Influence of Hybridizing Flax and Hemp-Agave Fibers with Glass

materials

Article Inﬂuence of Hybridizing Flax and Hemp-Agave Fibers with Glass Fiber as Reinforcement in a Polyurethane Composite

Pankaj Pandey 1, Dilpreet Bajwa 2, Chad Ulven 2 and Sreekala Bajwa 1,* 1 Department of Agricultural and Biosystems Engineering, North Dakota State University, Fargo, ND 58102, USA; [email protected] 2 Department of Mechanical Engineering, North Dakota State University, Fargo, ND 58102, USA; [email protected] (D.B.); [email protected] (C.U.) * Correspondence: [email protected]; Tel.: +1-701-231-7265; Fax: +1-701-231-1008

Academic Editor: Kim L. Pickering Received: 31 March 2016; Accepted: 13 May 2016; Published: 19 May 2016 Abstract: In this study, six combinations of flax, hemp, and glass fiber were investigated for a hybrid reinforcement system in a polyurethane (PU) composite. The natural fibers were combined with glass fibers in a PU composite in order to achieve a better mechanical reinforcement in the composite material. The effect of fiber hybridization in PU composites was evaluated through physical and mechanical properties such as water absorption (WA), specific gravity (SG), coefficient of linear thermal expansion (CLTE), flexural and compression properties, and hardness. The mechanical properties of hybridized samples showed mixed trends compared to the unhybridized samples, but hybridization with glass fiber reduced water absorption by 37% and 43% for flax and hemp-agave PU composites respectively.

Keywords: natural ﬁber composites; ﬂax; hemp-agave; glass; polyurethane; hybridization

1. Introduction In the last few decades there has been a greater thrust toward curbing our reliance on non-renewable resources and finding viable options to decrease the burden on the environment. With increasing environmental regulations, the development of new materials that utilize a larger proportion of renewable and “green” raw materials has gained momentum. One such emerging area in the composites world is known as ‘biocomposites’ and it is finding its niche applications in building, automobile, and packaging industries [1]. Biocomposites contain one or more bio-based raw materials such as natural fibers. Natural fibers typically have high flexibility, low cost, ease of processing, biodegradability, and recyclability [2]. Several natural fibers such as coir [3], flax [4], hemp [5], bamboo [6], kenaf [7], abaca [8], and jute [9] have proven to be effective as reinforcements in composites. These biocomposites are typically used in non-structural applications because of their relatively low strength properties. Polyurethanes (PUs) are a diverse class of polymers obtained from the reaction of polyols with isocyanates which give a highly crosslinked structure. They have successfully found applications in construction, automotive, and medical industries. The successful applications of PUs are credited to their ability to be produced as flexible to rigid structures [10]. The unique properties of PUs include their ease of processing, low viscosity, quick reaction time, no emission of volatile organic compounds, better economics, and excellent bonding with different fibers without the need of specific sizing of the fibers [11]. In spite of several advantages associated with the natural fibers, their inherent structure is a bottleneck to their successful application in polymer composites. While the natural fibers are polar

Materials 2016, 9, 390; doi:10.3390/ma9050390 www.mdpi.com/journal/materials Materials 2016, 9, 390 2 of 15 and hydrophilic in nature because of the presence of hydroxyl groups in cellulose, the polymers are non-polar and hydrophobic in nature. This incompatibility severely affects the bonding between these two groups of materials. The poor adhesion between a natural fiber and a polymer matrix affects the mechanical properties of the material, and leads to high affinity to water and poor dimensional stability [12]. However, hybridization of different fibers in a composite can offer a balanced solution to enhancing the mechanical properties by utilizing other strong fibers such as glass fiber. A hybrid polymer composite usually utilizes two or more types of fibers with a single polymer matrix or a single type of fiber with two or more polymer matrices. The hybridization of fibers provides a unique opportunity to design a composite material in which the shortcomings in one fiber are compensated by the other fiber [13]. The improvement in water absorption properties in hybrid epoxy composites with jute and glass fiber reinforcement compared to jute-epoxy composites and glass-epoxy composites [14] is an example. Similarly, impact strength of the polyester composite was improved by hybridizing glass fiber with oil palm fiber, and with increasing the number of layers of glass fiber and oil palm fiber [15]. Also, the hybridization of composites with two natural fibers, such as silk/cotton and jute/cotton, has been shown to improve the tensile properties with increases in fiber content [16] This paper presents a novel idea of combining natural fibers, such as flax and a hemp-agave mixture, with glass fiber in a polyurethane composite in order to improve the physico-mechanical properties of a natural fiber PU composite. To understand the impact of glass fiber hybridization with the natural fibers, both hybridized and unhybridized composite panels were manufactured and their physico-mechanical properties were compared.

2. Experimental Procedure

2.1. Fiber Mats The experiment involved manufacturing composite material samples with different types of fiber reinforcement. Different combinations of woven and non-woven glass fiber mats, flax fiber mats, and hemp-agave fiber mats were used as reinforcements. Fibers and fiber mats had distinctly different characteristics (Table1). Although woven glass mat had the highest aerial density, the flax mats were much thicker which made them more difficult to penetrate (Figure1).

Table 1. Characteristics of the ﬁber mats used in the study. Fiber length and thickness are measured for 15 ﬁbers from each mat.

