CA2477564A1 - Process to manufacture greener thermosetting composites of pre-shaped structure - Google Patents

Abstract

A process to prepare structural composites for automotive, aerospace, furniture and sports goods applications whereby composites are manufactured by resin transfer molding process with resins in liquid form either modified or unmodified were injected under pressure into a cavity of a preshaped design containing fibres in nonwoven, loose or mat form whereby fibres obtained mainly from natural origin but in few cases they are combined with synthetic fibres to a lesser extent are layered in the preshaped mold in single or multiple layers whereby the fibres.
The molded products exhibit excellent mechanical strength and durability suitable for applications in automotive exterior, interior parts, aerospace parts, furniture and sports goods applications.

Description

Process to manufacture Natural Fiber Reinforced Composites by Resin Transfer Molding Author: Mohini M. SAIN
44 Ovida Avenue, Toronto, ON, M9B 1 E4 Canada Invention In this work hemp fiber reinforced epoxy vinyl ester resin composites were manufactured using a Resin Transfer Molding (RTM) process. RTM composites with fiber contents, up to 40 % by volume, were manufactured. The wetting of the fibers was very good. The resin injection time was observed to increase dramatically at high fiber contents due to the low permeability of the mat. Surafec treatment of fibres provided improved moisture resistant properties and also enhanced composite properties.
Loose fibre, mats and woven as well as nowoven natural fibres were used with different design and construction. Examples of woven fabrics are jute, cotton and glass.
Examples of nonwovens are hemp, flax, hemp-polyester, flax-polyester, typical loose fibres were hemp, flax, soy, wheat, cotton, com. Natural fibres were used either alone in combination with a thermosetting resin or a hybrid system where glass fibre and /or glass mat was used in addition to the natural fibres substrates. Natural f fibres and their substartes any other form used were either untreated or surface treated with any functional chemicals.
Typical examples of functional treatment chemicals were malefic coplymers, maleimide coplymer, aryl- maleimide, their quaternary salts with varialble chrage density and aryl or malefic content, surface chemicals were also used with variable molecular weight. Other surface chemicals used were alkyl imines, and their combinations with rosin ester and rosin derivatives. Keeping a constant mold temperature is the key to obtain fast and homogeneous curing of the part. The experimental procedure designed in this research resulted in the production of parts with a good finish aa~d very promising mechanical properties. The performance of these samples was evaluated by measuring tensile strength and flexural strength.
Keywords: curing, natural composite, finite element, resin transfer molding, natural fibers, mechanical properties *Communication to be addressed to: Prof. Mohini Sain, Earth Science center, 33 Willcocks Street, University of Toronto, Toronto, ON, MSS 3B3, m.sain@utoronto.ca Introduction:
Natural and wood fibre composites are manufactured by combining wood or other natural fibres such as flax, hemp, jute or kenaf, with polymers including polyethylene, polypropylene, or polyvinyl chloride (PVC). Composites based on natural and wood fibres are one of the fastest growing maxkets in the plastics :industry. They can be used to produce products for building, automotive, infrastructure and consumer applications.
These types of composites present many advantages compared to synthetic fiber reinforced plastics such as low tool wear, low density, cheap cost, availability and biodegradability. For high performance composites bast fibers, extracted from the stems _..T _~_ of plants such as jute, kneaf, flax, and hemp, are widely accepted as the best candidates due to their very good mechanical properties. Hemp especially was shown to have very promising tensile properties for such applications t-3 Natural fibers consist mainly of cellulose fibers. These fibers are made of microfibrils in a matrix of lignin (or pectin) and hemicellulose. The strength and stiffness of the fibers are provided by hydrogen bonds and other linkages. The overall properties of the fibers depend on the individual properties of each of its components. Hemicellulose is responsible for the biodegradation, moisture absorption and thermal degradation of the fiber. On the other hand lignin (or pectin) is thermally stable but is responsible for the UV
degradation of the fiber. On average natural fibers contain 60-80% cellulose, 5-20%
lignin (or pectin) and up to 20% moisture'.
The thermal stability of the reinforcing fibers is a key parameter in composite processing, especially in the case of thermosetting resins and their exothermic curing behavior. Wielage et a1.4 studied the thermal stability of flax and hemp fibers using differential scanning calorimetric (DSC} and thermo-gravimetric (TGA) methods.
Their results suggest that hemp and flax fibers have the thermal stability to endure thermoset cure reactions encountered during composite manufacturing.
