Permeability and Mechanical Property ... - Wiley Online Library

Permeability and Mechanical Property Correlation of Bio Based Epoxy Reinforced with Unidirectional Sisal Fiber Mat Through Vacuum Infusion Molding Technique

S. Rajkumar,1,3* Jimi Tjong,1,2 S. K. Nayak,3 M. Sain1,4 1 Faculty of Forestry/Chemical Engineering, University of Toronto, Ontario M5S 3B3, Canada 2

Powertrain Engineering Research and Development Centre, Ford Motor Company, Windsor, Canada

3

Central Institute of Plastics Engineering and Technology (CIPET), Advanced Research School for Technology and Product Simulation, Guindy, Chennai 600032, India 4

Adjunct, King Abdulaziz University, Jeddah, Saudi Arabia

The present investigation is focused to study the permeability of natural fiber during vacuum infusion (VI) process and the effect of the surface treatments of natural fiber, fiber loading direction, resin flow direction and process parameter on the tensile properties of developed composites (sisal/bio based epoxy). The bio based resin exhibits good flow characteristics in NaOH and isocyanate treated fibers which may be attributed to change in polarity. The surface treatments appear to provide an appreciable enhancement in tensile strength through enhanced bonding between fiber and matrix. The longitudinal tensile strength has been found to be higher than that of the transverse direction and the flow along the fiber provides maximum tensile strength. It has also been demonstrated that VI process provides improved mechanical properties as compared to handlayup process. Morphological studies of fractured developed composites were performed by scanning electron microscopy (SEM) to understand the de-bonding of fiber/matrix adhesion. POLYM. COMPOS., 00:000–000, 2015. C 2015 Society of Plastics Engineers V

INTRODUCTION A bio material in composite industry is introduced to overcome the critical issues such as environmental pollution, ecological issues, degradability, maintenance, and weight reduction [1, 2]. The increasing demand in the field of bio material research for an in-depth understanding of processing phenomenon to convert it into a useful product is essential. These bio materials are fabricated by using Correspondence to: S. Rajkumar; e-mail: [email protected] DOI 10.1002/pc.23797 Published online in Wiley Online Library (wileyonlinelibrary.com). C 2015 Society of Plastics Engineers V

POLYMER COMPOSITES—2015

well established molding techniques such as hand-layup, press molding, pultrusion, resin transfer molding, and vacuum infusion (VI) molding [3, 4]. The lack of fiber wetting and insufficient resin distribution in traditional manufacturing processes leads to inferior composite performance. To efficiently use these techniques, conceptual understanding of natural fibers, its impact on process parameters and structure–property characteristics is enviable. Donnell et al. recommended that infusion molding process is more suitable for manufacturing of bio composite [5]. Therefore, this study is targeting to understand and develop the design fundamentals of resin impregnation and fiber wetting process to enhance the fiber/matrix interface. The permeability study of VI process is difficult task due to that the resin flow path is deviating from standard path which is attributed the unpredictable variations in the fiber preform mat [6]. Yula ma et al. observed that the effect of resin such as vinyl ester, corn syrup and epoxy on the permeability is less as compared to fiber architecture [7]. The flow profile is differing with the fiber-mat orientation and appears to be strongest along the inter-tow channels direction [8]. Reza Masoodi et al. observed that the swelling of fiber in bio based epoxy were more intense than conventional epoxy, which attributed to more interaction of cellulose fiber with bio based epoxy resin [9]. Mohseni Languri et al. also observed change in natural fibers diameter due to swelling and they observed, good flow-front prediction for natural fiber with the addition of swelling as a function of time in variable permeability model. Since the porosity is not constant in a swelling porous medium [10]. Dungan et al. claimed that the flow among the fiber filaments decrease the overall permeability by increasing penetration times due to the required wetting of the tows [11].