Mat Fiber Average Fiber Average Fiber Arial Weight Type Thickness Source Mats Length (mm) Diameter (µm) of Mat (g/m2) (mm) Non-woven SpaceAge Synthetics, Glass 37.08 40 450 1.25 (spunlaid) Fargo, ND, USA Woven SpaceAge Synthetics, Glass 304.80 100 600 1.50 (bi-axial) Fargo, ND, USA Non-woven Hemp-Flax, Flax1 (needle 63.25 20 500 4.00 Groningen, punched) Netherlands Non-woven Composite Innovation Flax2 (needle 77.56 40 550 5.50 Centre, Winnipeg, punched) MB, Canada Non-woven Composite Innovation Hemp- (needle 71.51 200 352 5.40 Centre, Winnipeg, Agave punched) MB, Canada Materials 2016, 9, 390 3 of 15 Materials 2016, 9, 390 3 of 15

(a) (b)

(e)

Figure 1.1. DifferentDifferent fiber fiber mats mats used used in thein study.the study. (a) non-woven (a) non-woven glass mat;glass ( bmat;) woven (b) glasswoven mat; glass (c) flax1;mat; ((dc)) flax1; flax2; ( de)) hemp-agave.flax2; (e) hemp-agave.

2.2. PU Resin The matrix plays a crucial role in transferring stressstress through interfacial layer to fiber, fiber, which is the main load bearing element element during during various various applications. applications. The The matrix matrix for for the the composite composite included included a amixture mixture of ofpolyol polyol resin, resin, ELASTOPOR ELASTOPOR®P ®12883RP 12883R RESIN RESIN (BASF (BASF Corporation, Corporation, Southfield, Southfield, MI, USA) MI, USA) and anda cross a cross linking linking agent agent ELASTOPOR ELASTOPOR®P 1230®P ISOCYANATE 1230 ISOCYANATE (BASF (BASF Corporation, Corporation, Southfield, Southfield, MI, USA), MI, USA),which whichyield yielda foamed a foamed based based polymeric polymeric methylen methylenee diphenyl diphenyl diisocyanate diisocyanate (MDI) (MDI) based based rigid polyurethane (PU)(PU) systemsystem withwith waterwater asas thethe sole blowing agent. The mix ratio of polyol to isocyanate used was 7777 polyol/100polyol/100 isocyanate isocyanate measured measured by by weight. The The density of the mixture before foaming was 1.15 g/cmg/cm33. .

2.3. Sample Preparation All the samples were made from PU resin and had an inner filterfilter core made of glass in addition to thethe differentdifferent typestypes ofof fiberfiber matsmats onon bothboth sides.sides. TheThe hybridizedhybridized compositecomposite panels were mademade withwith two layers of fiberfiber matsmats onon eacheach sideside ofof thethe filterfilter asas shownshown inin FigureFigure2 2.. In In contrast, contrast, thethe unhybridizedunhybridized composite panels were constructed similarly to the hybridized composite panels but without the woven mat layers. The number of fiber mat layers, fiber orientation, and formulation can be adjusted

Materials 2016, 9, 390 4 of 15 composite panels were constructed similarly to the hybridized composite panels but without the wovenMaterials mat layers.2016, 9, 390 The number of fiber mat layers, fiber orientation, and formulation can be adjusted4 of 15 to achieve specific thickness, density, and strength properties of the resulting composite material. to achieve specific thickness, density, and strength properties of the resulting composite material. These types of composites with glass as the woven and non-woven fiber reinforcement are currently These types of composites with glass as the woven and non-woven fiber reinforcement are currently availableavailable in the in marketthe market for for applications applications such such as as flooring flooring andand cabinetry.cabinetry.

