Embed Size (px)
Citation preview
POLYMER COMPOSITE MATERIALS BASED ON COCONUT FIBRES SUBTITLE OF THE PHD
Le Quan Ngoc TRAN
Dissertation presented in partial fulfilment of the requirements for the degree of Doctor of Engineering
December 2012
Members of the Examination Committee: Prof. Paul Sas, Chair Prof. Ignace Verpoest, Promoter Dr. Aart Willem Van Vuure, Promoter Prof. Christine Dupont-Gillain Prof. Jin Won Seo Prof. Bart Blanpain Prof. Peter Van Puyvelde Prof. Stepan Lomov
© 2009 Katholieke Universiteit Leuven, Groep Wetenschap & Technologie, Arenberg Doctoraatsschool, W. de Croylaan 6, 3001 Heverlee, België Alle rechten voorbehouden. Niets uit deze uitgave mag worden vermenigvuldigd en/of openbaar gemaakt worden door middel van druk, fotokopie, microfilm, elektronisch of op welke andere wijze ook zonder voorafgaandelijke schriftelijke toestemming van de uitgever. All rights reserved. No part of the publication may be reproduced in any form by print, photoprint, microfilm, electronic or any other means without written permission from the publisher.
ISBN 978-94-6018-615-8
D/2013/7515/3
Cover image: SEM image of fracture surface of coir epoxy composite, showing defibrillation of the coir fibres.
i
Acknowledgements
My career in composite materials started from a great occasion when I met Prof. Ignace
Verpoest in Vietnam in 2001. He opened a wide door for me to enter into the
interesting field of composite materials by receiving me into the Master program
EUPOCO. Since that time, I have learned from him not only valuable knowledge and
experience but also his kindness in supporting people. In the past four years of my PhD,
as my promoter, Prof. Ignace Verpoest has provided me patient guidance, enthusiastic
encouragement and useful inputs for this research work. I would like to take this
opportunity to express my deepest gratitude to him for all his supports.
I would like to express my great appreciation to my co-promoter Dr. Aart W. Van
Vuure for his knowledge, advice and available time for guiding me. Working in the
natural fibre composites group, I enjoy very much both his leadership and friendship. I
also thank for his patience in correcting my papers and the first draft of this dissertation.
I would like to offer my special thanks to Prof. Christine Dupont-Gillain for her kind
help in building up the method for wetting measurement of natural fibres, and giving
useful comments and inputs for papers and the thesis manuscript.
My grateful thanks are extended to Prof. Stepan Lomov, Prof. Bart Blanpain and Prof.
Peter Van Puyvelde, as members of advisory committee and examination committee,
for their advice on my research, reading and providing valuable remarks for the
manuscript. I would like to thank other members of the jury, Prof. Jin Won Seo, for her
effort to read my thesis and evaluate my work and Prof. Paul Sas for being the chairman
of my thesis defense.
The four years research consisting of many experiments and testing would have never
been successful without the technical assistance of Kris Van de Staey, Bart Pelgrims,
Manuël Adams, Danny Winant, Sylvie Derclaye, Yasmine Adriaensen and Michel
Genet. I greatly appreciate their help. Additional thanks to Gregory Pyka for his
training on using SEM-CT. I would also like to thank Aniko Lantos, Huberte Cloosen
and other MTM secretaries for their kind help in important administrative work.
ii
My warmest thanks to CMG friends who are always willing to offer me their help,
especially, my officemates Carlos Fuentes, Lina Osorio, Eduardo Trujillo, Ichiro Taketa
and Yasmine Mosleh for sharing nice time (in and out the office) during these years.
The thesis work of Linde De Vriese, Elisa Melcon Miguel, Delphine Depuydt and
Laurena Van Oproy is contributed to this work. They have had high motivation in
working with coir fibre composites, and obtained good initial results which help to have
further studies in this thesis. Thank you very much.
My acknowledgements are addressed to KU Leuven for providing I.R.O Scholarships
and Belgian Science Policy Department (BelSPO) for supporting our research. I also
wish to thank the staff involved in the BelSPO-MOST project Prof. Bui Chuong, Dr.
Truong Chi Thanh for their advice and providing the fibres for the research.
My family and I would have never had such a nice life in Leuven without the care and
support of Belgian and Vietnamese friends. I am really thankful to Mr. Jo Mariën and
his wife Claire Mariën. Con cảm ơn Chú Thiếm Kim rất nhiều về sự quan tâm giúp đỡ
con và gia đình trong suốt thời gian ở Bỉ. Cám ơn các chiến hữu Cần Thơ đã giúp đỡ
và chia sẻ vui buồn những lúc xa quê.
Finally, I want to express my deepest thank to my parents for their support and
encouragement throughout the years. My special thanks go to my wife Loan and my
little daughter Au Lam who was born in Leuven, for their love and support. They are
the driving force in my life. This thesis is dedicated to them.
iii
Abstract
The interest in using natural fibres in composite materials has greatly increased over
the past decades thanks to their good mechanical properties in combination with
environment-friendly characteristics. In this research, Vietnamese coir fibres are
studied and modified for use in composite materials. To be efficiently used in
composite materials, the microstructure and the mechanical properties of coir fibres
are first characterised. Secondly, the surface of natural fibres has a complex
morphology with chemical heterogeneity and relatively high roughness, which
strongly influences the fibre-matrix interfacial adhesion. Therefore, it is important to
acquire a systematic understanding of the fibre-matrix interfacial interactions in
composites. Lastly, unidirectional (UD) composites of coir fibre in both
thermoplastic and thermoset matrices are examined to evaluate the possible value of
coir fibre for composites.
The microstructure of technical coir fibres is examined using SEM and SEM-CT.
The results show that technical coir fibres comprise plenty of elementary fibres and
a lacuna at the centre. The elementary fibre is built up by two main cell walls which
consist of bundles of microfibrils aligned in a high angle to the fibre axis. Coir fibre
appears to have high porosity at 22 to 30%. The mechanical properties of coir fibre
are determined in tensile tests including single fibre tensile testing with optical strain
mapping and single fibre tensile testing using different test lengths. The results of
both methods indicate that coir fibres are not very strong and stiff, but have high
strain to failure.
An integrated physical-chemical-micromechanical approach is implemented to
investigate the fibre-matrix interfacial compatibility and adhesion of the coir fibre
composites. In this study, the interface between untreated and alkali treated coir
fibres and various thermoplastics is characterised. The differences of fibre surface
chemistry and properties of the matrices in terms of surface energy and potential
chemical reactions are considered. Wetting measurements of the fibres and the
matrices are carried out to obtain their static equilibrium contact angles in various
liquids, and these are used to estimate the surface energies comprising of different
components. The work of adhesion is calculated for each composite system,
accordingly. Also, fibre surface chemistry is examined by X-ray photoelectron
spectroscopy (XPS) to have more information about functional groups at the fibre
surface, which assists in a deeper understanding of the interactions at the composite
interfaces. To determine the quality of the composite interfaces, single fibre pull-out
tests and transverse three point bending tests are performed on UD composites to
iv
measure interfacial shear strength and interfacial strength (mode I) respectively. The
results suggest that the higher interfacial adhesion of coir fibres with polyvinylidene
fluoride compared with polypropylene can be attributed to higher fibre-matrix
physico-chemical interaction corresponding with the work of adhesion. Whilst the
improvement of interfacial adhesion for coir fibres with maleic anhydride grafted
polypropylene compared with polypropylene can probably be attributed to a
chemical adhesion mechanism. In addition to the specific results for coir fibre
composites, the integrated physical-chemical-micromechanical approach to
investigate and improve fibre-matrix interface has been developed. This knowledge
can be applied to study the interface of other natural fibre composite systems.
Mechanical properties of UD coir fibre composites with both thermoplastic and
thermoset matrices are assessed by tensile tests in fibre direction, flexural tests and
unnotched Izod impact tests. In agreement with the interface evaluation, higher
flexural strength and stiffness are found in the alkali treated fibre composites,
probably thanks to the better interfacial adhesion. The impact strength of coir
polypropylene composite is not significantly different from that of neat polymer,
while the coir fibres can improve the toughness of epoxy by minimum a factor of
three, when the impact strength is considered as toughness indicator.
An initial study on coir-bamboo fibre hybrid composites is carried out to investigate
the hybrid effect of tough coir fibre and brittle bamboo fibre in composites. With a
low bamboo fibre fraction, a hybrid effect with an increase of composite strain to
failure is obtained, which can be attributed to the high strain to failure of the coir
fibres. Meanwhile, the bamboo fibres provide high stiffness and strength to the
composites. The results show a potential for coir-bamboo hybrid composites, which
justifies further study on this topic.
v
Samenvatting
In de afgelopen decennia is de interesse in het gebruik van natuurlijke vezels voor
gebruik in composietmaterialen sterk toegenomen, vanwege hun goede mechanische
eigenschappen in combinatie met mileuvriendelijke karakteristieken. In dit
onderzoek worden Vietnamese cocosvezels onderzocht en gemodificeerd voor
gebruik in composietmaterialen. Om een efficiënt gebruik van de vezels toe te laten
in composietmaterialen, worden eerst de microstructuur en de mechanische
eigenschappen van de cocosvezels gekarakteriseerd. In de tweede plaats heeft het
oppervlak van natuurvezels een complexe morfologie met chemische heterogeniteit
en een relatief grote ruwheid. Deze factoren beïnvloeden sterk de vezel-matrix
interfase adhesie. Daarom is het belangrijk om een systematisch begrip te verwerven
van de vezel-matrix interfase interacties in composieten. Tenslotte worden
unidirectionele (UD) composieten van cocosvezel in zowel thermoplastische als
thermohardende matrices onderzocht, om een beoordeling te maken van de
mogelijke waarde van cocosvezels voor gebruik in composieten.
De microstructuur van technische cocosvezels is onderzocht met SEM en SEM-CT.
De resultaten laten zien dat de technische cocosvezels bestaan uit een reeks van
elementaire vezels met een lacuna in het centrum. De elementaire vezels zijn
voornamelijk opgebouwd uit twee celwanden die bestaan uit bundels van micro-
fibrillen die een grote hoek maken met de vezelas. Cocosvezels blijken een hoge
porositeit te hebben van 22 tot 30%. De mechanische eigenschappen van cocosvezel
worden bepaald met behulp van trekproeven, zowel met trekproeven op
enkelvoudige technische vezels met behulp van optische rekmetingen, als met
trekproeven op technische vezels met een reeks van testlengtes. De resultaten van
beide methoden geven aan dat cocosvezels niet zozeer sterk en stijf zijn, maar wel
een hoge breukrek hebben.
Een geïntegreerde fysisch-chemische-micromechanische aanpak werd gebruikt om
de vezel-matrix compatibiliteit en adhesie te onderzoeken in cocosvezel
composieten. In deze studie werd de interfase gekarakteriseerd van zowel
onbehandelde als met alkali behandelde cocosvezels in een reeks van
thermoplastische matrices. Verschillen in oppervlaktechemie van de vezels en
eigenschappen van de matrices in termen van oppervlakte-energie en mogelijke
chemische reacties werden beschouwd. Bevochtigings experimenten van de vezels
en de matrices werden uitgevoerd om hun statische evenwichts contacthoeken te
bepalen in verscheidene vloeistoffen. Met deze contacthoeken werden de
oppervlakte-energieën en de verschillende componenten hiervan bepaald, voor
vi
zowel vezels als matrices. Vervolgens wordt hiermee de theoretische adhesie arbeid
bepaald voor elk composiet systeem. Verder wordt de oppervlakte-chemie van de
vezels bepaald met behulp van Röntgen fotoelectron spectroscopy (XPS), om meer
informatie te verkrijgen over functionele groepen aan het vezeloppervlak. Hiermee
kunnen interacties in de composiet interfase beter begrepen worden.
Om de kwaliteit van de composiet interfase te bepalen worden pull-out testen
uitgevoerd op enkelvoudige technische vezels, alsmede transversale buigproeven
uitgevoerd op unidirectionele composieten. Dit om respectievelijk de afschuifsterkte
van het grensvlak te bepalen en de mode I interfase sterkte. De resultaten suggereren
dat de hogere interfase sterkte van cocosvezel met polyvinylidene fluoride
vergeleken met polypropyleen kunnen worden toegeschreven aan sterkere vezel-
matrix fysisch-chemische interactie, in overeenstemming met de theoretische
adhesie-energie. Tegelijkertijd wordt de verbetering in interfase adhesie voor
cocosvezel met maleinezuur anhydride gemodificeerde polypropeen toegeschreven
aan een chemisch adhesie mechanisme.
Naast specifieke resultaten voor cocosvezel composieten, werd in deze studie de
geïntegreerde fysisch-chemische-micromechanische aanpak ontwikkeld om de
vezel-matrix interfase te onderzoeken en te verbeteren Deze kennis kan gebruikt
worden om de interfase te onderzoeken in andere (natuurvezel) composieten.
De mechanische eigenschappen van unidirectionele cocosvezel composieten met
zowel thermoplastische als thermohardende matrix werden onderzocht door
trekproeven in vezelrichting, buigtesten en Izod impacttesten zonder kerf. In
overeenstemming met de interfase evaluatie, worden hogere buigsterkte en stijfheid
gevonden in alkali behandelde composieten, waarschijnlijk door betere interfase
adhesie. De impactsterkte van cocosvezel polypropeen composiet is niet significant
verschillend van die van onversterkte polypropeen, terwijl cocosvezel de taaiheid
van epoxy kan verbeteren met minimaal een factor drie (indien impactsterkte wordt
gebruikt als indicator van taaiheid).
Een initiële studie werd uitgevoerd op cocosvezel-bamboevezel hybride
composieten, om het hybride effect te onderzoeken in composiet van taaie
cocosvezels en sterke maar brosse bamboevezels. Met een lage bamboevezel fractie
wordt een positief hybride effect gevonden voor de composiet breukrek, wat kan
worden toegeschreven aan de hoge breukrek van de cocosvezels. Tegelijkertijd
geven de bamboevezels hoge stijfheid en sterkte aan de composieten. De resultaten
vii
laten het potentieel zien van cocos-bamboe hybride composieten, wat een verdere
studie van dit onderwerp ondersteunt.
viii
ix
List of abbreviations
3PBT Three-point bending test
CTE Coefficient of Thermal Expansion
IFSS Interfacial shear strength
MAPP Maleic grafted anhydride polypropylene
MFA Microfibril angle
MKT Molecular-Kinetic Theory
PET Polyethylene terephthalate
PP Polypropylene
PVDF Polyvinylidene fluoride
SEM Scanning electron microscope
SEM-CT X-ray tomography in SEM
T3PB Transverse three-point bending
Tcoir Treated coir fibre
Ucoir Untreated coir fibre
UD Unidirectional
XPS X-ray photoelectron spectroscopy
x
xi
List of symbols
Viscosity
Density
Elongation of fibre
Displacement caused by slippage and machine compliance
Measured displacement of clamps
Transverse E-modulus
Fibre modulus at infinite fibre length
E-modulus of fibre
E-modulus of fibre calculated for fibre solid material
E-modulus of matrix
Debonding force
Maximum applied load
Displacement frequency
Fibre volume fraction
Volume fraction of fibre solid material
Matrix volume fraction
Work of adhesion
Work of adhesion following acid-base approach
Work of adhesion following geometric mean approach
Fibre embedded length.
Longitudinal coefficient of thermal expansion of fibre
(or ) Liquid surface tension
Surface energy base component
(or ) Solid surface energy
Surface energy acid component
Surface energy acid-base component
Interfacial energy
xii
Surface energy Lifshitz – van de Waals component
Surface energy dispersive component
Surface energy polar component
Static/ equilibrium contact angle
Advancing contact angle
Receding contact angle
Fibre strength calculated for the fibre solid material
Matrix stress at fibre failure strength
Apparent interfacial shear strength
Debonding shear stress
Frictional stress
Tc Crystallisation temperature
Tg Glass transition temperature
Tm Melting temperature
V% Volume fraction
wt% Weight fraction
α Compliance factor
Load
Crack length
Fibre wetted perimeter
Measurement velocity
Shear-lag parameter
Contact angle/dynamic contact angle
Displacement length
xiii
Table of Contents
Acknowledgement ...........................................................................................................i
Abstract ........................................................................................................................ iii
Samenvatting .................................................................................................................. v
List of Abbreviations ..................................................................................................... ix
List of Symbols .............................................................................................................. xi
Table of Contents ........................................................................................................ xiii
Chapter 1. Introduction...................................................................................................... 1
1.1 General introduction................................................................................................ 2
1.2 Literature review ..................................................................................................... 4
1.2.1 Natural fibres................................................................................................... 4
1.2.2 Coir fibres ..................................................................................................... 11
1.2.3 Coir fibre composites .................................................................................... 19
1.2.4 Interface of natural fibre composites .............................................................. 21
1.2.5 Concluding remarks ...................................................................................... 29
1.3 Problem statement and the goal of thesis ............................................................... 29
Thesis structure ............................................................................................................. 32
References .................................................................................................................... 32
Chapter 2. Microstructure and mechanical properties of coir fibres.................................. 37
2.1 Introduction .......................................................................................................... 38
2.2 Materials and methods .......................................................................................... 38
2.2.1 Coir fibres ..................................................................................................... 38
2.2.2 Investigation of fibre microstructure using SEM and SEM-CT ...................... 41
2.2.3 Measurement of fibre density ........................................................................ 43
2.2.4 Single fibre tensile tests ................................................................................. 45
2.3 Results and discussion ........................................................................................... 48
2.3.1 Fibre surface and fibre internal microstructure ............................................... 48
2.3.2 Density of coir fibres ..................................................................................... 57
2.3.3 Tensile mechanical properties of coir fibres ................................................... 58
2.4 Conclusions .......................................................................................................... 63
References .................................................................................................................... 64
Chapter 3. Wetting analysis and surface characterisation of coir fibres ............................ 65
3.1 Introduction .......................................................................................................... 66
3.2 Materials and methods .......................................................................................... 68
3.2.1 Materials ....................................................................................................... 69
3.2.2 Dynamic contact angle measurement ............................................................. 71
3.2.3 Static equilibrium contact angle approximation ............................................. 73
3.2.4 Fibre surface energy estimation ..................................................................... 76
3.2.5 Fibre surface characterisation using X-ray photoelectron spectroscopy .......... 78
3.3 Results and discussion ........................................................................................... 80
3.3.1 Contact angle measurements.......................................................................... 80
3.3.1.1 Fibre wetted perimeter ............................................................................... 80
3.3.1.2 Advancing dynamic contac angles ............................................................. 82
xiv
3.3.1.3 Effect of liquid absorption on the contact angles ........................................ 86
3.3.1.4 Advancing static contac angles approximation using the MKT .................. 86
3.3.1.5 Static contac angles from relaxation experiments ....................................... 87
3.3.2 Surface energy of coir fibre ........................................................................... 90
3.3.3 Surface chemical analysis of coir fibre........................................................... 93
3.4 Conclusions .......................................................................................................... 94
References .................................................................................................................... 94
Chapter 4. Interfacial adhesion and compatibility of coir fibre composites....................... 97
4.1 Introduction .......................................................................................................... 98
4.2 Materials and methods ........................................................................................ 100
4.2.1 Materials ..................................................................................................... 100
4.2.2 Wetting analysis .......................................................................................... 101
4.2.3 Single fibre pull-out test .............................................................................. 105
4.2.4 Three point-bending test of UD composites ................................................. 110
4.3 Results and discussion ......................................................................................... 112
4.3.1 Surface enegies and the work of adhesion .................................................... 112
4.3.2 Fibre surface chemistry ............................................................................... 116
4.3.3 Fibre-matrix interfacial adhesion with pull-out test ...................................... 118
4.3.3.1 Load-displacement curves and apparent IFSS .......................................... 118
4.3.3.2 Two interfacial parameters fitting theoretical Fmax to the experimental data ...
................................................................................................................ 122
4.3.4 Transverse strength and interface properties of composites .......................... 126
4.3.5 IFSS verus transverse bending strength........................................................ 127
4.3.6 Work of adhesion in relation with practical adhesion ................................... 128
4.4 Conclusions ........................................................................................................ 129
References .................................................................................................................. 131
Chapter 5. Mechanical properties of unidirectional coir fibre composites ...................... 133
5.1 Introduction ........................................................................................................ 134
5.2 Materials and methods ........................................................................................ 134
5.2.1 Materials ..................................................................................................... 134
5.2.2 Production of composite samples ................................................................. 135
5.2.3 Test methods ............................................................................................... 139
5.2.4 Determination of coir fibre volume fraction ................................................. 141
5.2.5 Coir/bamboo hybrid composites .................................................................. 142
5.3 Results and discussion ......................................................................................... 143
5.3.1 Flexural properties of UD composites .......................................................... 143
5.3.1.1 Longitudinal properties ............................................................................ 143
5.3.1.2 Transverse properties ............................................................................... 148
5.3.2 Tensile properties of UD composites ........................................................... 151
5.3.3 Impac strength of UD composites ................................................................ 156
5.3.3.1 Impact strength of UD coir/PP and UD coir/epoxy composites ................ 156
5.3.3.2 Effect of fibre volume fraction and fibre treatment on the impact strength of
UD coir fibre epoxy composites ............................................................... 158
5.3.4 Tensile properties of UD coir/bamboo hybrid composites ............................ 159
5.4 Conclusions ........................................................................................................ 163
References .................................................................................................................. 164
xv
Chapter 6. Conclusions.................................................................................................. 165
6.1 General conclusions ............................................................................................ 166
6.1.1 Microstructure and mechanical properties of technical coir fibres ................ 166
6.1.2 Wetting measurements and surface energy estimation of the fibres .............. 167
6.1.3 Fibre-matrix interfacial compatibility and adhesion ..................................... 168
6.1.4 Performance of coir fibre composites........................................................... 168
6.2 Future work......................................................................................................... 169
Apendix A .................................................................................................................. 171
Apendix B .................................................................................................................. 173
Curriculum Vitae
List of publication
xvi
Introduction 1
Chapter 1
Introduction
Chapter 1 2
1.1 General introduction
Composite materials, by which is usually meant fibre reinforced polymers, are used
in a wide range of applications from aerospace, automotive and construction to
leisure and sporting goods, where high mechanical properties in combination with
light weight make them greatly attractive materials. Moreover, these materials excel
in chemical resistance, durability and design flexibility. Generally, a typical
composite material consists of a continuous phase, known as matrix, and a
reinforcement phase, typically in the form of fibres, distributed within it. As a rule,
the reinforcement fibres ensure the strength and rigidity of the material, whereas the
matrix keeps the fibres in desired orientation and maintains the shape of the part.
The matrix is also a medium for stress transfer between the fibres, and protects them
from environmental impacts such as chemicals, humidity and temperature. In this,
the fibre-matrix interface is an important element, where stress transfer from the
fibre to the matrix and vice versa takes place.
Besides using synthetic fibres, particularly carbon, glass and aramid fibres, natural
fibres such as flax, jute, coconut fibre (coir), hemp and bamboo have received a
growing interest for application in polymer composites during the last decades.
These fibres are available in large amounts, at low cost, have low energy utilisation
and are renewable and biodegradable. In most cases the specific properties of natural
fibre composites have been found to compare favourably with these of glass fibre
composites [1, 2]. In this research, the focus will be on coir fibres.
Generally, coir fibres are considered as a low-value product which is mainly used to
make mattresses, doormats or brushes. Other applications are coir nettings and
geotextiles for soil protection and erosion control, and rubberised coir mats used
in upholstery padding for automobiles. Nowadays, there are three good reasons to
use natural fibres, namely: economy, ecology and society; hence, coir fibres can be a
good candidate as reinforcement for composite materials. They are cheaper in cost
than other natural fibres, easily extracted, and available in large amounts. For basic
mechanical properties, flax, hemp, bamboo and jute can contribute their high
strength and stiffness to composites, while coir with high elongation to failure can
ameliorate the composite toughness [3].
Introduction 3
As mentioned above, the interfacial adhesion between fibre and matrix plays an
important role in the final composite mechanical properties. The knowledge of the
interface has been developed for the existing synthesized fibre composites, but
research has not really focused yet on natural fibre composites. Natural fibres are
usually extracted from different parts of the plant, which typically have different
surface chemical compositions, leading to different properties in terms of surface
energy and potential for chemical reactions. Apparently, the fibre surface is rough
and chemically heterogeneous, which affects the interfacial properties when the
fibres are used in composite materials. Therefore, a fundamental understanding of
the fibre-matrix interfacial compatibility and adhesion is necessary. The first
important concern is wetting between fibre and matrix to create a good fibre-matrix
contact. This strongly depends on the surface energies of the fibre and the matrix.
Subsequently, the fibre-matrix adhesion comprising different levels of interfacial
interactions, from molecular scale to bulk composite level, is an essential element to
be studied for natural fibre composites.
Terminology
The terms and definitions, which are used frequently in the thesis, are described in
the following glossary:
Elementary fibre is the structural unit of the plant, composing of cell walls and
formed out of cellulose crystalline microfibrils connected by amorphous lignin and
hemicellulose.
Technical fibre is the extracted fibre after a standard extraction process, which is
used as reinforcement for composites. A technical fibre consists of numerous
elementary fibres, and its configuration mainly depends on the biological structure
of the plant. Figure 1-1 displays the technical fibre in various plants. In case of
coconut, the technical coir fibre naturally presents as such as in the husk and it is
surrounded by organic tissues, while in case of flax or bamboo, its technical fibre
has a configuration depending on the fibre extraction method which separates a
bundle of elementary fibres to form a technical fibre.
Single fibre, in this thesis, is referred to as one technical fibre as it is used in some
tests.
Chapter 1 4
Figure 1-1. Representative images of a technical fibre (circles) (a) coir from coconut shell (b) flax
fibre from the stem [4] (c) bamboo fibre from the culm [5].
1.2 Literature review
The state of the art of the research on natural fibres, coir fibre composites and
interfaces in natural fibre composites is reviewed in the following sections. Firstly,
an overview of natural fibres and their characteristics is presented. Then, a
comprehensive description of coir fibres is followed, which comprises the fibre
extraction processes, the morphology and chemical composition of coir fibres, and
their physical and mechanical properties. Coir fibre composites will be reviewed
with a focus on the composite impact properties. Finally, there is a discussion on the
fibre-matrix interface adhesion and some fibre treatments for improvement of the
interface quality in natural fibre composites.
1.2.1 Natural fibres
Figure 1-2. Overview of natural fibres [2, 6, 7]
Introduction 5
Natural fibres generally are fibres which are not synthesised but obtained from
nature using different fibre extraction processes. Natural fibres can be divided in
subgroups based on their origins as plant fibres, animal or mineral fibres. Figure 1-2
shows the three subgroups highlighting some common fibres used in composite
materials.
Figure 1-3. Worldwide production of natural fibres, in million ton (Sources: FAOSTAT, 2009 and
FAO, 2009) [6].
The production volumes of natural fibres are shown in Figure 1-3. It can be seen that
cotton is the most important natural fibre with a high quantity in the market. Besides
this, a high market share is found for the other plant fibres such as jute, flax, coir,
hemp and sisal, which have been used in composite materials. Used as
reinforcement, the mechanical properties of the fibres are the main concern, which
are decided by the structure of the fibres and their chemical compositions. These
characteristics of common natural plant fibres will be described in the following
sections.
Chapter 1 6
1.2.1.1 Chemical composition
Figure 1-4. Schematic presentation of the hierarchy of a typical cell wall, from a simplified model
of a primary cell wall down to the microfibril structure of crystalline cellulose, to the cellulose
molecule with its monomer units. (After Akin [6])
Natural plant fibres of the stem, leaf, fruit or seed of the plant, typically have a cell
wall structure and comprise of cellulose, hemicelluloses, lignins and aromatics,
waxes and other lipids, pectin, ash and water-soluble compounds. Figure 1-4
presents a typical cell wall with main components and a schematic representation of
their organisation. Climatic conditions and age not only influence the structure of
the fibres but also the chemical composition [6, 8]. To have efficient processing and
quality improvement of the fibres, a good understanding of the fibre chemistry is
Introduction 7
necessary. In Table 1-1, the major chemical components of common natural fibres
are presented.
Table 1-1. Chemical composition of common natural plant fibres [6, 8-14]
Fibre/
Composition
(%) Coir Cotton Bamboo Hemp Jute Flax Sisal
Cellulose 32-53 82-96 26-43 57-92 51-84 60-81 43-88
Hemicelluloses 0.2-0.3 2-6 15-30 6-22 12-24 14-21 10-15
Lignin 40-45 0-1.6 21-31 2.8-13 5-14 2-5 4-14
Pectin 3-4 0-7 - 0.8-2.5 0.2-4.5 0.9-3.8 0.5-10
Wax - 0.6 - 0.7-0.8 0.4-0.8 1.3-1.7 0.2-2
Water soluble 4.5 0.4-1 - 0.8-2.1 0.5-2 3.9-10.5 1.2-6
Cellulose is the essential component of plant fibres. It is a linear condensation
polymer of glucose consisting of a linear carbohydrate polymer of β-1,4-linked
glucose units (d-anhydroglucopyranose units). The basic repeating unit of cellulose
is the dimer cellobiose, which comprises of two glucose units bound by the β-1,4
linkage as well as intermolecular hydrogen bonds. Figure 1-5 shows a typical
structure of cellulose. The properties of cellulose are decided by how glucose is
bound in the linear polymer. The cellulose structure consists of thousands of glucose
units, which can stack together to form crystal with intramolecular hydrogen bonds
providing a stable polymer with high tensile strength. Cellulose occurs in plant cell
walls as microfibrils (e.g. 2–20 nm diameter and 100–40000 nm long) providing a
linear and structurally strong framework. The mechanical properties of natural fibres
depend on its cellulose type, because each type of cellulose has its own crystalline
unit cell geometry and the geometrical conditions determine the mechanical
properties [6, 8].
Chapter 1 8
Figure 1-5. Schematic presentation of cellulose, showing the linear nature of the polymer made of
glucose units: (A) cellulose unit; (B) structure of the dimer cellobiose; (C) cellulose molecule with
β-1,4 linkage between C atoms 1 and 4 (After Akin [6]).
Hemicellulose is reported to be the second most abundant carbohydrate of plant cell
walls after cellulose. It comprises a heterogeneous group of polysaccharides which
remains associated with the cellulose after lignin has been removed, and differs from
cellulose in both composition and structure. Firstly, hemicelluloses contain several
different sugar units whereas cellulose contains only 1,4- -d glucopyranose units.
They exhibit a considerable degree of chain branching, whereas cellulose is a strictly
linear polymer. Moreover, the degree of polymerization of native cellulose is ten to
one hundred times higher than that of hemicellulose. Hence, hemicelluloses are
generally in amorphous form with lower molecular weight than cellulose. They are
quite hydrophilic and mainly responsible for the moisture sorption behaviour of the
fibres [6, 8]. Figure 1-6 shows a schematic illustration of hemicelluloses and
celluloses together in a cell wall.
Introduction 9
Figure 1-6. A schematic cell wall, in which cellulose and hemicellulose are arranged into layers in a
matrix of pectin polymers [6]
Lignin is a compound of complex hydrocarbon polymers with both aliphatic and
aromatic constituents, and has an amorphous structure. These compounds are very
diverse and present in many forms within plants and plant cell walls. In the structure
of a cell wall, lignin and hemicellulose are linked by covalent bonds, and celluloses
are often bonded by lignin or the lignin/hemicellulose complex [6, 15].
Pectin consists mainly of heteropolysaccharides, which consist essentially of
polygalacturon acid. Pectin amounts are often low in natural plant fibres, but they
are strategically located within the plant tissues as a matrix to hold tissues, including
fibres, together [6] (Figure 1-6).
Waxes consist of long chain alcohols which are insoluble in water as well as in
several acids. They are usually located on the cuticle of the plant or on the fibre
surface as a protective barrier which prevents drying and microbial entry inside the
plant. However, the waxy layers influence the processing and quality of natural
fibres, and are normally removed to obtain good quality cellulose fibres.
1.2.1.2 Physical structure and mechanical properties of natural fibres
A technical natural fibre commonly consists of several cells (referred to as
elementary fibres). The cell is mainly formed out of crystalline microfibrils based on
cellulose (major load-bearing components in plant cell walls), which are connected
into a cell wall layer, by amorphous lignin and hemicellulose. Hemicelluloses are
Chapter 1 10
assumed to be the mediators between cellulose and lignin, as they can bind to
cellulose via hydrogen bonds and even covalently to lignin [16]. Multiples of such
cellulose–lignin/hemicellulose layers stick together to build up the cell wall (Figure
1-6). This structure can be considered as a composite, in which the cellulose crystals
play a role as reinforcement in a matrix of lignin/hemicellulose compound.
The cell wall layers can be of different thickness, chemical organisation and
orientation of the cellulose microfibrils (microfibril angle – MFA). Figure 1-7
presents schematics of the fibre cell (elementary fibre) consisting of several layers
with different MFA. The thickness of the cell wall layers and their cellulose MFA
play a dominant role in the mechanical properties of plant fibres.
Figure 1-7. Schematics of possible cell wall organisation in (A) wood fibres, (B) bast fibres, (C)
monocotyledonous plant fibres and (D) seed fibres. Black lines indicate orientation of cellulose
microfibrils; stress-strain curves of fibre with different density (E) and MFA (F) [6].
The mechanical properties of plant fibres depend on the organisation of cell walls in
terms of cell wall/lumen ratio and the cellulose MFA in the dominant cell wall
layers. In relation with fibre cross-section, higher density fibres are stiffer and
Introduction 11
stronger than the lower density ones. The elastic modulus and strain at failure of
fibres are also dependent on the MFA. A small MFA, in which cellulose fibrils are
oriented almost parallel to the axial direction, leads to a high modulus of elasticity,
whereas the stiffness is considerably reduced for higher MFA. In Figure 1-7, it can
be seen that the stress-strain curve shows a stiff and elastic response with a brittle
fracture at low MFA. For large cellulose MFA the interaction of the cellulose fibrils
with the matrix becomes more crucial for the overall mechanical behaviour of the
cell wall. Typically, the stress–strain curves of tissues and fibres with high
microfibril angles show a biphasic or triphasic behaviour [17, 18], as shown in
Figure 1-7E and 1-7F.