Another important aspect is the moisture content of natural fibers. These fibers are hydrophilic and absorb water. The moisture content can go as high as 20%, but in most cases it will be in the range 5 to 10%. This can affect the final properties of the composites. During processing the presence of water can create voids in the matrix and also lead to a poor adhesion of the fibers with the hydrophobic resins°3. The hydrophilic nature of natural fibers can be a problem in the finished composites as well.
Li et al.s reviewed many papers concerning the mechanical properties of natural fibers. It was shown that the tensile properties of these fibers are not uniform along their length.
In their extensive report on "Composites Reinforced with cellulose based fibres"
Bledzki and Gassan2 gave some data for various natural fibers as well. As observed previously the characteristic values of natural fibers are comparable to those of glass fibers. The strength of natural fibers greatly depends on the process used to produce them.
In theory the elastic moduli of cellulose chains can reach values of 250 GPa.
However there is no existing procedure to separate these chains from the microfibrils and therefore obtain such values. Right now the pulp and paper industry is able to produce cellulose fibers with moduli around 70 GPa. Moreover some experimental data obtained from flax and pineapple fibers show that the tensile strength of these fibers is significantly more dependent on the length of the fiber than for the case of glass fibers.
Natural fibers seem to be less homogeneous than synthetic fibers. From these papers it can be concluded that even if natural fibers are well suited to replace glass fibers in composite materials many improvements can still be done concerning their mechanical properties.
Experimental data giving the tensile strength, flexural strength, modulus, impact force and compressive force are available in the literature for different types of natural-fiber composites.
Research on hemp fiber composite is still in its early stage and only few publications can be found in the literature. Kellerb worked on a biodegradable system based on thermoplastic resins. The mechanical properties of the resulting composites were found to be quite low compared to polypropylene. Pervaiz and Sain' studied the strength data for sheet molded polyolefin hemp fiber composites, and found that the tensile and impact strength of these materials were shown to be substantially lower than their glass fiber counterparts.
In this work a thermo-set resin will be the system of choice and hence the following paragraphs contain review of papers dealing with such polymers. Among the techniques available for the production of thermo-set composites, Resin transfer Molding is a very popular process in the automotive and aerospace industries to produce large and complex parts. Sebe et a1.$ manufactured hemp fiber/polyester composites using RTM.
They obtained good quality parts with high flexural properties, but the impact strength of these materials was found to be very low. Richardson and Zhang9 presented an experimental study of the mold filling process for a non woven hemp/phenolic resin system.
Fiber washing was shown to be a problem at low fiber concentration due to poor clamping.
Edge flow was observed during the mold filling as well. The use of performs larger than the mold solved this problem. The injection pressure and the fiber concentration were shown to be the critical parameters to achieve proper. mold filling. A few other publications presented natural fiber composites manufactured by RTM'o~".
Recently hemp fiber/unsaturated polyester composites were manufactured in our lab using a Resin Transfer Molding (RTM) process'z. These materials have promising mechanical properties. Surface modifications of the fibers were proposed in order to improve these mechanical properties as well as the fiber/matrix interface interaction'3.
The results did not provide substantial changes in the materials properties.
The strength tests gave promising results. In the presents study hemp fiber/Derakane composites were manufactured in our lab using the RTM process. The purpose of this work is enhance the mechanical properties by using epoxy vinyl resin and to provide a model predicting the cure behavior of a natural fiber composites; and also to optimize the RTM
process to obtain a high degree of cure in a minimum time. The next step of this work will be to propose some surface modification of the hemp fibers to improve the surface adhesion.
Experiments Resin Transfer Molding A number of polyester/natural fiber composites were manufactured in the lab.
The final dimensions of the parts were 380mm by 380 mm by 3.4 mm thick. The mold, made entirely of aluminum, was opened and closed manually with 16 screws distributed around the cavity. The two inlet ports were situated under the mold and a vent port was located on the top. It was kept at constant temperature during the curing reaction by water flowing inside its upper and lower sections. The water, circulated in a closed loop through a tank, was kept at constant temperature with a temperature controller connected to a thermocouple and an immersion heater of 2000 watts. To compensate for the heat produced during the exothermic crosslinking reaction cold water was kept running permanently in a copper coil placed in the tank. This system balanced itself around the preset temperature during the experiment. It should be noted that the thickness of the composite was defined by a frame placed between the upper and lower sections of the mold; it could therefore be modified in further experiments.

