CA2473319A1 - Modification of flax fibers and application in composites - Google Patents

Abstract

In recent years, the interest in using natural flax fibers in biocomposites has grown because They are lightweight, combustible, non-toxic, low cost and easy to recycle. On tue other hand, lack of good interfacial adhesion and poor resistance to moisture absorption make the use of natural fiber-reinforced composites less attractive. Chemical treatment of the fiber can stop the moisture absorption process, clean the fiber surface, chemically modify the surface or inerease the surface roughness. Short flax fibers, which were derived from Saskatchewan-grown flax straws, for use in fiber-reinforced composites. were mercerized and modified by silane treatment, benzoylation and peroxide treatment. Morphological and structural changes of the fibers were investigated by using scanning electron microscopy. The temperature variation method vas applied to obtain the melting point of treated flax fibers by the DSC
technique. The moisture absorption and bundle flax fiber tensile strength were also measured.

Natural flax fiber-reinforced composites have enhanced biodegradability, are lightweight, non-corrosive, temperature resistant and low environmental pollution. These advantages place natural fiber composites among the high performance composites having economic and environmental advantages. In the field of technical utilization of plant fibers, flax fiber-reinforced composites represent one of the most important areas, But high level of moisture absorption, poor wettability and insufficient adhesion between untreated fiber and the polymer matrix may lead to debanding with age. In order to improve the above qualities, various surface treatments of fibers like silane treatment, benzoylation and peroxide treatment were carried out which may result in improved mechanical performance of fiber composites. Short flax fibers were derived from Saskatchewan-grown flax straws, for use in fiber-reinforced composites Composites consisting of high-density polyethylene (HDPE) or linear low-density polyethylene (LLDPE) or HDPE/LLDPE, chemical treated fibers and additives were prepared by an extrusion process. The test samples were prepared by rotational molding. The effects of the different chemical treatments on the mechanical properties of natural fiber-reinforced composites were evaluated. The tensile fracture surfaces of the samples were characterized by scanning electron microscopy to determine whether the modified fiber-matrix interface had improved interfacial bonding.

Description

:'~i?~:':Ie:litt.'si'. (i~~!illl?lW. ilta?lc~t'iW ?~ ij~';ii~lt:c.{~:
Fi'tures: ~, ~, '~y, L-1. '~J'~-~, ~;
3' ~ . ~. ~t r-al~G.~U ;Gel LlnSCann~.ble items re;:~.ivzd v=i~u this application .
___ _..__ ____.._.__ ~F'~'IU~'~ °r~~inal documear; in File Prep.
Section on the I(3s',~i.oor~.__..._-. _..
Docuneat~ re;:u v~;=:. cede dema>?dc ae pou~ran~ ~tre balaties (Cotnraaade: L~~ doclTm.eal~ o>-~~~ain.'u_~ dart is secric~n de pr~parauoa des do~~izr~ a~~
IOzme ~t:~~e j rNTR,OD'CJC'x)tUN' Composites, particularly natural fiber-reii~.forced plastics have been getting att~entia» since 1941 (Joseph et al. ~U~4). The interest fn natural fiber-reinforced polymer composites is growing rapidly due to their high performance in mechanical properties, significQnt processing advantages, excellent chemical resistance, low cast and low density. They haws long served many useful purpoces, but the appli:catiarz of material technology for the utilization of natural fibers as a reinforcement in a polymer matrix has taken place in recent years.
The use of biofibers derived from annually renewable resources, as reinforcjng element in therntaplastic matrix composites provides positive environmental benefts and rtiw material utilization. Recent reports indicate that cellulose-based natural fibers can very well be used as reinforcement in polymer composites, replacing (to some extent) mare expensive: and non-renewable synthetic fibers uch as glass due to the potential for xecyciahility of the matterial forms (lVtunker et al, 1 X9.8). Cellulose-based natural fiber is a~sa a potential resource far rrtakang low cast composite materials. These advantages place the natural fiber composites among the high performance composites having economic and environmental advar~cages.
A major restriction in tlae successFul use of natural fibers in durable composite applications is their high moisture absorption and poet dimensional stability (swilling) (Panigrahi et a1. 2002). Flax fibers can be incorporated in polymers to form a biodegradable matrix. But a high level of moisture absorption and insufficient adhesion between untreated fibers and the polymer matrix may lead to biacampasites having high water absorption characteristics that reduce their utility in many applications. Chemical Ireatment of the fiber can help stop the moisture absorption process, chemically modify fiber surfaces and increase the surface roughness in order to increase the interfacial adhesion between the fiber and matrix, resuttmg tn iriupraved mechanical pcrfonnance of fiber-reinforced oamgosites.
The objective of this study is to deterrnino the cffccts of pre-treated flax fibers on the mcchanioal properties of fiber-reinforced LLDPE, HDPE and LLDPEIHDPE
composites. This research work should answer the question; do chemical treatments of flax fibetv have any influence on the composite properties?
1V.I~TERIA:~S AI~rD METHODS
IVI~teriai P"reparatian Flax fibers were derived from linseed flax grown in Saskatchewan and decorticated on a standard scutehing mill at Duraf her in Canora, SK, Canada. The f bars were first washed thoroughly with 2% detergent water and dried in an air oven at 7t1°C
for Z~ 11. The dried fibers were designated as untreated fbers. Then flax fibers were subjected to sequential extraction with 1::2 mixture of ethanol and ben.~ene -for ?2 h at Sit°C; folHowed by washing with double distilled water and air drying to remove waxes and water soluble polymers prior to chemical treatments.
Reagent grade chemicals are used for fiber surface rrcadifications, r~amsly, odium hydroxide (NaOI-~), benzoyl chloride, ethanol, dicumyl peroxide, acetone, alcohol and the coupling agent, triethaxy vinyl silane (Aldrich Chemical Co. Ltd;), In this series o.f experiments, high-density polyethylene, linear low~density polyethylene (HI?PE x7b1.27 and LLDPE 8460.29, Exxon Mobil, Torovto, ON) and LLDPFIHDPE