Table 1-2 shows the physical and mechanical properties of selected natural plant
fibres, which reflects the influence of the fibre structure on their mechanical
properties. For instance, the high MFA in coir fibres results at low stiffness and high
strain at failure. The high elongation at failure of coir fibres assists their relatively
high impact strength. It shows that nature is very smart, since the coconut fibres
need to prevent the coconut from breaking when it falls out of the tree.
Table 1-2. Physical characteristics and mechanical properties of common natural fibres (given
values from random single fibre or bundle tests) [6, 8-14]
Fibre/ Properties Coir Cotton Bamboo Hemp Jute Flax Sisal
Diameter (m) 100-460 12-20 200-400 16-50 30-150 11-20 50-200
Density (g/cm3) 1.1-1.3 1.5-1.6 1.4-1.5 1.4-1.6 1.3-1.5 1.4-1.5 1.0-1.5
MFA (o) 30-49 20-30 85-90 2-6.2 7-10 5-10 10-25
E-modulus (GPa), range
(most frequently published)
2.8-6
(5)
4.5-12.6
(8)
11-89
(30)
3-90
(65)
3-64
(30)
8-100
(70)
9-38
(12)
Tensile strength (MPa), range
(most frequently published)
95-270
(200)
220-840
(450)
140-1000
(500)
310-1110
(800)
190-800
(500)
343-1500
(700)
80-855
(600)
Elongation at break (%), range
(most frequently published)
15-50
(30)
2-10
(8) 2-3
1.3-6
(3)
0.2-3.1
(1.8)
1.2-4
(3)
1.9-14
(3)
1.2.2 Coir fibres
Coconut fibres are usually known under the name ‘coir’ fibres in literature, and are
obtained from the fruit of the coconut palm (Cocos nucifera L.) growing extensively
in tropical countries. Coconut palm is the most economically important cultivated
plant in over 93 countries situated in the tropical coastal ecosystem of the world,
providing more than 200 products or byproducts for human use. It occupies an area
Chapter 1 12
of approximately 12 million hectares globally, with an annual production of around
57 billion nuts [19]. The palms are mainly grown for the oil-rich copra (‘meat’)
contained inside the coconuts (Figure 1-8). In a mature coconut, the white meat (28
wt.%) is surrounded by a hard protective shell (12 wt.%) and a thick husk (35 wt.%).
This husk surrounding the large seed constitutes of 30 wt.% fibre and 70 wt.% pith
material (waste material from coir fibre industry, with high content of lignin) [20,
21]. Figure 1-8 shows the cross-section of a coconut consisting of the copra, the
core shell and the husk shell.
Figure 1-8. Coconuts from the palm and cross section of a coconut (adapted from [21]).
Traditionally, coir was extracted from husks that had been soaked for 6–9 months
(retted) in sea water or lagoon water and then beaten with a wooden mallet. The
fibres were used for production of ropes, yarns, mats, brushes and padding of
mattresses. Nowadays, the coir extraction processes have significantly improved, the
quality coir fibre being extracted either by wet processing (following retting
procedures) or mechanical decortications without soaking. The colour and properties
of coir fibre are not only dependent on the type of coconut palm, but also on harvest
time. White fibres are obtained from green coconuts which are harvested after about
6-7 months on the plant (the green coconuts have thin copra and mainly provide
coconut water for drinking). While brown fibre is obtained by harvesting fully
mature coconuts of 11-12 months when the nutritious layer in the seed is ready to be
processed into copra and desiccated coconut. The brown fibres are stronger but less
flexible than the white ones. Coir fibres are available in high quantity, and
considered as commodity in the world market. Their production is estimated at
around 1 million ton per year (FAOSTAT, 2009) at prices of order 30 to 40
Eurocents per kilo.
Introduction 13
1.2.2.1 Extraction of coir fibres
Extraction of coir fibres from coconut husk shells is mainly carried out in the
following steps: retting (the pre-treatment process in the traditional procedure),
extraction of coir fibre bundles, cleaning the coir bundles (removal of pith from coir)
and drying.
Retting is a microbial separation process, which consists essentially of soaking the
husk in water for a period. Depending on the condition of the husks and the nature
of the water, retting duration can vary from 2 to 9 months in the traditional process.
When the husks are mature and dry, the retting process takes nearly 6–9 months,
while it requires 2–3 months for green husks. Currently, the retting time is reduced
to 2-3 weeks thanks to an improved retting process, in which the husks are crushed
before soaking in water. Crushing the husks can help to increase the surface area in
contact with the water, and this accelerates the action of bacteria separating the fibre
bundles from pith tissues [6].
Extraction process
Following retting, the extraction process involves the breakdown and the separation
of the coir fibre bundles from the connecting tissues or pith in between the fibre
bundles and also from the outer exocarp (outer layer). Mechanical extraction of the
retted husks is often applied using various designed machines. The three following
types of machines are usually used for the extraction of various kinds of coir fibres
[6].
A decorticator is the first common machine for extracting the fibres from fresh
husks or husks that have been soaked for a few hours, and enables extraction of pith
tissues. The husks are mechanically beaten against a cylindrical cage made out of tor
steel bars. The rotary shaft fixed several plates consisting of sharp blades, facilitates
the holding and hammering of the husks (Figure 1-9). The disadvantage of this
machine is that the long coir fibres cannot be produced. Only mixed-grade coir is
produced by this machine.
Chapter 1 14
Figure 1-9. Coir fibre decorticator [22]
A defibre-ing machine can be used to produce bristle fibres which are considered as
long coir fibres. In the defibre-ing machine, the husk segments are gripped at the
edge of a large wheel, and then moved towards the picker drum. The sharp pins of
this drum remove the short fibres and pith, leaving the bristle coir (long fibres). The
first drum defibres half of the husk segment, which is then transferred to a second
wheel while the defibred part of the husk is held firmly by a conveyor chain. The
defibreing is completed by the second picker drum. The extracted fibres need to pass
through a cleaner drum or wash to remove the pith adhering to the bundles. This
type of machine is used for extracting the studied coir fibres in this thesis. So, the
details of extraction process will be presented in Chapter 2.
The last machine is a modified decorticator, in which the defibre-ing and the
decorticating processes are combined. Firstly, the husk is introduced to a picker
drum, in which the pith and exocarp of the husk are partly removed. And it is then
automatically transferred to a section similar to the decorticator for further removal
of pith by mechanical beating. This machine can be used with green husks, retted
brown husks or wetted husks, and produces mixed fibre bundles which have better
quality than the fibre bundles extracted by the decorticator alone.
Introduction 15
Cleaning and drying coir fibres
The received fibres after the extraction process are cleaned to remove the pith and
other attached tissues. The bristle coir or long fibre bundles have a smaller amount
of residual pith. Therefore, bristle fibre bundles are cleaned (hackled) by combing
through a set of steel spikes. While the mixed fibre bundles are fed into a cone-
shaped rotating screen sifter. By gravitational action, fibre bundles are separated
from the pith tissues. The coir fibre is then fed into a turbo cleaner, which consists of
fixed steel rods rotating at a high speed, for further cleaning. By centrifugal action,
remaining pith tissues and other waste attached to the fibre bundles are removed by
this mechanical process, and better-quality coir is obtained [6].
The cleaned fibre bundles are dried in a drying machine or under the sun to reduce
the moisture content to about 15%. For sun drying, it takes approximately 6 h,
during which time the coir is turned over several times to ensure a uniformly dried
product.
1.2.2.2 Morphology and chemical composition
Figure 1-10. A typical technical coir fibre and its fracture surface [21, 23].
The technical coir fibre (also referred to as coir fibre) typically is relatively
cylindrical consisting of numerous axially oriented elementary fibres which are
joined together, and often contains a central hollow channel which is named lacuna.
Coir fibres have a diameter in the range of 130-390 m, and the longest fibre length
is around 22 cm.
As a plant cell wall, the elementary fibre comprises several cell wall layers which
are built up by cellulose microfibrils and compound of hemicellulose and lignin. In
Chapter 1 16
Figure 1-11b, highly lignified elementary fibres are observed by lignin staining of
the coir fibre cross section. The diameter of elementary fibres was determined to be
on average 18 m with a relatively large lumen width of 12.5m. The length of
these elementary fibres was measured to be in the range of 0.9 to 1.06 mm [21, 24,
25].
Figure 1-11. Cross section of coir fibre (a) showing lacuna and lumens (b) a high content of lignin is
observed by lignin staining (in red) (adapted from [25, 26]).
Concerning chemical composition, coir fibre is composed of cellulose, lignin,
hemicellulose and a small amount of other substances. Table 1-3 shows the chemical
composition of coir fibres as reported by various literature sources. In comparison
with the other natural fibres (Table 1-1), coir fibres have a relatively high lignin
content. This result is consistent with the staining analysis of the fibre structure
shown in Figure 1-11b. It is also observed that the content of lignin is high in the
middle lamellae between elementary fibres.
Table 1-3. Chemical composition of coir fibres reported in literature.
Cellulose (%) Hemicellulose (%) Lignin (%) Reference
35-47 15-28 20-31 [27]
43 21 31 [28]
43-60 11.6-19 27.7-45 [29]
The surface of coir fibres has been observed by SEM (Figure 1-12a), which shows
randomly distributed organic tissues (attached pith) and ordered white dots (named
tyloses). The result of an energy dispersive X-ray spectroscopy measurement (EDS)
indicates that the white dots have high silica content, and can be removed by
chemical or mechanical treatments [30]. On the fibre surface, the presence of
longitudinally orientated cells with more or less parallel orientations is reported, and
the intercellular space is filled up by the binder lignin and fatty substances that hold
Introduction 17
cells firmly bonded in the fibre [31, 32]. The fibre surface is also reported to contain
a high fraction of waxes.
Figure 1-12. Coir fibre surface consisting of attached pith and rich in silica dots analysed by EDS
[21, 30].
1.2.2.3 Physical and mechanical properties of coir fibres
Physical properties
The colour of coir fibre varies from yellow to dark brown, mainly depending on the
coconut variety, maturity of the nuts, and the retting procedure. Having a high lignin
content, coir fibre is highly resistant to microbial attack and to sea water.
Coir fibres, as a natural fibre, absorb moisture from the surroundings when the dry
fibres are exposed to the atmosphere, they will take up moisture and reach
equilibrium. The moisture absorption results in changing the properties of coir such
as tensile strength, elastic recovery, electrical resistance, rigidity, etc. As a result of
absorption of water, the fibres tend to swell, altering their dimensions, and thus
leading to changes in the size, shape, stiffness. Similarly, when the fibres in
moisture equilibrium are exposed to a dry atmosphere, moisture is lost to the
Chapter 1 18
surroundings to establish a new equilibrium [6]. The amount of water in coir fibre
can be expressed in terms of moisture content or moisture regain, in which moisture
content indicates the amount of water present in a moist sample and moisture regain
is the amount of water that a completely dry fibre will absorb from the atmosphere
at a standard condition of 20 oC and a relative humidity of 65% (expressed as a
percentage of the dry fibre weight). Nawaratne [33] found that the moisture content
of fresh-water-retted fibre samples was 10.20% with a moisture regain of 11.31%,
while the moisture content in sea-water-retted coir was 7.92% with a moisture
regain of 8.60%. In comparison with the other common natural fibres (flax, jute,
hemp), coir fibres show relatively less moisture absorption.
Van Dam [24] also reported the thickness swelling of coir due to water absorption
when the fibres were placed in water. The result shows that the thickness swelling is
in the range from 22% to 34% after 15 minutes in water. The longitudinal swelling,
which increases the fibre length, was measured to be 0.9% after 15 minutes.
Mechanical properties
For application as composite reinforcement, the mechanical properties of the coir
fibres will strongly decide the final properties of the composite. As shown in Table
1-1, in comparison with the other natural fibres, coir fibre has low tensile strength
and E-modulus but high elongation at failure. This is explained by low cellulose
content in combination with a high MFA. The MFA in a range of 30-49° of the S2
cell wall layer determines the characteristics of the fibre [23, 34] (the second
secondary cell wall layer S2 is typically the thickest layer of the elementary fibre
cell wall). The increasing MFA decreases the fibre stiffness but increases the strain
at failure. This interrelation enables the fibres to adjust both stiffness and toughness
by shifting the cellulose fibril orientation in the cell wall [35, 36]. Nature is smart
since coir fibres have to prevent the nut from breaking when it falls out of the tree.
Thus, the strength of the fibres is not as important as the energy absorption at impact
[3].
Introduction 19
Figure 1-13. Typical stress-strain curve of coir fibre in a tensile test, and fibre fracture surface
showing the pull-out of elementary fibres, also giving indications of the high MFA [24].
Figure 1-13 presents a typical stress-strain curve of a coir fibre with a failure strain
of approximately 25%. The fracture surface of the fibre shows pulled out elementary
fibres, explaining the high plastic deformation of coir during a tensile test. The
mechanical properties of coir fibres are also dependent on fibre type (genetic variety
and maturity of the nut) which may result in different diameter, cellulose content,
cell wall thickness of the elementary fibres and MFA. In Table 1-4, the mechanical
properties of various types of coir fibres are shown.
Table 1-4. Tensile properties of different types of coir fibres.
Fibre species Tensile strength (MPa)
E-modulus (GPa)
Strain at failure (%)
Reference
Vietnamese white coir 162-192 3.44 26.1-42.4 [3]
Brown coir* 186-343 4.94 24.5-59.0 [3]
Philippine white coir 120-304 4.6-6 20-44 [24]
Brazilian coir 129-155 2.2-2.4 29.9-34.9 [34]
Indian coir 140-225 3-5 25-40 [37] *unknown origin
1.2.3 Coir fibre composites
Natural fibre composites have been studied and used in industry, especially in
automotive applications, thanks to their good mechanical properties combined with
their light weight. This results in specific mechanical properties comparable to those
of glass fibre composites. From the above literature survey, coir fibres show a high
potential for application in polymer composites which are applied in impact loading
and do not require high strength and stiffness. In literature, a variety of studies on
coir fibre composites can be found, considering different matrices and fibre
geometries. Both traditional matrices (thermoplastics and thermosets) and
Chapter 1 20
biodegradable polymers were used with different fibre forms such as short fibres,
unidirectional fibres, woven mats, etc. Table 1-5 presents the mechanical properties
of various coir fibre composites published in literature.
Table 1-5. Mechanical properties of untreated coir fibre composites with different fibre geometries
and matrices.
Matrix Fibre geometry Fibre
fraction
Strength
(MPa) E-modulus
(GPa)
Impact
strength
(kJ/m²) Reference
PP short fibre (3mm) 20 wt% 26.5 - 52 J/m [38]
PP random long fibre 40 V% 10 1.3 22 [1]
LDPE short fibres 25 V% 26.2 0.78 - [37]
Polyester short fibre mat 20 V% - 2.5 7.44 [39]
Polyester random mat 45 wt% 39.8 ± 3.0 3.6 ± 0.2 - [40]
Epoxy random mat 17.9 ± 2.3 - 11.5 ± 1.0 [41]
PBS short fibre (10mm) 50 wt% 15.2 ± 1.0 2.1 ± 0.1 - [42]
Gluten short fibre (40mm) 15 wt% 53.2 ± 1.3 3.0 ± 0.2 - [43]
As seen in Table 1-5, in composites coir fibres are mostly used in the form of short
fibres or random fibre mats. Consequently, the composite strength and stiffness is
relatively low. Tensile strength of the composite is highly influenced by the
orientation of the fibre layers. The strength of fibres most effectively contributes to
the composite strength when the fibres are perfectly aligned in unidirectional
direction. Obviously, there is a lack of published results on unidirectional (UD) coir
fibre composites.
Hill et al. studied the impact properties of random coir fibre polyester composites,
and presented the influence of fibre weight fraction on composite impact strength
[40]. The results showed that the impact strength of the composite increases with a
factor of three compared to neat resin. At low fibre loading (less than 20 wt%), no
increase of impact strength is seen and there is an approximately linear increase
thereafter, followed by a decrease at the highest fibre loading (45 wt%) (Figure 1-
14). The impact strength of a composite is influenced by many factors including the
toughness properties of the reinforcement, the nature of the interfacial region, and
the frictional work involved in pulling the fibres from the matrix. In this case, the
tough coir fibres and the fibre-matrix interfacial interactions played a key role in the
impact strength of the composite. When the fibre loading (of tough fibres) increases,
Introduction 21
typically toughness enhancement occurs in composites. However, at high fibre
loading, the resin is prevented to wet completely the fibre bundles, which changes
the interfacial properties. Less fibres can be loaded properly or participate in energy
absorption by pull-out and the composite toughness goes down.
Figure 1-14. The variation in the impact strength with fibre loading for composites reinforced with
(line A) acetylated coir, (line B) unmodified coir, (line C) acetylated oil palm fibres, and (line D)
unmodified oil palm fibres [40].
The nature of the interphase region is of high importance in determining the
toughness of the composite. If the fibre–matrix interfacial strength is too low, poor
stress transfer occurs leading to a weak composite. On the other hand, a strong
interfacial adhesion allows efficient stress transfer, but produces a composite
exhibiting poor toughness properties, because more localisation of damage occurs
with less fibre pull-out.
1.2.4 Interface of natural fibre composites
Natural fibres extracted from different plants and different parts of plants typically
have different surface physico-chemical properties. Most natural fibres are relatively
hydrophilic, have a rough surface and are physico-chemically heterogeneous. The
fibre surface properties strongly influence the fibre-matrix interactions in the
composite. The first important concern is wetting between fibre and matrix to create
a good fibre-matrix contact. Subsequently, a strong fibre-matrix adhesion ensures
that high stresses can be transferred across the interface without disruption. In
composite materials, the interfacial adhesion between fibre and matrix plays an
important role in the final mechanical properties.
Chapter 1 22
1.2.4.1 Interfacial adhesion and types of bonding
Generally, the adhesion at the interface can be described by the following main
interactions: 1) physico-chemical interactions, related to wettability and
compatibility of the fibre and the matrix plus physical adhesion (e.g. Van der Waals
forces); 2) chemical bonding (covalent bonds) and 3) mechanical interlocking
created on rough fibre surfaces, and other interactions such as molecular
entanglement, interdiffusion etc. [44]. Good interfacial adhesion initially requires a
good wetting between the fibre and the matrix, to achieve an extensive and proper
interfacial contact. The wettability mainly depends on the surface energies of the
two materials. Essentially the fibre-matrix interactions are controlled by the
functional groups on the surface of the fibre and the matrix in the interfacial
contacting area.
A general description of the fibre-matrix interfacial interactions is presented in
following sections. In addition, more details of the state of the art are also discussed
in Chapter 3 and 4.
Physico-chemical interactions and wetting
In literature, the wetting of a solid by a liquid is described by the physical attraction
between two materials depending on their surface energies. Accordingly, bonding
due to wetting involves short-range interactions of electrons on an atomic scale
which develop only when the atoms of the constituents approach within a few
atomic diameters or are in contact with each other (in equilibrium interatomic
distance) [44]. When a good contact between two materials is formed, the other
bonding mechanisms will occur to create interfacial adhesion.
Wetting can be quantitatively expressed in terms of the thermodynamic work of
adhesion, , of a liquid to a solid using the Dupre equation
(1-1)
where is the solid surface energy, is the liquid surface tension, and is the
interfacial energy. Here, represents a physical bond resulting from highly
localized intermolecular dispersion forces.
Young was the first to describe the equilibrium contact angle when a sessile drop of
liquid is in contact with a solid surface [45]. The relation between surface energies
of the solid and the liquid through the contact angle is expressed as follows:
Introduction 23
(1-2)
Figure 1-15. Schematic of a sessile drop equilibrium contact angle
It can be seen that if the liquid forms contact angles respectively greater than and
less than 90o, then the situation is either ‘non-wetting’ or favourable for wetting. In
case of the presence of a very high surface energy solid, then the contact angle will
approach zero, which means a complete spreading of the liquid on the solid [46].
The work of (physical) adhesion can be expressed in relation to the equilibrium
contact angle by combining Eq. 1-1 with Young’s equation, resulting in:
(1-3)
In composite materials, the physical interactions between fibre and matrix can be
investigated when the surface energies of the fibre and the matrix are known. In
Table 1-6, surface energies of some common used fibres and matrices are shown,
including their polar and dispersive fractions (when the surface energy of a solid is
described comprising polar and dispersive components). Based on the surface
energies, wetting parameters can be calculated to study fibre-matrix wettability and
adhesion. This topic will be presented and discussed in the following chapters of this
dissertation.
Chapter 1 24
Table 1-6. Surface energy, , and its polar component, , and dispersive component, , of some
fibres and matrices.
Fibre/Matrix
(mJ/m2)
(mJ/m2)
(mJ/m
2)
Reference
Carbon fibre
(unmodified) 37.5 10 27.5 [47]
Glass fibre (0.3% silane)
41.5 13 28.5 [48]
Aramid fibre 34.6 14.0 20.6 [49]
Flax 30.5 17.6 12.9 [50]
Hemp 35.2 15.2 20.0 [51]
Jute 30.8 9.3 21.5 [52]
Cellulose 32.3 11 21.3 [50]
PP 26.2 0.4 25.8 [52]
PA 6,6 44.8 9.8 36.3 [53]
PET 43.7 7.1 36.6 [53]
Cured epoxy 42.6 16.5 26.1 [47]
Chemical bonding
While physical interactions mainly depend on van der Waal forces, chemical
bonding mechanisms are based on primary bonds at the interface. Chemical reaction
to form chemical bonds at the interface is the common method to enhance the
interfacial strength of polymer composites. In this mechanism of adhesion, a
chemical bond is formed between a chemical group on the fibre surface and another
compatible chemical group in the matrix, the formation results from usually
thermally activated chemical reactions. For example, a silane group in an aqueous
solution of a silane coupling agent reacts with a hydroxyl group on the glass fiber
surface, while a group like vinyl on the other end of the coupling agent will react
with the epoxide group in the matrix [44].
In natural fibre composites, the chemical bonding at the interface usually occurs
between the hydroxyl groups of cellulose and lignin on the fibre surface linked to
functional groups in the matrix (e.g. maleic anhydride groups in maleic anhydride
grafted polypropylene).
Mechanical interlocking
Mechanical interlocking is promoted by the fibre surface roughness, by anchoring of
the matrices polymer on the fibre surface. The strength of this type of interface is
unlikely to be very high in transverse tension unless there are a large number of re-
Introduction 25
entrant angles on the fibre surface, but the strength in longitudinal shear may be
significant depending on the degree of roughness. In addition, there are many
different types of internal stresses which arise from shrinkage of the matrix material
and the differential thermal expansion between fibre and matrix upon cooling from
the processing temperature. Among these stresses, the residual clamping stress
acting normal to the fibre direction provides an extra stress on top of the mechanical
anchoring discussed above [44].
1.2.4.2 Surface properties and surface treatment/modification of natural fibres
A better understanding of the natural fibre surface is necessary for the development
of natural fibre composites. Based on this, matrix systems can be selected or
developed to reach the full potential of the composite. The natural fibre surface is
complex with heterogeneous substances composed of cellulose, hemicellulose and
lignin. The surface is influenced by bulk morphology, extractive chemicals and
processing conditions. For example, the flax surface is reported to be covered by a
waxy layer, while lignin is the main component on the surface of bamboo fibre [50,
54]. In order to enhance the fibre-matrix interfacial strength of natural fibre
composites, it is necessary to use a physical or chemical treatment to change the
surface structure of the fibres as well as the fibre surface energy.
Physical treatments
Some physical methods, like stretching [55], thermotreatment [56] and electric
discharge [57-59] can be used to change the structure and surface properties of the
fibres. Among these treatments, the electric discharge methods such as corona and
cold plasma are interesting techniques for surface oxidation activation. A corona
treatment process changes the surface energy of cellulose fibres [57] and in case of
wood surface activation increases the amount of aldehyde groups [58]. The same
effects are reached by cold plasma treatment. Depending on type and nature of the
used gases, a variety of surface modifications were achieved. Surface crosslinks
could be introduced, surface energy could be increased or decreased, reactive free
radicals could be produced [8, 57, 58].
For example, Gassan et al. [60] studied the corona discharge and ultraviolet (UV)
treatments on jute fibre for improving the mechanical properties of jute/epoxy
composites. The result showed that the corona treatment increased the polarity of the
fibre surface (from 10 mJ/m2 to 26 mJ/m
2), whereas the dispersive contribution
remained unchanged. The UV treatment on the fibre also resulted in a higher
Chapter 1 26
polarities of the fibre surface, which led to a better wettability of fibres and a higher
composite strength (the composite strength increases 30% after a 10
min treatment at a distance of 150 mm away from the UV lamp).
Chemical treatments
There are several chemical ways to treat or modify the surface of natural fibres.
When the fibre and the matrix are incompatible, it is often possible to bring about
compatibility by introducing a third material that plays a bridge between the other
two materials. The treatments should take into account the morphology of the
interphase, acid-base reactions at the interface, surface energies and wetting
phenomena [8].
Natural fibres have relatively hydrophilic properties in terms of surface energy.
Some investigations are concerned with methods to decrease the hydrophilicity. The
modification of wood-cellulose fibres with stearic acid hydrophobizes these fibres
and improves their dispersion in polypropylene [61].
Impregnation (or sizing) of fibres by a polymer which is compatible with the matrix
provides a better interfacial adhesion. In this method, polymer solutions or
dispersions of low viscosity are used. For instance, cellulose fibres are impregnated
with a butyl benzyl phthalate plastified polyvinylchloride (PVC/BBP) dispersion to
create PVC/BBP-coated fibres, which results in a compatible interface between the
fibres and polystyrene (PS) [62].
One of the important chemical modification methods is chemical coupling. The fibre
surface is treated with a compound that forms a bridge of chemical bonds between
fibre and matrix. There are several ways of chemical coupling that can be found in
literature, such as graft copolymerization, treatment with isocyanates, triazine
coupling agents etc. Among these methods, graft copolymerization is an effective
method of chemical modification of natural fibres [63, 64]. This reaction is initiated
by free radicals on the cellulose molecule. For example, the treatment of cellulose
fibres with hot maleic anhydride polypropylene (MAPP) copolymer, provides
covalent bonds across the interface [65]. The mechanism of reaction is shown in
Figure 1-16.
Introduction 27
Figure 1-16. Chemical reaction between cellulose fibre and MAPP [8].
Arbelaiz et al. [66] studied the effect of MAPP treatment in flax fibres on short flax
fibre polypropylene composites. The result showed that there was an improvement
on the fibre-matrix interface, and the 5% MAPP-treated fibres increased the
composite tensile and flexural strength of approximately 35% compared to untreated
flax fibre composites.
Brahmakumar et al. [37] investigated the effect of waxy surface of coir fibres on the
fibre-matrix interfacial bonding in coir fibre low-density polyethylene composites.
Removal of the waxy layer resulted in a weaker interfacial bonding, and decreased
the composite tensile strength by 40% and modulus by 60%. And by grafting a layer
of a C15 long alkyl chain molecule onto the wax-free fibre, the composite interfacial
compatibility and bonding was improved, leading to an improvement of the
longitudinal tensile strength and modulus of the coir fibre composite of about 300%
and 700%, respectively, by incorporating 25% fibre volume faction of 20 mm
long fibre.
Alkali treatment
Alkali treatment is a widely used method to change the structure and surface
properties of fibres. Depending on the type of alkali (NaOH, KOH and LiOH) and
its concentration, several modifications occur in the fibre such as swelling,
Chapter 1 28
transformation of cellulose inside the fibre (from cellulose-I to cellulose-II), and
changes in fibre surface chemistry by removing some substances (e.g. fats and
waxes).
Prasad et al. [39] reported that by soaking the fibres in 5% NaOH solution at 28 oC
for 72 to 76h, the waxes and tyloses from the fibre surface were removed, and
tensile strength of the treated fibre increased 15% compared to untreated coir fibres.
In another study by Rout et al. [31] on alkali treated coir fibres by immersing the
dewaxed coir fibres (the fibre surface were cleaned by a mixture of ethanol and
benzene for 72h) in 5% and 10% NaOH at 30 oC for 1h, the SEM images of the
treated fibre surface also showed a removal of tysoles. And similar results observed
by Prasad et al., the tensile strength of 10% alkali treated fibres was higher than that
of untreated coir fibres.
The mechanical properties of alkali treated coir fibre polyester composites were
investigated by Rout et at. [67]. It was reported that 2% NaOH
treated fibre polyester composites improved the tensile strength and flexural strength
by 26 and 15%, respectively, compared to untreated fibre polyester composites.
With further increase in NaOH concentration from 2 to 5%, the 5% NaOH
treated fibre composite improved the flexural strength by 17% in comparison to
untreated fibre composite. For 10% NaOH treated fibre composites, both the tensile
strength and flexural strength decreased. The enhancement in mechanical properties
in alkali treated (2 and 5%) fibre composites was attributed to the improved wetting
of alkali treated coir with polyester [39]. The decrease in mechanical properties in
the case of 10% alkali treated fibre composites was due to cell wall thickening
which leads to poor adhesion with polyester resin.
For other natural fibres, in a study of Borysiak et al., flax fibres were treated with
different NaOH concentrations. The results showed that there was a degradation of
the crystal structure and partial transformation into cellulose-II at too high alkali
concentration [68]. The fatty and waxy layers on the surface of fibres can be
removed using strong alkali solution (higher than 5 % concentration), but it also
results in an increase of water uptake and fibre swelling. A low NaOH concentration
can partially remove fats and waxes together with a smoothening of the surface [69,
70].
Introduction 29
1.2.5 Concluding remarks
The above literature review provides an overview of the state of the art in natural
fibres, coir fibre composites and interfaces in natural fibre composites. Natural
fibres show various advantages for use in composites such as eco-friendliness and
having comparable mechanical properties to glass fibres. However, there are also
some limitations such as large diverse properties, moisture absorption, complex
morphology etc...
Regarding coir fibres, which are the fibres studied in this research, the extraction of
technical fibre has been well developed, allowing use of the fibre as a commodity
product. These extracted fibres can be used in composite materials. Mechanical
properties of coir fibres are reported as low strength and stiffness but high
elongation at failure in comparison with other commonly used natural fibres (flax,
hemp, jute). Nevertheless, a more accurate technique for determining fibre strain
during a tensile test should be developed to confirm the tensile properties of the
fibres. This should also allow a better investigation of their composites.
For coir fibre composites, most research has focused on the properties and
application of short randomly oriented coir composites. To fundamentally
understand the behaviour of coir in polymer composites, investigation of
unidirectional fibre composites is preferred in this thesis. The high elongation to
failure of coir fibre may enhance the toughness of its composites, which is also
necessary to be explored.
The surface of natural fibres has a complex morphology with chemical heterogeneity
and relatively high roughness. This will strongly influence the fibre-matrix interface
when using the natural fibres in composites. The quality of the interface is usually
evaluated by interface mechanical tests (e.g. fibre pull-out test). One can find some
studies on wettability of natural fibre composites, but even so there is a lack of a
deep understanding of fibre-matrix compatibility and adhesion based on fibre
surface chemistry.
1.3 Problem statement and goal of the thesis
From the literature review, it can be deducted that natural plant fibres have a wide
diversity in species, chemical, physical and mechanical properties. To be efficiently
used in composite materials, the fibre characteristics need to be assessed, which
comprises fibre structure, mechanical properties of the fibre and fibre surface
Chapter 1 30
properties. The mechanical properties decide directly the mechanical performance of
the composite, while the interfacial adhesion in the composite is influenced by the
fibre surface properties.
The properties of Vietnamese coir fibres have not been reported in literature, and
will be determined in this study. To determine the mechanical properties of the fibre,
it is relatively easy to perform a tensile test. However, it is not possible to use an
extensometer due to the small dimensions involved. It is necessary to develop a
suitable way for measuring fibre strain.
Unlike synthetic fibres, most natural fibres are relatively hydrophilic, have a rough
surface and are physico-chemically heterogeneous. These characteristics will
strongly affect the fibre-matrix interfacial interactions in the composite. The first
important concern is wetting between fibre and matrix to create a good fibre-matrix
contact and compatibility. Subsequently, a strong fibre-matrix adhesion ensures that
high stresses can be transferred across the interface without disruption. To obtain a
systematic understanding of this topic, the fibre surface properties regarding surface
physico-chemistry and surface energy components need to be determined. This is
then followed by knowledge of the potential interfacial interactions and the
measured adhesion. This combined knowledge is a prerequisite to optimise the
interface of natural fibre composites in terms of wetting and adhesion. The
understanding can be used to improve the interface properties by intelligently
choosing the fibre treatment or matrix modification.
Introduction 31
The goals of the research comprise the following aspects as shown in Figure 1-17,
and each one focuses on a specific topic that is important for natural fibre
composites:
1. Investigation of the fibre microstructure and its effect on the mechanical
properties of the fibre.
2. Characterisation of the fibre surface properties in terms of surface chemistry
and surface energy (components), and how these properties affect the wetting
and interaction between fibre and matrix.
3. Understanding the fibre-matrix interfacial compatibility and adhesion by
means of wetting analysis and interface mechanical tests. Based on this
knowledge, the interface can be modified by fibre treatment or matrix
modification.
4. Exploration of the possible value of coir fibre composites by investigating the
mechanical properties of UD coir fibre composites, including impact
behaviour.
Figure 1-17. Scheme of the research
Chapter 1 32
Thesis structure
Based on the research goals, this dissertation is structured in the following chapters.
In Chapter 2, the coir fibre extraction is introduced, and the microstructure and
mechanical properties are investigated. The focus of this research is the wetting
analysis and the study of interfacial adhesion, which are presented in respectively
Chapter 3 and Chapter 4. In Chapter 3, the characterisation of fibre surface
chemistry in combination with contact angle measurements and estimation of fibre
surface energy (components) will be described. In Chapter 4, wetting evaluation and
interface mechanical tests with different matrices are carried out, with further
discussion on the fibre-matrix interfacial compatibility and adhesion. The
mechanical performance of coir fibre composites is studied in Chapter 5, where
unidirectional composites of coir fibre in both thermoplastic and thermoset matrices
are examined to find out the possible value of coir fibre for composites. In the last
chapter, Chapter 6, general conclusions are drawn and an outlook is presented for
further study on the topic.