Figure 2. The hand lay-up of different layers of mats from bottom to top used in the polyurethane Figure 2. The hand lay-up of different layers of mats from bottom to top used in the composite. polyurethane composite. The mass ratio of PU matrices varied from 62% to 77% between the different composites (Table 2). The mass ratioratio ofof PUthe matricesmatrix was varied determined from 62% after to 77%preliminary between trials the different were conducted composites to obtain (Table 2). The masscomplete ratio coverage of the matrix without was any determined runoff of the after PU preliminary resin. First, the trials mold were was conducted polished and to obtain then a completemold- coveragereleasing without agent any was runoff applied of theon the PU surface resin. First,before the placing mold the was mat polished lay-up andinto thenit. Second, a mold-releasing the mat agentlayers was of applied non-woven on the fiber, surface woven before fiber, placingand filter the were mat placed lay-up sequentially into it.Second, into the mold. the mat Next, layers the of non-wovenresin and fiber, the wovencross linking fiber, agent and filter for the were PU placedmatrix were sequentially mixed in into a cup the using mold. a mechanical Next, the resin stirrer and for the crossten linking seconds agent and forthen the poured PU matrix onto the were filter mixed layer. Finally, in a cup the using top two a mechanical layers of woven stirrer and for non-woven ten seconds and thenfiber pouredmats were onto quickly the filter placed layer. onto Finally,the PU matrix the top and two covered layers by of steel woven plates and before non-woven the crosslinking fiber mats werereaction quickly started placed to onto spread the out PU rapidly. matrix andNote covered the whole by arrangement steel plates of before the mat the lay-up crosslinking was covered reaction by a Teflon sheet (on the top and bottom) to prevent the panel from sticking to the steel plates. The started to spread out rapidly. Note the whole arrangement of the mat lay-up was covered by a Teflon mold was then placed on the lower movable platen of a hot press (Model 4122, Carver Inc., Wabash, sheet (on the top and bottom) to prevent the panel from sticking to the steel plates. The mold was then IN, USA) set at 65 °C. The material was then pressed at 0.29 MPa pressure for 20 min. To ensure placedcomplete on the curing, lower movablethe composite platen panels of a hotwere press post cured (Model at 4122,room temperature Carver Inc., for Wabash, 15 min. IN, USA) set at ˝ 65 C. TheA total material of six was different then set pressed of samples at 0.29 were MPa manufactured pressure for in the 20 lab min. (Table To ensure 2) with completedifferent types curing, the compositeof fiber reinforcement panels were that post resulted cured in at different room temperature amounts of forsubstrate. 15 min. The proportion of fiber, resin, and cross linking agent used for each type of composite sample was optimized after preliminary trials Tableaimed 2.atComposition obtaining evenly of the wet fiber panels reinforced without composite any leak material of PU resin samples outside manufactured the mold. inSample the lab. FG Commercial(flax1 and glass Glass fiber) Fiber (G’G’);was made Glass in Fiber a 203 (GG); × 203 Flax1 × 19 and mm Glass3 mold Fiber because (FG); Hemp-Agave of the relatively and Glass poor Fiberpenetrability (HG); Only of the Flax2 flax Fiber fiber (F); mat Only in 317.5 Hemp-Agave × 317.5 × Fiber25.4 mm (H).3 mold. Sample G’G’ (commercial glass fiber) was a commercially available material and was obtained in the size of 254 × 254 × 25.4 mm3. Sample Non-Woven PU Resin 3 Panel Density Woven Mat Filter Mold Size (mm ) 3 NameTable 2. CompositionMat of the fiber reinforced composite(Weight material %) samples manufactured in the (g/cmlab. ) G’G’Commercial Glass Glass Fiber (G’G’); Glass Glass Fiber Glass (GG); Flax1 and 67–70 Glass Fiber (FG); 254 ˆ Hemp-Agave254 ˆ 25.4 and Glass 0.45 GGFiber (HG); Glass Only Flax2 Fiber Glass (F); Only Hemp-Agave Glass Fiber (H). 65 317.5 ˆ 317.5 ˆ 25.4 0.22 FG Flax1 Glass Glass 65 203.2 ˆ 203.2 ˆ 19.1 0.33 HG Hemp-Agave Glass Glass 62 317.5 ˆ 317.5 ˆ 25.4Panel 0.22 Sample Woven PU Resin Mold Size FNon-Woven Flax2 Mat -Filter Glass 77 317.5 ˆ 317.5 ˆ 25.4Density 0.20 Name Mat (Weight %) (mm3) H Hemp-Agave - Glass 77 317.5 ˆ 317.5 ˆ 25.4(g/cm 0.163) G’G’ Glass Glass Glass 67–70 254 × 254 × 25.4 0.45 A totalGG of six differentGlass set of samplesGlass wereGlass manufactured 65 in the 317.5 lab × (Table317.5 ×2 25.4) with different 0.22 types FG Flax1 Glass Glass 65 203.2 × 203.2 × 19.1 0.33 of fiber reinforcement that resulted in different amounts of substrate. The proportion of fiber, resin, HG Hemp-Agave Glass Glass 62 317.5 × 317.5 × 25.4 0.22 and cross linkingF agentFlax2 used for each- type ofGlass composite sample77 was317. optimized5 × 317.5 × after25.4 preliminary 0.20 trials aimed at obtainingH Hemp-Agave evenly wet panels - without Glass any leak of 77 PU resin outside 317.5 × 317.5 the × mold. 25.4 Sample 0.16 FG (flax1

Materials 2016, 9, 390 5 of 15 and glass fiber) was made in a 203 ˆ 203 ˆ 19 mm3 mold because of the relatively poor penetrability of the flax fiber mat in 317.5 ˆ 317.5 ˆ 25.4 mm3 mold. Sample G’G’ (commercial glass fiber) was a commercially available material and was obtained in the size of 254 ˆ 254 ˆ 25.4 mm3.

2.4. Testing and Measurements The samples manufactured in the lab were tested for various mechanical and physical properties following speciﬁc ASTM standards. The details of such properties tested are provided below:

2.4.1. Void Content of Composites The void content in all the composite samples was determined according to the ASTM D2734 [17]. The densities of the ﬁbers and PU resin and their respective weight fractions used for different composite panels were used to determine the theoretical densities of the composite materials.

2.4.2. Water Absorption Test The water absorption of all composite samples was investigated in accordance with ASTM D570 [18]. Five samples per formulation were tested for percentage weight gain when submerged in water. The testing was performed on sample coupons of 76.2 mm length and 25.4 mm width. Composite samples were ﬁrst dried in an oven at 50 ˝C for 24 h, and weighed to record the dry weight. The samples were then immersed in a water bath set at room temperature and the sample weight gain was recorded every 24 h for the duration of 27 days. The percentage weight gain each day was calculated as the incremental water absorption.

2.4.3. Coefficient of Linear Thermal Expansion (CLTE) The CLTE was measured for five samples from each formulation in accordance with ASTMD 6341 [19]. All the samples were conditioned sequentially, first in a freezer set at ´23.4 ˝C, then at room temperature (23.5 ˝C), and finally in an oven set at 60 ˝C. For each of these three conditions, the samples were exposed for 48 h. At the end of each 48 h period, the length of the samples was recorded. The CLTE was measured as the slope of the curve of sample length versus temperature.

