Thesis for the Degree of Doctor of Philosophy

Development of thermoplastic biocomposites

based on aligned hybrid yarns for fast composite

manufacturing

Behnaz Baghaei

Copyright©Behnaz Baghaei

Swedish Centre for Resource Recovery

University of Borås

SE-501 90 Borås, Sweden

Digital version: http://urn.kb.se/resolve?urn=urn:nbn:se:hb:diva-764

ISBN 978-91-87525-79-7 (printed)

ISBN 978-91-87525-80-3 (pdf)

ISSN 0280-381X Skrifter från Högskolan i Borås, nr. 74

Cover Photo: Masoud Salehi

Printed in Sweden by Responstryck AB

Borås 2015

I

Abstract

The interest in natural fibres as reinforcement for composite materials has been steadily increasing

due to their attractive mechanical properties and the possibility of making more eco-friendly

materials. Currently, various alternatives are being introduced for commercial applications, as fibres

such as hemp, jute and flax exhibit properties, which make them appropriate for structural

composite components. Biocomposites offer reductions in weight and cost and have less reliance on

foreign oil resources, making them attractive. Several investigations have revealed that the full

utilisation of fibre mechanical properties in the final composites can be exploited, provided an

aligned fibre orientation is chosen. In fact, a major challenge for natural fibre reinforced composites

is to achieve high mechanical performance at competitive prices. The use of commingled/hybrid

yarns is one of the more promising methods for manufacturing structural thermoplastic composites.

Commingled yarns of thermoplastic and reinforcing fibres offer a potential for cost-effective

production of composite parts, thanks to reduced applied pressures and impregnation times during

processing. Besides economic advantages, there is also direct control over fibre placements and ease

of handling of fibres in yarn process. The yarn technologies provide homogenous distribution of

reinforcing fibre and matrix. Variation in natural fibre properties has been a major problem facing

composite manufacturers, compared to carbon and glass fibres that have well-defined production

processes. This issue can be addressed by regenerated cellulose fibres. These fibres can be

reproduced easily with high surface evenness and even quality, making it possible to get consistent

results, which is not possible with natural fibres. Combination of natural and regenerated cellulose

fibre brings together the best of both materials. The end result is a product with superior properties,

which could not be obtained by the individual components.

This thesis describes the development of aligned hybrid yarns with low fibre twist, for high

performance natural (hemp) and man-made (Lyocell) cellulose fibre-reinforced biocomposites,

suitable for use in structural or semi-structural applications. The properties of composites in terms

of fibre orientation, off-axis angle and alkali treatment were investigated, focusing on determining

void%, water absorption, mechanical and thermo-mechanical properties. The results show that

combining hemp and Lyocell in PLA composite leads to the reduction of moisture absorption and

can improve the mechanical properties. The mechanical properties of the composites were highly

affected by the fibre direction. The alkali treatment on hemp fibre improved the mechanical

properties of the composites.

Keywords: Hybrid yarns, Mechanical properties, Porosity, Weaving, Alkali treatment, Compression

moulding

II

III

List of Publications

This thesis is based on the results presented in the following publications:

I. Baghaei B, Skrifvars M and Berglin L. (2013) Manufacture and characterization of

thermoplastic composites made from PLA/hemp co-wrapped hybrid yarn prepregs. Composites

Part A: Applied Science and Manufacturing 50:93-101.

II. Baghaei B, Skrifvars M, Salehi M, Bashir T, Rissanen M and Nousiainen P. (2014) Novel

aligned hemp fibre reinforcement for structural biocomposites: Porosity, water absorption,

mechanical performances and viscoelastic behaviour. Composites Part A: Applied Science and

Manufacturing 61:1-12.

III. Baghaei B, Skrifvars M, Berglin L. (2015) Characterization of thermoplastic natural fibre

composites made from woven hybrid yarn prepregs with different weave pattern. Composites

Part A: Applied Science and Manufacturing 76:154-161.

IV. Baghaei B, M. Skrifvars, Rissanen M and Ramamoorthy SK. (2014): Mechanical and thermal

characterization of compression moulded polylactic acid natural fiber composites reinforced by

hemp and Lyocell fibers. Journal of Applied Polymer Science, 131(15) DOI:10.1002/app.40534.

V. Baghaei B, M. Skrifvars. Characterisation of polylactic acid biocomposites made from prepregs

composed of woven polylactic acid/hemp-Lyocell hybrid yarn fabrics. Submitted to Composites

Part A: Applied Science and Manufacturing

Statement of Contribution

The author of this thesis, Behnaz Baghaei, was main author of all the publications and performed

most of the experimental work, data analysis and manuscript writing in all publications. Dr Lena

Berglin prepared the textile fabrics for Publications III and V. Dr Marja Rissanen performed the

tensile testing of the fibres for Publications II and IV. Other co-authors assisted with data analysis

and manuscript writing.

SEM analysis was performed at Swerea IVF (Sweden) for Publication I and Chalmers University of

Technology (Sweden) for Publications II, III, IV and V.

Permission from publishers

Permission was obtained from:

- ELSEVIER to include Publications I, II and III in the printed version of this thesis

- John Wiley and Sons to include Publication IV in the printed version of this thesis

IV

Conference Contributions

1. Baghaei B and Skrifvars M. Investigation of pattern style of woven fabrics produced from

hybrid wrap spun yarns on fabricated composite, 20th International Conference on

Composite Materials in Copenhagen, Denmark, July 19-24, 2015 (oral presentation)

2. Baghaei B, Skrifvars M and Salehi M. Aligned hemp yarn reinforced biocomposites:

porosity, water absorption, thermal and mechanical properties, 16th European conference on

composite materials, Seville, Spain, June 22-26, 2014 (oral presentation)

3. Baghaei B, Skrifvars M and Berglin L. Tailoring of the mechanical and thermal properties

of hemp/PLA hybrid yarn composites, 3rd Avancell conference, Göteborg, Sweden, October

8-9, 2013 (Poster)

4. Baghaei B, Skrifvars M and Berglin L. Tailoring of the mechanical and thermal properties

of hemp/PLA hybrid yarn composites, IPLA Global Forum on Sustainable Waste

Management for the 21st Century Cities, Borås, Sweden, September 9-11, 2013 (Poster)

5. Baghaei B, Skrifvars M and Berglin L and Ramamoorthy SK. Hemp/PLA co-wrapped

hybrid yarns for structured thermoplastic composites, 50th

Anniversary Nordic Polymer

Days, Helsinki, Finland, May 28-31, 2013 (oral presentation)

V

Acknowledgements

I would like to thank everyone who has contributed immensely to my life in many remarkable ways.

First and foremost, I would like to express my deepest gratitude to my supervisor, Professor Mikael

Skrifvars, for his support during these past four years. I appreciate all his contributions of time, idea,

scientific advice and immense knowledge.

Besides my supervisor, I would like to thank my co-supervisors, Dr Lena Berglin and Dr Dan

Åkesson, for their insightful comments and encouragement.

I also wish to extend my gratitude to my examiner, Professor Tobias Richards, for making it

possible for me to present this dissertation.

Professor Mohammad Taherzadeh, I am so grateful for your valuable guidance.

The members of polymer group have contributed immensely to my personal and professional time

at Borås. The group has been a source of friendship as well as good advice and collaboration. To my

friends: Tariq, Fatimat, Haike, Sunil, Adib, I say thank you.

I would like to thank Dr Peter Therning, Tomas Wahnström, Sari Sarhamo, Susanne Borg, Louise

Holmgren and Thomas Södergren for your support throughout my studies.

I am also very grateful to my teachers, Dr.Magnus Lundin, Dr Kim Bolton and Dr Anita Pettersson.

I extend my gratitude to all PhD students and colleagues at the Swedish Centre of Resource

Recovery for being such wonderful friends and creating a pleasant working environment.

In addition, I thank my wonderful friends Khatereh, Abas, Narges, Ehsan, Noushin, Mahdi, Mehran,

Zohreh, Khosrow, Parvin, Nima, Shahram, Mehrsa, Johan, Kamran, Solmaz, Maryam, Gergely,

Marjan, Farhang, Samira, Farzad, Negar, Arman, Davood, Mina, Masoud, Faranak, Amin, Sima,

Raj, Rouzbeh and Sahar.

To my husband, Masoud, I express my deepest gratitude. You have always been there cheering me

up and standing by my side through the good times and bad. I would also like to thank my dear

parents, brother (Amirali), sister (Bahareh) and brother-in-law (Majid). They have always supported

me and encouraged me with their best wishes. I am thankful for all support from my parents-in-law.

I express my gratitude to Patrik Johansson and Joachim Almvång from Engtex Company for

making the fabrics for this project.

Finally, I would like to thank other institutions that were part of my research, either helping me with

instruments or providing support: Swerea AB, Sweden; Chalmers University, Sweden and Tampere

University of Technology, Finland.

This work was financed by Ångpanneföreningens (Åforsk) and Smart Textiles (University of Borås)

and their contributions are greatly acknowledged.

VI

VII

Abbreviations

NMMO N-Methylmorpholine N-oxide

NFC Natural fibre composites

NaOH Sodium hydroxide

SEM Scanning electron microscopy

WG Weight gain

Wt% Weight %

Vol % Volume %

l/d Aspect ratio

DMTA Dynamic mechanical thermal analysis

DSC Differential scanning calorimetry

Tg Glass transition temperature

TC Crystallisation temperature

Tm Melting temperature

ΔHcc Cold crystallisation enthalpy

ΔHm Melting enthalpy

Tex Mass of yarn in grams per 1,000 metres length

VIII

IX

TABLE OF CONTENTS

Abstract I

List of Publications III

Conference Contributions IV

Acknowledgements V

Abbreviations VII

1. INTRODUCTION 1

1.1 Justification 2

1.2 Aim of this study 3

1.3. Outline of the thesis 4

2. LITERATURE REVIEW 5

2.1 Natural fibres 5

2.1.1 Hemp 5

2.1.2 Regenerated cellulose fibre 6

2.2 Fibre surface modifications 7

2.2.1 Alkali treatment 7

2.3 Biodegradable polymers 8

2.3.1 Polylactic acid (PLA) 8

2.4 Biocomposites 8

2.4.1 Hybrid composites 9

2.4.2 Thermoplastic composites 10

2.4.3 Engineering applications of biocomposites 13

3. EXPERIMENTAL 15

3.1 Materials 15

3.1.1 Reinforcements 15

3.1.2 Matrices 15

3.1.3 Chemicals 15

3.2 Methods 15

3.2.1 Fibre alkali treatment 15

3.2.2 Manufacture of hybrid yarns 15

3.2.2.1 Co-wrapped hybrid yarn 15

3.2.2.2 Wrap spinning of hybrid yarns 16

3.2.3 Preparation of prepregs 17

3.2.4 Composite preparation 19

3.3 Characterisation 19

3.3.1 Single fibre tensile test 19

3.3.2 Composite density and porosity 19

3.3.3 Water absorption test 19

3.3.4 Tensile testing 20

3.3.5 Flexural testing 20

3.3.6 Impact testing 20

3.3.7 Dynamic mechanical thermal analysis 20

X

3.3.8 Differential scanning calorimetry 20

3.3.9 Scanning electron microscopy (SEM) 21

4. SUMMARY OF RESULTS 23

4.1 Effect of heat and alkali treatments on fibre properties (Publications II and IV) 23

4.2 Composites properties (Publications I, II, III, IV and V) 24

4.2.1 Composite porosity 24

4.2.2 Water absorption 25

4.2.3 Tensile properties 28

4.2.4 Flexural properties 30

4.2.5 Impact resistance 32

4.2.6 Dynamic mechanical thermal testing 34

4.2.7 Differential scanning calorimetry 37

5. DISCUSSION AND CONCLUSIONS 41

6. FUTURE STUDIES 43

REFERENCES 45

PUBLICATION I

PUBLICATION II

PUBLICATION III

PUBLICATION IV

PUBLICATION V

INTRODUCTION

1

1. INTRODUCTION

The idea of making composite materials is not a new one; nature has many examples wherein the

idea of composite materials is applied. For example, the coconut palm leaf is nothing but a basis

using the concept of fibre reinforcement. Wood is a fibrous composite; it is cellulose fibres in lignin

matrix. In addition to these naturally occurring composites, there are several other engineering

materials that are composites and that have been applied for a very long time [1]. In general the

composite materials consist of a matrix, reinforced with particles or fibres to create a material with

specific optimal properties. The two components control different aspects of the composite;

reinforcing fibres influence the strength and stiffness properties while the matrix acts to transfer

loads between fibres and protects against environmental conditions such as chemicals, heat and

moisture. The matrix binds the fibres together and is usually composed of metals, polymers or

ceramics [2].

Natural fibres have been applied as reinforcement in composites from very early on till the mid-

twentieth century; however, the interest in them had decreased with the development of synthetic

fibres such as carbon and glass fibres. Since mid-20th

century, there was an increased demand for

stronger and stiffer, still light weight, composites, in fields such as aerospace, transportation and

construction. This led to the incorporation of high performance fibres for reinforcement.

Consequently, the success of new synthetic materials almost ceased the development of natural

fibres. Nonetheless, there is a revivification of interest in natural fibres lately, due to environmental

reasons. They are coming back in high-tech world. The application of modern technologies,

knowledge and scientific tools for these natural materials has given rise to new developments in this

field [3, 4].

Currently, the use of thermoplastic resins in some of the composite application is clearly of higher

potential than the use of thermoset resins because of: better impact strength, easier recycling, faster

processing conditions (no time for curing is required), possibility of production in longer series,

lower cost, absence of toxic solvents and higher fracture toughness and elongation on fracture [5-7].

However, the high viscosity of molten thermoplastics, compared to thermosets makes the process

more problematic. Different impregnation methods have been proposed and applied.

A current trend in composites fabrication is to reduce the number of processes required to produce a

component rapidly and cost-effectively. Therefore, a series of required processes including

impregnation of reinforcement preforms, consolidation, forming into shape and solidification to be

done sequentially are now often combined into a single process, one that is highly integrated [8].

Among the possible methods for fabricating continuous fibre reinforced thermoplastic composites,

consolidation of preforms based on commingled/hybrid yarns, created on the principle of combining

reinforcing fibres and matrix fibres homogeneously during spinning, is an economical method for

the manufacture of complex shapes. The hybrid yarns can be further converted into a drapable

woven fabric preforms for forming the complex shaped-parts. The good distribution of

reinforcement and matrix in a non-molten state in commingled yarn leads to very short polymer

flow distance for impregnation, short cycle times and reduction in production cost [9, 10]. The

INTRODUCTION

2

quality of the component distribution in yarn is known to affect the mechanical properties of the

manufactured composites.

The fibre orientation (i.e. alignment of the fibres) must be controlled to ensure that the fibre

mechanical properties are efficiently utilised in order to attract industrial interest as an alternative to

the traditionally applied synthetic fibres (e.g. glass fibres). It is evident that PLA/hemp fibre

composites can compete with glass fibre composites regarding stiffness, whereas for tensile and

impact strength, the properties are still not on a satisfactory level [11-16]. Previous studies have

demonstrated that the full reinforcement potential of natural fibres can be exploited in

biocomposites if an aligned fibre orientation is used [16, 17]. Natural fibres are inherently

discontinuous; therefore, natural fibre reinforcements reported so far are based on twisted spun

staple yarns, which are produced by mechanical spinning methods, mainly ring spinning. These

spun yarns tend to be highly twisted, which leads to fibre misalignment due to their helical paths

around the yarn axis. Specifically, this misalignment contributes negatively to the mechanical

properties of the resultant composites. Another negative impact of yarn twist is that it tightens the

yarn structure, rendering resin impregnation difficult [18]. Therefore, in the textile industry, a broad

range of techniques for the alignment of natural fibres have been developed and optimised to

produce yarns with controlled fibre orientations by reducing or replacing twist in yarns.

To fully benefit from these advantages, the process parameters governing consolidation must be

identified and optimised. Consolidation of commingled yarns has been studied and modelled by

several authors [1-6] and has recently been reviewed by Svensson et al. [7]. In all cases, the

consolidation quality was characterised by the amount of residual porosity in the part. In general,

the polymer fibres melt rapidly and form liquid pools surrounding the reinforcing fibres. The

problem thus results in infiltration of dry fibre bundles surrounded by molten polymer pools. As

reinforcing fibres and polymer fibres are in general not distributed homogeneously within a yarn,

commingling quality becomes one of the most critical parameters [9].

1.1 Justification

There is growing interest in the use of natural fibre composites (NFCs) either in terms of industrial

applications or fundamental research studies. They are renewable, completely or partially recyclable

and relatively cheap. Plant fibres, such as hemp, sisal, flax, kenaf, ramie, etc., are often used as

reinforcement for composites due to their renewability, availability, low density, price and good

mechanical properties that make them an attractive ecological alternative to conventional fibres

such as carbon, glass and aramid fibres. Moreover, the European Union has set environmental

sustainability policies and regulations [19] aiming to reduce the land filling and apply renewable,

sustainable and biodegradable materials. Currently, NFCs are applied in transportation, military

applications, building and construction industries and packaging, etc. [20-23]. Despite the extensive

studies and development activities in the recent years aiming to optimise the properties and

performance of NFCs, current NFCs still cannot compete with glass fibre composites in terms of

mechanical performance [19]. In common with all engineering applications for NFCs, fibre

alignment should be optimised [22, 24] to suit the application. Unlike conventional continuous

reinforcement fibres, such as carbon and glass rovings, the natural fibres are of short length, being

INTRODUCTION

3

known as staple fibres in the textile industry. Usually, these natural fibres need to be processed into

continuous yarns using conventional spinning method before being made into directional

reinforcements for structural composite application, although research on unidirectional nonwoven

and woven preforms has been undertaken [17, 19]. Polylactic acid (PLA) is the most important bio-

thermoplastic. It shows also quite good properties, appropriate for applications that do not require

long-term durability or high mechanical performance at higher temperatures. The mechanical

properties of the PLA can be improved by using reinforcements like natural fibres in order to

increase its potential use in many industrial applications [25, 26]. The major attraction of these

biocomposites is that they are environmentally-friendly. These composites possess a high

biocontent and upon disposal can biodegrade with minimal harm to the environment, or at the end

of their life they can be recycled by energy recovery. Given the evident potential of the application

of thermoplastic biocomposites in high volume production, appropriate processing techniques are

needed. Due to the high melt viscosity of the thermoplastic matrix, the fibre impregnation is more

complex than for thermosets. Combining the reinforcement fibres and polymer at the fibre level

offers an improved distribution.

The idea of this research project was to modify the existing methods of commingled yarn

production from relatively short cellulose fibres and PLA (biopolymer) in order to improve

distribution and alignment of the fibres. The quality of the component distribution in the

commingled yarn affects the mechanical properties of the fabricated composites. The fibre

orientation (i.e. alignment of the fibres) must be controlled to ensure that the fibre mechanical

properties are efficiently utilised in order to attract industrial interest, as an alternative to the

traditionally applied synthetic fibres.

1.2 Aim of this study

The specific purpose of this thesis was to develop hybrid yarns with low fibre twist, optimised and

controlled fibre orientation for the fast processing, cost-effective and high performance

biocomposites suitable for use in load-bearing or semi-structural applications. I discussed the

development of two types of hybrid yarns: co-wrapped and aligned commingled yarns. The

influences of fibre content and wrap density on the properties of composites were reported. The

properties of composites, in terms of fibre orientation (aligned and random), off-axis angle and

alkali treatment were also investigated. The hybrid yarns were further processed into preforms in

the form of woven fabrics, which are consolidated and moulded in a single processing step. We also

studied the effect of two types of structures, 8-harness satin and basket, for the woven fabrics from

hybrid yarns on amount of porosity, moisture absorption, mechanical and thermo-mechanical

properties. In the context of this study, the mechanical characteristics of composites, in terms of the

reinforcing fibre characteristics, were examined. The natural (hemp) and man-made (Lyocell)

cellulose fibres were used. Different constituents can be used to tailor the composite characteristics

for diverse requirements, for instance, the high stiffness of the hemp fibres with the high elongation

at break and resin wetting and tenacity of the Lyocell fibres. Composites of PLA/hemp fibres,

PLA/hemp-Lyocell fibre mixtures, and PLA/Lyocell fibres were investigated.

INTRODUCTION

4

1.3. Outline of the thesis

The thesis is divided into six main chapters:

1. Chapter 1 introduces the motivation, justification and objectives of the research.

2. Chapter 2 reviews existing literatures related to the experimental work.

3. Chapter 3 explains the experimental work of the thesis.

4. Chapter 4 provides the summary of the obtained results.

5. Chapter 5 is the concluding chapter and discusses the results.

6. Chapter 6 suggests the avenues for future research.

LITERATURE REVIEW

5

2. LITERATURE REVIEW

2.1 Natural fibres

Textile fibre is a material mainly made from natural or synthetic sources. A general classification

for fibres is provided in Fig. 1.

Natural fibres generally can be classified as to their origin, that is, coming from plants, animals or

minerals.

Fig. 1 Classifications for Textile fibres

(Adapted and developed from Müeller et al. (2003) and Beldzki et al. (2002) [27, 28])

The plant-based fibres can be classified according to the part of the plant they are recovered from.

The precise composition of plant fibres, cellulose, hemicellulose, lignin, wax, etc. depends on

several factors such as the method of analysing, the growth environment and geographical location

of the plant and the level of maturity of the plant, etc.

2.1.1 Hemp

Hemp (Cannabis sativa L.) is a sustainable, multiuse, multifunctional, and high yielding industrial

crop, which can provide valuable raw material to meet the high-global demand for fibres. Hemp is

well known for requiring little or no pesticide use, and it needs low to medium quantities of

fertiliser [29]. The environmental friendly cultivation of hemp crop and the sustainability of its

products are the key drivers for future expansion of the hemp [30]. Currently, China, Europe and

Canada are the main hemp producing regions in the world [31]. Among natural fibres, bast fibres

have good mechanical properties, which make them an ideal candidate as the reinforcing media.

Hemp as a bast fibre has specific mechanical properties similar to the glass fibres, while it is

flexible. Furthermore, hemp fibres can be easily processed on the traditional opening and carding

LITERATURE REVIEW

6

machines. Many studies have been carried out in automotive market on olefin polymers reinforced

with hemp fibres. The above-mentioned reasons, along with industry wide acceptance, led to

selection of hemp fibres as reinforcement media in my research. Currently, hemp is the subject of a

European Union subsidy for non-food agriculture, and considerable strategies have been underway

for further development in Europe. Lear Corp., Johnson Controls and Ford are using hemp fibre for

reinforcing plastics in a number of automotive components [28].

2.1.2 Regenerated cellulose fibre

Polymers from renewable resources have gained attention due to their biodegradability and potential

for substituting petrochemicals in many applications. Cellulose, as a linear polysaccharide and the

most plentiful renewable polymeric material, has great properties and extensive applications.

Nonetheless, generally, processing of cellulose, natural polymer, is troublesome since it does not

melt or dissolve in common solvents because of its intramolecular or intermolecular hydrogen

bonds and crystalline structure. Furthermore, the traditional Viscose processes to produce cellulose-

regenerated materials result in critical environmental problems such as generating several

environmentally hazardous by-products, including CS2 and H2S. Thereby, additional facilities to

deal with the gaseous and aqueous waste emissions are required [32]. Accordingly, new

technologies have been investigated to avoid complex procedures and dangerous by-products [33].

Cellulose can be dissolved in solvents such as N-methylmorpholine-N-oxide (NMMO) [33],

dimethyl sulfoxide (DMSO)/paraformaldehyde (PF), N,N-dimethylacetamide (DMAc)/lithium

chloride (LiCl) and alkali solutions [34] as well as ionic liquids (ILs) [35]. Of these solvents, only

the use of NMMO in the process of commercialisation has resulted in technical success, namely, a

new man-made cellulose fibre: Lyocell [33, 36].

Lyocell is the first in a new generation of cellulosic fibres [37]. Lyocell is a 100% cellulosic fibre

made by a solvent spinning process. Demand for a process that is environmentally responsible and

utilises renewable resources as the raw materials is a major driving force for the development of

Lyocell fibre. Lyocell is derived from wood pulp, which is produced from sustainable managed

forests. The pulp is dissolved in NMMO, which is not classified as corrosive or toxic and is

biodegradable. Then the solution is spun to fibres and the solvent extracted as fibres pass through a

washing process [38].

After the fibre formation, the solvent is recovered from the fibre via multiple washing steps. Then it

is purified and returned to the concentration needed to dissolve cellulose, for reusing at the

beginning of the process. This is so called closed-loop technology. By paying careful attention at

each stage of recycling, the solvent recovery rates in excess of 99.5% can be achieved [38].

Lyocell is fully biodegradable. It has a relatively high strength, which facilitates its use in various

mechanical and chemical finishing treatments. The physical characteristics of Lyocell fibres aid in

its great blending characteristics with fibres such as silk, wool, linen and cashmere [37]. Lyocell

fibres are known to have good mechanical properties, wettability, high tenacity and good

drapability. Therefore, they have potential for applications such as reinforcements for composites

and can improve mechanical and physical properties [39].

LITERATURE REVIEW

7

2.2 Fibre surface modifications

The quality of the fibre–matrix interface is one of the most important factors, which control the

performance of the composite materials. In order to achieve optimum performance of the

biocomposites, an adequate degree of adhesion between the surface of hydrophilic cellulosic natural

fibres and the matrix is preferred. Since the interface determines the reinforcement efficiency of a

composite, strong bonding between matrix and fibre is necessary [22]. Because of the presence of

hydroxyl and other polar groups in several constituents of natural fibres, especially in the

amorphous region, the moisture uptake of bio-based composites is high, which causes weak

interfacial bonding between matrix and fibres, swelling and a plasticising effect, resulting in

dimensional instability and poor mechanical properties. Moreover, the presence of a natural waxy

substance on the natural fibre surface contributes to poor fibre-matrix bonding and ineffective

surface wetting.

Dewaxing, alkali treatment, isocyanate treatment, peroxide treatment, vinyl grafting,

cyanoethylation, bleaching, peroxide treatment, acetylation and treatment by coupling various

agents have achieved various levels of success in improving fibre–matrix adhesion in natural fibre

reinforced composites [40].

2.2.1 Alkali treatment

The optimum alkali treatment is an effective and low cost surface modification method for natural

fibres. The reaction of sodium hydroxide (NaOH) with natural fibre (Cell-OH) occurs as shown

below:

Cell − OH + NaOH → Cell − O−Na+ + H2O + surface impurities

During alkali treatment, a certain amount of lignin, impurities such as wax and natural oil covering

the external surface of the natural fibre cell wall may be removed; the cellulose structure will be

depolymerised and the short length crystallites will be exposed. A high alkali concentration may

depolymerise the cellulose, which can result in unfavourable effect on the fibre strength. Therefore,

the concentration of NaOH solution, treatment temperature and time play a vital role in obtaining

the optimum efficiency of the fibre.

Changes in performance of natural fibre composites as a result of alkali treatment have been

reported by many researchers [41-45]. The alkali treatment results in better fibre–matrix adhesion

because of the increase in the rough surface topography and fibrillar formation providing additional

sites of mechanical interlocking and more resin-fibre interpenetration at the surface. Furthermore,

this treatment leads to a decrease in the microfibril angle and an increase in degree of molecular

orientation. With removal of hemicellulose as a result of alkali treatment, the inter-fibrillar region

will be less dense and rigid, which makes the fibrils more capable of reorganising themselves along

the direction of tensile deformation [42].

The influence of alkali treatment on crystallisation of natural fibres has been studied by several

researchers [46, 47]. Alkali treatment results in better and closer packing of the cellulose chains or

rather increased crystallinity.

LITERATURE REVIEW

8

2.3 Biodegradable polymers

2.3.1 Polylactic acid (PLA)

There are plenty of thermoplastic polymers derived from renewable raw agricultural materials,

which are available to meet the ecological purpose of biodegradability. Polylactic acid, known as

PLA, is one of the fully biodegradable polymers that is commercially available and has received

particular attention for biocomposites manufacturing [48]. PLA is probably the most promising

matrix becuase its properties are close to the most widespread thermoplastic matrix of

polypropylene (PP) and polyethylene terephtalate (PET). PLA has many favourable markets due to

its biodegradability, making it one of the best candidates for substitution of PP and PET in some

applications. Despite the numerous advantages, however, there are considerable drawbacks which

still need to be overcome, such as its relatively poor impact properties arising from its inherent

brittleness, but also the moisture sensitivity, limited supply and higher cost of PLA, compared with

commodity polymers such as PET and PP [49]. PLA is commercially interesting because of its film

transparency, biocompatibility and good mechanical properties. Apart from several areas of

application in the packaging industry [1,2], currently PLA is applied as matrix in fibre reinforced

composites.

2.4 Biocomposites

Biocomposites are composite materials made from natural/bio fibre and petroleum based plastic or

renewable resource based plastic. Composites produced from synthetic fibres and biopolymers

could also be called biocomposites. The biocomposites derived by combining plant derived fibres

with bio-derived plastic are likely to be more eco-friendly [50, 51], thus, reducing the environmental

challenges. Being environmentally-friendly and sustainable are the major attractions for 100%

biocomposites. These biocomposites are suited for simply disposing or composting at the end of

their life without damaging the environment or can be recycled by energy recovery. These

composites are defined as composites that are designed with the lowest possible environmental

footprint [22].

Some of the biodegradable matrices (natural and synthetic) that can be applied for making

biocomposites are listed in Table 1 [52, 53].

LITERATURE REVIEW

9

Table 1 Biodegradable polymer matrices (Adapted and developed from (Stevens, 2002) [53])

Biodegradable polymers

Natural Polysaccharides: starch, cellulose, Chitin

Proteins: Collagene/gelatin, Casein, albumin, fibrogen, Silks

Polyesters: Polyhydroxyalkanoates

Other polymers: Lignin, Lipids, Shellac, Natural rubber

Synthetic

Poly(amides)

Poly(anhydrides)

Poly(amide-enamines)

Poly(vinyl alcohol)

Poly(vinyl acetate)

Polyesters: Poly(glycolic acid), Poly(lactic acid), Poly(caprolactone), Poly(orthoesters)

Poly(ethylene oxides)

Poly(phosphazenes)

2.4.1 Hybrid composites

One way of deliberately regulating the properties and performance of composite materials is to use

hybrid fibrous fillers (synthetic or natural fibre) into a single matrix. It will give these composite

materials specific properties, and allow for the expansion of the area of their application. Its

functionality depends on the balance between the properties of each individual component. The

properties of a hybrid composite depend directly on the fibre content, length of fibres, arrangement

and orientation of both the fibres and also fibre-matrix bonding [52]. Many interesting studies have

been conducted to explore different hybrid composites with different applications [54-56]. The

impact and tensile behaviour of epoxy resin reinforced by oil palm/glass fibre was investigated by

Bakar et al. (2005) [57]. Mishra et al. (2003) investigated the effect of the hybridisation of

sisal/glass and pineapple/glass-polyester composites on the moisture uptake characteristics of

composites. The authors observed that water absorption of the hybrid composites were less than that

of unhybridised composites [58]. In an interesting research, the hybrid biocomposites were made by

pineapple and sisal fibre reinforced thermosets and thermoplastics resins. The authors dealt with the

structure, composition and properties of hybrid biocomposites [51]. The mechanical performance of

short, randomly-oriented, sisal-banana hybrid fibre reinforced polyester composites was

investigated. A positive hybrid effect was observed in the flexural and tensile strength and modulus

of the hybrid composites [59]. The tensile properties of kenaf fibre/wood flour-polypropylene

hybrid composites were investigated by Mirbagheri et al. (2007) [60].

The possibility of improving the outdoor properties of NFCs needs to be investigated. Therefore,

studies on improving the mechanical properties and specifically reducing the water absorption in

NFCs need to be done. Making a composite system consisting of two different fibres with different

morphologies may be a good solution. The poor mechanical properties of bast fibre reinforced

composites can be improved by admixing cellulose fibres to bast fibres. Mieck et al. [61] studied the

effects of the admixture of Lyocell fibres with flax fibre reinforced polypropylene composites on

the mechanical properties. Further investigation was carried out by Graupner (2009). Composites of

the fibre mixtures of hemp/Lyocell were investigated [62].

LITERATURE REVIEW

10

2.4.2 Thermoplastic composites

Several fabrication methods have been developed for manufacturing thermoplastic biocomposites.

Film stacking [63, 64], injection moulding [14, 65] and compression moulding [14, 48, 66] are the

most widely applied production methods for fibre reinforced composite materials.

Many thermoplastic composites frequently suffer from inadequate fibre–matrix adhesion.

Furthermore, the use of thermoplastics introduces the problem of lack of adequate resin penetration

into the fibre tow. Due to the high viscosity of thermoplastic, it is very challenging to inject the

resin into a tightly woven textile structure and to fill its pores generated by the fibre interlacement.

This problem increases the porosity content in the composites and can be overcome by using high-

pressure injection pressure and heavier moulds. De-alignment of reinforcing fibres is another

problem caused by high matrix viscosity during consolidation. Reducing the melt flow distance

during the consolidation process can alleviate these problems. Mixing the fibres with the matrix

even before the preforming operation is a way to prepare composites with satisfactory matrix

dispersion within the fibre tows. There are several methods such as hot melt, solution, slurry,

emulsion, film, surface polymerisation, powder coating and commingling. Out of these techniques,

powder coating and commingling have the potential for producing prepregs with substantial

flexibility, being a critical necessity for textile processing [67].

Thermoplastic commingled/hybrid yarns are one of the possible intermediate products, so-called

“dry prepregs”, applied for continuous fibre reinforced thermoplastic composites. The commingled

yarns have emerged as a cost-effective method for manufacturing the different parts, especially

complex-shaped ones. The hybrid yarn has reinforcing and matrix-forming fibres combined together

in order to reduce the problems associated with high melt viscosity of thermoplastic matrices;

therefore, polymer flow distance for impregnation is reduced and the time needed for impregnation

or high-applied pressure are limited, leading to a reduction in production cost. Moreover, the yarns

have enough flexibility to go through textile preforming without getting damaged considerably, and

they can easily be converted into drapable woven fabrics being suitable for forming complex

shapes.

In this hybrid yarn, by applying heating the thermoplastic component melts and impregnates the

reinforcing component (fibre). After cooling, the system is transformed into a rigid composite [68].

To benefit from the mentioned advantages of hybrid yarn, the process parameters for yarn

production and consolidation should be optimised. Homogeneous distribution of reinforcing fibre

and polymer fibres or commingling quality is one of the most critical parameters since it governs

the consolidation quality, leads to a fast and complete impregnation of reinforcing fibres, and also

affects the amount of void content in the manufactured composites [9].

2.4.2.1 Types of hybrid yarns

Hybrid yarns can be prepared by different methods including co-wrapping, core spinning and

commingling.

LITERATURE REVIEW

11

In co-wrapping method, thermoplastic fibres are wrapped around a core of twistless reinforcing

fibres and thermoplastic filaments (Fig. 2), which brings individual filaments together and

consequently increases the structural integrity of the prepregs and also provides better protection for

the reinforcing fibres during further processing such as weaving or braiding [69]. However, the yarn

has poor impregnation due to inhomogeneous distribution of the reinforcing and matrix yarns and

requires higher pressure and processing temperature [70].

Fig. 2 Structure of wrap spun hybrid yarn

In core spinning (friction spinning), short staple fibres of thermoplastic matrix material are spun

around a core of high-performance continuous reinforcing fibres, as represented by Fig. 3. The

properties of these yarns are similar with those of the co-wrapped yarns. These yarns are very

flexible and make further processing easier [71].

Fig. 3 Structure of friction spun hybrid yarn

In commingling, the matrix and reinforcing fibres are mixed in a nozzle with the help of

compressed air. Fig. 4 shows the schematic of the structure of the commingled yarn. Among these

hybrid yarns, commingled yarn gives a good blending of matrix and reinforcing fibres; however,

this yarn tends to de-mingle under load and during preforming operations [72]. This process

provides soft, flexible and drapable yarn [73].

Fig. 4 Structure of commingled hybrid yarn

LITERATURE REVIEW

12

Apart from the above methods, thermoplastic hybrid yarns can be produced by the yarn

manufacturing processes such as rotor and ring spinning. However, the existing spinning techniques

cannot be applied to produce hybrid yarns without modifications at the process level [68].

The main application of natural fibres is today mainly in non-structural composites as they are

mostly available as randomly-oriented nonwovens [74-76]. The fibre orientation (i.e. alignment of

the fibres) must be controlled to ensure that the fibre mechanical properties are efficiently utilised in

order to attract industrial interest as an alternative to the traditionally applied synthetic fibres (e.g.

glass fibres). Previous studies have demonstrated that the full reinforcement potential of natural

fibres can be exploited in bio-composites if an aligned fibre orientation is used [17]. Natural fibres

are naturally discontinuous; therefore, natural fibre reinforcements reported so far are based on

twisted spun staple yarns, which are produced by spinning methods, mainly ring spinning. These

spun yarns tend to be highly twisted, which leads to fibre misalignment due to their helical paths

around the yarn axis. This misalignment contributes negatively to the mechanical properties of the

resultant composites. Another negative impact of yarn twist is that it tightens the yarn structure,

rendering resin impregnation difficult [77]. Therefore, in the textile industry, a broad range of

techniques for the alignment of natural fibres have been developed and optimised to produce yarns

with controlled fibre orientations by reducing or replacing twist in yarns. Goutianos et al. [77] tried

to produce flax yarns with the minimal level of twist for manufacturing aligned composites. Shah et

al. [78] used a sizing agent to substitute the use of twist in roving and yarn. Zhan et al. [17] studied

the effect of wrapped spun yarn with low twist for reinforcement purpose.

2.4.2.2 Textile processing of hybrid yarn

Textile production technology makes it possible to use hybrid yarns as the basis for obtaining an

intermediate prepreg, such as in the form of woven, weft knit or warp knit fabrics with the needed

properties. With subsequent processing under pressure and at elevated temperature, the melting

thermoplastic fibre component is converted into a polymer matrix, filling the mould and

impregnating the reinforcing fibres. This technology offers new possibilities for a shortened

production process for thermoplastic composites and articles made from them [79].

Fabric weaves are used in several applications in fibre reinforced polymers due to their good

mechanical properties, such as stiffness, strength and dimensional stability. With the weaving

technology, it is possible to fabricate high density woven structures with load-oriented fibre

positioning [80, 81]. The strength and stiffness of fabric reinforced composites are influenced not

only by the matrix and yarns properties, but also the structural parameters of materials such as

fabric count and weave pattern. The fabric count determines the number of weft and warp yarns per

cm, whereas the weave pattern specifies how the warp and the weft yarns are interlaced. Typical

weave patterns are plain, twill, basket and satin [82]. Fabrics that are produced by weaving a large

number of thick yarns such as warp and a few numbers of thin yarns as weft are called

unidirectional and are used in unidirectional composites. This weave type provides materials with

good processability, high stiffness and strength in the warp direction, which is specific for

unidirectional composites [83].

LITERATURE REVIEW

13

The structure and properties of the hybrid yarn in textile preforms has significant effect on preform

properties such as drape, flexibility, permeability and thickness. Ultimately, these properties are

important when making the composites from textile preforms.

2.4.3 Engineering applications of biocomposites

Over the last five decades, considerable research has been carried out in the field of bio-based

plastics and biocomposites, which demonstrates their significance. Nevertheless, part of product life

cycle is research and development. The actual engineering process begins when the developed

science is applied to a specific application. Bio-based composites offer great opportunities for an

increasing role as alternate material. The use of biocomposites has extended to almost all fields

including packaging, biomedical, building, civil, construction and automotive industry, etc. [84].