References
1. Wambua, P., J. Ivens, and I. Verpoest, Natural fibres: can they replace glass in
fibre reinforced plastics? Composites Science and Technology, 2003. 63(9): p.
1259-1264.
2. Mohanty, A.K., Sustainable bio-composites from renewable resources:
opportunities and challenges in the green materials world. Journal of polymers and
the environment, 2002. 10(1/2): p. 19.
3. Defoirdt, N., et al., Assessment of the tensile properties of coir, bamboo and jute
fibre. Composites Part a-Applied Science and Manufacturing, 2010. 41(5): p. 588-
595.
4. Charlet, K., et al., Characteristics of Hermes flax fibres as a function of their
location in the stem and properties of the derived unidirectional composites.
Composites Part A: Applied Science and Manufacturing, 2007. 38(8): p. 1912-1921.
5. Osorio, L., et al., Morphological aspects and mechanical properties of single
bamboo fibers and flexural characterization of bamboo/epoxy composites. Journal
of Reinforced Plastics and Composites, 2011. 30(5): p. 396-408.
6. Mussig, J. and C. Stevens, Industrial Applications of Natural Fibres: Structure,
Properties and Technical Applications: John Wiley & Sons.
7. Mohanty, A.K., Biofibres, biodegradable polymers and biocomposites: An
overview. Macromolecular materials and engineering, 2000. 276-277(1): p. 1.
8. Bledzki, A.K. and J. Gassan, Composites reinforced with cellulose based fibres.
Progress in Polymer Science, 1999. 24(2): p. 221-274.
9. Alexander, B., M. Supriya, and L. Thomas, Plant Fibers as Reinforcement for
Green Composites, in Natural Fibers, Biopolymers, and Biocomposites. 2005, CRC
Press.
10. Franck, R.R., Bast and other plant fibres. 2005, Woodhead Publishing.
Introduction 33
11. Lewin, M., Handbook of Fiber Chemistry. 2007, Marcel Dekker, Inc.
12. Mukherjee, P.S. and K.G. Satyanarayana, Structure and properties of some
vegetable fibres. Journal of materials science, 1984. 19(12): p. 3925-3934.
13. John, M.J., Recent developments in chemical modification and characterization of
natural fiber reinforced composites. Polymer composites, 2008. 29(2): p. 187.
14. Parameswaran, N. and W. Liese, On the fine structure of bamboo fibres. Wood
Science and Technology, 1976. 10(4): p. 231-246.
15. Kraessig, H., Cellulose chemistry and its applications, T. P. Nevell and S. H.
Zeronian, Eds., Halsted Press, John Wiley, New York, 1985, 552 pp. Journal of
Polymer Science Part C: Polymer Letters, 1987. 25(2): p. 87-88.
16. Fengel, D. and G. Wegener, Wood: Chemistry, Ultrastructure, Reactions. 1989:
Walter de Gruyter.
17. Page, D.H., F. El-Hosseiny, and K. Winkler, Behaviour of Single Wood Fibres
under Axial Tensile Strain. Nature, 1971. 229(5282): p. 252-253.
18. Navi, P., et al., Micromechanics of wood subjected to axial tension. Wood Science
and Technology, 1995. 29(6): p. 411-429.
19. Yogaratnam, N., Tapping the coconut coir fibre industry’s vast potential, in The
Nation Economist. 2009.
20. Thampan, P.K., Handbook on Coconut Palm. 1991, New Delhi: Oxford and IBH
Publishing Co.
21. Van Dam, J.E.G. and D. van, Process for production of high density/high
performance binderless boards from whole coconut husk: Part 1: Lignin as intrinsic
thermosetting binder resin. Industrial crops and products, 2004. 19(3): p. 207.
22. 2009; Available from: www.neda.gov.ph/knowledgeemporium/.
23. Silva, G.G., et al., Mechanical and thermal characterization of native brazilian coir
fiber. Journal of Applied Polymer Science, 2000. 76(7): p. 1197-1206.
24. Dam, J.E.G. and D. van, Process for production of high density/high performance
binderless boards from whole coconut husk: Part 2: Coconut husk morphology,
composition and properties. Industrial crops and products, 2006. 24(2): p. 96.
25. Conrad, K., Correlation between the distribution of lignin and pectin and
distribution of sorbed metal ions (lead and zinc) on coir (Cocos nucifera L.).
Bioresource technology, 2008. 99(17): p. 8476.
26. Satyanarayana, K.G., Structure property studies of fibres from various parts of the
coconut tree. Journal of materials science, 1982. 17(8): p. 2453.
27. Hill, C.A.S., A study of the potential of acetylation to improve the properties of
plant fibres. Industrial crops and products, 1998. 8(1): p. 53.
28. Khalil, H., The effect of acetylation on interfacial shear strength between plant
fibres and various matrices. European polymer journal, 2001. 37(5): p. 1037.
29. Han, J.S., Chemical composition of fibers. 1997.
30. Calado, V., The effect of a chemical treatment on the structure and morphology of
coir fibers. Journal of materials science letters, 2000. 19(23): p. 2151.
31. Rout, J., Scanning electron microscopy study of chemically modified coir fibers.
Journal of Applied Polymer Science, 2000. 79(7): p. 1169.
32. Bismarck, A., Surface characterization of natural fibers; surfaceproperties and the
water up-take behavior of modified sisal and coirfibers. Green chemistry, 2001.
3(2): p. 100.
33. Nawaratne, N.S., Use of coir fibre as a raw material for geotextiles. 2002, MSc
Thesis, University of Moratuwa: Sri Lanka.
Chapter 1 34
34. Tomczak, F., Studies on lignocellulosic fibers of Brazil. Part II: Morphology and
properties of Brazilian coconut fibers. Composites. Part A, Applied science and
manufacturing, 2007. 38(7): p. 1710.
35. Fratzl, P., I. Burgert, and H.S. Gupta, On the role of interface polymers for the
mechanics of natural polymeric composites. Phys. Chem. Chem. Phys., 2004. 6(24):
p. 5575-5579.
36. Burgert, I., Exploring the micromechanical design of plant cell walls. American
journal of botany, 2006. 93(10): p. 1391-1401.
37. Brahmakumar, M. and Brahmakumar, Coconut fibre reinforced polyethylene
composites: effect of natural waxy surface layer of the fibre on fibre/matrix
interfacial bonding and strength of composites. Composites Science and
Technology, 2005. 65(3-4): p. 563.
38. Haque, M.M., Physico-mechanical properties of chemically treated palm and coir
fiber reinforced polypropylene composites. Bioresource technology, 2009. 100(20):
p. 4903.
39. Prasad, S.V., C. Pavithran, and P.K. Rohatgi, Alkali treatment of coir fibres for coir-
polyester composites. Journal of Materials Science, 1983. 18(5): p. 1443-1454.
40. Hill, C.A.S. and H.P.S. Abdul Khalil, Effect of fiber treatments on mechanical
properties of coir or oil palm fiber reinforced polyester composites. Journal of
Applied Polymer Science, 2000. 78(9): p. 1685-1697.
41. Harish, S., et al., Mechanical property evaluation of natural fiber coir composite.
Materials Characterization, 2009. 60(1): p. 44-49.
42. Nam, T.H. and Nam, Mechanical and Thermal Properties of Short Coir Fibre
Reinforced Poly (Butylene Succinate) Biodegradable Composites. Journal of Solid
Mechanics and Materials Engineering. 5(6): p. 251.
43. Hemsri, S., et al., Wheat gluten composites reinforced with coconut fiber.
Composites Part A: Applied Science and Manufacturing. 43(7): p. 1160-1168.
44. Kim, J.K., Engineered Interfaces in Fiber Reinforced Composites. 1998.
45. Young, T., Phil. Trans. R. Soc. , 1805(95): p. 65-87.
46. Erbil, H.Y., Surface chemistry of solid and liquid interfaces. 2006, Wiley-
Blackwell.
47. Ramanathan, T., et al., Investigation of the influence of acidic and basic surface
groups on carbon fibres on the interfacial shear strength in an epoxy matrix by
means of single-fibre pull-out test. Composites Science and Technology, 2001.
61(4): p. 599-605.
48. Park, S.J. and J.S. Jin, Effect of silane coupling agent on mechanical interfacial
properties of glass fiber reinforced unsaturated polyester composites. Journal of
Polymer Science Part B: Polymer Physics, 2003. 41(1): p. 55-62.
49. Chen, P., et al., Influence of fiber wettability on the interfacial adhesion of
continuous fiber reinforced PPESK composite. Journal of Applied Polymer Science,
2006. 102(3): p. 2544-2551.
50. Aranberri-Askargorta, I., T. Lampke, and A. Bismarck, Wetting behavior of flax
fibers as reinforcement for polypropylene. Journal of Colloid and Interface Science,
2003. 263(2): p. 580-589.
51. Park, J.M., et al., Interfacial evaluation of modified Jute and Hemp
fibers/polypropylene (PP)-maleic anhydride polypropylene copolymers (PP-MAPP)
composites using micromechanical technique and nondestructive acoustic emission.
Composites Science and Technology, 2006. 66(15): p. 2686-2699.
Introduction 35
52. Park, J.M., et al., Interfacial evaluation and durability of modified Jute
fibers/polypropylene (PP) composites using micromechanical test and acoustic
emission. Composites Part B: Engineering, 2008. 39(6): p. 1042-1061.
53. Wu, S. Calculation of interfacial tension in polymer systems. 1971: Wiley Online
Library.
54. Fuentes, C.A., et al., Wetting behaviour and surface properties of technical bamboo
fibres. Colloids and Surfaces a-Physicochemical and Engineering Aspects, 2011.
380(1-3): p. 89-99.
55. Zeronian, S.H., H. Kawabata, and K.W. Alger, Factors affecting the tensile
properties of nonmercerized and mercerized cotton fibers. Textile research journal,
1990. 60(3): p. 179-183.
56. Semsarzadeh, M.A., Fiber matrix interactions in jute reinforced polyester resin.
Polymer composites, 1986. 7(1): p. 23-25.
57. Belgacem, M.N., P. Bataille, and S. Sapieha, Effect of corona modification on the
mechanical properties of polypropylene/cellulose composites. Journal of Applied
Polymer Science, 1994. 53(4): p. 379-385.
58. Sakata, I., et al., Activation of wood surface by corona treatment to improve
adhesive bonding. Journal of Applied Polymer Science, 1993. 49(7): p. 1251-1258.
59. Wang, Q., S. Kaliaguine, and A. Ait-Kadi, Catalytic grafting: A new technique for
polymer-fiber composites. III. Polyethylene-plasma-treated KevlarTM fibers
composites: Analysis of the fiber surface. Journal of Applied Polymer Science,
1993. 48(1): p. 121-136.
60. Gassan, J. and V.S. Gutowski, Effects of corona discharge and UV treatment on the
properties of jute-fibre epoxy composites. Composites Science and Technology,
2000. 60(15): p. 2857-2863.
61. Raj, R.G., et al., Compounding of cellulose fibers with polypropylene: Effect of fiber
treatment on dispersion in the polymer matirx. Journal of Applied Polymer Science,
1989. 38(11): p. 1987-1996.
62. Gatenholm, P., H. Bertilsson, and A. Mathiasson, The effect of chemical
composition of interphase on dispersion of cellulose fibers in polymers. I. PVC-
coated cellulose in polystyrene. Journal of Applied Polymer Science, 1993. 49(2): p.
197-208.
63. Sc, S.C.O.U.M., Structure/property relationships in textile fibres. Textile Progress,
1990. 20(4): p. 1-43.
64. Kroschwitz, J.I., Polymers: fibers and textiles: a compendium. 1990: Wiley.
65. Felix, J.M. and P. Gatenholm, The nature of adhesion in composites of modified
cellulose fibers and polypropylene. Journal of Applied Polymer Science, 1991.
42(3): p. 609-620.
66. Arbelaiz, A., et al., Mechanical properties of flax fibre/polypropylene composites.
Influence of fibre/matrix modification and glass fibre hybridization. Composites
Part A: Applied Science and Manufacturing, 2005. 36(12): p. 1637-1644.
67. Rout, J., et al., The influence of fibre treatment on the performance of coir-polyester
composites. Composites Science and Technology, 2001. 61(9): p. 1303-1310.
68. Borysiak, S. and J. Garbarczyk, Applying the WAXS method to estimate the
supermolecular structure of cellulose fibres after mercerisation. Fibres and Textiles
in Eastern Europe, 2003. 11(5): p. 104-106.
Chapter 1 36
69. Bismarck, A., et al., Surface characterization of natural fibers; surfaceproperties
and the water up-take behavior of modified sisal and coirfibers. Green Chem.,
2001. 3(2): p. 100-107.
70. Bismarck, A., et al., Surface characterization of flax, hemp and cellulose fibers;
Surface properties and the water uptake behavior. Polymer composites, 2002.
23(5): p. 872-894.
Microstructure and mechanical properties of coir fibres 37
Chapter 2
Microstructure and mechanical properties
of coir fibres
Chapter 2 38
2.1 Introduction
As a crop, coconuts have a wide diversity in size, colour, weight etc., which depend
on genetic variety and maturity of the nuts at harvest [1]. Hence, coir fibres which
are extracted from the coconut shells will depend on the properties of the coconuts,
and have a variety of fibre properties. van Dam et al. showed in their study on va-
rieties of Philippines’ coconuts that there was a small influence of cultivar on mor-
phology, structure and mechanical properties of the coir fibres, while the effect of
maturity on the chemical composition was larger than the variability between differ-
ent cultivars of the same age [2]. In this research, Vietnamese coir fibres will be
used as a reinforcement material. Therefore, it is necessary to investigate the mor-
phology, structure and mechanical properties of the fibres, which influence the
properties of their composites.
As mentioned in paragraph 1.2.1 of Chapter 1, a technical natural fibre usually
comprises numerous elementary fibres. Consequently, the mechanical properties of
the technical fibres are strongly influenced by the chemical composition and
organization of elementary fibres and by the micro fibril angles. In this chapter, the
microstructure of technical coir fibres will be examined using SEM and SEM-CT
and the mechanical properties of coir are determined in tensile tests. The relation
between fibre structure and fibre properties will be discussed based on the results of
the above analysis.
2.2 Materials and methods
2.2.1 Coir fibres
Coir fibres used in this research are long coir (fibre length in the range of 200-300
mm) which were provided by Can Tho University - Vietnam. The fibres were
mechanically extracted from husk shells of premature and mature coconuts (10-12
months on the plant) using a fibre extraction machine developed in the Mekong
delta, Vietnam. Concerning the fibre extraction process, the husk shells, after having
been separated from the nuts, were retted by exposing the shells to sunlight and by
spraying water several times a day during two weeks. Alternatively, a traditional
retting method was also applied in Vietnam, in which the shells were soaked in a
river for some months. For the reason of environmental pollution, this method is
seldom used nowadays. Following the retting process, the shells were compressed
and crushed by going between two rotating rollers, which help to partially separate
the coir fibres from the surrounding binding tissues. The compressed pieces of husk
Microstructure and mechanical properties of coir fibres 39
shell were then introduced to the extraction machine for extracting fibres, which
consists of two main steps. In the machine, the left half of the husk shell is gripped
by a large wheel; during rotating of the wheel, the right half of the husk shell is
moved towards a rotating picker drum, so that the shell is combed by the sharp nails
attached on the drum to remove pith and to separate the fibre from its bundle. In the
second step, the husk shell is transferred to a second wheel which helps to keep the
de-fibred right part of the husk and moves the left part to the second picker drum to
entirely defibre the whole husk shell. The extraction process is schematically shown
in Figure 2-1.
Chapter 2 40
Figure 2-1. Extraction process of coir fibre consists of retting, crushing and defibring.
Microstructure and mechanical properties of coir fibres 41
The extracted coir fibres comprise around 30% of the weight of the husk. The fibres
have a light colour and still contain some residual pith on their surface (Figure 2-2).
Figure 2-2. Coir fibre with residual pith attached to the surface.
2.2.2 Investigation of fibre microstructure using SEM and SEM-CT
SEM images of technical coir fibre surfaces and fibre cross-sections were taken
using a Philips XL 30 FEG scanning electron microscope. The images provide the
configuration of the fibre surface and the organisation of the elementary fibres inside
the technical fibre.
Figure 2-3. Schematic presentation of SEM-CT [3]. Coir fibre is positioned on rotation stage and
scanned by X-ray beam.
Chapter 2 42
To have more detailed information of the internal structure of the fibre, SEM-CT
was used to scan segments of the fibre. SEM-CT is a novel type of X-ray
tomography, in which the electron beam of the SEM hits a metal target to produce
X-ray radiation. The fibre is positioned on the precision rotation stage between the
target and the camera within the X-ray beam. The sample is rotated during scanning,
and a number of angular shadow projections of the sample internal structure is
acquired from the camera. A schematic of the SEM-CT set-up is shown in Figure 2-
3. A computer software package developed by SkyScan can be used to process the
scanned images, which allows to reconstruct a 3D structure of the sample in a non-
destructive way, displayed as virtual slices in any orientation or as a realistic three-
dimensional visual model which includes internal object details [3]. In this study, the
dimensions of technical fibre, elementary fibres and lacuna, as well as fibre porosity
will be characterised.
Figure 2-4. Internal structure analysis of coir fibre using SEM images and volumetric images from a
SEM-CT scan.
Microstructure and mechanical properties of coir fibres 43
For the analysis of the fibre porosity and the internal fibre structure, both SEM and
SEM-CT were used for characterisation on the same ten fibre samples, and the
results from the two methods were compared.
In the first method, SEM images of three different cross-sections were taken for
each fibre, as described in Figure 2-4. With the help of the software Leica QWin, the
area of fibre cross-section, lacuna and lumens were determined. These data were
also used for the calculation of the fibre porosity based on the ratio between the
porous area and the area of the fibre section.
The second method was X-ray tomography scanning of the fibre segments by SEM-
CT, with the SkyScan micro-CT attachment for the XL30 SEM. Titanium was used
as a target in combination with 30KV voltage from the electron beam to generate X-
rays. The entire fibre segment was scanned, and then a full volumetric image was
obtained after reconstructing the scanned images by using the SkyScan NRecon
software. With these sets of data, morphological measurement of the fibre in 2D and
3D was carried out with the help of the SkyScan CTanalyser software.
2.2.3 Measurement of fibre density
The density of the coir fibre was determined by the ratio of the weight of a fibre
sample and its volume. A gas pycnometer (Beckman 930) was used to measure the
volume of the sample. The principle of the measurement, as illustrated in Figure 2-5,
is as follows: two chambers are assumed to be equal in volume, and with the
coupling valve closed, any change in the position of one piston must be duplicated
by an identical stroke in the other in order to maintain the same pressure on each
side of the differential pressure indicator. When a sample with volume Vx is placed
in chamber B, with the coupling valve closed and when both pistons are advanced
the same amount from position 1 to position 2, the pressures will not remain the
same. The piston B is moved to position 3 where the pressures can be equalized.
Then the displacement d, from position 3 to position 2, corresponds to the volume of
the sample, Vx [4].
Chapter 2 44
Figure 2-5. Gas pycnometer and a schematic diagram of its operation
Coir fibre is a porous fibre which comprises a lumen in each elementary fibre and a
larger lacuna in the middle of the technical fibre. When the fibres are used in
composites, the whole fibre volume occupies a volume fraction of the composite,
but only the solid material of the fibre will carry load during loading of the
composite. Consequently, the density of the whole coir fibre and that of the fibre
solid fraction are important for the characterisation of fibre and composite
properties. In this experiment, coir fibres were cut to different fibre lengths of 4, 2, 1
and 0.5 mm and also to grinded fibre of approximately 0.05 mm length (considered
as solid particles). The samples were weighed to know the mass, and the sample
volume was determined using the pycnometer as described above. Accordingly, the
density of the samples could be calculated by using the measured mass and volume.
Microstructure and mechanical properties of coir fibres 45
2.2.4 Single fibre tensile tests
Single untreated technical fibres (which consist of a bundle of structurally bonded
elementary fibres) were tested in tension on a mini universal test machine. Because
of the small diameter of coir fibres (< 0.5 mm), it is practically difficult to measure
the fibre strain by using an extensometer, which is usually applicable for tensile
testing of larger samples. Therefore, two methods were used to determine the fibre
strain in this study.
In the first method, the test was performed on an Instron 5943 integrated with a
camera system for optical strain measurement (Figure 2-6). Speckles were created
on the fibre surface so that the camera system could map the fibre strain during
tensile loading. The recorded strain mapping data were analysed using Limess
software, and the calculated strain were then linked with the tensile load data to plot
the stress-strain behavior of the fibres. A 1 kN load cell was used for the test, and
the crosshead speed was set at 1 mm/min. It should be noted that the load
measurement accuracy of this new Instron machine is quite high ( +/- 0.5% of the
reading down to 1/500 of the load cell capacity) so this provides an accurate
measurement even at the low loading forces used. At least 15 fibres were tested in
this method.
The second method was based on correction of fibre slippage and machine
compliance. A variety of test span lengths (10, 15, 20, 25, 30 mm) were used for
performing the tensile test on a homemade mini tensile machine. For each span
length, a minimum of 15 fibres were tested. Based on the obtained data of load and
displacement at different span lengths, a theoretical correction (developed by
Defoirdt et al. [5], which is described in following paragraphs) for the fibre slippage
and machine compliance was used to determine the correct strain of the fibre
samples. The crosshead speed was set at 1 mm/min and a 200 N load cell was used
in this study.
Chapter 2 46
Figure 2-6. Set up of single fibre tensile testing with optical strain mapping
Concerning sample preparation, for both methods, the fibre sample was randomly
selected and glued into a paper frame, as shown in Figure 2-7. This keeps the fibre
as straight as possible and assures a good gripping. Before fixing the fibres in the
paper frame, the mass per length was measured for every fibre. The loaded cross-
sectional area of the fibre, which was used to convert applied force to stress, was
calculated using the mass, the length and the mean density of the coir fibres (in this
study the density of the solid coir material was used, which means that the
equivalent cross-section of the solid material was obtained).
Microstructure and mechanical properties of coir fibres 47
Figure 2-7. Paper frame for single fibre tensile test
As mentioned, in the second method, the measured displacement will consist of fibre
strain, slippage and machine compliance. A correction procedure for slippage and
machine compliance developed by Defoirdt et al. [5] was applied to correct strain at
failure and the E-modulus, described as follows:
The strain is expressed in Eq. 2-1:
where is the measured displacement of the clamps, is the elongation
of the fibre and is the displacement caused by slippage and machine
compliance.
The key element of this method is that the fibre modulus is determined at infinitely
long test length. At infinite fibre length, the displacement that is not caused by the
elongation of the fibre can be ignored. The procedure is that the measured (apparent)
modulus data from each test are plotted as function of 1/(test length). Then, by
(linear) extrapolation to 1/(test length) = 0, the fibre modulus at infinite fibre length
can be estimated ( )
To correct all the strain values for the effects of slippage and machine compliance,
the next step is to estimate a compliance factor αi for each test; αi captures the effects
of both slippage and machine compliance for each test and is assumed to be a
constant for each test.
It can be written for a certain stress σ at the first linear part of the stress-strain curve:
Chapter 2 48
It is assumed that the non-fibre strain is linear with the load put on the fibre:
where is the load put on the fibre (which corresponds to the chosen stress σ as
mentioned above) and is the factor that estimates the influence of slippage and the
test setup compliance. So, for every tested fibre can be calculated:
In the ideal case this factor should be the same for all tested fibres and for all
measured test lengths. In reality, there is quite some spread. In this work, an α value
was determined for each test length: all values are plotted versus the test length
and by a linear regression, an estimation of the value for each test length
can be determined. With this estimated value for the corrected strain can
be calculated:
With the corrected strain values, the corrected stress-strain curves can be drawn.
From these, as a consistency check, it can be verified if the E-moduli read from the
corrected graphs, correspond to Ee.
2.3 Results and discussion
2.3.1 Fibre surface and fibre internal microstructure
Fibre surface
Figure 2-8 shows a SEM picture of the surface of a typical technical coir fibre. In
the husk shell, coir fibres are positioned in parallel to each other, and surrounded by
porous organic tissue called pith (Figure 2-8a) which comprises approximately 70-
80% of the husk weight [2]. After extracting the fibre out of the husk, the pith may
Microstructure and mechanical properties of coir fibres 49
remain on the surface, and can be removed by further cleaning the fibre or by fibre
treatments.
In Figure 2-8b, a large number of arrays of globular protrusions on the fibre surface
are observed, which are located at regular intervals. These protrusions are called
‘tyloses’ and have been characterized to be rich in silicon content [6, 7]. They can
possibly be removed by mechanical or chemical treatment of the fibre surface,
leaving holes as shown in Figure 2-8c.
It can also be observed that the fibre surface consists of longitudinally oriented cells
with more or less parallel orientation (Figure 2-8c). Bismarck et al. suggested that
these cells are firmly held together by a binder of lignin and fatty substances which
are filling the intercellular space [8]. These characteristics of the coir surface will
influence the interfacial adhesion of coir fibre composites, which is studied and
presented in the following chapters.
Chapter 2 50
Figure 2-8. SEM images of technical coir fibre showing (a) coir fibres in husk shell surrounded by
organic tissue (b) surface of coir with arrays of protrusions (c) coir surface comprising of connected
cells (d) residual tissues remaining on the surface (e) protrusions
Microstructure and mechanical properties of coir fibres 51
Figure 2-9. SEM images of coir cross-section (a) cross-section of a typical coir fibre with presence
of lacuna and elementary fibres (b) a close up image of elementary fibres which shows lumens,
different cell walls and some micro fibrils.
Chapter 2 52
Internal structure and porosity
SEM images also reveal the internal structure of coir fibres as in Figure 2-9. A
typical cross section of a coir fibre (Figure 9a) indicates that a technical coir fibre
comprises of numerous elementary fibres with lumens inside. The hole, which is
approximately located in the centre of the fibre, is called lacuna. In the close up
image of elementary fibres (Figure 2-9b), it can be seen that each elementary fibre
consists of two cell wall layers which contain bundles of microfibrils, and the
middle lamella glues the elementary fibres together. The structure of coir fibres
follows the common cell wall structure of wood and plant fibres, but with much
larger MFA [9]. In the primary wall, the microfibrils seem to be oriented at around
45 degrees to the fibre direction, while the angle is larger (close to 90 degrees) in the
secondary wall. The secondary wall is somewhat thicker than the primary one. The
high angle of the microfibrils in coir fibre is also reported in literature [10].
Observably, coir fibre is a hollow fibre with quite big lumens and thin walls, and the
fibre cross-section is rather circular.
Figure 2-10. Image analysis to measure the porous area of the fibre cross-section using the software
Leica QWin.
Concerning the porosity of coir fibres, the results from both SEM image analysis
and SEM-CT scans are compared. In Figure 2-10, a SEM image of fibre cross
section is analysed using the software Leica QWin, and the fibre porous area is
detected and calculated. Assuming the fibre cross section, lumens and lacuna are
uniform along the fibre, the porosity of the fibre is then calculated based on the ratio
of the porous area and the total area of fibre cross section. Using the same principle,
the volume fraction of the lacuna in the fibre can also be determined.
The results of all analysed fibres are shown in Table 2-1. For each fibre, the data are
obtained based on the analysis of three different cross sections. To have an idea
Microstructure and mechanical properties of coir fibres 53
about the size of the tested fibres, the fibre diameter is approximately determined
based on fibre cross sectional area by simply assuming the fibre has a circular cross
section. With the SEM method, the results show that the fibre porosity is in the
range from 22 to 30%. Considering the fibre lacuna, its volume fraction in the fibre
is around 2 to 3%.
It should be noted that this analysis of fibre porosity has some limitations. In reality,
the lumens are not connected between the elementary fibres which are located in the
same line along the technical fibre. Therefore, the calculation of the volume of
lumens based on their cross sectional areas may give some overestimation. On the
other hand, the lumen of each elementary fibre is not a cylinder (the cross section of
lumen is not uniform, but its cross section is decreasing from the middle to the ends
of the elementary fibre). In this case, the volume of lumens can be underestimated
when a smaller cross section is analysed. Therefore, the hypothesis has been used
that by using 3 random cross-sections, a good approximation of the average lumen
size will be obtained (given also the relatively uniform cross-section of the fibres).
In the analysis of fibre structure using SEM-CT scans, a volumetric data set of
scanned fibre samples is reconstructed from scanned images. The structure can be
observed by orthogonal virtual slicing through the 3D structure (Figure 2-11, 2-12).
It can be seen that the elementary fibres are discontinuous and oriented uni-
directionally in the fibre direction. The lumens are also discontinuous and remain
inside every individual elementary fibre. The lacuna is a cylindrical channel in the
middle of the technical fibre. With the help of the software SkyScan CTAn, a 3D
model of the fibre can be built from the reconstructed data set, and internal structural
measurements such as fibre porosity and lacuna volume fraction are carried out. The
result of these analyses is also shown in Table 2-1.
Chapter 2 54
Table 2-1. Porosity of coir fibre determined from SEM image analysis and SEM-CT Scans (*fibre diameter is approximately calculated from the area of
fibre cross section by assuming it has circular shape)
Fibre Image analysis of fibre cross-section (SEM) SEM-CT Scan
Diameter *
(m) Cross-sectional
area (m
2)
Pore area (including
lacuna)
(m2)
Lacuna area (m
2)
Total fibre
porosity (%)
Lacuna
volume
fraction (%)
Total
fibre
porosity (%)
Lacuna
volume
fraction (%)
Elementary
fibre
diameter (m)
Elementary
fibre length (m)
1 282 ± 2.5 62572 ± 1115 15229 ± 1130 1745 ± 87 24.3 ± 1.4 2.8 ± 0.2 37.4 2.0 10.2-18.4 428-738
2 301 ± 6.2 71031 ± 2935 16356 ± 2786 1374 ± 395 23.1 ± 4.9 1.9 ± 0.5 27.0 3.6 7.9-15.8 364-617
3 219 ± 8.7 37657 ± 2948 10990 ± 878 774 ± 120 29.2 ± 1.3 2.1 ± 0.5 32.2 2.6 8.1-14.9 283-568
4 301 ± 13.9 71008 ± 6477 19217 ± 4652 2008 ± 976 26.8 ± 4.6 2.8 ± 1.3 37.1 2.5 6.4-15.0 455-960
5 192 ± 11.9 28895 ± 3581 8233 ± 712 960 ± 43 28.6 ± 2.1 3.4 ± 0.4 32.0 2.4 5.6-15.7 330-763
6 235 ± 6.6 43234 ± 2437 11297 ± 440 959 ± 89 26.2 ± 2.4 2.2 ± 0.2 29.6 1.8 6.9-17.8 457-869
7 276 ± 14.2 60024 ± 6213 18260 ± 1938 1656 ± 274 30.5 ± 3.0 2.8 ± 0.1 35.2 4.6 6.3-12.9 367-752
8 247 ± 3.2 47900 ± 1250 14264 ± 2157 1638 ± 586 29.8 ± 4.2 3.4 ± 1.3 33.4 3.3 8.4-14.9 336-781
9 158 ± 1.0 19684 ± 254 4142 ± 141 331 ± 111 21.1 ± 1.0 1.7 ± 0.6 39.8 4.0 7.6-18.6 366-551
10 259 ± 10.6 52881 ± 4298 16184 ± 1610 1497 ± 721 30.7 ± 3.9 2.9 ± 1.4 46.3 2.1 8.0-19.5 321-668
Microstructure and mechanical properties of coir fibres 55
Figure 2-11. Three orthogonal virtual slices through a 3D reconstructed internal structure of coir
fibre obtained from SEM-CT scanned images, (a) coronal image (b) 3D navigation (c) transaxial
image (d) sagittal image.
The porosity of the coir fibres (from 10 tested fibres) ranges from 27 to 40%, except
the value of fibre number 10, which is approximately 46%. In comparison with the
results obtained from the analysis of SEM images, the fibre porosity from this
analysis is higher. Considering the method, it works based on a densitometry
principle; the quality of the scanned images depends on the density difference
between fibre solid material and air. Because this difference is not large in case of
coir fibre, some errors are included. Besides, coir fibres consist of various thin
organic tissues, which may not be detected on the scanned images. Hence, the fibre
porosity analysed with this method is likely to be overestimated since the fibre solid
material is not fully determined.
Chapter 2 56
Figure 2-12. 3D model of typical coir fibre based on scanned data obtained from SEM-CT scans.
In summary, the two discussed methods offer good tools to study the porosity and
generally the internal structure of coir fibres. Based on the above discussion, it can
be hypothesized that the fibre porosity will be better estimated by image analysis on
SEM pictures of fibre cross sections, which means it will be in the range from 22 to
30%.
Length and diameter of elementary fibres
Using orthogonally sliced SEM-CT images of coir fibre (in longitudinal and
transverse direction), the length and diameter of elementary fibres can be estimated
as shown in Figure 2-13. In these images, the vertical section of elementary fibres is
seen to have a quasi elliptical shape. Hence, the length of the elementary fibre is
approximately equal to the major diameter of this ellipse shape. The diameter of the
elementary fibres is determined from the fibre cross section. The results are
presented in Table 2-1.
From the measurement of ten fibres, the length of elementary fibres is in the range
of 350 to 950 m, which is quite close to the reported values for Philippines’ coir,
ranging from 700 to 1100 m [11]. The result measured in this study may be
underestimated since the analysed vertical sections of elementary fibre may not be
Microstructure and mechanical properties of coir fibres 57
from the centre of each fibre. Therefore, it is recommended to rather refer to the
higher value in the range as the representative length.
For the average diameter of the elementary fibres, their values range from 6 to 19
m which depends on their location in the technical fibre. It is observed that the
elementary fibres located near the lacuna have a bigger diameter than those close to
the edge of the technical fibre. Again here, because the measured values are obtained
from a random cross-section, the values are estimations of the average cross-section;
the maximum cross-section of the elliptically shaped fibre will be closer to the
maximum values in the observed range.
Figure 2-13. Measuring the length and diameter of elementary fibres from SEM-CT sliced images.
The length of lumens (in black) give an estimation of elementary fibre length.