Albuquerque et al. demonstrated that NaOH treated jute fiber shows superior mechanical properties, due to better fiber surface wettability [12]. Soren Barsberg et al. derived a conclusion that, the generalization of wetting behavior of natural fiber is more complex due to macromolecular rearrangement swelling, extraction etc.,[13] Chih-h sin shih et al. highlighted that connecting the permeability to a particular fiber architecture is difficult [14]. The tensile strength of woven fiber composites depends on the fiber orientation, weave style, and the bonding between the fiber and matrix [15]. To develop natural fiber reinforced composite with superior mechanical strength, it is essential to reduce hydrophilic behavior of the fibers through surface modifications. Several studies were pertained to improve the properties of natural fiber using various chemical treatments [16, 17]. The chemically modified natural fiber offers improvement in mechanical strength and wettability between fibers and matrix, and reduces the moisture absorption which is due to enhanced bond interaction between the fibre and polymer [18, 19]. The NaOH chemical modification causes stronger fiber–matrix adhesion by increasing roughness of the fiber surface and better mechanical anchoring. Mechanical and thermal behavior of the composite gradually was improved by this treatment [20, 21]. Silane molecules form a chemical link between the fiber surface and the matrix through a siloxane bridge. Silane forms silanol in the presence of fiber moisture. The one end of silanol reacts with the cellulose hydroxyl group of the fiber and the other end reacts with the matrix functional group [22–24]. Kabir modified natural fibers with silane agent and stated that the silane covering a surface as layer that assist penetration in pores, higher tensile strength and reduction in moisture absorption [22]. Ly et al. treated natural fiber with isocyanate. In these composites, it was found that an enlightened performance is attainable with treated fiber [25]. The plain weave fiber composite shows higher tensile strength than randomly oriented composites [15]. Pinto et al. developed jute/epoxy laminated composites to investigate the modes of failures. In these composites, it was noticed that the major failure due to fiber/matrix de bonding and matrix fracture. Further, they found that the fiber/matrix de bonding due to the lack of adhesion and as an effect of fiber pull-out [26]. Oksman et al. achieved higher mechanical properties of sisal fibers composites manufactured by resin transfer molding process [27]. The understanding of the behavior of natural fiber reinforcement in the VI is less documented in the open literature. No significant work has been reported concerning the effect of chemical treatment on permeability of the natural fiber in the VI. There were no studies reported in literature on the effect of flow direction on tensile strength in infusion process. Hence, a sincere effort has been made to evaluate the effect of chemical modification on permeability of natural fibers in the VI process and finally on tensile properties of developed composites. This study also tries to give a hypothesis by correlating the permeability and tensile strength. The present investigation also addresses the effect 2 POLYMER COMPOSITES—2015

of fiber orientation, flow direction, and process parameters on tensile strength of composites. EXPERIMENTAL Materials The diglycidyl ether of bisphenol A (DGEBA) is used as a base component of epoxy resin. An amount of 30% epoxidized soybean oil (ESO) with 6.5 wt% epoxy oxirane content and specific gravity of 0.987 is added to the resin base. Triethylenetetramine (TETA) is used as a curing agent and the stoichiometry relation between the blended resin to curing agent is calculated to be 12.40. The woven sisal fiber mat with an areal density of 575 g/m2 is the reinforcement. The average filament diameter and thickness of the mat of sisal fiber are 0.35 mm and 1.2 mm, respectively. Fiber Surface Treatment To investigate the influences of chemical treatments in terms of the permeability and tensile strength, the fibers are treated chemically with different chemicals namely NaOH, silane and isocyanate. In alkali Treatment, the woven sisal mats were soaked in a 2 wt % solution of NaOH at room temperature for one hr. Then the mats were dried for 12 hr at 808C [18]. For silane treatment, 1 wt % of 3-aminopropyltriethoxysilane is dissolved in distilled water. The fibers are immersed in the solution for 30 min at room temperature to permit the hydrolysis of the silane. This is followed by drying of fibers in an oven at 1008C prior to its exposure for 12 hr at 808C [18, 22]. In isocyanate treatment, 6% of methylene diphenyl diisocyanate (MDI) solution is prepared using toluene solvent. The woven sisal fiber mat (150 g) is placed in the reaction kittle and the mat is immersed in the MDI solution of about 2000 ml. The resulting suspension is then heated to 508C for about 30 min. After completion of the reaction, the mat is decanted and dried at 708C for 2 hr [28]. Composite Fabrication: Vacuum Infusion The VI process experimental setup is shown in Fig. 1. This setup uses 500 3 400 3 50 mm3 dimension infusion mold which is coated by mold releasing agent. To improve the resin impregnation and surface finishing, gel coat is applied over an area of 410 3 310 mm2 on the infusion mold. The sisal fiber mat is prepared to the required size 400 3 300 mm2. The flow medium is placed on the upper face of the fiber mat, as it has to be released after infusion. Due to its high permeability the flow medium increases resin flow at the upper surface of the fiber, thus reducing impregnation time. The peel ply fabric is placed between the fiber mat and the flow medium. This peel ply fabric assists the removal of the flow medium. A vacuum bag is used to cover the whole top of the mold including the DOI 10.1002/pc