Some of the building and construction applications where bio-based composites are potentially used

include window frames, framing, fencing, walls, doors, flooring and ceiling panels. Roof structures

form biocomposites were manufactured from soy oil-based resin and cellulose fibres in the paper

sheets form, being made from recycled cardboard boxes [85]. Pulp from residual sisal and coir

fibres and eucalyptus waste has also been used as a replacement for asbestos in roofing components

[86]. The use of bamboo fibre as reinforcement in structural concrete elements has been studied

[87]. Building temporary housing can be another potential application of biocomposites. Usually,

temporary housings from wood plastic composites are set-up whenever a catastrophe occurs. When

the situation recovers to normalcy, the temporary housings are disassembled and the waste is put in

landfills. Thereby, to reduce this type of landfill waste, such applications necessitate applying the

biocomposites and more definitely green composites, which can possibly be composted or recycled

by energy recovery after their service life [84]. In the automotive industry, polyester matrix

reinforced by cotton fibres was used in the body of the East German ‘‘Trabant’’ car [88]. Another

example is the use of flax fibres in car disk brakes as a replacement for asbestos fibres. Mercedes

used jute-base reinforced door panels for their vehicle production in 1996. The bast fibres are

mainly used in automotive industry since they exhibit greatest strength and weight savings of

between 10% and 30% and corresponding cost savings. Many recent studies have shown that

thermoplastics reinforced by hemp fibre mat are promising candidates in automotive industries

where high specific stiffness is required [89]. Almost all the major car companies (Audi Group,

BMW, Mercedes, Volkswagen, Ford and Opel) now use bio-based composites in several

applications such as interior trim components like dashboards and door panels. For example, in

2000, Audi launched the A2 midrange car in which door trim panels were manufactured from

flax/sisal mixture mat reinforcing polyurethane.

LITERATURE REVIEW

14

[Type a quote from the document or the summary of an interesting point. You can position the text box

anywhere in the document. Use the Drawing Tools tab to change the formatting of the pull quote text box.]

EXPERIMENTAL

15

3. EXPERIMENTAL

3.1 Materials

3.1.1 Reinforcements

Hemp in the form of baled loose staple fibres (genus species Cannabis Sativa L) was supplied by

Hempage AG (Adelsdorf, Germany). According to the manufacturer’s information, the average

diameter of the hemp fibre was 20–40 µm, and the mean fibre length was 30 mm (Publications I, II,

III, IV and V).

The Lyocell staple fibres were supplied by Lenzing AG (Lenzing, Austria). The average length and

diameter of the fibres were 38 mm and 13.4 µm, respectively (Publications IV and V).

3.1.2 Matrices

Two different types of PLA were used as the matrix in the hybrid reinforcements:

A PLA staple fibre, provided by Trevira GmbH (Hattersheim, Germany), had a fineness of 1.7 dtex

and a mean fibre length of 38 mm. Based on the manufacturer’s information, the PLA fibres were

made from Nature Works®

PLA 6202D from, Cargill Dow LLC (Minnetonka, USA). This

thermoplastic polymer has a density of 1.24 g/cm3, a melt temperature of 160–170 °C, and a glass

transition temperature of 60–65 °C (Publications II, III, IV and V).

An air-textured PLA continuous yarn with 72 filaments was provided by Torcitura Lei-Tsu S.r.l

(Bergamo, Italy). According to the supplier, the yarn was made from the PLA grade 6201D from

Nature Works®

, Cargill Dow LLC (Minnetonka, USA). This thermoplastic has a glass transition

temperature of 55–60 °C (Publications I, II, III and V).

3.1.3 Chemicals

Reagent grade NaOH was obtained from Scharlau and Sigma Aldrich and used for fibre alkali

treatment (Publication II). Absolute ethanol (> 99.8%) was supplied by Sigma Aldrich (Publications

II, III, IV and V) and used for porosity measurement test.

3.2 Methods

3.2.1 Fibre alkali treatment

The hemp fibres were first treated by 4 wt% sodium hydroxide (NaOH) solution for 1 h, then rinsed

with distilled water until it was neutral and finally dried at room temperature for 48 h (Publication

II).

3.2.2 Manufacture of hybrid yarns

3.2.2.1 Co-wrapped hybrid yarn

The PLA/hemp hybrid yarns were produced by using a laboratory yarn twist machine from

DirecTwist, AGTEKS Co. Ltd. (Istanbul, Turkey). In the co-wrapping, the PLA yarn filaments were

wrapped around low twist and very fine hemp yarns. These PLA filaments will melt during the

consolidation process and will be incorporated in the matrix. Thus, they need to be very fine yarns

so that the interfacial adhesion between the reinforcing hemp yarns and the PLA matrix is not

EXPERIMENTAL

16

affected to any significant extent. However, inhomogeneous distribution of the reinforcing hemp

and the PLA matrix yarns may lead to poor impregnation in the consolidated composite; therefore,

it was decided to put PLA filaments in the core part of the hybrid yarn together with the hemp yarn.

The core yarn was assembled by putting aligned hemp yarns and PLA filaments into a bundle

without increasing the hemp yarn twist. The yarn structure obtained from the co-wrap spinning used

is shown in Fig. 5 (a, b and c). The wrap density, or number of wraps per unit length (1 m), ranged

from 150 (Fig. 5 b) to 250 (Fig. 5 c) in order to study the effect of wrap density on the properties of

the composite. The name of the hybrid yarn indicates the composition. For example, in the hybrid

yarn PLA/45 Hemp-150, one PLA filament is wrapped 150 times around 1 m of the core part

(composed of four hemp yarns and two PLA filaments), thus, giving a nominal ratio of 45 mass%

hemp fibre content in the hybrid yarn. The linear densities of the hybrid yarns were determined by

weighing a 10 m long yarn sample that was dried in a vacuum chamber (0.9 mbar; 20 °C) for at least

18 h, and calculating the tex value.

Fig. 5. a) Structure of co-wrapped hybrid yarn, b) yarn with 150 wrapping density and c) yarn with 250 wrapping

density (Publication I)

3.2.2.2 Wrap spinning of hybrid yarns

Previous studies have demonstrated that the full reinforcement potential of reinforcing fibres can be

exploited in composites if an aligned fibre orientation is applied [16, 17]. Therefore, the yarn

structure, which has the reinforcing fibres straight and parallel to the yarn axis, is preferred. Wrap

spinning can be used to produce such a yarn [17]. PLA/hemp hybrid wrap spun yarns were

produced by using a laboratory spinning machine from Mesdan S.p.A. (Brescia, Italy) and a

laboratory yarn twist machine from DirecTwist, AGTEKS Co., Ltd. (Istanbul, Turkey). The hemp

and PLA fibres arrived at our laboratory in baled loose fibre form. The PLA fibre and the hemp

fibre were weighed to the desired proportion (30 mass%), and the fibre mixture was then fed into

the carding machine. The blended PLA/hemp web was carded three times to parallelise the fibres

a)

b)

c)

EXPERIMENTAL

17

and achieve sliver uniformity. Then the sliver was fed through a roving frame, where the strands of

fibre were further elongated. The sliver was drawn twice after carding to achieve the required

roving linear density. Although it was possible to create hybrid yarns with low twist, the cohesion of

the fibres was very low because PLA/hemp roving had a false twist, which means that they could

not form a roving of sufficient integrity. Moreover, the roving is too weak to be able to be collected

alone in the roving machine. In order to collect the roving without causing breakage in the roving

machine, the processable PLA filaments were used as a processing carrier for the PLA/hemp roving

in the final step. After the spinning, the roving was wrapped by PLA filaments in the twisting

machine. The wrap yarn was spun to the nominal count of 550 tex, and it had a wrapping intensity

of 200 turns/m. These wrappings provide better protection for the reinforcing fibres during further

processing, such as weaving [68] or making a prepreg. The yarn structure obtained from the wrap

spinning used is shown in Fig. 6.

Fig. 6. Structure of wrap spun hybrid yarn (Publications II, III and V)

3.2.3 Preparation of prepregs

3.2.3.1 Nonwoven fabric manufacturing process

The fibres in their loose form were manually mixed and fed to a carding machine from Mesdan

S.P.A (Brescian, Italy). The blended fibre web was carded three times to parallelise the fibres and to

achieve mat uniformity (Publication IV).

3.2.3.2 Bidirectional hybrid yarn prepreg

Before compression moulding, multilayer 0/90 bidirectional hybrid yarn prepregs were prepared by

winding the hybrid yarn around a 19×19-cm steel rectangular frame (Fig. 7). A 0/90 bidirectional

layup of the prepreg was chosen, due to the ease of the winding technique. Each prepreg had 12

layers of wound yarns, to achieve a composite thickness of about 1–3 mm (Publication I).

EXPERIMENTAL

18

Fig. 7 Prepreg frame with the PLA/hemp hybrid yarns in a 0/90 configuration (Publication I)

3.2.3.3 Unidirectional woven fabrics from hybrid yarns

In order to manufacture samples of thermoplastic composites, we made woven unidirectional (UD)

fabrics from hybrid yarns and PLA filaments on a handloom weaving machine. The obtained woven

fabrics had the PLA/hemp hybrid yarn in the transverse direction of manufacture of the fabric (weft

direction) and were supported by the 18 tex PLA filaments as the warp yarns in the longitudinal

direction (8 threads per cm). The fabrics were woven in two different weave patterns: 8-harness

satin (4 threads per cm) and basket weave (3 threads per cm) as seen in Fig. 8 a and b. These fabrics

have similar characteristics as unidirectional fabric in the weft direction, as the thin warp yarn,

reduces the crimp considerably. The woven fabrics obtained can be considered as prepregs for

subsequent processing to make composites by compression moulding.

Fig. 8. Schematic diagram of a) 8-harness satin and b) basket weave patterns (Publications III and V)

a)

b)

EXPERIMENTAL

19

3.2.4 Composite preparation

The prepreg mats were first put in a vacuum chamber (0.9 mbar; 70 °C) for at least 18 h, to dry

before the compression moulding. The prepreg was covered with a Teflon sheet to prevent sticking

of the matrix to the surface of the mould before being placed into a pre-heated steel mould with a

20×20-cm square cavity, and 10 mm depth. The steel mould was of our own design and machined

by a local machine shop. The composites were then prepared by pressing the prepreg at a

temperature of 175 °C and a pressure of 1.04 MPa for 15 min (Publication I), 195

°C and a pressure

of 1.7 MPa for 15 min (Publications II, III and V) and 195 °C and a pressure of 1.7 MPa for 20 min

(Publication IV) in a hydraulic compression moulding machine from Rondol Technology Ltd.

(Staffordshire, UK). Neat PLA sheets to be used as reference material were made by compression

moulding under the same processing conditions of the same PLA filaments formed to a similar

prepreg.

3.3 Characterisation

Before performing the testing, the specimens were conditioned for at least 24 hours at 23 °C and

50% relative humidity, according to DIN EN ISO. Specimens for different tests were cut by GCC

LaserPro Spirit laser cutting machine, according to the ISO test standards.

3.3.1 Single fibre tensile test

All staple fibres were tested on a Favigraph single fibre tensile tester from Textechno GmbH

(Mönchengladbach, Germany) equipped with a 20 cN load cell. The test speed was 20 mm/min, and

the gauge length was 20 mm and 20 replicated tests were done for each fibre type. The single fibres

were treated in a hot press at 190 °C and 1.7 MPa for 15 min in order to simulate the effect of

composite processing conditions (presented in Section 3.2.4) on the Lyocell and hemp

characteristics and then tested for tensile properties.

3.3.2 Composite density and porosity

Composite density was evaluated in order to assess the volumetric composition. Composite

densities were determined by the buoyancy method (Archimedes’ principle) using ethanol as the

displacement medium. To avoid absorption of liquid during immersion, the specimens were covered

by a varnish containing paraffin. Based on the data obtained regarding densities and mass fractions,

the volume fractions of fibre, matrix and porosity were calculated by using the proposed method by

Madsen et al. [5].

3.3.3 Water absorption test

Water absorption analysis was done on the composite specimens according to ASTM D570-98. The

specimens were dried in an oven for 24 h at 60°C, and then cooled to room temperature in a

desiccator. The initial weight of these specimens was denoted as W0. The specimens were then

immersed in two different baths: one with distilled water at room temperature and one with distilled

water bath at 80°C. The amount of water absorbed was measured every 24 hours for 10 days. At

each measurement, the specimen was taken out of the water and the surface wiped dry before the

EXPERIMENTAL

20

weight was recorded as W. For each composite, five replicate samples were measured. The

percentage of apparent weight gain (WG) was then calculated using the following equation.

WG = [(W-Wo)/Wo] × 100 % (1)

3.3.4 Tensile testing

The tensile testing was performed according to the ISO 527 standard test method, using a universal

H10KT testing machine equipped with a model 100R mechanical extensometer, both supplied by

Tinius Olsen Ltd. (Salford, UK). The loading rate was 10 mm/min and the load range was 1 kN. Six

specimens were analysed for each composite laminate. The specimens were cut in 20 mm wide

dumbbell shape with an overall length of 150 mm and a 60 mm length of the narrow, 10 mm wide

parallel-sided portion. The gauge length was 50 mm.

3.3.5 Flexural testing

The flexural testing was performed according to ISO 14125, with the same testing machine as for

tensile testing. At least five specimens were tested for every composite composition. The loading

rate was 10 mm/min and the load range was 1 kN. The sample dimensions were: length 60 mm,

width 15 mm, while the thickness varied depending on the sample.

3.3.6 Impact testing

Charpy impact testing was done on un-notched specimens in accordance with ISO 179 using a

pendulum type Zwick test instrument from Zwick GmbH (Ulm, Germany). A total of 10 specimens

were tested edgewise to determine the mean impact resistance.

3.3.7 Dynamic mechanical thermal analysis

DMTA was done using a Q-series TA instrument supplied by Waters LLC (Newcastle, DE, USA).

The analysis was run in the dual cantilever bending mode and the sample dimensions were: length

50 mm, width 8 mm, while the thickness varied depending on the sample. The temperature interval

was 15–150 °C with a heating rate of 3 °C /min using a frequency of 1 Hz and amplitude of 15 lm.

3.3.8 Differential scanning calorimetry

DSC analysis was done on a DSC Q2000 supplied by TA Instruments (New Castle, DE, USA).

Samples of 8–10 mg were heated in a nitrogen-purge stream at a rate of 10 °C/min from 20 to 200

°C, cooled to 0 °C, and then heated again from 20 to 200°C. The data from the first scan were used

in order to see the effect of processing conditions on the thermal properties. For each PLA sample,

three replicates were scanned. The final results were the averages of three DSC runs. The

percentage crystallinity (XDSC) of PLA was calculated using the following equation [90]:

XDSC %= ∆𝐻𝑓−∆𝐻𝑐𝑐

∆𝐻𝑓𝑜 ×

100

𝑤 (2)

Where, ∆Hfo = 93 J/g for 100 % crystalline PLA, ∆Hf is the enthalpy of melting, ∆Hcc is the cold

crystallisation enthalpy and w is the weight fraction of PLA in the composite.

EXPERIMENTAL

21

3.3.9 Scanning electron microscopy (SEM)

The fibre surface and the composite fracture surface obtained from the tensile testing were studied

using low-vacuum scanning electron microscopy with a Quanta 200 ESEM FEG instrument from

FEI, (Oregon, USA) with an operating voltage of 5–15 kV. For low vacuum imaging, no specific

preparation was required.

EXPERIMENTAL

22

[Type a quote from the document or the summary of an interesting point. You can position the text box

anywhere in the document. Use the Drawing Tools tab to change the formatting of the pull quote text box.]

SUMMARY OF RESULTS

23

4. SUMMARY OF RESULTS

4.1 Effect of heat and alkali treatments on fibre properties (Publications II and IV)

In order to evaluate the effect of moulding conditions, the mechanical properties of heat and alkali

treated single hemp fibres were determined, see Table 2.

The alkali treatment improves fibre tensile strength and modulus, which helps to improve the

properties of the composites [91, 92]. This treatment possibly orients fibrils along the direction of

tensile forces by removing hemicelluloses from fibre (Fig. 9), resulting in better load sharing

between the fibrils [93].

Heat treatment greatly reduces mechanical properties. Actually, cellulosic fibres are mixtures of

organic materials, and heat treatment at elevated temperatures can cause physical and chemical

changes. The physical changes are associated with enthalpy, weight, colour, strength, crystallinity,

and orientation of microfibril angle [94]. The chemical changes are related to the decomposition of

several chemical elements. Heat treatment results in the weight loss of moisture plus weight loss

due to thermal degradation. The thermal degradation of cellulosic fibres results in change in colour

and deterioration in mechanical properties of the fibres [94-96].

Table 2 Tenacity, modulus, elongation and linear density for hemp and Lyocell fibres before and after heat and alkali

treatments (Publications II and IV).

Treatment Fibre Tenacity

cN/dtex

E-Modulus

cN/dtex

Elongation

%

Linear

density

dtex

As received Hemp 4.66 (1.43) 97.71 (24.92) 3.93 (0.66) 4.15

Lyocell 3.27 (0.63) 55.87 (23.06) 8.70 (2.01) 1.36

After hot pressing Hemp 3.75 (1.10) 84.62 (26.60) 3.37 (0.88) 4.27

Lyocell 2.12 (0.92) 48.46 (20.29) 4.75 (1.84) 1.38

After alkali treatment Hemp 5.18 (1.30) 99.15 (32.64) 4.05 (0.86) 3.75

SUMMARY OF RESULTS

24

Fig. 9 SEM micrographs of the surface morphology of: (a) untreated and (b) alkali treated hemp fibres (Publication II)

4.2 Composites properties (Publications I, II, III, IV and V)

4.2.1 Composite porosity

Porosity is one of the most critical issues during the manufacturing process of composites. The

presence of porosities can significantly reduce the tensile, compressive, inter laminar shear and

structural strengths of a composite [97].

In Publication I, co-wrapped hybrid yarns were produced from hemp and PLA yarns in four

nominal mass ratios: 10/90, 20/80, 35/65 and 45/55 with two wrapping densities: 150 and 250

wraps/metre. In total, eight composites were manufactured with fibre volume fractions in the range

8–38 vol.% and with porosity in the range 6–9 vol.%. The physical composition of the composites

made are summarised in Table 3. From the results, it is possible to conclude that the higher the fibre

volume fraction, the higher the amount of porosity, which is obviously due to the physical structure

of the prepreg. Similar trends have also been seen in other studies [5, 98]. Furthermore, it appears

that the use of lower wrapping density causes a slight increase in the porosity. In the prepared

composite, the porosity fraction was still high even though the fibre volume fraction was low, which

could be due to several factors. Porosity is difficult to avoid in natural fibre composites and

influences on the composite properties. Thus, research on how to decrease the amount of porosity is

warranted.

In Publication II, we describe the developed natural fibre hybrid yarns with low twist, for use in

high performance composites with lower amount of porosity. The composites were produced from

PLA/hemp yarn prepregs with three off-axis fibre orientations (0°, 45° and 90°) and PLA/alkali

treated hemp yarn prepregs in a nominal mass ratio of 30 mass% reinforcement. Because the fibre

angle does not have any effect on the composite composition, only the composite of 0° off-axis

fibre was tested. The formation of a stronger interface by improved hemp fibre and PLA adhesion

could be the result of the reduced porosity of the composite from alkali treated hemp composites,

compared with the untreated hemp/PLA composites. This can be explained by the fact that alkali

treatment increases the available hydroxyl groups [91] due to removal of the noncellulosic materials

a) b)

SUMMARY OF RESULTS

25

covering the cellulose hydroxyl groups and cleans the surface impurities that lead to the formation

of the rough fibre surface, which would largely increase the surface area of the fibre [99].

In Publication III, composites were produced from satin PLA/hemp hybrid fabrics and basket

PLA/hemp hybrid fabrics in a nominal mass ratio of 30 mass% hemp reinforcement in order to

investigate the effect of different weave patterns on the properties of the composites. From the

results, it is possible to conclude that the composite made by unidirectional satin fabric gave

significantly lower porosities, compared to the winded hybrid yarn laminates (composite described

in Publication II). The porosity content for the composite made from satin fabric (0.96 %) was

clearly lower, compared to the composite made from basket fabric (4.55 %). It can probably be

attributed to the fact that satin weaves allow individual fibres to be woven in the closest proximity

and can therefore produce fabrics with a close ‘tight’ weave, compared to basket weave. Moreover,

basket weave has higher quantity of interlace points per unit area, compared to satin. Perhaps voids

might be present between the interlace points that are higher in basket systems than those of satin

[100].

In Publications IV and V, are composites produced from unidirectional satin hemp/PLA,

Lyocell/PLA and Lyocell-hemp/PLA fabric and the nonwovens discussed. All the composites had a

nominal mass ratio of 30 mass% reinforcement in order to investigate the effect of mixing different

fibres and their alignment on the composites properties. From the results, it is clear that the

composite made by unidirectional satin fabric gave significantly lower porosities, compared to the

nonwoven laminates [101, 102]. The tortuosity of the resin flow path is an important factor in the

process of void movements and their elimination. The nonwoven mat fibre has a more tortuous flow

path, compared to the composite made from hybrid yarn prepregs;, hence, it is more difficult for the

entrapped air to flow out of the system. Moreover, it can be observed that the Lyocell-based

composite had a lower void content, which is obviously due to the higher interfacial adhesion.

4.2.2 Water absorption

The apparent weight gain (WG) as a function of time at different temperatures for the manufactured

composites are shown in Fig. 10 a, b and c. It can be observed that the fibre-based composites had

significantly higher water absorption than neat PLA due to the hydrophilic nature of hemp and

Lyocell with polar groups such as hydroxyl and carboxyl groups on the fibre surfaces. Among all

composites, the composite from PLA/hemp nonwoven showed the highest water absorptions. It can

probably be attributed to the fact that it has less close and compact packing of hemp fibres in the

PLA matrix, compared to the composite made from PLA/hemp yarn. The compact and close

packing of PLA/hemp yarns reduces the porosity inside the composites, which in turn contributes to

the reduction in water absorption [103]. It is evident that the effect of weave pattern is significant

for the water absorption for the composites. Although the composite made from the winded hybrid

yarn prepreg has larger moisture absorption than that of the pure PLA, it exhibits lower moisture

absorption than the composites with woven fabrics (satin and basket). Actually, the matrix pockets

and yarn undulation within the weaves provide an easier diffusion path for the moisture to travel

within the weave, compared to a composite composed of unidirectional yarns (Publications II, III,

IV and V).

SUMMARY OF RESULTS

26

Table 3. Composition of the fabricated composites calculated from the density measurements.

Sample Density in

g/cm3

Fibre mass

fraction in %

Fibre

volume

fraction

in %

Matrix

volume

fraction in

%

Porosity

volume

fraction in

%

PLA 1.24 0.0 0.0 100.0 -

PLA/10 Hemp-150 1.192 10.1 9.1 85.3 5.60

PLA/10 Hemp-250 1.193 11.3 8.1 86.5 5.30

PLA/20 Hemp-150 1.196 22.3 18.8 74.0 7.20

PLA/20 Hemp-250 1.197 23.3 18.0 75.0 6.90

PLA/35 Hemp-150 1.210 34.9 29.0 62.9 8.00

PLA/35 Hemp-250 1.219 35.5 28.7 64.0 7.30

PLA/45 Hemp-150 1.224 45.5 38.1 53.2 8.70

PLA/45 Hemp-250 1.226 46.1 37.7 53.9 8.40

Alkali hemp/PLA yarn

(off-axis angle 0°)

1.294 30.0 26.23 72.53 1.23

Hemp/PLA yarn

(off-axis angle 0°)

1.271 30.0 25.77 71.25 2.97

Satin hemp/PLA fabric 1.298 30.0 26.31 72.74 0.96

Basket hemp/PLA fabric 1.251 30.0 25.35 70.10. 4.55

Satin Lyocell/PLA fabric 1.273 27.02 71.32 1.66

Satin Lyocell-hemp/PLA fabric 1.296 15.0 hemp

15.0 Lyocell

26.27 72.63 1.09

Nonwoven PLA/hemp 1.0914 30.0 23.0 61.2 10.8

Nonwoven PLA/Lyocell 1.2765 30.0 25.2 71.5 3.3

Nonwoven PLA/hemp-Lyocell 1.1222 15.0 hemp

15.0 Lyocell

22.4 76.4 2.1

Among all composites in this study, composites made from satin PLA/Lyocell fabric showed the

lowest water absorptions, which may be due to their better interfacial adhesion that decreases the

thickness of the interphase area between the fibres and the matrix, thus, decreasing the water

absorption through the interphase and further into inner parts of the structure [104, 105].

As a result of alkali treatment, the interfacial adhesion in PLA/hemp fibres improved, which is

necessary for the reduction of interfacial wicking of the water molecules. Thus, the moisture

absorption of the composites can be reduced by chemical treatments [103, 105, 106] (Publication

II).

Furthermore, as shown in Fig. 10 c, it can be seen that for all the composites investigated, WG

increases monotonically by time, initially before reaching a maximum. It can be observed that the

water immersion temperature does have an influence on the water absorption curves. Increasing the

immersion temperature from room temperature to 80 ˚C led to increased water absorption of the

neat PLA and composites as well as decreased the saturation time. The weight gain was found to

decrease after passing through a maximum. For the composites, this could be due to the neat PLA

layer peeling on the surface as a result of biodegradation and dissolving with time, together with the

removal of some substances from the hemp fibre surface during the immersion [107, 108]

(Publications II and IV).

SUMMARY OF RESULTS

27

a)

b)

SUMMARY OF RESULTS

28

Fig. 10 Apparent weight gain against time for the manufactured composites at: a) room temperature, and b) 80 ˚C

4.2.3 Tensile properties

An overview of the tensile strength, modulus and elongation at break of the composites, compared

to the values of the neat PLA matrix are given in Fig. 11 a and b and Table 4. It can be seen that an

increase in fibre content improves the tensile strength and modulus as expected. It is also obvious

from the results that composites made from the hybrid yarn with a wrapping density of 250 showed

a consistent improvement in tensile modulus over the composites with 150 wrapping density, across

the whole range of fibre fraction ratios investigated. Increasing wrapping density improved the

tensile modulus of the composites because the tortuosity of the hybrid yarns decreases as wrap twist

increases, and consequently the reinforcement alignment would be improved [109] (Publication I).

As can be seen in Table 4, the composites fabricated from PLA/hemp yarns (both untreated and

treated) had considerably higher tensile strength and modulus in the principal fibre direction (i.e.

longitudinal direction) than in the corresponding nonwoven mats. As expected, the composite

strength and stiffness decreased with increasing fibre orientation angle. The composite made from

the nonwoven demonstrated higher tensile properties in the principal fibre direction, compared with

hemp/PLA yarn composite with fibre orientation angles 45˚ and 90˚. Furthermore, the results

showed that tensile strength of composites made with fibre orientation angles 45˚ and 90˚ were

lower than pure PLA. The highest strength values were reached by treated hemp/PLA yarn with

fibre orientation angle 0˚. The decrease in elongation at break in the case of the composites is

principally due to the structural integrity of the matrix (PLA), which is destroyed by the loading of

the natural fibre and leads to faster fracture than pure matrix [110] (Publication II).

The composites from the woven fabric show better tensile properties, compared to the winded yarn

laminate composite. In fact, tightly woven fabrics are usually the right choice to maintain fibre

orientation during the fabrication process, which was not possible in the winded yarn laminate. The

c)

SUMMARY OF RESULTS

29

tensile modulus and strength of composite for basket 2/2 is lower, compared to satin due to its fabric

structure. The basket weave has a higher crimp, compared to the satin weave due to its more

interlacement (Publication III).

Specimens made from nonwoven reinforcement show lower values than specimens made from

fabric. The strength values of hemp-PLA, Lyocell-PLA and Lyocell-hemp PLA specimens differ

noticeably. The improvement of the tensile properties of hemp-reinforced PLA composites by

addition of Lyocell fibres could be attributed to the much higher fineness of the Lyocell fibres,

which cause a larger specific bonding surface and a better matrix–fibre adhesion [99] (Publications

IV and V).

Fig. 11. Tensile properties of PLA/hemp composites: a) tensile modulus, and b) tensile strength (Publication I)

SUMMARY OF RESULTS

30

Table 4 Tensile properties of PLA-based composites (Publications II, III, IV and V)

Sample Tensile

strength

(MPa)

Young’s

modulus

(GPa)

Elongation at

break (%)

Publication

PLA 41.21 (2.25) 2.91 (0.39) 1.92 (0.26) II, III, IV and

V

Hemp/PLA yarn (off-axis angle 0°) 72.75 (6.26) 8.77 (1.44) 1.24 (0.36) II

II

II

Hemp/PLA yarn (off-axis angle 45°) 34.75 (4.63) 4.62 (1.04) 0.66 (0.18)

Hemp/PLA yarn (off-axis angle 90°) 22.01 (3.38) 3.70 (0.65) 0.53 (0.11)

Satin hemp/PLA fabric 88.06 (7.70) 10.23 (2.67) 1.35 (0.30) III

III Basket hemp/PLA fabric 81.09 (5.62) 9.20 (1.81) 1.58 (0.22)

Alkali hemp/PLA yarn (off-axis angle 0°) 77.08 (2.60) 10.27 (1.36) 1.58 (0.26) II

Hemp/PLA nonwoven 53.63 (1.22) 5.60 (0.95) 1.10 (0.30) II

Hemp/Lyocell/PLA nonwoven 60.63 (8.36) 6.31 (0.51) - IV

IV Lyocell/PLA nonwoven 80.87 (2.94) 6.86 (0.40) -

Satin hemp/Lyocell/PLA fabric 96.01 (2.05) 11.15 (0.61) - V

V Satin Lyocell/PLA fabric 101.23 (1.27) 11.42 (0.84) -

Note: Data in table are mean (SD).

4.2.4 Flexural properties

The results of the flexural test are shown in Fig. 12 (a and b) and Fig. 13 (a and b). The composite

made from yarn with 250 wrapping density gave slightly higher values than the composite made

from yarn with 150 wrapping density, which is probably due to the lower porosity and more aligned

fibres (Fig. 12) (Publication I).

From Fig. 13 a and b, it can be seen that the composites made from PLA/hemp yarn (0˚ direction)

showed an evident improvement in flexural modulus over the composites produced from PLA/hemp

nonwoven. The flexural properties of unidirectional composites followed the reduction trend in their

value when the fibre orientation changed from 0˚ to the off-axial direction. Alkali treatments

enhanced the flexural strength and modulus of the composites, which can be attributed to the greater

interfacial bonding between treated hemp fibres and PLA [111-113] (Publication II).

Different weave pattern has significant influence on the flexural strength of the composites. Satin

fabric composite exhibited the highest flexural strength due to the presence of more aligned fibres in

the axial direction and lower number of interlacing points, which has less chance for stress

concentration [114] (Publication III).

From the results, it can be seen that the highest values were achieved by the composite reinforced

SUMMARY OF RESULTS

31

by satin Lyocell/PLA fabric. Fibre reinforcement resulted in a clear and significant increase of the

modulus values, compared to pure PLA. Between satin hemp-Lyocell/PLA and satin hemp/PLA

composites, no significant differences in modulus were determined, but their strength values differ

clearly (Publications IV and V).

Fig. 12. Flexural properties of PLA/hemp composites

a) b)

a)

SUMMARY OF RESULTS

32

Fig. 13 Flexural properties of PLA-based composites

4.2.5 Impact resistance

The results of the Charpy impact test are shown in Fig. 14 (a, b and c). It is apparent from Fig. 14

that the impact strength of the PLA/hemp composites increases with increasing mass proportion of

fibre. However, the value at 10 mass% fibre content is still lower than the value for neat PLA.

When longer fibres are introduced into the PLA matrix, the impact strength greatly increases due to

the frequent occurrence of long fibre pull-outs [113]. In our work, the higher impact strength of

composites, compared to neat PLA can be attributed to the fact that the reinforcing staple fibres are

in the form of yarn and can act as long fibres. The composites made from yarn with wrapping

density of 250 showed relatively lower impact strengths, compared to the composite from yarn with

wrapping density of 150 (Publication I).

A significant reinforcement effect was determined for the treated PLA/hemp yarn composite. In this

case, debonding occurred, followed by fibre breakage rather than by interfacial debonding, usually

associated with high energy absorption [115, 116]. The increased PLA crystallinity of the alkali

treated hemp fibre composites, compared with untreated hemp fibre composites could be another

factor leading to increased impact strengths [117, 118]. The value of the PLA/hemp nonwoven

composites was lower than that of the neat PLA matrix. Further increments of fibre angle caused a

decrease in impact strength of the composites, which is likely to be due to decreased dissipation of

energy by less fibre pull-out. The impact strength of the nonwoven composites was higher than that

of the yarn composites with 45˚ and 90˚ off-axis (Publication II).

It can be seen from the results (Fig. 14 b) that the variation in impact strength of the composites was

not significantly influenced by the weaving architecture, as the variation is meagre. The higher

impact energies for fabric laminate, compared with winded yarn laminate are mainly because of the

presence of a high number of axially oriented fibres [119] (Publication III).

b)

SUMMARY OF RESULTS

33

The improvement of the impact strength of hemp fibre-reinforced PLA was caused by the addition

of 50% Lyocell to the hemp fibres (Fig. 14 c). In fact, the higher impact energies for satin fabric,

compared with nonwoven laminate are mainly because of the presence of a greater number of

axially oriented fibres, since woven fabric composites provide more balanced properties in the

fabric plane than randomly oriented composites [32]. The significant increase in impact strength of

Lyocell/PLA composites can be referenced to higher elongation at break [120, 121] (Publications

IV and V).

a)

b)

SUMMARY OF RESULTS

34

Fig. 14 Impact strength of the manufactured composites

4.2.6 Dynamic mechanical thermal testing

Fig. 15 shows how fibre treatment and orientation influenced the storage modulus and damping

factor (tan δ) of PLA and its respective composites. It is evident from Fig. 15 a and b that

incorporation of hemp fibres results in an increase in the storage modulus of the biocomposites,

which reveals effective stress transfer from the fibre to the matrix at the interface and good adhesion

between them. The storage modulus of the composites showed remarkable dependence on fibre

orientation. The storage modulus was highest for the hemp/PLA (0) yarn composite, followed by

the hemp/PLA (45) yarn composite and the hemp/PLA (90) yarn composite. Composite from satin

and basket weaves had a better storage modulus, compared to the winded yarn laminate composite,

which is in line with the results from tensile, flexural and impact tests showing that composites from

satin and basket weave are superior to winded yarn laminate composite.

Fig. 15 b and c show the tan δ value of the composites as the function of temperature and the glass

transition temperature corresponding to the maximum tan δ peak. The glass transition temperatures

of composites are higher, compared to pure PLA. This indicates that with the presence of hemp

fibres, the polymer relaxation is delayed and segmental motion of the PLA polymer chains is

restricted due to increased crystallinity.

The area under this peak for PLA composites seems to be smaller especially for alkali treated hemp

fibre composite. A possible explanation is that there is a strong interaction between the fibre and the

matrix. The magnitude of tan δ values is also seen to increase in the case of composite from

nonwoven, compared to the composites from yarn and it could be because of its higher porosity. In

reality, the storage modulus favours long fibres (large l/d), and damping favours short fibres (small

l/d). Our experimental investigations indicated that composites with fibre off-axis 90° showed better

damping than composites with 45° and 0.

c)

SUMMARY OF RESULTS

35

a)

b)

SUMMARY OF RESULTS

36

Fig. 15. DMTA analysis of PLA reinforced with hemp and Lyocell fibres. a-b) storage modulus vs temperature; c-d)

tanδ curves (Publications II and III)

c)

d)

SUMMARY OF RESULTS

37

4.2.7 Differential scanning calorimetry

The DSC heating thermograms of neat PLA and produced composites are shown in Fig. 16, and the

corresponding transition temperatures are tabulated in Table 5. The data indicate that the Tg and Tm

of the PLA composites decreased less than 2 ˚C with the addition of hemp fibre to the PLA matrix;

however, the ΔHm and Tc of the PLA composites decreased significantly in the presence of hemp.

These results suggest that hemp does not significantly affect the crystallisation properties of the

PLA matrix. In the first article, the crystallisation temperature of the PLA/hemp composite

decreased by up to about 13–15°C (for pure PLA compared to composites with 45 mass% fibre

content), which signifies that the hemp fibres have a negative effect on crystallisation (Publication I).

However, the data from Publication II indicate that the crystallinity and Tc of PLA was found to

increase as a result of the presence of hemp, since the fillers acted as nucleating agents for the

crystallisation of the polymers [99, 122]. Also, it is apparent that the crystallinity of PLA in the

treated hemp/PLA composites increased, compared with that of the untreated hemp/PLA

composites. This could be due to the fact that after alkali treatment, the impurities including wax

and pectin were removed from the fibres, which in turn increased the number of nucleating sites of

the fibres [14, 122] (Publication II).

The data indicate that the Tg of PLA in its composites did not show much difference. However, Tm

was slightly affected by the introduction of hemp and Lyocell fibres with a few decreases.

Furthermore, the Tc decreased from pure PLA to composite containing hemp-Lyocell fibres. It can

be clearly seen that the degree of crystallisation increases from 4.6% for neat PLA up to 8.3% for

PLA/hemp, 16.9% for PLA/Lyocell and 11.8% for PLA/hemp-Lyocell. The double melting peaks

suggest the occurrence of the crystal’s reorganisation during the heating run (Publication IV).

a)

SUMMARY OF RESULTS

38

Fig. 16. DSC curves of neat PLA and PLA composites

b)

c)

SUMMARY OF RESULTS

39

Table 5. Calorimetric data for PLA-based composites for the first heating run (10˚C/min)

Sample ∆Hcc

(J/g)

∆Hm

(J/g)

XDSC

(%)

Tg

(°C)

TC

(°C)

Tm

(°C)

Publication

PLA (6201D, processed at 175 ºC) - 52.29 55.71 54.03 112 168.82 I

I

I

I

I

I

PLA/10 Hemp-250 3.92 34.94 37.06 57.13 100.34 166.76

PLA/20 Hemp-250 4.28 28.84 33.02 57.17 98.50 167.13

PLA/35Hemp-250 4.50 26.21 31.67 56.37 98.54 166.43

PLA/45 Hemp-250 5.02 21.39 28.18 57.18 97.58 166.52

PLA (6202D, processed at 195 ºC) 28.9 35.9 7.5 59.9 109.9 167.4 II

II

II

Untreated hemp/PLA 13.17 22.73 14.7 61.3 110.5 167.5

Treated hemp/PLA 19.3 30.3 16.9 61.8 111.8 167.9

PLA (6202D, processed at 190 ºC) - - 4.6 59.2 112.1 167.4 IV

IV

IV

IV

PLA/hemp - - 8.3 59.4 108.6 166.8

PLA/Lyocell - - 16.9 57.6 108.7 166.7

PLA/hemp-Lyocell - - 11.8 57.5 106.4 166.3

SUMMARY OF RESULTS

40

[Type a quote from the document or the summary of an interesting point. You can position the text box

anywhere in the document. Use the Drawing Tools tab to change the formatting of the pull quote text box.]