2.3.2 Density of coir fibres
Figure 2-14 presents the density of coir fibre as function of the length of the tested
fibre sample, from pycnometer measurements. The results show that the density of
solid fibre material measured on grinded powder (estimated length of 0.05 mm) is
Chapter 2 58
approximately 1.3 g/cm3 (in the range of its constituents’ density: the cellulose and
lignin density are 1.53 g/cm3 and 1.06-1.33 g/cm
3 respectively). The density
decreases to 0.9 g/cm3 with increasing fibre length, and this value should be
considered as the density of structural coir fibre. The results can be explained
considering the internal porous structure of coir fibre. At very short length powder,
there is no porosity in the fibre sample, and the fibre solid volume is measured by
the pycnometer. The enclosed porosity of the fibre sample increases with increasing
fibre length, and the measured volume is the fibre solid volume plus the internal
enclosed air volume of the lumen.
Based on this result, the average (lumen) porosity of coir fibres can be derived by
the ratio of the fibre air volume and the total fibre volume, which gives about 31%.
This result is quite consistent with that obtained from image analysis from SEM and
SEM-CT.
Figure 2-14. Density of coir fibre as function of the length of the fibre samples; results from
pycnometer measurements
2.3.3 Tensile mechanical properties of coir fibres
Figure 2-15 presents typical stress-strain curves from tensile tests on single technical
coir fibres, where the fibre strain is obtained from the image analysis of speckles
created on the fibre surface. As described in paragraph 2.2.4, the coir fibres were
tested by the tensile machine with an optical strain mapping system. The movement
of speckles on the fibre surface was captured by the camera, and analysed using the
software Vic-2D. The strain of each analysed fibre is then linked to the
corresponding tensile load recorded by the tensile testing machine to obtain the
stress-strain curves (the stress is determined by the ratio of the tensile load and the
0,5
0,7
0,9
1,1
1,3
1,5
0,0 0,5 1,0 1,5 2,0 2,5 3,0 3,5 4,0 4,5
Den
sity
(g/
cm3)
Fibre length (mm)
Microstructure and mechanical properties of coir fibres 59
fibre solid cross-section which is calculated by using the fibre length and the solid
fibre density of 1.3 g/cm3). The curves show that coir fibre behaves in a linear
elastic manner at low stress, and then shows plastic behaviour until fibre failure at
very high strain to failure (Figure 2-15a). Figure 2-15b shows tensile stress-strain
curve of the coir fibre under cyclic tensile load. It can be seen that there are
remaining strain when the applied load is stopped at certain strain (around 5% and
13%), which indicates the plastic behaviour of the coir fibre. Moreover, when
comparing the E-modulus in the reloading cycles (E2 and E3) to the initial value E1,
the result shows E3> E2> E1. This suggests that the microfibrils slide and reorient to
the loading axis under the cyclic loading, which results in the increasing of the fibre
stiffness.
Figure 2-15. (a) Typical tensile stress-strain curves of single coir fibre. (b) Stress-strain curve of
coir fibre, in which unloading and reloading is applied at a certain strain.
Chapter 2 60
The tensile modulus, strength and strain to failure of the coir fibres were calculated
and are shown in Table 2-2. It can be seen that the coir fibres have high elongation
to failure, but are not so strong and stiff. As known from the analysis of the fibre
internal structure, coir fibres have a high MFA, which explains the low stiffness in
fibre direction and the high elongation thanks to reorientation of the microfibrils
under tensile loading. The properties of the fibres are comparable with previously
reported results [5, 11].
Table 2-2. Tensile E-Modulus, strength and strain to failure of coir fibre by two different testing
methods (optical strain mapping and corrected fibre strain from a range of test lengths); the fibre
density used to determine fibre cross-sectional area was1.3 g/cm3.
Optical strain mapping (5 mm test length)
E-Modulus (GPa) 4.6 ± 1.1
Strength
(MPa) 234.2 ± 57.4
Strain at failure (%) 18.0 – 36.7
Range of test lengths
Extrapolated E-Modulus
(GPa) 4.54
E-Modulus from corrected stress-strain curves
(GPa) 4.9 ± 0.9
Strength (MPa) 204.6 ± 39.8
Measured strain at failure
(smallest test length ÷ longest test length) (%)
44.7 ± 11.4
÷ 34.4 ± 4.9
Corrected strain at failure
(smallest test length ÷ longest test length) (%)
41.0 ± 10.7
÷ 33.5 ± 5.0
With the procedure using a correction for slippage and machine compliance, by
using different test span lengths, the extrapolated E-Modulus at infinite test length
was first obtained by plotting a trendline of the measured E-modulus at different test
lengths as shown in Figure 2-17a. An extrapolated modulus value of 4.54 GPa was
obtained (see also Table 2-2). Next, the factor αi was calculated for each tested fibre.
Figure 2-16 shows average α values for the different test lengths, where the α value
depends on the test length. In the study of Defroidt et al. [5], it is suggested that α is
rather caused by slippage than by test setup compliance. At shorter test lengths, the α
values are higher which means the measured extra strain is determined more by
slippage in the clamps than by test setup compliance, which should be assumed as
constant. Because of the observed variation of the α values, a linear regression line
was constructed to obtain the most probable α value for each test length, to correct
the fibre strain values (see Table 2-3).
Microstructure and mechanical properties of coir fibres 61
Figure 2-16. Alpha values in function of the test length
Table 2-3. Alpha value for every test length determined from the trendline.
Test length
(mm)
Alpha value
on the trendline
10 0.046
15 0.043
20 0.040
25 0.037
30 0.034
The corrected strain values were used to construct corrected stress-strain curves,
from which once more E-moduli were read. Measured and corrected E-moduli are
presented in function of the used test lengths in Figure 2-17a. The measured
modulus is clearly depending on the test length which means that slippage and test
setup compliance influence the moduli. After correction, the corrected E-moduli are
as they should be independent on the test length, and the correction is larger at
shorter test length. The mean value of the re-constructed E-Modulus is around 4.9
GPa which is quite consistent with the value obtained from the optical strain
mapping method. The re-constructed values are a bit higher than the extrapolated
modulus of 4.54 GPa, which was the baseline value for the correction procedure. In
principle, the corrected values should be exactly the same as the extrapolated value,
but the correspondence is believed to be acceptable.
y = -0,0006x + 0,0519 R² = 0,4458
0
0,02
0,04
0,06
0,08
0,1
0 10 20 30 40
Alp
ha
- α
test length (mm)
Chapter 2 62
Figure 2-17. Tensile properties of coir fibre: (a) uncorrected and corrected E-modulus as function of
1/test length (b) uncorrected and corrected strain (c) strength.
Microstructure and mechanical properties of coir fibres 63
In the same manner as the E-modulus, the measured strain to failure is dependent on
the test length. It can be seen that the corrected strain to failure stays dependent on
the test length (Figure 2-17b), which is logically linked to the failure probability
theory. The longer the fibres, the higher the chance of defects and the earlier they
break. The strain to failure at 10 mm test length is approximately 40% which is
comparable with the value obtained from the optical strain mapping and literature
values at similar test length [5, 11].
The fibre strength is expected to decrease like the strain to failure when the test
length increases following the probability of breakage theory, but it was not
observed. The strength seems to stay in a range from 170 to 240 MPa, which is
situated in the middle of literature values [11, 12]. Apparently, the defect sensitivity
of coir fibres is relatively low. This is logical, as failure is proceeded by massive
plastic deformation.
Concerning the test methods, optical strain mapping provides a fast and precise way
to determine the fibre elongation during tensile loading, in comparison with the
procedure using a correction for slippage and machine compliance, by using
different test span lengths. On the other hand, the data at different test lengths in the
latter method give more information about the defect sensitivity of the fibres.
2.4 Conclusions
The characterisation of the coir fibre surface using SEM provides useful information
about the fibre surface, which will help to improve or modify the fibre-matrix
interfacial adhesion when the fibres are used in composites. It is obvious that there
are arrays of rich silicon protrusions, and pith tissues still partially remain on the
fibre surface. The SEM images of fibre cross sections show that technical coir fibres
comprise plenty of elementary fibres (in the range of 200-300 elementary fibres) and
a lacuna at the centre. The elementary fibre is built up by two main cell walls which
consist of bundles of microfibrils aligned in a high angle to the fibre axis (high
microfibril angle). Coir fibre appears to have high porosity at 22 to 30%.
SEM-CT is a good tool for analysing the internal structure of coir fibre. The fibre
porosity and the dimensions of lumen, lacuna and elementary fibres were
determined by using 3D information and three orthogonal virtual slices of the
scanned fibre. The results confirm that coir fibre has high porosity.
Chapter 2 64
Single fibre tensile testing with optical strain mapping offers a fast and reliable tool
to measure tensile properties of coir fibres. The test using different test lengths gives
more information about the defect sensitivity of the fibres and shows that the defect
sensitivity of coir fibres is relatively low. The results of both methods indicate that
coir fibres are not very strong and stiff (strength and stiffness are approximately 234
MPa and 4.6 GPa respectively), but have high strain to failure (20-40%), which may
increase toughness of some brittle matrices when they are used in composites.
References
1. Ohler, J.G., Modern coconut management: palm cultivation and products. 1999:
Intermediate Technology Publications.
2. Van Dam, J.E.G. and D. van, Process for production of high density/high
performance binderless boards from whole coconut husk: Part 1: Lignin as intrinsic
thermosetting binder resin. Industrial crops and products, 2004. 19(3): p. 207.
3. Skyscan Micro-CT in SEM. Available from: http://www.skyscan.be.
4. Beckman® INSTRUMENTS, INC. Analytical Chemistry, 1962. 34(4): p. 64A.
5. Defoirdt, N., et al., Assessment of the tensile properties of coir, bamboo and jute
fibre. Composites Part a-Applied Science and Manufacturing, 2010. 41(5): p. 588-
595.
6. Rout, J., Scanning electron microscopy study of chemically modified coir fibers.
Journal of Applied Polymer Science, 2000. 79(7): p. 1169.
7. Calado, V., The effect of a chemical treatment on the structure and morphology of
coir fibers. Journal of materials science letters, 2000. 19(23): p. 2151.
8. Bismarck, A., et al., Surface characterization of natural fibers; surfaceproperties
and the water up-take behavior of modified sisal and coirfibers. Green Chem.,
2001. 3(2): p. 100-107.
9. Persson, K., Micromechanical modelling of wood and fibre properties. 2000:
Division of Structural Mechanics, Lund Institute of Technology.
10. Martinschitz, K., et al., Changes in microfibril angle in cyclically deformed dry coir
fibers studied by in-situ synchrotron X-ray diffraction. Journal of materials science,
2008. 43(1): p. 350-356.
11. Dam, J.E.G. and D. van, Process for production of high density/high performance
binderless boards from whole coconut husk: Part 2: Coconut husk morphology,
composition and properties. Industrial crops and products, 2006. 24(2): p. 96.
12. Munder, F., Mechanical and thermal properties of bast fibers compared with
tropical fibers. Molecular Crystals and Liquid Crystals, 2006. 448(1): p. 197.
Wetting analysis and surface characterisation of coir fibres 65
Chapter 3
Wetting analysis and surface characterisation
of coir fibres
Chapter 3 66
3.1 Introduction
In composite materials, the interfacial properties of fibre and matrix play an
important role in the final mechanical performance. The interactions at the interface
generally consist of physical adhesion, chemical bonding and mechanical
interlocking.
Good interfacial adhesion initially requires a good wetting between the fibre and the
matrix, to achieve an extensive and proper interfacial contact; and the wettability
mainly depends on the surface energy of the two materials. High surface energy of
both fibre and matrix contributes to a high work of adhesion, while the matching of
surface energy components results in a low interfacial energy which indicates a good
fibre-matrix interfacial compatibility. These interactions are essentially controlled by
the functional groups on the surface of the fibre and the matrix in the interfacial
contacting area.
The surface energy of the fibre and the matrix can be estimated using its contact
angles in different probe liquids. Moreover, studying wetting between the fibre and
the test liquids provides an understanding of fibre hydrophilicity and fibre surface
polarity.
Wetting is the consequence of the change in nature of interfaces driven by free
energy minimisation [1]. The quantitative measure of solid–liquid interactions is the
contact angle. The equilibrium contact angle, defined by Young [2] at the
intersection of the three phases, gas–liquid–solid, was initially developed for a drop
of liquid on a smooth solid surface (Figure 3-1a).
In case of a single fibre, contact angles are reported to be measured either directly or
indirectly, typically by using the modified Wilhelmy technique [3-5]. The direct
technique is carried out by introducing a drop of liquid on the fibre (Figure 3-1b).
The contact angle is usually calculated following the theoretical model proposed by
Carroll [6], Yamaki and Katayama [7]. Some difficulties are observed
experimentally, such as improper drop making and high curvature variation at the
interface. For the indirect way, the Wilhelmy technique is usually recommended.
The principle is to use a microbalance to record the wetting force, which is the
capillary force exerted by the liquid on the fibre. The contact angle is then deduced
from the recorded force in relation to the liquid surface tension and the fibre
perimeter.
Wetting analysis and surface characterisation of coir fibres 67
Figure 3-1. Contact angle between a liquid and a solid (a) Young equilibrium contact angle (b)
contact angle of a liquid and a fibre
Besides the consideration of measurement techniques, wetting measurements on
plant fibres present additional complexities, which are typically not found for
synthetic fibres and which may affect the measurement: liquid absorption/diffusion
into the surface layers/cell walls; diffusion of low-molecular-weight compounds
from the surface layers into the liquid; physico-chemical heterogeneity of the
constituents (e.g., cellulose, hemicelluloses, and lignin) of the surface layers;
viscoelastic response of the surface layers to the liquid [8].
In some specific studies, the measurement of the contact angle is performed in
dynamic conditions. There, the wetting of a liquid on a solid surface is commonly
known as the moving of the three-phase line or wetting line across the surface of the
solid. Associated with this moving wetting line, dynamic contact angles and wetting
velocity are the main parameters used to quantify the dynamics of wetting. It is
experimentally reported that the dynamic contact angle is usually found to depend
on both the speed and the direction of the wetting line displacement. The dynamic
angles differ from the corresponding static value and may refer to either an
advancing or receding contact line. The measured angle will depend on the history
of the system and varies according to whether the contact line is being advanced or
receded. This phenomenon is known as contact angle hysteresis (Figure 3-2), and
occurs in most real systems where solid surfaces are often rough and chemically
heterogeneous. In this complex situation, the static contact angle is even unlikely to
be single-valued, but may show “advancing” and “receding” limits [9, 10].
Chapter 3 68
Figure 3-2. Schematic presentation of the velocity-dependence of the contact angle (after Dussan
[11]).
The fibre surface chemistry provides important information on the chemical
functionality that determines the interactions with the chemical functionality of the
matrix. The level of hydrophilicity and the surface energy of fibres are also the result
of chemical groups on the surface. For surface chemical characterisation, X-ray
photoelectron spectroscopy (XPS) has shown to be a useful tool to investigate the
surface chemistry of several kinds of cellulose fibres, which assists to determine a
quantitative elemental composition and certain functional groups present on the
surface [12-15].
In this chapter, the wetting measurement procedure to determine stable and
reproducible advancing static contact angles of coir fibres is reported. The
experiments are carried out by considering the effects on the contact angle results of
irregular wetted length along the fibre perimeter and liquid absorption, which
commonly appear with natural fibres. Coir fibre static contact angles are used to
estimate the fibre surface energy. XPS is used to analyse the fibre surface chemistry,
linked with the wetting results, to obtain a deeper understanding of the coir fibre
surface.
3.2 Materials and Methods
Contact angle measurements of the coir fibres were carried out using the Wilhelmy
technique, which allows to determine dynamic contact angles of various test liquids
on the fibres. However, the significance of the contact angle as a measure of
wettability is based on equilibrium thermodynamic arguments, for which static
systems are most frequently studied. Hence, it would be more reliable if static
equilibrium contact angles are introduced in the wetting parameters calculation.
Wetting analysis and surface characterisation of coir fibres 69
Following sections will explain the method to obtain the equilibrium static contact
angle from the data of dynamic angles.
3.2.1 Materials
Fibres
Coir fibre was supplied by the Can Tho University of Vietnam, where it was
extracted from the husk shell of coconut from the coconut palm (Cocos nucifera L.).
The extraction process was a purely mechanical method as described in chapter 2.
The extracted fibres were soaked in hot distilled water at 70°C for 2 h, and then
smoothly washed with alcohol to remove greases which may attach on the fibre
surface during the fibre extraction process, rinsed with deionised water and dried
under vacuum at 90 °C. The thus cleaned fibres are considered as untreated fibres. In
order to prevent the effect of a rough fibre surface due to unevenly distributed
organic material (Figure 3-3a), coir with a clean surface (Figure 3-3b) was selected
with the help of a light microscope.
Figure 3-3. SEM images of coir cross-section and coir surface: (a) organic residues on coir surface,
(b) clean surface, (c) fibre cross-section, (d) longitudinally orientated cells
Polyethylene terephthalate (PET) monofilaments (diameter 800 μm) from
Goodfellow were used as a reference synthetic fibre to evaluate the measurement
methods and to compare with coir fibres. PET fibres were washed with a detergent
(RBS-35 from Chemical Products) for 1 h with a magnetic stirrer to remove organic
residues on the surface, and rinsed in deionised water at 90 °C for 1 h. The cleaned
fibres were then dried under vacuum at 90 °C for 2 h and conserved under silicagel.
Test liquids
Test liquids used in this study are described in Table 3-1, providing data on surface
tension, density, and viscosity (from products datasheet and [16-18]).
(a)
(b)
(c)
(d)
Chapter 3 70
Table 3-1. Summary of surface tension comprising of polar and dispersive components following the Owens-Wendt approach, and Lifshitz – van de
Waals and acid-base components following the van Oss-Good approach; density, viscosity and purity of test liquids used in this study.
Test liquid
Surface tension (mJ/m2)
(g/cm3)
(mPas) Purity Supplier
disperse-polar
(Owens-Wendt) acid-base (van Oss-Good)
n-Hexane 18.4 0 18.4 18.4 18.4 0 0 0.66 0.32 99.6% Acros
Diiodomethane 50.8 0 50.8 50.8 50.8 0 0 3.32 2.76 >98% Merck
Ethylene glycol 48.0 19.0 29.0 48.0 33.9 0.97 51.6 1.11 16.1 ≥99% Merck
Formamide 58.0 19.0 39.0 58.0 35.5 11.3 11.3 1.13 3.3 99% Sigma
Benzyl alcohol 39.0 8.7 33.3 - - - - 1.04 8 99.5% Acros
Ultrapure water 72.8 51.0 21.8 72.8 26.25 48.5 11.16 1 1 18.2 Ω·cm
resistivity Millipore
Wetting analysis and surface characterisation of coir fibres 71
3.2.2 Dynamic contact angle measurement
The dynamic contact angle measurement of the coir fibres was carried out according
to the Wilhelmy method, using a Krüss K100 tensiometer with 1 μg weight
resolution. The fibre sample was attached to a beam of an electrobalance and was
held vertically at a fixed position during the measurement, while a beaker containing
the test liquid was raised and lowered via a driving platform. When the test liquid
touched the fibre, a force was detected on the balance. In contact with the liquid, the
fibre was scanned in both advancing and receding directions, at a constant velocity,
from which a force-position plot was constructed with help of the software LabDesk
(Figure 3-4). Theoretically, the force on the balance is the sum of the wetting force,
the weight of the probe and the buoyancy force, as given by:
(3-1)
where is the total measured force on the electrobalance, is the fibre wetted
perimeter, is the liquid surface tension, is the contact angle at the three-phase
contact line, is the mass of the fibre, is the acceleration of gravity, is liquid
density, is the fibre cross sectional area and is the immersion depth. The weight
of the fibre probe can be measured beforehand and set to zero on the balance, while
the effect of buoyancy can be removed by extrapolating the force back to zero
immersion depth. The remaining force is the wetting force only:
(3-2)
The contact angle was then calculated from the received force data at any depth. For
both advancing and receding processes, the obtained angles were called advancing
contact angle and receding contact angle respectively. Due to the effect of coir fibre
surface roughness test liquid may remains on the fibre surface after the advancing
process, which leads to an unstable receding angle (close to 0 degree). Therefore,
only the advancing contact angles were studied.
Chapter 3 72
Figure 3-4. Advancing contact angle measurement following Wilhelmy method (a) Tensiometer (b)
schematic of wetting measurement (c) recorded advancing and receding forces
The measurements were performed at a temperature of 20 °C and humidity of
approximately 60%. Different measurement speeds ranging from 2 μm/s to
8000 μm/s were used to study both dynamic wetting and static contact angle
approximation, the latter is described in the following parts of this chapter. The
immersion depth used for experiments was from 2 mm to 5 mm depending on the
applied measurement speeds. As mentioned, water, diiodomethane, ethylene glycol
and formamide were used as probe liquids in both contact angle measurements and
the surface energy calculation.
For a given measurement of the fibre with a test liquid, the calculation of contact
angles was mainly affected by two parameters, namely the fibre wetted perimeter
and the measured force. In contrast to synthetic fibres, natural fibres do not have a
uniform geometry along the fibre. Therefore, the fibre perimeter may change and
this may lead to an incorrect result for the contact angle. Besides, the liquid
absorption into the fibres may also create an extra weight during the measurement.
These above considerations have been taken into account and the effects have been
minimized in the later experiments.
Wetting analysis and surface characterisation of coir fibres 73
3.2.2.1 Determination of fibre wetted perimeter
The method to determine the fibre wetted perimeter was based on a tensiometric
measurement. A low surface tension liquid, n-Hexane, was used for wetting
measurements with the fibre. At relatively low speed, it is assumed that the fibre is
totally wetted by the liquid. Then the contact angle is zero. Eq. 3-2 becomes:
(3-3)
Following Eq. 3-3, the fibre wetted perimeter, , can be determined from the wetting
force and liquid surface tension. Thus, in principle, the fibre wetted perimeter can be
measured at any position along the fibre. Another method was also carried out to
make a comparison. The microscope images of cross-sections of fibre samples were
then examined by image analysis. For each fibre, images of various cross-sections
were taken by a scanning electron microscope (SEM 30 XL FEG). With help of the
software Leica Qwin, the contour of each cross-section was determined.
3.2.2.2 Estimation of the effect of liquid absorption
Using the electrobalance of the tensiometer, the fibre sample was weighed before
and after the wetting measurement. The absorbed mass into the fibre was determined
to analyse the effect on the final contact angle results. The experiments were carried
out with the five test liquids.
3.2.3 Static equilibrium contact angle approximation
Since the significance of the contact angle as a measure of wettability is based on
equilibrium thermodynamic arguments, static systems are most frequently studied
[9]. It would be more reliable if static equilibrium contact angles are introduced in
the surface energy calculation. In this study, two methods were used to determine
the static angle. The first method is based on the relationship between the advancing
dynamic contact angles and the wetting velocity following the Molecular-kinetic
theory (MKT). As a second method to determine static contact angles, a variant of
the Wilhelmy method was used for wetting measurements with the different test
liquids.
Chapter 3 74
3.2.3.1 Molecular-Kinetic Theory (MKT) of dynamic wetting and static contact
angle approximation
Several studies of dynamic wetting, centred on using the contact angle as a
geometrical boundary condition for the moving fluid/fluid1 interface at the solid
surface, are described elsewhere [9]. For the aim of this study, the MKT of dynamic
wetting is applied to study the relation between the dynamic contact angle and the
static contact angle.
According to the MKT [19], the macroscopic behaviour of the wetting line depends
on the overall statistics of the individual molecular displacements that occur within
the three phase zone. The wetting line is modelled by the displacements of length
of the molecules from one adsorption site to another. The molecules can move
forward with a frequency , or backward with a frequency (Figure 3-5). The
net frequency is . Then the velocity of the wetting line is given by
. At equilibrium, , is zero and . For the wetting
line to move, work must be done to overcome the energy barriers to molecular
displacement in the preferred direction. Blake and Haynes assumed this work is
provided by out-of-balance surface tension force , where
is the surface tension of the liquid in contact with vapour, is the static contact
angle and is the dynamic contact angle. Combining these ideas and applying the
Eyring’s theory of absolute reaction rates for transport in liquids, the following
relationship between and is obtained as:
[
]
where is the number of adsorption sites per unit area, is Boltzmann’s constant,
is the absolute temperature.
The equilibrium frequency is related to the activation free energy of wetting
as
(
) (
)
1 In case of a surrounding vapour, usually air, this is also modelled as a fluid.
Wetting analysis and surface characterisation of coir fibres 75
where is Planck’s constant and is Avogadro’s number.
If the adsorption sites are evenly distributed, then √ . Eq. 3-4 then contains
just two free molecular parameters, namely and . These parameters cannot
usually be determined a priori, and so are obtained by fitting experimental data.
Figure 3-5. Dynamic wetting according to the MKT [20]
In current application of the model, and are two fitting parameters. By fitting of
the correlation plot of the measured dynamic contact angle versus the velocity using
Eq. 3-4, the advancing static contact angle (Figure 3-2) was obtained. The
procedure followed by Vega et al. [21] was adapted to fit the data using the MKT.
Dynamic wetting of PET was previously studied and modelled using the MKT
theory by Blake [9]. Therefore, PET was used for two purposes, one in which the
wetting behaviour of the synthetic fibre and coir fibre were compared and secondly
PET was used as a reference to evaluate MKT fitting procedures.
3.2.3.2 A modification of Wilhelmy method
As a second method to determine static contact angles, a variant of the Wilhelmy
method was used for wetting measurements with the different test liquids. During
the advancing phase, the vertical movement of the liquid beaker was stopped at a
certain immersion length of the fibre during 360 s to allow relaxation of the liquid
meniscus to approach a static condition. The last data point of the wetting force at
the end of the relaxation process was used to calculate the static contact angle. In
Chapter 3 76
dynamic phase, various measurement speeds ranging from 2 m/s to 500 m/s were
applied. The relaxation time was selected based on measurements in sorption test, in
which coir fibre was immersed in test liquids to measure sorption mass as a function
of time. The sorption mass mainly was a sum of force created by wetting
(transformed into mass) and absorbed liquid into the fibre. The results showed that
the recorded mass initially fluctuated due to the movement of wetting line on the
rough surface of the fibres. Then, it typically reached a stable value after
approximately 200 s (Figure 3-6); which indicates that the static condition of wetting
had been reached, and the absorption process may be completed.
Figure 3-6. Typical sorption measurement curves in water of coir fibres having different perimeters,
showing that static equilibrium is reached after approximately 200 seconds.
The advancing static contact angle values obtained from the two above
approximations were used to estimate coir fibre surface energy and its components.
3.2.4 Fibre surface energy estimation
Contact angle measurements of a solid yield data that reflect the thermodynamics of
the solid–liquid interactions. These data can be used to estimate the surface energy
of the solid. Several approaches for the surface energy calculation were proposed by
Zisman [22], Fowkes [23], Wu [24] and Van Oss-Good [25, 26]. In this work, two
methods are used to determine fibre surface energy. The first method based on a
Wetting analysis and surface characterisation of coir fibres 77
geometric-mean approach was first given by Fowkes and later by Owens and Wendt
[27], and the later one is the acid-base approach developed by van Oss, Good and
Chaudhury [25, 26].
3.2.4.1 Geometric-mean approach (Owens and Wendt method)
In this method, the surface energy of the solid, , is divided into two components,
dispersive2,
, and polar,
, using a geometric-mean approach to combine their
contributions:
(3-6)
The interfacial energy of two phases can be approximated by:
(√
√
) (3-7)
By combining with the Young equation, , one obtains:
√
√
(
√
√
)
√
If one has obtained contact angle data on the fibre for a series of test liquids with
known surface tension components, the two unknowns and
are
simultaneously solved by linear fitting using Eq. 3-8, referred to as the Owens–
Wendt equation.
3.2.4.2 Acid-base approach (van Oss-Good method)
In the approach proposed by van Oss, Good and Chaudhury, it suggests that a solid
surface consists of two terms: the Lifshitz – van de Waals component3,
,and an
acid-base component, , which includes electron acceptor,
, and electron
donor, , components as shown in Eq. 3-9
and √
(3-9)
2 dispersive component is responsible for London dispersion interactions, and polar component contributes to
polar interactions including the hydrogen bond or acid-base interactions. 3 the Lifshitz – van de Waals interactions comprise dispersion, dipolar and induction interactions.
Chapter 3 78
When the solid is in contact with a liquid, the relation between their surface energy
components can be described by:
(√
√
√
) (3-10)
To determine the three unknown surface components, the contact angle
measurements are carried out using various test liquids (at least three liquids) with
known surface tension components, thus, a linear set of different equations is
obtained in the form:
(√
√
√
) (3-11)
The set of equations can be written in the matrix form:
(3-12)
where
[ √
√ √
√ √
√
√ √
√
]
[
]
[ √
√
√
]
In the current study, the same four test liquids are used as in Owens-Wendt method.
The approximate solution of the overdetermined system (four equations) can be
solved to obtain the three surface energy components of coir fibres. The software
SurfTen 4.3 developed by Claudio Della Volpe and Stefano Siboni, University of
Trento, Italy was used to perform the calculation of the surface components. More
details of mathematical approach of the calculation can be found in [28].
3.2.5 Fibre surface characterization using X-ray photoelectron
spectroscopy (XPS)
XPS analyses were performed on a Kratos Axis Ultra spectrometer (Kratos
Analytical, UK) equipped with a monochromatized aluminium X-ray source
(powered at 10mA and 15 kV). One single fibre was cantilevered fixed on a flat
stainless steel trough with a piece of double sided isolative tape. This way of
Wetting analysis and surface characterisation of coir fibres 79
mounting insured that the fibre surface only was analysed but not its surrounding.
The troughs, holding each fibre sample, were then inserted in the multispecimen
holder. The pressure in the analysis chamber was about 10−6
Pa. The angle between
the normal to the sample surface and the direction of photoelectrons collection was
about 0o. Analyses were performed in the hybrid lens mode with the slot aperture
and the iris drive position set at 0.5, the resulting analysed area was 700m×300m.
The pass energy of the hemispherical analyser was set at 160 eV for the survey scan
and 40 eV for narrow scans. In the latter conditions, the full width at half maximum
(FWHM) of the Ag 3d5/2 peak of a standard silver sample was about 0.9 eV.
Charge stabilisation was achieved by using the Kratos Axis Ultra device. The
electron source was operated with a filament current between 1.9 and 2.1A and a
bias of −1.1 eV. The charge balance plate was set between −3.3 and −3.9 V.
The following sequence of spectra was recorded: survey spectrum, C 1s, O 1s, N 1s,
Ca 2p, Si 2p, and C 1s again to check for charge stability as a function of time and
the absence of degradation of the sample during the analysis. The C–(C, H)
component of the C 1s peak of carbon was fixed to 284.8 eV to set the binding
energy scale.
The data analysis was performed with the CasaXPS program (Casa Software Ltd.,
UK). Mole fractions were calculated using peak areas normalised on the basis of
acquisition parameters after a linear background subtraction and consideration of
experimental sensitivity factors and transmission factors (depending on kinetic
energy, analyser pass energy and lens combination) provided by the manufacturer. C
1s spectra were decomposed with a Gaussian/Lorentzian (70/30) product function,
by constraining the FWHM’s of all components to be equal.
Untreated coir as used in the wetting measurements and n-Hexane modified coir
were characterised. A surface modification procedure of coir was carried out by
soaking untreated coir in n-Hexane for 24 h at ambient temperature, followed by
washing with deionised water, then drying under vacuum at 90 oC for 2 h.
Chapter 3 80
3.3 Results and discussion
3.3.1 Contact angle measurements
3.3.1.1 Fibre wetted perimeter
Figure 3-7. A typical wetting force–position curve in the experiment determining fibre wetted
perimeter using n-Hexane.
Figure 3-7 shows a typical plot of wetting force vs. immersion position during a
wetting measurement of a coir fibre in n-Hexane. The extrapolated value of the force
at the position of zero immersion depth was used for calculating the wetted
perimeter (at zero buoyancy).
Figure 3-8. Image analysis of fibre cross-section using the software Leica Qwin to determine fibre
wetted length, a coir (left) and PET (right)
y = -0.0002x + 0.0157
R2 = 0.8676
0.01
0.012
0.014
0.016
0.018
0 2 4 6 8 10 12
position (mm)
Fo
rce (
mN
)
PET COIR
lacuna
lumen
Wetting analysis and surface characterisation of coir fibres 81
In Figure 3-8, an image of a coir fibre cross-section was examined, in which the
fibre wetted perimeter is the sum of the fibre contour and the contour of the fibre
lacuna. It is reported that a lacuna occurs in coir fibres with different sizes
depending on fibre diameter and position along the fibre and that it would be bigger
in thicker fibres. The lumen of single fibre cells (elementary fibres) has a length and
width in the range of 0.71–1.06 mm [29, 30] and 6–19 m respectively. Therefore,
the effect of the lumen on the wetting measurement is small, and can be eliminated.
Thus, the wetted perimeter is mainly determined by the contour of the fibre and the
lacuna. In case of the PET fibre, the wetted perimeter is identical to the fibre
perimeter.
The wetted perimeter of PET fibres and five samples coir fibres is shown in Table 3-
2. For PET fibres, there is a good agreement between the results of two methods
with less than 3% difference. For the coir fibres, a quite good agreement is also
found. However, the results of the image analysis are systematically higher than
those of the n-Hexane wetting, in the order of 3–7%. A possible explanation for this
slight discrepancy is that the wetting of n-Hexane may not occur over the whole
length of the fibre lacuna, while image has been selected where the contour of the
fibre lacuna is included in the images analysis (although not necessarily the largest
lacuna cross-section may have been found). It is likely that the precise value of the
wetted perimeter is somewhere in between the values measured with the two
methods. When comparing the two methods, the wetting procedure is less time
consuming than the image analysis.
Table 3-2. Fibre wetted perimeter (mm) from different methods.
Fibre Methods
Coir
Fibre
No.
n-Hexane
wetting Image analysis
1 0.715 0.003 0.767 ± 0.014
2 0.442 0.003 0.462 ± 0.027
3 0.913 0.004 0.961 ± 0.027
4 0.777 0.003 0.799 ± 0.043
5 0.952 0.004 0.982 ± 0.042
PET 2.621 0.002 2.549 0.022
Chapter 3 82
3.3.1.2 Advancing dynamic contact angles
Figure 3-9. Typical dynamic contact angle measurements of coir and PET fibres.