K5

df 2 u3 k ð1-u Þ2

(2)

where df is the fiber diameter, k is the Kozeny constant and u is the porosity of the mat. Kozeny constant is an approximate value only. This model is only valid for flow along the fiber mats and the permeability predicted from this model is isotropic [14]. The basic Carman model prediction can be improved by modifying the Carman–Kozeny equation by varying kozeny constant in the different direction, as in equation 3 K5 FIG. 1. Experimental setup and permeability measurement. [Color figure

sealant tape. Further, it is necessary to confirmed that the vacuum bag being used should be 254mm greater than the sealant tape to accommodate the shrinkage. Mold is connected to vacuum pump and resin inlet on its sides. Air is evacuated from the mold by the application of 2735 mm vacuum pressure of mercury. In the midst of this, the blended resin (DGEBA/ESO) that cures at room temperature is poured into a beaker. This is followed by addition of TETA curing agent stir mixed for few minutes. Subsequently, the resin beaker is moved inside the inlet reservoir tank and connected to the inlet tubing for resin infusion. The resin is provided ample time to infuse completely over the mat followed by closing off resin supply. The vacuum is permitted to persist until the resin has gelled. In addition to these VI process, the hand layup method is used for comparison. In hand layup, the mold surface is cleaned, waxed and gel-coated, and then the fabric is laid on the mold. Resin impregnation is done manually.

un11 Cð12 uÞn

(3)

Where n and C are experimental parameters. This model does not represent the behavior of the fiber reinforcements with complete accuracy, and this modified carman-kozeny model may not accurately fit for all data sets, until the n and C parameters are modified for each fiber reinforcement mat. It is a complex model to relate fiber architecture of each mat to the values of the other parameters. Some authors also found differences in permeability values with different resin of similar viscosity, a phenomenon which is not accounted for above models [30–32]. Rajkumar et al. developed permeability equation which includes porosity and contact angle to improve the flow prediction accuracy. They observed that there is a reasonable percentage of improvement in the flow prediction. Since in an infusion process the driving force for the micro flow is the capillary pressure developed in the intra tow region that should be added in flow prediction equation. K5

c/ 2 1 h2 a1b 1 1 h2 2 21 21

(4)

Permeability Measurements Permeability is a key parameter for flow prediction in infusion molding process [29]. Well established and reliable models frequently used in literatures are adopted for this study [30]. The Darcy’s law is the most commonly used for describing the resin flow through porous reinforcement given in Eq. 1. K 5ðQ:l:DLÞ=ðA:DPÞ

(1)

where K is permeability (m2), l is viscosity of the resin (Pa s), Q is the volumetric flow rate (m3/s), DL is the preform length (m), A is mold cavity transverse area (m2), and DP is the pressure gradient (Pa). Subsequently, the carman–kozeny developed a flow model on array of channels of varying cross section. The carman–kozeny relationship is mainly used to predict the flow behavior of resin in the fiber porous medium, while relating the porosity and permeability. DOI 10.1002/pc

where u is porosity, h is contact angle, a, b, and c are empirical constants [33]. Contact Angle Measurement The contact angle which is a parameter to determine the wettability of surface was measured by goniometer. The wettability of epoxidized soya bean oil blended resin on differently treated sisal fibers were investigated. Tensile Testing Tensile Testing was performed according to ASTM D3039 using an INSTRON 3382 testing machine. Rectangular test specimens cut to 25 mm 3 255 mm as specified by the standard, were tested at a crosshead speed of 2 mm/min. For statistical significance, five to six samples were characterized for each case. POLYMER COMPOSITES—2015 3

FIG. 2. ESO blended resin on the sisal fiber (a) Untreated (b) NaOH treated (c) Silane treated.