2.4.4. Flexural Properties The flexural properties of the samples were tested according to the ASTM D790 [20], which is a three point flexural test method for unreinforced and reinforced plastics. Due to the limitations on the mold size, a span-to-depth ratio different than the 16:1 recommended in the standard was used to determine modulus of elasticity (MOE) and modulus of rupture (MOR). Instead, a span-to-depth ratio of 10.5:1 was used for samples GG (Glass Fiber), HG (Hemp-Agave and Glass Fiber), F (Flax2), and H (Only Hemp-Agave Fiber). For the commercial panels G’G’ (Commercial Glass Fiber), and flax1 samples, an 8:1 and 6.4:1 span-to-depth ratio were used, respectively, because of their smaller sample sizes. The cross head motion rate was calculated according to the span length, width, and thickness of the samples as specified in the standard. A universal testing machine (312 Frame, Test Resources, Shakopee, MN, USA) was used for testing the materials.

2.4.5. Compression Properties All the samples were tested for compression properties in accordance with ASTM D695 [21]. All the specimens were cut as speciﬁed in the standard. The cross head motion rate of 1.3 mm/min was used. The compressive strength and modulus of elasticity were calculated from the stress-strain data recorded by the universal testing machine (312 Frame, Test Resources, Shakopee, MN, USA). Materials 2016, 9, 390 6 of 15

2.4.6. Hardness The surface hardness of the samples was tested in accordance with ASTM D1037 [22] using the universal testing machine (312 Frame, Test Resources, Shakopee, MN, USA). The modiﬁed Janka-ball test method used a ball with an 11.3 mm in diameter. The specimen size of 76.2 ˆ 152.4 ˆ 25.4 mm3 was penetrated by the “ball” until it reached half of its diameter and the load value was recorded at that point. Two penetrations each were made on both sides on the ﬂat faces of the panel. A cross head motion rate of 6 mm/min was used.

2.5. Light Microscope Images The samples tested during ﬂexural bending were analyzed using stereo light microscope SZM 7045 (AmScope, Irvine, CA, USA) to study their morphology. The images of the fracture surface of the samples were collected at a magniﬁcation of 25ˆ.

2.6. Statistical Analysis In order to establish a comparison of means between treatments, ANOVA tests were performed. Two sample t-tests were performed between all formulations to compare treatment means for the various properties tested. Both ANOVA and two sample t-tests were performed with the software Minitab 17 (Minitab Inc., Pennsylvania State University, Central County, PA, USA). The error bar in the graphs is the standard deviation of the mean for the dataset. The strength properties obtained by hardness, ﬂexural, and compression tests were ﬁrst normalized by the respective panel densities before doing statistical analysis as material density can affect strength properties of the material, and different formulations showed large variability in material density.

3. Results and Discussion The sample manufacturing process was difficult for the samples containing flax or hemp-agave mat. The natural fiber mats interfered with the penetration and distribution of the PU resin. Further, the resin-crosslinking mix did not spread uniformly to wet and encapsulate the fibers, particularly in large molds, leaving the outermost layers of the natural fibers partially exposed. This was one of the reasons for using the smaller molds. Additionally, flax and hemp-agave mats were not available in preferred specification including lower density and thickness. Although flax is grown locally for grain and the crop residue left in the field is a good source of strong fibers, there are no flax fiber processing industries in the US.

3.1. Void Content The presence of voids in a composite material leads to a reduction in its mechanical and physical properties. The formation of voids can be attributed to the trapped air or other volatiles present in the composite materials [23]. The void content (%) of composite samples made with different combinations of fibers is presented in Table3. The highest amount of void (7.9%) was exhibited by the flax alone composite (F) showing the incompatibility between the PU resin and the flax2 mat. However, when the flax1 mat was hybridized with the glass fiber mat, a void content of 4.8% was observed, which depicts a better interaction of the flax1 fiber mat with the PU resin than the flax2 mat. The lowest void content of 4.1% was observed for the hemp alone (H) composite which increased to 4.8% when hybridized with the glass fiber (HG). This increase in void content shows that when the hemp fiber mat was used as the outermost layer, the resin couldn’t interact with the hemp skin layers as well as it did in the hemp alone samples. The void content in the glass fiber composite (GG) was 4.8% whereas the commercial glass fiber composites (G’G’) showed only 1.4% void content. Materials 2016, 9, 390 7 of 15

Table 3. The void content of ﬂax, hemp-agave, and glass ﬁber reinforced composites.