Prior to a typical experiment the surfaces of the mold were cleaned with the Frekote~ PM mold cleaner and then coated with the Frekote~ B-15 Sealer and the Frekote~ 700-NC mold release agent. Once these coatings were cured layers of natural fibers' mats having the mold's size were placed in the cavity. The mold was tightly closed and a vacuum of 725 mm of mercury was created in the cavity through the vent port connected to an aspirator placed on a tap. At this point the fibers were dried for 2 hours by circulating water at 55°C. The mold was then cooled down with cold water. In the meantime the resin was mixed with the initiator and placed in the injection pot. From there the resin was injected in the mold with compressed air at a constant gauge pressure of 2.00x105 Pa. This pressure was kept constant in the pressure pot by continuously adjusting manually the compressed air valve. The injection time of course varied with the amount of fibers present in the mold. Once the resin was observed at the outlet, the vent port was closed. A small flask was placed between the vent port and the tap for safety, to prevent any resin from flowing to the tap water. The resin was left flowing at the inlet fox more minutes to make sure that the mold was filled completely. Then the inlet ports were closed as well and hot water at constant temperature was circulated in the mold. The composite was cured under these conditions for an hour. Meanwhile the injection pot and all the tubes were cleaned with acetone to avoid any clogging due to cured resin.
Materials In this study the polymer used was Derakane[TM] 8084, epoxy vinyl ester resin obtained from the Dow Chemical Co. and it contained 4S wt% dissolved styrene.
This resin is manufactured for use in closed mold processes such as RTM. It is a low viscosity resin, which starts reacting by addition of an initiator. In this case the chosen initiator was MEKP DDM-9 from Ato-FINA. The resin manufacturer recommends using an initiator concentration between 1.5% and 3% by weight. Therefore three different concentrations were investigated during the pilot study: 1%, 1.5% and 2%. Following this study it was decided that a MEKP concentration of 1.5% should give the best results. To allow for curing to occur at 30°C, the resin was mixed with 0.3% by weight of 6%
cobalt naphthenate catalyst (Sigma Aldrich Co.). Additionally, 0.025% of 99% N, N
dimethyl aniline (Sigma Aldrich Co.) was used as an accelerator, while 1.5% of methyl ethyl ketone peroxide (9% active oxygen) was the initiator.
The fibers used in this study were manufactured by Flax craft, Inc. The Bastmat 100 was a 4mm mat made of 67% hemp fibers and 33% kenaf fibers. Hybrid fiber mats consisting of hemp fibers sandwiched between slim glass fiber mats (randomly oriented) were prepared. The mats were pre-press at a high temperature (above 80°C) and at a defined thickness to reduce their spring back behavior and allow more fibers to be placed in the mold.
Fibre treatment with Silane and Sizing Chemicals A 1 % by weight aqueous solution of 3-aminopropyltriethoxysiane (from Sow Corning) was prepared using distilled water. The solution was then poured in a bottle and sprayed on the hybrid fiber mats until soaking them. The mats were left on the bench for 30 min to allow the hydrolysis of the silane. Then the fibers were dried first in an oven at 100°C
for one hour followed by 12 hours at 80°C. Fibers were also separately treated with styrene malefic anhydride copolymer with different molecular weight and malefic anhydride content . Similar treatment was also carried out with rosin ester and polyethylene-imide (PEI). Finally, treatment of fibres were also carried out with maleated imide cationic polymers such as styrene maleimide and its quaternary salts with different charge density and molecular weights.
Composite Processing The composites with various fiber contents were prepared using the RTM
process.
Once treated the fiber mats were first placed in the mold and dried under vacuum. The water evaporating from the fibers could be observed in the liquid trap connected to the vent port. Once the fibers were dried for 2 hours, resin mixed with initiator was injected at a constant gauge pressure of 2.00 x105 Pa. This pressure was chosen by trial and error to provide the shortest injection time possible coupled with a proper wetting of the fibers (knowing that the maximum gauge pressure in the system should not exceed 2.5 x105 Pa for safety). Then the composite was cured at 40°C for an hour. Finally each part obtained was post cured in an oven at 105°C to ensure complete and homogeneous curing of the polyester matrix.
The mold being entirely made of aluminum the flow front could not be observed during the resin's injection. The injection time increased dramatically with increasing fiber content. Time collection was started with the injection of the resin and the values in Table 3 correspond to the time when the resin was first observed exiting the mold through the vent port. These results were expected since an increase in fiber's content decreased the permeability of the fabric in the mold; hence increasing its resistance to resin's flow.
More work is needed to quantify these permeability and compare them to the ones obtained with synthetic fibers and also hemplepoxy vinyl ester resin.
The natural fibber composites manufactured by the RTM process in this work were found to be of good quality. An excellent wetting of the fibres was obtained and the drying of the fibers prior to resin's injection permitted to avoid the formation of small gas bubbles in the part due to water evaporation.
Mechanical Properties The tensile and flexural strength of the composites were determined using a SATEC T10000 Materials Testing System. The tensile properties of the materials were measured following the ASTM standard method D638-99. The flexural properties were obtained according to the ASTM standard method D790-99. The size and shape of the different samples were chosen depending of their thickness as mentioned in these methods.
Results and Discussion Tensile Properties The tensile strength tests were performed using samples made from parts that reached final degree of cure. Figure 2 presents the tensile strength of Silane treated Stypol composite and Silane treated Derakane composite samples. As expected the tensile strength of the samples increased with change in resin from Stypol to Derakane. A
substantial increase was observed between the Stypol resin and the Derakane sample with 20.6% fibers, from 54.86 MPa to 64.52 MPa. After break, very less fiber pull out could be observed on the specimens with Derakane, proving that the fiber-matrix adhesion was substantially improved. For information the tensile strength of a glass fiberlunsaturated polyester composite i.e. Stypol of similar volume fraction and prepared using the same process was added in Figure 13. It can be seen that the natural fiber composites manufactured in this work have tensile strengths approximately 20% lower than their glass fiber/Stypol counterpart.
The tensile modulus of the Silane treated Stypol composite and Silane treated Derakane composite samples axe shown in Figure 14. The Stypol hemp fiber composites have the same tensile modulus as the Derakane one. Once again the glass fiber/Stypol results were added for information showing that the Derakane fiber composites had tensile modulus very close to the synthetic fiber one.
Flexural Properties The flexural strengths are reported in Figure 15. The Figure shows the same trend as the results of tensile strengths. The flexural strength of the samples increased with change in resin from Stypol to Derakane. An increase was observed between the Stypol resin and the Derakane sample with 20.6% fibers, from 12,9.96 MPa to 132.76 MPa. The explanations for these results are that the flexural properties are influenced by the fiber/matrix interface interaction as well. Once again the strength value for the glass fiber sample is 1.4 times greater than that for the natural fiber composites.
The modulus results can be seen in Figure 16. The flexural modulus for Derakane composites exhibits similar modulus when compared to Stypol composite.