(NOVA Chemicals Ltd,, Calgary, AB) were used as major matrix materials for reinforcement.

Experimental Design and Aada Analysis Methnds The experimental dosign is a factorial arrangement of treatments conducted in a randomized fashion. Table 1 shows the outline of the experimental design fdr one xype of treated fiber. The same design was used ,for the other chemically treated fibers.
Taiile 1. Experimental design for one type of treated fiber.
Filler Plastic Test ~DPE

$enxoylation treated fibers ' LLI Tensile Strength JpE

( 10% perceatogc of fiber -r at Yield S~Yoparceatnge oteoupling ~,r,;Y]~'~jDPE
agent) l~Iain Treatments: 'r'~ (Silane treatment); Tz (Betuoylatian);
T3 (peroxide treatment); T~ (Untreated) = ~
Sub-treatments; S, (HDPE); S2 (LL,I~I'E); 53 (LLDPE/~DI'E) - 3 Treatments eombirtation = 4 types of fiber X ~ types ofplastiG = 12 ~Tis'i~ ~1s2~ TlS3y ~~51~ T2~2~ T3~3s T3~1~ T'352, T3s3r T4~1~ ,~4~~-'2r T4°~-3 Five samples will 6e tested for each sample, according to the appropriate ASTIVI standard.
Tensile Test = I 2 X S replicates = 60.
Fiber Surface Treatment Generally, the first step is the mercerization process (pre-treatment process.) for all of fiber surface treatments which causes changes in the crystal structure of cellulose. Then the different chemicals can be used ot1 the fibers surface in order to improve the interfsae properties.

Shane treatment: Fibers v~ere pre-tread with 5% NaOI-1 for about half an hour in order to activate the OI~I groups of the cellulose and lignin in the fiber. Fibers were then washed many times in distilled water and finally dried.
The pre-treated fibers were dipped in alcohol water mixture (6~0:4a) containing tri ethoxy vinyl silane coupling agent. The pH of the solution was maintained between 3.5 and ~, using the METREPATC Phydrion buffers and pH indicator strips. Fibers were washed in double distilled water and dried in the oven.
Henxoylatlon: An amount of pre-treated fiber were soaked in 18% Na4H solution for half an hour, filtered and washed with water. ~fhe treated fibers were suspended in I
O°!a l~aG~H solution and agitated with benzoyl chloride. The mixture eras kept for i S min, lyItertad, washed thoroughly with water and dried between filter papers: The isolater3 fibers were thcx~ spr~.lced in ethanol for 1 h to remove the umrcatcd bcnzoyl chloride and finally were washed with water and dried.
Peroxide treatment: Fibers were coated with dicumyl peroxide from acetone solution after alkali pre-treatments. Saturated solution of the peroxide in acetone was used.
High terryperature is favored for decomposition of the peroxide.
Composite Preparation Pre-treated and untreated fibers were ground by the ~rindin~ mill (Falling Number, Huddinge, Sweden) and oven dried aI ~0°C for 24 h to reduce the moisture content to less than 2°J°. Mixtures of thexrn.oplt~stic powder and 10% by weight of flax ~bcrs were pre-dist~~ibuted by using a food t,Iender {blaring Products Corporation, New Yorlc, N~. This is done tv aid in homogeneous mixing of fibers and thermoplastics during extrusion process. The silane-coupling agents were also be added in a proportion of ~% by weight as "resin additive".
The blend was fed to the laboratory mixing extruder (LMT} (Dynisco, Franklin, IvIA) using a bane! to die temperature pra~le of 1?S°C with a screw speed of 1~0 rpm. Blends prepared in this manner were extruded using a strand die. Extruded strands were then pelletized. The pellets were ground using a grinding mill (Retsch GmbH 5657 hIAlIN, t~J'est ~rermany) attd the ground product was used in rotational rnoldin~. The powder of ~bc:rlthermoplastic was obtziined by extrusion, pelletizing and grinding of this blend.
Bincomposites Manu~'acturing lay ftotation.al Molding Dog-bone shaped test samples were prepared using a rotational molding machine (Parkland Plastics in Saskatoon. SIC, Canada). It is a carousel-typo molding rnochine with fow separate ass that can cash rotate ax two separate axes; while completely closed in an oven at DSO °C for 30 min.
Microstructure of Fiberreitiforced 'Composites As a supplementary tool, the microstructure of the modified fiber-polyzrfer rnatri.K interface was examined using a ~ scanning electron microscope (SFN1505 FhiIips, Holland) at the accelerating voltage of 30 KV. The sample surfaces wet~e vacuum coated with a thin layer of gold on the surface of interest using a Sputter Coatex 51508 (Edwards, USA) to provide electrical conductivity and did not significantly affect the resolution.
Scanning electron .micragraph of fiber-rcinfotecd composites may show the interfacial bonding between flax Fber and polymer matrix to indicate the extent of i'~ber-matrix adhesion.
S