DISCUSSION AND CONCLUSIONS

41

5. DISCUSSION AND CONCLUSIONS

Recently, the great research efforts into the development of materials derived from renewable

resources, such as biocomposites have been initiated thanks to concerns over environmental and

economic problems. However, most of the studies and researches to date in the field of

biocomposites have been directed towards non-structural applications. This is mainly attributed to

earlier research efforts showing that these composites do not possess the properties essential to

make them suitable for applications involving structural elements. Through the development and

characterisation of a novel biocomposite system, this project aims to overcome some of the current

challenges for these materials application under different mechanical conditions.

Unidirectional-aligned hemp fibre reinforced PLA composites were produced for this research,

applying selected processing procedures to allow for efficient fibre reinforcement and an

improvement in overall composite performance and properties. The aligned hemp/PLA yarns were

produced through two developed types of hybrid yarn process: co-wrapping and commingling. In

Publication I, the properties of 0/90 bi-directional composites made from PLA/hemp co-wrapped

hybrid yarn preforms as reinforcement, instead of using only hemp yarns impregnated with a melted

PLA matrix, were assessed. Composites were fabricated by compression moulding of 0/90 bi-

directional prepregs and characterised regarding porosity, mechanical strength and thermal

properties. Here, we used continuous PLA filaments, which were used to wrap a low twist hemp

yarn. The composites made from the hybrid yarn with higher wrapping density showed

improvements of mechanical properties due to lower porosity. However, the porosities of

composites were between 6–9 vol.%, and the porosity fraction was still high even though the fibre

volume fraction was low. Mechanical tests showed that the tensile and flexural strengths of the

composites markedly increased with the fibre content, reaching 59.3 and 124.2 MPa when

reinforced with 45 mass% fibre, which is approximately 2 and 3.3 times higher, compared to neat

PLA. Impact strength of the composites decreased, initially up to 10 mass% fibre; however, higher

fibre loading (up to 45 mass%) caused an increase in impact strength up to 26.3 KJ/m2, an

improvement of about 2 times higher than neat PLA.

In Publication II, we investigated the development of new hybrid yarns with low twist for high

performance natural fibre-reinforced composites suitable for use in structural or semi-structural

applications and with lower amount of porosity. We investigated the influences of different

orientation especially off-axial direction of hemp fibre as well as alkali treatment. The superior

mechanical properties of the treated fibre composites were attributed to the greater interfacial

bonding between treated hemp fibres and PLA. The results showed that the mechanical properties

of the composites were influenced by the fibre direction. Tensile, flexural and impact values of the

composites showed the decreasing trend for off-axial composites, compared to 0˚ axial-oriented

composite. The damping properties were highly affected by the testing direction; it was increased at

an off-axis angle of 45°and 90°. From water absorption test results, it was found that higher

temperature generally increased the WG% of the neat PLA and all of the composites, as well as

shortening the saturation time.

DISCUSSION AND CONCLUSIONS

42

Porosity is difficult to avoid in natural fibre composites and influences on the composite properties,

yet how to control the porosity has so far only received limited attention. Thus, research on how to

decrease the amount of porosity is warranted. Therefore, the effects of the choice of reinforcing

filler on the structure and properties of PLA-based composites have been studied, focusing on the

porosity, water absorption, mechanical and thermo-mechanical properties. The obtained results

showed that combining hemp and Lyocell in a PLA-based composite can significantly improve the

impact strength at ambient temperature, flexural and tensile strength and modulus, compared with

hemp fibre reinforced PLA since Lyocell-based composites had lower void content due to the

higher interfacial adhesion. The complexity of the surface chemistry and the irregularity of the

morphology of plant fibres is one of the most important considerations; in addition, the presence of

luminal cavities could be another factor. Although Lyocell is expensive, it is reproducible by

artificial production, and the admixture of Lyocell fibres to hemp fibres leads to less variation in

quality of the fabricated composites. Publications III and IV in this thesis confirmed that a

composite system consisting of two different fibres (Lyocell and hemp) with different morphologies

results in composites with better mechanical properties and lower moisture absorption, and their use

in outdoor applications can then be established.

In Publications III and V, we also studied the manufacture of hybrid textile yarns containing

reinforcing and thermoplastic components within their structure, the technology for processing them

into woven fabrics (prepregs), and the manufacture of thermoplastic composites based on them. The

effect of two types of structures for the fabrics, namely 8-harness satin and basket, was studied. The

results showed that the unidirectional hemp/PLA composite made by satin-weave architecture fabric

possessed the highest mechanical strength, compared to the composites manufactured with basket

weave architecture fabric and winded hybrid yarn laminate. This improvement in mechanical

strength was correlated to that of decrease in void content and fibre misalignments. The best overall

properties in this research were achieved with satin fabric from aligned Lyocell/PLA yarn, leading

to a tensile strength of 101.2 MPa, Young’s modulus of 11.4 GPa, flexural strength of 158.2 MPa,

flexural modulus of 9.7 GPa, and impact strength of 47.4 kJ/m2, followed by satin fabric from

Lyocell/hemp mixture.

This study, therefore, indicated that hemp fibres can be a successful solution for many load-bearing

and structural applications such as construction and automotive.

FUTURE STUDIES

43

6. FUTURE STUDIES

Although the research completed in the current study has successfully demonstrated that

unidirectional hemp/PLA and hemp-Lyocell/PLA biocomposites can be manufactured with high

levels of performance characteristics, additional studies on the durability, biodegradation, the

possibility of improving the outdoor properties, and long term performance of the composites are

necessary if they are to be used in load-bearing and structural applications. Hence, understanding

the degradation mechanisms under the lifetime of NFCs is of great importance.

Failure in structural composites occurs often thanks to mechanical fatigue. Certain facts such as the

behaviour of these composite materials under fatigue loading and their failure mechanisms should

be investigated. Indeed, fatigue analysing is very time consuming; therefore, modelling may help to

save time, energy and resources. This can be done by adopting the model being developed for

synthetic composite materials and applied for NFCs.

FUTURE STUDIES

44

[Type a quote from the document or the summary of an interesting point. You can position the text box

anywhere in the document. Use the Drawing Tools tab to change the formatting of the pull quote text box.]

REFERENCES

45

REFERENCES

[1] Chawla KK, SpringerLink. Composite Materials: Science and Engineering. New York, NY:

Springer New York; 1998.

[2] Chawla KK. Composite materials: science and engineering. New York: Springer

Science+Business Media; 2012.

[3] Hull D, Clyne TW. An introduction to composite materials. Cambridge: Cambridge Univ. Press;

1996.

[4] Thomas S, Pothan LA, Cherian BM. Advances in natural fibre reinforced polymer composites:

macro to nanoscales. International Journal of Materials & Product Technology. 2009;36(1-4):317-33.

[5] Madsen B, Thygesen A, Lilholt H. Plant fibre composites – porosity and volumetric interaction.

Composites Science and Technology. 2007;67(7):1584-600.

[6] Pickering KL, Sawpan MA, Jayaraman J, Fernyhough A. Influence of loading rate, alkali fibre

treatment and crystallinity on fracture toughness of random short hemp fibre reinforced polylactide

bio-composites. Composites Part A: Applied Science and Manufacturing. 2011;42(9):1148-56.

[7] Svensson N, Shishoo R, Gilchrist M. Manufacturing of Thermoplastic Composites from

Commingled Yarns-A Review. Journal of Thermoplastic Composite Materials. 1998;11(1):22-56.

[8] Wysocki M, Toll S, Larsson R, Skolan för t, Lättkonstruktioner, Kth, et al. Press forming of

commingled yarn based composites: The preform contribution. Composites Science and

Technology. 2007;67(3):515-24.

[9] Bernet N, Michaud V, Bourban PE, Månson JAE. Commingled yarn composites for rapid

processing of complex shapes. Composites Part A. 2001;32(11):1613-26.

[10] Kravaev P, Stolyarov O, Seide G, Gries T. A method for investigating blending quality of

commingled yarns. Textile Research Journal. 2013;83(2):122-9.

[11] Islam MS, Pickering KL, Foreman NJ. Influence of Hygrothermal Ageing on the Physico-

Mechanical Properties of Alkali Treated Industrial Hemp Fibre Reinforced Polylactic Acid

Composites. Journal of Polymers and the Environment. 2010;18(4):696-704.

[12] Oksman K. Mechanical properties of natural fibre mat reinforced thermoplastic. Applied

Composite Materials 2000;7(5-6):403-14.

[13] Gamstedt EK, Berglund LA, Peijs T, Department of Engineering S, Mathematics, Polymeric

Composite M, et al. Fatigue mechanisms in unidirectional glass-fibre-reinforced polypropylene.

Composites Science and Technology. 1999;59(5):759-68.

[14] Sawpan MA, Pickering KL, Fernyhough A. Improvement of mechanical performance of

industrial hemp fibre reinforced polylactide biocomposites. Composites Part A. 2011;42(3):310-9.

[15] Hu R, Lim J-K. Fabrication and Mechanical Properties of Completely Biodegradable Hemp

Fiber Reinforced Polylactic Acid Composites. Journal of Composite Materials. 2007;41(13):1655-69.

[16] Behnaz Baghaeia MS, Lena Berglin. Manufacture and characterisation of thermoplastic

composites made from PLA/hemp co-wrapped hybrid yarn prepregs. Composites Part A: Applied

Science and Manufacturing. 2013.

[17] Zhang L, Miao M. Commingled natural fibre/polypropylene wrap spun yarns for structured

thermoplastic composites. Composites Science and Technology. 2010;70(1):130-5.

[18] Goutianos S, Peijs T. The optimisation of flax fibre yarns for the development of high-

performance natural fibre composites. Advanced Composites Letters. 2003;12(6):237-41.

REFERENCES

46

[19] Gu H, Miao M. Optimising fibre alignment in twisted yarns for natural fibre composites.

Journal of Composite Materials. 2014;48(24):2993-3002.

[20] Shen L, Worrell E, Patel M. Present and future development in plastics from biomass. Biofuels,

Bioproducts and Biorefining. 2010;4(1):25-40.

[21] Huang XN, Hull D. Effects of fibre bridging on GIC of a unidirectional glass/epoxy composite.

Composites Science and Technology. 1989;35(3):283-99.

[22] Baillie C. Green composites: polymer composites and the environment. Cambridge, England;

Boca Raton, Fla: CRC Press; 2004.

[23] Chandramohan D, Marimuthu K. A REVIEW ON NATURAL FIBERS. International Journal

of Research and Reviews in Applied Sciences. 2011;8(2).

[24] Hodzic A, Shanks B. Natural fibre composites: materials, processes and properties.

Philadelphia, PA: Woodhead Pub; 2014.

[25] Pickering KL, Sawpan MA, Jayaraman J, Fernyhough A. Influence of loading rate, alkali fibre

treatment and crystallinity on fracture toughness of random short hemp fibre reinforced polylactide

bio-composites. Composites Part A. 2011;42(9):1148-56.

[26] Tserki V, Matzinos P, Panayiotou C. Novel biodegradable composites based on treated

lignocellulosic waste flour as filler. Part II. Development of biodegradable composites using treated

and compatibilized waste flour. Composites Part A. 2006;37(9):1231-8.

[27] Mueller DH, Krobjilowski A. New Discovery in the Properties of Composites Reinforced with

Natural Fibers. Journal of Industrial Textiles. 2003;33(2):111-30.

[28] Beldzki AK, Sperber VE, Faruk O, ebrary I, Rapra Technology L. Natural and wood fibre

reinforcement in polymers. Shawbury, U.K: Rapra Technology Ltd; 2002.

[29] Fortenbery TR, Bennett M. Opportunities for Commercial Hemp Production. Review of

Agricultural Economics. 2004;26(1):97-117.

[30] Müssig J. Industrial application of natural fibres: structure, properties, and technical

applications. Chichester, West Sussex, U.K; Hoboken, N.J: Wiley; 2010.

[31] Salentijn EMJ, Zhang Q, Amaducci S, Yang M, Trindade LM. New developments in fiber

hemp (Cannabis sativa L.) breeding. Industrial Crops and Products. 2014.

[32] Sun H, Miao J, Yu Y, Zhang L. Dissolution of cellulose with a novel solvent and formation of

regenerated cellulose fiber. Applied Physics A. 2015;119(2):539-46.

[33] Fink HP, Weigel P, Purz HJ, Ganster J. Structure formation of regenerated cellulose materials

from NMMO-solutions. Progress in Polymer Science. 2001;26(9):1473-524.

[34] Roy C, Budtova T, Navard P. Rheological Properties and Gelation of Aqueous

Cellulose−NaOH Solutions. Biomacromolecules. 2003;4(2):259-64.

[35] Swatloski RP, Spear SK, Holbrey JD, Rogers RD. Dissolution of Cellose with Ionic Liquids.

Journal of the American Chemical Society. 2002;124(18):4974-5.

[36] Jiang G, Huang W, Li L, Wang X, Pang F, Zhang Y, et al. Structure and properties of

regenerated cellulose fibers from different technology processes. Carbohydrate Polymers.

2012;87(3):2012-8.

[37] Woodings C, Textile I, Textile I. Regenerated cellulose fibres. Boca Raton, FL: CRC Press;

2001.

[38] Blackburn RS, Textile I. Biodegradable and Sustainable Fibres. GB: Elsevier; 2005.

REFERENCES

47

[39] Johnson RK, Zink-Sharp A, Renneckar SH, Glasser WG. Mechanical properties of wetlaid

lyocell and hybrid fiber-reinforced composites with polypropylene. Composites Part A.

2008;39(3):470-7.

[40] Drzal, Misra, Mohanty. Surface modifications of natural fibers and performance of the

resulting biocomposites: An overview. Composite Interfaces 2001;8(5):313-43.

[41] Bisanda ETN, Ansell MP. The effect of silane treatment on the mechanical and physical

properties of sisal-epoxy composites. Composites Science and Technology. 1991;41(2):165-78.

[42] Gassan J, Bledzki AK. Possibilities for improving the mechanical properties of jute/epoxy

composites by alkali treatment of fibres. Composites Science and Technology. 1999;59(9):1303-9.

[43] Mohanty AK, Khan MA, Hinrichsen G. Influence of chemical surface modification on the

properties of biodegradable jute fabrics—polyester amide composites. Composites Part A.

2000;31(2):143-50.

[44] Mohanty AK, Khan MA, Hinrichsen G. Surface modification of jute and its influence on

performance of biodegradable jute-fabric/Biopol composites. Composites Science and Technology.

2000;60(7):1115-24.

[45] Mohanty AK, Khan MA, Sahoo S, Hinrichsen G. Effect of chemical modification on the

performance of biodegradable jute yarn-Biopol® composites. Journal of Materials Science.

2000;35(10):2589-95.

[46] Sharma HSS, Fraser TW, McCall D, Shields N, Lyons G. Fine structure of chemically

modified flax fibre. Journal of the Textile Institute. 1995;86(4):539-48.

[47] Sreenivasan S, Bhama Iyer P, Krishna Iyer KR. Influence of delignification and alkali

treatment on the fine structure of coir fibres (Cocos Nucifera). Journal of Materials Science.

1996;31(3):721-6.

[48] Oksman K, Skrifvars M, Selin JF. Natural fibres as reinforcement in polylactic acid (PLA)

composites. Composite Sceinec and Technology. 2003;63(9):1317-24.

[49] Perego G, Cella GD. Mechanical Properties. Poly(Lactic Acid): John Wiley & Sons, Inc.;

2010. p. 141-53.

[50] Netravali AN, Chabba S. Composites get greener. Elsevier B.V; 2003. p. 22-9.

[51] Mohanty AK, Misra M, Hinrichsen G. Biofibres, biodegradable polymers and biocomposites:

An overview. Macromolecular Materials and Engineering. 2000;276-277(1):1-24.

[52] John MJ, Thomas S. Biofibres and biocomposites. Carbohydrate Polymers. 2008;71(3):343-64.

[53] Stevens ES. Green Plastics: An Introduction to the New Science of Biodegradable Plastics:

Princeton: Princeton University Press; 2002.

[54] Reis PNB, Ferreira JAM, Antunes FV, Costa JDM. Flexural behaviour of hybrid laminated

composites. Composites Part A. 2007;38(6):1612-20.

[55] Burgueño R, Quagliata MJ, Mohanty AK, Mehta G, Drzal LT, Misra M. Hybrid biofiber-based

composites for structural cellular plates. Composites Part A. 2005;36(5):581-93.

[56] Rout J, Misra M, Tripathy SS, Nayak SK, Mohanty AK. The influence of fibre treatment on the

performance of coir-polyester composites. Composites Science and Technology. 2001;61(9):1303-

10.

REFERENCES

48

[57] Hariharan ABA, Khalil HPSA. Lignocellulose-based Hybrid Bilayer Laminate Composite: Part

I - Studies on Tensile and Impact Behavior of Oil Palm Fiber-Glass Fiber-reinforced Epoxy Resin.

Journal of Composite Materials. 2005;39(8):663-84.

[58] Mishra S, Mohanty AK, Drzal LT, Misra M, Parija S, Nayak SK, et al. Studies on mechanical

performance of biofibre/glass reinforced polyester hybrid composites. Composites Science and

Technology. 2003;63(10):1377-85.

[59] Joseph PV, Joseph K, Rabello MS, Mattoso LHC, Thomas S. Environmental effects on the

degradation behaviour of sisal fibre reinforced polypropylene composites. Composites Science and

Technology. 2002;62(10):1357-72.

[60] Mirbagheri J, Tajvidi M, Hermanson JC, Ghasemi I. Tensile properties of wood flour/kenaf

fiber polypropylene hybrid composites. Journal of Applied Polymer Science. 2007;105(5):3054-9.

[61] Mieck KP, Reußmann T, Hauspurg C. Correlations for the Fracture Work and Failing Weight

Impact Properties of Thermoplastic Natural/Long Fiber Composites. Materialwissenschaft und

Werkstofftechnik. 2000;31(2):169-74.

[62] Graupner N. Improvement of the Mechanical Properties of Biodegradable Hemp Fiber

Reinforced Poly(lactic acid) (PLA) Composites by the Admixture of Man-made Cellulose Fibers.

Journal of Composite Materials. 2009;43(6):689-702.

[63] Yuan Y, Guo M, Wang Y. Flax Fibers as Reinforcement in Poly (Lactic Acid) Biodegradable

Composites. Intelligent Computing and Information Science. Berlin, Heidelberg: Springer Berlin

Heidelberg; 2011. p. 547-53.

[64] Ouagne P, Bizet L, Baley C, Bréard J. Analysis of the Film-stacking Processing Parameters for

PLLA/Flax Fiber Biocomposites. Journal of Composite Materials. 2010;44(10):1201-15.

[65] Bax B, Müssig J. Impact and tensile properties of PLA/Cordenka and PLA/flax composites.

Composites Science and Technology. 2008;68(7-8):1601-7.

[66] Plackett D, Løgstrup Andersen T, Batsberg Pedersen W, Nielsen L. Biodegradable composites

based on l-polylactide and jute fibres. Composites Science and Technology. 2003;63(9):1287-96.

[67] Kaldenhoff R. WB. New developments and applications of textile reinforcements for

composite materials. Indian Journal of Fiber and Textile Research. 1997;22(4):255-8.

[68] Alagirusamy R, Fangueiro R, Ogale V, Padaki N. Hybrid Yarns and Textile Preforming for

Thermoplastic Composites. Textile Progress. 2006;38(4):1-71.

[69] Sawhney APS, Ruppenicker GF, Kimmel LB, Robert KQ. Comparison of Filament-Core Spun

Yarns Produced by New and Conventional Methods. Textile Research Journal. 1992;62(2):67-73.

[70] Lauke B, Bunzel U, Schneider K. Effect of hybrid yarn structure on the delamination behaviour

of thermoplastic composites. Composites Part A. 1998;29(11):1397-409.

[71] R. H. Core-spun yarns – market of the future. Melliand International. 2001;7(4):283-4.

[72] Long AC, Wilks CE, Rudd CD. Experimental characterisation of the consolidation of a

commingled glass/polypropylene composite. Composites Science and Technology.

2001;61(11):1591-603.

[73] Ye L, Friedrich K, Kästel J, Mai Y-W. Consolidation of unidirectional CF/PEEK composites

from commingled yarn prepreg. Composites Science and Technology. 1995;54(4):349-58.

REFERENCES

49

[74] Karus M KM, Ortmann S. Use of Natural Fibres in Composites in the German and Austrian

Automotive Industry—Market Survey 2002: Status, Analysis and Trends. Journal of Industrial

Hemp.8(2):73-8.

[75] CM. C. Woodfiber-plastic composites in the United States – History and current and future

markets. 3rd International Wood and Natural Fibre Composites Symposium. Kassel,

Germany2000. p. 1-7.

[76] Parikh DV, Calamari TA, Sawhney APS, Blanchard EJ, Screen FJ, Myatt JC, et al.

Thermoformable Automotive Composites Containing Kenaf and Other Cellulosic Fibers. Textile

Research Journal. 2002;72(8):668-72.

[77] Goutianos S PT. The optimisation of flax fibre yarns for the development of high-performance

natural fibre composites. Advanced Composites Letters.12(6):237-41.

[78] Shah DU SP CM. Mechanical characterization of vacuum infused thermoset matrix composites

reinforced with aligned hydroxyethylcellulose sized plant bast fibre yarns. 4th International

Conference on Sustainable Materials, Polymers and Composites. Birmingham, UK2011.

[79] Stolyarov ON, Stolyarov IN, Kryachkova TA, Kravaev PG. Hybrid Textile Yarns and

Thermoplastic Composites Based on Them. Fibre Chem. 2013;45(4):217-20.

[80] Salman SD, Sharba MJ, Leman Z, Sultan MTH, Ishak MR, Cardona F. Physical, Mechanical,

and Morphological Properties of Woven Kenaf/Polymer Composites Produced Using a Vacuum

Infusion Technique. International Journal of Polymer Science.

[81] Azrin Hani AR, Shaari MF, Mohd Radzuan NS, Hashim MS, R A, M M. Analysis of woven

natural fiber fabrics prepared using self-designed handloom. International Journal of Automotive

and Mechanical Engineering. 2013;8(1):1197-206.

[82] Vasiliev V ME. Mechanics and Analysis of Composite Materials. Oxford, UK: Elsevier

Science; 2001.

[83] E. M. Mechanics and analysis of fabric composites and structures. AUTEX Research Journal.

2004;4(2):60-71.

[84] Pilla S. Handbook of Bioplastics and Biocomposites Engineering Applications. US: John

Wiley & Sons Inc; 2011.

[85] Dweib MA, Hu B, Shenton Iii HW, Wool RP. Bio-based composite roof structure:

Manufacturing and processing issues. Composite Structures. 2006;74(4):379-88.

[86] Agopyan V, Savastano Jr H, John VM, Cincotto MA. Developments on vegetable fibre–

cement based materials in São Paulo, Brazil: an overview. Cement and Concrete Composites.

2005;27(5):527-36.

[87] Ghavami K. Bamboo as reinforcement in structural concrete elements. Cement and Concrete

Composites. 2005;27(6):637-49.

[88] Mohanty AK, Misra M, Drzal LT, Ebooks C. Natural Fibers, Biopolymers, and Biocomposites.

London: CRC Press; 2005.

[89] Pervaiz M, Sain MM. Sheet-Molded Polyolefin Natural Fiber Composites for Automotive

Applications. Macromolecular Materials and Engineering. 2003;288(7):553-7.

[90] Pilla S, Gong S, O'Neill E, Rowell RM, Krzysik AM. Polylactide-pine wood flour composites.

Polymer Engineering & Science. 2008;48(3):578-87.

REFERENCES

50

[91] Joseph K, Thomas S, Pavithran C. Effect of chemical treatment on the tensile properties of

short sisal fibre-reinforced polyethylene composites. Polymer. 1996;37(23):5139-49.

[92] Arbelaiz A, Fernández B, Ramos JA, Mondragon I. Thermal and crystallization studies of short

flax fibre reinforced polypropylene matrix composites: Effect of treatments. Thermochimica Acta.

2006;440(2):111-21.

[93] Bledzki AK, Reihmane S, Gassan J. Properties and modification methods for vegetable fibers

for natural fiber composites. Journal of Applied Polymer Science. 1996;59(8):1329-36.

[94] Yang P, Kokot S. Thermal analysis of different cellulosic fabrics. Journal of Applied Polymer

Science. 1996;60(8):1137-46.

[95] Prasad BM, Sain MM, Roy DN. Properties of ball milled thermally treated hemp fibers in an

inert atmosphere for potential composite reinforcement. Journal of Materials Science.

2005;40(16):4271-8.

[96] Thwe MM, Liao K. Durability of bamboo-glass fiber reinforced polymer matrix hybrid

composites. Composites Science and Technology. 2003;63(3):375-87.

[97] PK. M. Fiber reinforced composites materials, manufacturing and design. 3rd edition ed: CRC

Press; 2007.

[98] Thygesen A, Thomsen AB, Daniel G, Lilholt H. Comparison of composites made from fungal

defibrated hemp with composites of traditional hemp yarn. Industrial Crops and Products.

2007;25(2):147-59.

[99] Islam MS, Pickering, K. L. & Foreman, N. J. Curing kinetics and effects of fibre surface

treatment and curing parameters on the interfacial and tensile properties of hemp/epoxy composites.

Journal of Adhesion Science and Technology. 2009;23(16):2085-107.

[100] Ismail H, Nasir M, Mariatti M. Influence of Different Woven Geometry and Ply Effect in

Woven Thermoplastic Composite Behaviour-Part 2. International Journal of Polymeric Materials.

2000;47(2):499-512.

[101] Biresaw G, Carriere CJ. Interfacial tension of poly(lactic acid)/polystyrene blends. Journal of

Polymer Science Part B: Polymer Physics. 2002;40(19):2248-58.

[102] Sarazin P, Li G, Orts WJ, Favis BD. Binary and ternary blends of polylactide,

polycaprolactone and thermoplastic starch. Polymer. 2008;49(2):599-609.

[103] George G, Joseph K, Nagarajan ER, Tomlal Jose E, George KC. Dielectric behaviour of

PP/jute yarn commingled composites: Effect of fibre content, chemical treatments, temperature and

moisture. Composites Part A: Applied Science and Manufacturing. 2013;47:12-21.

[104] Almgren KM, Gamstedt EK, Berthold F, Lindstrom M. Moisture Uptake and Hygroexpansion

of Wood Fiber Composite Materials With Polylactide and Polypropylene Matrix Materials. Polymer

Composites. 2009;30(12):1809-16.

[105] Toro P, Quijada R, Murillo O, Yazdani‐Pedram M. Study of the morphology and mechanical

properties of polypropylene composites with silica or rice‐husk. Polymer International.

2005;54(4):730-4.

[106] Yuan Q, Wu D, Gotama J, Bateman S. Wood Fiber Reinforced Polyethylene and

Polypropylene Composites with High Modulus and Impact Strength. Journal of Thermoplastic

Composite Materials. 2008;21(3):195-208.

REFERENCES

51

[107] Kumar R, Yakubu MK, Anandjiwala RD. Biodegradation of flax fiber reinforced poly lactic

acid. Express Polymer Letters. 2010;4(7):423-30.

[108] Mofokeng JP, Luyt AS, Tábi T, Kovács J. Comparison of injection moulded, natural fibre-

reinforced composites with PP and PLA as matrices. Journal of Thermoplastic Composite Materials.

2012;25(8):927-48.

[109] Miao M, How YL, Cheng KPS. The Role of False Twist in Wrap Spinning. Textile Research

Journal. 1994;64(1):41-8.

[110] Nam TH, Ogihara S, Tung NH, Kobayashi S. Effect of alkali treatment on interfacial and

mechanical properties of coir fiber reinforced poly(butylene succinate) biodegradable composites.

Composites Part B. 2011;42(6):1648-56.

[111] Huda MS, Drzal LT, Mohanty AK, Misra M. Effect of fiber surface-treatments on the

properties of laminated biocomposites from poly(lactic acid) (PLA) and kenaf fibers. Composites

Science and Technology. 2008;68(2):424-32.

[112] Huda MS, Drzal LT, Mohanty AK, Misra M. Effect of chemical modifications of the

pineapple leaf fiber surfaces on the interfacial and mechanical properties of laminated

biocomposites. Composite interfaces. 2008;15(2-3):169-91.

[113] Tokoro R, Vu DM, Okubo K, Tanaka T, Fujii T, Fujiura T. How to improve mechanical

properties of polylactic acid with bamboo fibers. Journal of Materials Science. 2008;43(2):775-87.

[114] Saiman MP, Wahab MS, Wahit MU. The Effect of Fabric Weave on the Tensile Strength of

Woven Kenaf Reinforced Unsaturated Polyester Composite. In: Ahmad MR, Yahya MF, editors.

Proceedings of the International Colloquium in Textile Engineering, Fashion, Apparel and Design

2014 (ICTEFAD 2014): Springer Singapore; 2014. p. 25-9.

[115] Hill CAS, Abdul Khalil HPS. Effect of fiber treatments on mechanical properties of coir or oil

palm fiber reinforced polyester composites. Journal of Applied Polymer Science. 2000;78(9):1685-

97.

[116] Kannan TG, Wu CM, Cheng KB, Wang CY. Effect of reinforcement on the mechanical and

thermal properties of flax/polypropylene interwoven fabric composites. Journal of Industrial

Textiles. 2013;42(4):417-33.

[117] Park SD, Todo M, Arakawa K, Koganemaru M. Effect of crystallinity and loading-rate on

mode I fracture behavior of poly(lactic acid). Polymer. 2006;47(4):1357-63.

[118] Perego G, Cella GD, Bastioli C. Effect of molecular weight and crystallinity on poly(lactic

acid) mechanical properties. Journal of Applied Polymer Science. 1996;59(1):37-43.

[119] Prakash V. Shock and Impact Response of Glass Fiber-Reinforced Polymer Composites.

Polymer Composites: Wiley-VCH Verlag GmbH & Co. KGaA; 2012. p. 17-81.

[120] Qiang Yuan, Donglyang Wu, Gotama J, Bateman S. Wood Fiber Reinforced Polyethylene

and Polypropylene Composites with High Modulus and Impact Strength. Journal of Thermoplastic

Composite Materials. 2008;21(3):195-208.

[121] Graupner N, Herrmann AS, Müssig J. Natural and man-made cellulose fibre-reinforced

poly(lactic acid) (PLA) composites: An overview about mechanical characteristics and application

areas. Composites Part A. 2009;40(6):810-21.

REFERENCES

52

[122] Eichhorn SJ, Groom L, Hughes M, Hill C, Rials TG, Wild PM, et al. Review: Current

international research into cellulosic fibres and composites. Journal of Materials Science.

2001;36(9):2107-31.

Paper I

Composites: Part A 50 (2013) 93–101

Contents lists available at SciVerse ScienceDirect

Composites: Part A

journal homepage: www.elsevier .com/locate /composi tesa

Manufacture and characterisation of thermoplastic composites madefrom PLA/hemp co-wrapped hybrid yarn prepregs

1359-835X/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.compositesa.2013.03.012

⇑ Corresponding author. Tel.: +46 33 435 4497; fax: +46 33 435 4008.E-mail address: mikael.skrifvars@hb.se (M. Skrifvars).

Behnaz Baghaei a, Mikael Skrifvars a,⇑, Lena Berglin b

a School of Engineering, University of Borås, SE-501 90 Borås, Swedenb The Swedish School of Textiles, University of Borås, SE-501 90 Borås, Sweden

a r t i c l e i n f o

Article history:Received 10 October 2012Received in revised form 10 March 2013Accepted 15 March 2013Available online 26 March 2013

Keywords:A. Natural fibre compositesB. Mechanical propertiesC. Thermal analysisD. Compression moulding

a b s t r a c t

PLA/hemp co-wrapped hybrid yarns were produced by wrapping PLA filaments around a core composedof a 400 twists/m and 25 tex hemp yarn (Cannabis sativa L) and 18 tex PLA filaments. The hemp contentvaried between 10 and 45 mass%, and the PLA wrapping density around the core was 150 and 250 turns/m. Composites were fabricated by compression moulding of 0/90 bidirectional prepregs, and character-ised regarding porosity, mechanical strength and thermal properties by dynamic mechanical thermalanalysis (DMTA) and differential scanning calorimetry (DSC). Mechanical tests showed that the tensileand flexural strengths of the composites markedly increased with the fibre content, reaching 59.3 and124.2 MPa when reinforced with 45 mass% fibre, which is approximately 2 and 3.3 times higher com-pared to neat PLA. Impact strength of the composites decreased initially up to 10 mass% fibre; whilehigher fibre loading (up to 45 mass%) caused an increase in impact strength up to 26.3 kJ/m2, an improve-ment of about 2 times higher compared to neat PLA. The composites made from the hybrid yarn with awrapping density of 250 turns/m showed improvements in mechanical properties, due to the lowerporosity. The fractured surfaces were investigated by scanning electron microscopy to study the fibre/matrix interface.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

In recent years, natural fibre-reinforced polymer compositeshave been attracting attention from the viewpoint of reducingthe impact on the natural environment. Due to the increase inenvironmental consciousness, manufacturers are looking for moreecologically friendly bio-based materials for interior and outdoorproducts, and in these applications natural fibre-reinforcedcomposites could have great potential. Currently, the use ofthermoplastic resins in composites is clearly of higher potentialthan the use of thermoset resins because of their easier recycling,faster processing conditions (no time for curing is required), thepossibility of production in longer series, their lower cost, andthe absence of toxic solvents [1,2].

There are many thermoplastic polymers derived from renew-able raw materials, which are also biodegradable. Polylactic acid(PLA) is one such candidate, and it shows rather good propertiesthat are suitable for applications that do not require long-termdurability or elevated mechanical performance at higher tempera-tures. In order to make their possible use in many technical appli-cations more attractive, the mechanical properties of the PLA canbe enhanced by using reinforcements [2,3]. The reinforcement of

PLA with lignocellulosic fibres seems to be a logical alternative inorder to increase their mechanical performance and to preservethe environmentally friendly character of the final material [4].Hemp fibres can be considered to be a good choice for reinforcingpolymer composites, due to their high stiffness, strength, and as-pect ratio [5]. They also have an extremely high fibre yield per unitdensity and they are disease- and pest-resistant, enabling produc-tion methods with a low impact on the environment [6]. When thelife cycle for natural fibres such as hemp are evaluated they tend tohave a neutral CO2 balance, i.e. they release at decomposition asmuch CO2 as they have taken up during growth [7]. This has beenconfirmed by life cycle assessment of hemp fibres used in automo-tive applications [8,9]. Wötzel et al. [8] compared in a LCA hempreinforced fibre composite side panels and acrylonitrile butadienestyrene (ABS) copolymer composite side panels. The study showedthat the hemp fibre reduced the energy consumption by 45%compared to the ABS composite. Additionally the hemp fibre usedonly 5.3% of the cumulative energy demand, and same was foundfor the CO2.

Highly ordered textile reinforcements, such as interlaced wovenfabrics and unidirectional fabrics made from natural-fibre yarns,perform considerably better than random non-woven mats innatural-fibre composites [10]. A number of companies have intro-duced natural fibre composites, of these Libeco Lagae (Belgium),Lineo (Belgium) and NPSP (Netherlands) can be mentioned.

94 B. Baghaei et al. / Composites: Part A 50 (2013) 93–101

Furthermore, in a recently published technical report the mechan-ical and specific properties of flax and hemp used in polymer rein-forcement were discussed, and the suitability of flax and hempreinforcement was assessed, especially regarding the use in unidi-rectional reinforcements, fabrics, non-crimp fabrics, mats,prepregs, and compounds [11].

At present, the commercially available plant-fibre yarns are notintended for structural composites, but mainly for textiles, whichhave entirely different demands on the yarns. Thus, work is neededto tailor-make the best plant-fibre yarn for reinforcement ofcomposites. This also includes investigation of the possibility ofcombining plant-fibre yarns with the matrix polymer in fibre forminto one hybrid yarn (a composite preform), and how to do it(twisting or blending [12,13]). Previous studies [2,14,15] havedemonstrated that reinforcing hemp fibre increases the tensilestrength, Young’s modulus, impact strength, and flexural modulusof PLA biocomposites, which is a good indication of the compatibil-ity of hemp fibre with PLA. It is well known that fibres provide thehighest strength and stiffness when they are continuous andaligned in the direction of the applied load. Natural fibres arenaturally discontinuous and conventional spun staple yarns tendto be highly twisted, which leads to fibre misalignment and poorresin wet-out. The structured natural-fibre composites reportedso far are based on twisted yarns produced by long-establishedconventional spinning methods, mainly ring spinning [12,13,16].

In this paper, we report our work on improving the orientationof hemp fibres in composites by using our recent development ofco-wrapped yarn structures. This novel co-wrapped yarn consistsof low twist and very fine hemp yarns next to PLA filaments inthe core part, which are wrapped by PLA filaments. By varyingthe composition of hybrid yarn, it is possible to vary the hemp fibrecontent from 10 to 45 mass%. An exciting recent advancement hasbeen a new family of aligned natural-fibre reinforcements, whichhas overcome these issues by using ‘low twist’ yarns. We alsoreport the influence of fibre content and wrap density (numberof wraps per unit length) on the properties of composites.

2. Materials and methods

2.1. Materials

An air-textured PLA yarn with 72 filaments, provided byTorcitura Lei-Tsu S.r.l (Bergamo, Italy) was used as a the matrixmaterial. According to the supplier, the yarn was made from thePLA grade 6201D from NatureWorks�, Cargill Dow LLC (Minne-tonka, USA). This thermoplastic has a density of 1.24 g/cm3, a melttemperature of 160–170 �C, and a glass transition temperature of55–60 �C. The bleached hemp yarn (Cannabis sativa L), suppliedby Hempage AG, (Adelsdorf, Germany), was used in the core partof hybrid co-wrapped yarns. This staple fibre yarn have an averagefibre length in the range of 25–35 mm. The mechanical propertiesof the yarns used are shown in Table 1.

2.2. Methods

2.2.1. Manufacture of hybrid yarnsThe PLA/hemp hybrid yarns were produced by using a

laboratory yarn twist machine from DirecTwist, AGTEKS CO. Ltd.,(Istanbul, Turkey). In the co-wrapping, the PLA yarn filaments were

Table 1Mechanical properties of hemp and PLA yarns.

Raw yarn Linear density in tex Breaking tenacity in N/tex El

Hemp 25 0.35 ± 0.06 1.PLA 18 0.24 ± 0.01 7.

wrapped around low twist and very fine hemp yarns. These PLA fil-aments will melt during the consolidation process and will beincorporated in the matrix. Thus, they need to be very fine yarnsso that the interfacial adhesion between the reinforcing hempyarns and the PLA matrix is not affected to any significant extent.However, inhomogeneous distribution of the reinforcing hempand the PLA matrix yarns may lead to poor impregnation in theconsolidated composite [17]; it was therefore decided to put PLAfilaments in the core part of the hybrid yarn together with thehemp yarn. The core yarn was assembled by putting aligned hempyarns and PLA filaments into a bundle without increase the hempyarn twist. The yarn structure obtained from the co-wrap spinningused is shown in Fig. 1. The wrap density, or number of wraps perunit length (1 m), ranged from 150 to 250 in order to study theeffect of wrap density on the properties of the composite. The com-position of the hybrid yarns obtained is given in Table 2. The nameof the hybrid yarn indicates the composition. For example, in thehybrid yarn PLA/45 Hemp-150 one PLA filament is wrapped 150times around 1 m of the core part (composed of four hemp yarnsand two PLA filaments), thus giving a nominal ratio of 45 mass%hemp fibre content in the hybrid yarn. The linear densities of thehybrid yarns were determined by weighing a 10-m long yarn sam-ple which was dried in a vacuum chamber (0.9 mbar; 20 �C) for atleast 18 h, and calculating the tex value (1 tex = 1 g/1000 m).