Scanning Electron Microscopy (SEM) The fracture surface of the specimens were investigated using SEM (EVO15, Carl Zeiss, Germany) operating in the high vacuum mode at accelerating voltage of 10–15 kV. RESULTS AND DISCUSSION

contact angle between fiber and matrix. This may be also due to roughness behavior on the surface. In isocyanate treated surface shows appreciable wetting behavior to ESO blended resin which may be the reason of interaction between the non-polar fiber surfaces and resin. Different observation are recorded for silane treated fibers wetted with ESO blended resin, it may be due to polarity difference between SiAO linkage of silane and NACAO linkage in isocyanate [35, 36].

Contact Angle The contact angles between sisal fiber mat and matrix are shown in Fig. 2.The contact angle measurement provides information about the surface wettability, which indicates the degree of wetting when a fiber and matrix interact. The decrease of contact angle leads to the appreciation in wettability. Contact angle is measured by placing a drop of blended epoxidized soy bean oil on the untreated and various treated sisal fiber mat surface respectively. It is clearly evident that the surface modification of natural fiber creates a difference in the interaction of fiber and matrix. The initial contact angle of the untreated fiber is very high. It may be attributed to fact that the interaction between non-polar resin molecules and polar OH group on sisal fiber mat which limits resin spreading and it remains as a droplet on the mat surface [34, 35]. Inversely, In NaOH treatment, the polar OH groups are reduced from the surface of the sisal fiber which enables the fiber to exhibit a non-polar behavior [24, 34]. Due to the fact that non-polar-non polar interaction, it reduce the 4 POLYMER COMPOSITES—2015

Permeability Study Permeability of natural fibers mainly important in low pressure injection techniques like VI where void formation and filling time can be increased dramatically when the permeability decreases. The experimental analysis involves resin flow front measurements with filling time during the resin impregnation. In untreated sisal fiber mat shows lesser permeability, which may be attributed to weak interaction between non-polar ESO blended resin and polar OH group on the untreated sisal fiber. When sisal fiber is treated by NaOH, the initial contact angle decreases to 408 wetted with ESO blended resin. NaOH treatment increases surface roughness resulting in better mechanical interlocking and the amount of cellulose exposed on the fiber surface. This increases the number of possible active sites. Consequently, the surface free energy may increase, probably due to the decrease of contact angle and allows better fiber wetting. Thus, increase the permeability of sisal fiber mat. Besides, during NaOH treatment the polar OH group on the fiber DOI 10.1002/pc

TABLE 1. Permeability and contact angle of untreated and treated sisal fibers.

Treatment

Permeability (m2)

Fiber volume fraction (%)

Contact angle (Radians)

Untreated NaOH Silane Isocyanate

7.38310210 1.4631029 6.07310210 1.213 1029

33 35 33 30

1.13 0.69 1.30 0.69

surface may reduce which allow better wetting. The similar trend is observed for permeability of isocyanate treated fibers. The permeability of mat is decreased after silane treatment which indicates high repelling action attained and polarity difference between interaction of fiber and matrix. The Table 1 shows identical behavior between wettability and permeability. It is well known fact that the mold filling time is related to permeability, resin injection pressure, geometry of the mold, resin viscosity, and other parameters. Once wettability is improved, the permeability will increase this will leads to the quicker filling as shown in Fig. 3 [33, 37] it showed that mold filling time is decreased with the increase of permeability. The Fig. 4 presents two models for comparison the experimental results and find out the accuracy of the model. The improved kozeny model (Eq. 3) is generally used to find the permeability, while relating the porosity and permeability. The values of C (4.8 3 108 m22, 0.0103) and n (1.48, 2.04) are adopted from the literature [14, 38].

FIG. 3. A comparison of flow front as a function of time for untreated and chemically treated sisal fiber. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

The next model includes contact angle, it shows the effect of the contact angle in the prediction results. The present study reveals that the degree of capillary pressure is more in natural fiber as compared to synthetic fiber. This may be due to that the imperfect and hollow configuration of natural fibers provides more capillary channels for micro-flow. It is also evident from Fig. 4 that prediction of flow is better in the case of untreated fiber whereas treated fibers (NaOH, Silane, Isocyanate) the flow prediction is poor. This may be attributed to the fact that static contact angle taking place within the system. Instead of static contact angle, if there is a dynamic contact angle then flow prediction may be better for treated samples. The silane treatment and isocyanate treated fiber

FIG. 4. Comparisons of experimental data with model predictions (a) Untreated, (b) NaOH, (c) Silane and (d) Isocyanate. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