Conclusions In this work the manufacture of natural fiber composites using a Resin Transfer Molding was investigated. The drying process before the resin's injection permitted to obtain a good wetting of the fibers as well as to avoid <~.ny formation of gas bubbles during curing. In this work data concerning the curing behaviour of Derakane 8084, an epoxy-vinyl ester resin were presented. In order to achieve high fiber contents with hemp fibres in a process such as RTM the need of pre-pressing stage at 100oC was asserted.
This additional step reduced greatly the spring back behaviour of the fibres, making the closure of the mold much easier.
The natural fiber composites obtained by this process were found to be of high quality.
No voids could be observed within the parts. The tensile, flexural properties were found to increase with change in resin from Stypol to Derakane. It was observed that the optimum properties were not reached in this study and that the fiber content higher than 20 vol% should yield better mechanical properties.
Acknowled e~ ment:
The author sincerely acknowledges the financial support from Network of Centre of Excellence-Auto 21-Canada, NSERC and Industry partners.
References [1] Saheb, D.N., Jog, J.P., "Natural fiber polymer composites: a review", Advances in Polymer Technology, 18, 4, 1999, pp. 351-363 [2] Bledzki, A.K., Gassan, J., "Composites reinforced with cellulose based fibres", Progress in Polymer Science, 24, 1999, pp. 221-274 [3] Prasad, B. M., Sain, M., "Mechanical properties of thermally treated hemp fibers in inert atmosphere for potential composite reinforcement", Materials Research Innovations, 7, 4, 2003, pp. 231-238 [4] Wielage, B., Lampke, T. R., Marx, G., Nestler, K., Starke, D., "Thermo gravimetric and differential scanning calorimetric analysis of natural fibers and polypropylene", Thermochimica Acta, 337, 1999, pp. 169-177 [5] Li, Y., Mai, Y-W., Ye, L., "Sisal fibre and its composites: a review of recent developments", Composites Science and Technology, 60, 2000, pp. 2037-2057 (6] Kelley, A., "Compounding and mechanical properties of biodegradable hemp fiber composites", Composites Science and Technology, 63, 2003, pp. 1307-1316 [7] Pervaiz, M., Sain, M., "Sheet-molded polyolefin natural fiber composites for automotive applications", Macromolecular Materials and Engineering, 288, 2003, pp 556-[8] Sebes, G., Cetin, N. S., "RTM hemp fiber--reinforced polyester composites".
Applied Composite Materials, 7, 2000, pp. 341-349 [9) Richardson, M. O. W., Zhang, A. Y., " Experimental investigation and flow visualization of the resin transfer mould filling process for non woven hemp reinforced phenolic composites", Composites part A, 31, 2000, pp 1303-1310 [10) Williams, G. L, Wooi, R. P., "Composites from natural fibers and soy oil resins", Applied Composite Materials, 7, 2000, pp. 421-432 [ 11 ] O'Dell, J. L., "Natural fibers in resin transfer molded composites", Proceedings Wood Fiber Plastics Composites Symposium, Forest Prod Soc., Madison, WI, 1997, pp.