2.2.2. Preparation of prepreg and compositesBefore compression moulding, multilayer 0/90 bidirectional

hybrid yarn prepregs were prepared by winding the hybrid yarnaround a 19 � 19-cm steel rectangular frame (Fig. 2). A 0/90 bidi-rectional layup of the prepreg was chosen, due to the ease of thewinding technique. Each prepreg had 12 layers of wound yarns,to achieve a composite thickness of about 1–3 mm. The prepregmats were first put in a vacuum chamber (0.9 mbar; 20 �C) for atleast 18 h, to dry before the compression moulding. The prepregwas covered with a Teflon sheet to prevent sticking of the matrixto the surface of the mould before being placed into a pre-heatedsteel mould with a 20 � 20-cm square cavity, and 10 mm depth.The steel mould was of own design, and machined by a localmachine shop. The composites were then prepared by pressingthe prepreg at a temperature of 175 �C and a pressure of1.04 MPa for 15 min in a hydraulic compression moulding machinefrom Rondol Technology Ltd. (Staffordshire, UK). Neat PLA sheetsto be used as reference material, were made by compressionmoulding under the same processing conditions of the same PLAfilaments formed to a similar prepreg.

2.2.3. Characterisation of compositesTo compare the mechanical and thermal properties of compos-

ites, tensile testing, dynamic mechanical thermal analysis (DMTA),differential scanning calorimetry (DSC), and scanning electronmicroscopy (SEM) were carried out. The specimens were storedat ambient conditions after processing, and then before testing,the specimens were conditioned for at least 24 h at 23 �C and50% relative humidity according to DIN EN ISO 291.

Statistical comparisons, based on linear regression analysis atthe 95% confidence level, were performed on the experimental datato test the effects of the various wrapping density and fibre loadingon the mechanical properties. Analysis of data was done usingMinitab (version 15).

ongation at break in% Initial modulus in N/tex Twists per metre

04 ± 0.19 33.14 ± 4.49 40020 ± 0.31 4.49 ± 0.83 –

Fig. 1. Structure of co-wrapped hybrid yarn.

Fig. 2. Prepreg frame with the PLA/hemp hybrid yarns in a 0/90 configuration.

B. Baghaei et al. / Composites: Part A 50 (2013) 93–101 95

2.2.3.1. Composite density and porosity. Composite density wasevaluated in order to assess the volumetric composition. Compos-ite and hemp fibre densities were determined by the buoyancymethod (Archimedes’ principle) using ethanol as the displacementmedium. Before the specimens were immersed in ethanol, theyhad been covered by a varnish containing paraffin to avoid absorp-tion of liquid during immersion. Based on the data obtainedregarding densities and mass fractions, the volume fractions offibre, matrix, and porosity were calculated by using the methodproposed by Madsen et al. [18]. The composite porosity contentwas estimated as the volume fraction not taken up by the fibreand matrix components.

2.2.3.2. Tensile test. The tensile testing was performed according tothe ISO 527 standard test method for the fibre-reinforced plasticcomposites, using a universal H10KT testing machine equippedwith a model 100R mechanical extensometer, both supplied byTinius Olsen Ltd. (Salford, UK). The loading rate was 10 mm/minand the load range was 1 kN. Six specimens were analysed for eachcomposite laminate. Cutting of the specimens was done with alaser machine. The specimens were cut in 20-mm-wide dumbell-shape with an overall length of 150 mm and a 60-mm length ofthe narrow, 10-mm-wide parallel-sided portion. The gauge lengthwas 50 mm and the initial distance between the grips was115 mm. The extensometer was attached to the central portionof the test specimen with clips.

2.2.3.3. Dynamic mechanical thermal analysis. DMTA on thecomposites was done using a Q-series TA instrument supplied byWaters LLC, Newcastle, DE, USA. The DMTA was run in the dual-cantilever bending mode and the sample dimensions were: thick-ness 1–2 mm, length 50 mm, and width 8 mm. The temperatureinterval was 15–150 �C with a heating rate of 3 �C/min using afrequency of 1 Hz and amplitude of 15 lm.

2.2.3.4. Flexural testing. The flexural testing was performedaccording to ISO 14125, with the same testing machine as for ten-sile testing. At least 5 specimens were tested for every compositecomposition. The loading rate was 10 mm/min and the load rangewas 1 kN. The specimen dimension was 60 mm � 15 mm(length �width), while the thickness varied depending on thesample. The outer span was taken to be 40 mm and the range ofdisplacement was 20 mm.

Table 2Composition of hybrid yarns and their properties.

Sample no. Wrap part Core part

PLA/45 hemp-150 1PLA 4H, 2PLAPLA/45 hemp-250PLA/35 hemp-150 2PLA 3H, 2PLAPLA/35 hemp-250PLA/20 hemp-150 3PLA 2H, 2PLAPLA/20 hemp-250PLA/10 hemp-150 4PLA 1H, 2PLAPLA/10 hemp-250

2.2.3.5. Impact testing. Charpy impact testing was done onun-notched specimens in accordance with ISO 179 using a pendu-lum type Zwick test instrument from Zwick GmbH (Ulm,Germany). A total of 10 specimens were tested edgewise to deter-mine the mean impact resistance.

2.2.3.6. Differential scanning calorimetry. The DSC analysis was doneon a DSC Q2000 supplied by TA Instruments, New Castle, DE, USA.Samples of 8–10 mg were heated in a nitrogen-purge stream at arate of 10 �C/min from 20 �C to 200 �C, cooled to 0 �C, and thenheated again from 20 �C to 200 �C. The data from the first scanwere used in order to see the effect of processing conditions onthe thermal properties. For each PLA sample, three replicates werescanned. The final results were the averages of three DSC runs. Thepercentage crystallinity (XDSC) of PLA was calculated using thefollowing equation [19]:

XDSC% ¼ DHf � DHcc

DH�f� 100

w

where DH�f = 93 J/g for 100% crystalline PLA, DHf is the enthalpy ofmelting, DHcc is the cold crystallisation enthalpy, and w is the massfraction of PLA in the composite.

2.2.3.7. Scanning electron microscopy (SEM). Composite fracturesurface morphologies were studied using low-vacuum scanningelectron microscopy using a JEOL JSM 6610LV instrument, JEOLLtd. (Tokyo, Japan) with an operating voltage of 5–10 kV. As lowvacuum SEM imaging was done, no specific preparation wasrequired.

3. Results and discussion

3.1. Influence of yarn structure

In this study, co-wrapped hybrid yarns were produced fromhemp and PLA yarns in four nominal mass ratios: 10/90, 20/80,35/65, and 45/50 with two wrapping densities: 150 and250 wraps/m. The mass fraction of the hemp fibre obtained in

Wrap density Linear densityin tex

Mass-percentageof hemp

150 191.5 46.2250 192.5 45.5150 190.0 35.5250 193.0 34.9150 185.8 23.3250 191.6 22.3150 182.7 11.3250 187.3 10.1

96 B. Baghaei et al. / Composites: Part A 50 (2013) 93–101

the hybrid yarn with a wrapping density of 250 was slightly lowerthan for the hybrid yarn with wrapping density of 150 because ofthe denser PLA filament wrapping (Table 2). The same natural-fibremass fractions were assumed for the compression-moulded com-posites since there was no loss of matrix during preparation ofcomposites. In total, 8 composites were manufactured with fibrevolume fractions in the range 8–38 vol-% and with porosity inthe range 6–9 vol-%. In the calculations, we used a density of1.48 ± 0.01 g/cm3 for the hemp yarn. The physical composition ofthe composites made are summarised in Table 3. From the results,it is possible to conclude that the higher the fibre volume fraction,the higher the amount of porosity—which is obviously due to thephysical structure of the prepreg. Similar trends have also beenseen in other studies [18,20]. Furthermore, it appears that theuse of lower wrapping density causes a slight increase in the poros-ity. However, it must be considered that the fibre volume fractionof the composite with a wrapping density of 150 is slightly higherthan for the composite with a wrapping density of 250, so thismight explain the increased degree of porosity. In the preparedcomposite, the porosity fraction is still high even if the fibrevolume fraction is low, which could be due to several factors.The complexity of the surface chemistry and the irregularity ofthe morphology of plant fibres is one of the most important consid-erations, in addition to the presence of luminal cavities [18]. Inaddition, in this case we also used PLA fibres as the matrix, andPLA used for melt spun fibres usually has a high molecular weight,and therefore lower melt viscosity. Composite porosity is known tobe linearly related to the logarithm of matrix viscosity [21]. Thishinders the penetration of PLA resin formed by melting of thePLA fibres between and around the hemp fibres to form a uniformmatrix, and it causes some of the porosity. Furthermore, porosity isalso affected by the processing techniques used. The effects of theprocessing factors such as preheating temperature on the mechan-ical properties of composites have been studied. Preheating wasfound to be the most important factor for good-quality moulding.However, shrinkage problems may be encountered when preheat-ing the prepreg yarns. The high draw ratio used in the meltspinning of the PLA fibres causes contraction of the matrix fibresand hence distortion of the fibre architecture when these fibresrelax on heating; therefore, preheating of the hybrid yarns at lowtemperature can also contribute to development of porosity [22].Svensson et al. [22] examined laminates made from co-wrappedyarns using aramid, carbon, and glass fibres as reinforcementsand spun polyamide and polyimide as matrix. They found thatlaminates manufactured from the co-wrapped yarns containedvoids and resin-rich pockets.

3.2. Tensile properties

An overview of the tensile modulus and strength of thecomposites compared to the values of the neat PLA matrix is given

Table 3Composition of the fabricated composites calculated from the density measurements.

Sample no. Density in g/cm3 Fibre mass fraction in% Fibre volum

PLA 1.24 0.0 0.0PLA/10 hemp-150 1.192 (0.007) 10.1 9.1 (0.1)PLA/10 hemp-250 1.193 (0.025) 11.3 8.1 (0.1)PLA/20 hemp-150 1.196 (0.004) 22.3 18.8 (0.1)PLA/20 hemp-250 1.197 (0.004) 23.3 18.0 (0.1)PLA/35 hemp-150 1.210 (0.009) 34.9 29.0 (0.2)PLA/35 hemp-250 1.219 (0.002) 35.5 28.7 (0.1)PLA/45 hemp-150 1.224 (0.006) 45.5 38.1 (0.2)PLA/45 hemp-250 1.226 (0.004) 46.1 37.7 (0.1)

Note: Data in table are mean (SD).

in Fig. 3a and b. It can be seen that an increase in fibre contentimproves the tensile strength, although the standard deviationfor each group of specimens was high, as shown by the error barsin Fig. 3. The statistical analysis of the data confirmed however asignificant increase in the modulus and the strength with theaddition of hemp fibre to the PLA-matrix (p-value < 0.001). Froma statistical viewpoint, the addition of hemp fibre to the PLA matrixwas found to have statistically significant contributions (at 5%significance level) on the response modulus and strength, and thisrelation is effectively modelled by a linear regression model. Inaddition, the statistical analysis demonstrated normal distributionof the values. The hemp yarn was bleached, which obviouslyenhances the quality of the fibre surface. This promotes betterinterfacial fibre-matrix adhesion and leads to better compositeproperties. It is also obvious from the results that composites madefrom the hybrid yarn with a wrapping density of 250 showed aconsistent improvement in tensile modulus over the compositeswith 150 wrapping density, across the whole range of fibre fractionratios investigated. At a low fibre fraction ratio (<25% by mass), theimprovement was about 4.6%; at a higher fibre fraction ratio (>40%by mass), the improvement increased to about 14.7%. This increasecould be attributed to improved fibre alignment. Increasing wrap-ping density improved the tensile modulus of the compositesbecause the tortuosity of the hybrid yarns decreases as wrap twistincreases, and consequently the reinforcement alignment would beimproved. At a low wrap twist, the buckling moment generated bythe wrapping filament is the main cause of yarn tortuosity. As wraptwist increases, the wrapping pitch (1/T) becomes smaller so thatthe bending span is shortened and the yarn core responds less tothe buckling action of the wrapping filament [23].

3.3. Flexural properties

The results of the flexural testing are shown in Fig. 4a and b. Boththe flexural strength and the modulus of hemp-reinforced PLA com-posites were all higher than for neat PLA, which is especially evi-dent for the 45 mass% fibre content with wrapping density of 250,with a flexural strength (124.2 MPa) 3.3 times higher than that ofneat PLA (37.7 MPa). The statistical analysis of the data confirmedsignificant increase in the flexural modulus and strength with theaddition of hemp fibre to the PLA matrix (p-value < 0.001). In addi-tion, the analysis demonstrated normal distribution of the results. Itis noteworthy that the above increase in flexural strength withrespect to the fibre content, which can be assigned to a good adhe-sion between the fibre and the matrix, is consistent with previouslypublished results on natural fibre/PLA composites. Hu and Lim [24]reported that the flexural strength of PLA/hemp composite in-creased with increase in hemp fibre content. Cao et al. [25] foundthat flexural strength and modulus increased with increase in fibrecontent for PLA/bagasse fibre composites. Okubo et al. [26] investi-gated PLA/bamboo and micro-fibrillated cellulose composites and

e fraction in% Matrix volume fraction in% Porosity volume fraction in%

100.0 –85.3 (0.5) 5.6 (0.6)86.5 (0.2) 5.3 (0.9)74.0 (0.2) 7.2 (0.3)75.0 (0.2) 6.9 (0.3)62.9 (0.5) 8.0 (0.7)64.0 (0.1) 7.3 (0.1)53.2 (0.3) 8.7 (0.4)53.9 (0.2) 8.4 (0.3)

Fig. 3. Tensile properties of PLA/hemp composites: (a) tensile modulus, and (b)tensile strength.

Fig. 4. Flexural properties of PLA/hemp composites: (a) flexural modulus, and (b)flexural strength.

B. Baghaei et al. / Composites: Part A 50 (2013) 93–101 97

Shih and Huang [27] conducted research on PLA/banana fibre com-posites, and they found improved flexural strength compared topure PLA. Ma and Joo [28] investigated PLA/jute composites andfound that the flexural strength and modulus of jute/PLA compos-ites increased with increase in the content of jute fibre. However,Shibata et al. [29] studied PLA/abaca composite and observed thatflexural strength decreased with increase in abaca fibre content.Similar trends have also been reported by Sawpan et al. [30] forPLA/hemp composite and by Serizawa et al. [31] for PLA/kenaf com-posite. In addition, the above finding is consistent with the tensilestrength results with PLA/hemp composites regarding fibre con-tent. The results show that there were differences in tensile andbending Young’s modulus values. It can be explained by the differ-ent oading modes in tensile and bending, and the fact that themanufactured composites have local defects which act as stressconcentration points, which causes local weakness. Furthermore,qualitatively, this can be explained from statistically consideringflaws and fractures and the fracture energy available in flexuralsamples under a constant rate of deflection as compared to tensilesamples under the same load conditions [32]. In the bending modeonly the inner (concave face) and outer (convex face) edges of thesample are at the largest stress of compressive and tensile respec-tively, and if these fibres are free from defects, the flexural strengthwill be controlled by the strength of these intact ‘fibres’. However, ifthe identical specimen is subjected to tensile forces then all the fi-bres in the composite are under the same stress and the failure willoccur when the weakest fibre reaches its limiting tensile stress.Therefore it is common for flexural strengths and modulus to behigher than tensile strengths and modulus for the same material[32]. A small difference in composite flexural strength and moduluswas found for the hybrid yarns with the two different wrapping

densities. At a fibre ratio of about 45% by mass, composites madefrom yarn with wrapping density of 250 had 10% higher flexuralstrength and 9% higher flexural modulus than composite fabricatedfrom yarn with 150 wrapping density. The composite made fromyarn with 250 wrapping density gave slightly higher values thanthe composite made from yarn with 150 wrapping density, whichis probably due to the lower porosity and more aligned fibres.

3.4. Impact resistance

The results of the Charpy impact test are shown in Fig. 5. It isapparent that the impact strength of the PLA/hemp compositesincreases with increasing mass proportion of fibre. But the valueat 10 mass% fibre content is still 8.6% lower than the value forpure PLA. Fibre pull-out is a major energy-absorbing mechanismsince the matrix deformation in the composites with 45 mass%fibre content will be lower than in the composites with10 mass% fibre content due to the lower composite failure strain[33]. The effect of fibre reinforcement on the impact strength ofcomposites is more complicated than bending and tensilestrengths, since the impact strength is attributed to the energyconsumption during failure. Higher interfacial strength doesnot always give higher impact strength. Medium or lower inter-facial strength is sometimes appropriate to increase the dissipa-tion of energy during fracture due to fibre pull-out. In this case,longer fibres are preferable. For fibre-reinforced composites, thelonger the fibre length, the more the energy consumption thatoccurs during fibre pull-out, as long as no breakage of fibre oc-curs. When longer fibres are introduced into the PLA matrix,the impact strength greatly increases due to the frequent occur-rence of long fibre pull-outs [34]. In our work, the higher impactstrength of composites compared to pure PLA can be attributedto the fact that the reinforcing staple fibres (which are between

Fig. 5. Impact strength of the composites relative to their proportion of fibre mass.

98 B. Baghaei et al. / Composites: Part A 50 (2013) 93–101

25 and 35 mm long) are in the form of yarn and can act as longfibres. It is notable that the above increase in impact strengthwith respect to the fibre content is consistent with some otherwork on natural fibre/PLA composites [35,36]. The compositesmade from yarn with wrapping density of 250 showed relativelylower impact strengths compared to the composite from yarnwith wrapping density of 150. This can be attributed to betterfibre–matrix adhesion, which was also obvious in the flexuraland tensile testing. Higher fibre–matrix adhesion resulted inshorter average pull-out lengths, and therefore caused lowerimpact resistance or strength [33].

Fig. 6. DMTA analysis of thermoplastic PLA reinforced with various proportions ofhemp fibre. (a) Storage modulus vs. temperature; and (b) tand curves.

3.5. Dynamic mechanical thermal testing

DMTA is a technique for applying stress to a material andmeasuring its response. Since polymers are viscoelastic by nature,the analysis will give information about the storage modulus andthe loss modulus. As the storage modulus is conceptually equiva-lent to the modulus of traditional mechanical testing, it gives ameasurement of the stiffness of the material. In this study, onlythe composites made from the hybrid yarns with a wrapping den-sity of 250 were analysed. The results are shown Fig. 6. The storagemodulus (at 20 �C) increased from 3.07 GPa for pure PLA up to6.2 GPa for the composite with 45 mass% fibre (Fig. 6a). It isevident that incorporation of hemp fibres results in an increasein the storage modulus of the biocomposite which reveals effectivestress transfer from the fibre to the matrix at the interface andgood adhesion between the fibre and the matrix. These resultsare consistent with previously published results on natural fibre/PLA composites [37–39].

The dampening, or tand, is the ratio between the loss modulusand the storage modulus and gives information about the internalfriction of the material. For a composite, the molecular motion inthe interface will contribute to the dampening. The dampening willconsequently give information about the adhesion of the interface.There is a general trend of the data in Fig. 6b, with reduced tandvalues with higher fibre ratio. The same trend has also beenobserved by other authors for various reinforcements [37–39]. Apossible explanation is that there is a strong interaction betweenthe fibre and the matrix. For a composite with a low fibre ratio,the polymer chains are relatively free to move. Increasing the fibre

Fig. 7. DSC heating thermograms for fabricated PLA/hemp composites from co-wrapped hybrid yarn with (a) wrapping density of 250, and (b) wrapping density of150.

B. Baghaei et al. / Composites: Part A 50 (2013) 93–101 99

ratio will decrease the mobility of the polymer chains and conse-quently reduce the dampening. This was discussed by Pothanet al. [40]. The peak in dampening takes place in the region ofthe glass transition where the material changes from a glass to arubbery state. The glass transition temperature is often recorded

Table 4Calorimetric data for PLA/hemp composites for the first heating run (10 �C/min).

Sample DHcc in J/g DHm in J/g

PLA – 52.29PLA/10 hemp-150 4.78 29.88PLA/10 hemp-250 3.92 34.94PLA/20 hemp-150 5.45 29.23PLA/20 hemp-250 4.28 28.84PLA/35 hemp-150 6.09 25.40PLA/35 hemp-250 4.50 26.21PLA/45 hemp-150 6.25 21.93PLA/45 hemp-250 5.02 21.39

Fig. 8. (a–e) SEM images (at different magnifications) of tensile fracture surfaces of PLA

at the maximum of the tand. Using this method of determiningthe glass transition temperature, the recorded Tg of the samplesranged from 66 �C for pure PLA up to 73 �C for almost the highestfibre ratio (45 mass%). It is obvious that incorporating fibres en-hances the Tg. The slight shift in Tg to higher temperature (by a

XDSC in% Tg in �C Tc in �C Tm in �C

55.71 54.03 112 168.8235.54 56.78 100.24 166.4837.06 57.13 100.34 166.7632.05 56.96 99.50 167.4033.02 57.17 98.50 167.1331.95 56.73 99.09 166.9431.67 56.37 98.54 166.4330.65 56.37 99.12 166.1128.18 57.18 97.58 166.52

/hemp composite with 45 mass% hemp/fibre content and wrapping density of 150.

100 B. Baghaei et al. / Composites: Part A 50 (2013) 93–101

few degrees) indicates that the mobility of polymer chains is af-fected. Mathew et al. [41] discussed that the shift to higher tem-perature usually indicates restricted movement of moleculesbecause of better interaction between the fibre and the polymermatrix.

3.6. Differential scanning calorimetry

The DSC heating thermograms of composites produced areshown in Fig. 7a and b. The glass transition temperature (Tg),crystallisation temperature (Tc), melting temperature (Tm), coldcrystallisation enthalpy (DHcc), and melting enthalpy (DHm)obtained from the DSC studies are summarised in Table 4. The dataindicate that the Tg and Tm of the PLA composites decreased less than2 �C with the addition of hemp fibre to the PLA matrix, however theDHm and Tc of the PLA composites decreased significantly in thepresence of hemp. These results suggest that hemp does not signif-icantly affect the crystallisation properties of the PLA matrix. Twomain factors control the crystallisation of polymeric compositesystems [42]. Firstly, the additives have a nucleating effect that re-sults in an increase in the crystallisation temperature, which has apositive effect on the degree of crystallisation. Secondly, additiveshinder the migration and diffusion of polymer molecular chains tothe surface of the growing polymer crystal in the composites, result-ing in a reduction in the crystallisation temperature, which has anegative effect on crystallisation. In this study, the crystallisationtemperature of the PLA/hemp composite decreased by up to about13–15 �C (for pure PLA compared to composites with 45 mass% fibrecontent), which signifies that the hemp fibres have a negative effecton crystallisation. Almost similar results were obtained in the case offabricated composites with wrapping density of 150 and 250. It isnoteworthy that the above decrease in crystallinity of PLA withrespect to fibre content is consistent with some other work onnatural fibre/PLA composites [27,37,39,43,44]. However, crystallisa-tion properties of PLA/hemp composites have been reportedelsewhere. For example, Masirek et al. [45], Pickering et al. [2], andSawpan et al. [14] studied hemp-reinforced PLA composites andthe crystallinity of PLA was found to increase as a result of thepresence of hemp, since the fillers acted as nucleating agents forthe crystallisation of the polymers.

3.7. Scanning electron microscopy analysis

SEM micrographs of the tensile fractured surface of the PLA/hemp biocomposites with a hemp fibre content of 45 mass% anda wrapping density of 150 are shown at different magnificationsin Fig. 8. It was observed from fracture surfaces that in some areasthe fibres and matrix did not adhere to each other adequatelyenough, showing little imprints of fibres on the matrix, as can beseen in Fig. 8a and b. Pulled-out fibres and the corresponding holesare visible in the composite (Fig. 8c). Furthermore, it can be seenthat the surfaces of the pulled-out fibres are partially clean(Fig. 8d and e). There are gaps between the fibres and the PLA,which could either occur because of debonding during mechanicaltesting or because of poor wetting during production of composite,which both indicates poor matrix/fibre adhesion. These observa-tions suggest that the adhesion between the matrix and the fibreis not optimal. It should be noted that unmodified hemp fibreswere used in the composites, and modification of fibres woulddefinitely lead to better adhesion between the matrix and naturalfibres. Heinemann and Fritz [46] predicted a good adhesionbetween natural fibres and PLA because of the hydrophilic natureof PLA. PLA has slightly polar oxygen atoms, which could formhydrogen bonds with the hydroxyl groups of the natural fibres.However, because of the results of this work and the work by otherauthors, it can be assumed that these hydrogen bonds only have a

small influence on fibre/matrix adhesion which leads to not opti-mal adhesion between the matrix and natural fibres.

4. Conclusions

The main objective of the present study was assessment of theproperties of 0/90 bidirectional composite made from PLA/hempco-wrapped hybrid yarn preforms as reinforcement, instead ofusing only hemp yarns impregnated with a melted PLA matrix.We investigated the mechanical and thermo-mechanical proper-ties of hemp-reinforced PLA composites. Compared to neat PLA,the tensile and flexural modulus and the strength of the PLA–hempcomposites were significantly higher as a result of the increased fi-bre content. Impact strength of the composites decreased initiallyup to 10 mass% fibre loading, but even higher fibre loading causedan improvement in impact strength. From the DMTA results, it isevident that incorporation of the fibres gives a considerableincrease in storage modulus and a decrease in tand values. Theseresults show the reinforcing effect of hemp on PLA matrix. Fromthe general trend in the results obtained, it can be affirmed thatco-wrapped hybrid yarn with lower wrapping density leads tolower mechanical properties in the composite. The studyperformed with DSC revealed that the glass transition temperatureand the melting point of PLA were not affected significantly afterreinforcement with hemp. The crystallisation temperature of thehemp-reinforced PLA composites decreased compared to purePLA, which indicates that the hemp fibres hinder the migrationand diffusion of PLA molecular chains to the surface of the nucleusin the composites. No noteworthy differences in calorimetric datafrom DSC for composites were observed between the hybrid yarnpreforms with different wrapping density. Future work willconcentrate on efforts to evaluate the biodegradability of thesedeveloping and promising composites.

Acknowledgments

Financial support from ÅForsk, Sweden, for this work is grate-fully acknowledged. The authors thank Jan Johansson, SwereaIVF, Mölndal, Sweden, for his assistance in performing the SEManalysis. Facius Nielsen of JN Textile Trading ApS, Denmark, isgratefully acknowledged for supplying the PLA filaments.

References

[1] Madsen B. Properties of plant fiber yarn polymercomposites. Roskilde: Technical University of Denmark; 2004.

[2] Pickering K, Sawpan M, Jayaraman J, Fernyhough A. Influence of loading rate,alkali fibre treatment and crystallinity on fracture toughness of random shorthemp fibre reinforced polylactide bio-composites. Compos. A2011;42(9):1148–56.

[3] Tserki V, Matzinos P, Panayiotou C. Novel biodegradable composites based ontreated lignocellulosic waste flour as filler. Part II. Development ofbiodegradable composites using treated and compatibilized waste flour.Compos. A 2006;37(9):1231–8.

[4] Girones J, Lopez J, Mutje P, Carvalho A, Curvelo A, Vilaseca F. Natural fiber-reinforced thermoplastic starch composites obtained by melt processing.Compos Sci Technol 2012;72(7):858–63.

[5] Mutjé P, Lòpez A, Vallejos M, López J, Vilaseca F. Full exploitation of Cannabissativa as reinforcement/filler of thermoplastic composite materials. Compos AAppl Sci Manuf 2007;38(2):369–77.

[6] Kostic M, Pejic B, Skundric P. Quality of chemically modified hemp fibers.Bioresour Technol 2008;99(1):94–9.

[7] Wambua P, Ivens J, Verpoest I. Natural fibres: can they replace glass in fibrereinforced plastics. Compos Sci Technol 2003;63(9):1259–64.

[8] Wötzel K, Wirth R, Flake M. Life cycle studies on hemp fibre reinforcedcomponents and ABS for automotive parts. Die Angew Makromol Chem1999;272(1):121–7.

[9] Schmidt W, Beyer H. Life cycle study on a natural fibre reinforced component.SAE Trans 1998;107:2095–102.

[10] Goutianos S, Peijs T, Nystrom B, Skrifvars M. Development of flax fibre basedtextile reinforcements for composite applications. Appl Compos Mater2006;13(4):199–215.

B. Baghaei et al. / Composites: Part A 50 (2013) 93–101 101

[11] Reux F, Verpoest I. Flax and hemp fibres: a natural solution for the compositeindustry. 1st ed. Paris, France: JEC, composites; 2012.

[12] Baets J, Plastria D, Ivens J, Verpoest I. Determination of the optimal flax fibrepreparation for use in UD flax–epoxy composites. In: Whysall C, editor.Proceedings of the 4th international conference on sustainable materials,polymers and composites. Studio Venue (Birmingham), UK: NetComposites;2011.

[13] Shah D, Schubel PJ, Clifford MJ. Modelling the effect of yarn twist on the tensilestrength of unidirectional plant fibre yarn composites. J Compos Mater2012(13 March).

[14] Sawpan M, Pickering K, Fernyhough A. Improvement of mechanicalperformance of industrial hemp fibre reinforced polylactide biocomposites.Compos A 2011;42(3):310–9.

[15] Sawpan M, Pickering K, Fernyhough A. Characterisation of hemp fibrereinforced Poly(lactic acid) composites. Int J Mater Prod Technol 2009;36(1–4):229–40.

[16] Zhang L, Miao M. Commingled natural fibre/polypropylene wrap spun yarnsfor structured thermoplastic composites. Compos Sci Technol2010;70(1):130–5.

[17] Alagirusamy R, Fangueiro R, Ogale V, Padaki N. Hybrid yarns and textilepreforming for thermoplastic composites. Text Prog 2006;38(4):1–71.

[18] Madsen B, Thygesen A, Lilholt H. Plant fibre composites – porosity andvolumetric interaction. Compos Sci Technol 2007;67(7):1584–600.

[19] Pilla S, Gong S, O’Neill E, Rowell R, Krzysik A. Polylactide-pine wood flourcomposites. Polym Eng Sci 2008;48(3):578–87.

[20] Thygesen A, Thomsen A, Daniel G, Lilholt H. Comparison of composites madefrom fungal defibrated hemp with composites of traditional hemp yarn. IndCrops Prod 2007;25(2):147–59.

[21] Madsen B, Lilholt H. Physical and mechanical properties of unidirectional plantfibre composites—an evaluation of the influence of porosity. Compos SciTechnol 2003;63(9):1265–72.

[22] Svensson N, Shishoo R, Gilchrist M. Manufacturing of thermoplasticcomposites from commingled yarns – a review. J Thermoplast ComposMater 1998;11(1):22–56.

[23] Miao M, How YL, Cheng KPS. The role of false twist in wrap spinning. Text Res J1994;64(1):41–8.

[24] Hu R, Lim J-K. Fabrication and mechanical properties of completelybiodegradable hemp fiber reinforced polylactic acid composites. J ComposMater 2007;41(13):1655–69.

[25] Cao Y, Shibata S, Fukumoto I. Fabrication and flexural properties of bagassefiber reinforced biodegradable composites. J Macromol Sci Part B Phys2006;45(4):463–74.

[26] Okubo K, Fuji T, Yamashita N. Improvement of interfacial adhesion in bamboopolymer composite enhanced with micro-fibrillated cellulose. JSME Int J Ser ASolid Mech Mater Eng 2005;48(4):199–204.

[27] Shih Y-F, Huang C-C. Polylactic acid (PLA)/banana fiber (BF) biodegradablegreen composites. J Polym Res 2011;18(6):2335–40.

[28] Ma H, Joo C. Structure and mechanical properties of jute–polylactic acidbiodegradable composites. J Compos Mater 2010;45(14):1451–60.

[29] Shibata M, Ozawa K, Teramoto N, Yosomiya R, Takeishi H. Biocomposites madefrom short abaca fiber and biodegradable polyesters. Macromol Mater Eng2003;288(1):35–43.

[30] Sawpan M, Pickering K, Fernyhough A. Flexural properties of hemp fibrereinforced polylactide and unsaturated polyester composites. Composites PartA 2012;43(3):519–26.

[31] Serizawa S, Inoue K, Iji M. Kenaf-fiber-reinforced poly(lactic acid) used forelectronic products. J Appl Polym Sci 2006;100(1):618–24.

[32] Dominick V, Nick R, Donald V, Marlene E, editors. Plastics Institute of Americaplastics engineering manufacturing and data handbook. Springer-Verlag;2001.

[33] Van D, Bos H, Molenveld K. Flax fibre physical structure and its effect oncomposite properties: impact strength and thermo-mechanical properties. DieAngew Makromol Chem 1999;272(1):71–6.

[34] Tokoro R, Vu DM, Okubo K, Tanaka T, Fujii T, Fujiura T. How to improvemechanical properties of polylactic acid with bamboo fibers. J Mater Sci2008;43(2):775–87.

[35] Ganster J, Fink H-P. Novel cellulose fibre reinforced thermoplastic materials.Cellulose 2006;13(3):271–80.

[36] Bax B, Müssig J. Impact and tensile properties of PLA/Cordenka and PLA/flaxcomposites. Compos Sci Technol 2008;68(7–8):1601–7.

[37] Huda MS, Drzal LT, Misra M, Mohanty AK, Williams K, Mielewski DF. A studyon biocomposites from recycled newspaper fiber and poly(lactic acid). Ind EngChem Res 2005;44(15):5593–601.

[38] Zhang Q, Shi L, Nie J, Wang H, Yang D. Study on poly(lactic acid)/natural fiberscomposites. J Appl Polym Sci 2012;125(S2):E526–E533.

[39] Majhi SK, Nayak SK, Mohanty S, Unnikrishnan L. Mechanical and fracturebehavior of banana fiber reinforced polylactic acid biocomposites. Int J PlastTechnol 2010;14(1):57–75.

[40] Pothan L, Oommen Z, Thomas S. Dynamic mechanical analysis of banana fiberreinforced polyester composites. Compos Sci Technol 2003;63(2):283–93.

[41] Mathew A, Oksman K, Sain M. Mechanical properties of biodegradablecomposites from poly lactic acid (PLA) and microcrystalline cellulose (MCC).J Appl Polym Sci 2005;97(5):2014–25.

[42] Kokta B, Dembele F, Daneult C, Carraher Jr, Sperling L, editors. Polymer scienceand technology, vol. 33. New York: Plenum; 1985.

[43] Lee S-H, Wang S. Biodegradable polymers/bamboo fiber biocomposite withbio-based coupling agent. Compos A Appl Sci Manuf 2006;37(1):80–91.

[44] Lee S-H, Wang S. Biodegradable polymers/bamboo fiber biocomposite withbio-based coupling agent. Composites Part A 2006;37(1):80–91.

[45] Masirek R, Kulinski Z, Chionna D, Piorkowska E, Pracella M. Composites ofpoly(L-lactide) with hemp fibers: morphology and thermal and mechanicalproperties. J Appl Polym Sci 2007;105(1):255–68.

[46] Heinemann M, Fritz H-G. Polylactid – struktur, eigenschaften und. 22.anwendungen. 19th Stuttgater Kunststoff-Kolloquium; 2005.

Paper II

Composites: Part A 61 (2014) 1–12

Contents lists available at ScienceDirect

Composites: Part A

journal homepage: www.elsevier .com/locate /composi tesa

Novel aligned hemp fibre reinforcement for structural biocomposites:Porosity, water absorption, mechanical performances and viscoelasticbehaviour

http://dx.doi.org/10.1016/j.compositesa.2014.01.0171359-835X/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Tel.: +46 33 435 4497; fax: +46 33 435 4008.E-mail address: mikael.skrifvars@hb.se (M. Skrifvars).

Behnaz Baghaei a, Mikael Skrifvars a,⇑, Masoud Salehi a, Tariq Bashir a, Marja Rissanen b, Pertti Nousiainen b

a School of Engineering, University of Borås, SE-501 90 Borås, Swedenb Department of Materials Science, Tampere University of Technology, P.O. Box 589, FI-33101 Tampere, Finland

a r t i c l e i n f o

Article history:Received 28 August 2013Received in revised form 29 January 2014Accepted 31 January 2014Available online 8 February 2014

Keywords:A. Fabrics/textilesB. Mechanical propertiesB. PorosityE. Compression moulding

a b s t r a c t

This paper examines the thermal and mechanical behaviour as well as moisture absorption of alignedhemp composites using hemp/PLA wrap spun yarns. Uniaxial composites were fabricated with 30 mass%hemp using compression moulding. The properties of composites in terms of hemp fibre orientation(aligned and random), off-axis angle and alkali treatment were investigated. It was found that the testingdirection influenced the mechanical properties of the composites. Compared with all the fabricated com-posites, the aligned alkali hemp/PLA yarn composite possessed the best mechanical properties, includingtensile, flexural and impact strengths, lower porosity and water absorption. The water absorption for allcomposites was higher than for neat PLA, both at room temperature and 80 �C. The PLA in its treated com-posites had higher crystallinity, which was attributed to effective heterogeneous nucleation induced byhemp. Based on SEM observation and theoretical analysis of DMTA data, there was a favourable interfa-cial adhesion in all composites.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

The potential of natural fibres as reinforcement in compositematerials is well recognized due to their attractive mechanicalproperties which enhance the possibility of producing eco-friendlymaterials. Natural fibres such as hemp and flax are already used inthe automotive industry to reduce weight, cost and environmentalimpact. Hemp is an upcoming European industrial crop [1], withgood mechanical fibre properties which can be cultivated with alow consumption of fertilizers and almost no pesticides [2]. Con-cerning the matrix in the composites, the industrial trend for nat-ural fibre composites is giving more importance to a thermoplasticmatrix, rather than a thermosetting matrix [1,3].

Polylactic acid (PLA) is the most important biothermoplastic forapplications requiring biodegradability. It shows also quite goodproperties appropriate for applications that do not require long-term durability or high mechanical performance at higher temper-atures. The mechanical properties of the PLA can be improved byusing reinforcements like natural fibres in order to increase its po-tential use in many industrial applications [4,5]. Therefore, because

of the attractive properties of hemp fibre, it was used as reinforce-ment for PLA composite in the presented study.

The main application of natural fibres is today mainly in non-structural composites as they are mostly available as randomly ori-ented nonwovens [1,3,6]. The fibre orientation (i.e. alignment ofthe fibres) must be controlled to ensure that the fibre mechanicalproperties are efficiently utilized in order to attract industrialinterest as an alternative to the traditionally applied synthetic fi-bres (e.g. glass fibres). It is evident that PLA/hemp fibre compositescan compete with glass fibre composite regarding stiffness,whereas for tensile and impact strength, the properties are stillnot on a satisfactory level [7–12]. Previous studies have demon-strated that the full reinforcement potential of natural fibres canbe exploited in bio-composites if an aligned fibre orientation isused [12,13]. Natural fibres are naturally discontinuous; thereforenatural fibre reinforcements reported so far are based on twistedspun staple yarns, which are produced by spinning methods,mainly ring spinning. These spun yarns tend to be highly twisted,which leads to fibre misalignment due to their helical paths aroundthe yarn axis. This misalignment contributes negatively to themechanical properties of the resultant composites. Another nega-tive impact of yarn twist is that it tightens the yarn structure, ren-dering resin impregnation difficult [14]. Therefore, in the textileindustry a broad range of techniques for the alignment of natural

2 B. Baghaei et al. / Composites: Part A 61 (2014) 1–12

fibres have been developed and optimised to produce yarns withcontrolled fibre orientations by reducing or replacing twist inyarns. Goutianos and Peijs [14] tried to produce flax yarns withthe minimal level of twist for manufacturing aligned composites.Shah et al. [15] used a sizing agent to substitute the use of twistin roving and yarn. Zhan and Miao [13] studied the effect ofwrapped spun yarn with low twist for reinforcement purpose.