Type of Composites Void Content (%) Commercial Glass Fiber (G’G’) 1.4 Glass Fiber (GG) 4.8 Flax1 and Glass Fiber (FG) 4.8 Hemp-Agave and Glass Fiber (HG) 5.2 Only Flax2 Fiber (F) 7.9 Only Hemp-Agave Fiber (H) 4.1

3.2. Water Absorption

MaterialsWater 2016 absorption, 9, 390 in composites is important to study as it may cause debonding, disruption7 of 15 of integral structure, and loss of strength [24], particularly in applications where the material may be exposedexposed to water. All All composite composite samples samples containi containingng natural natural fibers fibers show showeded much much higher higher water water absorptionabsorption than than those those containing containing onlyonly glassglass fibers fibers throughout throughout the the data data collection collection period period (Figure (Figure 3). 3). TheThe initial initial rate rate of of water water absorption absorption waswas higherhigher for for unhybridized unhybridized natural natural fiber fiber composite composite samples. samples. HempHemp composite composite samples samples (H) (H) showed showed thethe highest mo moistureisture gain gain of of 76%, 76%, followed followed by flax by flax composite composite samples (F) with 60% moisture gain. The natural fibers are inherently hydrophilic and using the flax samples (F) with 60% moisture gain. The natural fibers are inherently hydrophilic and using the and hemp fibers as the skin layers made them highly susceptible to the moisture absorption. Natural flax and hemp fibers as the skin layers made them highly susceptible to the moisture absorption. fibers swell when they absorb water, which can lead to micro-cracking of the material, further Naturalhampering fibers swellthe integrity when theyof the absorb interfacial water, surface. which The can swelling lead to micro-crackingstresses and dimensional of the material, instability further hamperingof the composite the integrity when of thethe interfacial natural fiber surface. absorbs The wa swellingter will stresses subsequently and dimensional lead to failure instability of the of the compositecomposites when [25]. the When natural hybridized fiber absorbs with water glass willfiber, subsequently a drastic drop lead of to 43% failure and of 37% the in composites moisture [25]. Whenabsorption hybridized was with observed glass fiber,for composites a drastic dropcontaining of 43% hemp-agave and 37% in and moisture flax, respectively. absorption wasThis observeddrop in for compositesmoisture containing uptake was hemp-agave expected because and flax,of the respectively. high resistance This ofdrop glass infibers moisture to moisture. uptake The was lowest expected becausemoisture of the absorption high resistance of 12% ofand glass 2% were fibers observed to moisture. in unhybridized The lowest moistureglass fiber absorption composite samples of 12% and of 2% wereGG observed and G’G’, in respectively. unhybridized glass fiber composite samples of GG and G’G’, respectively.

90 80 70 60 50 40 30 Moisture gain (%) 20 10 0 0 5 10 15 20 25 30 Time (h1/2)

G'G' GG FG HG F H

Figure 3. Water absorption of composite samples at room temperature plotted against square root of Figure 3. Water absorption of composite samples at room temperature plotted against square root of time for composite samples containing commercial sample with glass fiber (G’G’), glass fiber (GG), time for composite samples containing commercial sample with glass fiber (G’G’), glass fiber (GG), flaxflax and and glass glass fiber fiber (FG), (FG), hemp hemp and and glass glass fiberfiber (HG), flax flax alone alone (F), (F), and and hemp hemp alone alone (H). (H).

3.3.3.3. Coefficient Coefficient of Linearof Linear Thermal Thermal Expansion Expansion (CLTE)(CLTE) CompositeComposite samples samples containing containing onlyonly natural fiber fiber reinforcements reinforcements such such as flax as flax and and hemp hemp fibers fibers showedshowed negative negative CLTE CLTE while while all all other other samples samples showed showed positivepositive valuesvalues (Figure 44).). This This is is expected expected as as natural fibers are expected to shrink under elevated temperature while glass fibers are expected to expand. The highest absolute value of CLTE was observed for samples containing only glass fibers (G’G’) and samples containing only flax fiber reinforcement (F). It is notable that the hybrid composites showed the lowest CLTE, which is preferred in applications such as flooring and cabinetry, or where the material will be exposed by large temperature fluctuations. A low CLTE would ensure high dimensional stability when exposed to temperature extremes.

Materials 2016, 9, 390 8 of 15 natural fibers are expected to shrink under elevated temperature while glass fibers are expected to expand. The highest absolute value of CLTE was observed for samples containing only glass fibers (G’G’) and samples containing only flax fiber reinforcement (F). It is notable that the hybrid composites showed the lowest CLTE, which is preferred in applications such as flooring and cabinetry, or where Materials 2016, 9, 390 8 of 15 the material will be exposed by large temperature fluctuations. A low CLTE would ensure high dimensionalMaterials 2016 stability, 9, 390 when exposed to temperature extremes. 8 of 15 15

15 10 10 5

cm/(cm·°C)] 5 6 − 0 cm/(cm·°C)] 6 − 0 G'G' GG FG HG F H -5 G'G' GG FG HG F H -5

expansion [10 -10 Coefficient of Coefficient linear thermal

expansion [10 -10 Coefficient of Coefficient linear thermal -15 -15 Samples Samples Figure 4. Coefficient of thermal linear expansion for composite samples with commercial glass fiber Figure 4. Coefficient of thermal linear expansion for composite samples with commercial glass fiber aloneFigure (G’G’), 4. Coefficient glass fiber of thermal(GG), flax linear and expansion glass fiber for (FG), composite hemp-agave samples and with glass commercial fiber (HG), glass flax fiber fiber alone (G’G’), glass fiber (GG), flax and glass fiber (FG), hemp-agave and glass fiber (HG), flax fiber alonealone (F), (G’G’), and hempglass fiberfiber (GG),alone flax(H) andmeasured glass fiber with (FG), ASTM hemp-agave D6341. and glass fiber (HG), flax fiber alonealone (F), (F), and and hemp hemp fiber fiber alone alone (H) (H) measured measured with ASTM D6341. D6341. 3.4. Flexural Properties 3.4.3.4. Flexural Flexural Properties Properties DuringDuring the the flexural flexural test, test, no no samplesample brokebroke underunder the the recommended recommended strain strain value value of of5%. 5%. The The load load During the flexural test, no sample broke under the recommended strain value of 5%. The load vs.vs position. position curve curve depicts depicts the the change change inin thethe load valu valuee with with change change in in the the position position of ofthe the composites composites vs. position curve depicts the change in the load value with change in the position of the composites (Figure(Figure5). 5). It It can can be be observed observed that that the the commercial commercial glassglass samples behaved behaved much much better better than than other other samples(Figure under5). It can the beflexural observed load. that Initially, the commerc flax hybridizedial glass samplessamples behavedbehaved wellmuch but better this trendthan other didn’t samples under the flexural load. Initially, flax hybridized samples behaved well but this trend didn’t transcendsamples underinto higher the flexural specific load. flexural Initially, properti flax hybridizedes than glass samples fiber basedbehaved composite well but samplesthis trend (GG). didn’t transcendtranscend into into higher higher specific specific flexural flexural properties properties than glass glass fiber fiber based based composite composite samples samples (GG). (GG).

