or reduce thermal stability

number of polymer-nanoparticle interactions increases with decreasing the

Figure 2. Schematic illustration of different types of polymer/clay composites.

MODIFICATION OF CLAY LAYERS

PREPARATION METHODS FOR

dispersion of clay within the polymer matrix, as dispersion mainly depends on

Figure 3. Transmission electron microscopy micrographs of cross-sectional views of

Nowdays, commercial TPUs are usually manufactured by a reactive

The strong industrial and academic interest in TPU/clay NCs has

segment based TPU-based NCs (prepared by solvent casting) containing two

temperature, especially for NCs containing 4 wt.% of LuO. They reported, on

Figure 4. Stress-strain curves of pure PDMS/PHMO-based TPU (i.e., Elast-Eon E5-

Generally, including clay in the polymer matrix improves thermal

Using differential scanning calorimetry (DSC), Cruz and Viana [103]

segments and interphases due to their thermodynamic incompatibility. Based

Generally, adding organoclay into TPU matrices improves their barrier

Flame Retardant Properties

carbonaceous-silicate char that controls the flammability [116]. It was known

A few studies have revealed that the release of quaternary alkyl

TPU/CARBON NANOTUBE COMPOSITES

exposed to increased temperatures, infrarred radiation and electrical current.

The introduction of 2 wt.% of MWCNT amide derivate into

hard segment domain of TPU/MWCNT-NH2 was higher, while that for

Flame Retardant Properties

Bourbigot et al. [153] studied the flame retardancy of TPU CNT

It is important to possess a three-dimensional network of conductive

Fahim et al. [173] prepared TPU/graphene NCs with differing amounts of

graphene oxide (GO), thermally reduced graphene oxide (TRG) and

Studies on graphene TPU-based NCs have suggested improved thermal

Lee et al. [167] prepared waterborne polyurethane (WTPU) with

Recently, more researchers report the application of graphene and

was not as effective as solution mixed thermally reduced graphene oxide in

Flame Retardant Properties

Mechanical properties of CNF-based polymer composites depend on the

which could act as stress concentration regions leading to weakness in

increase in melting temperature of soft segments with added CNF could be

stabilizers to prevent agglomeration [212]. Metal nanoparticles must be

According to the literature, adding Ag nanoparticles into TPU matrix has a

interfacial interactions between Ag nanoparticles and the HWPU matrix,

Adding Ag nanoparticles into the TPU matrix significantly improves the

Furthermore, for TPU-extracellular matrix/Ag bio-NCs, small changes (up to