Tensile Test Specimens were conditioned for 7 days at 23°C and 50% relative humid;ty prior to performing tensile tests. Composites hrving 10% fiber by weight loading were p~epar~ed and properties wore evaluated by mechanical tests. The appropriate ASTM methods wexe Followed, and at Least five replicate specimens were tested far each property and the results were presented as average of tested specimens. The test conducted eat standard laboratory atmosphere of 23 °C
and SOQ1° relative humidity.
An Instron Universal testing maehinc (SATEC Systems, Inc., Crrove City, Pty;) was used to perform the tensile strength test at a crosshead speed of ~ mrnlmin as described in ASTIvI
procedure Dfi38-99 (ASTM, 1999), and each test was performed until tensile failure occurred.
Each sample was manufactured by wc~y of rntational rnoldfng, and the familiar dog-bone shape was ucili~ed 1-n the testing procedure.
dig I. Tensfte test rising Inctron testing machine 'V~ater Absorption of Composites Water absorption characteristics of composites arc altered by the addition of additives such as flax fiber fillers because these additives show a greater aftnity to watex.
The samples were dried in an oven at 70°C for 24 h and nrrmediately weighed. In order to measure the moisture absorption of composites, all samples were immersed in water for abort 24 h xt room temperature as described in ASTM procedure D~70-~9 (ASTIvi, t 9~9).
Excess water on the surface of the samples was re~no~ed before weighing. The moisture absarption was calculated according to the following:
Wet wcig~c -- Initial wei~xt increasein weight (percent) = x 100 Initial weight RESLY~'~''S Alvl~D Dh~GIUSS~U~1 Composite Microstructure ribcr-matrix interface plays an important role in eampasite properties. .~
strong fiber-matrix interface band is critical far high mechanical properties- of corrtposites. Figure 2-5 show the SEM photographs of fiber-matrix interae~ipn studies of uzttreated and surface treated flax fiber-reinforced composites. Scanning electron micrographs of the tensile fracture of treated composites revealed the failure mechanisms. Fiber breakage was the main failure criteria observed. Untreated flax composite indicated that theca was very pour adhesion between fiber and matrix. While chemically treated flax composites showed better fiber-matrix interaction as observed from the gflod dispersion of fibers in the matrix system, thereby predicting micropores at the interface.

Fig 2. SEfvI rrzicrogragh of LLDPE with 10% Fig 3. SEM micrograph of L~,D~"E
with untrcatcd flax in eompos9ces 1U% Shane treated flax in com~asites Fig 4. SEM tnicrograph of LLDPE ~i#h x0% Fig 5. SEM microgranh of LLDP)~ writh Bettxoylation treated flax in composites I0% Prcrxide treated flax in composites Tensile Test Tensile testing was performed with varying methods of Chemical treatment iaa order to develop a sense of horv the chemical treatments of flax fiber affected the tensile strength. Figure 6 describes the tensile strength at yield of fiber-reinforced Lr:I7PE, HD~'E
and LLl3PElHDPE
composites. Compared to the untreated fiber-based composite having IO% by weight fiber loading, tensile strength at yield was improved by these treatments. This is probably ~tue to the increased fiber-matrix adhesion. The variation in tensile properties could be expE~ined on the basis of the changes in chemical interactions at fiber-matrix interface on various treetrnents. The tensile stren~h of flax fiber-reinforced composites is determined both by the tensile strength of the fiber attd by the presence of weak lateral fiber bonds (Mohanty et al.
20fl.1). The variations in the tensile strength at yield of the composites on different modifie$tion were attributed to the changes in tha chemical structure and bondahility of tho fiber. Also tho tensile properties of natural fber-reinforced plastic corn~osites could be improved by the use of silazie cot~plin~
agent.
_ ~-~ ~~ h~
"rt ~ ~ N G O n1 ~! [~.. ~ M 01 h M ,.., h~ 00 D0. ~..' O~..C ~ r h ~' ~i3 .-. ~p va ,,ti ~p aD r' b rr Q?
-1a b w A
D untreated ~ s~azte treatment Cl Benroyl~,tivn Cl Peroxide txeatmerlt Ftg G. Comparison of tensile stt~ength ~t yield of 10% fiber with different t.hermoptast'ies.