[ 12] Rouison, D., Sain, M., Couturier, M., "Resin Transfer Molding of natural fiber reinforced composites: cure simulation°', Composites Science and Technology, under press [13] Rouison, D., Couturier, M., "The effect of surface modification on the mechanical properties of hemp fiberlpolyester composites", To be published.
[ 14] Promotional Literature, DERAKANE epoxy vinyl ester resins: product and usage guide. Dow Chemical Company, USA.
List of Figures:
Fi ure 1: Diagram of resin transfer molding equipment F~i ure 2: Tensile strength of Silane treated composites (20 vol% fibres) Figure 3: Tensile modulus of Silane treated composites (20 vol% fibres) Fi a 4: Flexural strength of Silane treated composites (20 vol% fibres) Fi ~ur~e 5: Flexural modulus of Silane treated composites (20 vol% fibres) Tensile Flexural Im act Strength (J/r strengthmodulus strengthmodulus mothunnoctct M a G M a G _ a a 20% hemp, glass58 1.6 96 5.7 %

fiber 20% hemp, glass65 1.5 133 6.3 175 198 %

fiber 30% hemp, glass80 1.9 191 ~ 9.9 ~ 240 ~ 260 1 %

fiber RESIN TRANSFER MOLDING (RTM) PROCESS
Achievement:
1 The technique has been designed for optimization of RTM process to achieve the composites with high mechanical performance from hemp/glass fibers and synthetic thermoses resin. The resin injection pressure and temperature have been optimized.
2 The curing system has been standardized by selecting appropriate resin, catalyst, accelerators, retarders, coupling agent with their concentrations Typical example of the RTM composite with 26%hemp fiber and 7% glass fiber exhibits the optimum tensile strength of 80 MPa, Flexural strength of 200 MPa and modulus of 10 GPa and impact strength of 250 J/m.