Our previous study [12] investigated the mechanical propertiesof composites manufactured from PLA/hemp co-wrapped hybridyarn prepregs. Here we used continuous PLA filaments, which wereused to wrap a low twist hemp yarn. The composites made fromthe hybrid yarn with higher wrapping density showed improve-ments of mechanical properties due to lower porosity. However,the porosities of composites were between 6 and 9 vol.%, and theporosity fraction was still high even if the fibre volume fractionwas low, which could be due to several factors. Porosity is difficultto avoid in natural fibre composites and influences on the compos-ite properties, yet how to control the porosity has so far only re-ceived limited attention. Thus, research on how to decrease theamount of porosity is warranted.

In the current paper, we discuss the development new hybridyarns with low twist for high performance natural fibre-reinforcedcomposites suitable for use in structural or semi-structural appli-cations and with lower amount of porosity. The overall was tostudy the mechanical properties of these novel aligned hemp fibreyarn composites and investigate the effect of a range of relevantparameters such as prepreg type such as nonwovens and hybridyarn prepregs with different off-axis angles (0�, 45� and 90�) andfibre treatment.

2. Materials and methods

2.1. Materials

Two types of staple fibres were used in this study: hemp andPLA fibres. The PLA staple fibre, provided by Trevira GmbH (Hat-tersheim, Germany), had a fineness of 1.7 dtex and a mean fibrelength of 38 mm. Based on the manufacturer’s information; thePLA fibres were made from PLA Polymer 6202D from Nature-Works�, Cargill Dow LLC (Minnetonka, USA). This thermoplastichas a density of 1.24 g/cm3, a melt temperature of 160–170 �C,and a glass transition temperature of 60–65 �C. The hemp in theform of baled loose staple fibres (genus species Cannabis SativaL) was supplied by Hempage AG (Adelsdorf, Germany). Accordingto the manufacturer’s information, the average diameter of thehemp fibre was 20–40 lm and had a mean fibre length of30 mm. The hemp fibres were treated by 4 wt% NaOH solutionfor 1 hr, rinsed with distilled water until it was neutral and finallydried at room temperature for 48 h. In addition, a 18-tex PLA mul-tifilament yarn, provided by Trevira GmbH (Hattersheim, Ger-many), was used as wrapping yarn in the wrap spun yarns.

Fig. 1. Structure of wrap spun hybrid yarn.

2.2. Methods

2.2.1. Wrap spinning of hybrid yarnsThe preferred yarn structure has the reinforcing fibres straight

and parallel to the yarn axis. Wrap spinning can be used to producesuch a yarn [13]. PLA/hemp hybrid wrap spun yarns were producedby using a laboratory spinning machine from Mesdan S.p.A., (Bre-scia–Italy) and a laboratory yarn twist machine from DirecTwist,AGTEKS Co., Ltd., (Istanbul, Turkey). The hemp and PLA fibres ar-rived at our laboratory in baled loose fibre form. The PLA fibreand the hemp fibre were weighed to the desired proportion(30 mass%) and the fibre mixture was then fed into the carding ma-chine. During carding, the longer PLA fibre supported the shorter

hemp fibre and provided the necessary fibre-to-fibre cohesion toget a web suitable for further processing. The blended PLA/hempweb was carded three times to parallelize the fibres and achievesliver uniformity. Then the sliver was fed through a roving frame,where the strands of fibre were further elongated. The sliver wasdrawn twice after carding to achieve the required roving lineardensity. Although it was possible to create hybrid yarns with lowtwist, the cohesion of the fibres was very low because PLA/hemproving had a false twist, which means that they could not form aroving of sufficient integrity. Moreover, the rowing is too weak tobe able to be collected alone in the roving machine. In order to col-lect the roving without causing breakage in the roving machine,the processable PLA filaments were used as a processing carrierfor the PLA/hemp roving in the final step. Then the roving waswrapped by PLA filaments in the twisting machine. The wrap yarnwas spun to the nominal count of 550 tex, and it had a wrappingintensity of 200 turns/m. These wrappings provide better protec-tion for the reinforcing fibres during further processing, such asweaving [16] or making a prepreg. The yarn structure obtainedfrom the wrap spinning used is shown in Fig. 1.

2.2.2. Preparation of prepreg and compositesBefore compression moulding, multilayer unidirectional pre-

pregs were prepared by winding the hybrid yarn around a19 � 19 cm2 rectangular steel frame. The off-axis fibres were ori-ented at different angles including 0�, 45� and 90�. In order toinvestigate the effect of random fibre orientation on composites,non-woven PLA/hemp prepreg was produced through cardingand the blended PLA/hemp web was carded three times. The pre-preg mats were first dried in a vacuum chamber (0.9 mbar;70 �C) for at least 18 hours before compression moulding. The pre-preg was then covered by a Teflon sheet to prevent sticking of thematrix to the surface of the mould, and then it was placed into apre-heated steel mould with a 20 � 20-cm2 square cavity, and10 mm depth. The composites were formed by pressing the pre-preg at 195 �C and at 1.7 MPa for 15 min. The thickness of pro-duced composite samples was between 2 and 3 mm. Neat PLAsheets to be used as reference material were made by meltingPLA fibres under the same processing conditions. Specimens forthe mechanical testing were cut by GCC LaserPro Spirit laser cut-ting machine according to the standards given below. Before per-forming the testing, the specimens were conditioned for at least

B. Baghaei et al. / Composites: Part A 61 (2014) 1–12 3

24 hours at 23 �C and 50% relative humidity according to DIN ENISO.

2.2.3. Single fibre tensile testThe hemp fibres were tested on a Favigraph single fibre tensile

tester from Textechno GmbH (Mönchengladbach, Germany)equipped with a 20 cN load cell. The gauge length was 20 mm,the test speed was 20 mm/min and 20 replicated tests were done.To simulate the effect of composite processing conditions on thehemp characteristics, single hemp fibres were treated in a hotpress at 190 �C and 1.7 MPa for 15 min, and then again tested fortensile properties. These processing conditions are identical tothose used for the fabrication of the composites presented in Sec-tion 2.2.2.

2.2.4. Composite density and porosityThe densities of the composites were determined by the buoy-

ancy method using ethanol as the displacement medium. In orderto avoid absorption of liquid during immersion in ethanol, thespecimens were covered by a varnish containing paraffin. Usingthe method proposed by Madsen et al. [17], the fibre volume frac-tion was calculated from the fibre weight fraction take in porosityinto account.

2.2.5. Water absorption testWater absorption analysis was done on the composite speci-

mens according to ASTM D570-98. The specimens were dried inan oven for 24 h at 60 �C, and then cooled them down to room tem-perature in a desiccator. The weight of these specimens was de-noted as W0. The specimens were immersed in two differentbaths: one of distilled water at room temperature and one distilledwater bath at 80 �C. The amount of water absorbed was measuredevery 24 hours for 10 days. At each measurement, the specimenwas taken out of the water and the surface wiped dry before theweight was recorded as W. For each composite, five replicatedsamples were measured. The percentage of apparent weight gain(WG) was then calculated using Eq. (1).

WG ¼ ½ðW �W0Þ=W0� � 100% ð1Þ

2.2.6. Mechanical testingTensile testing was done according to the ISO 527, using a uni-

versal H10KT testing machine equipped with a mechanical exten-someter (model 100R long travel extensometer) supplied by TiniusOlsen Ltd. (Salford, UK). The testing parameters were: 10 mm/minfor loading rate, and 1 kN for loading range. The extensometer wasattached to the central portion of the test specimen with clips.

Flexural testing was performed on the same testing machineaccording to ISO 14125. At least five specimens were tested foreach batch of samples. The loading rate was 10 mm/min and theload range was 1 kN. The outer span was taken to be 64 mm fordiscontinuous-fibre-reinforced composites and 40 mm for unidi-rectional composite, and the range of displacement was 20 mm.

To investigate the toughness of the composites, an un-notchedCharpy impact strength test was carried out according to ISO179 179, using a pendulum type Zwick test instrument from ZwickGmbH (Ulm, Germany). A total of 10 specimens were tested edge-wise to determine the mean impact resistance.

2.2.7. Dynamic mechanical thermal analysisThe time-temperature dependency of the mechanical proper-

ties was determined by DMTA, using a Q-series instrument sup-plied by TA Instruments, Newcastle, DE, USA. The DMTA was runin the dual-cantilever bending mode, whereas the temperaturerange was from 30 to 150 �C with a heating rate of 3 �C/min, using

a frequency of 1 Hz and amplitude of 15 lm. At least 3 specimenswere tested for every composite composition.

2.2.8. Differential scanning calorimetryThe DSC analysis was done on a DSC Q2000 supplied by TA

Instruments, Newcastle, DE, USA. Samples of �10 mg were heatedat 10 �C/min under nitrogen between 20 �C to 200 �C, cooled to0 �C, and then heated again from 20 �C to 200 �C. The data fromthe first scan were used. The percentage crystallinity (XDSC) ofPLA was calculated using Eq. (2) [18].

XDSC% ¼ DHf � DHcc

DH0f

� 100w

ð2Þ

where DH0f ¼ 93 J=g for 100% crystalline PLA, DHf is the enthalpy of

melting, DHcc is the cold crystallisation enthalpy, and w is theweight fraction of PLA in the composite.

2.2.9. Scanning electron microscopy (SEM)Fibres and composite fracture surface obtained from the tensile

testing and upper surface morphologies were studied using low-vacuum scanning electron microscopy with a Quanta 200 ESEMFEG instrument from FEI, (Oregon, USA) with an operating voltageof 10–15 kV. For low vacuum imaging, no specific preparation wasrequired.

3. Results and discussion

3.1. Single fibre tensile test

In order to evaluate the effect of moulding conditions, themechanical properties of heat and alkali treated single hemp fibreswere determined, see Table 1. The alkali treatment improves fibretensile strength and modulus which helps to improve the proper-ties of the composites [19,20]. This treatment possibly orients fi-brils along the direction of tensile forces by removinghemicelluloses from fibre, resulting in better load sharing betweenthe fibrils [21]. Heat treatments greatly reduce mechanical proper-ties. This could be due to thermal degradation of cellulose chains[22].

3.2. Composite porosity

Composite fabrication by compression moulding requires thatthe fibre and the thermoplastic matrix are well mixed before beingprocessed into composites. Such fibre/matrix mixing can be doneby fibre opening and web formation which is a technique well sui-ted for non-woven fibre mats. However, in the case of hybrid yarnsfor preparing unidirectional prepregs, fibre/matrix mixing can alsoachieved by fibre opening and web formation, which is followed byspinning. A spinning method known as wrap spinning can be usedto produce such a yarn, consisting of low twist staple fibreswrapped by a matrix filament [13]. In the current study, compos-ites were produced from PLA/hemp nonwoven fibre mats, PLA/hemp yarn prepregs with 3 off-axis fibre orientations (0�, 45�and 90�) and PLA/alkali treated hemp yarn prepregs in a nominalmass ratio of 30% reinforcement. When 70% or more PLA fibrewas used, the staple fibre mixture could be successfully spun intoa hybrid yarn with uniform distribution of the fibres in composite.Because the fibre angle does not have any effect on the compositecomposition, only the composite of 0� off-axis fibre was tested. Intotal, the manufactured composites had porosity in the range 1.23–16.7 vol.%. In the calculations, we used a density of 1.48 ± 0.01 g/cm3 for the hemp yarn. The obtained physical composition of thecomposites produced are summarised in Table 2. From the results,it is possible to conclude that the composite made from PLA/hemp

Table 1Hemp fibre properties before and after pressing and treatment.

Treatment Tenacity in cN/dtex E-Modulus in cN/dtex Elongation in % Linear density dtex

As received 4.66 (1.43) 97.71 (24.92) 3.93 (0.66) 4.15After hot pressing 3.75 (1.10) 84.62 (26.60) 3.37 (0.88) 4.27After alkali treatment 5.18 (1.30) 99.15 (32.64) 4.05 (0.86) 3.75

Table 2Composition of the fabricated composites calculated from the density measurements.

Sample Density (g/cm3)

Fibre mass fraction(%)

Fibre volume fraction(%)

Matrix volume fraction(%)

Porosity volume fraction(%)

PLA 1.2490(0.0002)

0.0 0.0 100.0 –

Alkali hemp/PLA yarn (off-axis angle0�)

1.2943(0.0019)

30.0 26.23 (0.03) 72.53 (0.10) 1.23 (0.15)

Hemp/PLA yarn (off-axis angle 0�) 1.2713(0.0268)

30.0 25.77 (0.54) 71.25 (1.50) 2.97 (2.04)

Hemp/PLA nonwoven 1.0914(0.0154)

30.0 21.12 (0.31) 61.16 (0.9) 16.71 (1.17)

Note: Data in table are mean (SD).

4 B. Baghaei et al. / Composites: Part A 61 (2014) 1–12

nonwoven had the higher porosity. The formation of voids can beattributed to many factors, the main one being entrapped air inthe prepreg, which is not released during compression moulding.The size of voids is affected by the available pore size and their po-sition along the melted resin flow path [23]. The tortuosity of theflow path is an important factor in the process of void movementsand their elimination. The nonwoven mat fibre has a more tortuousflow path compared with the composite made from hybrid yarnprepregs and, hence, it is more difficult for the entrapped air toflow out of the system. The formation of a stronger interface by im-proved hemp fibre and PLA adhesion could be the result of the re-duced porosity of the composite from alkali treated hempcomposites compared with the untreated hemp/PLA composites.This could be explained as an effect of the alkali treatment. Thiswill increase the amount of available hydroxyl groups [19] dueto removal of the noncellulosic materials covering the cellulose hy-droxyl groups. The fibre surface is also cleaned from impurities,which leads to the exposure of the neat and rough fibre surfacewith many pits (Fig. 2), which will largely increase the surface areaof the fibre. In addition to increasing the number of hydroxylgroups for hydrogen bonding, increased surface roughness wouldalso enable better mechanical interlocking with PLA [24]. The lackof interfacial interaction leads to porosity and increases theamount of moisture absorption [21,25].

Fig. 2. SEM micrographs of the surface morphology of

3.3. Water absorption characteristics

The apparent weight gain (WG) as a function of the square rootof immersion time (

ffiffi

tp

) at different temperatures for the manufac-tured composites are shown in Fig. 3a and b. In Fig. 3a, it can beseen that for all the composites investigated, WG increases mono-tonically by

ffiffi

tp

. This increase in WG is consistent with some otherstudies on natural fibre composites [3,6,12]. The absence of theinduction period characterized by zero weight gain at the initialstage of water immersion treatment, which is explained in termsof the different material packing between the skin (which is in con-tact with the hot metal mould surface) and core regions, indicatesthat the hemp fibres were uniformly dispersed in the PLA matrix inthe current study [26]. It can also be seen that after reaching themaximum value, WG decreases when immersion time is increased.Before reaching the maximum WG, the slopes of the WG vs.

ffiffi

tp

plotswere higher for composites compared with neat PLA. It can be ob-served that the fibre-based composites showed significantly higherwater absorption than did neat PLA due to the hydrophilic natureof hemp with polar groups such as hydroxyl and carboxyl groupson the fibre surfaces. These results are in agreement with pub-lished data, showing that the lignocellulosic fibres displayed ahigher tendency to absorb water than does the hydrophobic PLA[6,12]. SEM images of the fracture surface morphology of compos-

: (a) untreated and (b) alkali treated hemp fibres.

0

5

10

15

20

25

0 1 2 3 4 5 6 7 8 9 10111213141516

App

aren

t w

eigh

t ga

in (

wt%

)

Square root of immersion time (hr1/2)

Hemp/PLA nonwoven

Hemp/PLA (0) yarn

Alkali hemp/PLA yarn

Neat PLA

0

5

10

15

20

25

0 1 2 3 4 5 6 7 8 9 10111213141516

App

aren

t w

eigh

t ga

in (

wt%

)

Square root of immersion time (hr1/2)

Hemp/PLA nonwoven

Hemp/PLA (0) yarn

Alkali hemp/PLA yarn

Neat PLA

(a)

(b)

Fig. 3. Apparent weight gain against square root of immersion time for the differentPLA/hemp composites at: (a) room temperature and (b) 80 �C. For neat PLA no datawas recorded after 5 days, due to severe softening of the specimens.

B. Baghaei et al. / Composites: Part A 61 (2014) 1–12 5

ites made of the nonwoven hemp–PLA and the alkali treated hemp/PLA are shown in Fig. 4. Compared with all composites, compositefrom PLA/hemp nonwoven showed the highest water absorptions.It can probably attributed to the fact that it has less close and com-pact packing of hemp fibres in the PLA matrix compared with thecomposite made from PLA/hemp yarn (Fig. 4a). The compact andclose packing of PLA/hemp yarns reduces the porosity inside thecomposites, which in turn contributes to the reduction in waterabsorption [27]. As a result of alkali treatment, the interfacial adhe-sion in PLA/hemp fibres improved (Fig. 4b), which is necessary forthe reduction of interfacial wicking of the water molecules. Thus,the moisture absorption of the composites can be reduced bychemical treatments [14,27,28]. Furthermore, as shown in Fig. 3b, it can be seen that for all the composites investigated, WG in-

Fig. 4. SEM images of the fracture surface morphology of composites m

creases monotonically byffiffi

tp

initially before reaching a maximum.It can be observed that the water immersion temperature doeshave an influence on the water absorption curves. Increasing theimmersion temperature from room temperature to 80 �C led to in-creased water absorption of the neat PLA and composites as well asa shortening of the saturation time. After 5 days at 80 �C, a massloss of almost 20% was observed for neat PLA. At this point, theneat PLA sample was very soft, and it was difficult to decant waterand rinse the sample without losing some material; therefore, thewater immersion for neat PLA was discontinued after 5 days. Theeffect of immersion temperature was sharper on PLA/hemp non-woven than on PLA/hemp yarn composites. The weight gain wasfound to decrease after passing through a maximum. For the com-posites, this could be due to the neat PLA layer peeling on the sur-face as a result of biodegradation and dissolving with time togetherwith the removal of some substances from the hemp fibres duringthe immersion as can be seen in Fig. 5a and b, SEM images of theupper surface of the alkali hemp/PLA specimens before and afterbeing subjected to water immersion treatment [29,30].

3.4. Tensile properties

The tensile strength, modulus and elongation at break of thecomposites are compared with the values of the neat PLA matrixin Table 3. It was found that the composites fabricated from PLA/hemp yarns (both untreated and treated) had considerably highertensile strength and modulus in the principal fibre direction (i.e.,longitudinal direction) than in the corresponding nonwoven mats.Compared to the tensile modulus of the pure PLA matrix, there isan improvement of 255% for treated hemp/PLA yarn, 203% for un-treated hemp/PLA with fibre orientation angle 0�, and 86% for thePLA/hemp nonwoven composites. As expected, the compositestrength and stiffness decreased with increasing fibre orientationangle. When the external load is directed at a significant angle tothe primary fibre axes, large reductions in key composite proper-ties such as ultimate tensile strength can occur [31]. The tensileproperties in the perpendicular direction (90�) for the PLA/hempyarn composites were in almost the exact opposite order to thosein the principal fibre direction (0�). This was entirely within expec-tation because the prepreg with the highest fibre orientation alongthe principal fibre direction would have the lowest fibre orienta-tion in the perpendicular direction [32,33]. The composite madefrom the nonwoven demonstrated higher tensile properties inthe principal fibre direction compared with hemp/PLA yarn com-posite with fibre orientation angles 45� and 90�. Also the resultsshowed that tensile strength of composites made with fibre orien-tation angles 45� and 90� were lower than pure PLA. The reason for

ade of: (a) nonwoven hemp–PLA and (b) alkali treated hemp/PLA.

Fig. 5. SEM images of the upper surface of the alkali hemp/PLA specimens: (a) before and (b) after being subjected to water immersion treatment.

Table 3Tensile properties of PLA/hemp composites.

Sample Tensile strength (MPa) Young’s modulus (GPa) Elongation at break (%)

PLA 41.21 (2.25) 2.91 (0.39) 1.92 (0.26)Hemp/PLA yarn (off-axis angle 0�) 72.75 (6.26) 8.77 (1.44) 1.24 (0.36)Hemp/PLA yarn (off-axis angle 45�) 34.75 (4.63) 4.62 (1.04) 0.66 (0.18)Hemp/PLA yarn (off-axis angle 90�) 22.01 (3.38) 3.70 (0.65) 0.53 (0.11)Hemp/PLA nonwoven 53.63 (1.22) 5.60 (0.95) 1.10 (0.30)Alkali hemp/PLA yarn (off-axis angle 0�) 77.08 (2.60) 10.27 (1.36) 1.58 (0.26)

Note: Data in table are mean (SD).

Fig. 6. Flexural properties of PLA/hemp composites.

6 B. Baghaei et al. / Composites: Part A 61 (2014) 1–12

this may be attributed to the fact that the on-axis properties arestrongly dependent on fibre properties, and that off-axis propertiesare strongly dependent on matrix properties, particularly on theirstiffness and load-carrying ability, which are typically related totheir porosity content. For example, when a composite with porousmatrices were tested off-axis, relatively weak strengths were ob-served because the porous matrix could not carry significant loadcompared to the pure matrix which has low porosity [34]. In thecase of treated hemp/PLA yarn composites, the stronger interfaceformed due to increased potential hydrogen bonding [7]. The high-est strength values were reached by treated hemp/PLA yarn withfibre orientation angle 0� with 77.1 MPa, followed by untreatedhemp/PLA yarn 0�, PLA/hemp nonwoven, PLA/hemp with fibre ori-entation angle and hemp/PLA yarn with fibre orientation angle 90�.As can be seen in Table 3, the hemp fibres evidently improve thetensile modulus, strength and reduce the elongation at break ofthe PLA. The decrease in elongation at break for the compositesis due to the destruction of the structural integrity of the PLA ma-trix, due to the loading of the natural fibre, and which leads to fas-ter fracture than for a pure matrix [35].

3.5. Flexural properties

The results of the flexural test are shown in Fig. 6. As can beseen, the flexural modulus of the composites is higher than thatof neat PLA, which is especially evident for the treated hemp/PLAyarn with a flexural modulus of 7.1 GPa. It is also obvious that com-posites made from PLA/hemp yarn (0� direction) showed an evi-dent improvement in flexural modulus over the compositesproduced from PLA/hemp nonwoven. However, the compositesfrom nonwoven exhibited the optimum tensile and flexuralstrength of 52.3 and 70.9 MPa, respectively. The reason for thereduction of mechanical properties could be the fact that all fibreswith off-axis orientation failed at the particular point, and cannot

contribute furtherer in the composite failure. So, the remainingnumber of on-axis (0� direction) fibres almost exhibited more than50% of the mechanical properties of uniaxial composites [36,37].The flexural properties of unidirectional composites followed thereduction trend in their value when the fibre orientation changedfrom 0� to the off-axial direction. The lowest value of flexuralstrength was observed for the composites with fibre orientationangle 90�. Flexural strength and modulus value were 36.1 MPaand 4.5 GPa, respectively. Alkali treatments enhanced the flexuralstrength and modulus of the composites, which is consistent withother studies [38–40]. The superior flexural properties of the trea-ted fibre composites can be attributed to the greater interfacialbonding between treated hemp fibres and PLA [41,42]. These re-sults are consistent with the tensile properties described previ-ously. Moreover, our research work found the highest flexural

B. Baghaei et al. / Composites: Part A 61 (2014) 1–12 7

strength compared with the tensile strength for different fibre ori-entation. The main reason could be that the volume of materialbeing subjected to the maximum stress is smaller for flexural testthan for the tensile. So, the presence of the critical defects is muchlower than for tensile test [37,43,44].

Fig. 7. Impact strength of the composites relative to their proportion of fibre mass.

Fig. 8. DMTA analysis of PLA reinforced with hemp fibres of various orienta

3.6. Impact resistance

The results of the Charpy impact test are shown in Fig. 7. For theneat PLA matrix, impact strength of 11.5 kJ/m2 was measured. Asignificant reinforcement effect was determined for the treatedPLA/hemp yarn composite with an impact value of 18.8 kJ/m2

and for untreated PLA/hemp yarn composite (0�) with the valueof 16.0 kJ/m2, while the value of the PLA/hemp nonwoven compos-ites (9.7 kJ/m2) was lower than that of the neat PLA matrix which issimilar to flexural strength analysis. Further increment of fibres an-gle caused a decrease in impact strength of the composites, whichis likely to be due to decreased dissipation of energy by less fibrepull-out. For an equivalent amount of hemp fibres (30 mass%),the impact strength of the nonwoven composites was 128% and237% higher than that of the yarn composites with 45� and 90�off-axis. Similar to the tensile properties, it can be observed thatthe alkali treatments enhanced the impact strength. This findingis in agreement with other studies [38–40,45]. In the case of alkalitreated fibres, the dissipation of energy by fibre pull-out is muchless, and debonding occurred followed by fibre breakage ratherthan by interfacial debonding usually associated with high energyabsorption [37,46]. Moreover, the increased PLA crystallinity of the

ion and type. (a) storage modulus vs. temperature and (b) tand curves.

8 B. Baghaei et al. / Composites: Part A 61 (2014) 1–12

alkali treated hemp fibre composites compared with untreatedhemp fibre composites could be another factor leading to increasedimpact strengths [47,48]. The impact resistance of the nonwovencomposites was lower than that of neat PLA. The effect of fibrereinforcement on the impact strength of composites is more com-plicated than bending and tensile strength. The impact strength isattributed to the energy consumption during failure. The additionof fibres probably creates regions of stress concentration, requiringless energy to initiate a crack. In the case of yarn composites, be-cause the reinforcing fibres are in the form of yarn and can act aslong fibres, more energy consumption occurred during fibre pull-out, as long as no breakage of fibre occurred. When longer fibresare introduced into the PLA matrix, the impact strength greatly in-creases due to the frequent occurrence of long fibre pull-outs [40].

3.7. Dynamic mechanical thermal testing

Fig. 8a and b shows how fibre treatment and orientation influ-enced the storage modulus and damping factor (tand) of PLA andits respective composites. The storage modulus (at 30 �C) increasedfrom 2.5 GPa for neat PLA up to 6.5 GPa for treated hemp/PLA, fol-lowed by 4.9 GPa for untreated hemp/PLA and 3.9 GPa PLA/hempnonwoven composite (Fig. 8a). As expected the storage modulusof the composites showed remarkable dependence on fibre orien-tation. The storage modulus was highest for the hemp/PLA (0) yarncomposite, followed by the hemp/PLA (45) yarn composite and thehemp/PLA (90) yarn composite. The increase in the stiffness of fi-bre-containing samples, revealing effective stress transfer fromthe fibre to the matrix at the interface, can be interpreted as goodadhesion between the fibres and the matrices [7,30,49,50]. The in-crease in storage modulus is more prominent in the glassy region(below Tg). The sharp decrease in the storage modulus around

Table 4Thermo-mechanical and crystallization parameters used for the estimation of the constra

Sample XDSC % based on DSC Tand at Tg based

PLA 7.50 1.69Hemp/PLA nonwoven 14.70 0.49Hemp/PLA (90) yarn 14.70 0.46Hemp/PLA (45) yarn 14.70 0.18Hemp/PLA (0) yarn 14.70 0.13Treated hemp/PLA yarn 16.90 0.10

Fig. 9. DSC curves for neat PLA

57–61 �C corresponds to the a-relaxation of the amorphous re-gions in PLA [24]. This fall in modulus at higher temperature espe-cially for the composites can be associated with the easyinterlaminar failure of composites at higher temperature, the chainmobility of matrix and thermal expansion taking place in the ma-trix resulting in reduced intermolecular forces [51,52]. The storagemodulus started to increase again around 100 �C, which is the re-sult of the cold crystallization typical of PLA [53], and which wasalso observed in the DSC results. The decrease in modulus around140 �C indicates the softening of the sample before the onset ofmelting. The dampening, or tand, is the ratio between the lossmodulus and the storage modulus and gives information aboutthe internal friction of the material. The peak of tand curve inthe glass transition region is the most dominant feature, corre-sponding to high damping due to initiation of motion in long seg-ments of the main polymer chain [54]. For a composite, themolecular motion in the interface will contribute to the damping.The dampening will consequently give information about theadhesion of the interface. A larger area under the a-relaxation peakin the tand curves of a polymer indicates that the molecular chainsexhibit a higher degree of mobility i.e. better damping properties[55]. The area under this peak for PLA composites (Fig. 8b) seemsto be smaller especially for alkali treated hemp fibre composite.A possible explanation is that there is a strong interaction betweenthe fibre and the matrix. The fibre/matrix interfacial adhesion canbe indirectly quantified by estimating the damping term as it is atrue indicator of the molecular motions in a material. When a com-posite material is subjected to deformation, the deformation en-ergy is dissipated mainly in the matrix and at the interface. Ifmatrix, fibre volume fraction and fibre orientation are identical,then the damping term can be used to assess the interfacial prop-

ined region.

on DMA Wc according to Eq. (4) C % according to Eq. (3)

0.84 7.500.60 33.440.59 34.980.36 60.500.30 67.370.24 73.39

and PLA/hemp composites.

B. Baghaei et al. / Composites: Part A 61 (2014) 1–12 9

erties between fibre and matrix. For a weak interface, more energyis dissipated during testing [56,57].

The peak point values of tand curves of different samples werecompared and ordered as neat PLA, composites made by hemp/PLAnonwoven, hemp/PLA (90) yarn, hemp/PLA (45) yarn, hemp/PLA(0) yarn and alkali treated hemp/PLA yarn. The height reduction

Table 5Calorimetric data for PLA/hemp composites for the first heating run (10 �C/min).

Sample DHcc (J/g) DHm (J/g)

PLA 28.9 35.9Untreated hemp/PLA 13.17 22.73Treated hemp/PLA 19.3 30.3

IIImmpprriinnttss ooffffiibbrreess

PPuulllleedd--oouuttffiibbrreess

(a) (

(c) (

(e)

Fig. 10. SEM images (at different magnifications) of tensile fracture surfaces of PLA/hemPLA (90), (d) nonwoven hemp/PLA and (e) alkali hemp/PLA.

for the tand peak for the composites shows that there is a reduc-tion in the amount of the mobile polymer chains during the glasstransition. The magnitude of tand values is also seen to increasein the case of composite from nonwoven compared to the compos-ites from yarn. Actually, the storage modulus favours long fibres(large l/d), and damping favours short fibres (small l/d). It is also

XDSC (%) Tg (�C) TC (�C) Tm (�C)

7.5 59.9 109.9 167.414.7 61.3 110.5 167.516.9 61.8 111.8 167.9

IImmpprriinnttss ooffffiibbrreess

VVooiiddss

b)

d)

p composite made from: (a) hemp/PLA (0) yarn, (b) hemp/PLA (45) yarn, (c) hemp/

10 B. Baghaei et al. / Composites: Part A 61 (2014) 1–12

interesting to note that, for high aspect ratios l/d, damping for ran-domly oriented short fibre composites is higher than that of eitheraligned short fibre or continuous fibre composites [58–60]. In addi-tion, material and structural integrity determined by defects, suchas microstructure (voids, impurities and imperfections in resin/fi-bre bonding or debonding) are the points of stress concentrationwhich increase the damping [61]. Higher damping of compositefrom nonwoven compared to composite from yarn could be be-cause of its higher porosity. Damping properties of compositesfrom yarn were improved by increasing fibre off-axis angle whichis consistent with Gibson et al. study [58]. They found that the the-oretical optimum damping property of discontinuous aligned shortfibre composites when the axial load is parallel to the fibre direc-tion. Our experimental investigations indicated that compositewith fibre off-axis 90� showed better damping than compositewith 45� and 0. The glass transition temperature is often recordedat the maximum of the tand. Using this method of determining theglass transition temperature, the Tg seems to increase (by a few de-grees) as hemp fibres are added to PLA. This shows that the poly-mer relaxation is delayed in the presence of fibres which restrictthe segmental motion of the PLA polymer chains due to increasedcrystallinity [30].

DMA measurements can be used to detect changes in themolecular mobility of the polymer segments in the vicinity ofhemp fibres. The volume of the constrained polymer regions inthe PLA-based composites has been estimated from the dynamicmechanical measurements. The volume fraction of the constrainedregion, C is given by Eqs. (3) and (4) [62,63],

C ¼ 1�Wc

W0ð1� C0Þ ð3Þ

W ¼ p tan dp tan dþ 1

ð4Þ

where Wc and W0 denote energy loss ratio of the composite and thepure polymer at Tg, and C0 the degree of crystallinity for the purePLA (Eq. (2)). The results are presented in Table 4. The addition ofhemp fibres brings changes in the mobility of polymer matrix andit has been observed that the maximum restriction in segmentalmobility is obtained with treated hemp fibres. This may be due tothe strong intercalation of the polymer matrix into the treated fi-bres [64].

3.8. Differential scanning calorimetry

The DSC thermograms of neat PLA, untreated and alkali treatedhemp fibre reinforced composites are given in Fig. 9 and the corre-sponding transition temperatures are tabulated in Table 5. The dataindicate that the crystallinity and Tc of PLA was found to increase as aresult of the presence of hemp, since the fillers acted as nucleatingagents for the crystallisation of the polymers [24,65]. Also, it isapparent that the crystallinity of PLA in the treated hemp/PLA com-posites increased compared with that of the untreated hemp/PLAcomposites. This could be due to the fact that after alkali treatment,the impurities including wax and pectin were removed from the fi-bres, which in turn increased the number of nucleating sites (i.e.crystalline portion of cellulose) of the fibres [10,65]. The crystalliza-tion peak of composite is found to become large with addition oftreated fibre which that peak was small in the case of neat PLA. Itis shown that the surface treatment increases the degree of regular-ity of the molecular chain [38]. The composites exhibit an increase inTg compared with that of neat PLA. The results obtained are in agree-ment with those of DMA. The melting behaviour of PLA is also influ-enced by the presence of untreated and treated hemp fibres. Theneat PLA has a single melting point around 167 �C divided into twosmall peaks (around 161 and 167 �C) in the case of untreated and

treated hemp, indicating the formation of two different crystal types[66].

3.9. Fractography analysis of damaged specimens

It was seen on the tensile specimens that the off-axis samplestend to break in the inter-fibre cleavage region and extend towardthe fibre direction. Thus, most of fibre breakages happened alongthe fibre orientation direction.

The morphologies of the fractured surfaces of tensile specimenswere investigated using SEM (Fig. 10). Debonding between fibreand matrix as well as fibre imprinting was found for compositesamples with 45� and 90� orientation arrangements (Fig. 10b andc). Furthermore, it can be seen from Fig. 10d that the compositesfabricated from nonwoven have higher porosity content insidethe composites because of less close and compact packing of hempfibres in the PLA matrix compared with the composite made fromPLA/hemp. From the surface images, there is a difference in theadhesion between PLA matrix and hemp fibre. Fig. 10a and e showsthat untreated hemp reinforced composites had more fibres pulledout with relatively clean fibre surface which is an indication of lowadhesion between them; however, the treated hemp fibres werecoated by layers of the PLA matrix that significantly reduce thegaps between them.

4. Conclusions

Unidirectional hemp reinforced PLA composites were producedfrom PLA/hemp yarn using compression moulding. We investigatedthe influences of different orientation especially off-axial directionof hemp fibre as well as alkali treatment; focusing on the determin-ing void%, water absorption, mechanical and thermo-mechanicalproperties. The properties of PLA composites were significantly im-proved compared to the neat PLA matrix. The alkali treated hemp/PLA yarn gave maximum improvement in mechanical propertiescompared to untreated hemp/PLA yarn. The best overall propertieswere achieved with aligned alkali treated hemp/PLA yarn leadingto a tensile strength of 77.1 MPa, Young’s modulus of 10.3 GPa, flex-ural strength of 100.9 MPa, flexural modulus of 7.1 GPa, and impactstrength of 18.8 kJ/m2. The results showed that the mechanicalproperties of the composites were highly affected by the fibre direc-tion. Tensile, flexural, and impact values of the composites showedthe decreasing trend for off-axial composites compared to 0� axial-oriented composite. Furthermore, the nonwoven PLA/hemp com-posites had an evident improvement in mechanical properties overthe composites produced from yarn with off-axis angle of 45�and90�. The thermo-mechanical tests showed that composites contain-ing alkali treated hemp fibres had improved storage modulus due toenhanced interfacial bonding. The damping properties were highlyaffected by the testing direction; it was increased at an off-axis angleof 45�and 90�. Composites made from PLA/hemp nonwoven showedan evident improvement in viscoelactic properties over the compos-ites produced from yarn with off-axis angle of 45�and 90�. The PLA inthe composites had high orientation degree and crystallinity whichwas attributed to effective heterogeneous nucleation induced byhemp fibres, however, the degree of crystallinity of alkali treatedhemp/PLA composite was higher. From water absorption test re-sults, it was found that higher temperature generally increased theWG% of the neat PLA and all of the composites, as well as shorteningthe saturation time. At room temperature, neat PLA had the lowestWG% value followed by PLA/hemp nonwoven, untreated hemp/PLAyarn and alkali treated hemp/PLA yarn composites. Future work willconcentrate on efforts to evaluate the biodegradability of thesedeveloping and promising composites.

B. Baghaei et al. / Composites: Part A 61 (2014) 1–12 11

Acknowledgment

The authors thank Tommy Martinsson, Swedish School of Tex-tiles, University of Borås (Sweden) for his assistance in preparinghybrid yarns.

References

[1] Karus M, Kaup M, Ortmann S. Use of natural fibres in composites in the germanand austrian automotive industry—market survey 2002: status, analysis andtrends. J Indus Hemp 2003;8(2):73–8.

[2] Robinson R. The great book of Hemp. Main, USA: Inner Traditions Bear andCompany; 1996.

[3] Clemons CM. Woodfiber-plastic composites in the United States – history andcurrent and future markets. In: 3rd International wood and natural fibrecomposites symposium. Kassel, Germany 2000. p. 1–7.

[4] Pickering KL, Sawpan MA, Jayaraman J, Fernyhough A. Influence of loading rate,alkali fibre treatment and crystallinity on fracture toughness of random shorthemp fibre reinforced polylactide bio-composites. Compos A2011;42(9):1148–56.

[5] Tserki V, Matzinos P, Panayiotou C. Novel biodegradable composites based ontreated lignocellulosic waste flour as filler. Part II. Development ofbiodegradable composites using treated and compatibilized waste flour.Compos A 2006;37(9):1231–8.

[6] Parikh DV, Calamari TA, Sawhney APS, Blanchard EJ, Screen FJ, Myatt JC, et al.Thermoformable automotive composites containing kenaf and other cellulosicfibers. Text Res J 2002;72(8):668–72.

[7] Islam MS, Pickering KL, Foreman NJ. Influence of hygrothermal ageing on thephysico-mechanical properties of alkali treated industrial hemp fibrereinforced polylactic acid composites. J Polym Environ 2010;18(4):696–704.

[8] Oksman K. Mechanical properties of natural fibre mat reinforcedthermoplastic. Appl Compos Mater 2000;7(5–6):403–14.