LLD.PE HDPE LLI7PEIHD~'E

Water Absarptian of Compasrtcs During chemical treatment of the flax fibar, the hcrrlicellulase and lignin were separated and cellulose was used for the biocomposite. Figure 7 shows the moisture absorption of composites at the room temperature. The moisture absorption of the chemically treated flax fiber-based composites was lower than that of the untreated fiber-based composites. S~trnng intermolecular fiber-matrix bonding decreases the rate of moisture absorption in biocornpositc. h shaves that chemical treatments of flax fiber can decrease the water absorption of the biocomposites.
0.2 ~, N
0.15 ~ ~ _, ~ ,~ °
m ~cJ' o~..e O_ M _ ~ ~ ; O
G ga~~°° '~.ooQ° 'd'O
W oo ; y sn ~ ~ ~ r, . a a o. x °:~. ~ ~ °. ,af h~~~ .0 p R~ ~ Y ~ t I , ~J i a.os ~ ~ ;,~~~ : 2~z k~b~ ,~~Gi h ~':
-~H x. 4 ,;.
~~LDPE HaPE LLDP~JHDPE
17 untreated ~ silane treatment ~ Benza~~tic~n 0 Peroxicd~ treat~x~ent Fig 7. Moistarc absorption of cosupositcs.

CONCJL~JSiCfNS
Fiber-matrix interface plays an important role in composite properties. The ability to control the chemical arid mechanical properties of the fiber-matrix interface is crucial.
Morphological studies showed that the chemical treatments improved the fber-m~tr'sx adhesion and the dispersion of the particles.
Compared zo the untreated fiber-based composite; tensile properties rwere improved with a suitable fiber surface treatment. Silane; Benzoylation, and peroxide treated fiber-based composites offered superior physical and mechanical properties. lVfechartical properties of natural fiber-reinforced plastic composites could be improved by the uce of silane eouplin~
agent.
The hydrophilic nature of biofibers leads to biocomposites having high water absorption characteristics that can be overcome by treating these fibers with suitable chemicals to decrease the hydroxyl groups of the fibers. The water absorp~tior~ and swelling of the treated flt,~ fiber composites is lower than that of composites based on untreated flax fibers.
In the present study, chemically treated flax fiber is used as a supplement to plasiics, and reinforcement in thernioplastic matrix in rotational molding process. The flax fiber is already being produced and can be obtained a relatively low cost compared to glass fiber reinforcements. .
Thus, natural flber~reinforced bit~degradable matrix composites (biocornposites} will get morc attention in the future.
lI

The authors would like to ackna~wledge the Department of Civil Engineering and biology at the University of Saskatchewan for the use of their facilities arid equipment.
Financial support of this study was given by Saskatchewan Flax Development Commission and the Agriculture bevelopment 1~und of Saskatchewan Agriculture, Food and Rural Rcvitali2ation.
The support of Parkland Plastics and the University of Saskatchewan is also acknowledged.
11~F~~tENCES
ASTM standard B 256-97, 1997. Sca~ard Test Methods. for l~eterminittg the Izad Peztdulum Impact Resistance of Plastics. 1997 Annual $ook of ASfiM Standards, Textiles.
7(1 ): l-20 ASTNI standard D 63~-99, 1999. Standard Test l~fethad for Tensile properties of Plastics. T~99 Annual Book of ASTM Standards, ~"eattiles. 7(1):46-S8 Joseph; K., L:I~.C. Mattoso, R..1:). Toledo, S.Thomas,.L.li_ de.Carvalho, L.flthen, S. Kala and B.
Tames. 200. Natural fiber rei»fareerl therrnoplasfic Composites. Natural Polymors and Agrofibers Composites 159:01 lvIohanty; A.I~., M. Ir~iisra and 1,.T. Drzal. 2001. Surface rnodftcation of natural fibers arid performance of the rssul'rig liiocoxnposites: An Qvetview. Composite ~";nter~'ac~ex 8 (5): 313-343.
lvluttker, M., R Holtrnann and lxf. Michaeli. I99:S:, Improvement of the fiberlmatrix:-adhesion of natural fiber reinforced polymers. Proceeding of she d3r~ Inror»crrianal' SAtYI~.h' s~mpc~sium, Iviay 31-yune 4.
fanigrahi, S., L.G. Tabil, W:f. Crerar .and S.Sokansanj. X002. Application of Saskatchewan grown flax fiber in rotational molding of polymer composites: Paper N~o. CS~E
02-302.
Saskatoon, SK: Canadian Society of agricultural Engineering.