1800 1800 1600 1600 1400 1400 1200 1200 1000 1000 800

Load (N) 800

Load (N) 600 600 400 400 200200 00 024681012024681012 Position (mm)(mm)

G'G'G'G' GGGG FG HGHG FF HH

Figure 5. FigureFigure 5. Flexural5. Flexural Flexural load-position load-position load-position curvescurvescurves of the composites containingcontaining containing commercialcommercial commercial glass glass glass fiber fiber fiber (G’G’), (G’G’), (G’G’), glassglassglass fiber fiber fiber (GG), (GG), (GG), flax flax flax and and and glass glass glass fiberfiber fiber (FG), (FG),(FG), hemphemp and glass glass fiberfiber fiber (HG),(HG), (HG), flax flax flax alone alone alone (F) (F) (F)and and and hemp hemp hemp alone alone alone (H). (H). (H).

TheThe flexural flexural MOE MOE and and MORMOR valuesvalues werewere calculated asas specifiedspecified by by the the ASTM ASTM D790 D790 standard, standard, whichwhich were were further further normalized normalized byby thethe respective densities ofof thethe panelspanels to to eliminate eliminate the the effect effect of of densitydensity across across all all thethe formulations.formulations. BothBoth types of glass fiberfiber reinforcedreinforced unhybridizedunhybridized composite composite samplessamples (G’G’ (G’G’ and and GG) GG) exhibitedexhibited thethe highesthighest specificic flexuralflexural modulusmodulus compared compared to to both both natural natural fiberfiber reinforced reinforced samples samples (Figure (Figure 6).6). TheThe formationformation of anan interphaseinterphase region region between between a aglass glass fiber fiber and and

Materials 2016, 9, 390 9 of 15

The flexural MOE and MOR values were calculated as specified by the ASTM D790 standard, which were further normalized by the respective densities of the panels to eliminate the effect of density acrossMaterials all the 2016 formulations., 9, 390 Both types of glass fiber reinforced unhybridized composite samples9 of 15 (G’G’ Materials 2016, 9, 390 9 of 15 and GG) exhibited the highest specific flexural modulus compared to both natural fiber reinforced a PU matrix allows for the better dispersion and proper wetting of fibers into the matrix leading to samples (Figure6). The formation of an interphase region between a glass fiber and a PU matrix allows for bettera PU matrixmechanical allows properties for the better of the dispersion composites and [26]. prop Theer additionwetting ofof fibersglass fiberinto theresulted matrix in leading lowering to the betterofbetter modulus dispersionmechanical in hybridized and properties proper flax of wettingcomposite the composites of fiberssamples [26]. into by The the42%, matrixaddition whereas leading of the glass hybridization to fiber better resulted mechanical of inhemp lowering fiber properties of theresultedof compositesmodulus in a in55% [hybridized26 improvement]. The addition flax compositein modulus of glass samples (p fiber < 0.05), resultedby when 42%, compared whereas in lowering the with hybridization of just modulus hemp fiber of in hemp hybridizedreinforced fiber flax compositecompositeresulted samples in samples. a 55% by improvement 42%, whereas in modulus the hybridization (p < 0.05), when of hemp compared fiber with resulted just hemp in a 55% fiber improvement reinforced in moduluscomposite (p < 0.05), samples. when compared with just hemp fiber reinforced composite samples. 4500 4500 A 4000 A 4000 3500 A A

)] 3500 3 3000 )]

3 3000 2500 B 2500 B 2000 C 2000 C 1500 C [MPa/(g·cm 1500 C D [MPa/(g·cm 1000 1000 D 500 Specific Flexural Stiffness Specific Flexural 500 Specific Flexural Stiffness Specific Flexural 0 0 G'G' GG FG HG F H G'G' GG FG HG F H Samples Samples