[9] Gamstedt EK, Berglund LA, Peijs T, Department of Engineering S, Mathematics,Polymeric Composite M, et al. Fatigue mechanisms in unidirectional glass–fibre-reinforced polypropylene. Compos Sci Technol 1999;59(5):759–68.

[10] Sawpan MA, Pickering KL, Fernyhough A. Improvement of mechanicalperformance of industrial hemp fibre reinforced polylactide biocomposites.Compos A 2011;42(3):310–9.

[11] Hu R, Lim J-K. Fabrication and mechanical properties of completelybiodegradable hemp fiber reinforced polylactic acid composites. J ComposMater 2007;41(13):1655–69.

[12] Behnaz Baghaeia MS, Lena Berglin. Manufacture and characterisation ofthermoplastic composites made from PLA/hemp co-wrapped hybrid yarnprepregs. Compos A: Appl Sci Manuf 2013.

[13] Zhang L, Miao M. Commingled natural fibre/polypropylene wrap spun yarnsfor structured thermoplastic composites. Compos Sci Technol2010;70(1):130–5.

[14] Goutianos S, Peijs T. The optimisation of flax fibre yarns for the development ofhigh-performance natural fibre composites. Adv Compos Lett2003;12(6):237–41.

[15] Shah DU Schubel PJ, Clifford MJ. Mechanical characterization of vacuuminfused thermoset matrix composites reinforced with alignedhydroxyethylcellulose sized plant bast fibre yarns. In: 4th Internationalconference on sustainable materials, polymers and composites. Birmingham,UK; 2011.

[16] Alagirusamy R, Fangueiro R, Ogale V, Padaki N. Hybrid yarns and textilepreforming for thermoplastic composites. Text Progress 2006;38(4):1–71.

[17] Madsen B, Thygesen A, Lilholt H. Plant fibre composites – porosity andvolumetric interaction. Compos Sci Technol 2007;67(7):1584–600.

[18] Pilla S, Gong S, O’Neill E, Rowell RM, Krzysik AM. Polylactide-pine wood flourcomposites. Polym Eng Sci 2008;48(3):578–87.

[19] Joseph K, Thomas S, Pavithran C. Effect of chemical treatment on the tensileproperties of short sisal fibre-reinforced polyethylene composites. Polymer1996;37(23):5139–49.

[20] Arbelaiz A, Fernández B, Ramos JA, Mondragon I. Thermal and crystallizationstudies of short flax fibre reinforced polypropylene matrix composites: effectof treatments. Thermochim Acta 2006;440(2):111–21.

[21] Bledzki AK, Reihmane S, Gassan J. Properties and modification methods forvegetable fibers for natural fiber composites. J Appl Polym Sci1996;59(8):1329–36.

[22] Prasad BM, Sain MM, Roy DN. Properties of ball milled thermally treated hempfibers in an inert atmosphere for potential composite reinforcement. J MaterSci 2005;40(16):4271–8.

[23] Saunders RA, Lekakou C, Bader MG. Compression in the processing of polymercomposites 1. A mechanical and microstructural study for different glassfabrics and resins. Compos Sci Technol 1999;59(7):983–93.

[24] Islam MS, Pickering KL, Foreman NJ. Curing kinetics and effects of fibre surfacetreatment and curing parameters on the interfacial and tensile properties ofhemp/epoxy composites. J Adhes Sci Technol 2009;23(16):2085–107.

[25] Felix JM, Gatenholm P. The nature of adhesion in composites of modifiedcellulose fiberss and polypopylene. J Appl Polym Sci 1991;42(3):609–20.

[26] Lu X, Zhang MQ, Rong MZ, Yue DL, Yang GC. Environmental degradability ofself-reinforced composites made from sisal. Compos Sci Technol2004;64(9):1301–10.

[27] George G, Joseph K, Nagarajan ER, Tomlal Jose E, George KC. Dielectricbehaviour of PP/jute yarn commingled composites: effect of fibrecontent, chemical treatments, temperature and moisture. Compos AAppl Sci Manuf 2013;47:12–21.

[28] Klein W. The technology of short-staple spinning. Manchester: TextileInstitute; 1998.

[29] Kumar R, Yakubu MK, Anandjiwala RD. Biodegradation of flax fiber reinforcedpoly lactic acid. Exp Polym Lett 2010;4(7):423–30.

[30] Mofokeng JP, Luyt AS, Tábi T, Kovács J. Comparison of injection moulded,natural fibre-reinforced composites with PP and PLA as matrices. JThermoplast Compos Mater 2012;25(8):927–48.

[31] Cady C, Heredia FE, Evans AG. Inplane mechanical properties of severalceramic matrx composites. J Am Ceram Soc 1995;78(8):2065–78.

[32] Miao M, Shan M. Highly aligned flax/polypropylene nonwoven preforms forthermoplastic composites. Compos Sci Technol 2011;71(15):1713–8.

[33] Van de Weyenberg I, Ivens J, De Coster A, Kino B, Baetens E, Verpoest I.Influence of processing and chemical treatment of flax fibres on theircomposites. Compos Sci Technol 2003;63(9):1241–6.

[34] Carelli EAV, Fujita H, Yang JY, Zok FW. Effects of thermal aging on themechanical properties of a porous–matrix ceramic composite. J Am Ceram Soc2002;85(3):595–602.

[35] Nam TH, Ogihara S, Tung NH, Kobayashi S. Effect of alkali treatment oninterfacial and mechanical properties of coir fiber reinforced poly(butylenesuccinate) biodegradable composites. Compos B 2011;42(6):1648–56.

[36] Mallick PK. Performance fiber-reinforced composites: materials,manufacturing and design. New York: CRC Press; 2008.

[37] Kannan TG, Wu CM, Cheng KB, Wang CY. Effect of reinforcement on themechanical and thermal properties of flax/polypropylene interwoven fabriccomposites. J Ind Text 2013;42(4):417–33.

[38] Huda MS, Drzal LT, Mohanty AK, Misra M. Effect of fiber surface-treatments onthe properties of laminated biocomposites from poly(lactic acid) (PLA) andkenaf fibers. Compos Sci Technol 2008;68(2):424–32.

[39] Huda MS, Drzal LT, Mohanty AK, Misra M. Effect of chemical modifications ofthe pineapple leaf fiber surfaces on the interfacial and mechanical propertiesof laminated biocomposites. Compos Interfaces 2008;15(2–3):169–91.

[40] Tokoro R, Vu DM, Okubo K, Tanaka T, Fujii T, Fujiura T. How to improvemechanical properties of polylactic acid with bamboo fibers. J Mater Sci2008;43(2):775–87.

[41] Sawpan MA, Pickering KL, Fernyhough A. Effect of fibre treatments oninterfacial shear strength of hemp fibre reinforced polylactide andunsaturated polyester composites. Compos A 2011;42(9):1189–96.

[42] Sawpan MA, Pickering KL, Fernyhough A. Flexural properties of hemp fibrereinforced polylactide and unsaturated polyester composites. Compos A2012;43(3):519–26.

[43] Wisnom MR. The relationship between tensile and flexural strength ofunidirectional composites. J Compos Mater 1992;26(8):1173–80.

[44] Whitney JM, Knight M. The relationship between tensile strength and flexurestrength in fiber-reinforced composites: flexure-and tensile-coupon data onunidirectional graphite-epoxy composites are compared to a weibull two-parameter statistical-strength model. Exp Mech 1980;20(6):211–6.

[45] Karthikeyan A, Balamurugan K. Effect of alkali treatment and fiber length onimpact behavior of coir fiber reinforced epoxy composites. J Sci Ind Res2012;71(9):627–31.

[46] Hill CAS, Abdul Khalil HPS. Effect of fiber treatments on mechanical propertiesof coir or oil palm fiber reinforced polyester composites. J Appl Polym Sci2000;78(9):1685–97.

[47] Park SD, Todo M, Arakawa K, Koganemaru M. Effect of crystallinity andloading-rate on mode I fracture behavior of poly(lactic acid). Polymer2006;47(4):1357–63.

[48] Perego G, Cella GD, Bastioli C. Effect of molecular weight and crystallinity onpoly(lactic acid) mechanical properties. J Appl Polym Sci 1996;59(1):37–43.

[49] Zhang Q, Shi L, Nie J, Wang H, Yang D. Study on poly(lactic acid)/natural fiberscomposites. J Appl Polym Sci 2012;125(S2):E526–33.

[50] Majhi SK, Nayak SK, Mohanty S, Unnikrishnan L. Mechanical and fracturebehavior of banana fiber reinforced polylactic acid biocomposites. Int J PlastTechnol 2010;14(1):57–75.

[51] George J, Thomas S, Bhagawan SS. Effect of strain rate and temperature on thetensile failure of pineapple fiber reinforced polyethylene composites. JThermoplast Compos Mater 1999;12(6):443–64.

[52] Joseph K, Joseph PV, Thomas S, Pillai CKS, Prasad VS, Groeninckx G, et al. Thethermal and crystallisation studies of short sisal fibre reinforcedpolypropylene composites. Compos A 2003;34(3):253–66.

[53] Tabi T, Sajo IE, Szabo F, Luyt AS, Kovacs JG. Crystalline structure of annealedpolylactic acid and its relation to processing. Exp Polym Lett2010;4(10):659–68.

[54] Earl JS, Shenoi RA. Hygrothermal ageing effects on FRP laminate and structuralfoam materials. Compos A 2004;35(11):1237–47.

[55] Pilla S, Kramschuster A, Yang LQ, Lee J, Gong SQ, Turng LS. Microcellularinjection-molding of polylactide with chain-extender. Mater Sci Eng C-BioSupramol Syst 2009;29(4):1258–65.

[56] Afaghi-Khatibi A, Mai Y-W. Characterisation of fibre/matrix interfacialdegradation under cyclic fatigue loading using dynamic mechanical analysis.Compos A 2002;33(11):1585–92.

[57] John MJ, Anandjiwala RD. Chemical modification of flax reinforcedpolypropylene composites. Compos A 2009;40(4):442–8.

12 B. Baghaei et al. / Composites: Part A 61 (2014) 1–12

[58] Gibson RF, Chaturvedi SK, Sun CT. Complex moduli of aligned discontinuousfibre-reinforced polymer composites. J Mater Sci 1982;17(12):3499–509.

[59] Sun CT, Gibson RF, Chaturvedi SK. Internal material damping of polymermatrix composites under off-axis loading. J Mater Sci 1985;20(7):2575–85.

[60] Sun CT, Wu JK, Gibson RF. Prediction of material damping in randomlyoriented short fiber polymer matrix composites. J Reinf Plast Compos1985;4(3):262–72.

[61] Chandra R, Singh SP, Gupta K. Damping studies in fiber-reinforced composites– a review. Compos Struct 1999;46(1):41–51.

[62] Karevan M, Kalaitzidou K. Formation of a complex constrained region at thegraphite nanoplatelets-polyamide 12 interface. Polymer (United Kingdom).2013;54(14):3691–8.

[63] Kojima Y, Usuki A, Kawasumi M, Okada A, Fukushima Y, Kurauchi T, et al.Mechanical properties of nylon 6-clay hybrid. J Mater Res 1993;8(05):1185–9.

[64] Thomas PC, Jose ET, Gejo George, Sabu Thomas, Kuruvilla Joseph. Dynamicmechanical and rheological properties of nitrile rubber nanocomposites basedon TiO2, Ca3(PO4)2 and layered silicate. J Compos Mater 2013;0(0):1–15.

[65] Eichhorn SJ, Groom L, Hughes M, Hill C, Rials TG, Wild PM, et al. Review:current international research into cellulosic fibres and composites. J Mater Sci2001;36(9):2107–31.

[66] Aydın M, Tozlu H, Kemaloglu S, Aytac A, Ozkoc G. Effects of alkali treatment onthe properties of short flax fiber–poly(lactic acid) eco-composites. J PolymEnviron 2011;19(1):11–7.

Paper III

Composites: Part A 76 (2015) 154–161

Contents lists available at ScienceDirect

Composites: Part A

journal homepage: www.elsevier .com/locate /composi tesa

Characterization of thermoplastic natural fibre composites made fromwoven hybrid yarn prepregs with different weave pattern

http://dx.doi.org/10.1016/j.compositesa.2015.05.0291359-835X/� 2015 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Tel.: +46 33 435 4497; fax: +46 33 435 4008.E-mail address: mikael.skrifvars@hb.se (M. Skrifvars).

Behnaz Baghaei, Mikael Skrifvars ⇑, Lena BerglinSwedish Centre for Resource Recovery, Academy for Textile, Engineering and Business, University of Borås, SE-501 90 Borås, Sweden

a r t i c l e i n f o a b s t r a c t

Article history:Received 25 February 2015Received in revised form 13 May 2015Accepted 29 May 2015Available online 6 June 2015

Keywords:A. Fabrics/textilesB. Mechanical propertiesE. WeavingE. Compression moulding

This paper focuses on the effect of weave structure on mechanical behaviour and moisture absorption ofthe PLA/hemp woven fabric composites made by compression moulding. The unidirectional woven fabricprepregs were made from PLA (warp) and PLA/hemp wrapped-spun hybrid yarn (weft) with two differentweave patterns; 8-harness satin and basket. Unidirectional composites with 30 mass% hemp contentwere fabricated from these prepregs, and compared to winded PLA/hemp hybrid yarn laminates withsame composition. The composite from the satin fabric had significantly lowest porosities and bestmechanical properties compared to the composite made from the winded hybrid yarn and basket fabric.The tensile, flexural, and impact strength were 88 MPa, 113.64 MPa, and 24.24 kJ/m2, respectively. Theeffect of weave pattern on water absorption is significant. Although the composite from hybrid yarn lam-inate has larger water absorption than that of the pure PLA, it exhibits lower moisture absorption thanboth weaves.

� 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Currently, the increasing global environmental awareness hascaused a paradigm shift toward producing environmental friendlymaterials. Therefore, in recent years, annually renewable naturalfibres such as hemp, flax, jute, kenaf and sisal have attracted moreinterest as reinforcements for both thermosetting and thermoplas-tic polymer composites [1,2].

One of the advantages of using thermoplastic composite is itsshort cycle production process which makes it possible for exten-sive application in several industries including automobile manu-facture, construction of aircraft and ship components and otherhigh volume products. In contrast to thermosetting composites,thermoplastic composites have better impact strength, easier recy-clability and in the case of PLA, even biodegradability. On the otherhand, due to the high melt viscosity of the thermoplastic matrix,the fibre impregnation is more complex than for thermosets.Different impregnation methods have been proposed and appliedin order to solve this problem. One promising technique is the fab-rication of hybrid yarns being based on the principle of mixingreinforcing fibres and thermoplastic matrix fibres homogeneouslyduring yarn spinning. Hybrid yarns are non-uniform yarns, whichcontain components with different properties: a reinforcing

component and a thermoplastic component [3–6]. Recently, hybridyarn manufacturing has been developed for rapid and cost effectivemethod of continuous fibre reinforced thermoplastic composites.Hybrid yarns offer an ideal opportunity to achieve short cycletimes due to the very short flow paths of the viscous thermoplasticmelt [7]. To achieve high mechanical performance of thermoplasticcomposites the homogeneous fibre/matrix distribution is neces-sary, which can be achieved with hybrid yarns.

Handling of hybrid yarn for making unidirectional composite isproblematic. Hybrid yarns can be woven into a wide variety of con-formable and drapable fabrics. Textile production technologymakes it possible to use hybrid yarns as the basis for obtainingan intermediate prepreg. With subsequent processing under pres-sure and at elevated temperature, the melting thermoplastic fibrecomponent is converted to a polymer matrix filling the mouldand impregnating the reinforcing fibres. This technology offersnew possibilities for a shortened production process for thermo-plastic composites and articles made from them [8–10].

Fabric weaves are using in several applications in fibre rein-forced polymers due to their good mechanical properties, such asstiffness, strength and dimensional stability. With the weavingtechnology, it is possible to fabricate high density woven struc-tures with load-oriented fibre positioning [11–13].

The strength and stiffness of fabric reinforced composites areinfluenced by not only the matrix and yarns properties, but alsostructural parameters of materials such as fabric count and weave

B. Baghaei et al. / Composites: Part A 76 (2015) 154–161 155

pattern. The fabric count determines the number of weft and warpyarns per cm, whereas the weave pattern specifies how the warpand the weft yarns are interlaced. Typical weave patterns are plain,twill, basket and satin [14].

Fabrics which are produced by weaving a large number of thickyarns as warp and a few numbers of thin yarns as weft are calledunidirectional and are used in unidirectional composites. Thisweave type provides materials with good processability, high stiff-ness and strength in the warp direction, which is specific for unidi-rectional composites [15].

Recently different studies have reported limited improvementin the mechanical properties of the composites mostly madedirectly from hybrid yarns. There are limited studies on the com-posites manufactured from fabric using hybrid yarn especially bio-composite [5,7,16–18].

The main focus of this study was to develop and characterizereinforcement fabrics made of novel aligned hybrid yarn and usethis fabric as a prepreg for manufacturing composite. In this study,unidirectional hemp-fibre/PLA composites were made by compres-sion moulding of fabrics composed of hybrid yarns [19], and fur-ther subjected to different tests in order to investigate the effectsof the pattern style of the woven fabrics (satin and basket patterns)on the mechanical properties.

2. Materials and methods

2.1. Materials

The studied composites consisted of a polylactide matrix rein-forced with hemp fibres. The components were prepared at aweight ratio of 70/30 (polymer/fibre).

NatureWorks™ PLA Polymer 6202D by the company CargillDow LLC (Minnetonka, USA) was used as the matrix material.This thermoplastic has a glass transition temperature (Tg) of60–65 �C, a melting point (Tm) of 160–170 �C and a density of1.24 g/cm3. The used PLA was in filament and fibre form. A 18tex PLA multifilament yarn and Ingeo™ staple fibre were providedby Trevira GmbH (Hattersheim, Germany). The staple fibres had afineness of 1.7 dtex and a mean fibre length of 38 mm. Hemp staplefibres (genus species Cannabis sativa L.), supplied by Hempage AG(Adelsdorf, Germany), with an average diameter of 20–40 lmand a mean fibre length of 30 mm were used as reinforcement.

Fig. 1. Structure of wrap spun hybrid yarn.

2.2. Methods

2.2.1. Hybrid yarn manufactureA wrap spinning method was used to make a hybrid yarn struc-

ture with the reinforcing hemp fibres straight and parallel to theyarn axis [5]. Within the framework of this study, the hybrid yarn(nominal count of 550 tex) consists of hemp staple fibres used as areinforcement and PLA staple fibre used as the thermoplastic com-ponent. This hybrid yarn was manufactured with wrap spinningtechnology using a laboratory spinning machine from MesdanS.p.A. (Brescia – Italy) and a laboratory yarn twist machine fromDirecTwist, AGTEKS CO. Ltd. (Istanbul, Turkey). The PLA-hemp fibreweight ratio was 30:70 [19]. A very uniform and aligned rovingwas produced in a carding and roving frame. No twist was addedto the roving during the drawing, this is important, as a twist willhave a negative influence on composite strength. The obtained rov-ing had however poor integrity, it was too weak to be able to becollected alone without causing yarn breakage in the rovingmachine. Therefore, PLA filaments were used as a processing car-rier for the PLA/hemp staple fibres in the roving in the final man-ufacturing step. The roving was further stabilised by using acontinuous PLA filament thread, which was wounded around the

roving strand (wrapping intensity of 200 turns/m) in the twistingmachine. This imparted strength to the strand to prevent it fallingapart and gave a better protection for the reinforcing staple fibresduring further textile processing, such as weaving [20] or making aprepreg, The hybrid wrap yarn thus consists of two components,one twist-free PLA/hemp staple fibres component in the yarn core,and a filament wound around the core. The yarn structure is shownin Fig. 1.

2.2.2. UD woven fabrics from hemp/PLA hybrid yarnsIn order to manufacture samples of thermoplastic composites,

we made woven unidirectional (UD) fabrics from hybrid yarnsand PLA filaments on a hand loom weaving machine. The obtainedwoven fabrics had the PLA/hemp hybrid yarn in the transversedirection of manufacture of the fabric (weft direction) and weresupported by the 18 tex PLA filaments as the warp yarns in the lon-gitudinal direction (8 threads per cm). The fabrics were woven intwo different weave patterns; 8-harness satin (4 threads per cm)and basket weave (3 threads per cm) as seen in Fig. 2. The con-structional particulars of fabric used in this study are presentedin Table 1. These fabrics have similar characteristics as an unidirec-tional fabric in the weft direction, as the thin warp yarn reducesthe crimp considerably. The woven fabrics obtained can be consid-ered as prepregs for subsequent processing to make composites bycompression moulding.

2.2.3. Composite manufacturing and testingMechanical tests for tensile strength, flexural strength and

impact behaviour were conducted on UD composite laminates asa preliminary characterization of the fabrics made from hempand PLA in various architecture patterns; satin (8HS) and basket(2/2). Each laminate had 6 layers of 20 � 20 cm2 prepreg to achievea composite thickness of about 3–4 mm. The laid up prepregs werefirst put in a vacuum chamber (0.9 mbar; 70 �C) for at least 18 h, todry before the compression moulding. The UD composite lami-nates were consolidated at a temperature of 195 �C and a pressureof 1.7 MPa for 15 min in a laboratory hydraulic compressionmoulding machine from Rondol Technology Ltd. (Staffordshire,UK). These parameters result in optimal impregnation and verylow voids in the composite. Neat unreinforced PLA sheets to be

Fig. 2. Schematic diagram of (a) 8-harness satin and (b) basket weave patterns. (For interpretation of the references to colour in this figure legend, the reader is referred to theweb version of this article.)

Table 1Details of constructional particulars of fabrics.

Sample Weave style Weight (GSM) Weft yarn Warp yarn Wrap count (ends/cm) Weft count (ends/cm)

1 Satin 1⁄8 344.0 (22.9) Wrapped spun hemp/PLA hybrid yarn (550 tex) 18 tex PLA 8 42 Basket 482.0 (48.9) Wrapped spun hemp/PLA hybrid yarn (550 tex) 18 tex LA 8 3

156 B. Baghaei et al. / Composites: Part A 76 (2015) 154–161

used as reference material were made by compression mouldingunder the same processing conditions of the same PLA fibres andfilaments formed to a similar prepreg. Before performing themechanical testing, the specimens were conditioned for at least24 h at 23 �C and 50% relative humidity according to DIN EN ISO291. The tensile and flexural strength tests on the UD compositesheets were done using an universal H10KT testing machineequipped with a mechanical extensometer (model 100R long travelextensometer) supplied by Tinius Olsen Ltd. (Salford, UK) accord-ing to ISO 527 and ISO 14125. The impact test was carried out withthe help of a pendulum arm-type impact tester, which works onthe principle of the Charpy Impact Test. Standard method of testingwas applied as stated in the ISO 179. The specimens used in themechanical testing were cut from the laminates using laser cutting,and they were tested in the weft direction.

2.2.4. Composite density and porosityFibre volume fraction was measured using the density buoy-

ancy method in which the mass of sample (coated with paraffin)is recorded in air and ethanol. Therefore, having obtained compos-ite densities and knowing the densities of the constituent parts i.e.the hemp fibre and PLA resin, the percentage content of the fibresand porosity were calculated [3].

2.2.5. Water absorption testThe effect of water absorption on hemp fibre reinforced com-

posites was calculated using ASTM D570-98. The percentageweight gain of the samples (WG) was calculated by the weight dif-ference between the sample immersed in distilled water (W1) anddry sample (W0), using Eq. (1):

WG ¼ ½ðW1 �W0Þ=W0� � 100% ð1Þ

B. Baghaei et al. / Composites: Part A 76 (2015) 154–161 157

2.2.6. Determination of dynamic mechanical propertiesThe time–temperature dependency of the mechanical proper-

ties was determined by dynamic mechanical thermal analysis,with a DMA Q800 (dual cantilever) supplied by TA Instruments,Newcastle, DE, USA. The temperature range was from 30 to150 �C with a heating rate of 3 �C/min and at frequency of 1 Hzand the sample dimensions were: thickness 2–3 mm, length50 mm, and width 8 mm.

2.2.7. Scanning electron microscopy (SEM)Fractured surfaces of the composites obtained from the tensile

testing were examined using a low-vacuum scanning electronmicroscopy with a Quanta 200 ESEM FEG instrument from FEI,(Oregon, USA) with an operating voltage of 10–12.5 kV.

3. Results and discussion

3.1. Composite porosity

Porosity is one of the most critical issues during the manufac-turing process of composites. The presence of porosities can signif-icantly reduce the tensile, compressive, inter laminar shear andstructural strengths of a composite [21]. In the current study, com-posites were produced from satin PLA/hemp hybrid fabrics andbasket PLA/hemp hybrid fabrics in a nominal mass ratio of 30%hemp reinforcement in order to investigate the effect of differentweave patterns on the properties of the composites.Unidirectional composites previously made from identicalPLA/hemp hybrid yarns, produced by winding the yarn around ametal frame, were used as reference for comparison to the madewoven hybrid fabric based composites [19]. Data for compositesmade from PLA/hemp nonwoven mats from a previous study arealso used as reference in this study in order to show the effect offibre alignment on the porosity content [22]. Table 2 demonstratesthe quality of the produced composites from the current study andcompared to results from the previous studies [19,22]. Both thecomposite density and constituents of the composites were evalu-ated. From the results, it is possible to conclude that the compositemade by unidirectional satin fabric gave significantly lower porosi-ties compared to the winded hybrid yarn laminates, and the valueis in the acceptable range [19,22]. Generally, less than 2% porosityis considered as high quality composite [23].

Creation of voids during composite processing is a function ofseveral construction factors including volume fractions of fibres,fibre distribution, fibre diameter and laminate tightness. Theporosity content for the composite made from satin fabric(0.96%) was clearly lower compared to the composite made frombasket fabric (4.55%) and much lower compare to the compositemade from a nonwoven mat (16.71%) [19,22]. It can probably beattributed to the fact that satin weaves allow individual fibres tobe woven in the closest proximity and can therefore produce fab-rics with a close ‘tight’ weave compared to basket weave. Thetightly woven fabrics are typically the choice to maintaining fibreorientation during the manufacturing process and minimizingresin void size due to the shorter impregnation distance.

Table 2Composition of the fabricated composites calculated from the density measurements.

Sample Density (g/cm3) Fibre mass fraction (%) Fibre vol

PLA 1.2490 (0.0002) 0.0 0.0Satin hemp/PLA fabric 1.2978 (0.0032) 30.0 26.31 (0.Basket hemp/PLA fabric 1.2508 (0.0090) 30.0 25.35 (0.Hemp/PLA yarn laminate 1.2713 (0.0268) 30.0 25.77 (0.Hemp/PLA nonwoven 1.0914 (0.0154) 30.0 21.12 (0.

Note: Data in table are mean (SD).

Therefore, the laminar and crosswise flow of resin along andbetween the fibres, in between the yarns and fabric layers is easier[24,25]. Moreover, basket weave has higher quantity of interlacepoints per unit area compared to satin. Perhaps voids might be pre-sent between the interlace points which are higher in basket sys-tems than those of satin [26]. Therefore, the composite made ofthe satin fabric is of high quality, as shown in Fig. 3a and b, becauseof its lower porosity.

3.2. Water absorption characteristics

Fig. 4 shows apparent weight gain (WG) as a function of time forneat PLA and PLA/hemp reinforced composite samples immersedin water at room temperature (23 �C). All the composites havethe same overall fibre weight fraction (30%). As expected the neatPLA shows the lowest water absorption, which is due to the lack ofhydrophilic fibres. It is seen that the effect of weave pattern is sig-nificant for the composites. Although the composite made from thewinded hybrid yarn prepreg has larger moisture absorption thanthat of the pure PLA, it exhibits lower moisture absorption thanboth weaves (satin and basket). Actually the matrix pockets andyarn undulation within the weaves provides an easier diffusionpath for the moisture to travel within the weave, compared to acomposite composed of unidirectional yarns. If the volume of thewavy region is larger, the WG of a composite would be higher[27,28]. It is observed that basket weave has higher WG than satin.

3.3. Tensile properties

Table 3 shows the tensile properties of unidirectional (hempfibres in the longitudinal direction) hemp/PLA composites. It canbe seen that the measured values in the case of the winded yarnlaminate composite is significantly lower compared to the wovenfabric composites. Actually tightly woven fabrics are usually thechoice to maintain fibre orientation during the fabrication process,which was not possible in the winded yarn laminate. The tensilemodulus and strength of composite for basket 2/2 is lower com-pared to satin due to its fabric structure, which makes some partsof the composite weaker. The basket weave has a higher crimpcompared to the satin weave due to its more interlacement. (seeFig. 2) Less crimp will result in greater strength on the compositesas its free inter-fibres contribute toward the force. After the weav-ing process of the fabric, contraction occurs which causes floatingof adjacent yarns to join together making a jammed structurewhereas the end point of interlacement creates a gap withoutany reinforcement. This gap will give in the composite a resin richarea, which is the weakest point because of no reinforcementmaterials. The basket 2/2 weave has more number of interlacingsper surface area, which increase the number of stress concentra-tion points and resin rich areas which significantly affect the ten-sile properties. The jammed fabric structure has influence on theproperties of the composites since in this area the resin penetra-tion between the fibres is low. In the case of satin weave each adja-cent floating yarn has a different interlacing point which makes it

ume fraction (%) Matrix volume fraction (%) Porosity volume fraction (%)

100.0 –07) 72.74 (0.18) 0.96 (0.25)18) 70.10 (0.51) 4.55 (0.71)54) 71.25 (1.50) 2.97 (2.04)31) 61.16 (0.9) 16.71 (1.17)

Voids

(b)(a)

Fig. 3. SEM images of the fracture surface of the hemp/PLA specimens from: (a) satin-weave architecture fabric and (b) basket-weave architecture fabric.

0 1 2 3 4 5 6 7 8 9

10

0 1 2 3 4 5 6 7 8 9 10

App

aren

t wei

ght g

ain

(wt%

)

Day

Hemp/PLA yarn laminate

Basket hemp/PLA fabric

Satin hemp/PLA fabric

Neat PLA

Fig. 4. Apparent weight gain against time for the different PLA/hemp composites atroom temperature.

Table 3Tensile properties of PLA-based composites.

Sample Tensile strength(MPa)

Young’s modulus(GPa)

Elongation atbreak (%)

PLA 41.21 (2.25) 2.91 (0.39) 1.92 (0.26)Satin hemp/PLA

fabric88.06 (7.70) 10.23 (2.67) 1.35 (0.30)

Basket hemp/PLAfabric

81.09 (5.62) 9.20 (1.81) 1.58 (0.22)

Hemp/PLA yarnlaminate

72.75 (6.26) 8.77 (1.44) 1.24 (0.36)

Note: Data in table are mean (SD).

0 2 4 6 8 10 12

0 20 40 60 80

100 120 140

Flex

mod

ulus

in G

pa

Flex

stre

ngth

in M

Pa Flex StrengthFlex modulus

Fig. 5. Flexural properties of PLA/hemp composites.

15

20

25

30

35

tren

gth

in k

J/m

2

158 B. Baghaei et al. / Composites: Part A 76 (2015) 154–161

evenly distributed and the even yarns floating distribution withsmall gap makes it free for the resin to distribute evenly [29].

The highest strength values (88.06 ± 7.70 MPa) were registeredfor composites reinforced with hemp/PLA satin fabric, followed byhemp/PLA basket fabric and winded hemp/PLA yarn laminate. Ascan be seen from the results, the tensile strength and modulus ofthe PLA were clearly improved by adding hemp fibres, however;the elongation at break was reduced. The reduction in elongationat break in the composites is mainly due to the destruction ofthe PLA structural integrity by the loading of the natural fibrewhich leads to faster fracture of the PLA matrix [30].

0

5

10

PLA Satin hemp/PLA

fabric

Basket hemp/PLA

fabric

Hemp/PLA yarn laminate

Impa

ct s

Fig. 6. Impact strength of the composites relative to their proportion of fibre mass.

3.4. Flexural properties

Fig. 5 shows the flexural strength of hemp/PLA composites.One-Way ANOVA test showed that mean flexural strength andflexural modulus values of materials investigated were signifi-cantly different (P-value < 0.05). The flexural strength for all com-posites was significantly higher than for the unreinforced

specimens (neat PLA). Flexural modulus was lowest for the neatPLA (4.41 GPa) and highest for the satin composite sample(7.63 GPa), and had the same trend as for the flexural strength.

Different weave pattern has significant influence on the flexuralstrength of the composites. Satin fabric composite exhibited thehighest flexural strength. It is obviously due to the presence ofmore aligned fibres in the axial direction, which enabled it to holdthe flexural stress developed over the surface of the material.Moreover, the satin fabric composite had a less chance for stressconcentration due to lower number of interlacing points [29,31].

3.5. Impact resistance

Fig. 6 shows the impact energy of hemp/PLA yarn reinforcedcomposites for different laminates. The impact properties of the

B. Baghaei et al. / Composites: Part A 76 (2015) 154–161 159

composites followed the same trend as obtained in the flexural andtensile testing. The impact energy was around 24.2, 23.8 and16 kJ/m2 for satin fabric, basket and winded yarn laminatecomposite, respectively. It can be seen from the results that thevariation in impact strength of the composites was not signifi-cantly influenced by the weaving architecture, as the variation ismeagre. The higher impact energies for satin and basket laminatecompared with winded yarn laminate are mainly because of thepresence of a high number of axially oriented fibres since wovenfabric composites provide more balanced properties in the fabricplane than unidirectional composites [32].

3.6. Dynamic mechanical thermal testing

Storage modulus is related to the Young’s Modulus under ten-sile loading. The incorporation of fibres plays significant role onincreasing the stiffness of a composite. Fig. 7a shows the effect of

Fig. 7. DMTA analysis of thermoplastic PLA reinforced with of hemp

weave patterns on the storage modulus values with temperature.From the results, it is clearly seen that addition of fibres hastremendously increased the storage modulus value of the compos-ite when compared with pure PLA. The increase in storage moduluswas attributed by the incorporation of hemp fibres and it is clearlyseen for both types of weave pattern laminates and winded yarnlaminate. This increase is due to good fibre/matrix adhesion andgreat degree of stress transfer at the interface. Composite fromsatin and basket weaves had a better storage modulus of 8.5 and7 GPa compared to the winded yarn laminate composite which isin line with the results from tensile, flexural, and impact tests, thatcomposites from satin and basket weave are superior to windedyarn laminate composite.

Fig. 7b shows the tand value of the composites as the functionof temperature and the glass transition temperature correspondingto the maximum tand peak. The results indicate that the incorpo-ration of fibres in the matrix has considerably reduced the

fibre. (a) Storage modulus vs. temperature; and (b) tand curves.

Fig. 8. SEM images of the fracture surface morphology of composites made of: (a)satin-weave architecture fabric and (b) basket-weave architecture fabric.

160 B. Baghaei et al. / Composites: Part A 76 (2015) 154–161

damping property of the PLA matrix. The tand value for the satinand basket weave types of composites are found to be a little lowerand wider compared to winded yarn laminate which indicates thestress transfer from hemp fibres to PLA matrix occurs easily with-out the failure of matrix [33]. Lower damping property shows thecomposite absorb more amount of the energy when loaded.

The glass transition temperature for satin and basket type ofwoven and yarn laminate composite are found to be higher(71.5–74 �C) compared to pure PLA (68 �C). This indicates that withthe presence of hemp fibres, the polymer relaxation is delayed andsegmental motion of the PLA polymer chains is restricted due toincreased crystallinity and good interaction between the fibreand the polymer matrix [34,35]. According to the previous study[22], the PLA in the composites had high degree of orientationand crystallinity which was attributed to effective heterogeneousnucleation induced by hemp fibres.

3.7. Fractography analysis of damaged specimens

SEM analysis is carried out to investigate the fracture behaviourof the composites (Fig. 8a and b). Figures reveal that the interaction

of polymer matrix (PLA) with hemp fibres is strong since the fibrepull-out is too low. Moreover, there is uniform distribution of uni-directional hemp fibres in the PLA matrix.

4. Conclusions

We report the manufacture of hybrid textile yarns containingreinforcing (hemp) and thermoplastic (PLA) components withintheir structure, the technology for processing them into woven fab-rics (prepregs), and the manufacture of thermoplastic compositesbased on them. The effect of two types of structures for the fabrics,8-harness satin and basket, on amount of porosity, moistureabsorption, mechanical and thermo-mechanical properties of com-posites were studied. The results showed that the unidirectionalhemp/PLA composite made by satin-weave architecture fabric pos-sessed the highest tensile, flexural and impact strength comparedto the composites manufactured with basket weave fabric andwinded hybrid yarn laminate. This improvement in mechanicalstrength was correlated to the decrease in void content and fibremisalignments. This means that satin composite is the strongest,stiffest, and toughest. Mechanical properties of the compositesshowed the decreasing trend for winded hybrid yarn laminatecomposites compared to woven fabric composite. However, mois-ture absorption test did not show a similar trend. Thethermo-mechanical tests showed that composites manufacturedfrom satin and basket weave types had lower damping propertiescompared to yarn laminate which indicates the stress transfer fromthe hemp fibres to PLA matrix occurs easily without the failure ofmatrix. Composite from satin and basket weaves had a better stor-age modulus compared to yarn laminate composite which sup-ported all the results from tensile, flexural, and impact tests.

Acknowledgments

The authors would like to acknowledge Jan Johansson, SwereaIVF, Mölndal, Sweden, for his assistance with the impact testing.

References

[1] Khondker OA, Ishiaku US, Nakai A, Hamada H. A novel processing technique forthermoplastic manufacturing of unidirectional composites reinforced withjute yarns. Composites Part A. 2006;37(12):2274–84.

[2] Abdelmouleh M, Boufi S, Belgacem MN, Dufresne A. Short natural-fibrereinforced polyethylene and natural rubber composites: Effect of silanecoupling agents and fibres loading. Composites Science and Technology.2007;67(7):1627–39.

[3] Madsen B, Thygesen A, Lilholt H. Plant fibre composites – porosity andvolumetric interaction. Composites Science and Technology. 2007;67(7):1584–600.

[4] Padaki N, Fangueiro R, Ogale V, Alagirusamy R. Hybrid Yarns and TextilePreforming for Thermoplastic Composites. Textile Progress. 2006;38(4):1–71.

[5] Zhang L, Miao M. Commingled natural fibre/polypropylene wrap spun yarnsfor structured thermoplastic composites. Composites Science and Technology.2010;70(1):130–5.

[6] Choi B-D, Diestel O, Offermann P. Commingled CF/PEEK hybrid yarns for use intextile reinforced high performance rotors. In: ICCM 12, France; 1999. p. 528.

[7] Mäder E, Rausch J, Schmidt N. Commingled yarns – processing aspects andtailored surfaces of polypropylene/glass composites. Composites Part A2008;39(4):612–23.

[8] Wakeman MD, Zingraff L, Bourban PE, Månson JAE, Blanchard P. Stampforming of carbon fibre/PA12 composites – a comparison of a reactiveimpregnation process and a commingled yarn system. Compos Sci Technol2006;66(1):19–35.

[9] Thanomsilp C, Hogg PJ. Interlaminar fracture toughness of hybrid compositesbased on commingled yarn fabrics. Compos Sci Technol 2005;65(10):1547–63.

[10] Stolyarov ON, Stolyarov IN, Kryachkova TA, Kravaev PG. Hybrid textile yarnsand thermoplastic composites based on them. Fibre Chem 2013;45(4):217–20.

[11] Abounaim M, Hoffmann G, Diestel O, Cherif C. Thermoplastic composite frominnovative flat knitted 3D multi-layer spacer fabric using hybrid yarn and thestudy of 2D mechanical properties. Compos Sci Technol 2010;70(2):363–70.