IN'FRQD~CTIOiV
Flax fibers ire abundantly available in Canada, well known far their low cost and high strength characteristics. The interest in using natural flax fibers as reinforcement in hiacomposites has increased dramaticaliy and also represents one ofthe most important uses: Hut cellulasic fibers are hygrosCapic in nature; tnaisture absorption can result in swelling of ihc fibers which rnay lead to micro-~crackiag crf the composite and degradation of mechanical properties. This problem can be overcome by treating these Fbers with suitable c~emicats to decrease the hydroxyl groups which tray be involved in the hydrogen banding within the cellulose molecules. Chemical treatments may activate these groups or can introduce new moieties that can effectively interlock with the matrix. A number of fiber surface treatments Iihe silana treatment, benr_oylation and pero~idc treatment were carried out which may result in innproved mechanical performance of the f fiber and cornpasite. By Irmiting the substitution reaction on the surface of the fibers, good mechanical properties were obtained and e~ degree of biodegradability was maintained (Scandola et al. 2000). Research on a cast effective modification of natural fibers is a necessity since the main attraction for today's ~ market of biacomposites is the competitive cost of biofibers.
The possibility of fortlting mechanical and chemical bonding at the fiber surface is mainly dependent on the surface morphology and chemical composition of the f begs (Edwards et al.
200I). Therefore, the microscopic analysis of fiber surface topology arid morphology is of utmost importance in fibrous composites. The structural and chemical changes that have occurred on the surfaces of flax fibers upon treatment were characterized by scanning electron raicroscopy (BEM).
rlax fibers are incorporated in thermoplastic materials to form a biodegradable matrix in »,~"~p~"_~~,wu..N._.~w..~_..__ __ _.._..__ _._.
_____~.m,~..,~~~~,.~~"~.:~,,",.;,-.~.~.~,~..~~~~_.___.~____.u.
_;.

order to impart desired characteristics such as increased strength into the composite. Mechanical performance of a fiber..reanforced composite is mainly determined by the properties pf fiber. 1t is very important to know the strength of the fibers before being combined into the thermoplastic matri-x to better understand how the anal composite product behaves, While much effort has been denoted to studding the mechanical behavior of the materials, a comparable understanding of the thermal behavior is lacking: 'I'o determine the rnoictng point of treated flax fibers, a di~'erentia.l scanning calorimeter (13SC) is the most widely used of al!
therrno-analytical techniques and ideal for research and quality control applications.
The present study investigvtes the different modification methods of flax fib8~rs and their effect on the mechanical performance and thermal behayiar of the fiber. The specific e~bjective of this work is to apply the SEM to analyze the surface morphology of flax fibers upon trcatrr~cnt;
to apply the temperature variation ~ncthod tv obtain the melting temperature of treated ~Iax fibers; to use the Instron 1 Oll testing machine to measure the fiber bundle tensile stxep~th and to use the environmental test chamber to compare the moisture absorption of treated and untreated fibers.
___ _.._____.._____ .... _ .. w..d...~,~"vr~~... __.____...__~. .......:....
.. ~~...._.~,____.__ _~_...~...._.. _ _.

MA'Tl~~t.IALS AID METHf,~DS
Material Preparation Flax fibers were derived from linseed flax gown in Saskatchewan and decaxt~cated an a standard seutching mill at Durafibre in Canora, Saskatchewan, Canada. The fibers, were First washed thoroughly with 2% detergent water and dried in an air oven ax ?0°C for 2~ h. The dried fibers were designated as untreated fibers. Then flax fibers were $ub,~ected to sequential extraction with 1:2 mixture of ethanol and benzene for 72 h at 5~°C, followed by washing with double distilled water and air drying to remove waxes and water soluble polymevs prior to chemical treatments. Reagent ~rat3e chemicals were used for Iiber surfacE
madificatiorts> namely, sodium hydroxide (NaOH), benzoyl chloride, ethanol, di.cumyl peroxide, acetone, arid alcohol.
The structure of silane coupling agents, triethoxy vinyl silane (Aldrich Chemical Co. Ltd.) is shown in Figure 1, ft,~,~0 ~~zh's Si Cf~'a HfCsO
~'ig.1. Structure of silane entapling agents (Tri ethoxy vinyl silane)»
Finer Suri'acc'~'Sreatment Generally, the first step is the mercerization pracess {pre-treated process) for all of the fiher surface treatments. Mercerization causes the changes in the crystal structure of cellulose and then the different chemicals can be used on the fibers surface in order to improve the interfacial properties, SiIar~e treatment: Fibers were pre-treated with 5°fo NaOH for about half an hour in order to activate tho DH groups of the cellulose and lignin in the fiber. Fibers were then wa:~hed many times in distilled water and finaliy dried.
The pre-treated fibers were Clipped in alcohol water mixture (60:40) containing tri ethoxy vinyl silane coupling agent. The p~ of the solution was maintained between 3.5 and ~1, using the aVIETREPAK Phydrion buffers and pH indicator strips. Fibers were washed ixs dou~~lc distilled water and dried in the oven.
t'llkoxy silanes are able to form bonds with hydroxyl groups. Silanes urtdergb hydrolysis;
condensation and the bond formation stage. Silanols can form polysiloxane structures by reaction with hydroxyl group of the fibers (Sreekala et al. 2000). The possible reactions are shown in Figures 2 and 3. In presence of moisture, hydrolysable all:oxy group le~.ds to the formation of silanals.
~ c~~$
Chip=CSI-Si-(~C2H5 ~Z-~-->Cki2=C1:-i-S-~0-hI
OCxHs O-H
Fig. 2. Hydrolysis of silatte (Sreekala et al. 2000).