DOI 10.1002/pc

POLYMER COMPOSITES—2015 5

findings may be attributed to the effect of transverse (z direction) permeability and unsaturated permeability may influence more than the longitudinal and saturated permeability. Effect of Fiber Loading Direction on Tensile Strength

FIG. 5. Tensile properties of treated and untreated sisal/bio based composites.

demonstrate similar behavior in flow prediction. The contact angle model clearly emphasizes its strength in all aspects as seen from Fig. 4a–d. The adding contact angle in permeability model leads to better accuracy and precision as compared to kozeny model. The contact angle model appears to possess the best fit value. Therefore we can conclude by saying that contact angle model gives good accuracy as compared kozeny model.

It is clear from Fig. 6 that, in transverse direction the tensile strength is significantly lower than that of the longitudinal direction. This may be attributed to that, the bundle of fibers is aligned perpendicular to the load direction and these fibers are weak in the transverse direction, they cannot contribute efficiently to the strength in that direction [26]. These composite generally exhibit behavior between the isostrain and isostress conditions. Strength and stiffness are generally much higher along the fiber direction (isostrain) than perpendicular to the fiber direction (isostress).In the isostrain case, the contribution of a constituent is directly proportional to its volume fraction. In the case of isostress, the fibers are much less effective in raising the composite modulus. Unidirectional composites are inherently anisotropic in that maximum strength and modulus are achieved along the direction of fiber alignment. Effect of Flow Direction on Tensile Strength

Tensile Strength of the Composites The Effect of Chemical Treatment on Tensile Strength. The effect of different chemical treatments on the tensile strength of sisal/ESO based composites is shown Fig. 5 It is clear that in all cases an improvement is observed. The NaOH treatment takes out certain portion of hemicelluloses, lignin, wax, and other impurities present on the fiber surface. Consequently, the fiber bundles are disintegrated to single fibers. Further, it reduces fiber diameter, thereby increases the aspect ratio which leads to the development of a rough surface that results in enhanced interfacial bonding between fiber and matrix and an increase in tensile properties. The tensile strength improvement in the case of silane treated fiber composite due to the amino silane helps to bond with hydroxyl groups through hydrolysable alkoxy groups and amino groups capable of interacting with the epoxy resin. It emphasized that, the silane creates a bridge between the natural fiber and the resin that may improve the interfacial adhesion between the fiber and matrix. The formation of strong covalent bonds between the isocyanate and hydroxyl groups of cellulose may leads to significant improvement in strength for isocyanate treated fiber [35]. The hydrophilic nature of the sisal fiber can be reduced by isocyanate treatment. The isocyanate treatment makes fiber hydrophobic, compatible and highly dispersible in the matrix. This will result in a strong interfacial bond between the fiber and the matrix. In the silane treated fiber based composites have shown lesser permeability and higher tensile strength. These 6 POLYMER COMPOSITES—2015

The key characteristic of unidirectional fiber mats is that the fiber tows are aligned in one direction. In unidirectional fiber mats, when resin is injected in to a mold, it firstly passes through the fiber inter-tow channels without impregnating the fiber tows. After the front has passed, resin from the surrounding gap region continues to impregnate the fiber tows. This delayed impregnation of tows leads to the unsaturated flow in liquid molding process [39]. According to the unsaturated flow theory, if these fiber mats are placed such that their tows are in the x direction with flow taking place in this direction as well, then the resin will flow easily through the inter-tow channels aligned with the fiber. The above fact leads to good wettability and higher permeability of the system. Good wetting can enhance the adhesive

FIG. 6. Effect of fiber loading on tensile properties. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

DOI 10.1002/pc

FIG. 8.

Effect of process parameters on tensile properties.

FIG. 7. Effect of flow direction on tensile properties. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

resin, less impregnation of fibers and subsequently low wetting and therefore low strength of composite. bond strength between fiber and matrix, which is leads to better tensile strength [12]. This is clear evidence from Fig. 7 that the fiber have higher permeability, good wetting characteristics and formation of strong bonds have to be assured for higher tensile strength. Decreasing in tensile strength in y-direction is because of low flowability of

FIG. 9.