FigureFigure 6. Specific 6. Specific flexural flexural stiffness stiffnessresults results for composite samples samples containing containing commercial commercial GlassGlass fiber fiber Figure 6. Specific flexural stiffness results for composite samples containing commercial Glass fiber alonealone (G’G’), (G’G’), Glass Glass fiber fiber alone alone (GG), (GG), Flax Flax and and Glass Glas fibers fiber (FG), (FG), Hemp-agave Hemp-agave and Glass fibers fibers (HG), (HG), Flax Flaxalone fiber (G’G’), alone Glass (F), andfiber Hemp alone fiber(GG), alone Flax (H).and Glass fiber (FG), Hemp-agave and Glass fibers (HG), fiberFlax alone fiber (F), alone and (F), Hemp and fiberHemp alone fiber alone (H). (H). The highest specific flexural strength (flexural strength normalized by material density) of 64.81 The highest specific flexural strength (flexural strength normalized by material density) of MPa/g/cmThe highest3 was observed specific flexural for G’G’ strength samples (flexural (Figure 7).strength The GG normalized samples showed by mate arial decrease density) of of 51% 64.81 in 3 3 64.81specificMPa/g/cm MPa/g/cm flexural was wasobservedstrength observed forcompared G’G’ for samples G’G’ to G’G’ samples (Figure samp les. (Figure7). TheThe GG 7high). samples The strength GG showed samples of G’G’ a decrease showed samples of amay decrease51% be in of 51%creditedspecific in specific flexuralto better flexural strengthmanufacturing strength compared compared processes, to G’G’ to such G’G’ samp as samples.les. even The mixing high The and highstrength distribution strength of G’G’ of of G’G’samples resinsamples and may cross be may be creditedlinkingcredited tobetter agent, to better manufacturing and manufacturing better encapsulation processes, processes, such of such fibers as evenas evenin mixing the mixing commercial and and distribution distribution sample. of ofAll resin resin natural and and cross crossfiber linking agent,reinforcedlinking and better agent, composite encapsulation and better samples encapsulation of showed fibers in significantly the of commercial fibers lowerin the sample. ( pcommercial < 0.05) All specific natural sample. flexural fiber All reinforced strengthnatural compositethanfiber reinforced composite samples showed significantly lower (p < 0.05) specific flexural strength than samplesG’G’ showedand GG samples. significantly There lower was an (p improvement< 0.05) specific of 67% flexural in specific strength flexural than strength G’G’ observed and GG for samples. G’G’ and GG samples. There was an improvement of 67% in specific flexural strength observed for Therehybridized was an improvement hemp samples of whereas 67% in hybrid specific flax flexural samples strength did not observedresult in significant for hybridized change hemp(p > 0.05) samples forhybridized specific flexuralhemp samples strength. whereas hybrid flax samples did not result in significant change (p > 0.05) whereasfor specific hybrid flexural flax samples strength. did not result in significant change (p > 0.05) for specific flexural strength.

80 A 80 70 A 70 60 )]

3 60

)] 50 3 50 B 40 B 40 30 C,D C C [MPa/(g·cm 30 C,D D 20 [MPa/(g·cm D E 20 E 10 Specific Flexural Strength Specific Flexural 10 Specific Flexural Strength Specific Flexural 0 0 G'G' GG FG HG F H G'G' GG FG HG F H Samples Samples

Figure 7. Specific flexural strength for composite samples containing commercial Glass fiber alone Figure 7. Specific flexural strength for composite samples containing commercial Glass fiber alone (G’G’),(G’G’),Figure Glass Glass7. fiberSpecific fiber alone flexuralalone (GG), (GG), strength Flax Flax and forand Glass composite Glass fiber fibe (FG),samplesr (FG), Hemp-agave Hemp-agave containing commercial andand GlassGlass fibersfibersGlass (HG), fiber alone Flax Flax fiber alonefiber(G’G’), (F), alone and Glass Hemp(F), fiber andfiber Hempalone alone (GG), fiber(H). aloneFlax and(H). Glass fiber (FG), Hemp-agave and Glass fibers (HG), Flax fiber alone (F), and Hemp fiber alone (H).

Materials 2016, 9, 390 10 of 15

Materials 2016, 9, 390 10 of 15 The flexural fracture behavior of samples depends mainly on the tensile strength of the material as the initiationThe flexural of cracking fracture alwaysbehavior occur of samples on the depends tension mainly side, whichon the tensile then travels strength across of the the material width to as the initiation of cracking always occur on the tension side, which then travels across the width to the compression side until the sample fails. To understand the fracture behavior, all the samples were the compression side until the sample fails. To understand the fracture behavior, all the samples were flexurally loaded until they failed. Of all the fiber reinforced samples tested, only the commercial glass flexurally loaded until they failed. Of all the fiber reinforced samples tested, only the commercial fiberglass reinforced fiber reinforced sample brokesample apart broke completely apart completely (Figure (Figure8). 8).