O-H
Cel lulose -- _ _ _ ..p,~
Fibcrs_ _' _ _ _ He~cellulose -O~ H + C~-IZ=CH- ~i-O-,~I ---~'--~~ Lignin--_____ _p,,~Lr H O-H

Cellulose------4--~ ~i-Ci-I=CHz , o_l~
.,. O_H
Tiber-----Hcmicellulose -O- Si-CH~CHZ
', O-H

Lignin-_____ .. p-... ~i_CH==CH, ~'ig. 3. Hypothetical reacti4n of fibers and silane (Srcekata et al. 2440).
I~enzoyl~tion: An amount of pre-treated fiber were soaked in 1E°/Q
NaOT~ solution for half an hour; fltered and washed with watc~-. Tho treated fibers were suspended in 10°/n IVaOH solution snd agltatcd with ber~zoyI chloride. The mixture was kept for 15 min, filtered, washed tharou~hly with water and dried between otter papers. The isolated fibers were then soaked in ethanol for 1 h to remove the untreated benzoyl chloride and finally were washed with water and dried.
The reaction between the cetlulosic -OI-I group of fiber and benzoyl chloride is shown in Figure 4.

Fibre - OH -r NaOH -----~" Fibre - O'Na~ -~- H2O
O
Fibre- O'1~'a'~ + C1C ~ / ---=- Fibre- O - C ~ ~ +hJaCl Fig 4. A possible reaction between eelludnsic-UH groups and bcnzoyl ehxorfde (Joseph et al. 2000).
Peroxide treatment: Fibers were coated with dieumyl peroxide from acetone solution after alkali pre-treatments. Saturated solution of the peroxide in acetone was used.
High temperature is favored for derompositiort of the peroxade.
The decomposition of the peroxide and the subsequent reaction at the interface is expected at the time of curing of composites. This is shown in Fa~ure 5.
RO - OR ~ 2R0.
RO. + Cellulose - H ---~ R ° OH + Gellulase.
Fig S. Peroxide treatment rea~efion (~ree~ala et al. 2000j.
DSC ~xpet~irnenta~ Procedure DSC is a thermoanalytical technique in which heat flow is measured as a Function of temperature or tune. DSC is useful in charaeterizin~ thermal properties of raw materials, mixtures of materials or finished products and also provides information quickly and easily on a minimum amount of sample.
One treated fiber sample of 7 to 10 mg was placed in an aluminum sample pan and sealed with the crucible SCalil7$ ~7TCS5; thl5 WiIS placed lnLO Itle DSc ~~I~ (TA
Instr'uTr~L'ntS, InC.
Newcastle; DE}. '.Che sample was heated at a controlled rate (5°Clrnin) from room temperature to .
X00°C and a plat of heat flow versus temperature was produced. The resulting therma~rarn was then analyzed. For each sample at least three experiments were performed and the results averaged the melting point evaluated.
Fiber Surface lWlorphology As a supplementary tool, the microscopic examination of modified and untreated fibers' surfaces were carried out by using the scanning electron microscope (SEM505 Philip, Holland) at an accelerating voltage of IU I~~': The sample surfaces were vacuum coated with ~ thin layer of gold on the surface using a sputter caster SI~aB (Edwaxds, USA} to provide electrical conductivity arid did not significantly affect the resolution.
Fiber Bundle Tensile Teat Flax fiber bundle tensile strength tests were preforrned by using a computerwcontrolled Instron Model 1011 (Instron Corporation, Cantr~n, 1VIA} with a gauge length of 40 mm and a crosshead speed of 5 mmlmin (Figure 6). For every set of chemical treatment, a minimum of fifty specimens were tested for determining the fiber tensile strength. The tests were cc~nduoted at standard laboratory atmosphere of 23°C and 51 % relative hurr~idity, The Tnstron was set up to display a force displacement-loading curve and to read the load at maximum or the break point. Looking at the shape of this curve is a method to check the accuracy o~ each individual test. yf the sample was tensioned unevenly, more than one peak will be seen on the graph. If the fiber have slipped between the clamp and the grip fixture, then no distinct break point will be seen on the graph and the test has been compromised ('Ward et al.
2002).