[12] Großmann K, Mühl A, Löser M, Cherif C, Hoffmann G, Torun AR. New solutionsfor the manufacturing of spacer preforms for thermoplastic textile-reinforcedlightweight structures. Prod Eng 2010;4(6):589–97.

B. Baghaei et al. / Composites: Part A 76 (2015) 154–161 161

[13] Torun AR, Hoffmann G, Mountasir A, Cherif C. Effect of twisting on mechanicalproperties of GF/PP commingled hybrid yarns and UD-composites. J ApplPolym Sci 2012;123(1):246–56.

[14] Vasiliev V, Morozov E. Mechanics and analysis of composite materials. Oxford(UK): Elsevier Science; 2001.

[15] Morozov E. Mechanics and analysis of fabric composites and structures.AUTEX Res J 2004;4(2):60–71.

[16] Schafer J, Stolyarov O, Ali R, Greb C, Seide G, Gries T. Process-structurerelationship of carbon/polyphenylene sulfide commingled hybrid yarns usedfor thermoplastic composites. J Ind Text 2015.

[17] Kravaev P, Stolyarov O, Seide G, Gries T. Influence of process parameters onfilament distribution and blending quality in commingled yarns used forthermoplastic composites. J Thermoplast Compos Mater 2014;27(3):350–63.

[18] Prabhakaran RTD, Toftegaard H. Environmental effect on the mechanicalproperties of commingled-yarn-based carbon fibre/polyamide 6 composites. JCompos Mater 2014;48(21):2551–65.

[19] Baghaei B, Skrifvars M, Salehi M, Bashir T, Rissanen M, Nousiainen P. Novelaligned hemp fibre reinforcement for structural biocomposites: porosity,water absorption, mechanical performances and viscoelastic behaviour.Compos Part A: Appl Sci Manuf 2014;61:1–12.

[20] Alagirusamy R, Fangueiro R, Ogale V, Padaki N. Hybrid yarns and textilepreforming for thermoplastic composites. Text Prog 2006;38(4):1–71.

[21] Mallick PK. Fiber reinforced composites materials, manufacturing and design.3rd ed. CRC Press; 2007.

[22] Baghaei B, Skrifvars M, Rissanen M, Ramamoorthy SK. Mechanical and thermalcharacterization of compression moulded polylactic acid natural fibercomposites reinforced with hemp and lyocell fibers. J Appl Polym Sci2014;131(15). n/a–n/a.

[23] Prabhakaran RT, Andersen T, Markussen CM, Madsen B, Lilholt H. Tensile andcompression properties of hybrid composites – a comparative study. In: The19th international conference on composite materials; 2013. p. 1029–35.

[24] Indu Shekar R, Kotresh TM, Krishna Prasad AS, Damodhara Rao PM,Ananthakrishnan T, Satheesh Kumar MN, et al. Hybrid fabrics for structuralcomposites. J Ind Text 2011;41(1):70–103.

[25] Sakaguchi M, Nakai A, Hamada H, Takeda N. The mechanical properties ofunidirectional thermoplastic composites manufactured by a micro-braidingtechnique. Compos Sci Technol 2000;60(5):717–22.

[26] Ismail H, Nasir M, Mariatti M. Influence of different woven geometry and plyeffect in woven thermoplastic composite behaviour-Part 2. Int J Polym Mater2000;47(2):499–512.

[27] Choi HS, Ahn KJ, Nam JD, Chun HJ. Hygroscopic aspects of epoxy/carbon fibercomposite laminates in aircraft environments. Composites Part A 2001;32(5):709–20.

[28] Tang X, Whitcomb JD, Li Y, Sue H-J. Micromechanics modeling of moisturediffusion in woven composites. Compos Sci Technol 2005;65(6):817–26.

[29] Saiman MP, Wahab MS, Wahit MU. The effect of fabric weave on the tensilestrength of woven Kenaf reinforced unsaturated polyester composite. In:Ahmad MR, Yahya MF, editors. Proceedings of the International Colloquium inTextile Engineering, Fashion, Apparel and Design 2014 (ICTEFAD2014). Singapore: Springer; 2014. p. 25–9.

[30] Nam TH, Ogihara S, Tung NH, Kobayashi S. Effect of alkali treatment oninterfacial and mechanical properties of coir fiber reinforced poly(butylenesuccinate) biodegradable composites. Composites Part B 2011;42(6):1648–56.

[31] Kannan TG, Wu CM, Cheng KB. Open hole flexural and izod impact strength ofunidirectional flax yarn reinforced polypropylene composites as a function oflaminate lay-up. Polym Compos 2013;34(11):1912–20.

[32] Malhotra SK, Joseph K, Thomas S. Polymer composites, volume 1: macro- andmicrocomposites. John Wiley & Sons; 2012.

[33] Kretsis G. A review of the tensile, compressive, flexural and shear properties ofhybrid fibre-reinforced plastics. Composites 1987;18(1):13–23.

[34] Mofokeng JP, Luyt AS, Tábi T, Kovács J. Comparison of injection moulded,natural fibre-reinforced composites with PP and PLA as matrices. JThermoplast Compos Mater 2012;25(8):927–48.

[35] Mathew AP, Oksman K, Sain M. Mechanical properties of biodegradablecomposites from poly lactic acid (PLA) and microcrystalline cellulose (MCC). JAppl Polym Sci 2005;97(5):2014–25.

Paper IV

Mechanical and Thermal Characterization of Compression MouldedPolylactic Acid Natural Fiber Composites Reinforced with Hemp andLyocell Fibers

Behnaz Baghaei,1 Mikael Skrifvars,1 Marja Rissanen,2 Sunil Kumar Ramamoorthy1

1School of Engineering, University of Borås, SE-501 90 Borås, Sweden2Department of Materials Science, Tampere University of Technology, FI-33101 Tampere, FinlandCorrespondence to: M. Skrifvars (E - mail: mikael.skrifvars@hb.se)

ABSTRACT: This research evaluates the effects of PLA/PP blend ratio and Lyocell/hemp mixture ratio on the morphology, water

absorption, mechanical and thermal properties of PLA-based composites. The composites were fabricated with 30 mass % hemp

using compression moulding. As a reference composites made from PP were also studied. Combining of hemp and Lyocell in PLA

composite leads to the reduction of moisture absorption and can improve the impact, tensile, flexural properties when compared

with PLA/hemp. Composite based on the PLA/PP blend-matrix could not improve the tensile and flexural properties compared with

PLA/hemp, however; the lighter composite with better impact properties was obtained. The crystallization temperature of the PLA-

PP/hemp increased compared with pure PLA. This result was also confirmed by the SEM micrographs. The moisture absorption of

PLA-PP/hemp was higher than PLA/hemp. Based on theoretical analysis of DMTA data, there was favorable adhesion in all compo-

sites. VC 2014 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2014, 131, 40534.

KEYWORDS: biopolymers and renewable polymers; composites; fibers; mechanical properties; molding

Received 18 December 2013; accepted 31 January 2014DOI: 10.1002/app.40534

INTRODUCTION

Poly(lactic acid) (PLA) is a biodegradable polymer produced

from lactic acid, which is made by fermentation of carbohydrate

sources, such as corn. It is considered as one of the most prom-

ising biomedical and packing materials with a broad market

prospect.1 Many of the PLA properties are compared to those

polyethylene (PE), polypropylene (PP), polystyrene (PS), and

polyethylene terephthalate (PET) (such as stiffness, tensile

strength, and gas permeability), turning PLA into a potential

substitutes to petroleum-based products.2 However, to date the

use of PLA in the engineering field is limited due to hydropho-

bicity, brittleness, low impact resistance, high cost as well as to

sensitivity to the temperature. To enhance the thermal stability

and mechanical properties of PLA, chemical modification, phys-

ical blending with some polymers such as poly(glycolic acid),

poly(hydroxyalkanoates), and poly(caprolactone), and using

natural fiber reinforcements can be done.1,3,4

Several researches on PLA blends with other polymers have

been carried out in order to modify the properties for PLA5–10

or to reduce the cost.11,12 Because PLA is biodegradable, blend-

ing of PLA with nonbiodegradable polymers, such as PP, HDPE,

LDPE, PS, and PET can improve the resistance of PLA to

hydrolysis, also the degradability of conventional polymers can

be improved by blending with PLA.13,14 PLA and PP are two

quite different candidate matrix materials with different advan-

tages, where PLA is a renewable biopolymer, and PP is a more

hydrophobic low-cost commodity thermoplastic and has high

toughness, and low density.15,16 In addition, PLA and PP have

similar processing temperatures (200–230�C).17 A brief attempt

has been made to blend and produce products such as fibers

from PLA and PP.14,18 However, there are limited numbers of

studies on the mechanical properties of the composites pro-

duced from PLA/PP bend.19

Low weight, low cost, recyclability, and biodegradability are

advantages of natural fibers compared to synthetic com-

pounds.20,21 These are also renewable and have relatively high

strength and stiffness. The reinforcement of PLA with lignocel-

lulosic fibers seems to be a viable alternative to increase their

mechanical performance and to preserve the environmentally

friendly character of the outcome.22

Hemp fibers can be considered an appropriate choice for rein-

forcing polymer composites due to their high stiffness, strength,

and aspect ratio.23 They also have an extremely high fiber yield

per unit density, and they are disease and pest-resistant,

VC 2014 Wiley Periodicals, Inc.

WWW.MATERIALSVIEWS.COM J. APPL. POLYM. SCI. 2014, DOI: 10.1002/APP.4053440534 (1 of 10)

enabling production methods with a low impact on the envi-

ronment.24 There are already some studies aimed at optimizing

hemp-fibers-reinforced PLA composites. Researchers have inves-

tigated the effects of different manufacturing processes, fiber

pre-treatments, fiber loading and fiber orientation on PLA/

hemp composites.25,26 It was found that PLA/hemp composites

mostly have good stiffness and tensile characteristics. However,

the impact properties are often worse than those of the pure

matrix, which is typical for a natural fiber-reinforced PLA com-

posite. Bax et al.27 investigated PLA/flax composites and

observed that the impact strength increased with increased flax

fiber content, but the impact strength of the composites was

clearly lower than that of the pure PLA sample. Oksman et al.28

studied the Charpy impact behavior of PLA/flax composites (40

wt % fiber), it was found that the composite impact strength

decreased by >25% compared to pure PLA. This effect was also

observed by M€ussig.29 He investigated the mechanical properties

of hemp fiber (40 wt %) reinforced PLA. The impact strength

was halved compared with pure PLA.

In contrast with bast fibers (like hemp fibers) with an elonga-

tion at break of approximately 5%,17 man-made cellulose

fibers have a higher elongation at break (>8–10%).30 Because

of the high elongation at break for the regenerated cellulose

fibers such as Rayon, Cupro or Lyocell, compared to the elon-

gation at break for the matrix, composites with high impact

strength values can be created. Lyocell fibers are known to

have good mechanical properties, wettability, high tenacity,

and good drapability as well as being environmentally friendli-

ness. Therefore, they have potential applications as reinforce-

ments for composites and can improve mechanical and

physical properties.31

In the context of this study, the mechanical characteristics of

composites in terms of the reinforcing fiber and matrix charac-

teristics were examined. Composites of PLA and hemp fibers,

PLA and hemp-Lyocell fiber mixtures, a PLA/PP blend and

hemp fibers, PLA and Lyocell fibers as well as PP and hemp

were investigated. The aim was to manufacture several compo-

sites based on two different kinds of reinforcing fiber and two

different matrices to investigate the characteristics of the com-

posites related to the fiber and matrix used. Different constitu-

ents can be used to tailor the composite characteristics for

diverse requirements, for instance the high stiffness of the hemp

fibers with the high elongation at break and resin wetting and

tenacity of the Lyocell fibers also, by combining the good

mechanical properties of PLA, including stiffness and strength

with the positive characteristics of polypropylene, including

high toughness and being cheap polymer.

MATERIALS AND METHODS

Materials

The PLA staple fibers, provided by Trevira GmbH (Hattersheim,

Germany), had a fineness of 1.7 dtex and a mean fiber length of

38 mm. Based on the manufacture’s information, the PLA fibers

were made from PLA Polymer 6202D from NatureWorksVR , Car-

gill Dow LLC (Minnetonka, USA). It had a density of 1.24 g

cm23, a melt temperature of 160–170�C, a glass transition tem-

perature of 60–65�C, 98% L-lactide, and a molecular weight of

97,000. The hemp, in the form of baled loose staple fibers

(genus species Cannabis Sativa L) was supplied by Hempage AG

(Adelsdorf, Germany). According to the manufacture’s info, the

average diameter of the hemp fiber was 20–40 lm and had a

mean fiber length of 30 mm. The Lyocell staple fibers were sup-

plied by Lenzing AG (Lenzing, Austria). The average length and

diameter of the fibers were 38 mm and 13.4 lm, respectively.

The PP fibers were supplied by FiberVisions (Varde, Denmark)

and had a fineness of 3.3 dtex and a mean fiber length of

66 mm.

Single Fiber Tensile Test. The hemp and Lyocell fibers were

tested on a Favigraph single-fiber tensile tester from Textechno

GmbH (M€onchengladbach, Germany) equipped with a 20-cN

load cell and with gauge length of 20 mm. The test speed was

20 mm min21 and the averages from 20 tests are reported in

Table I. To evaluate the effect of the composite processing con-

ditions on the hemp and Lyocell characteristics, single fibers

were treated in a hot press machine at 190�C and 1.7 MPa for

15 min, identical to the conditions used for the fabrication of

the composites, and then tested for tensile properties.

Table I summarizes the measured properties of the hemp and

Lyocell fibers used in this investigation.

Methods

Preparation of Prepreg and Composites. Five prepreg mats

with different fiber compositions were prepared:

� 70 mass % PLA 1 30 mass % hemp

� 70 mass % PLA 1 30 mass % Lyocell

� 70 mass % PLA 1 15 mass % hemp 1 15 mass % Lyocell

� 35 mass % PLA 1 35 mass % PP 1 30 mass % hemp

� 70 mass % PP1 30 mass % hemp

The needed amount of PLA, PP, Lyocell and the hemp staple

fibers were first weighed in their loose form and then they were

manually mixed, and fed to a carding machine from Mesdan

S.P.A (Brescian, Italy). During carding, the longer PLA and PP

fibers supported the shorter hemp and Lyocell fibers and

Table I. Tenacity, Modulus, Elongation, and Linear Density for Hemp and Lyocell Fibers Before and After Heat Treatment

Treatment Fibre Tenacity (cN/dtex) E-modulus (cN/dtex) Elongation (%) Linear density (dtex)

As received Hemp 4.66 (1.43) 97.71 (24.92) 3.93 (0.66) 4.15

Lyocell 3.27 (0.63) 55.87 (23.06) 8.70 (2.01) 1.36

After hot pressing Hemp 3.75 (1.10) 84.62 (26.60) 3.37 (0.88) 4.27

Lyocell 2.12 (0.92) 48.46 (20.29) 4.75 (1.84) 1.38

ARTICLE WILEYONLINELIBRARY.COM/APP

WWW.MATERIALSVIEWS.COM J. APPL. POLYM. SCI. 2014, DOI: 10.1002/APP.4053440534 (2 of 10)

provided the fiber-to-fiber cohesion which resulted in a web

suitable for further processing. The blended fiber web was

carded three times to parallelize the fibers and to achieve mat

uniformity.

The obtained prepreg mats were dried in a vacuum chamber

(0.9 mbar; 70�C) for at least 18 h before compression molding.

After drying the mats were stored under dry conditions. The

prepreg was covered by a Teflon sheet to prevent sticking of the

matrix to the surface of the mould before it was placed into a

preheated steel mould with a 20 3 20 cm2 cavity and 10-mm

depth. The steel mould was of own design, and machined by a

local machine shop. The mould was then placed in a hydraulic

compression-molding machine from Rondol Technology (Staf-

fordshire, UK). Compression molding was done at a tempera-

ture of 195�C and a pressure of 1.7 MPa for 20 min. Neat PLA

sheets, to be used as reference material, were made by melting

PLA fibers under the same processing conditions.

Test Methods. The specimens were stored at ambient conditions

after processing. Then, before testing, the specimens were condi-

tioned for at least 24 h at 23�C and 50% relative humidity

according to DIN EN ISO 291. The test specimens were cut by

laser according to the standard dimensions, given below.

Composite density and porosity. The densities of the compo-

sites were determined by the buoyancy method (Archimedes’

principle), using ethanol as the displacement medium. Before

the specimens were immersed in ethanol, they had been covered

by a varnish containing paraffin to avoid absorption during

immersion. The fiber volume fraction was calculated from the

fiber weight fraction with allowance for porosity using the

method proposed by Madsen et al.32

Water absorption test. Water absorption analysis was done on

composite specimens according to ASTM D570-98. The specimen

dimension was �36 3 10 mm2. The specimens were first dried in

an oven for 24 h at 60�C, and then put in a desiccator in order

to cool down to room temperature. The measured weights of

these specimens were denoted as W0. The specimens were then

immersed in two different baths, one of distilled water at room

temperature and one of hot water at 80�C. The amount of water

absorbed was measured every 24 h for 10 days. At each measuring

point the specimen was removed from the bath, and the surface

was wiped dry and the weight was recorded as W. The percentage

of apparent weight gain (WG) was calculated using eq. (1).

WG5 W 2W0ð Þ=W0½ � 3 100% (1)

Tensile test. The tensile testing was done according to ISO 527

in a universal H10KT testing machine equipped with a mechan-

ical extensometer (model 100R long travel extensometer),

attached to the central part of the specimen by clips. The testing

parameters were a loading rate of 10 mm min21 and a loading

range of 1 kN. The average tensile values were collected from

six separate measurements. The testing machine and the exten-

someter were supplied by Tinius Olsen (Salford, UK).

Dynamic mechanical thermal analysis. The viscoelastic proper-

ties of the specimens were measured using a dynamic mechani-

cal analyser (DMTA Q-series TA instrument supplied by Waters

LLC, Newcastle, DE). Rectangular specimens with the dimen-

sions 50 mm 3 8 mm and 1–2 mm thickness were used using

the dual cantilever method. The measurements were performed

at a frequency of 1 Hz and amplitude of 15 lm. The tempera-

ture range was from 30 to 150�C at a scanning rate of 3�Cmin21. The storage modulus (E0) and loss factor (tan d) of the

specimens were measured as a function of temperature.

Flexural testing. The flexural test was performed using the same

testing machine as for tensile testing according to ISO 14125

standard test method for fiber–reinforced plastic composites. At

least five specimens were assessed for each batch of samples. The

loading rate was 10 mm min21 and the load range was 1 kN.

The specimen dimension was 80 3 15 mm2 (length 3 width),

and the thickness varied depending on the sample. The outer

span was 64 mm and the range of displacement was 20 mm. At

least five specimens were assessed for each sample reported the

average values.

Impact testing. The Charpy impact strength of the composites

was tested according to ISO 179. A pendulum type Zwick test

instrument from Zwick GmbH (Ulm, Germany) was used to

measure the unnotched, rectangular specimens (80 mm 3 10

mm 3 1–2 mm). For each material, ten specimens were tested

edgewise.

Differential scanning calorimetry. The DSC analysis was done

on a DSC Q2000 supplied by TA Instruments, New Castle, DE.

Samples of �10 mg were heated at a rate of 10�C min21 in a

nitrogen-purge stream from 20 to 200�C, then cooled to 0�C,

and heated again from 20 to 200�C. The data from the first

scan were used. For each sample, three replicates were scanned

in order to get average values. The percentage crystallinity

(XDSC) of PLA and PP was calculated using eq. (2)33:

XDSC %5DHf 2DHcc

DHof

3100

w(2)

where DHof 5 93 J g21 for 100% crystalline PLA and 209 J g21

for 100% crystalline PP DHf is the enthalpy of melting, DHcc is

the cold crystallization enthalpy, and w is the weight fraction of

PLA in the composite.

Scanning electron microscopy (SEM). Composite fracture sur-

face morphologies were studied using low-vacuum scanning-

electron microscopy, a JEOL JSM 6610LV instrument, JEOL

(Tokyo, Japan) with an operating voltage of 5–10 kV. For low

vacuum imaging, no specific preparation was required.

RESULTS AND DISCUSSION

Single Fiber Tensile Test

To evaluate the effect of the high temperature compression

molding conditions, the tensile properties of heat treated hemp

and Lyocell staple single fibers were determined and compared

to untreated single fibers, see Table I. Heat treatments greatly

reduce the mechanical properties. Cellulosic fibers are mixtures

of organic materials (cellulose, hemicellulose, and lignin) and

heat treatment at elevated temperatures can cause physical and

chemical changes. The physical changes are associated with

enthalpy, weight, color, strength, crystallinity, and orientation of

microfibril angle.34 The chemical changes are related to the

ARTICLE WILEYONLINELIBRARY.COM/APP

WWW.MATERIALSVIEWS.COM J. APPL. POLYM. SCI. 2014, DOI: 10.1002/APP.4053440534 (3 of 10)

decomposition of the molecular structure. Heat treatment

results in the weight loss of moisture plus weight loss due to

thermal degradation. The thermal degradation of cellulosic

fibers results in change in color and deterioration in mechanical

properties of the fibers.35,36

Influence of Different Natural Fibers

In this study, five different composites were produced by com-

pression molding of prepregs containing 30 mass % reinforce-

ment (hemp respective Lyocell) and 70 mass % matrix, (PP

respective PLA). The physical characteristics of the composites

made are summarized in Table II. The porosity ranged from 2.1

to 14.9 vol %, and the higher porosity was seen for the PP/

hemp and PLA-PP/hemp composites. The porosity fraction is

rather high even if the fiber volume fraction was as low as 30

mass %, and this could be due to several factors. The complex-

ity of the surface chemistry and the irregularity of the morphol-

ogy of plant fibers is one of the most important considerations

as well as to the presence of luminal cavities.32 It can be

observed that the Lyocell-based composite had the lower void

content which is obviously due to the higher interfacial

adhesion.

Water Absorption Characteristics

Figure 1 shows the apparent weight gain (WG) as a function of

the immersion time at different temperatures for the manufac-

tured composites. In Figure 1(a), it can be seen that for all the

composites investigated, WG increases monotonically with time.

This increase in WG is consistent with other studies of natural

fiber composites.37–39 The absence of the induction period char-

acterized by zero weight gain at the initial stage of water

immersion treatment, could be explained by the different mate-

rial packing between the skin and core regions, which indicates

that the hemp and Lyocell fibres were uniformly dispersed in

the PLA and PP matrix.16

It can also be seen that the slopes of the WG vs. time plots were

steeper for all composites compared with the slopes for neat PP

and PLA. It can be observed that the composites exhibited sig-

nificantly higher water absorption than for neat PLA and PP

due to the hydrophilic nature of Lyocell and hemp because of

the presence of polar groups such as AOH and ACOOH in the

fibers. These results are in accordance with published data

showing that the lignocellulosic fibres displayed a higher tend-

ency to absorb water than does the PLA and PP.39–41 Among all

composites in this study, PLA/Lyocell and PLA/hemp-Lyocell

showed the lowest water absorptions (up to 6.7 and 10.3 wt %,

respectively), which may be due to their lower porosity and also

better interfacial adhesion which decreases the thickness of the

interphase area between the fibers and the matrix, thus decreas-

ing the water absorption through the interphase and further

into inner parts of the structure.42,43 From the results, it can be

Table II. Composition of the Fabricated Composites Calculated from the Density Measurements

Sample Density (g cm23)

Fibre massfraction (%)

Fibre volumefraction (%)

Matrix volumefraction (%)

Porosity volumefraction (%)Hemp Lyocell

PLA 1.2490 (0.0002) – – 0.0 100.0 –

PLA/hemp 1.0914 (0.0154) 30.0 – 23.0 (0.6) 61.2 (0.9) 10.8 (0.2)

PLA/Lyocell 1.2765 (0.0112) – 30 25.2 (0.2) 71.5 (0.6) 3.3 (0.9)

PLA/hemp-Lyocell 1.1222 (0.0169) 15.0 15.0 22.4 (0.3) 76.4 (1.2) 2.1 (0.1)

PLA-PP/hemp 1.0171 (0.0402) 30 – 20.6 (0.8) 65.9 (12.6) 11.9 (3.5)

PP/hemp 0.8758 (0.0154) 30 – 17.8 (0.4) 67.4 (1.4) 14.9 (1.8)

PP 0.9103 (0.0019) – – 0.0 100.0 –

Note: Data in the brackets are mean (SD).

Figure 1. Apparent weight gain against time for the manufactured com-

posites at: (a) room temperature and (b) 80�C.

ARTICLE WILEYONLINELIBRARY.COM/APP

WWW.MATERIALSVIEWS.COM J. APPL. POLYM. SCI. 2014, DOI: 10.1002/APP.4053440534 (4 of 10)

concluded that the PP/hemp had higher moisture absorption

compared with PLA/hemp. It could be explained by the lower

interfacial compatibility between hemp fiber and PP matrix,

which can result in incomplete wetting resulting in microcracks,

offering channels for better moisture transport.44 The PLA-PP/

hemp composite showed higher moisture absorption than PLA-

hemp composite.

As shown in Figure 1(b), after 5 days at 80�C, the neat PLA

sample was very soft and it was difficult to decant water and

rinse the sample without losing some material. The experiment

for neat PLA was therefore discontinued after 5 days. Further-

more, it can be seen that the water immersion temperature

influences the water absorption curves. Increasing the immer-

sion temperature from room temperature to 80�C increased the

water absorption of the neat PLA and composites, as well as

shortened the saturation time. The immersion temperature

affected PLA/hemp-Lyocell more than it did for PLA/Lyocell

composites. The weight gain was found to decrease after passing

through a maximum. For PLA composites, this was due to the

formation of a neat resin peel on the surface resulting from

degradation and dissolving with time together with the removal

of some substances from the hemp and Lyocell fiber during the

immersion.45–47 In contrast, even though the apparent weight

gain for the PP composite also decreased after passing through

a maximum, the PP dissolution did not take place. In fact, it

has been reported that the PP is very resistant to moisture

attack within even at elevated temperature.37,39,46 It is of interest

to understand the mechanisms that cause the decrease in weight

gain. The hemp fibers are responsible for the weight loss

observed, which is due to the removal of certain fractions from

the hemp fibers during the water immersion.

Tensile Properties

An overview of the tensile strength and modulus of the compo-

sites compared with the values of the neat PLA and PP matrix is

given in Figure 2(a,b). Compared with the tensile modulus of the

pure PLA matrix [Figure 2(a)], there is an improvement of 138%

for PLA/Lyocell, 117% for PLA/hemp-Lyocell, 106% for PLA/

hemp and 59% for the PLA-PP/hemp composites. The tensile

modulus have been decreased by the admixture of the PP fibers

in the PLA/hemp composite compared with the PLA/hemp com-

posite since the tensile modulus of PP is lower than that of PLA.

As can be seen in Figure 2(b), the tensile strength of PLA com-

posites was significantly higher than the tensile strength of the

neat PLA and neat PP matrix, which was 39.7 and 24.6 MPa,

respectively. Neat PLA has better mechanical properties than neat

PP.28 In the comprehensive outline, the highest tensile strength

values were reached by PLA/Lyocell with 80.9 MPa, PLA/hemp-

Lyocell with 60.6 MPa, followed by PLA/hemp with 45.7 MPa,

PLA-PP/hemp with 29.8 MPa and PP/hemp with 26.9 MPa.

The improvement of the tensile properties of hemp-reinforced

PLA composites by addition of Lyocell fibers could be attributed

to the lower amount of porosity in the composite and the

much higher fineness of the Lyocell fibers, which promote a

better fiber–matrix adhesion.

Flexural Properties

The results of the flexural testing are shown in Figure 3(a,b).

The flexural modulus of the composites were all higher than for

neat PLA and neat PP, which is especially evident for the PLA/

Lyocell which had a flexural modulus (5.8 GPa) three times

higher than that of neat PLA (1.9 GPa) [Figure 3(a)]. Moreover,

the superior flexural properties of the PLA composites with

hemp-Lyocell compared with the PLA composites with hemp

can be attributed to the better adhesion between Lyocell fibers

with PLA matrix as well as lower porosity. It is also obvious

from the results that composites made from PLA-PP/hemp

showed an evident improvement in flexural modulus across the

PP/hemp composite, but lower than that of PLA/hemp compos-

ite. As shown in Figure 3(b), the flexural strength showed an

increase of 43.5% for the PLA/Lyocell composite and 23.5% for

the PLA/hemp-Lyocell composite compared with neat PLA.

However, the flexural strengths of the other composites were all

lower than of neat PLA.

Impact Resistance

The results of the Charpy impact test are shown in Figure 4. For

the neat PLA matrix, an impact strength value of 11.5 kJ m22

Figure 2. Tensile properties of the manufactured composites: (a) tensile modulus, and (b) tensile strength.

ARTICLE WILEYONLINELIBRARY.COM/APP

WWW.MATERIALSVIEWS.COM J. APPL. POLYM. SCI. 2014, DOI: 10.1002/APP.4053440534 (5 of 10)

was measured. Similar values have also been seen in other stud-

ies.27,28 A significant reinforcement effect was determined for the

PLA/Lyocell composites with an impact value of 26 kJ m22 and

for the PLA/hemp-Lyocell composite with an impact value of

21.7 kJ m22, while the value of the PLA/hemp composites

(9.7 kJ m22) was lower than for the neat PLA matrix. This can

be explained by a higher elongation at break of Lyocell compared

with hemp fibers resulting in an improvement of the impact

strength in the composites.48,49 Hence, the highest impact value

was reached by PLA/hemp-Lyocell due to the higher elongation

characteristics of the Lyocell fibers, while using hemp fibers

resulted in a brittle character of the PLA composites. The behav-

ior of the hemp fiber is in accordance with the known behavior

of natural fibers under an impact load.

The impact strength of pure PP was around 18.9 kJ m22. Simi-

lar values have also been seen in other studies.50,51 However, the

impact strength of PP-based composites was lower than that of

neat PP. It has been reported that the un-notched impact resist-

ance of thermoplastic natural fibers composites generally show a

decreasing trend. The behavior of fiber reinforcement under

impact load is more complicated than under bending and ten-

sile load since the impact strength is attributed to the energy

consumption during failure, and also attributed to the addition

of fibers which probably creates regions of stress concentration

which require less energy to initiate a crack.

A mixture of PLA-PP/hemp resulted in a further improvement

of the impact strength up to 9.8 kJ m22 compared with PLA/

hemp with 8.8 kJ m22.

Dynamic Mechanical Thermal Testing

Figure 5(a,b) shows the dynamic viscoelastic curves for neat

PLA and composites made from PLA. The storage modulus (at

30�C) increased from 2.5 GPa for pure PLA, up to 3.7 GPa for

PLA/Lyocell followed by 3.2 GPa for PLA/hemp-Lyocell, 3.1

GPa for PLA/hemp and 2.6 GPa for PLA-PP/hemp composite

[Figure 5(a)]. The storage modulus increased when hemp and

Lyocell fibers were introduced into PLA. The increase in the

stiffness of fiber-containing samples, which reveals effective

stress transfer from the fiber to the matrix at the interface, can

be interpreted as good adhesion between the fibers and the mat-

rices.1,52–54 The sharp decrease in the storage modulus (around

57–61�C for most of the samples) corresponds to the

a-relaxation of the amorphous regions in PLA.55

The storage modulus started to increase again at temperatures

of around 90–100�C, which is a result of the cold crystallization

of PLA and this peak shifted to a lower temperatures with the

addition of PP. The cold crystallization was also observed in the

DSC curves. This result suggested that the incorporation of PP

increased the cold-crystallization ability of PLA.56

The dampening, or tan d, is the ratio between the loss modulus

and the storage modulus and provides information about the

internal friction of the material and the adhesion of the inter-

face. For a composite, the molecular motion in the interphase

will contribute to the dampening. A larger area under the

a-relaxation peak in the tan d curves of a polymer indicates

that the molecular chains exhibit a higher degree of mobility

thus better damping properties.57 The area under this peak for

PLA-based composites [Figure 5(b)] seems to be smaller com-

pared with neat PLA. A possible explanation is that there is a

good interaction between the fiber and the matrix. Adding

Figure 3. Flexural properties of the composites: (a) flexural modulus, and (b) flexural strength.

Figure 4. Impact strength of the manufactured composites.

ARTICLE WILEYONLINELIBRARY.COM/APP

WWW.MATERIALSVIEWS.COM J. APPL. POLYM. SCI. 2014, DOI: 10.1002/APP.4053440534 (6 of 10)

fibers to the matrix will decrease the mobility of the polymer

chains, and consequently reduce the dampening, as reported by

Pothan et al.58 The glass-transition temperature is often

recorded at the maximum of the tan d. The recorded Tg of the

samples were 65.6�C for pure PLA, 67.5�C for PLA/Lyocell,

68�C for PLA/hemp-Lyocell, 68.8�C for PLA-PP/hemp and

69.7�C for PLA/hemp. It is obvious that incorporating fibers

increases the Tg for both PLA and PP. The slight shift in Tg to a

higher temperature (by a few degrees) indicates that the mobil-

ity of polymer chains is affected. Mathew et al.59 discussed that

the shift to higher temperature usually indicates restricted

movement of molecules because of better interaction between

the fiber and the polymer matrix.

The broad transition (between 90 and 110�C) for all the fiber-

reinforced PLA relates to the cold crystallization of PLA in this

temperature region.

Differential Scanning Calorimetry

To examine the effect of the added fibers on the crystallinity of

PLA, DSC analysis was performed. The DSC heating and cooling

thermograms for all composites produced are shown in Figure

6(a,b). The glass transition temperature (Tg), crystallization tem-

perature (Tc), and melting temperature (Tm) obtained from the

DSC studies are summarized in Table III. The data indicates that

Tg and Tm was slightly affected by the introduction of hemp and

Lyocell fibers with a few decrease. Furthermore the Tc decreased

from 112.1�C of pure PLA to 106.4�C of composite containing

hemp-Lyocell fibers. It can be clearly seen that the degree of crys-

tallization increases from 4.6% for neat PLA up to 8.3% for

PLA/hemp, 16.9% for PLA/Lyocell and 11.8% for PLA/hemp-

Lyocell. It is noteworthy to see that the above increase in the

crystallinity of PLA with respect to fiber content is consistent

with other studies of natural fiber/PLA composites.3,60,61 The

double melting peaks suggest that occurrence of the crystal’s reor-

ganization during the heating run.62–64

In comparison with PP, the incorporation of hemp fibers shifts

the crystallization temperature to the higher temperature. This

Figure 6. DSC thermograms from the cooling and the second heating run

for the fabricated composites.

Figure 5. DMTA analysis of PLA and PP reinforced with hemp and Lyo-

cell fibers. (a) Storage modulus vs. temperature; (b) tan d curves.

Table III. Differential Scanning Calorimetric Data for the Composites

from the Cooling and Second Heating Run (10�C min21)

Sample XDSC (%) Tg (�C) TC (�C) Tm (�C)

PLA 4.6 59.2 112.1 167.4

PLA/hemp 8.3 59.4 108.6 166.8

PLA/Lyocell 16.9 57.6 108.7 166.7

PLA/hemp-Lyocell 11.8 57.5 106.4 166.3

PLA-PP/hemp 5.3 58.8 117.9 166.5

PP/hemp 1.4 – 117.3 164.4

PP 0.6 – 112.5 166.6

ARTICLE WILEYONLINELIBRARY.COM/APP

WWW.MATERIALSVIEWS.COM J. APPL. POLYM. SCI. 2014, DOI: 10.1002/APP.4053440534 (7 of 10)

increase suggests that hemp fibers possibly act as heterogeneous

nucleating agents causing the PP to crystalize at higher temper-

ature.65 As evident from the crystallization thermogram Tc [Fig-

ure 6(b)], the incorporation of PP within the PLA matrix

increased Tc of the PLA from 112 to 117.9�C. Moreover, the

crystallinity of the PLA in its blend with PP slightly increased.

Scanning Electron Microscopy Analysis

Figure 7 shows the SEM micrographs of the fracture surface for

t tensile specimens for different composites. Compared to PLA/

hemp composites [Figure 7(a)], less pulled-out fibers and corre-

sponding holes are visible for the PLA/Lyocell-hemp and PLA/

Lyocell composites [Figure 7(b,c)]. These observations suggest

that the adhesion between the PLA matrix and the Lyocell fiber

is quite good. It should be noted that the finer Lyocell fibers

were used which lead to the bigger specific surface between fiber

and matrix and better adhesion between them.

As shown in Figure 7(d) the PP and PLA polymers do not have

the well-defined spherical shapes and the separation of the two

components or pores.

CONCLUSIONS

The effects of the choice of matrix and reinforcing filler on the

structure and properties of PLA-based composites have been

studied, focusing on the water absorption, mechanical and

thermo-mechanical properties. Based on the mechanical tests,

the obtained results showed that combining hemp and Lyocell

in a PLA-based composite can improve significantly the impact

strength at ambient temperature, flexural and tensile strength

and modulus, compared with hemp fiber reinforced PLA. More-

over, the moisture absorption was reduced by up to 47.4%.

From the DMTA results, it is evident that incorporation of the

hemp and Lyocell fiber gives a considerable increase in storage

modulus and a decrease in tan d values. The study performed

using DSC revealed that the melting point of PLA was not

affected significantly after reinforcement with the hemp-Lyocell

mixture but the glass transition temperature increased a few

degrees. The PLA in the composites had high orientation degree

and crystallinity which was attributed to effective heterogeneous

nucleation induced by hemp and Lyocell fibers, however, the

degree of crystallinity of PLA/hemp composite was higher.

Figure 7. SEM images of tensile fracture surfaces of (a) PLA/hemp, (b) PLA/Lyocell, (c) PLA/hemp-Lyocell, and (d) PLA-PP/hemp composites.

ARTICLE WILEYONLINELIBRARY.COM/APP

WWW.MATERIALSVIEWS.COM J. APPL. POLYM. SCI. 2014, DOI: 10.1002/APP.4053440534 (8 of 10)

Although Lyocell is expensive, it is reproducible by artificial

production and the admixture of Lyocell fibers to hemp fibers

leads to less quality variation in the fabricated composites.48

On the contrary, the combination of PP and PLA had a destruc-

tive effect on the mechanical properties. Composites based on

the PLA/PP blend-matrix containing hemp had only enhanced

impact strength of the composite, compared with the PLA/

hemp composite but not the flexural, tensile and damping

properties. The crystallization temperature of the hemp-

reinforced PLA-PP composites decreased compared with pure

PLA indicating that the incorporation of PP improved the

cold-crystallization ability of PLA. The changes in the crystal

structure and melting points PLA-PP/hemp composite indicate

partial compatibility between the two polymers. This result was

also confirmed by the SEM micrographs, in which no well-

defined spherical shapes and separation of the two polymers

were observed. The PLA-PP/hemp composite showed higher

water absorption than the PLA/hemp did. In addition to

improving the impact properties of PLA, the PLA-PP/composite

are lighter, with relatively lower price and can be engineered to

have controlled degradability for different applications.

REFERENCES

1. Park, G. S. Polym. J. 1986, 18, 209.

2. Matthew, H. N. In Plastics Technology; Gardner Publica-

tions, Inc: Cincinnati, 1998; p 13.

3. Pickering, K. L.; Sawpan, M. A.; Jayaraman, J.; Fernyhough,

A. Compos. A 2011, 42, 1148.

4. Tserki, V.; Matzinos, P.; Panayiotou, C. Compos. A 2006, 37,

1231.