Typical dynamic contact angle measurements of coir and PET fibres in water at a
measurement speed of 25 m/s are shown in Figure 3-9, after buoyancy subtraction.
The advancing angles of coir are observably stable, while there is a big scatter of the
receding contact angles. The unstable receding contact angles seem to be because of
two main effects. Firstly, the fibre roughness influences the receding measurement,
in which the test liquid remains on some areas of the fibre surface after the
advancing process. A another reason can be the effect of the nature of the cosine
function on the calculation of contact angle which makes the angle close to 0 degree
highly sensitive to the changes of the measured force. In case of PET, both
advancing and receding angles are steady. The difference of advancing and receding
angles is known as contact angle hysteresis and is explained by the nature of the
fibre surface [9], mainly due to the effect of adsorbed liquid when the liquid is
receding. The effect seems to be stronger with coir fibre, which has a higher surface
roughness, and more chemical and topographical heterogeneities than the PET fibre.
Observation of coir fibres by SEM, shows a circular cross-section and a relatively
smooth surface (Figure 3-3a and 3-3b). This structure of the coir surface apparently
leads to stable advancing angles, but it does lead to fluctuating receding angles. This
is an important reason why only advancing contact angles have been analysed.
0
20
40
60
80
100
120
0 2 4 6 8 10
co
nta
ct
an
gle
(0)
posiotion (mm)
coir1 coir2 coir3 PET
advancing
receding
Wetting analysis and surface characterisation of coir fibres 83
The dynamic contact angles of coir fibres in water, diiodomethane, ethylene glycol,
formamide and benzyl alcohol at different measurement speeds ranging from 2 m/s
to 8000 m/s were determined using the Wilhelmy method. The results indicate that
the advancing angles are speed-dependent in all the test liquids. The angles vary
from 80o to 105
o in water, 51
o to 64◦ in diiodomethane, 45
o to 95
o in ethylene glycol,
49o to 82
o in formamide and from 29
o to 48
o in benzyl alcohol respectively (Figure
3-10). This behaviour of angle speed-dependence also occurs with PET having a
contact angle in water ranging from 82o to 110
o in the same speed range as above.
Similar results with PET were also reported in the study of Blake [9].
Chapter 3 84
Wetting analysis and surface characterisation of coir fibres 85
Figure 3-10. Dynamic advancing contact angle at various measurement speed of (a) PET in water
(b) coir in water (c) coir in diiomethane (d) coir in ethylene glycol (e) coir in formamide (f) coir in
benzyl alcohol. Solid curves are non-linear regression of experimental data following MKT with
two fitting parameters and .
Chapter 3 86
3.3.1.3 Effect of liquid absorption on the contact angles
Since the contact angle is calculated from the measured wetting force, any incorrect
value of the force would lead to an error for the contact angle result. Table 3-3
shows the amount of liquid absorption in the fibres and its effect on the calculated
angles. The results indicate that the effect is small with water, diomethane and
ethylen glycol, and bigger with formamide and benzyl alcohol. The contact angles in
water, diiomethane and ethylene glycol seem to be not altered by the absorption.
However, they change approximately 1o in case of formamide and 2
o in case of
benzyl alcohol. Among these test liquids, the absorption of benzyl alcohol is quite
high. However, its influence is not pronounced since the absorbed mass is only
approximately 2 to 5% of the total wetting mass. Another consideration is that the
contact angles are calculated via a cosine function in the Wilhelmy equation.
Consequently, the nature of the cosine function makes the angles close to 90o less
sensitive to a change in the force than those close to 0o or 180
o. In this way, the
bigger variation in case of benzyl alcohol can also be explained by its low contact
angle with coir.
Table 3-3. The effect of absorbed liquid into coir fibre on its dynamic contact angle with various
test fluids; mass range is presented due to different perimeter of test samples
liquids absorption
(%)
absorbed mass
(mg)
(in range)
mass created
by wetting (mg)
(in range)
contact angle
(0)
contact angle
variation
(0)
number
of
samples
water 5.2 ± 1.1 0.013 - 0.023 0.175 - 0.956 82.94 ± 4.44 0.18 ± 0.02 5
diiodomethane 5.0 ± 2.1 0.012 - 0.019 0.698 - 1.986 54.35 ± 2.38 0.25 ± 0.10 5
ethylenglycol 3.0 ± 0.7 0.015 – 0.045 1.909 – 3.665 50.82 ± 3.40 0.59 ± 0.21 5
formamide 8.1 ± 4.9 0.016 – 0.138 1.642 – 2.550 61.68 ± 3.16 0.91 ± 0.59 5
benzyl alcohol 14.4 ± 5.8 0.021 - 0.073 1.366 - 3.225 24.31 ± 3.59 1.92 ± 0.89 5
3.3.1.4 Advancing static contact angle approximation using the Molecular-
kinetic theory
The dynamic advancing angles and MKT fitting curves of PET and coir in water are
presented in Figure 3-10. Using a characteristic length of 1.16 nm and an
equilibrium displacement frequency of 1.554 x 105 s
-1 for PET, the MKT fitted
well the experimental contact angle data with R2 = 0.94 (Figure 3-10a). In a previous
study of Blake, which models the dynamic wetting of water on PET at low speeds,
values of of 1.06 nm and of 2.5 x 105 s
-1 were found, which are quite similar.
Wetting analysis and surface characterisation of coir fibres 87
As noted in literature [9, 21], some level of discrepancy can be explained by
differences in fibre surface roughness and polymer crystallinity.
The same fitting procedure is applied for the wetting of water on coir by using
values of of 1.21 nm and of 0.124 x 105 s
-1 giving a good agreement between
experimental data and the MKT theoretical curve with R2 = 0.95. The fitting
parameters and are in the same order of magnitude when comparing between
coir and PET. The change of K0 is more pronounced than for but is not unusual.
Vega et al worked with Nylon fibre [21] and saw that increased by a factor of 70
when Nylon was studied with five different liquids. Here, The MKT fitting curve
provides a static contact angle of 77.3 degrees with water, showing relatively
‘hydrophobic’ properties of the coir fibre surface. The jump frequency in case of
coir fibre in water is lower than that in case of PET fibre, but the obtained static
contact angles are not very different. The low indicates a better wetting of coir
fibre in water than PET fibre (coir fibre surface is more polar than PET surface). So,
the polar properties of coir fibre may be underestimated due to a high static contact
angle in water.
Figure 3-10 also presents the MKT fit of the wetting of diiodomethane, ethylene
glycol, formamide and benzyl alcohol on coir. In diiodomethane, a good fitting is
obtained with of 2.18 nm, of 0.038 x 105 s
-1 and R
2 = 0.92 providing a static
contact angle of 48.2 degrees. A static angle of 47.1 degrees is obtained in
formamide when using of 1.15 nm, of 0.209 x 105 s
-1 with R
2 = 0.99. In
formamide, with of 1.33 nm, of 0.030 x 105 s
-1 with R
2 = 0.95, the static
contact angle is 47.7. And a good fit is obtained with R2 = 0.91 using of 2.19 nm
and K0 of 0.137 x 105 s
-1 in case of benzyl alcohol. The static angle of coir in benzyl
alcohol following the MKT fit curve is 25.1 degrees. On the same solid surface, the
displacement length of the molecules from one adsorption site to another is related
to the molecular size of liquid. In the five test liquids, it is observably logical that
is smallest in case of water and biggest in case of benzyl alcohol.
3.3.1.5 Static contact angle from relaxation experiments
Figure 3-11 shows a typical result of the static contact angle approximation at
different speeds in dynamic contact angle measurements. At a given speed, the
dynamic contact angles are measured in dynamic wetting after which the movement
is stopped. A relaxation process at maximum immersion depth is then followed in
order to approach a static condition after the relaxation of the liquid meniscus. At
the end of the process, an apparent static angle is obtained.
Chapter 3 88
Figure 12 shows the relationship between the dynamic angles and the static angles,
as determined by the static angle approximation procedure. It can be observed that
the dynamic angles are speed dependent, while the static angles are steady at
different applied measurement speeds. The difference between the dynamic angles
and the static angles is smallest at lowest speed, and increases with increasing speed.
The obtained static angles are all distributed around the same value in different
measurements performed at a range of speeds. The mean values of the static angles
in different liquids are also presented in Table 3-4. It is likely that these angles are
stable with small deviation, and must be the advancing static angles. Moreover,
there is a good agreement between the advancing static contact angles of MKT
fitting and these of this method.
advancing self relaxation
static angle
Figure 3-11. Typical experimental results of static contact angle approximation during stationary
immersion of coir in water.
Wetting analysis and surface characterisation of coir fibres 89
Figure 3-12. Advancing contact angles () and static contact angles corresponding to different
measurement speeds in dynamic phase (∎) of (a) PET in water (b)coir in water (c) coir in
diiodomethane (d) coir in ethylene glycol (e) coir in formamide (f) coir in benzyl alcohol.
(c) coir in diiodomethane (d) coir in ethylene glycol
(e) coir in formamide (f) coir in benzyl alcohol
(a) PET in water (b) coir in water
Chapter 3 90
3.3.2 Surface energy of coir fibre
Based on the above approaches, the results of advancing static angles can be used to
estimate the coir surface energy, consisting of different components following two
methods (Owens-Wendt and van Oss-Good). The contact angles in the liquid set of
water – diiodomenthane – ethylene glycol – formamide, which shows a broad range
of surface tensions and diversity in their surface tension components, are used for
estimation of fibre surface energy.
Following the Owens-Wendt method, the static contact angles of coir fibre in
various test liquids can be transformed into values of the term
√ and plotted as a function of the term √
√
(Eq. 3-8), as shown in
Figure 3-13. From the slope and the intercept of the linear relationship, respectively
the polar and dispersive parts of the coir fibre surface tension are calculated, as can
be seen in Table 3-4. Since the static contact angles of the test liquids on the coir
fibre can be obtained from two different ways (MKT fitting and relaxation), the
surface energies were calculated using data of two above described methods.
Figure 3-13. Owens-Wendt plot to estimate the surface energy of coir fibre using the static contact
angles obtained from MKT method () and the static relaxation method (□).
Wetting analysis and surface characterisation of coir fibres 91
The surface energy of coir fibre comprising three components (acid-base approach)
is also determined, and presented in Table 3-4. The calculation of the acid-base
surface energy components was performed by using SurfTen 4.3 software developed
by Claudio Della Volpe and Stefano Siboni, at the University of Trento, Italy.
Considering the series of contact angles obtained from the different methods, the
total surface energy of coir fibre is around 40 mJ/m2 (Owens-Wendt method) with a
high dispersive fraction of 34-35 mJ/m2. The surface energy is close to the range
often quoted for hydrophobic materials (from 28 to 34 mJ/m2) [31]. Several studies
on coir fibres indicate that mainly waxes exist on the coir surface. Moreover, the
coir surface demonstrates the presence of longitudinally orientated cells with more
or less parallel orientations (Chapter 2). The intercellular space is filled up by the
binder lignin and fatty substances that hold cells firmly bonded in the fibre [32, 33].
Thus, the surface energy of coir should be influenced by a combination of waxes,
fatty substances and lignin. Bartell and Zuidema provide values of the surface
energy dispersive part of waxes of 26.5 mJ/m2 and for the polar part values of zero
[34], while the surface energy of lignin film is found to be 52.5 mJ/m2 with
dispersive component of 40-43.5 mJ/m2 in the work of Lee and Lunar [35]. Our
estimated surface energy of coir is somewhat higher than that of waxes and lower
than that of lignin. It seems to reflect well the properties of the coir surface that
exists not only of waxes with non-polar surface tension but also of lignin and fats
with polar components.
The surface energy of coir fibre is approximately 37.5 mJ/m2 according to acid-base
method, and consists low acid fraction of 0.3 mJ/m2 and high base contribution in
range of 3-9 mJ/m2, while the Lifshitz-van der Waals component is around 35
mJ/m2. The result indicates that coir fibre surface is rather hydrophobic and has
negative charge (high Lewis basicity).
Chapter 3 92
Table 3-4. Summary of advancing static contact angles in different test liquids and estimated surface energy of coir fibre, according to 2 methods to
determine the static contact angles
Methods Static contact angle (
0)
Surface energy (mJ/m2)
disperse-polar (Owens-Wendt) acid-base (van Oss-Good)
Water Diiodomethane Ethylen
glycol Formamide
LWS
MKT fit 77.3 ± 0.3 48.2 ± 0.3 47.1 ± 0.5 47.7 ± 1.2 40.4 ± 1.4 35.1 ± 1.3 5.3 ± 0.5 37.5 ± 0.2 35.5 ± 0.2 0.33 ± 0.03 3.17 ± 0.12
Wilhelmy static
approximation 75.6 ± 5.9 50.9 ± 2.1 46.6 ± 3.0 45.7 ± 3.9 40.2 ± 3.6 33.8 ± 1.6 6.4 ± 0.7 37.3 ± 1.4 34.1 ± 1.2 0.28 ± 0.18 9.19 ± 2.38
Table 3-5. Relative atomic percentages, O/C ratio, and decomposition of C 1s peaks obtained by XPS on untreated and modified coir fibres
Fibres C O N Si O/C Binding energy (eV)
(%) (%) (%) (%) 284.8 ± 0.1 286.3 ± 0.1 287.5 ± 0.3 288.8 ± 0.1
C1 (%) C2 (%) C3 (%) C4 (%)
(C-C/C-H) (C-O) (C=O/O-C-O) (O-C=O)
untreated coir 74.9 ± 3.3 21.8 ± 4.5 1.7 ± 0.4 0.9 ± 0.7 0.3 ± 0.1 66.2 ± 10.4 23.1 ± 5.9 6.2 ± 3.0 4.5 ± 2.4
n-hexane modified coir 69.7 ± 1.4 26.6 ± 1.5 3.1 ± 0.5 0.4 ± 0.1 0.4 ± 0.1 53.1 ± 1.0 30.2 ± 0.8 9.2 ± 0.6 7.6 ± 0.9
Wetting analysis and surface characterisation of coir fibres 93
3.3.3 Surface chemical analysis of coir fibre
Relative atomic percentages of the elements detected by XPS, together with the
oxygen to carbon atomic ratio of untreated and modified coir fibres are provided in
Table 3-5. It is observed that a high proportion of carbon in untreated coir may
represent a hydrocarbon rich waxy layer on the surface. In the same fashion, the low
oxygen-carbon ratio also indicates a high proportion of aliphatic and aromatic
carbons [36].
After the modification of the coir surface with n-hexane, the carbon percentage
decreases while the O/C ratio increases. The O/C value of 0.38 is close to that
reported for thio lignin (O/C of 0.38) and dioxane lignin (range of 0.31-0.36), but
still far different from cellulose having an O/C ratio of 0.83 [37, 38]. Therefore, the
surface of n-hexane modified coir likely has a greater proportion of lignin. It is
probable that waxes and fatty substances on the coir surface are washed away by n-
hexane, to expose the lignin which binds the elementary fibres.
Figure 3-14. Typical C 1s spectra, decomposed into four components C1-C4 for (a) untreated coir
(b) n-hexane modified coir
In Figure 3-14, typical results of C 1s spectra for untreated and n-hexane modified
coir are compared. The C 1s peak is decomposed into four sub-peaks C1-C4
representing: carbon solely linked to carbon or hydrogen C-C or C-H (C1), carbon
singly bound to oxygen or nitrogen C-O or C-N (C2), carbon doubly bound to
oxygen O-C-O or C=O (C3) and carbon involved in ester or carboxylic acid
functions O=C-O (C4), as also shown in Table 3-5. For both untreated and modified
coir fibres, C1 is higher than C2, C3 and C4. The high value of C1 indicates the
C-O/C-N
C-(C,H)
O-C-O /C=O O=C-O
C-O/C-N
C-(C,H)
O-C-O /C=O
O=C-O
(a) (b)
Chapter 3 94
presence of unoxidised carbon atoms at the surface, which can be attributed to
hydrocarbon in extractives and lignin. In the untreated coir, the high proportion of
C1 carbon (66.2%) suggests a combination of hydrocarbon rich waxes and lignin.
This is supported by the low proportion of C2, C3 and C4. The modified fibre shows
lower C1 and higher C2, C3 and C4 than in case of the untreated one. This again
points at a large amount of lignin present at the surface after removing the waxes by
the n-hexane as discussed above, and maybe also a possible partial exposure of
cellulose (rich in C2 function).
3.4 Conclusions
The wetting behaviour of coir fibre was characterised by dynamic wetting
measurement in various test liquids using the Wilhelmy technique. The dynamic
advancing contact angles corresponding to different measurement speeds were well
fitted by the Molecular-Kinetic Theory, which allows modelling the dynamic
wetting of coir and determining its static contact angle. Another experimental
approach for static contact angle measurement was carried out based on the
Wilhelmy technique, in which the relaxation of the liquid meniscus after stopping
the fiber movement was monitored. The obtained static contact angles were stable
and reproducible. Moreover, there was a good agreement between the two methods
for static contact angle approximation. The values of the static angles were further
used to estimate the fibre surface energy.
The estimated surface energy of coir fibres, comprising of high dispersive and low
polar contributions, pointed to a surface with rather hydrophobic properties. Surface
chemical analysis of the fibre by XPS indicated a high proportion of hydrocarbon
rich material, which could be attributed to waxes, fatty substances and lignin, in
agreement with the results from the wetting measurements.
The study of wetting and surface chemistry offers a deeper understanding of the coir
fibre surface, which will assist in the selection of fibre treatments or matrix
modifications for future improvements of the interface in coir fibre composites.
References
1. Stokes, R.J., Fundamentals of interfacial engineering. 1997, Wiley-VCH.
2. Young, T., Phil. Trans. R. Soc. , 1805(95): p. 65-87.
3. Le, C.V., Measuring the contact angles of liquid droplets on wool fibers and
determining surface energy components. Textile research journal, 1996. 66(6): p.
389.
Wetting analysis and surface characterisation of coir fibres 95
4. Silva, J.L.G. and H.A. Al-Qureshi, Mechanics of wetting systems of natural fibres
with polymeric resin. Journal of Materials Processing Technology, 1999. 93: p. 124-
128.
5. Aranberri-Askargorta, I., T. Lampke, and A. Bismarck, Wetting behavior of flax
fibers as reinforcement for polypropylene. Journal of Colloid and Interface Science,
2003. 263(2): p. 580-589.
6. Carroll, B.J., The accurate measurment of contact angle, phase contact area, drop
volume, and La Place excess pressure in drop-on-fibre system. J. Colloid Interface
Sci., 1976. 57: p. 488–495.
7. Yamaki, J.I., New method of determining contact angle between monofilament and
liquid. Journal of Applied Polymer Science, 1975. 19(10): p. 2897.
8. Barsberg, S. and L.G. Thygesen, Nonequilibrium phenomena influencing the
wetting behavior of plant fibers. Journal of Colloid and Interface Science, 2001.
234(1): p. 59-67.
9. Berg, J.C., Wettability. 1993, New York: Marcel Dekker.
10. Blake, T.D., The physics of moving wetting lines. Journal of Colloid and Interface
Science, 2006. 299(1): p. 1-13.
11. Dussan, E.B., On the spreading of liquids on solid surfaces: static and dynamic
contact lines. Annual Review of Fluid Mechanics, 1979. 11(1): p. 371.
12. Belgacem, M.N., et al., Surface characterization of cellulose fibres by XPS and
inverse gas chromatography. Cellulose, 1995. 2(3): p. 145-157.
13. Johansson, L.S., et al., On surface distributions in natural cellulosic fibres. Surface
and Interface Analysis, 2004. 36(8): p. 706-710.
14. Sgriccia, N., M.C. Hawley, and M. Misra, Characterization of natural fiber
surfaces and natural fiber composites. Composites Part a-Applied Science and
Manufacturing, 2008. 39(10): p. 1632-1637.
15. Topalovic, T., et al., XPS and contact angle study of cotton surface oxidation by
catalytic bleaching. Colloids and Surfaces a-Physicochemical and Engineering
Aspects, 2007. 296(1-3): p. 76-85.
16. Rulison, C., So you want to measure surface energy?, Kruss Technical Note #306,
Kruss USA.
17. Della Volpe, C., Acidbase surface free energies of solids and the definition of scales
in the Goodvan OssChaudhury theory. Journal of Adhesion Science and
Technology, 2000. 14(2): p. 235.
18. Volpe, C.D., Some reflections on acid-base solid surface free energy theories.
Journal of Colloid and Interface Science, 1997. 195(1): p. 121.
19. Blake, T.D. and Blake, Kinetics of liquidliquid displacement. Journal of Colloid and
Interface Science, 1969. 30(3): p. 421.
20. Blake, T.D. and J. De Coninck, The influence of solid-liquid interactions on
dynamic wetting. Advances in Colloid and Interface Science, 2002. 96(1-3): p. 21-
36.
21. Vega, M.J., et al., Experimental investigation of the link between static and dynamic
wetting by forced wetting of nylon filament. Langmuir, 2007. 23(21): p. 10628-
10634.
22. Zisman, W.A., Relation of the equilibrium contact angle to liquid and solid
constitution. Advances in Chemistry Series, 1964. 43(1): p. 51.
Chapter 3 96
23. Fowkes, F.M., Additivity of Intermolecular Forces at Interfaces .1. Determination
of Contribution to Surface and Interfacial Tensions of Dispersion Forces in Various
Liquids. Journal of Physical Chemistry, 1963. 67(12): p. 2538-&.
24. Wu, S., Polym Sci Part C: Polym Symp 1971, 34, 19.
25. Van Oss, C.J. and Vanoss, The role of van der Waals forces and hydrogen bonds in
"hydrophobic interactions" between biopolymers and low energy surfaces. Journal
of Colloid and Interface Science, 1986. 111(2): p. 378.
26. Van Oss, C.J., Interfacial Lifshitz-van der Waals and polar interactions in
macroscopic systems. Chemical reviews, 1988. 88(6): p. 927.
27. Owens, D.K., Estimation of the surface free energy of polymers. Journal of Applied
Polymer Science, 1969. 13(8): p. 1741.
28. Della Volpe, C., The solid surface free energy calculation: I. In defense of the
multicomponent approach. Journal of Colloid and Interface Science, 2004. 271(2):
p. 434.
29. Dam, J.E.G. and D. van, Process for production of high density/high performance
binderless boards from whole coconut husk: Part 2: Coconut husk morphology,
composition and properties. Industrial crops and products, 2006. 24(2): p. 96.
30. Jensen, W.A., Botanical Histochemistry: Principles and Practice. 1962, San
Francisco: W.H. Freeman.
31. Van Hazendonk, J., et al., A simple experimental method for the measurement of the
surface tension of cellulosic fibres and its relation with chemical composition.
Colloids and surfaces A: Physicochemical and engineering aspects, 1993. 81: p.
251-261.
32. Rout, J., Scanning electron microscopy study of chemically modified coir fibers.
Journal of Applied Polymer Science, 2000. 79(7): p. 1169.
33. Bismarck, A., Surface characterization of natural fibers; surfaceproperties and the
water up-take behavior of modified sisal and coirfibers. Green chemistry, 2001.
3(2): p. 100.
34. Bartell, F. and H. Zuidema, Wetting characteristics of solids of low surface tension
such as talc, waxes and resins. Journal of the American Chemical Society, 1936.
58(8): p. 1449-1454.
35. Lee, S. and P. Luner, The wetting and interfacial properties of lignin. Tappi, 1972.
55(1): p. 116-121.
36. Panthapulakkal, S. and M. Sain, Agro-residue reinforced high-density polyethylene
composites: Fiber characterization and analysis of composite properties.
Composites Part A: Applied Science and Manufacturing, 2007. 38(6): p. 1445-1454.
37. Dorris, G.M. and D.G. Grey, The surface analysis of paper and wood fibers by
ESCA. I. Application to cellulosics and lignin,. Cellulose Chem. Technol., 1978. 12:
p. 9-23.
38. Dorris, G.M. and D.G. Grey, The surface analysis of paper and wood fibers by
ESCA. II. Surface composition of mechanical pulp. Cellulose Chem. Technol.,
1978. 12: p. 721-734.
Interfacial adhesion and compatibility of coir fibre composites 97
Chapter 4
Interfacial adhesion and compatibility
of coir fibre composites
Chapter 4 98
4.1 Introduction
Fibre polymer composites provide their advantages based on the combination of the
properties of the constituents, in which the fibres act as high strength and stiffness
component and the surrounding matrix keeps them in a desired location and
orientation, providing the structural integrity. In such system, the mechanical
properties of the composites cannot be achieved with either fibre or matrix playing
alone. Therefore, the composites work with the presence of fibre-matrix interface.
The interface is a surface formed by the boundary of the fibre and matrix, which
maintains the bonds between them for the load transfer, in order to make the effect
of fibre reinforcement available.
The composite structure can be considered at four structural levels: molecular level,
micro level, meso level and bulk composite [1], as shown in Figure 4-1. At the
molecular level, the interaction between the fibre and the matrix is determined by
chemical groups presented on the surface of both phases. At this level, the interfacial
adhesion depends on the physico-chemical interaction (e.g. van der Waals forces,
acid-base interactions, hydrogen bond) and chemical bonds (covalent bonds). The
physico-chemical interaction is quantitatively characterised by the work of adhesion.
At the micro level, considering a single fibre composite, the interfacial adhesion is
mainly described in term of various parameters such as interfacial tensile strength,
interfacial shear strength, debonding stress, critical energy release rate, etc., which
characterise the stress transfer through the interface.
The meso level takes into account the fibre distribution in the composite,
determining the stress transfer between the fibres. The last level characterises a part
of composite as bulk material; thus is not the focus of this chapter.
Regarding to the investigation of the interface, both molecular and micro levels are
mainly considered in the literature. At the first level, the fundamental adhesion
related to surface properties and adhesion bonds is studied from the viewpoint of
chemistry and molecular physics. From an engineering point of view, the micro
level is important for investigating the interfacial adhesion, which determines stress
transfer efficiency and the debonding stress (interfacial strength). In fact, molecular
interactions at the interface directly affect interfacial strength [2, 3], and hence the
characterisation of the interface adhesion at different levels provides a consistent
understanding of the composite interface, which strongly decides the final properties
of the composites.
Interfacial adhesion and compatibility of coir fibre composites 99
Figure 4-1. Structural levels of fibre reinforced composite in considering fibre-matrix interfacial
interactions (a) molecular level (b) micro level (c) meso level (adapted from [1])
To understand the interactions at the molecular level, wetting parameters including
contact angle, surface energy and work of adhesion have to be assessed. At the
higher levels, the quality of the composite interface also can be characterised by
mechanical tests which are performed on either single fibre micro-composites or
bulk laminate composites. In the former, a single fibre is embedded in a matrix
block of different shapes and sizes. Then, interfacial shear strength (IFSS) is
determined in various ways which comprise of pull-out, fragmentation and micro-
indentation tests. Regarding bulk laminates, several testing techniques have been
developed for unidirectional fibre composites such as transverse tensile and bending
tests, short beam shear tests and the Iosipescu shear test. In these tests, the interface
quality is characterised by either the transverse interfacial tensile strength (mode I)
or interlaminar shear strength (ILSS).
Every abovementioned test has shown its advantages and limitations, which are
mostly concerned with test sample preparation and properties of fibre and matrix,
especially of the fibre surface [4]. In natural fibre composites, the dependency of the
measured fibre-matrix interface properties on the particular test method can be
aggravated due to their irregular geometry and surface condition. Therefore, a
combination of test methods will provide a better understanding of the composite
interface.
Fibre Matrix
(a) (b) (c)
Chapter 4 100
The aim of this chapter is to study the interfacial adhesion between untreated and
alkali treated coir fibre composites and various thermoplastics. Based on the wetting
analysis described in the previous chapter, the contact angles of fibres and matrices
in various test liquids are determined, which are used to estimate their surface
energies and surface energy components. The fibre-matrix work of adhesion and
interfacial energy are calculated to predict the physical adhesion and compatibility
of the composites. Flexural transverse three-point bending tests on unidirectional
composites and single fibre pull-out tests are performed for determining the
interfacial tensile strength and interfacial shear strength (IFSS), to examine the
interface quality and to obtain a deeper understanding of the interfacial interactions.
4.2 Materials and methods
4.2.1 Materials
Fibres
Untreated and alkali treated coir fibres were used in this study. The cleaning
procedure for the untreated fibre was described in chapter 3. The treated coir fibres
were obtained by soaking the fibres with 5% NaOH solution for 2h at room
temperature; they were then washed thoroughly with de-ionized water and dried in a
vacuum oven at 90°C. The alkali treatment was expected to remove waxes and fatty
substances from the untreated fibre surface.
Matrices
Three thermoplastic matrices were used, supplied as films: polypropylene,
polyvinylidene fluoride and maleic anhydride grafted polypropylene. The
polypropylene (PP) was an unmodified grade supplied by Propex GmbH (Germany).
Polyvinylidene fluoride (PVDF) provided by Solvay (Belgium) was used to study
the influence of surface energy differences on the composite interface. To study the
effect of chemical bonding at the interface on the interfacial adhesion, 0.3% maleic
anhydride grafted polypropylene (MAPP) (supplied by Dupont, Switzerland)) was
used for comparison. Table 1 shows the thermal properties and tensile properties of
the matrices.
Interfacial adhesion and compatibility of coir fibre composites 101
Table 4-1. Thermal and mechanical properties of thermoplastics (which are measured or taken from
data sheet)
Matrix
Tg
(oC)
Tc
(oC)
Tm
(oC)
Density
(g/cm3)
CTE
(10-6
/K)
Tensile properties
E-
Modulus
(GPa)
Strength
(MPa)
Strain
at
failure
(%)
PP - 115.7 160.6 0.9 62.7-73.2 1.6-1.8 55-65 > 300
PVDF -30 142.5 171.6 1.78 54.4-67.3 0.8-1.2 22-30 > 300
MAPP - 100.3 147 0.89 112.2-175.8 2.2-2.6 78 5-10
4.2.2 Wetting analysis
4.2.2.1 Contact angle measurements
Contact angle measurements of the coir fibres were carried out using the Wilhelmy
technique, which allows to determine dynamic contact angles of various test liquids
on the fibres. As discussed in Chapter 3, the significance of the contact angle as a
measure of wettability is based on equilibrium thermodynamic arguments, for which
static systems are most frequently studied [5]. Hence, it would be more reliable if
static equilibrium contact angles are introduced in the surface free energy
calculation.
In order to obtain the static contact angle, the molecular-kinetic theory (MKT) was
used to model dynamic wetting of the fibres following Eq. 4-1. By using
experimental data of dynamic angles, the static equilibrium angle can thus be
determined [6]. More details were presented in Chapter 3.
[
]
where is the equilibrium molecular jump frequency, is the distance between
two adsorption sites, (or ) is the surface tension of the liquid in contact with
vapour, is the static equilibrium contact angle and is the dynamic contact angle
corresponding to measurement velocity ; n is the number of adsorption sites per
unit area, is Boltzmann’s constant and T is the temperature.
Chapter 4 102
For the matrices, the equilibrium contact angles were estimated from their advancing
( ) and receding angles ( ), which were measured using the Wilhelmy method
on film samples. Considering that the sample surfaces were smooth and chemically
homogeneous, a model of arithmetic mean of the corresponding cosines of
advancing and receding angles as shown in Eq. 4-2 was used for the calculation of
the equilibrium angles [7-9].
(4-2)
The measurements were carried out on the polymer films, with a measurement speed
of 25 m/s and in the same test liquids as used for the fibres.
The test liquids used for the measurement of both fibres and matrices were ultrapure
water (18.2 cm resistivity), diiodomethane (Merck), ethylene glycol and
formamide (Sigma-Aldrich), and their physical properties were shown in Chapter 3.
4.2.2.2 Estimation of surface energy and work of adhesion
Surface energy
Surface energies of the fibres and matrices were estimated using the data of contact
angles of various test liquids on the fibres and matrices. Following the procedure
presented in Chapter 3, both the Owens-Wendt [10] and the van Oss-Good [11, 12]
methods were applied to determine the surface energies. In the former approach,
total surface energy, , is the sum of dispersive , , and polar ,
, components.
While, in the latter approach, it comprises of a Lifshitz – van de Waals component,
, and an acid-base component,
, which includes electron acceptor, and
electron donor, components as shown in Eq. 4-3 and Eq. 4-4
(4-3)
and √
(4-4)
Work of adhesion
As discussed above, the physical interactions between fibres and matrix are always
present at the interface, and they contribute to the interfacial adhesion. When
chemical bonds are neglected, the physical adhesions, which are related to surface
energies of two phases, play an important role to the interfacial strength. In order to
establish interfacial adhesion, sufficient contact is required between the fibres and
Interfacial adhesion and compatibility of coir fibre composites 103
the matrix which forms during composite processing. This ‘wetting’ is furthermore
also related to the thermodynamic work of adhesion, which can be used as a
quantitative measure for the fibre-matrix physical interactions.
The work of adhesion is the work required to disjoin a unit area of the solid-
liquid interface by creating a unit area of liquid-vacuum and solid-vacuum interface
[5], and defined by the Dupre equation:
(4-5)
where is the solid surface energy, is the liquid surface tension, and is
interfacial energy.
The work of adhesion can be expressed in relation to the equilibrium contact angle
by combining Eq. 4-5 with Young equation [13], and results in:
(4-6)
Nevertheless, Eq. 4-6 cannot be used to determine the work of adhesion directly
from the measured matrix/fibre contact angles, since it is very difficult to measure
directly the contact angle between a fluid (a viscous thermoplastic melt) and a fibre.
Therefore, the solid surface energies of matrices ( ) and fibre ( ) are determined
separately and for the work of adhesion it is assumed that the matrix surface energy
in the melt is similar to the solid.
The work of adhesion can be estimated using the geometric mean approach
(Owens-Wendt) for the dispersive-polar model or the acid-base approach (van
Oss-Good) for three surface energy components model.
Geometric mean approach
Berthelot [14], at the end of the 19th
century, assumed that the work of adhesion was
equal the geometric mean of the work of cohesion of a solid and the work of
cohesion of a liquid :
√ (4-7)
Fowkes, in 1964, proposed that the surface energy of a solid and of a liquid is a sum
of different components associated with specific interactions [15, 16]: the dispersive,
Chapter 4 104
, polar,
, hydrogen (related to hydrogen bonds),
, induction, , acid-base,
, components, and
refers to all remaining interaction components.