DOI 10.1002/pc

Comparing VI and Hand-Layup Process on Tensile Strength of Composite From Fig. 8, It can be observed, the VI sample shows the better mechanical properties then hand-layup technique. It may be attributed to decreasing of voids due to applied

SEM photographs of untreated and treated fiber composites: (a) Untreated, (b) NaOH (c) Isocyanate and (d) Silane.


compression force (vacuum pressure) and maintained until the resin cures completely for infusion process. In addition, the resin infusion process involves vacuum, there is a little or no moisture content while the resin sucked into the mold. Reducing the moisture content of the infusion sample during the curing phase will contributed to better mechanical properties of composites. The hand layup technique is carried out in environment that exposed to the moistures and gaseous as a result it will effect to the composites during the curing process.

dinal flow also supports the enhancement of tensile strength as compared to transverse flow because of unsaturated flow theory. The SEM micrographs of treated samples demonstrated that there is an excellent interfacial bonding between the reinforcement and matrix. This might be the reason that there is appreciable enhancement tensile strength of the treated samples as compared to untreated sample. We can conclude by present study that VI is a promising technique to fabricate bio-based composites with superior mechanical properties.

SEM Characterization

NOMENCLATURE

As it can be observed from Fig. 9, the SEM characterization of sisal reinforced composites shows the brittle character of the fracture at a macroscopic level with significant presence of pull-out representing weak fiber/ matrix interface and presence of fibrillation. Many hollow portions after the fracture can be seen in the micrograph, indicating that the phenomenon of fiber “pull-out” occurred to a large extent. The reduction in tensile strength of untreated composite is possibly due to more number of voids present and large gap between the fiber and the matrix [15]. After NaOH treatment, the gap between the sisal fiber and epoxy is closer than in the untreated sisal fiber composite (see Fig. 9b) due to fact that the epoxy resin will easily permeate into the gaps of fibrillation and subsequently strongly joint the fibers together. A good link between the isocyanate treated fibers and the epoxy could be seen from the fracture of the composite. Figure 9c presents lesser voids due to fiber pull out and closed packing between the fiber and resin which results higher strength compare to NaOH treated fiber. The occurrence of enriched resin surrounding the fibers restricts the sliding motion of fibers by shearing action at the interface. The closely packed interfacial bonding between fiber and matrix is evident from Fig. 9d. Better bonding and absence of fiber pullout is mainly attributed to higher strength of silane treated fiber. Thus, it can concluded that the improvement in tensile strength may be attributed to excellent interfacial adhesion between sisal and matrix. CONCLUSION Bio based resin exhibit good flowability in NaOH and Isocyanate treated fiber composites compare to untreated fibers due to polarity changes. Significant enhancement in the tensile properties of the developed composites has been observed after various surface treatment processes which may be because of excellent interfacial bonding between sisal fiber and matrix. VI samples depict better mechanical properties as compared to hand-layup due to less number of voids and moisture absorption. It has also been seen that the tensile strength is more in longitudinal direction of fiber as compare to transverse direction of the fiber. This may be because of the isostrain behavior of the composite. Longitu8 POLYMER COMPOSITES—2015

DGEBA Diglycidyl ether of bisphenol A ESO Epoxidized soybean oil MDI Methylene diphenyl diisocyanate SEM Scanning electron microscopy TETA Triethylenetetramine VI Vacuum infusion REFERENCES 1. O. Faruk, A.K. Bledzki, H.P. Fink, and M. Sain, Prog. Polym. Sci., 37, 1552 (2012). 2. O. Faruk, A.K. Bledzki, H.P. Fink, and M. Sain, Macromol. Mater. Eng., 299, 9 (2014). 3. M.P. Ho, H. Wang, J.H. Lee, C.K. Ho, K.T. Lau, J. Leng, and D. Hui, Compos. B, 43, 3549 (2012). 4. G. Francucci, E.S. Rodrıguez, and A. Vazquez, Compos. A, 41, 16 (2009). 5. A.O. Donnell, M.A. Dweib, and R.P. Wool, Compos. Sci. Technol., 64, 1135 (2004). 6. J.B. Alms, S.G. Advani, and J.L. Glancey, Compos. A, 42, 57 (2011). 7. Y. Ma and R. Shishoo, J. Compos. Mater., 33, 729 (1999). 8. T. Roy and K.M. Pillai, Polym. Compos., 26, 756 (2005). 9. R. Masoodi and K.M. Pillai, J. Reinf. Plast. Compos., 31, 285 (2012). 10. E.M. Languri, R.D. Moore, R. Masoodi, K.M. Pillai, and R.Sabo, “An approach to model resin flow through swelling porous media made of natural fibers,” In proc. 10th International Conference on Flow Processes in Composite Materials (FPMCM10), Monte Verit a, Ascona, Switzerland. 11–15 July 2010. 11. F.D. Dungan and A.M. Sastry, J. Compos. Mater., 36, 1581 (2002). 12. A.C. De Albuquerque, K. Joseph, L.H. de Carvalho, and J.R.M. d’Almeida, Compos. Sci. Technol., 60, 833 (2000). 13. S. Barsberg and L.G. Thygesen, J. Colloid. Interface Sci., 234, 59 (2001). 14. C.H. Shih and L.J. Lee, Polym. Compos., 19, 626 (1998). 15. A. Alavudeen, N. Rajini, S. Karthikeyan, M. Thiruchitrambalam, and N. Venkateshwaren, Mater. Des., 66, 246 (2015). 16. A.M.M. Edeerozey, H.M. Ahil, A.B. Azhar, and M.I.Z. Ariffin, Mater. Lett., 61, 2023 (2007). 17. M. Tajvidi, R.H. Falk, and J.C. Hermanson, J. Appl. Polym. Sci., 101, 4341 (2006). DOI 10.1002/pc