Fig. b ~'ibcr bundle teasile strength test.
Unit Break.is calculated bar the following equation:
UB=FID........ ............,....................................... ..(l) where. F= maximum breaking load {mN) D= linear density or tax (mgjrn) t113= unit break (mN'/tex) The average unit beak of flax fiber bundle was estimated by use of this fiber bundle tensile test. Such as assuming the breaking point is located at the same spot.
This testing procedure uses linear density to in effect measure the aroa.
Moisture Absorption friar to testing, the fibers were dried in an oven at 70°G for 2~ h.
fiach sample gas planed in a conditioning device for ~Z lt. Conditioning was conducted in the extviroxunental test ahambcr (Anglelantoni, ACS, IVIassa Niartar~a, Italy) at 23°C and relative humidity values af' 33, 66 and 100%, respectively The weight of fibers was measured at dif~'erent time intervals and the moisture absorption was calculated by the weight difFerence.
After weighing an ara analytical- balance, the moisture absorption was calculated according to equation 2.
Increase in weight (percent) = M' r ~' x 100 ....... ... . .. .... . ... . ..
. .. .. . . . . . .. . .. .. (2) Md whew= Mr-- mass of tl~e sample after conditioning (g) (wet weight) M'o= mass of the sample before conditioning (g).(dry weight 1 __ ~x~su»TS alvn n~scussxr~N
physical Modifieatioas:1~'ibers Surface lVlorphoiogy .
Scanning electron miexoscopic analysis examined the surface topology of un#reated and treated fibers. It is important to mention that the changes of surface topo;~raphy affect the interfacial adhesion. Aporaus structure is observed for untreated fibers.
There is saong evidence ~chac phy$ica3 microstructure changes occurred at the fiber surface.
It can be clearly observed that silane treatment gave surface coating to the fibers; drtd surface features of f begs wexe not clearly visible. Since the flax fibers exhibited micropores on its surface, the coupling agent penetrated iota the pores and formed a rnechanicall~r interlocked coating on its surface. Benzoylatiarl treatment led to major changes on the fiber surface. Smooth fiber surface is seen due to the mass like substances deposited on the 'surface of the fiber. The surface topography is entirely modified after dicumyt peroxide treatment. The fibrall~r structure of the individual ultimate fibers is revealed from the photograph and may be duE to tho leaching out of waxes and pectic substances. Ivlicropores; particles adhering to the surface; groove like portions and protruding structures made the fiber surface very rough.