(4-8)
According to Fowkes, the work of adhesion between the solid and the liquid only
depends on the dispersive interactions, based on Eq. 4-7 it can be expressed as:
√
√
Afterwards, Owens and Wendt changed the idea of Fowkes by assuming that the
sum of all components in Eq. 4-8, except dispersive component , can be
considered as associated with the polar interaction . Accordingly, the work of
adhesion following the geometric mean approach was determined as follows
(√
√
)
Acid-base approach
Later on, van Oss, Chaudhury, and Good proposed the surface energy comprises
three components as shown in Eq. 4-4. Then, the work of adhesion according to the
acid-base approach was expressed as:
(√
√
√
)
It should be noted that the work of adhesion is a thermodynamic quantity referring
to the reversible work needed to create two new surfaces from a defect free
interface. The work of adhesion can correlate well with the interfacial strength [17],
but it is not sufficient to characterise the interfacial strength which is affected by
irreversible processes (e.g. inelastic deformation, voids at the interface).
Interfacial energy
The interfacial energy is the reversible work of forming a unit of solid-liquid
interface. It was proposed that minimizing the interfacial energy would yield a more
stable system and hence increase the adhesive strength. There are various
Interfacial adhesion and compatibility of coir fibre composites 105
expressions of the interfacial energy, which related to different approaches of
surface energy components of the two phases. Here, the description of the interfacial
energy is presented following the Owens and Wendt and the acid-base approaches.
In the Owens and Wendt approach, the surface energy of the solid and the liquid
comprises the dispersive and the polar contributions, and the interfacial energy is
described as in Eq. 4-12. While, in the acid-base approach, the three components of
the surface energy are considered, and the description of the interfacial energy is as
in Eq. 4-13.
(√
√
)
(√
√
√
√
√
) (4-13)
The interfacial energy is minimised when the surface energy components of the two
materials are equal, in which two phases have the same dispersive and polar
components in Owens-Wendt approach, and the acidic component of one phase is
equal to the basic component of the other in acid-base approach [18]. The interfacial
energy can be considered as an indicator of fibre-matrix interfacial compatibility. A
low interfacial energy indicates high interfacial fibre-matrix compatibility.
4.2.3 Single fibre pull-out test
4.2.3.1 Sample preparation
The fibre-matrix adhesion in the composite was evaluated using single fibre pull-out
tests. To prepare the pull-out test samples, coir fibres were fixed on an aluminium
frame. At both fibre ends, the fibres were passed through silicon bars with the
intention to keep the fibres free from matrix in these zones during the later
processing stage. The whole set up was placed in a mould in which matrix films
were stacked around the fibres. The set up is shown in Figure 4-2. Compression
moulding was performed using a Pinette hot press. The processing parameters were
175 0C for PP and 185
0C for PVDF and MAPP respectively, and 10 bar pressure for
all samples. After moulding, the silicon bars were removed. In the obtained samples,
the embedded fibre length inside the matrix was controlled by drilling holes through
the fibre and matrix at a defined distance from the entrance point of the fibre, as
shown in Figure 4-2c.
Chapter 4 106
Figure 4-2. Sample preparation for pull-out test (a) compression moulding used for embedding coir
fibre in thermoplastic matrices (b) pull-out test sample in aluminium frame (c) holes made in the
sample to determine fibre embedded lengths.
4.2.3.2 Evaluation of interfacial strength using a stress-based model
The pull-out test was performed on the pull-out samples at different embedded
lengths using a universal mini Instron testing machine. The pull-out load as a
function of displacement was obtained, and the peak force was used to calculate the
apparent interfacial shear strength (apparent IFSS, ), for each sample following
Eq. 4-14
⁄ (4-14)
where is the fibre diameter (which is in the range of 250-350 m for the studied
coir fibres) and is the fibre embedded length.
The above apparent IFSS would be constant over the whole embedded fibre length
in the case of a ductile fracture of the interface, where plastic yielding at the
interface takes place under loading. In this case the interfacial shear stress is uniform
and independent of the embedded fibre length. However, in case of brittle interface
fracture, an inhomogeneous stress field appears and the interfacial shear stress is as
non-uniform. Therefore, the apparent IFSS calculated from Eq. 4-14 can only simply
distinguish between “good” or “poor” interfacial bonding [19].
To have an adequate characterization of the fibre-matrix interface properties, a
detailed analysis is required, where the actual mechanism of interfacial failure is
taken into account. In literature, two main approaches for analysis the interfacial
Interfacial adhesion and compatibility of coir fibre composites 107
debonding can be found: a stress-based model and an energy-based model. In the
first model, a local interfacial shear stress is determined, while in the second model a
value for the interface fracture toughness or critical energy release rate is derived
from the experimental data. In this study, the stress-based model will be used for
analysis of interfacial strength. In this context, the local interfacial shear stress, ,
to create debonding initiation is used instead of average or apparent shear stresses.
Also, the separation of the contribution of bonding and friction is considered [1, 20,
21].
To characterise the interfacial strength by , the debonding force, , should be
known. The value can be determined by continuous visual monitoring of crack
initiation in case of transparent matrices, or determining the ‘kink’ points of the
force-displacement curves. In some cases both above methods are impossible and
only the maximum value of applied load, , is obtained. Then, the analysis of
will be carried out based on modelling of the measured , which will be
described in the following sections.
Calculation of from the ‘kink’ force
As mentioned, the ‘kink’ in the force-displacement curve can be recognised in some
cases, as shown in Figure 4-3, in which the external load reaches a certain critical
value, , where the fibre starts to debond from the matrix through interfacial crack
propagation. In the following stage, the force continues increasing while the debond
grows in a stable way until the peak force ( ) is reached. The force indeed must
increases because a frictional load in debonded regions appears and has to be added
to the adhesion load from the intact part of the interface. Afterwards, the peak force
drops and the whole embedded length suddenly and fully debonds.
Chapter 4 108
Figure 4-3. Typical force-displacement curve in pull-out test which shows a ‘kink’ at debonding
force , and maximum force (after Zhandarov [20])
The local interfacial shear stress in relation to the debonding force is described
in a model developed by Zhandarov et al. [1, 20-22] as follows
where is the shear-lag parameter as determined by Nayfeh [23]
[
(
)]
with is radius of the fibre, is tensile modulus of the fibre and is shear
modulus of the fibre, and are tensile modulus and shear modulus of the
matrix respectively, and and are the fibre and matrix volume fraction (within
the sample) respectively.
For the pull-out test samples in this study, the fibre volume fraction is calculated
based on the geometry of the sample where the fibre is considered to be embedded
in a matrix cylinder with diameter equal to the thickness of sample.
The Nayfeh shear-lag parameter is different from the Cox parameter commonly used
in shearlag models. Nairn et al. reported that model using the Nayfeh parameter
gives correct results for finite specimens [24, 25]
Interfacial adhesion and compatibility of coir fibre composites 109
is a residual stress due to thermal shrinkage when the composite is formed at high
temperature [26], described as follows
( )
with and are the longitudinal coefficient of thermal expansion (CTE) of the
fibre and the matrix respectively, and is the temperature difference between test
temperature and the stress-free temperature. The stress-free temperature is usually
taken as crystallisation temperature of semi-crystalline polymers or glass transition
temperature of amorphous polymers.
Estimation of and from fitting theoretical to experimental data
The stress-based model assumes that the process of interfacial crack growth is
governed by the local shear stress at the interface. For any crack length, , the shear
stress at the crack tip is assumed to be constant [21]:
(4-18)
In this consideration, the current load, , applied to the fibre end can be related to
the current crack length as follows:
(4-19)
Based on shear lag model of stress transfer from the fibre to the matrix, Zhandarov
et al. have developed several models [21, 22, 27] to describe Eq. 4-20. The
following expression has been proven to be sufficiently accurate for the analysis.
{ [ ] [ ] [
[ ]
] }
where is the crack length and is the frictional stress in the already debonded
regions.
In pull-out tests, the recorded peak force indicates the total debonding of the
fibre from the matrix, which is attributed by both the adhesion and friction in the
system. Hence, it is necessary to investigate the interfacial properties by mean of the
interfacial parameters and . A procedure to estimate and is proposed by
Zhandarov et al. [26], which is a two-parameter ( and ) fit of measured and
Chapter 4 110
theoretical peak load as a function of embedded length using a standard
least-squares method. The theoretical peak load is obtained with the help of Eq. 4-20
which gives a relation between current applied load as a function of the crack length
for a sample with a given embedded length. The details of the procedure are as
follows:
(1) Estimation the value for and . In this research, the value of calculated
from ‘kink’ force was used as a reference for selecting the fitting .
(2) For a given embedded length , the interval [0, ] was divided into n
subintervals.
(3) For each of the (n + 1) dividing points, values (i = 0,1,...n)
were calculated using Eq. 4-20. The dividing point, m, in which had
highest value corresponding to a certain crack length , is selected. Then,
the pair (Fm, am) was defined.
(4) If 0 < m < n , the interval [am-1, am+1] was chosen. If m = 0 or m = n, the
intervals [0, a1] or [an-1, an] were selected respectively. Then the selected
interval was divided into subintervals again, and step (3) was repeated.
Following this method, the crack length and its corresponding
maximum force was defined.
(5) The calculation was repeated for different values, and then was
plotted as a function of for the studied system.
(6) To fit the experimental data with the theoretical values, a non-linear least-
squares method was used. The ‘best fit’ was obtained when the pair and
values that minimised the sum:
∑ [
( ) ( )]
(4-21)
The was also calculated by introducing the calculated value of and the
measured into Eq. 4-20, where the crack length was assumed to be equal to the
value that corresponds to the theoretical at the same fibre embedded length.
4.2.4 Three-point bending test (3PBT) of UD composites
When unidirectional composites are tested with the fibres in transverse direction, the
matrix and interface properties will dominate the final composite properties. The
transverse strength of the composite represents the fibre-matrix interfacial adhesion,
or the cohesion of component materials (fibre or matrix). Therefore, using
transverse three point bending test in combination with the investigation of the
Interfacial adhesion and compatibility of coir fibre composites 111
fracture surface of the tested samples, the interface quality of the composite can be
characterised. In this work, transverse and longitudinal three point bending tests of
coir UD composites were performed according to ASTM D790M. And in this way,
the transverse and shear interfacial strength can be measured approximately.
UD composites of untreated and treated coir fibres with the selected matrices were
prepared by compression moulding. Before the production of the composites,
prepregs of coir fibre and polymer films were made manually to ensure the fibres
were in good unidirectional alignment. The prepregs and extra polymer films with a
certain stacking sequence were then placed in a closed mould, after the desired fibre
volume fraction had been calculated. Processing parameters were the same when
making pull-out test samples. Figure 4-4 shows the prepreg of coir fibre
thermoplastic matrix, UD coir fibre composites samples and 3PBT on the composite
samples.
Figure 4-4. Sample preparation and testing in three-point bending test on UD coir fibre composites.
Chapter 4 112
4.3 Results and discussion
4.3.1 Surface energies and the work of adhesion
The dynamic contact angles of both untreated and alkali treated fibre in the four test
liquids are velocity-dependent, reflecting the effect of angle hysteresis as observed
and discussed in Chapter 3. By fitting the dynamic angles with the MKT method, the
static (equilibrium) contact angle can be obtained. The same fitting procedure was
applied for both untreated and treated fibres in four different liquids: water,
diiodomethane, ethylene glycol and formamide. The results of the static contact
angles following the MKT fitting are presented in Table 4-2.
In case of the matrices, the dynamic contact angle measurements provided both
stable advancing and receding angles (Figure 4-5). Hence, equilibrium contact
angles were calculated following Eq. 4-2, and the results are shown in Table 4-2
Figure 4-5. Typical dynamic contact angles of PP in water showing stable advancing and receding
contact angles.
Surface energies of the fibres and matrices are estimated and shown in Table 4-3 and
Figure 4-6. According to the Owens-Wendt approach, it can be concluded that the
untreated fibres seem to be hydrophobic with a low polar fraction of the surface
energy. On the other hand, 5% alkali treated fibres have higher surface energy with
an increased polar fraction. In similar fashion, the acid-base components of the
treated fibres are much higher than these of the untreated fibres. And both fibres
have negative charge on the surface with a higher base component.
0
10
20
30
40
50
60
70
80
0 0,5 1 1,5 2 2,5 3 3,5
Con
tact
an
gle
[°]
Position [mm]
Interfacial adhesion and compatibility of coir fibre composites 113
Figure 4-6. Surface energies of coir fibres and matrices described by the Owens-Wendt approach
and the van Oss-Good approach.
For the matrices, the surface energies of PP and MAPP are quite similar to reported
values in literature [28, 29]. A small polar fraction is found in the surface energy of
PP, possibly caused by contaminants present during film processing. It is seen that
the surface energy of MAPP is not far different from that of PP, since grafting a
small amount of maleic anhydride on PP does not affect much the wetting behaviour
and surface energy. As expected, the surface energy of PVDF is higher than that of
PP with a high polar fraction including negative and particularly also positive
charges [30].
Chapter 4 114
Table 4-2. Static/Equilibrium contact angles of untreated and alkali treated coir fibres, and matrices in water (H2O), Diiodomethane (DM), Ethylene
glycol (EG) and Formamide (FM)
Liquid
Untreated
coir
Alkali treated
coir PP PVDF MAPP
H2O 77.3 ± 0.3 70.9 ±1.1 97.7 ± 1.9 73.9 ± 1.7 85.9 ± 1.3 85.7 ± 1.1 69.6 ± 1.3 77.7 ± 0.8 100.7 ± 1.3 64.4 ± 0.9 82.9 ± 0.8
DM 48.2 ± 0.3 51.5 ± 0.5 67.8 ± 1.7 45.5 ± 2.1 57.4 ± 1.3 63.4 ± 0.7 46.2 ± 1.8 55.2 ± 0.9 77.6 ± 0.9 37.4 ± 2.6 59.7 ± 1.0
EG 47.1 ± 0.5 41.9 ± 1.2 70.9 ± 1.6 48.8 ± 2.6 60.5 ± 1.4 53.9 ± 1.2 32.7 ± 2.0 44.3 ± 1.0 85.0 ± 1.8 41.8 ± 0.9 65.4 ± 1.0
FM 47.7 ± 1.2 40.2 ± 0.7 77.2 ± 2.5 66.2 ± 1.2 71.8 ± 1.4 64.2 ± 1.9 48.6 ± 2.1 56.8 ± 1.4 94.0 ± 1.3 54.9 ± 1.7 75.3 ± 1.0
Table 4-3.Surface energies of coir fibres and matrices comprising of polar and dispersive components following the Owens-Wendt approach, and
Lifshitz – van de Waals and acid-base components following the van Oss-Good approach.
Fibre/Matrix
Surface energy (mJ/m2)
disperse-polar (Owens-Wendt) acid-base (van Oss-Good)
LWS
Untreated coir 40.4 ± 1.4 35.1 ± 1.3 5.3 ± 0.5 37.5 ± 0.2 35.5 ± 0.2 0.33 ± 0.03 3.17 ± 0.12
Alkali treated coir 42.2 ± 1.9 33.5 ± 1.7 8.7 ± 0.9 39.6 ± 0.5 34.2 ± 0.3 0.64 ± 0.09 11.27 ± 0.54
PP 30.7 ± 1.8 27.1 ± 1.6 3.6 ± 0.6 30.9 ± 0.8 30.0 ± 0.7 0.12 ± 0.07 1.87 ± 0.34
PVDF 37.2 ± 0.5 30.8 ± 0.5 6.4 ± 0.2 35.1 ± 0.6 31.6 ± 0.5 0.88 ± 0.11 3.39 ± 0.29
MAPP 28.6 ± 2.9 23.6 ± 2.6 5.0 ± 1.3 28.8 ± 0.6 28.3 ± 0.6 0.02 ± 0.02 3.15 ± 0.29
Interfacial adhesion and compatibility of coir fibre composites 115
Table 4-4. Calculated work of adhesion and interfacial energy of the composites following
both the Owens-Wendt and van Oss-Good approaches.
Composite
Owens-Wendt van Oss-Good
(mJ/m2
)
(mJ/m2
)
(mJ/m2
)
(mJ/m2
)
Untreated coir /PP 70.4 ± 1.7 0.7 68.1 ± 0.3 0.3
Treated coir /PP 71.5 ± 2.1 1.4 68.6 ± 0.5 1.9
Untreated coir /PVDF 77.4 ± 1.0 0.2 72.4 ± 0.2 0.2
Treated coir /PVDF 79.2 ± 1.3 0.2 75.0 ± 0.3 -0.3
Untreated coir /MAPP 67.9 ± 2.4 1.1 65.9 ± 0.2 0.4
Treated coir /MAPP 69.4 ± 2.9 1.4 66.0 ± 0.4 2.4
Using the results of the surface energies of the fibres and the matrices, the work of
adhesion and the interfacial energy for each composite system were calculated and
are shown in Table 4-4 and Figure 4-7. Both the Owens-Wendt and van Oss-Good
approaches provide the same tendencies for the work of adhesion and the interfacial
energy. The work of adhesion shows a higher value for coir fibre in PVDF in
comparison with that in PP and MAPP (approximately 12-14 % higher), which is
mainly thanks to the higher surface energy and polar component of PVDF. It also
can be seen that the alkali treatment somewhat improves the work of adhesion of all
fibre-matrix systems, which can be partially attributed to the higher surface energy
and polar component of the fibres. In coir fibre-PVDF, the surface energy
components of both untreated and treated fibres are quite well matched (equal and
high) leading a high work of adhesion and a low value of interfacial energy; the
compatibility is even a little better in case of the treated fibre. For PP and MAPP
systems, the improvement in work of adhesion for treated fibres is not significant
since the compatibility is relatively low, caused by mismatching surface energies
and relatively high interfacial energy.
Chapter 4 116
Figure 4-7. Work of adhesion of fibre-matrix pairs calculated following the Owens-Wendt and van
Oss-Good approaches.
4.3.2 Fibre surface chemistry
Using the XPS technique as described in Chapter 3, the surface chemistry of alkali
treated fibre is characterised and in the compared with the result on untreated fibres
presented in Chapter 3. Table 4-5 and Figure 4-8 present the surface chemical
composition of the untreated and treated fibres, consisting of relative atomic
percentages of the elements, oxygen-carbon ratio and decomposition of the C 1s
peak into four sub-peaks C1–C4. These represent: carbon solely linked to carbon or
hydrogen C–C or C–H (C1), carbon singly bound to oxygen or nitrogen C–O or C–
N (C2), carbon doubly bound to oxygen O–C–O or C=O (C3) and carbon involved
in ester or carboxylic acid functions O=C–O (C4). For untreated coir fibre, a high
proportion of C1 and low proportion of C2-C4 suggests a combination of
hydrocarbon rich waxes and lignin existing on the fibre surface as was discussed in
detail in a previous study [31]. After treatment with alkali, C1 decreases while C2-
C4 increase, which shows that more lignin as binder of elementary fibres (and
possibly also cellulose) is exposed on the surface of the treated fibres after mainly
the waxes are removed [32]. A correlation is found with the result of the wetting
measurements, where higher surface energy and polarity are determined after
removing waxes by alkali treatment.
Interfacial adhesion and compatibility of coir fibre composites 117
Table 4-5. Relative atomic percentages, O/C ratio, and decomposition of C 1s peaks of untreated and alkali treated coir fibres using XPS.
Fibres C O N Si O/C Binding energy (eV)
(%) (%) (%) (%) 284.8 ± 0.1 286.3 ± 0.1 287.5 ± 0.3 288.8 ± 0.1
C1 (%) C2 (%) C3 (%) C4 (%)
(C-C/C-H) (C-O)
(C=O/O-C-
O) (O-C=O)
Untreated coir 74.9 ± 3.3 21.8 ± 4.5 1.7 ± 0.4 0.9 ± 0.7 0.29 ± 0.07 66.2 ± 10.4 23.1 ± 5.9 6.2 ± 3.0 4.5 ± 2.4
Alkali treated coir 72.9 ± 0.3 23.2 ± 0.5 2.2 ± 0.3 1.3 ± 0.4 0.32 ± 0.01 48.0 ± 5.5 34.2 ± 5.0 11.5 ± 1.9 6.4 ± 4.1
Figure 4-8. Typical C 1s spectra comprising of the decomposition into four components C1-C4 for (a) untreated coir (b) alkali treated coir.
Chapter 4 118
4.3.3 Fibre-matrix interfacial adhesion with pull-out test
4.3.3.1 Load-displacement curves and apparent IFSS
Figure 4-9. Pull-out test sample with (a) marked point to determine fibre embedded length (b) SEM
investigation of embedded surface after pull-out.
As shown in Figure 4-9a, before performing the pull-out test a marked point was
created on the fibre at the border between the embedded segment and the free part,
which helps to determine the fibre embedded length for further calculation of the
interfacial stress. After the pull-out test, the surface of embedded fibre was
investigated by SEM images. A typical SEM image of fibre embedded surface in a
sample of untreated coir/PP is shown in Figure 4-9b, where a clean surface is
observed (without attached matrix), which represents an adhesion failure at fibre-
matrix interface.
Interfacial adhesion and compatibility of coir fibre composites 119
Figure 4-10. Typical force-displacement curves in pull-out tests of different coir fibre matrix
systems, which shows a debonding force and maximum force .
Figure 4-10 presents typical force-displacement curves recorded during pull-out
tests of coir fibres in PP, MAPP and PVDF matrices. Using the maximum recorded
force, the apparent IFSS of all systems is calculated following Eq. 4-14, and plotted
as a function of the embedded length as shown in Figure 4-11. It is obvious that the
apparent IFSS is dependent on the embedded length, which can be interpreted as
dominant brittle fracture behaviour of the interface [19, 33-35]. The IFSS of the
fibres in PVDF and MAPP decreases more rapidly than in case of PP. It seems that
the interface fracture in case of the PVDF and MAPP matrices is more brittle than in
case of PP. As mentioned, it is not possible to assess the interfacial adhesion by only
using the apparent IFSS at a certain embedded length or average values for the IFFS.
However, it can be qualitatively said that there is a higher interfacial strength of coir
fibres in PVDF and MAPP than in PP.
Chapter 4 120
Figure 4-11. Apparent interfacial shear strength at different embedded fibre lengths of untreated
coir (a) and alkali treated coir (b) fibres in PP, MAPP and PVDF.
In theses fibre-matrix systems, it is possible to detect the ‘kink’ force (debonding
force ) as shown in Figure 4-10, which are used for determination of the interface
parameter . In Table 4-6, the calculated from the ‘kink’ force of different fibre-
matrix pairs is presented. The results show the highest value of in coir/PVDF
systems and the lowest value in coir/PP pairs. It indicates that coir/PVDF and
coir/MAPP systems have a stronger interfacial strength than coir/PP systems, and
the interfacial strength of coir/PVDF is somewhat higher than that of coir/MAPP.
Considering the effect of alkali treatment, the interfacial adhesion of treated fibre in
the three matrices is 5-15% higher than that of untreated fibres.
Interfacial adhesion and compatibility of coir fibre composites 121
Table 4-6. Deponding IFSS ( ) and interfacial friction stress ( ) calculated using individual force-displacement curve and estimated from 2
parameter fitting the maximum force as a function of embedded length, and apparent IFSS from the single fibre pull-out test.
Composite
Average
apparent
IFSS
(MPa)
Calculated values (directly from force-
displacement curves) Two-parameter fit
(MPa)
(*)
(MPa)
(**)
(MPa)
(1/mm)
in range
(MPa)
in range
(MPa)
(MPa)
Untreated coir /PP 2.41 ± 0.86 8.79 ± 1.62 1.11 ± 0.49 0.55 ± 0.17 1.855 - 2.460 3.93 - 4.50 9.0 1.0
Treated coir /PP 2.85 ± 1.31 10.23 ± 1.64 1.38 ± 1.24 0.58 ± 0.34 1.972 - 2.422 3.97 - 4.51 10.4 1.0
Untreated coir /PVDF 3.30 ± 2.12 18.82 ± 3.04 1.11 ± 0.68 0.86 ± 0.45 3.160 - 4.575 5.60 - 6.53 15.8 1.0
Treated coir /PVDF 4.23 ± 2.64 19.22 ± 3.96 1.15 ± 0.72 1.96 ± 0.79 3.080 - 4.045 5.85 - 6.62 19 1.0
Untreated coir /MAPP 5.64 ± 2.67 13.86 ± 2.57 3.37 ± 1.89 1.13 ± 0.78 2.019 - 3.319 5.36 - 6.62 15.8 1.5
Treated coir /MAPP 5.58 ± 2.36 15.50 ± 2.68 3.24 ± 1.75 0.75 ± 0.27 2.211 - 3.317 5.02 - 5.89 16.8 2.0
(*) is calculated using measured with the value of calculated
(**) is calculated using the friction force of the applied force – displacement curves
Chapter 4 122
4.3.3.2 Two interfacial parameters ( and ) fitting theoretical to the
experimental data
Nayfeh parameter and thermal residual stress
For a given fibre-matrix system, the parameter is not only also dependent on the
mechanical properties of the fibre and the matrix, but also depends on sample
geometry, hence the value of is calculated for each tested sample, and presented
for every fibre-matrix systems. Accordingly, also the thermal residual stress is not a
constant value. The calculated and for each fibre-matrix pair is shown in Table
4-6. It can be seen that the of coir/PVDF and coir/MAPP are rather higher than
that of coir/PP, which also influences the value of of these systems.
Algorithm for theoretical
It is assumed that the crack growth at the fibre-matrix interface is governed by the
local shear stress. For a certain fibre embedded length, a theoretical for a
selected pair and can be obtained by generating different values of crack
length a. Figure 4-12 shows a typical curve of crack length as a function of applied
force following the equilibrium conditions as expressed by Eq. 4-20, which
illustrates that the crack starts (a > 0) at certain applied force, and stably grows
because of the interfacial friction, the applied force is continuously increasing until a
crack length, a, there the crack propagation becomes unstable. By this way, the
theoretical for various fibre embedded lengths can be determined, which used
for fitting to experimental .
Figure 4-12. Crack length as a function of external applied force in pull-out test of untreated coir/PP
system with 1mm embedded length, and is maximum applied force.
Interfacial adhesion and compatibility of coir fibre composites 123
Before carrying out the fitting, the selection of the values and is important for
a sufficient fit. The chosen value of was the same magnitude of the calculated
from the ‘kink’ force. From the same force-displacement curves of the pull-out test,
the friction force can be derived (Figure 4-10), which is used for calculation of
the friction stress. This value is then a reference for selecting in the fitting
procedure. It should be noted that could not be detected for all tested samples;
only in some curves the drop of the applied force after reaching the maximum value
was clearly observed. The calculated friction stress for different fibre-matrix
systems is shown in Table 4-6.
The best fitting of the theoretical curves to the measured values are shown in
Figure 4-13 and 4-14 with fitting parameters and , as presented in Table 4-6.
As can be seen, these theoretical curves describe well the behaviour for all
fibre-matrix systems. There is also a good correlation between the two methods of
determination of and . It can be seen that the two methods provide quite similar
results of , which are somewhat higher with the fitting method than with the direct
calculation from . Comparing the interfacial adhesion of different fibre-matrix
pairs, the results of indicate higher interfacial adhesion of coir fibres in PVDF
and MAPP than in PP, and an improvement of interfacial strength when using
treated coir fibres.
Chapter 4 124
Figure 4-13. Experimentally measured maximum force versus embedded length of untreated coir
fibres in PP (), in PVDF () and in MAPP (∎). Dotted lines represent data fitting by the theoretical
function using two fitting parameters and .
Interfacial adhesion and compatibility of coir fibre composites 125
Figure 4-14. Experimentally measured maximum force versus embedded length of alkali treated
coir fibres in PP (), in PVDF () and in MAPP (∎). Dotted lines represent data fitting by the
theoretical function using two fitting parameters and .
Chapter 4 126
4.3.4 Transverse strength and interface properties of composites
Table 4-7. Transverse bending and longitudinal tensile strengths in 3PBT on UD composites, and
efficiency factor of the longitudinal tensile strength.
Composite
Transverse
bending strength
(MPa)
Longitudinal
strength
(MPa)
Efficiency factor of
Longitudinal strength
Untreated coir /PP 3.1 ± 0.6 66.4 ± 5.8 0.59
Treated coir /PP 4.4 ± 0.7 71.5 ± 6.2 0.61
Untreated coir /PVDF 16.6 ± 2.7 82.8 ± 1.8 0.66
Treated coir /PVDF 21.5 ± 2.8 103.4 ± 2.4 0.85
Untreated coir /MAPP 21.0 ± 1.3 53.8 ± 2.3 0.72
Treated coir /MAPP 19.1 ± 1.2 49.5 ± 2.9 0.71
The transverse bending strength and the longitudinal tensile strength of UD
composites measured by 3PBT are presented in Table 4-7. The fracture surface of
the tested samples were also investigated (Figure 4-15), and show a clean surface of
fibres indicating adhesion failure at the fibre-matrix interface. Therefore the
transverse strength can be considered representative for interfacial tensile strength of
the composites. As can be seen, the higher transverse strength indicates a better
interfacial adhesion in the case of coir fibres with PVDF and MAPP as compared to
PP. There is an improvement in interfacial strength for treated fibres with PP and
PVDF in comparison with that of untreated coir, while the interfacial strength is
similar for both untreated and treated fibre with MAPP. It seems that the change of
fibre surface properties by the treatment highly affects the physical adhesion in case
of PP and PVDF, but has less influence on the chemical bonding in case of MAPP.
Figure 4-15. Typical SEM images of the facture surface of coir fibre composites in transverse
3PBT.
Interfacial adhesion and compatibility of coir fibre composites 127
The efficiency factor of the longitudinal tensile strength is also calculated and
presented in Table 4-7 and Figure 4-16. The efficiency factor is used to compare the
influence of interfacial adhesion on the composite strength. It is the ratio of
experimental longitudinal strength over the calculated value following the rule of
mixtures. It can be seen that there is a good agreement between the interfacial
adhesion and the composite strength. Higher adhesion in coir/PVDF and coir/MAPP
than that in coir/PP leads to an increase of composite strength in correspondence
with the interfacial strength (Figure 4-16).
Figure 4-16. Transverse bending strength and efficiency factor of longitudinal tensile strength of
different coir fibre composites.
4.3.5 IFSS versus transverse bending strength
In Figure 4-17, the transverse bending strength is plotted as a function of the local
debonding IFSS, which presents a good correlation between the two methods of
interfacial adhesion evaluation. A small exception is the case of MAPP, where pull-
out results shows higher interfacial adhesion of the treated fibre system than for the
untreated fibre, whereas the transverse bending strength is a bit lower. This result is
probably affected by fibre roughness caused by the treatment which is not detected
properly in transverse 3PB (tensile loading of the interface).
0,00
0,10
0,20
0,30
0,40
0,50
0,60
0,70
0,80
0,90
1,00
0
5
10
15
20
25
30
Ucoir - PP Tcoir - PP Ucoir - MAPP Tcoir - MAPP Ucoir - PVDF Tcoir - PVDF
Effi
cie
ncy
fac
tor
Tran
sve
rse
be
nd
ing
stre
ngt
h (
MP
a)
Composites
Transverse bending strength Efficiency factor Long. tensile strength
Chapter 4 128
Figure 4-17. The correlation between local IFSS from the single fibre pull-out test and transverse
bending strength from the 3PB test of UD composite for different fibre-matrix composites.
4.3.6 Work of adhesion in relation with practical adhesion
Figure 4-18. The transverse bending strength of UD composites as function of the work of adhesion
calculated following the van Oss-Good approach.
In Figure 4-18, the transverse strength of the UD composites is plotted as a function
of the calculated work of adhesion. The work of adhesion directly reflects the
significance of the surface energies of the fibre and matrix, where a higher work of
adhesion results in stronger physical (physico-chemical) interactions. The interfacial
energy, depending on the matching of surface energy components of fibre and
matrix, also has an influence on the work of adhesion, where lower interfacial
0
4
8
12
16
20
24
28
60 65 70 75 80 85 90
Tran
sver
se b
en
din
g st
ren
gth
(M
Pa)
Waab (mJ/m2)
Ucoir/PP
Tcoir/PP
Ucoir/PVDF
Tcoir/PVDF
Ucoir/MAPP
Tcoir/MAPP
Interfacial adhesion and compatibility of coir fibre composites 129
energy contributes to higher work of adhesion. Therefore, in an alternative (but
related) approach, interfacial energy should be minimised to increase the
thermodynamic stability of the interface.
When comparing the interfacial adhesion of coir fibre with PP and PVDF, it can be
seen that PVDF has higher surface energy, which components are much better
matched (equal and high) to both untreated and treated fibre surface energies than in
the case of PP. This results in a significantly higher interfacial adhesion for the coir
fibre PVDF composites compared to the PP composites (Figure 4-18). Since there is
likely no chemical bonding involved in the interface interactions, physical adhesion
is proposed as the main interaction leading to the difference in interfacial adhesion
for these two composite systems. The modification of fibre surface energy by alkali
treatment contributes to a further improvement of adhesion by increasing the fibre
surface energy and minimising the differences between the surface energy
components of the fibre and the PVDF matrix.
When comparing PP and MAPP, the surface energies are not so different, which
results in similar work of adhesion of coir fibre – PP and coir fibre - MAPP. Thus,
the physical adhesion is suggested to be comparable in the two systems. According
to the results of the transverse strength and IFSS measurements, the interfacial
adhesion of coir fibre with MAPP is approximately five times higher than that of
coir fibre with PP, which may be attributed to covalent bonds between maleic
anhydride groups of MAPP and hydroxyl groups of lignin on the fibre surface.
Summarising, the fibre-matrix adhesion can be improved by increasing the work of
adhesion using a high surface energy and a compatible matrix as in case of PVDF,
where the physical adhesion plays the main role. Alternatively, modification of the
matrix, offering chemical interaction across the interface can be used to obtain better
interfacial adhesion, which in this case is mainly dominated by covalent bonds.