Claims (18)

1. Claim 1: A process to prepare structural composites for automotive, aerospace, furniture and sports goods applications whereby composites are manufactured by resin transfer molding process with resins in liquid form either modified or unmodified were injected under pressure into a cavity of a pre-shaped design containing fibres in nonwoven, loose or mat form whereby fibres obtained mainly from natural origin but in few cases they are combined with synthetic fibres to a lesser extent are layered in the preshaped mold in single or multiple layers whereby the fibres are either used as is without any surface modification or their surface were modified either by chemical , heat or enzymatic treatment to improve adhesion and to reduce moisture ingress in the molded composites, the layered fibres being impregnated with liquid resin of adequate viscosity after drying in-situ within the mold by taking off the additional moisture from fibre by applying vacuum before resin impregnation, allowing to heat the resin impregnated fibre to solidify the resin inside the mold to obtain product parts of required shape, size, performance and durability requirement.
2. Claim 2. A process as recited in claim 1, where in the resin system is a thermosetting material may comprise of Silane Modified Polyester resin, Epoxy vinyl ester resin, Silane modified epoxy ester resin, Acrylic resin, silane modified acrylic resin, Soy resin, silane modified soy resin and contains not less than 30 % by weight of the resin composition.
3. Claim 3. A process as recited in claim 1, where in the said moldable composition comprises cellulosic fibres selected from pulp fibres, bast fibres, leaf fibres, and agricultural waste fibres such as corn, wheat and rice straw and comprises not more than 70 percentage by weight of the said composition.
4. Claim 4. A moldable composition prepared as in claim 1, and 3 wherein the cellulosic fibres comprise of wood pulp fibres and bast fibres.
5. Claim 5. A said moldable composition as cited in claim 1,3, and 4 where in wood pulp fibres selected from thermomechanical pulp (TMP), kraft pulp and bleached kraft pup (BKP) from hard wood or soft wood or a combination of the same.
6. Claim 6. A said moldable thermoplastic composition as cited in claim 1,3 and 4, where in the bast fibres can selected from hemp, flax, jute and kenaf and / or a combination of the same.
7. Claim 7. A process as recited in claim 1, where in the synthetic fibres comprise of any chemical artificial fibres such as polypropylene, carbon, kevalr, boron, and glass fibres.
8. Claim 8. The said composite product of claim 7, where in the inorganic fibres can selected from any chemical artificial fibres such as polypropylene, carbon, kevlar, boron, and glass fibres.
9. Claim 9. The said composite product of claim 4 to 6, where in the organic fibres are treated with an enzyme or chemicals.
10. Claim 10. The said composite product of claim 4 to 6, where in the organic fibres are treated or untreated were used in loose form or mat form, more specifically loose form were used in the core and mat form were used on the surface of the layered structure.
11. Claim 11. The said composite product of claim 4 to 7, where in the treatment chemicals are from any of the following functional chemicals such as silanes, imides, maleimides, alkanes, alkene dimers, rosin acid and esters, alkyl imines.
12. Claim 12. The said composite product of claim 4 to 8, where in the specific treatment chemicals are Microthene Powder, Polyvinyl acetate, Polyethyleneimine, Styrene Maleic Anhydride, Styrene Maleic Imide, Alkyl ketene Dimer, Rosin Acid, Trimethoxy (3,3,3-trifluoropropyl) silane, (3-Glycidyloxy propyl) trimethoxy silane, Dimethyl (pentafluorophenyl) styryl silane, Chloro-dimethyl (3,3,3-trifluoropropyl) silane
13. Claim 13. The said composite product of claim 4 to 7, where in the treatment enzymes are from any of the following category : such as endo-glucanase, hemicellulase.
14. Claim 14. The said composite product of claim 4 to 7, where in the treatment enzymes are from any of the following category : such as endo-glucanase, hemicellulase.
15. Claim 15. The composite of claims 7-13, where in the said liquid resin material has been modified with silane chemicals to enhance mechanical properties of the resin in the presence of fibre.
16. Claim 16. The composite of claims 7-14, where in the said resin transfer molded composite materials have flexural strength and modulus 140-250% and 300- 700% greater than that of the said resin materials and have flexural modulus exceeding 10GPa.
17. Claim 17. The composite of claim 7-15, where in the said composites have impact strength improved by more than 70% by using loose fibre in the layered design.
18. Claim 18. The said composite product of claim 1-17 can be used for structural applications in the automotive, sports goods, furniture and aerospace industry.