Thermal.A:nalysis: lVlelti~g Point The rncteriaIs tested were assumed to be thexmaliy hamageneous. The heat flaw shawed the rise of temperature as fttnctian ref time. Analysis of the DSC thermogram showed that the melting range of the flax i'ibers is displayed as an endothermic peak. The primary disadvantal;e is fiber's low melting point. Comparing the melting paint of untr~cated and treated flax. ~tbers, it is easy to observe that the latter had higher melting point as shown in Figure 8, Atl increased melting point mar lead to improved thermal properties of flax fiber-reinforced composites.
- ~ ~ ~ ~ r ,.t-.,.: .,.",.m.,. ... .d .. ....t 's:
~.: .
.1 ..,. ~., ~..
y .. .
anas~tsd . ' iilepe ireas~nont ' ~ ~~uaa'~io~ tresime~ ' neto~dd~ crasanares .
Ftg. 8. Melting Point of uatrcated and treated flax fibers.
Fiber $undtc Tensile Strength Test Fiber bundle strength was tested and the results are shown in Figure 9 with a gauge length of 40rnm at standard laboratory atmosphere of 23°C and 51 %, relative humidity. Data shown are averages based on fifty tests. The data show chat the higher strength of silanc and peroxide treated fibers rnay he a result of the removal of surface imperfections after the treatment. The increased uniformity of the fibers would Sive an increase to strength; its points of unconformity are rernowed during the treatment and this changes the deformation behavior of~tlle fibers. On the other hand, benzoylatinn could not give a high unit break of flax fiber trundle due to the brealca~e of the bond structure.
,m\~.~:. ~ .t mar;sY _ ,.:.ls. ..~'t~fk~:;~ t .l ,- I <t .. . ' ~~' ::r Y~fr~~. ..
::i.,, ~'~.f ' .' .., rvLy : .. ...,..
' S . .:L.:.
!Y+ ,. ....
.~
~ri~m r~ fir- ~5 ~'~ r ~.r. ,y~ x ' ~.a ~ rr ~ r!~ , j.~~ ~n~f ..~:,;' ', ti,a~..r~lit~ t 1 tt ~.tx;h 3.y., ~~'~~~~~~. J ~~r ~~' ar;.
~itrtt~'iS~ ~if ~.r:: 4, Sx f1~3~.~1~'a~.rr,.~Kn"i\Vr ,.. S.. ~f!~~ 4V1 .Y. ~' b~....i - vY. r "~
t f ~ v.: J t ..' '.
> rt: t ' ; i ' ' Jt j f ,22,1:0T, ~~ ~y ~ ~ ' ' ~ .
< < t..
~f~~ a.~ a ~ 3a . - ~
x' . ~ ~T~ . ~ ,. ''~
.. ~ ,.. .
;a ~
..
,~ °* ~ .
~..w.f. .-~.~a_~.,. ~.., ~ .,'J,N..w . ,,:-,. ~. r ~ c~tl~w~iller(tni'~lsos? .
~< . .
. ~i~ated~..: :#~,~~ra#t~t~c, . ' kae~rci~otiaaxraa~ussnc~.v-:.poru~Idraeoiniano'.. -.... . :x . .
.. .,. :. .... ; :,.. ; w ~ ~. y ,°
Fig. 9 Average ~~it break of flax fiber bundle;
lVroisture Absorption During chemical treatment of l~he nax fiber; the htxrliccllutase and lignin lucre separated and cellutose was used for the biocornposate. Before making the composite, the moisture absorption of flax fibers should be reduced. JFigure 10 shows the moisture absarption of untreated and treated flax fibers at different relative I~umidity: The rnoisturc absorption of the chcrnically treated flax fiber is lower than that of untreated fla ~c fihers.
It shows thist chemical treatments can decrease the moisture ~bsorptxon of the fibers and we hope it may also lead to the biocornposites having low moisture absorption characteristics.
r:
G
d ~ ~ ;, Q ~ ~ ~y ~~ ~ ~i rn ' .O ~ ~ :. '.
c? o0 O ~, .3 - r '~7 t ~.t v ~ ~ i SV ~L,,t y.!~Y~ P'~~i, . . 'r ~ ~'.'.' k t.°~ I
p Untreated ~ Shane ~r~atment p ~enzoy~tinn p Peroxide treatment Fig.10 Moisture absorption of untreated at~d treatedf flax fibers.
~3 66 1~0 Morphological and structural changes of the fibers were investigated by using scanning electron microscopy. The coupling agents were found to be effective to improve the surface properties of flax fiber and form a mechanically interlocked coating axi its surface. 'fherefi~re, physical microstructure changes occurred to the fiber surface by chemical treatment.
A validation of the method was made on the samples witI~ dii~'erentiat scanning;
calorimetry (DSC). The results obtained in this work show that the treated flax filers have higher melting point than the untreated one.
Silane and peroxide treatment on flax fiber bundle Iead to a higher tensile strength than that of the untreated fiber bundle. Comparatively lower tensile strength is observed in benxoylation treated fibers.
Ftax fiber is highly hydrophilic due to the presence of hydroxyl groups from ~celluiase and lignin. Chemical treatment can reduce the hydrophi.lieity of the fiber by treating these fibers with suitable chemicals td decrease the hydroxyl groups iry the ixbers.
From the results of these experiments, it is quite evident that flax fiber has a very promising future and aan be used ,as a substituzc for glass fibers. Surface modifications of hydrophilic natural fibers have achieved some degree of success in rraaking a superior interface, mechanical properties and thermal properties, but lower cost surface modif cation~
needs to be emphasized for biocomposites to replace glass fiber corr~posites in many applicatiorts in the future. ?~latural fiber-reinforced composites should be developed and characterised so as to produce cost~competifive biaeomposites for industrial applications.
ACKNOWIr~AGIVX~NTS
The authors would like to acknowledge the Department of Biology at the University of Saskatchewan for the use of its facilities and equipment. Financial support of this study was Fund of Saskatchewan. Agriculture, Food end Rural Revitalization.
,~2ElF'ERENr~ES
Edwards, H.G.M., D.VtI. Farwell and l~. Webster. 1997. AFT Roman microseop~ of untreated natural plant fibers. .S~ectrnclrimic~a Actu, Pare A: Molecular and h'iomoleculur Spectrvscapy S3A (13): 2383-2392.
Joseph; lC., L.H.C. lvtattosa, R.T~. Toledo; S:Thomas, L:I-I. de Carvalho, L.Plthen, S. Kola and B. .lames. 2000.1'~latW al fiber re~nf~reed thermoplastic Composites. Natural Polymers and Agro~hers Composites. 159:201 Scandala, M.., G Frisoni, M. Baiar~o. 2000. Chemically modified cellulosic reinfoFCements.
2a9thACSNcttional Meeting, San F'raneisco, CA, lvlareh 26-30.
Sreekala, M.S., Ivf.('~. Kumaran, S. Joseph and M. ,Tacob. 200Q. C7il palm fibers reinft~reed phenol formaldehyde composites: I~tluence of fibers surface modifications on the mechanical performance. Applied composite materials ?:295-329.
Ward, 3.. L.fs.Ta'bii, $. Panigeahi, 't~J'.J.Crerar, T. Powell, A.:T.Kovaes, Alvin Ulrich, 2402.
Tensile testing of flax fibers: Presented' at the ~Sl9ElCSAE ~lTorth Central Inner-sectiuncal Conference, Saskatoon, SK; Se~ptam6er 27-2$, 2002.

1 All publications mentioned inthis specification are indicative of the level of skin in the 2 art of this invention. Alt publications are herein incorporated by reference to the same s xtent as 3 if each publication was specifically and individually indicated to be incorporated by refer ence.
4 The terms and expressions used are, unless otherwise defined herein, used as terms of d.acription and not limitation. There is no intention, in using such terms and expressions, of excluii ing 6 eduivalents of the features illustrated and described.

Claims

1. A method of preparing a natural fiber-reinforced plastic comprising:
chemically modifying natural fibers, said chemical modification being selected from the group consisting of silane treatment, benzoylation and peroxide treatment;
grinding the modified fibers;
mixing the modified fibers with a thermoplastic powder;
extruding blended fibers and powder in strands; and pelletizing the strands.