5. Li, Y.; Shimizu, H. Eur. Polym. J. 2009, 45, 738.

6. Sarazin, P.; Li, G.; Orts, W. J.; Favis, B. D. Polymer 2008, 49, 599.

7. Wang, N.; Yu, J.; Chang, P. R.; Ma, X. Carbohydr. Polym.

2008, 71, 109.

8. Huneault, M. A.; Li, H. Polymer 2007, 48, 270.

9. Wang, N.; Yu, J.; Chang, P. R.; Ma, X. Starch St€arke 2007,

59, 409.

10. Wang, N.; Yu, J.; Ma, X. Polym. Int. 2007, 56, 1440.

11. Biresaw, G.; Carriere, C. J. J. Polym. Sci. B Polym. Phys.

2002, 40, 2248.

12. Hamad, K.; Kaseem, M.; Deri, F. Polym. Bull. 2010, 65, 509.

13. Reddy, N.; Nama, D.; Yang, Y. Polym. Degrad. Stab. 2008, 93, 233.

14. Hamad, K.; Kaseem, M.; Deri, F. J. Polym. Res. 2011, 18, 1799.

15. Peijs, T.; Garkhail, S.; Heijenrath, R.; van den Oever, M.;

Bos, H. Macromol. Symp. 1998, 127, 193.

16. Hornsby, P. R.; Hinrichsen, E.; Tarverdi, K. J. Mater. Sci.

1997, 32, 1009.

17. Pearce, M. L. A. E. M. Handbook of Fiber Chemistry; Mar-

cel Dekker: New York, 1998.

18. Wojciechowska, E.; Fabia, J.; Slusarczyk, C.; Gawlowski, A.;

Wysocki, M.; Graczyk, T. Fibres Text. East. Eur. 2005, 13, 126.

19. Kim, H.-S.; Kim, H.-J. Fiber. Polym. 2013, 14, 793.

20. Mishra, S.; Mohanty, A. K.; Drzal, L. T.; Misra, M.;

Hinrichsen, G. Macromol. Mater. Eng. 2004, 289, 955.

21. Jayaraman, K. Compos. Sci. Technol. 2003, 63, 367.

22. Girones, J.; Lopez, J. P.; Mutje, P.; Carvalho, A. J. F.;

Curvelo, A. A. S.; Vilaseca, F. Compos. Sci. Technol. 2012, 72,

858.

23. Mutj�e, P.; L�opez, A.; Vallejos, M. E.; L�opez, J. P.; Vilaseca, F.

Compos. A. 2007, 38, 369.

24. Kostic, M.; Pejic, B.; Skundric, P. Bioresour. Technol. 2008,

99, 94.

25. Hu, R.; Lim, J.-K. J. Compos. Mater 2007, 41, 1655.

26. Masirek, R.; Kulinski, Z.; Chionna, D.; Piorkowska, E.;

Pracella, M. J. Appl. Polym. Sci. 2007, 105, 255.

27. Bax, B.; M€ussig, J. Compos. Sci. Technol. 2008, 68, 1601.

28. Oksman, K.; Skrifvars, M.; Selin, J. F. Compos. Sci. Technol.

2003, 63, 1317.

29. M€ussig, J. N-FibreBase KongressNaturfaserverst€arkte

Kunststoffe (NFK) – Wood-plastic-composites (WPC) Bio

Kunststoffe; nova-Institut: H€urth, Germany, June 27–28,

2006. (in German).

30. Ganster, J.; Fink, H.-P. Cellulose 2006, 13, 271.

31. Johnson, R. K.; Zink-Sharp, A.; Renneckar, S. H.; Glasser,

W. G. Compos. A 2008, 39, 470.

32. Madsen, B.; Thygesen, A.; Lilholt, H. Compos. Sci. Technol.

2007, 67, 1584.

33. Pilla, S.; Gong, S.; O’Neill, E.; Rowell, R. M.; Krzysik, A. M.

Polym. Eng. Sci. 2008, 48, 578.

34. Yang, P.; Kokot, S. J. Appl. Polym. Sci. 1996, 60, 1137.

35. Prasad, B. M.; Sain, M. M.; Roy, D. N. J. Mater. Sci. 2005,

40, 4271.

36. Wielage, B.; Lampke, T.; Marx, G.; Nestler, K.; Starke, D.

Thermochim. Acta 1999, 337, 169.

37. Thwe, M. M.; Liao, K. Compos. Sci. Technol. 2003, 63, 375.

38. George, J.; Bhagawan, S. S.; Thomas, S. Compos. Sci. Tech-

nol. 1998, 58, 1471.

39. Arbelaiz, A.; Fern�andez, B.; Ramos, J. A.; Retegi, A.; Llano-

Ponte, R.; Mondragon, I. Compos. Sci. Technol. 2005, 65,

1582.

40. Finkenstadt, V. L.; Liu, L.; Liu, C.-K.; Evangelista, R.;

Cermak, S. C.; Hojilla-Evangelista, M.; Willett, J. L. Ind.

Crop. Prod. 2007, 26, 36.

41. Tawakkal, I.; Talib, R. A.; Abdan, K.; Ling, C. N. Bioresources

2012, 7, 1643.

42. Qin, L.; Qiu, J.; Liu, M.; Ding, S.; Shao, L.; L€u, S.; Zhang,

G.; Zhao, Y.; Fu, X. Chem. Eng. J. 2011, 166, 772.

43. Toro, P.; Quijada, R.; Murillo, O.; Yazdani-Pedram, M.

Polym. Int. 2005, 54, 730.

44. Almgren, K. M.; Gamstedt, E. K.; Berthold, F.; Lindstrom,

M. Polym. Compos. 2009, 30, 1809.

45. Mofokeng, J. P.; Luyt, A. S.; T�abi, T.; Kov�acs, J. J. Thermo-

plast. Compos. Mater. 2012, 25, 927.

46. Chow, C. P. L.; Xing, X. S.; Li, R. K. Y. Compos. Sci. Technol.

2007, 67, 306.

ARTICLE WILEYONLINELIBRARY.COM/APP

WWW.MATERIALSVIEWS.COM J. APPL. POLYM. SCI. 2014, DOI: 10.1002/APP.4053440534 (9 of 10)

47. Kumar, R.; Yakubu, M. K.; Anandjiwala, R. D. Express

Polym. Lett. 2010, 4, 423.

48. Graupner, N. J. Compos. Mater. 2009, 43, 689.

49. Graupner, N.; Mussig, J. J. Biobased Mater. Bio. 2009, 3, 249.

50. Yuan, Q.; Wu, D.; Gotama, J.; Bateman, S. J. Thermoplast.

Compos. Mater. 2008, 21, 195.

51. Lapcik, L.; Jindrova, P.; Lapcikova, B.; Tamblyn, R.; Greenwood,

R.; Rowson, N. J. Appl. Polym. Sci. 2008, 110, 2742.

52. Islam, M. S.; Pickering, K. L.; Foreman, N. J. J. Polym. Envi-

ron. 2010, 18, 696.

53. Majhi, S. K.; Nayak, S. K.; Mohanty, S.; Unnikrishnan, L.

Int. J. Plast. Technol. 2010, 14, 57.

54. Andreopoulos, A. G.; Tarantili, P. A. J. Appl. Polym. Sci.

1998, 70, 747.

55. Mathew, A. P.; Oksman, K.; Sain, M. J. Appl. Polym. Sci.

2006, 101, 300.

56. Liu, P.; Ouyang, Y.; Xiao, R. J. Appl. Polym. Sci. 2012, 123,

2859.

57. Pilla, S.; Kramschuster, A.; Yang, L. Q.; Lee, J.; Gong, S. Q.;

Turng, L. S. Mater. Sci. Eng. 2009, 29, 1258.

58. Pothan, L. A.; Oommen, Z.; Thomas, S. Compos. Sci. Tech-

nol. 2003, 63, 283.

59. Mathew, A. P.; Oksman, K.; Sain, M. J. Appl. Polym. Sci.

2005, 97, 2014.

60. Masirek, R.; Kulinski, Z.; Chionna, D.; Piorkowska, E.;

Pracella, M. J. Appl. Polym. Sci. 2007, 105, 255.

61. Sawpan, M. A.; Pickering, K. L.; Fernyhough, A. Compos. A

2011, 42, 310.

62. Lee, S.-H.; Wang, S. Compos. A 2006, 37, 80.

63. Song, Y.; Liu, J.; Chen, S.; Zheng, Y.; Ruan, S.; Bin, Y.

J. Polym. Environ. 2013, 21, 1117.

64. Wang, Y.; Mano, J. F. Eur. Polym. J. 2005, 41, 2335.

65. Niu, P.; Wang, X.; Liu, B.; Shengru, L.; and Jie, Y. J. Comp.

Mater. 2012, 46, 203.

ARTICLE WILEYONLINELIBRARY.COM/APP

WWW.MATERIALSVIEWS.COM J. APPL. POLYM. SCI. 2014, DOI: 10.1002/APP.4053440534 (10 of 10)

Paper V

1

Characterisation of polylactic acid biocomposites made

from prepregs composed of woven polylactic acid/hemp-

Lyocell hybrid yarn fabrics

Behnaz Baghaei

1 and Mikael Skrifvars

*,1

1 Swedish Centre for Resource Recovery, Academy for Textile, Engineering and Business

University of Borås, SE-501 90 Borås, Sweden

* Correspondence to Mikael Skrifvars: Tel: +46 33 435 4497; Fax: + 46 33 435 4008

E-mail address: mikael.skrifvars@hb.se

2

Abstract

This paper describes the mechanical properties and water absorption characteristics for

biocomposites made from woven PLA/hemp/Lyocell prepregs. The aim was to improve

the properties with the addition of Lyocell fibre into a hybrid yarn. Well-aligned hybrid

yarns composed of hemp/PLA, hemp-Lyocell/PLA, respective, Lyocell/PLA were made

by wrap spinning. Unidirectional satin fabrics were made by weaving with PLA (warp)

and the hybrid yarns (weft). Uniaxial composites were fabricated with 30 fibre mass-%

using compression moulding. The composites were investigated for tensile, flexural and

impact properties. Combining hemp with Lyocell in a PLA matrix improves the

mechanical properties, compared to hemp/PLA composites. The composite made from

the satin Lyocell/PLA fabric gave the best mechanical properties. The type of fibre

reinforcement compositions did not significantly affect the water absorption of the

biocomposites. Scanning electron microscopy showed that fibre pull-outs appear more

often in hemp/PLA composites than in composites also including Lyocell fibre.

Keywords:

A. Fabrics/textiles

B. Mechanical properties

E. Weaving

E. Compression moulding

3

1. INTRODUCTION

Natural fibre reinforced polymers are an emerging material that has great potential to be

applied in lightweight engineering applications. Polymer matrices can be divided into

two categories: thermoplastic and thermosetting polymers. The thermosets have

traditionally been used as reinforced, but today thermoplastics find more applications as

reinforced composites. Since the early 1990s, hybrid composite materials made from

natural fibres and thermoplastic matrix polymers have been commonly processed in

compression moulding [1]. Thermoplastic composites have several advantages over

thermoset composites in terms of processing and mechanical properties. Thermoplastic

polymers can be reshaped upon the application of heat, and this can be done repeatedly.

On the other hand, thermosetting polymers undergo chemical reactions during curing,

which cross-link the polymer molecules, causing the resins to become permanently hard

and rigid. However, thermoplastic molecules are associated with physical

intermolecular forces, and they can move under stress. Thereby, thermoplastic

composites usually have higher strains at failure than thermoset composites; thus, they

possess better fracture toughness and fatigue endurance. Thermoplastic composites

usually need a shorter and simpler processing cycle which involves no chemical

reactions, and they are therefore used in composite applications requiring short process

time [2].

Bio-based polymers are increasingly replacing the petroleum-based polymers in

many industrial applications. The most important thermoplastic biopolymer is polylactic

acid (PLA), which is produced from lactic acid obtained from carbohydrate feedstock.

PLA is today commercially used in biodegradable packing materials and other

applications with short lifespan. Nevertheless, various studies have shown that PLA is

also suitable as matrix for embedding of fibres in composites [3, 4].

4

Several research studies have been carried out on cellulose fibre-reinforced PLA

composites, but there is still a need for more optimisation of these materials to meet the

demands required in automotive components and construction materials. Beside annual

plant natural fibres, man-made fibres based on regenerated cellulose such as Lyocell or

viscose can be used to reinforce PLA. Lyocell is obtained by dissolving native cellulose

under specific conditions in N-methyl-morpholine-N-oxide, known as the NMMO

process. The dissolved cellulose can be regenerated as a fibre out of this solution [5, 6].

Many studies have been carried out on cellulose fibre-reinforced PLA composites to

improve the mechanical properties [7-9].

Apart from the choice of reinforcing fibres in composites to improve the composite’s

mechanical properties, methods that combine the thermoplastic resins with reinforcing

fibres is also important since thermoplastic resins have very high melt viscosities, which

makes in-process melt impregnation of the reinforcement difficult. Various techniques

for combining natural fibres with thermoplastic resins have been developed to overcome

these problems. Generally, the technique giving the shortest average matrix flow

distance through the layup is preferable. Matrix sheets can be placed between layers of

fibre mats or woven cloths. Thermoplastic filaments can be filament wound together

with the reinforcing yarns by filament winding, or they can be mechanically blended

into the reinforcement yarns, to a commingled fibre/matrix yarn, also named as

commingled hybrid yarns [10-12].

Recently there has been much interest in developing unidirectional continuous fibre-

reinforced (UD) composites since they provide the best performance when the direction

of applied load is parallel to the orientation of fibre. In UD composites, it is necessary to

handle the orientation and the position of the reinforcing fibres embedded in the

composite. Hybrid yarns with oriented reinforcing fibres can be positioned parallel in

5

the fabric to form a unidirectional preform. This will result in a composite with optimal

mechanical properties in the yarn direction, which is desirable for many applications.

For such applications, low crimp unidirectional woven yarn fabrics [13] or very thin

tape-shaped UD composite preforms can be used [14].

Our previous study [13] investigated the effect of fabric weave structure on

mechanical behaviour and moisture absorption of a PLA/hemp woven fabric composite

made by compression moulding. The unidirectional woven fabric prepregs were made

from PLA and PLA/hemp wrapped-spun hybrid yarn with two different weave patterns:

8-harness satin and basket. The results showed that the composite made by satin-weave

architecture fabric possessed the highest tensile, flexural and impact strength, compared

to the composites manufactured with basket weave fabric. This improvement was

correlated to the decrease in void content and fibre misalignments.

In the present study, we have continued the development of hybrid thermoplastic

reinforcements, by including Lyocell fibres in the yarn. The main objective is to

compare the mechanical characteristics of uniaxial composites fabricated from

reinforcement made from hemp/PLA, hemp-Lyocell/PLA and Lyocell/PLA wrap spun

hybrid yarns. The hybrid yarns have been made from parallel non-twisted commingled

fibres in the yarn core, which is mechanically stabilised by a wrap-filament wound

around the core. Based on the results from our previous study, we selected a satin-

weave architecture prepreg fabric, which was processed by press heating and

consolidation into a UD composite laminate.

6

2. MATERIALS AND METHOS

2.1. Materials

Hemp fibres (genus species Cannabis Sativa L) with a medium fibre length of 30 mm

were provided by Hempage AG (Adelsdorf, Germany). Lyocell fibres with a length of

38 mm and a fineness of 1.3 dtex were supplied by Lenzing (Lenzing AG, Austria).

A thermoplastic biopolymer NatureWorks™ PLA polymer 6202D from Cargill Dow

LLC (Minnetonka, USA) was selected as the matrix component. The biopolymer was

supplied both in staple fibre and continuous filament forms by the company Trevira

GmbH (Hattersheim, Germany). The staple fibres had a fineness of 1.7 dtex and a mean

fibre length of 38 mm. The filament had a fineness of 18 tex.

2.2. Fibre processing and production of composites

The hybrid yarns were produced according to the wrap spinning process reported

earlier in our previous studies [13, 15]. The wrap spun yarn consists of parallel staple

fibres being wrapped helically by a continuous PLA filament [11]. The core is a well-

mixed strand of hemp/PLA, hemp-Lyocell/PLA or Lyocell/PLA staple.

Within the framework of this study, the hybrid yarns (nominal count of 550 tex)

consist of the following mixtures: hemp 30% + PLA 70%, Lyocell 30 mass% + PLA

70% and hemp 15% + Lyocell 15% + PLA 70%. Data from the composite from hemp

30% + PLA 70% were used from previous study [13]. In order to compare the effect of

random fibre orientation on composites on mechanical properties, data from the

nonwovens with the same compositions from previous study [16] are reported in this

article as well. As reference, webs of pure PLA fibres were made.

To produce the thermoplastic prepregs, unidirectional woven fabrics were made from

hybrid yarns (weft yarn) and PLA filaments (warp yarn) on a hand loom weaving

machine. The fabrics were woven with 8-Harness satin pattern. Fig. 1 shows the

7

unidirectional hemp/PLA fabric. The fabric specifications are presented in Table 1.

These fabrics are used as prepregs for subsequent processing to composites by

compression moulding. Prior to compression moulding, the fabrics were dried for 24 h

at 70°C and 0.9 mbar in a vacuum oven.

Compression moulding was carried out with a manual hydraulic press from Rondol

Technology Ltd., Staffordshire, UK. The pressure was set to 1.7 MPa, which was

maintained for 15 min at 195°C. Before demoulding, the composites were cooled down

to approximately room temperature. After demoulding, the composites were cut into

specimens with GCC LaserPro Spirit laser cutting machine for the different mechanical

testing experiments.

2.3. Mechanical testing

Tensile and flexural tests of the composites were performed by using a universal

H10KT testing machine equipped with a mechanical extensometer (model 100R long

travel extensometer) supplied by Tinius Olsen Ltd. (Salford, UK) based on ISO 527 and

ISO 14125. Charpy impact tests were carried out at ambient conditions on a QC-639D

mechanical impact tester (Cometech testing machines, Taichung Hsien, Taiwan)

according to the ISO 179. The morphology of fractured surfaces of the composites

obtained from the tensile testing was observed by scanning electron microscopy (SEM),

using a low-vacuum scanning electron microscopy with a Quanta 200 ESEM FEG

instrument from FEI, (Oregon, USA) with an operating voltage of 10-12.5 kV.

The density of the composites was experimentally determined by the buoyancy-

flotation method (Archimedes) in ethanol; fibre volume fraction and porosity were

calculated using density of hemp and PLA fibres [17].

The effect of water absorption on hemp, Lyocell and hemp/Lyocell fibres reinforced

hybrid composites were investigated in accordance with ASTM D570-98. Water

8

absorption tests were conducted by immersing the composite specimens in distilled

water at room temperature. The specimens were weighed regularly from 24 hours to 240

hours exposure, at an interval of 24 hours. The moisture absorption was calculated by

the weight difference. The percentage weight gain of the samples was measured at

different time intervals.

Statistical comparisons, based on a one-way analysis of variance (ANOVA) at the

95% confidence level, were performed to test the effects of different fibres on the

mechanical properties.

3. RESULTS AND DISCUSSION

3.1. Composite porosity

One of the most common manufacturing induced defects in composites is voids [18,

19]. In general, the voids have detrimental effects on the mechanical strength of

composite laminates [20], and they make them more susceptible to moisture penetration

and environmental conditions. The effect is more pronounced in compressive, bending

and inter-laminar shear loading, which are related to the matrix dominated mechanical

properties [21, 22]. In the current study, we studied the composites produced from

unidirectional satin hemp/PLA, Lyocell/PLA and Lyocell-hemp/PLA fabrics and from

nonwovens. All composites had the same reinforcement loading (30 mass-%) in order to

investigate the effect of combining different fibre types and their alignment on the

composites properties. The obtained data were compared to corresponding data for

composites made from PLA/hemp nonwoven mat and hemp/PLA fabric from a previous

study [13, 16]. The density and constituents of the composites were evaluated and are

shown in . According to the data, it is clear that the composite made by unidirectional

satin fabric gave significantly lower porosities, compared to the composites made from

nonwoven mats, and the values are in the acceptable range [15, 16]. Generally, the voids

9

are closely related to the used processing conditions and the laminate tightness. They

can also be formed due to the discontinuous resin matrix in the composites, which is

caused by the PLA fibre’s failure to form a continuous phase of matrix in the

composites or by the uneven distribution of the PLA staple fibres [23, 24]. This fact is

more evident in the case of the composites from nonwoven mats.

3.2. Water absorption characteristics

Fig. 2 illustrates the measured gravimetric water absorption for the PLA

biocomposites. Generally, the three types of fibre reinforcement compositions (hemp,

Lyocell and mixture of hemp and Lyocell) studied do not significantly affect the weight

percentage of water absorbed by the PLA biocomposite. However, it was noted that the

water uptake of the biocomposite made from nonwoven mats was significantly higher

than for the composites made from satin fabrics. This higher water absorption for

nonwoven mat reinforced PLA was expected, due to the higher void content. As shown

in , there are more voids and gaps present in the nonwoven reinforced composite. This

condition worsened the moisture uptake behaviour since voids favour moisture

absorption. The voids result in a larger amount of poorly bonded surface area between

the fibre and the matrix [19, 25].

3.3. Tensile properties

presents tensile properties of PLA composites reinforced with hemp, Lyocell and

glass fibres, as typically reported in the literature. This demonstrates that glass fibre

composites are superior to hemp fibre composites irrespective of fibre orientation, and

the tensile strength in particular is larger for glass fibre composites [26, 27]. Hence, the

table reveals the current technical performance for PLA/hemp and PLA/Lyocell

composites where the E-modulus is acceptable, but the tensile strength needs to be

somewhat improved. It has also been observed that for all composite characteristics,

10

there are clear differences between the values measured for the polymer composites

from fabric reinforcement and nonwoven mat due to the fibre orientation. Specimens

taken from nonwoven reinforcement show lower values than specimens taken from

fabric reinforcements. The strength values of hemp-PLA, Lyocell-PLA and Lyocell-

hemp PLA specimens differ noticeably. The improvement of the tensile properties of

hemp-reinforced PLA composites by addition of Lyocell fibres could be attributed to

the much higher fineness of the Lyocell fibres, which cause a larger specific bonding

surface and a better matrix-fibre adhesion [16, 28].

3.4. Flexural properties

Fig. 3 shows the flexural properties of PLA-based composites. In all diagrams, the

values of the pure PLA are presented next to the composite characteristics. The results

of the flexural strength revealed similar trends as the results of the tensile tests.

However, flexural strength values were higher than tensile strength values. This effect

has also been noticed by Mieck et al. [8] for 35 mass-% kenaf/PLA composites and by

Graupner et al. [9] for 30 mass-% Lyocell/PLA composites. Improved bending strength

values were achieved with both hemp and Lyocell fibre reinforcements. From the

results, it is evident that the highest values were achieved by the composite reinforced

by satin Lyocell/PLA fabric with 158.2 MPa followed by the satin hemp-Lyocell/PLA

fabric with 145.1 MPa and satin hemp/PLA fabric with 119 MPa.

With regard to the flexural modulus values, our measurements show a modulus of

2.91 GPa for neat PLA. Fibre reinforcement resulted in a clear and significant increase

of the modulus values. For satin Lyocell/PLA, the highest flexural modulus, a value of

9.72 GPa was found. One-Way ANOVA test showed that mean flexural strength and

flexural modulus values of materials investigated were significantly different (P-

value˂0.05); however, between satin hemp-Lyocell/PLA and satin hemp/PLA

11

composites, no significant differences in modulus were determined, but their strength

values differ clearly.

3.5. Impact resistance

Fig. 4 represents results of Charpy impact tests. It can be seen that for satin

hemp/PLA, hemp-Lyocell/PLA and Lyocell/PLA composites, the impact strength

increased by factor 1.1, 2.7 and 3.1, respectively, compared to pure PLA. Graupner et

al. [29] tested composites of nonwoven PLA/Lyocell and PLA/hemp-Lyocell and found

a similar tendency. In our test, the Charpy impact resistance of Lyocell/PLA composite

increased by factor 3.1, compared to factor 0.6 of Graupner. In fact, the higher impact

energies for satin fabric, compared with nonwoven laminate are mainly due to the

presence of a higher number of axially oriented fibres, as woven fabric composites

provide more balanced properties in the fabric plane than randomly oriented composites

[30]. The significant increase in impact strength of Lyocell/PLA composites can be

related to higher elongation at break [29]. The improvement of the impact strength of

hemp fibre-reinforced PLA was caused by the addition of 50 mass-% Lyocell to the

hemp fibres. The impact strength increased from 24.3 kJ/m2 for satin hemp/PLA fabric

composite up to 42.3 kJ/m2 for the satin hemp-Lyocell/PLA fabric composite.

3.6. Scanning electron microscopy

Fig. 5 shows the SEM micrographs of the fracture surface for the tensile specimens

for different composites. The SEM micrographs show that fibre pull-outs appear more

often with hemp than with Lyocell fibre. It can be interpreted as that the interfacial

adhesion between the Lyocell fibres and PLA is quite good. Actually, Lyocell fibres

are finer than hemp fibres being led to the larger specific surface between matrix and

fibres and consequently better adhesion between them. Thereby, composite from hemp

and Lyocell blending is a much stronger composite than the hemp/PLA composite.

12

4. CONCLUSIONS

The purpose of this study was the optimisation of the mechanical characteristics and

water absorption resistance of hemp fibre reinforced composites by mixing with Lyocell

fibres. According to the findings from our previous research, combining hemp and

Lyocell in nonwoven PLA composite can improve the mechanical properties when

compared with PLA/hemp reinforcements [16]. In the present study, we discuss

production of aligned hybrid yarns containing reinforcing (hemp and Lyocell and

hemp/Lyocell mixture) and thermoplastic (PLA) material, the technology for processing

them into unidirectional-satin woven fabrics (intermediates), and the manufacture of

UD composites by compression moulding. Satin-weave architecture was chosen

because of lower void content and fibre misalignments [13]. The obtained results

showed that the mechanical properties of the composites were significantly improved,

compared to the pure PLA matrix. Combining hemp and Lyocell in a hybrid yarn

(hemp-Lyocell/PLA hybrid yarn) can significantly improve the impact strength, flexural

and tensile strength and modulus, compared to the composite from hemp/PLA hybrid

yarn, which was manufactured in our previous study [13, 15]. In the general outline, the

best overall properties were achieved with satin Lyocell/PLA fabric composites leading

to a tensile strength of 101.2 MPa, Young’s modulus of 11.4 GPa, flexural strength of

158.2 MPa, flexural modulus of 9.7 GPa and impact strength of 47.4 kJ/m2. It has also

been observed that there are clear differences between the values measured for the

polymer composites from woven fabric reinforcement (woven from hybrid yarn) and

nonwoven mat, produced previously [16], due to the fibre orientation. Specimens taken

from nonwoven reinforcement show lower mechanical strength than specimens taken

from woven fabric.

13

Water absorption test indicated that moisture absorption decreased for composites

being manufactured from woven fabrics, compared to the nonwoven composites.

However, the studied woven fibre reinforcement compositions (hemp, Lyocell and

mixture of hemp and Lyocell) did not significantly affect the water absorption into the

composite.

References

[1] Carus M, Gahle C. Injection moulding with natural fibres. Reinforced Plastics. 2008;52(4):18,24–2,5.

[2] Hansmann H, Wismar H. Compendium Composites, Thermoplastic Resins, ASM

Handbook/extraction. FB MVU, Werkstofftechnologien/Kunststofftechnik, Germany 2003.

[3] Reinhardt M, Kaufmann J, Kausch M, Kroll L. PLA-Viscose-Composites with Continuous Fibre

Reinforcement for Structural Applications. Procedia Materials Science. 2013;2:137–43.

[4] Hans-Josef E, Andrea S-R. Manufacture and Chemical Structure of Biopolymers. Engineering

Biopolymers: Carl Hanser Verlag GmbH & Co. KG; 2011. p. 71–148.

[5] Fink HP, Weigel P, Purz HJ, Ganster J. Structure formation of regenerated cellulose materials from

NMMO-solutions. Progress in Polymer Science. 2001;26(9):1473–524.

[6] Wilkes AG. The viscose process. Regenerated Cellulose Fibres. Cambridge: Woodhead Publishing;

2001. p. 37–61.

[7] Thomas S NN, Mohan S, Francis E. Natural polymers, biopolymers, biomaterials, and their

composites, blends and IPNs. Portland: Ringgold Inc; 2012.

[8] Mieck KP RT, Hauspurg C. Correlations for the fracture work and falling weight impact properties of

thermoplastic natural/long fibre composites. Materialwiss Werkstofftech. 2000;31(2):169–74.

[9] Graupner N, Mussig J. A comparison of the mechanical characteristics of kenaf and lyocell fibre

reinforced poly and poly composites. Composites Part A. 2011;42(12):2010.

[10] Baghaei B, Skrifvars M, Berglin L, Institutionen I, University of B, School of E, et al. Manufacture

and characterisation of thermoplastic composites made from PLA/hemp co-wrapped hybrid yarn

prepregs. Composites Part A. 2013;50:93.

[11] Zhang L, Miao M. Commingled natural fibre/polypropylene wrap spun yarns for structured

thermoplastic composites. Composites Science and Technology. 2010;70(1):130–5.

[12] Madsen B, Hoffmeyer P, Lilholt H. Hemp yarn reinforced composites – II. Tensile properties.

Composites Part A. 2007;38(10):2204–15.

[13] Baghaei B, Skrifvars M, Berglin L. Characterization of thermoplastic natural fibre composites made

from woven hybrid yarn prepregs with different weave pattern. Composites Part A: Applied Science and

Manufacturing. 2015;76:154–61.

[14] Baets J, Plastria D, Ivens J, Verpoest I. Determination of the optimal flax fibre preparation for use in

unidirectional flax–epoxy composites. Journal of Reinforced Plastics and Composites. 2014;33(5):493–

502.

[15] Baghaei B, Skrifvars M, Salehi M, Bashir T, Rissanen M, Nousiainen P, et al. Novel aligned hemp

fibre reinforcement for structural biocomposites: Porosity, water absorption, mechanical performances

and viscoelastic behaviour. Composites Part A: Applied Science and Manufacturing. 2014;61:1–12.

[16] Baghaei B, Skrifvars M, Rissanen M, Ramamoorthy SK. Mechanical and thermal characterization of

compression moulded polylactic acid natural fiber composites reinforced with hemp and lyocell fibers.

Journal of Applied Polymer Science. 2014;131(15):n/a-n/a.

[17] Madsen B, Thygesen A, Lilholt H. Plant fibre composites – porosity and volumetric interaction.

Composites Science and Technology. 2007;67(7–8):1584–600.

[18] Strong AB. Fundamentals of composites manufacturing; materials, methods and applications, 2d ed.

Portland: Ringgold Inc; 2008.

[19] Frimpong S, Bowles KJ. Void effects on the interlaminar shear strength of unidirectional graphite-

fiber-reinforced composites. Journal of Composite Materials. 1992;26(10):1487–509.

[20] Jeong H. Effects of Voids on the Mechanical Strength and Ultrasonic Attenuation of Laminated

Composites. Journal of Composite Materials. 1997;31(3):276–92.

[21] Almeida S, Santacreu A. Environmental Effects in Composite Laminates with Voids. Polymer

Composites. 1995;3(3):193–204.

14

[22] Suarez JC, Molleda, F. and Güemes, A. Void Content in Carbon Fiber/Epoxy Resin Composites and

its Effects on Compressive Properties. ICCM-9. Madrid, Spain1993. p. 589–96.

[23] R Indu Shekar, Kotresh T, AS Prasad, PM Rao, T Ananthakrishnan, MN Kumar MN, et al. Hybrid

Fabrics for Structural Composites. Journal of Industrial Textiles. 2011;41(1):70–103.

[24] Sakaguchi M, Nakai A, Hamada H, Takeda N. The mechanical properties of unidirectional

thermoplastic composites manufactured by a micro-braiding technique. Composites Science and

Technology. 2000;60(5):717–22.

[25] ML Costa, S Almeida, Rezende M. The influence of porosity on the interlaminar shear strength of

carbon/epoxy and carbon/bismaleimide fabric laminates. Composites Science and Technology.

2001;61(14):2101–8.

[26] Mieck KP, Lützkendorf R, Reussmann T. Needle-Punched hybrid nonwovens of flax and ppfibers—

textile semiproducts for manufacturing of fiber composites. Polymer Composites. 1996;17(6):873–8.

[27] Oksman K. Mechanical Properties of Natural Fibre Mat Reinforced Thermoplastic. Applied

Composite Materials. 2000;7(5):403–14.

[28] Graupner N, Herrmann AS, Müssig J. Natural and man-made cellulose fibre-reinforced poly(lactic

acid) (PLA) composites: An overview about mechanical characteristics and application areas. Composites

Part A. 2009;40(6):810–21.

[29] Graupner N. Improvement of the Mechanical Properties of Biodegradable Hemp Fiber Reinforced

Poly(lactic acid) (PLA) Composites by the Admixture of Man-made Cellulose Fibers. Journal of

Composite Materials. 2009;43(6):689–702.

[30] Malhotra SK, Joseph K, Thomas S. Polymer Composites, Volume 1: Macro- And Microcomposites:

John Wiley & Sons; 2012.

[31] Gamstedt EK, Berglund LA, Peijs T. Fatigue mechanisms in unidirectional glass-fibre-reinforced

polypropylene. Composites Science and Technology. 1999;59(5):759–68.

[32] Company R. Glass reinforced PLA compounds. Reinforced Plastics. 2011;55(4):17–.

[33] Islam MS, Pickering KL, Foreman NJ. Influence of Hygrothermal Ageing on the Physico-

Mechanical Properties of Alkali Treated Industrial Hemp Fibre Reinforced Polylactic Acid Composites.

Journal of Polymers and the Environment. 2010;18(4):696–704.

[34] Sawpan MA, Pickering KL, Fernyhough A. Improvement of mechanical performance of industrial

hemp fibre reinforced polylactide biocomposites. Composites Part A. 2011;42(3):310–9.

[35] Hu R, Lim J-K. Fabrication and Mechanical Properties of Completely Biodegradable Hemp Fiber

Reinforced Polylactic Acid Composites. Journal of Composite Materials. 2007;41(13):1655–69.

15

Fig. 1 Surface view of woven satin hemp-Lyocell/PLA prepreg fabric.

Fig. 2. Apparent weight gain with time for the different PLA/hemp composites at room temperature.

0

5

10

15

20

25

0 1 2 3 4 5 6 7 8 9 10

Ap

pa

ren

t w

eigh

t g

ain

(w

t%)

Day

Neat PLA

Lyocell-hemp/PLA satin

fabrics

Lyocell/PLA satin fabrics

Hemp/PLA satin fabric

Lyocell-hemp/PLA nonwoven

Hemp/PLA nonwoven

Lyocell/PLA nonwoven

16

Fig. 3. Flexural properties of PLA/hemp composites.

Fig. 4. Impact strength of PLA/hemp composites.

0

2

4

6

8

10

12

0

20

40

60

80

100

120

140

160

E-M

od

ulu

s in

GP

a

Fle

xu

ral

stre

ngth

in

MP

a Strength

E-Modulus

05

101520253035404550

Imp

act

str

eng

th in

kJ

/m2

17

Fig. 5. SEM images of tensile fracture surfaces of (a) hemp/PLA, b) hemp-Lyocell/PLA and c) Lyocell/PLA

composites.

Hemp

Lyocell

b)

Fibre pull-out

a)

18

Fig. 5 Continued.

c)

19

he

gv

ghh

PL Table 1. Specifications of PLA/hemp woven fabric used for making composite.

Sample Weft yarn Warp

yarn

Weave

type

Yarn linear

density (tex)

Density (per

cm)

Weight

(g/m2)

warp weft warp Weft

1 Hemp/PLA hybrid

yarn

PLA

Filament

8-harness

satin

18 550 8 4 344.0

(22.9)

2 Hemp-Lyocell/PLA

hybrid yarn

PLA

Filament

8-harness

satin

18 550 8 4 496.0

(18.2)

3 Lyocell/PLA hybrid

yarn

PLA

Filament

8-harness

satin

18 550 8 4 519.0

(25.3)

Note: Data in table are mean (SD).

Table 2. Composition of the fabricated composites calculated from the density measurements.

Sample Density

(g/cm3)

Fibre mass

fraction (%)

Fibre

volume

fraction (%)

Matrix

volume

fraction (%)

Porosity

volume

fraction (%)

Hemp Lyocell

PLA 1.2490

(0.0002)

0.0 0.0 0.0 100.0 -

Satin hemp/PLA fabric 1.2978

(0.0032)

30.0 0.0 26.31 (0.07) 72.74 (0.18) 0.96 (0.25) [13]

Satin Lyocell/PLA fabric 1.2725

(0.0360)

0.0 30.0 27.02 (0.03) 71.32 (2.02) 1.66 (0.57)

Satin Lyocell-hemp/PLA

fabric

1.2960

(0.0031)

15.0 15.0 26.27 (0.062) 72.63 (0.17) 1.09 (0.23)

Hemp/PLA nonwoven 1.0914

(0.0154)

30.0 0.0 23.0 (0.6) 61.2 (0.9) 10.8 (0.2) [16]

Lyocell/PLA nonwoven 1.2765

(0.0112)

0.0 30.0 25.2 (0.2) 71.5 (0.6) 3.3 (0.9) [16]

Lyocell-hemp/PLA

nonwoven

1.1222

(0.0169)

15.0 15.0 22.4 (0.3) 76.4 (1.2) 2.1 (0.1) [16]

Note: Data in table are mean (SD).

20

Table 3. Tensile properties of hemp and Lyocell/PLA-based composites, as typically reported in

literature. For comparison, reported properties of glass fibre composites are included.

Fibre

orientation

Fibre type Fibre

loading

E-modulus

(GPa)

Tensile

strength (MPa)

Reference

Random Glass reinforced PP

composite

Vf = 0.20 5.4 77 [27]

Aligned Vf = 0.60 45.0 1020 [31]

Random Glass reinforced PLA

composites

Wf = 30 - 114 [32]

Aligned Hemp-alkali treated Wf = 30

11 85 [33]

Hemp-untreated Wf = 30 8 63 [33]

Random

(*MD)

Hemp-untreated Vf = 0.21

(Wf = 30)

5.60 53.63

[16] Random (MD) Lyocell Wf = 30 6.86 80.87

Random Lyocell-hemp mixture Wf = 30 6.31 60.63

Aligned Off-axis

angle 0˚

Hemp-untreated

Vf = 0.25 8.77 72.75

[15] 45˚ 4.62 34.75

90˚ 3.70 22.01

Aligned Hemp-alkali treated Vf = 0.26 10.27 77.08

Aligned

Hemp-untreated Wf = 30 7.5 68

[34] Hemp-alkali treated Wf = 30 8.2 71

Hemp-silane treated Wf = 30 7.9 73

Hemp-alkali/silane

treated

Wf = 30 8 75

Random Hemp-untreated Wf = 30 5.6 41

[35]

Hemp-alkali treated Wf = 30 7.6 39

Aligned Hemp-untreated Wf = 35 5.8 50 [17]

Aligned Hemp-untreated Wf = 30 10.23 88.06 This study

Aligned Lyocell Wf = 30 11.42 101.23 This study

Aligned Lyocell/hemp mixture Wf = 30 11.15 96.01 This study

*MD = fibre orientation predominantly in length direction