18. D. Rouison, M. Couturier, and M. Sain, The effect of surface modification on the mechanical properties of hemp fiber /polyester composites. In: SAE 2004 World Congress and Exhibition, 1–5 March 2004, Detroit, MI, paper no.2004201-0728, pp. 63–77. 19. J. Rout, M. Misra, and S.S. Tripathy, Compos. Sci. Technol., 61, 1303 (2001). 20. M.S. Sreekala and S. Thomas, Compos. Sci. Technol., 63, 861 (2003). 21. N. Lopattananon, Y. Payae, and M. Seadan, J. Appl. Polym. Sci., 110, 433 (2008). 22. M.M. Kabir, Effects of Chemical Treatments on Hemp Fibre Reinforced Polyester Composites, PhD Thesis, University of Southern Queensland, Australia (2012). 23. J. George, M.S. Sreekala, and S. Thomas, Polym Eng Sci., 41, 1471 (2001). 24. X. Li, L.G. Tabil, and S. Panigrahi, J. Polym. Environ., 15, 25 (2007). 25. B. Ly, W. Thielemans, and A. Dufresne, Compos. Sci. Technol., 68, 3193 (2008). 26. M.A. Pinto, V.B. Chalivendra, Y.K. Kim, and A.F. Lewis, Eng. Fract. Mech., 114, 104 (2013). 27. K. Oksman, J. Reinf. Plast. Compos., 20, 621 (2001). 28. J. George, S.S. Bhagawan, and S. Thomas, Compos. Sci. Technol., 58, 1471 (1998).

DOI 10.1002/pc

29. A. Hammami and B.R. Gebart, Polym. Compos., 21, 28 (2000). 30. K. Han, S. Jiang, C. Zhang, and B. Wang, Compos. A, 31, 79 (2000). 31. T.M. Schmidt, T.M. Goss, S.C. Amico, and C. Lekakou, J. Reinf. Plast. Compos., 28, 2839 (2009). 32. Nuno andre curado mateus correia.Analysis of the vacuum infusion Moulding process. Ph.D thesis, University of Nottingham, 2004. 33. S. Rajkumar, J. Tjong, S.K. Nayak, and M. Sain, Can J Chem Eng., 93, 1364 (2015). 34. S. Rajkumar, J. Tjong, S.K. Nayak, and M. Sain, J. Reinf. Plast. Compos., 34, 807 (2015). 35. K. Joseph, S. Thomas, and C. Pavithran, Polymer, 37, 5139 (1996). 36. A.K. Gupta, M. Biswal, S. Mohanty, and S.K. Nayak, Adv. Mech. Eng., 4, 418031 (2012). 37. A.I.S. Brigida, V.M.A. Calado, and L.R.B. Gonc alves, Carbohydr. Polym., 79, 832 (2010). 38. E. Rodrıguez, F. Giacomelli, and A. Vazquez, J. Compos. Mater., 38, 259 (2004). 39. H. Tan, T. Roy, and K.M. Pillai, Compos. Part A-Appl. S, 38, 1872 (2007).