4.4 Conclusions
Wetting analysis consisting of contact angle measurements and fibre surface energy
estimations was conducted to predict the composite interfacial compatibility and
adhesion by determining the work of (physical) adhesion and the interfacial energy.
Using the equilibrium contact angles, the determination of surface energies and
work of adhesion follows the equilibrium thermodynamic conditions of a static
wetting situation, providing reliable results for studying fibre-matrix interactions.
The results of the characterization of the fibre surface chemistry using XPS were
consistent with these of the wetting measurements. The combination of these
Chapter 4 130
techniques offers a deeper understanding of the fibre surface, which assists in the
selection of fibre treatments or matrix modifications to improve the interfacial
adhesion and compatibility.
Practical adhesion in single fibre composites and UD composites was evaluated
using single fibre pull-out tests and transverse three-point bending tests. The results
of IFSS from the pull-out tests showed a brittle fracture behaviour of the coir fibre-
matrix interface, because the apparent IFSS was dependent on the embedded fibre
length. For the comparison of the fibre-matrix interfacial adhesion, the local
debonding IFSS was calculated from the ‘kink’ force in the pull-out force-
displacement curves and secondly by fitting experimental data of the maximum
force using models developed by Zhandarov. Both methods gave very similar
results. The interfacial adhesion in tension mode of the composites was directly
examined by transverse 3-point bending (T3PB), which provided a good correlation
between these interfacial strength and the results of the local IFSS from the pull-out
test. Both results of IFSS and transverse strength showed the influence of the
interfacial adhesion on the composite strength, which was also reflected in the
longitudinal strength efficiency factor.
In thermoplastic composites, adhesion is often dominated by the surface energy
forces, because no covalent bonds can be formed by the molten polymer at the
interface. In this way PVDF was identified as an interesting matrix for coir fibres.
The strongly improved interface strength in coir/PVDF, as compared to coir/PP, can
indeed be attributed to an increase in work of adhesion. This is further improved by
an alkali treatment of the fibres, which increases the fibre polarity. When MAPP is
used as a matrix, the mechanism is different; the analysis shows that no
improvement in physical adhesion may be expected compared to PP, but the strong
increase in interface strength must be attributed to chemical adhesion due to
activation of the anhydride groups at the processing temperature.
In this study, there has been a good agreement between the results of the wetting
analysis and those of the composite interface mechanical tests. The combination of
different characterisations, from wetting analysis and fibre surface characterization
to practical fibre-matrix adhesion measurements, has offered a deeper understanding
of the interfacial adhesion and compatibility in coir fibre composites.
Interfacial adhesion and compatibility of coir fibre composites 131
References
1. Zhandarov, S. and E. Mader, Characterization of fiber/matrix interface strength:
applicability of different tests, approaches and parameters. Composites Science and
Technology, 2005. 65(1): p. 149-160.
2. Nardin, M. and J. Schultz, Relationship between fiber–matrix adhesion and the
interfacial shear strength in polymer-based composites. Compos Interfaces, 1993.
1: p. 172-192.
3. Pisanova, E., S. Zhandarov, and E. Mader, How can adhesion be determined from
micromechanical tests? Composites Part a-Applied Science and Manufacturing,
2001. 32(3-4): p. 425-434.
4. Kim, J.K., Engineered Interfaces in Fiber Reinforced Composites. 1998, Elsevier
Science.
5. Berg, J.C., Wettability. 1993, New York: Marcel Dekker.
6. Blake, T.D. and Blake, Kinetics of liquidliquid displacement. Journal of Colloid and
Interface Science, 1969. 30(3): p. 421.
7. De Jonghe, V., et al., Contact angle hysteresis due to roughness in four
metal/sapphire systems. Journal de chimie physique, 1990. 87(9): p. 1623-1645.
8. Andrieu, C., C. Sykes, and F. Brochard, Average spreading parameter on
heterogeneous surfaces. Langmuir, 1994. 10(7): p. 2077-2080.
9. Dettre, R.H., Wetting of low-energy surfaces. Wettability, 1993. 49: p. 1.
10. Owens, D.K., Estimation of the surface free energy of polymers. Journal of Applied
Polymer Science, 1969. 13(8): p. 1741.
11. Van Oss, C.J. and Vanoss, The role of van der Waals forces and hydrogen bonds in
"hydrophobic interactions" between biopolymers and low energy surfaces. Journal
of Colloid and Interface Science, 1986. 111(2): p. 378.
12. Van Oss, C.J., Interfacial Lifshitz-van der Waals and polar interactions in
macroscopic systems. Chemical reviews, 1988. 88(6): p. 927.
13. Young, T., Phil. Trans. R. Soc. , 1805(95): p. 65-87.
14. Berthelot, D., Sur le mélange des gaz. Comptes rendus hebdomadaires des séances
de l'Académie des Sciences, 1898. 126 (1703–1706): p. 1857–1858.
15. Fowkes, F.M., Attractive forces at interfaces. Industrial & Engineering Chemistry,
1964. 56(12): p. 40-52.
16. Fowkes, F.M., Donor-Acceptor Interactions at Interfaces. The Journal of Adhesion,
1972. 4(2): p. 155-159.
17. Nardin, M., et al., Study of the carbon fiber–Poly(Ether–Ether–Ketone) (PEEK)
interfaces, 4: influence of fiber–matrix adhesion on the mechanical properties of
unidirectional composites. Polymers for Advanced Technologies, 1991. 2(4): p.
171-176.
18. Connor, M. and Connor, A criterion for optimum adhesion applied to fibre
reinforced composites. Journal of materials science, 1997. 32(19): p. 5059.
19. Hampe, A. and C. Marotzke, Adhesion of polymers to reinforcing fibres. Polymer
International, 1992. 28(4): p. 313-318.
20. Zhandarov, S. and E. Mader, Indirect estimation of fiber/polymer bond strength and
interfacial friction from maximum load values recorded in the microbond and pull-
out tests. Part II: Critical energy release rate. Journal of Adhesion Science and
Technology, 2003. 17(7): p. 967-980.
Chapter 4 132
21. Zhandarov, S., E. Pisanova, and E. Mader, Is there any contradiction between the
stress and energy failure criteria in micromechanical tests? Part II. Crack
propagation: Effect of friction on force-displacement curves. Composite Interfaces,
2000. 7(3): p. 149-175.
22. Zhandarov, S.F. and E.V. Pisanova, The local bond strength and its determination
by fragmentation and pull-out tests. Composites Science and Technology, 1997.
57(8): p. 957-964.
23. Nayfeh, A.H., Thermo-Mechanically Induced Interfacial Stresses in Fibrous
Composites. Fibre Science & Technology, 1977. 10(3): p. 195-209.
24. Liu, C.H. and J.A. Nairn, Analytical and experimental methods for a fracture
mechanics interpretation of the microbond test including the effects of friction and
thermal stresses. International journal of adhesion and adhesives, 1999. 19(1): p.
59-70.
25. Nairn, J.A., Analytical fracture mechanics analysis of the pull-out test including the
effects of friction and thermal stresses. Advanced Composites Letters, 2000. 9(6): p.
373-383.
26. Zhandarov, S.F., E. Mader, and O.R. Yurkevich, Indirect estimation of
fiber/polymer bond strength and interfacial friction from maximum load values
recorded in the microbond and pull-out tests. Part 1: local bond strength. Journal of
Adhesion Science and Technology, 2002. 16(9): p. 1171-1200.
27. Zhandarov, S., E. Pisanova, and K. Schneider, Fiber-stretching test: a new
technique for characterizing the fiber-matrix interface using direct observation of
crack initiation and propagation. Journal of Adhesion Science and Technology,
2000. 14(3): p. 381-398.
28. Troger, J., et al., Determination of the surface tension of microporous membranes
using wetting kinetics measurements. Colloids and surfaces A: Physicochemical and
engineering aspects, 1998. 134(3): p. 299-304.
29. Aranberri-Askargorta, I., T. Lampke, and A. Bismarck, Wetting behavior of flax
fibers as reinforcement for polypropylene. Journal of Colloid and Interface Science,
2003. 263(2): p. 580-589.
30. Lee, S., J.S. Park, and T.R. Lee, The Wettability of Fluoropolymer Surfaces:
Influence of Surface Dipoles. Langmuir, 2008. 24(9): p. 4817-4826.
31. Tran, L.Q.N., et al., Wetting analysis and surface characterisation of coir fibres
used as reinforcement for composites. Colloids and Surfaces a-Physicochemical and
Engineering Aspects, 2011. 377(1-3): p. 251-260.
32. Van Dam, J.E.G. and D. van, Process for production of high density/high
performance binderless boards from whole coconut husk: Part 1: Lignin as intrinsic
thermosetting binder resin. Industrial crops and products, 2004. 19(3): p. 207.
33. Marotzke, C., The elastic stress field arising in the single-fiber pull-out test.
Composites Science and Technology, 1994. 50(3): p. 393-405.
34. Piggott, M. and Y. Xiong, Visualization of debonding of fully and partially
embedded glass fibres in epoxy resins. Composites Science and Technology, 1994.
52(4): p. 535-540.
35. Marotzke, C. and L. Qiao, Interfacial crack propagation arising in single-fiber pull-
out tests. Composites Science and Technology, 1997. 57(8): p. 887-897.
Mechanical properties of UD coir fibre composites 133
Chapter 5
Mechanical properties
of unidirectional coir fibre composites
Chapter 5 134
5.1 Introduction
In Chapter 2, the study of the mechanical properties of coir fibres showed that the
fibres have a low stiffness, around 4.6 GPa, and a low strength of approximately 230
MPa, but show a high elongation to failure, in the range from 20 to 40%. With these
properties, coir fibres are expected to be able to toughen a polymer matrix when
used in composites.
In this chapter, the mechanical properties of UD coir fibre composites with both
thermoplastic and thermoset matrices are assessed by tensile tests in the fibre
direction, flexural tests and unnotched Izod impact tests. The toughening effect of
coir fibres on a brittle matrix is further investigated by studying the correlation
between the fibre volume fraction and the impact strength of composites. Besides,
an initial study on coir/bamboo fibre hybrid composites is carried out to investigate
the hybrid effect of tough coir fibres and brittle bamboo fibres in composites.
5.2 Materials and methods
5.2.1 Materials
Coir fibres
Untreated coir fibres and coir fibres treated with 5% alkali were used in this study.
The procedure for preparation of the fibres was described in Chapter 4. The tensile
mechanical properties of untreated coir fibres are shown in Table. 5-1. The
mechanical properties of the treated fibres were assumed to be equal to the values of
untreated fibres, since the applied alkali treatment was light (5% NaOH for 2h at
20oC) which would mainly change the fibre surface properties but not much the
mechanical properties of the fibres.
Bamboo fibres
Technical bamboo fibres of the species Guadua angustifolia were extracted from
bamboo culms using a novel, purely mechanical extraction process developed by
KU Leuven, giving a maximum fibre length between 20 and 35 cm. The bamboo
fibres were obtained from well defined locations in Colombia. There was not any
retting or chemical treatment applied before extracting the fibres [1].
Mechanical properties of UD coir fibre composites 135
Matrices
As shown in Chapter 4, thermoplastic matrices were supplied as films. The
Polypropylene (PP) used was an unmodified grade supplied by Propex GmbH
(Germany). Polyvinylidene fluoride (PVDF Solef 1008) was provided by Solvay
(Belgium), and 0.3% maleic anhydride grafted polypropylene (MAPP Bynel 50) was
supplied by Dupont (Switzerland). Regarding thermosets, epoxy resin EPIKOTE
828 LV was selected, with Dytek DCH-99 as hardener. The curing procedure for the
epoxy was 70°C for 1h and subsequently at 150°C for another hour. The mechanical
properties of the matrices are shown in Table 5-1.
Table 5-1. Mechanical properties of studied fibres and matrices.
5.2.2 Production of composite samples
5.2.2.1 Alignment of coir fibres in UD fibre layer
Like most natural fibres, the extracted coir fibres are delivered in a bundle and
slightly twisted. To make good UD (uni-directional) composites, it is required that
the fibres are properly aligned in one direction. In this work, a procedure for fibre
alignment was developed, where the coir fibres were soaked in water, then combed
and evenly spread in a thin layer of UD fibres (with thickness of 2-4 technical
fibres). This wet layer was placed between two plastic plates to keep the UD form of
the fibre layer, during drying at 70 oC for 3 days in an oven. After drying, the UD
fibre layers were used for making prepregs with thermoplastics, or directly placed in
a mould for producing thermoset composites. Figure 5-1 shows a schematic
presentation of the various steps from fibre alignment to production of the UD
composites.
Material
E-Modulus
(GPa)
Strength
(MPa)
Strain to failure
(%)
Density
(g/cm³)
Reference
Coir fibre 4.6 ± 1.1 234.2 ± 57.4 18.0 – 36.7 0.9-1.3 testing
Bamboo 42-46 775-860 1.7-1.9 1.4 [1]
PP 1.6-1.8 55-65 > 300 0.9 Flexural test
MAPP 0.8-1.2 22-30 > 300 0.89 Flexural test
PVDF 2.2-2.6 78 5-10 1.78 Data sheet
Epoxy 2.73 70.1 4.1 1.16 [2]
Chapter 5 136
Figure 5-1. Schematic presentation of making UD coir fibre composites with thermoplastic matrix
(d) or thermoset matrix (e), using layers of well aligned fibres.
5.2.2.2 UD coir fibre thermoplastic composites by compression moulding
Prepregs
UD coir fibre composites with PP, MAPP and PVDF matrices were produced using
prepregs. To make a prepreg, a UD coir fibre layer was placed in a sandwich, where
the fibres were clamped in between thermoplastic films, as seen in Figure 5-2. An
iron was used to apply the temperature (around 200 oC) and pressure to consolidate
the fibre and the matrix to form a prepreg.
Figure 5-2. Making prepreg of UD coir fibre thermoplastics.
Mechanical properties of UD coir fibre composites 137
Composite processing
There are three types of UD composite samples which were made, for flexural
testing, tensile testing and Izod impact tests. The moulds used for making these
samples are shown in Figure 5-3. Tensile samples were made of 15 mm x 250 mm x
2 mm (width x length x thickness) following ASTM 3039. Impact samples were 10
mm x 80 mm x 4 mm according to ISO 179. For the flexural samples, composite
plates of 50 mm x 100 mm x 2 mm were first prepared, and then samples of 12.5
mm x 50 mm x 2 mm were cut from the plates, following ASTM 790M.
Figure 5-3. Moulds for making thermoplastic composite samples for (a) flexural, (b) tensile and (c)
impact tests
For composites processing, prepregs were cut into the desired dimensions fitting the
moulds, and then filled into the moulds in designated stacking sequences. The
thickness of the samples was controlled by placing aluminium stoppers at both edges
of the mould channels between the upper and lower mould. For tensile and impact
samples, six samples of each type could be produced at one time using six channels
in the moulds. The fibre volume fraction of the composite samples was estimated by
the weight of the fibres and the matrix films.
The closed mould set-ups were then placed into the Pinette hot press (Figure 5-4) for
composites fabrication, under processing parameters of 175 oC for PP and 185
oC for
PVDF and MAPP respectively, at 10 bar pressure and for 15 minutes, after that the
mould was cooled down at cooling rate of 5 oC/min until 80
oC, and then moved to a
cold press at room temperature under the same pressure for a faster cooling until
room temperature.
Chapter 5 138
Figure 5-4. Pinette hot press for the production of thermoplastic composites.
5.2.2.3 UD coir fibre epoxy composites with vacuum assisted resin infusion
(VARI)
With a thermoset resin such as epoxy, UD composites were produced using the
VARI technique. Figures 5-1.e and 5-5 show the set up for producing coir fibre
epoxy composites using the VARI technique.
Firstly, UD layers of coir fibres, which were obtained from the above described
alignment process, were laid up as a dry stacked laminate and fixed on an
aluminium (bottom mould) plate by adhesive paper tape. An upper mould was then
put on top of the fibre laminate; and the thickness of the composite was controlled
by placing stoppers in the cavity between the upper and bottom moulds. Two tubes
for resin inlet and resin outlet (vacuum tube) were positioned at two ends of the fibre
layers. To assist the flow of resin during the infusion process, breathers were used
around the fibres. A peel ply was placed on top of the upper mould to avoid damage
to the covering vacuum bag by the mould. The whole set up was then covered by a
vacuum bag which was attached to the bottom plate by a sealant tape.
Mechanical properties of UD coir fibre composites 139
The fibre impregnation process was started by applying vacuum via the vacuum
tube, while the resin inlet was closed. When a suitable vacuum level was reached,
the resin inlet was opened to let the resin flow and wet out the fibres. A curing
process was carried out at 70 oC for 1h and then at 150
oC for another hour.
5.2.3 Test methods
5.2.3.1 Flexural 3PBT
Three point bending tests (3PBT) in longitudinal and transverse fibre direction were
carried out for both untreated and alkali treated coir fibres in thermoplastic matrices
(PP, MAPP, PVDF) and in epoxy, following the ASTM 790M standard. Test
samples were prepared with the dimensions of 12.5 mm x 50 mm x 2 mm (width x
length x thickness). The tests were performed on an Instron universal testing
machine with a load cell of 1 KN, crosshead speed of 0.85 mm/min and span length
of 32 mm, to guarantee loading in pure bending (span/thickness > 16) (Figure 5-6).
Figure 5-5. VARI technique for production of coir epoxy composites
Chapter 5 140
Figure 5-6. Three-point bending test set-up
5.2.3.2 Tensile test
Tensile tests were performed according to the standard ASTM D3039, on composite
samples of 15 mm x 200 mm x 2 mm, to which composite endtabs were glued. A
load cell of 5 kN was used and a crosshead speed of 1 mm/min was applied. The
gauge length between the two clamps was set at 100 mm, while an extensometer
with gauge length of 50 mm was employed for measuring the sample strain. Figure
5-7 shows the set up for the tensile test and some tested samples.
Various composite systems of coir fibre including untreated coir in PP, PVDF and
epoxy were characterised. Six samples for each type of composite were tested.
Figure 5-7. Tensile test and test samples
Mechanical properties of UD coir fibre composites 141
5.2.3.3 Izod impact test
Following the ISO 179 standard, the Izod impact test was performed on unnotched
samples of 10 mm x 80 mm x 4 mm. UD composites of untreated coir fibre with PP
and epoxy were studied, in which the effect of fibre volume fraction on impact
properties was investigated for the coir/epoxy system.
Figure 5-8 displays the impact test machine and set up for Izod impact tests on UD
coir fibre composites.
Figure 5-8. Impact test machine and schematic of Izod impact test on UD composite sample.
5.2.4 Determination of coir fibre volume fraction
As reported in Chapter 2, coir fibres have a high porosity of approximately 30%. To
determine correctly the fibre volume fraction in their composites, it is necessary to
distinguish between fibre volume fraction, , and fibre solid fraction, (which is
the volume fraction of the solid material of fibre in the composite). In this study, the
fibre volume fraction of composites was calculated from the weight of the used
fibres and the fibre density. Regarding the fibre density, as shown in Chapter 2, the
Chapter 5 142
value of 1.3 g/cm3 refers to the density of the solid material of the fibres, and the
value of 0.9 g/cm3 is the overall density of the long fibres (porous fibres). Therefore,
depending on which value of density is used, the or will be determined. In a
general way, the relation between and can be described as
(5-1)
5.2.5 Coir/bamboo hybrid composites
From Chapter 2, it is known that coir fibres are less stiff and strong, but potentially
tough fibres. To explore further the potential uses of coir fibres in composites, the
concept of combination (hybridisation) of coir fibres with other fibres (e.g. stiff and
strong, but brittle fibres) is considered. In this preliminary study, bamboo fibres were
selected for making UD coir/bamboo hybrid composites, due to their strong and stiff
but brittle properties [1].
In this study, the hybrid effect in coir/bamboo polypropylene composites was
characterised at the macro level, where fibres are mixed at fibre layer scale. However,
thin coir and bamboo prepregs (thickness of 1-3 technical fibres) were used for
making the hybrid composite samples with the intention of approaching a good
mixing at single fibre level, which is considered as hybridisation at micro scale;
theoretical studies [3-5] predict a better stress transfer in hybrid composites when the
fibres are mixed at micro level.
The hybrid samples were prepared by stacking coir/PP and bamboo/PP prepregs in a
sequence of 2 layers of coir/PP prepreg at the outside and 1 layer of bamboo/PP
prepreg in the middle. The compression moulding technique was used for production
of the composites with processing parameters of 175 oC and 10 bar pressure (the
same as for coir/PP).
To characterise the properties of the hybrid composites, tensile tests were performed
with the same test set up as used for the mono composites, as described above.
Mechanical properties of UD coir fibre composites 143
5.3 Results and discussion
5.3.1 Flexural properties of UD composites
5.3.1.1 Longitudinal properties
Figure 5-9. Typical stress-strain curves in longitudinal 3PBT.
Figure 5-9 shows typical stress-strain curves obtained from 3PBT on UD coir fibre
composites with four different matrices (PP, PVDF, MAPP and epoxy). It can be
seen that the coir/epoxy composite is rather brittle compared with the
coir/thermoplastic systems. The flexural properties in longitudinal fibre direction
including E-modulus, strength and strain at failure of the composites are also
presented in Table 5-2. Because the fibre volume fractions of the composites are
different, an efficiency factor and normalised values (to the same fibre volume
fraction) of E-modulus and strength will be used to have a good comparison
between different systems.
0
20
40
60
80
100
120
0 2 4 6 8 10
Stre
ss (
Mp
a)
Strain (%)
Ucoir/PP
Ucoir/PVDF
Ucoir/MAPP
Ucoir/epoxy
Chapter 5 144
Table 5-2. Flexural properties of untreated and alkali treated coir fibre composites; theoretical values calculated following rule of mixtures
Composites
(%)
(%)
Longitudinal
E-Modulus
(GPa)
Longitudinal
Strength
(MPa)
Strain at failure*
(%)
(GPa)
(MPa)
Untreated coir /PP 46 32.2 2.02 ± 0.20 66.35 ± 5.84 5.0 ± 0.2 2.74 112.9
Treated coir /PP 50 35 2.05 ± 0.17 71.48 ± 6.19 4.8 ± 0.2 2.83 117.5
Untreated coir /PVDF 48 33.6 2.34 ± 0.36 82.76 ± 1.75 6.7 ± 0.4 3.20 124.6
Treated coir /PVDF 45 31.5 2.80 ± 0.40 103.37 ± 2.35 7.3 ± 0.6 3.16 121.7
Untreated coir /MAPP 33 23.1 1.53 ± 0.12 53.83 ± 2.28 6.7 ± 0.2 1.94 74.5
Treated coir /MAPP 30 21 1.47 ± 0.17 49.46 ± 2.88 6.6 ± 0.4 1.86 70.0
Untreated coir /Epoxy 56 39.2 2.61 ± 0.28 86.94 ± 9.60 3.7 ± 0.5 3.40 105.0
Treated coir /Epoxy 64 44.8 2.83 ± 0.43 105.44 ± 9.43 4.3 ± 0.6 3.48 117.3
(*) For ductile matrix (thermoplastic) composites, strain at failure is referred to as strain at maximum stress.
Mechanical properties of UD coir fibre composites 145
Efficiency factor based on rule of mixtures
The efficiency factor is defined as the ratio of the actual mechanical property and the
theoretical property. The theoretical value can be calculated using the rule of
mixtures as follows
The theoretical E-modulus of a composite, , is calculated according to the rule
of mixtures as shown in Eq. 5-2
(5-2)
where and are the volume fractions of fibre and matrix respectively; and
are the E-modulus of fibre and matrix respectively.
For porous coir fibre and the modulus of the fibre is only calculated
for the solid material, hence is used; then Eq. 5-2 becomes
(5-3)
, so
(5-4)
Theoretical strength of coir thermoplastic composites
In the coir fibre composites with PP and MAPP and, the fibre failure strain is lower
than the matrix failure strain. So, the strength of the fibres will determine the failure
of the composites; hence the estimation of theoretical strength can be calculated as
(5-5)
where is the fibre strength (calculated only for the fibre solid material), and
is the matrix stress at fibre failure strength.
In this case, the strain at maximum stress of the composites is set as the fibre failure
strain (approximately 30%)
(5-6)
Chapter 5 146
Theoretical strength of coir/epoxy composites
In case of the coir/epoxy and coir/PVDF systems, the failure strain of the matrices is
lower than that of the coir fibre. When the composite is loaded beyond matrix
failure, the strength of the fibres is dominant for carrying the applied load after
matrix cracking (with high fibre volume fraction). For this system, the theoretical
strength is calculated following Eq. 5-7.
(5-7)
And the strain at maximum stress of the composites is also assigned to the fibre
failure strain as described in Eq. 5-6.
Table 5-3. Efficiency factor of flexural E-modulus, strength and strain at failure for coir fibre
composites; and composite E-modulus and strength normalised to Vf = 50% using the efficiency
factors.
Composite
Efficiency
factor of
E-modulus
Efficiency
factor of
Strength
Efficiency
factor of
Strain at
failure
Normalised
E-Modulus
(Vf = 50%)
Normalised
Strength
(Vf = 50%)
Untreated coir /PP 0.74 0.59 0.17 2.08 69.05
Treated coir /PP 0.73 0.61 0.16 2.05 71.48
Untreated coir /PVDF 0.73 0.66 0.22 2.35 84.05
Treated coir /PVDF 0.88 0.85 0.24 2.85 107.50
Untreated coir /MAPP 0.79 0.72 0.22 1.91 72.26
Treated coir /MAPP 0.79 0.71 0.22 1.92 70.65
Untreated coir /Epoxy 0.78 0.89 0.12 2.56 77.63
Treated coir /Epoxy 0.82 0.94 0.14 2.70 82.34
Table 5-3 and Figure 5-10 present the efficiency factors of longitudinal E-modulus
and strength and failure strain for the composite systems. The efficiency factor of
the E-modulus is a measure for quality of the fibre alignment and fibre-matrix
wetting in the samples. From the results, it is seen that the coir/PP systems have
lower values compared to the other systems, which can be explained by inefficient
fibre stress transfer at the fibre-matrix interface due to incomplete wetting (as
reported in Chapter 4). Comparing untreated and treated fibre composites, the
efficiency factors are similar in case of PP and MAPP composites, while there is an
improvement by fibre treatment in PVDF and epoxy systems. As seen in Chapter 4,
Mechanical properties of UD coir fibre composites 147
a higher fibre-matrix compatibility between treated fibre and PVDF was obtained
when the polarity of the fibre surface increased by removing a waxy layer on the
surface with alkali treatment. The same behaviour was also found in the treated
fibre/epoxy system [6].
Figure 5-10. Efficiency factor of longitudinal E-modulus, strength and strain at failure of the
composites.
The efficiency factor of the strength is an indicator for the quality of the fibre/matrix
adhesion. It can be seen that the strength efficiency factor of coir fibre thermoplastic
composites are quite low compared to that of coir/epoxy composites. The poor
interfacial adhesion of the coir fibre thermoplastic composites leads to an inefficient
stress transfer at the interface, which may result in an early failure of the composites.
Among the four matrices evaluated, the efficiency factor of strength for the coir/PP
system is quite low because of bad interfacial adhesion which was proved in the
investigation of the composite interface in Chapter 4.
The efficiency factor of failure strain is very low for all composites due to the
premature failure of the composites (or the limitation of strain measurement in
3PBT) . The result shows that the coir/epoxy systems have much lower values
compared to coir/PVDF and coir/MAPP composites (except coir/PP systems, which
have very poor interfacial adhesion). The higher values in coir/PVDF and
coir/MAPP likely thank to a high failure strain of matrices.
0,74 0,73 0,79 0,79
0,73
0,88
0,78 0,82
0,59 0,61 0,72 0,71 0,66
0,85 0,89 0,94
0,17 0,16 0,22 0,22 0,22 0,24
0,12 0,14
0,0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
1,0
Effi
cie
ncy
fac
tor
Composites
Efficiency factor of E-Modulus Efficiency factor of Strength Efficiency factor of failure strain
Chapter 5 148
Using the efficiency factors, the experimental values of E-modulus and strength of
composites with different fibre volume fraction are scaled to normalised values at a
same fibre volume fraction of 50%, as shown in Figure 5-11 and Table 5-3. The
results indicate that coir/PVDF and coir/epoxy are stiffer compared to coir/PP and
coir/MAPP systems, which is correlated to the stiffness of the matrices. For the
composite strength, coir/PVDF composites give the strongest systems. The effect of
fibre treatment can also be observed with the improvement of strength in treated
fibre systems with PVDF and to a lesser extent epoxy matrix.
Figure 5-11. Normalised E-modulus and strength at a fibre volume fraction of 50%.
5.3.1.2 Transverse properties
Table 5-4 presents the transverse flexural properties of UD untreated and alkali
treated coir fibre composites.
Transverse E-modulus
It can be seen that the transverse modulus of all systems is somewhat lower than
their matrix modulus. Like other natural fibres, coir fibre has an anisotropic structure
with cellulose chains in elementary fibres oriented along the fibre axis, which results
in high longitudinal modulus and lower transverse modulus. In literature, the
69,1 71,5 72,3 70,7
84,1
107,5
77,6 82,4
2,1 2,1 1,9 1,9
2,4
2,9 2,6
2,7
0,0
0,5
1,0
1,5
2,0
2,5
3,0
3,5
4,0
4,5
5,0
0
20
40
60
80
100
120
E-m
od
ulu
s (G
Pa)
Stre
ngt
h (
MP
a)
Composites
Normalised Strength Normalised E-Modulus
Mechanical properties of UD coir fibre composites 149
transverse modulus of coir fibre could not be found. Transverse properties of jute
fibre were presented by Cichoki Jr et al. [7]. They reported that the longitudinal fibre
modulus of jute fibres is approximately seven times higher than the transverse
modulus. Assuming the same factor applies to coir fibre1, the transverse modulus of
coir fibre can be estimated at around 0.7 GPa. This value was used for estimation of
the theoretical transverse modulus of the composite, following the rule of mixtures
(Eq. 5-9). The theoretical values obtained are shown in Table 5-4.
The rule of mixtures for the transverse modulus of composites is given by
considering an in series loading of the components:
For the coir/epoxy systems, it can be seen that the theoretical modulus of the
composites is lower than the experimental value (Table 5-4). It is probable that the
transverse modulus of the fibre is underestimated, also considering the high
microfibril angle of coir fibre. To have a precise analysis, the fibre transverse
modulus needs to be determined.
When comparing the transverse modulus between untreated and treated fibre
composites, it is obvious that the value of the treated fibre composite is typically
higher than that of the untreated fibre system with the same matrix. This can be
explained by the presence of a weak layer of waxes on the fibre surface, at the fibre-
matrix interface in case of the untreated fibre composites, which leads to a lower
composite modulus. Another reason maybe debonding that occurs at the interface
even at very low stresses, in the region where E-modulus is normally measured.
1 Coir fibres have much larger microfibril angle than jute fibres, the transverse modulus should be higher.
However, the presence of lumens and lacuna also plays a big role in the lower transverse modulus.
Chapter 5 150
Table 5-4. Transverse properties of the composites
Composite
Transverse
E-Modulus
(GPa)
Theoretical
Transverse
E-Modulus
(GPa)
Transverse
Strength
(MPa)
Untreated coir /PP 0.88 ± 0.21 1.32 3.1 ± 0.6
Treated coir /PP 0.94 ± 0.08 1.29 4.4 ± 0.7
Untreated coir /PVDF 1.22 ± 0.14 1.47 16.6 ± 2.7
Treated coir /PVDF 1.71 ± 0.24 1.51 21.5 ± 2.7
Untreated coir /MAPP 0.73 ± 0.07 1.00 21.0 ± 1.3
Treated coir /MAPP 0.84 ± 0.04 1.00 19.1 ± 1.2
Untreated coir /Epoxy 1.83 ± 0.14 1.39 20.4 ± 2.5
Treated coir /Epoxy 2.09 ± 0.10 1.30 19.5 ± 1.6
Transverse strength
The transverse strength of the composites is displayed in Table 5-4 and Figure 5-12.
When the unidirectional composites are tested with the fibres in transverse direction,
the matrix and interface properties will dominate the final composite properties.
Therefore, the interface quality of the composite can be characterised.
The results show low values of transverse strength in coir/PP systems, which are
situated much below the value of the neat matrix. From the SEM image of the
fracture surface (Figure 5-12), it is observed that the failure occurred at the fibre-
matrix interface, which shows clearly imprinted fibre channels in the matrix.
In coir/MAPP and coir/PVDF systems, a higher transverse strength was measured,
and also the observation of the fracture surface suggested that the failure was at the
interface. As discussed in Chapter 4, the higher interfacial physical adhesion and
fibre-matrix compatibility in coir/PVDF composites is attributing to the increase of
their interface strength, and chemical adhesion leads to a high interface strength in
case of coir/MAPP composites.
On the other hand, the SEM image of the fracture surface in the coir/epoxy
composites (Figure 5-12) shows fibre breakage and a clean surface of brittle epoxy.
The measured strength of this system was around 20 MPa, which is lower than the
epoxy matrix strength. These observations suggest that perhaps stress concentrations
Mechanical properties of UD coir fibre composites 151
at the interface lead to failure of the interface and the brittle epoxy, and then crack
propagation breaks the fibres.
Figure 5-12. Fracture surface of 3PB samples in transverse fibre direction, and the transverse
strength of the composites.
5.3.2 Tensile properties of UD composites
Besides the investigation of the UD composites by flexural testing, some composite
systems were examined in tensile loading. Figure 5-13 presents typical tensile stress-
strain curves of untreated coir/PP, coir/PVDF and coir/epoxy UD composites. It can
be seen that the coir/epoxy fails at low strength and strain, even though it has a
higher fibre volume fraction compared to coir/PP and coir/PVDF composites. For
the composites with PP and PVDF, the strain at failure is around 3.5%, which is still
Chapter 5 152
lower than the expected value (given the high strain to failure of the fibres (20-40%)
and the higher strain to failure of the matrices (5-10% for PVDF, much higher for
PP).
Figure 5-13. Typical tensile stress-strain curves of untreated coir fibre in PP, PVDF and epoxy
matrices.
The tensile properties of the composites are summarised in Table 5-5. The efficiency
factors for the E-modulus, strength and strain at failure are also calculated, in which
the theoretical values of the modulus, strength and failure strain are determined
following the rule of mixtures. The results show that the strength efficiency factors
in tension are much lower than those in 3PBT, which indicates a premature failure
of the composites during tensile testing. From the typical fracture surfaces of tested
samples shown in Figure 5-14, it is obvious that the failure of the coir/PP and
coir/PVDF composites include fibre pulled out of the matrix, whereas it seems to go
across the sample in the case of coir/epoxy composites.
Moreover, from SEM images of the fracture surface, it is observed that the
composite fails adhesively at the fibre-matrix interface in case of coir/PP and
coir/PVDF (Figure 5-14a), in which a lots of fibres are pulled out of the matrix and
the imprint of the fibre remains on the matrix. It is probably that the fibres are not
perfectly aligned in the composites. Under tensile loading, debonding occurs at the
interface due to fibre reorientation while the matrix is plastically deforming. When
the interface fails, the fibres and the matrix individually carry the load which causes
the system to fail at low strength (strength efficiency around 40%).
Mechanical properties of UD coir fibre composites 153
In case of coir/epoxy composites, the fracture surface of the composite sample
(Figure 5-14c) shows that epoxy is a brittle matrix with a clear fracture surface, and
the fibre breaks without pulling out of the matrix, which suggests a good interfacial
adhesion. In this system, the matrix fails in brittle way before the fibres start
breaking, and this matrix fracture including sharp cracks induces the fibre failure,
which then leads the whole composite to fail easily.
Figure 5-14. Fractured tensile samples of UD coir fibre composites, from left to right: (a) fracture
surface of coir/PP and (b) coir/PVDF fracture surface showing many fibres pull-out; (c) fracture
surface of coir/epoxy showing brittle matrix and good interfacial adhesion.
Chapter 5 154
Table 5-5. Tensile properties and efficiency factors of E-modulus, strength and failure strain of the UD composites; normalised E-modulus and strength
at a fibre volume fraction of 50%.
Composites
(%)
E-Modulus
(GPa)
Strength
(MPa)
Strain at
failure
(%)
Theo. E-
Modulus
(GPa)
Theo.
Strength
(MPa)
Eff.
factor of
E-
modulus
Eff.
factor of
Strength
Eff.
factor of
strain at
failure
Normalised
E-Modulus
(Vf = 50%)
Normalised
Strength
(Vf = 50%)
Ucoir /PP 44 2.41 ± 0.11 43.0 ± 0.8 3.3 ± 0.3 2.70 110.6 0.89 0.39 0.11 2.52 45.7
Ucoir /PVDF 35 2.42 ± 0.15 46.2 ± 3.3 3.6 ± 0.4 3.04 112.0 0.80 0.41 0.12 2.57 52.3
Ucoir /Epoxy 49 2.95 ± 0.20 34.9 ± 2.8 1.3 ± 0.1 3.28 85.8 0.90 0.41 0.04 2.96 35.6
Tcoir /Epoxy 53 3.28 ± 0.21 52.4 ± 5.1 1.8 ± 0.3 3.33 92.8 0.98 0.56 0.06 3.25 49.4
Mechanical properties of UD coir fibre composites 155
Figure 5-15 displays the E-modulus and strength of the composites, normalised to
the same fibre volume fraction of 50% for a fair comparison between the
composites. It can be seen that the coir/epoxy composites are stiffer than the coir/PP
and coir/PVDF composites. On the other hand, the coir/PVDF is stronger compared
to coir/PP and even coir/epoxy. The low strength of coir/epoxy is due to premature
failure in the brittle epoxy polymer. When comparing the untreated and alkali treated
fibre composites in epoxy, higher E-modulus and strength are obtained for treated
coir/epoxy, which may be thanks to a better wetting, leading to less interface
defects.
Figure 5-15. Normalised E-modulus and strength of the composites at Vf = 50%.
45,7
52,3
35,6
49,4
2,5 2,6 3,0
3,2
0,0
1,0
2,0
3,0
4,0
5,0
6,0
0
10
20
30
40
50
60
Ucoir/PP Ucoir/PVDF Ucoir/epoxy Tcoir/epoxy
E-m
od
ulu
s (G
Pa)
Stre
ngt
h (
MP
a)
Composites
Normalised Strength Normalised E-Modulus
Chapter 5 156
5.3.3 Impact strength of UD composites
5.3.3.1 Impact strength of UD coir/PP and UD coir/epoxy composites
Figure 5-16. Impact strength of UD coir fibre composites with PP and epoxy matrix, and of the neat
matrices.
Izod impact tests were carried out on unnotched UD untreated coir fibre composites
in PP and epoxy. The impact strength of the composites is compared to that of the
neat matrix, as shown in Figure 5-16. It can be seen that there is no improvement of
toughness when coir fibres are reinforcing PP (with Vf = 40%); the impact strength
of the composite is even somewhat lower than that of neat PP. This can be explained
by the fact that the toughness of coir fibre and PP are not significantly different.
Material toughness can be defined as the amount of energy per volume that a
material can absorb before fracture, which also can be determined by the area under
the stress-strain curve (till failure). As seen in Figure 5-17, the toughness of coir
fibre and PP is similar, hence, it is likely there is no toughening effect of the fibre in
its composite with PP (although various mechanisms can be invoked to introduce
toughness like increasing fibre pull-out energy through friction which will depend
on fibre-matrix adhesion, which is low in this case).
47,1
1,5
38,3
12,1
0
5
10
15
20
25
30
35
40
45
50
55
untreated coir-PP untreated coir-epoxy
Imp
act
stre
ngt
h (
KJ/
m2)
Composites
neat matrix
composites
Mechanical properties of UD coir fibre composites 157
Figure 5-17. Schematic presentation of stress-strain curves of coir fibre, epoxy and PP.
In case of the coir/epoxy composite (with Vf = 34.5%), the impact strength of the
composite is much higher than the value for the epoxy. Observably, the coir fibres
can improve the toughness of the epoxy by minimum a factor of five when the
impact strength is considered as toughness indicator. The toughening mechanism
can be observed with SEM images of the fracture surfaces of the composites, as
shown in Figure 5-18b: even without debonding, it seems that the coir fibres can
fully envelop their energy absorbing fracture mechanisms, namely defibrillation of
the elementary fibres inside the coir fibres.
Figure 5-18. Fracture surface of UD composites in Izod impact test (a) coir/PP (b) coir/epoxy (c)
defibrillation of coir fibre.
Chapter 5 158
5.3.3.2 Effect of fibre volume fraction and fibre treatment on the impact
strength of UD coir fibre epoxy composites
Further study on the impact properties of coir fibre epoxy composites was carried
out by investigating the effect of fibre volume fraction and fibre treatment on the
composite impact strength. The relation between fibre volume fraction and the
impact strength of the untreated and alkali treated fibre epoxy composites are
presented in Figure 5-19. The impact strength of neat epoxy is shown at fibre
volume fraction of zero. It can be seen that the impact strength increases with
increasing fibre volume fraction. For the untreated fibre composite, the impact
strength reaches the highest value around 35%, and declines above 40% fibre
volume fraction. In case of the treated fibre composite, the impact strength of the
composite with low fibre volume fraction is situated at more or less the same value
as for the untreated fibre composite. At higher volume fractions (above 40%) the
impact strength is higher for the alkali treated fibre composites.
Figure 5-19. The relation between fibre loading and the impact strength of UD untreated and alkali
treated coir fibre epoxy composites.
It is known that the impact strength of a composite is influenced by many factors
including the toughness properties of the reinforcement, the nature of the interfacial
region, and the frictional work involved in pulling the fibres from the matrix. In this
case, the tough coir fibres and the fibre-matrix interfacial interactions may play a
key role in the impact strength of the composite. When the fibre loading increases,
Mechanical properties of UD coir fibre composites 159
typically toughness enhancement occurs in composites. However, at high fibre
loading the epoxy resin is likely prevented to completely wet out the fibre bundles,
which changes the interfacial properties. This results in less fibres being loaded
properly or participating in energy absorption by pull-out, and the composite
toughness goes down. By treatment of the fibre, the wetting and interfacial adhesion
are improved, which allows more efficient stress transfer, leading to an increase of
energy absorption by the defibrillation of the coir fibres.
5.3.4 Tensile properties of UD coir/bamboo hybrid composites
Tensile behaviour of the hybrid composite
Figure 5-20. Typical tensile stress-strain curves of UD coir-bamboo/PP hybrid composites,
displayed together with stress-strain curves of UD mono coir/PP and UD mono bamboo/PP
composites.
UD coir and bamboo fibre hybrid composites in PP matrix were produced with fibre
volume fractions of the coir and bamboo fibres of approximately 30% and 8%
respectively. The tensile stress-strain curves of the UD coir-bamboo hybrid
composites are presented in Figure 5-20.
It can be seen that the composites show an almost linear-elastic behaviour until a
peak stress, and then the stress dramatically decreases to a certain value. From this
point on, the stress reduces slowly in a plastic manner. From this behaviour, it is
suggested that the coir fibres and the bamboo fibres together carry the tensile load
until reaching the peak stress, at which point most bamboo fibres (with a low fibre
load of 8%) fail, leading to a drop in stress. From this point on, the remaining coir
fibres continue to bear some stress until the whole composite fails. When comparing
Chapter 5 160
the hybrid composites with the mono composites (Figure 5-20), the bamboo/PP
composite fails in a brittle manner at high strength but low strain (<1%), and the
coir/PP system shows a failure (as discussed in paragraph 5.3.2) at low strength and
somewhat higher strain; the E-modulus and strength of the hybrid composite is
situated at intermediate values and there is furthermore some residual stress after the
peak stress till higher strain values. This demonstrates a hybrid effect when
combining strong bamboo fibres with high elongation coir fibres. Moreover, the
failure strain of the bamboo fibres in the hybrid composite (~1.2%) is higher than in
the mono-composite (~0.8%), suggesting that the presence of the coir fibres has a
beneficial effect on the failure strain of the bamboo fibres. A possible explanation
could be the stronger thermal contraction of the coir fibres during cooling after
compression moulding, leading to a mild compressive residual strain in the bamboo
fibres.
Figure 5-21. Fracture of the hybrid composite (a) typical sample fracture in tensile test (b) a cross-
section of the composite showing coir and bamboo fibre are distributed in 3 layers; (c) and (d)
fracture surface of the composite in tensile test.
Mechanical properties of UD coir fibre composites 161
In Figure 5-21b, the cross-section of the hybrid composite shows the distribution of
the coir fibres and the bamboo fibres, which are still positioned in three different
layers. It means that the hybrid effect in the composite is taking place at meso level.
The composite fracture shows many pulled out coir fibres and the presence of a few
broken bamboo fibres. It is likely that the pull-out of the coir fibres delayed the
failure of the composite as observed in its stress-strain curve.
Table 5-6. Tensile properties of coir-bamboo/PP hybrid composite, and of mono bamboo/PP
composite [8].
Composites
(%)
(%)
E-Modulus
(GPa)
Strength
(MPa)
Strain at
failure
(%)
Coir-bamboo /PP 30 8 7.8 ± 0.9 87.6 ± 4.4 2.2 ± 0.8
bamboo /PP 0 45 25.5 148.3 0.8
The foregoing is confirmed in table 5-6, which once more summarises the tensile
mechanical properties of the hybrid composites. The strain to failure of the hybrid
composite is clearly higher than that of the mono bamboo/PP composite. The
analysis of the tensile properties is further carried out by comparison with theoretical
values determined by the rule of mixtures.
Figure 5-22. Mono-material properties used as input to calculate the properties following the rule of
mixtures of the coir-bamboo/PP hybrid composite.
Chapter 5 162
Rule of mixtures for the hybrid composite
The theoretical E-modulus of the composite is calculated as:
(5-10)
As illustrated by Figure 5-22, the theoretical strength is estimated as follows:
(i) If is very low compared to : the strength of the composite is
determined by coir fibre strength. In this case, the coir fibres can carry
load after the failure of the bamboo fibres, then
(5-11)
(ii) If is high. Then, the composite strength is dependent on bamboo
fibre strength.
(5-12)
where and
are the stress in the coir fibre and the stress in
the PP respectively at the failure strain of the bamboo fibre (Figure 5-22).
With the fibre volume fraction of coir fibre and of bamboo fibres are 30%
(approximately 21% fibre solid volume fraction) and 8% respectively, the
theoretical strength of the composite calculated following Eq. 5-11 is 87.4 MPa,
which is lower than the value calculated following Eq. 5-12 (100.8 MPa). The result
shows the bamboo fibre load is high enough to determine the hybrid composite
strength. Hence, the theoretical strength of the composite will be calculated
according to Eq. 5-12.
Table 5-7. Theoretical E-modulus and strength of the hybrid composite estimated by the rule of
mixtures, and the efficiency factors of E-modulus and strength.
Composite
Theoretical
E-modulus
(GPa)
Efficiency
factor of
E-modulus
Theoretical
Strength
(MPa)
Efficiency
factor of
Strength
Coir-bamboo /PP 5.95 1.31 100.8 0.87
Mechanical properties of UD coir fibre composites 163
The theoretical E-modulus and strength of the hybrid composite is calculated
following Eq. 5-10 and 5-12, and shown in Table 5-7. The efficiency factors (the
experimental values normalised to the theoretical values) are also estimated. It can
be seen that the strength efficiency factor is surprisingly high (0.87) compared to the
values of the mono coir/PP (0.39) and bamboo/PP (0.43) systems. As discussed
above, there likely exists a beneficial effect of the residual strain in bamboo fibres,
leading to an important increase in failure stress, and hence a higher contribution to
the overall strength of the hybrid composite. For the E-modulus efficiency factor,
the value is higher than 1 which is still unexplained. Possibly it is related to an
inaccurate determination of fibre volume fraction.
In summary, the investigation of coir-bamboo hybrid composites in this work has
been an initial study. Only one configuration has been studied, which shows the
potential of creating a hybrid effect by combination of coir and bamboo fibres.
Further study should be directed at different fibre mixing levels and variation of
fibre loading to obtain a deeper understanding of the hybrid effect.
5.4 Conclusions
UD composites of untreated and 5% alkali treated coir fibres in both thermoplastic
and thermoset matrices were studied in this chapter. For manufacturing the
composites, a fibre alignment procedure was developed which provided a straight
and clean UD fibre preform for making good UD composites.
To characterise the mechanical properties of the composites, both flexural and
tensile tests were carried out on the UD composites. The flexural longitudinal
strength of coir/PVDF and coir/epoxy systems is significantly higher than in case of
coir/PP and coir/MAPP composites. Moreover, the transverse strength and the
longitudinal strength efficiency factors of the composites suggest that coir/PP
composites have low interfacial adhesion compared to the other composite systems,
which is in agreement with the study of interfacial adhesion in Chapter 4.
Improvement of the mechanical properties and interfacial strength were obtained by
the treatment of the fibres with alkali. Some composite systems were investigated in
tension. The results showed that untreated coir/PP and untreated coir/PVDF
composites failed at low strength due to weak interfacial adhesion, whereas the
coir/epoxy system had a premature failure, possibly due to the brittle matrix. All the
systems appeared highly defect sensitive in tensile loading. For the epoxy system,
the strength of the composite was improved by using alkali treated fibres.
Chapter 5 164
The impact properties of the UD composites were studied in two systems, namely
coir/PP and coir/epoxy. The result for the Izod impact strength showed that the
toughness of PP cannot be improved by adding coir fibres; while, for the brittle
epoxy, coir fibres can improve the toughness with minimum a factor of five. The
effects of fibre loading and fibre treatment on the composite impact strength were
also investigated for the coir/epoxy composites. The results showed an optimum
value of fibre loading to obtain highest impact strength. Above this value, the impact
strength decreased. The fibre treatment improved the composite interface, which
then also improved the composite impact strength.
An initial study on coir-bamboo fibre hybrid composites in PP was conducted. With
a low bamboo fibre fraction, a hybrid effect with an increase of composite strain to
failure was obtained, which can be attributed to the high strain to failure of the coir
fibres; the bamboo fibres provided high stiffness and strength to the composites. The
results show a potential for coir-bamboo hybrid composites, which justifies further
study.
References
1. Osorio, L., et al., Morphological aspects and mechanical properties of single
bamboo fibers and flexural characterization of bamboo/epoxy composites. Journal
of Reinforced Plastics and Composites, 2011. 30(5): p. 396-408.
2. Truong, T.C., The mechanical performance and damage of multiaxial multi-ply
carbon fabric reinforced composites, in Department of metallurgy and applied
materials science, Faculty of engineering sciences. 2005, Katholieke Universiteit
Leuven.
3. Swolfs, Y., L. Gorbatikh, and I. Verpoest, A 3D finite element analysis of static
stress concentrations around a broken fibre, in 15th European Conference on
Composite Materials (ECCM). 2012: Venice, Italy.
4. Swolfs, Y., et al., Interlayer hybridization of unidirectional glass fibre composites
with self-reinforced polypropylene, in 15th European Conference on Composite
Materials (ECCM). 2012: Venice, Italy.
5. Taketa, I., Analysis of failure machanisms and hybrid effects in carbon fibre
reinforced thermoplastic composites. 2011, Katholieke Universiteit Leuven:
Leuven, Belgium.
6. Tran, L.Q.N., et al., Investigating the interfacial compatibility and adhesion of coir
fibre composites. ICCM 18 proceeding, Korea 2011.
7. Cichocki Jr, F. and J. Thomason, Thermoelastic anisotropy of a natural fiber.
Composites Science and Technology, 2002. 62(5): p. 669-678.
8. Vander Velpen, H., Characterization of discontinuous UD bamboo fibre composites
in Department MTM. 2010, Katholieke Universiteit Leuven.
Conclusions 165
Chapter 6
Conclusions
Chapter 6 166
6.1 General conclusions
This doctoral thesis has presented a study on the structure and properties of natural
coir fibres, the mechanical properties of their composites, and especially the
interfacial interaction between the natural fibre and polymers which is significantly
important for using natural fibres in composite materials. The result of the research
has largely followed the study plan and achieved the thesis objectives.
As commonly accepted, the consideration for using natural fibres in composites is
based on a few main aspects: their favourable specific properties for composite
applications, economy, ecology, and society. Following these considerations, coir
fibres were shown as a suitable candidate; these are tough fibres and potentially can
perform as toughening reinforcement for brittle polymer composites. They can be
considered as cheap fibres for composites and are available in large amounts. For a
developing country like Vietnam, which has its own coir fibre production, the use of
coir fibre in composites will indirectly contribute to improving the low income of
local workers who are directly producing the fibres.
The most important contribution of this thesis is a developed procedure for studying
the interface of natural coir fibre composites. An integrated physical-chemical-
micromechanical approach to improve fibre-matrix interfacial compatibility and
adhesion was implemented. This knowledge can be applied in not only coir fibre
composites but also for other (natural) fibres used in composite materials. It can be
used to optimise the interface properties through fibre treatments and matrix
modifications.
In addition to above major results, the conclusions of this thesis are summarised in
terms of output, as follows:
6.1.1 Microstructure and mechanical properties of technical coir fibres
The internal structure of technical coir fibres was characterised using SEM and
SEM-CT. The result shows a technical coir fibre comprises plenty of elementary
fibres with the lumens inside, and a lacuna at the centre of the fibre. As a result, coir
fibre appears to have high porosity at approximately 30%. The elementary fibre is
built up by two main cell walls which consist of bundles of microfibrils aligned in a
high angle to the fibre axis (around 45 degrees in the primary wall, and close to 90
degrees in the secondary wall). Concerning the characterisation technique, SEM-CT
is a good tool for analysing the internal structure of coir fibre. The fibre porosity and
Conclusions 167
the dimensions of lumen, lacuna and elementary fibres were determined by using a
3D model of the scanned fibre. This technique can also be applied for
characterisation of other natural fibres.
The surface of coir fibre was observed by SEM and there are arrays of silicon rich
protrusions, which can possibly be removed by mechanical or chemical treatment of
the fibre surface. Furthermore the fibre surface consists of longitudinally oriented
cells with more or less parallel orientation; it is suggested in literature that these
cells are firmly held together by a binder of lignin and fatty substances which are
filling the intercellular space. This is confirmed in this study by the analysis of fibre
surface chemistry using XPS, in which XPS indicates a heterogeneous surface with
a high proportion of hydrocarbon rich material consisting of waxes, fatty substances
and lignin. Moreover, on fibre treatment with alkali, the waxes are largely removed
and leave a relatively homogenous surface with more exposed lignin as binder of
elementary fibres. The characterisation of the coir fibre surface provides useful
information which will help to improve or modify the fibre-matrix interfacial
adhesion when the fibres are used in composites.
The mechanical properties of the coir fibres were assessed by single fibre tensile
testing, in which the fibre strain was determined by two methods: with optical strain
mapping and using different test lengths. The results of both methods indicate that
coir fibres are not very strong and stiff, but have high strain to failure. This is
explained by the high microfibrillar angle in the fibres leading to the low stiffness in
fibre direction and to high elongation thanks to reorientation of the microfibrils
under tensile loading.
6.1.2 Wetting measurements and surface energy estimation of the fibres
A wetting measurement procedure was established to determine stable and
reproducible static contact angles of coir fibres, in which the effects on the contact
angle results of irregular wetted length along the fibre perimeter and liquid
absorption were carefully considered. Regarding this, the dynamic contact angles of
coir fibre were determined following the Wilhelmy method. Using the Molecular-
kinetic theory, the wetting behaviour of coir fibre is also modelled by fitting the
dynamic advancing contact angles corresponding to different measurement speeds,
which also provides the static contact angle. The values of the static angles were
further used to estimate fibre surface energy.
Chapter 6 168
The surface energy of coir fibres was estimated following two approaches, namely
the geometric-mean approach and the acid-base approach, which describe the fibre
surface energy comprising on the one hand polar and dispersive components and on
the other hand a Lifshitz – van de Waals component and acid-base components .
Both approaches suggest the coir fibre surface has high dispersive and low polar
contributions, which points to a surface with rather hydrophobic properties. This
result is also in agreement with the analysis of fibre surface chemistry by XPS.
6.1.3 Fibre-matrix interfacial compatibility and adhesion
Wetting analysis provides the surface energies of the coir fibres and various
matrices, which are used to calculate the fibre-matrix work of adhesion and
interfacial energy to predict the physical adhesion and compatibility of the
composites. Next, practical adhesion in single fibre composites and UD composites
was evaluated using single fibre pull-out tests and transverse three-point bending
tests. In this work, untreated and alkali treated coir fibres and various thermoplastics
were investigated.
From the wetting analysis, the results show that the work of adhesion of both
untreated and treated fibres with PVDF is higher than in case of PP and MAPP. On
the other hand, the results of pull-out and transverse 3PB tests show much higher
interfacial adhesion of coir fibres with both PVDF and MAPP in comparison with
PP. This suggests that the higher interfacial adhesion of coir fibres with PVDF
compared with PP is thanks to higher fibre-matrix physico-chemical interaction
corresponding with the work of adhesion, while the improvement of interfacial
adhesion between coir fibres and MAPP versus coir fibres and PP is likely
dominated by chemical bonding.
There has been a good agreement between the results of the wetting analysis and
those of the composite interface mechanical tests. The combination of different
characterisation techniques has offered a deeper understanding of the interfacial
adhesion and compatibility in coir fibre composites.
6.1.4 Performance of coir fibre composites
Mechanical properties of UD untreated and alkali treated coir fibre composites in
both thermoplastic and thermoset matrices were assessed by flexural tests, tensile
test and unnotched Izod impact tests.
Conclusions 169
The results from flexural testing show that the coir fibre composites with PVDF and
epoxy are stronger and stiffer compared to the coir fibre PP and MAPP composites.
The transverse strength of the composites indicates a low interfacial adhesion of
coir/PP composite in comparison with the other systems, which is consistent with
the results from the study of the composite interface. In addition, the mechanical
properties of the composites are improved by the fibre treatment, possibly thanks to
the better interfacial adhesion. The tensile testing of the composites was not highly
successful to determine the full mechanical properties due to the premature failure of
the composite samples, which shows a high defect sensitivity of the composites.
The impact strength of the UD coir/PP and coir/epoxy composites were obtained by
Izod impact testing, which shows that the toughness of PP cannot be improved by
adding coir fibres; while, for the brittle epoxy, coir fibres can improve the toughness
with minimum a factor of five. For coir/epoxy composites, the investigation of the
effect of fibre loading on the composite impact strength shows there is an optimum
value of fibre loading to obtain highest impact strength. Moreover, the impact
strength is also improved by alkali treatment of the fibres.
An initial study on coir-bamboo fibre hybrid composites with PP matrix was carried
out, where the coir fibre and bamboo fibre were mixed at meso level by layer by
layer stacking of UD fibre prepregs. The result shows that a positive hybrid effect is
obtained when a low bamboo fibre fraction is used, which leads to a higher
composite strain at failure compared to mono bamboo fibre composite.
6.2 Future work
The study of the microstructure of the coir fibres showed that they have a high
porosity with lumens in their elementary fibres. This hollow structure of the fibres is
expected to give good damping capacity. In future research, the coir fibre
composites should be characterised by vibration and acoustic damping tests.
The mechanical properties of coir fibre were obtained from single fibre tensile tests
using different methods to measure reliable values for the fibre elongation. Many
tests need to be done to have a statistically distributed result. Therefore, dry and
impregnated fibre bundle tensile tests are recommended for determination of the
fibre properties. With these tests, the statistical data for a larger population will be
obtained, which help to understand the mechanical properties of the fibres and to
conveniently analyse the final composites’ behaviour.
Chapter 6 170
In wetting analysis of the fibres, the static equilibrium contact angles of the fibre in
various test liquids were determined by two methods, which are fitting the dynamic
angles by Molecular-kinetic theory and the relaxation of the liquid meniscus during
static contact angle measurement with a tensiometer. Another method will be
considered, in which a sound vibration will be used to force the test liquid into
equilibrium state during a static contact angle measurement, which is supposed to
provide more reliable equilibrium contact angles.
Alkali treatment was used to modify the fibre surface for studying the composite
interfacial adhesion and the mechanical properties of the composites. More
treatments should be applied on the fibre for further understating the fibre-matrix
interfacial interaction and the composite properties.
It was found that the coir fibre can ameliorate the toughness of brittle epoxy. This is
an interesting result which should be further studied with other matrix systems.
The preliminary work on coir-bamboo fibre hybrid composites showed a potential
use and application of coir fibre in composites. An intensive study should be focused
on this topic, in which several hybrid composites will be considered, by
hybridisation with bamboo fibre or flax fibre. The hybrid effect of the composites
will be characterised on both the meso level (fibres are mixed at the scale of fibre
layers), and micro level (mixing at single fibre level). Different fibre volume
fractions will also be considered in analysis of the hybrid effect.
Appendix A 171
Appendix A
Scanning set up and parameters for analysis of coir fibre using
Skyscan Micro-CT in SEM (SEM-CT)
The Skyscan Micro-CT attachment for the SEM Philips XL 30 FEG in the
Department MTM allows visualisation and measurement of the 3D internal structure
of an object, which can be applied for analysis of microstructure of natural fibres,
like coir fibre. The set up of the scanning equipment is shown in Figure 1, in which
a Titanium target is used in combination with a SEM electron beam for producing
X-rays. The resolution of scanning images can be changed by adjusting the distance
between the target and the sample.
Figure 1. Set up for SEM-CT scan of coir fibre.
Coir fibre sample Ti target
Appendix A 172
Sample specification for scanning with this equipment:
Spatial resolution: from 350 nm to 8m
Sample length: up to 8-10 mm
Sample cross-section: 0.18 - 4mm
Scanning parameters used for coir fibres:
Source voltage of SEM: 30 kV
Source current: 113-122 µA
Image resolution: 1.8, 2.3 m (image pixel size)
Exposure time: 4000 ms
Gain: 5 times
The Skyscan software NRecon is used to reconstruct cross-section images from
scanning projection images. The reconstructed set of slices can be viewed in the
Skyscan Data Viewer program. And the analysis of fibre microstructure including
morphology measurements, 2D/3D distances and angle measurements can be carried
out using the software CTanalyser.
Appendix B 173
Appendix B
Single fibre tensile testing with optical strain mapping
1. Sample preparation for 2D mapping of fibre deformation
The deformation of the fibre is measured using the image correlation technique,
which tracks the movement of small speckles in a speckle pattern on the fibre
surface. Firstly, the speckles were created on the fibre surface by spraying black
paint (coir fibre has a light colour). Due to the small area of the fibre surface, it is
necessary to obtain a small size of speckles for a sufficient tracking in lateral image
processing. In case of coir fibre, the fibre elongation is high, hence a small test
length (5 mm) was used to prevent the loss of tracking due to a large movement of
the speckles (Figure 1)
Figure 1. Set up for fibre tensile test with strain mapping.
5 mm
Speckles on coir
fibre surface
Appendix A 174
2. Image correlation for fibre strain measurement
Images recorded by the camera during tensile loading are correlated for analysis of
fibre displacement using the software Vic-2D. The analysis procedure is as follows:
Open the recorded images by selecting Speckles Images in the common task.
Select the analysed area on the fibre surface (using rectangle function R).
Adjust the size of subset (small values for small sample size)
Then, run the correlation process (using the green arrow button)
A series of data files is created after finishing the correlation process. These
files are used for the strain analysis (select Data as shown in Figure 2). In
the Plotting tools, the measurement of fibre strain can be set by selecting
variable exy.
Appendix B 175
Select an area of interest (red rectangle line in Figure 2). Then, use function
‘inspect rectangle’, followed by ‘extract’ to get the result of fibre strain exy.
The distribution of fibre strain is presented as in Figure 2.
Figure 2. Strain distribution of analysed coir fibre using image correlation.
Appendix A 176
Curriculum Vitae
Personal data
Name: Le Quan Ngoc Tran
Date of birth: 23/01/1978
Nationality: Vietnamese
Address: Schapenstraat 37/105
3000 Leuven
Belgium
GSM: +32 487 19 56 39
e-mail: [email protected]
Education
2008 – present Ph.D., Materials Engineering, Katholieke Universiteit Leuven,
Belgium.
Dissertation: Polymer composites based on coconut fibres
2002 – 2004 M.Sc., Materials Engineering: Polymers and Composites, Katholieke
Universiteit Leuven, Belgium.
Thesis: Internal structure and mechanical properties of random long
glass fibre composite.
1994 – 1999 B.Sc., Chemical Engineering, Ho Chi Minh City University of
Technology, Vietnam.
Thesis: Polymer blends of natural rubber and polyvinyl chloride.
Internship
June 2006 Training on “Synthesis of nanomaterials” at ARC Centre for
Functional Nanomaterials, The University of Queensland, Australia.
July-August 2004 Technical training on “Composites processing techniques” at Arplam
NV, Arplama Group, Brugge, Belgium.
October-November
2000
Training on “Construction materials from natural fibres” at Technishe
Universitaat Dresden, Germany.
Professional experience
2008 – present
Research assistant in Composite Materials Group, KU Leuven, Belgium.
2004 – 2008
Researcher and lecturer in Polymers and Composites at Can Tho
University, Vietnam.
Publications
Journal papers
1. L.Q.N. Tran, C.A. Fuentes, C. Dupont-Gillain, A.W. Van Vuure, I. Verpoest;
Understanding the interfacial adhesion and compatibility of coir fibre thermoplastic
composites; Composites Science and Technology (2012). (Submitted).
2. C.A. Fuentes, L.Q. N. Tran, C. Dupont-Gillain, A.W. Van Vuure, I. Verpoest; Interfaces
in natural fibre composites: effect of surface energy and physical adhesion; Journal of
Biobased Materials and Bioenergy (2012). (Accepted)
3. L.Q. N. Tran, C.A. Fuentes, C. Dupont-Gillain, A.W. Van Vuure, I. Verpoest; Wetting
analysis and surface characterisation of coir fibres used as reinforcement for composites;
Colloids and Surfaces A 337 (2011) 251-260.
4. C.A. Fuentes, L.Q. N. Tran, C. Dupont-Gillain, A.W. Van Vuure, I. Verpoest; Wetting
behavior and surface properties of bamboo fibres; Colloids and Surface A 380 (2011) 89.
5. N. Defoirdt, S. Biswas, L. De Vries, L.Q.N. Tran, J. Van Acker, Q. Ahsan, L. Gorbatikh,
A. Van Vuure, I. Verpoest; Assessment of the tensile properties of coir, bamboo and jute
fibre; Comp. Part A 41 (2010) 588-595.
Conference proceedings
1. L.Q. N. Tran, C.A. Fuentes, C. Dupont-Gillain, A.W. Van Vuure, I. Verpoest; Coir fibre
composites: from fibre properties to interfacial adhesion and mechanical properties of
composites; ECCM 15 European Conference on Composite Materials, Venice, Italy; June
2012.
2. L.Q.N. Tran, C.A. Fuentes, C. Dupont-Gillain, A.W. Van Vuure, I. Verpoest;
Investigation of the interfacial compatibility and adhesion of natural (coir) fibre
thermoplastic composites; SAMPE Benelux student seminar, Ermelo, Netherlands;
January 2012.
3. L.Q.N. Tran, C.A. Fuentes, C. Dupont-Gillain, A.W. Van Vuure, I. Verpoest;
Investigating the interfacial compatibility and adhesion of coir fibre composites ; ICCM
18 International Conference on Composite Materials, Jeju, South Korea; August 2011.
4. L.Q. N. Tran, C.A. Fuentes, C. Dupont-Gillain, A.W. Van Vuure, I. Verpoest; Interfacial
adhesion and mechanical properties of unidirectional coir fibre composites; Ecocomp 4th
International Conference on Sustainable Materials, Polymer and Composites;
Birmingham, UK; July 2011.
5. L.Q. N. Tran, C.A. Fuentes, C. Dupont-Gillain, A.W. Van Vuure, I. Verpoest; Wetting
behaviour and surface characteristics of coconut (coir) fibres used as reinforcement for
composites; ECCM 14 European Conference on Composite Materials, Budapest,
Hungary; June 2010.
6. L.Q. N. Tran, C.A. Fuentes, C. Dupont-Gillain, A.W. Van Vuure, I. Verpoest; Wetting
behaviour and surface energy of coconut (coir) fibres; Natural Fibres 09’ International
Conference, London, UK; London 2009.