Figure 8. Images of composite samples tested for flexural properties showing the failure surface. Figure 8. Images of composite samples tested for flexural properties showing the failure surface. (a) commercial glass (G’G); (b) glass (GG); (c) hybrid flax and glass (FG); (d) hybrid hemp and glass (a) commercial glass (G’G); (b) glass (GG); (c) hybrid flax and glass (FG); (d) hybrid hemp and glass (HG); (e) only flax (F); and (f) only hemp (H). (HG); (e) only flax (F); and (f) only hemp (H). In the commercial samples, the outermost non-woven glass fiber layer on the tension side was wellIn the dispersed commercial in the samples,PU matrix theas no outermost fiber pull non-wovenout was observed glass (Figure fiber layer 9a,a1). on When the tensionthese samples side was wellwere dispersed flexurally in the loaded, PU matrix the outermost as no fiber non-woven pull out was glass observed fiber bears (Figure the 9loada,a 1up). When to its thesemaximum samples werecapacity flexurally and loaded, then transfers the outermost this load non-woven to the wove glassn glass fiber fiber. bears However, the load the up woven to its maximum glass fiber capacitywas andfound then transfers to be poorly this mixed load towith the the woven matrix glass which fiber. was However,apparent from the woventhe clean glass long fiberglass wasfibers found pulled to be poorlyout mixedwith no with adherence the matrix to the which matrix. was On the apparent other hand, from the the glass clean reinforc long glassed samples fibers manufactured pulled out with no adherencein our lab did to thenot matrix.break completely On the other though hand, all the the commercial glass reinforced samples samplesdid. Also, manufactured the mechanism inof our lab didcrack not propagation break completely in GG samples though observed all the commercialwas different samplesthan in the did. commercial Also, the samples. mechanism The crack of crack initiated from the tension side and the outermost layer of non-woven glass fiber failed after reaching propagation in GG samples observed was different than in the commercial samples. The crack initiated its maximum capacity. The crack propagated through the non-woven glass fiber to the next interface from the tension side and the outermost layer of non-woven glass fiber failed after reaching its maximum of woven glass and the PU matrix (Figure 9b,b1). The woven glass fiber was found to be strong capacity.enough The to crack prevent propagated the crack through from transcending the non-woven through glass it; fiberrather, to the the crack next interfacewas found of to woven be running glass and the PUalong matrix the surface (Figure due9b,b to1). lack The of woven bonding glass between fiber was the foundtwo layers. to be During strong enoughthis time, to the prevent core of the the crack fromglass transcending filter, being through weaker it; in rather, strength the than crack woven was foundglass fiber, to be initiated running a alongcrack thedue surfaceto the continuous due to lack of bondingincreasing between load the from two the layers. compression During side. this time,The fa theilure core of the of the sample glass was filter, then being continued weaker from in strength the thancore woven towards glass the fiber, compression initiated a side. crack due to the continuous increasing load from the compression side. The failureThe of crack the sample propagation was then during continued flexural from bending the core of flax towards and hemp the compression fiber and their side. hybridized samplesThe crack is shown propagation in Figure during 9c–f,c1–f flexural1. The crack bending initiated of in flax all andof these hemp samples fiber on and the their tension hybridized side. In flax fiber reinforced samples (F), the fiber pull out from the outermost layer was the origin of the samples is shown in Figure9c–f,c 1–f1. The crack initiated in all of these samples on the tension side. In flaxfailure fiber mode. reinforced Non-woven samples natural (F), fiber thefiber mats pullconsisted out from of fibers the of outermost different lengths layer waswhich the can origin explain of the the bigger fiber pull outs in the image. For flax and glass fiber reinforced samples (FG), a change in failure mode. Non-woven natural fiber mats consisted of fibers of different lengths which can explain direction of the propagation of crack was observed with the increase in load value as was observed the bigger fiber pull outs in the image. For flax and glass fiber reinforced samples (FG), a change in in GG samples. The crack initiation started on the tension side but the flax fiber having high directionextensibility of the propagationdid not allow of propagation crack was observed through withit to thethe increasestronger inwoven load valueglass fiber. as was The observed crack in GG samples.appears to The run crackalong initiationthe flax and started woven on glass the fiber tension layer, side and but as thethe flaxload fiberkept on having increasing high extensibilityfrom the did notcompression allow propagation side, the crack through again it initiates to thestronger but from woventhe glass glass filter fiber. core The(it being crack weaker appears than to woven run along the flaxglass and fiber). woven The crack glass then fiber propagated layer, and from as the the load core keptupwards on increasing to the woven from glass the fiber compression and woven side, the crack again initiates but from the glass filter core (it being weaker than woven glass fiber). The crack then propagated from the core upwards to the woven glass fiber and woven layers, and failure of these Materials 2016, 9, 390 11 of 15 Materials 2016, 9, 390 11 of 15 upperlayers, layers and of failure fibers of appear. these upper The samelayers mechanismof fibers appear. of crack The same propagation mechanism appears of crack during propagation hemp and its hybridizedappears during samples. hemp and The its hemp hybridized fiber pull samples. out fromThe hemp matrix fiber was pull the out main from reason matrix forwas thethe failuremain of hempreason reinforced for the samples.failure of Thehemp crack reinforced initiated sample at thes. tensionThe crack side initiated and propagated at the tension through side theand core layerpropagated of the glass through filter with the nocore change layer inof propagationthe glass filter direction. with no However, change in when propagation hybridized direction. with glass fiber,However, the crack when initiation hybridized started with from gla thess fiber, tension the sidecrack but initiation it could started not propagate from the tension in the sameside but direction it could not propagate in the same direction as it encountered the woven glass fiber layer. The crack as it encountered the woven glass fiber layer. The crack appeared to change its direction and started appeared to change its direction and started propagating along the interface between woven glass propagating along the interface between woven glass fiber and hemp fiber. fiber and hemp fiber.

500 µm

(a) (a1)

500 µm

(b) (b1)

500 µm

(d) (d1)

Figure 9. Cont. Materials 2016, 9, 390 12 of 15 Materials 2016, 9, 390 12 of 15 Materials 2016, 9, 390 12 of 15

500 µm

500 µm (e) (e1)

(e) (e1)

500 µm 500 µm

(f) (f1) (f) (f1)

FigureFigure 9. Light 9. Light microscopy microscopy images images captured captured from from the the tensiontension sideside of of all all the the flexurally flexurally broken broken samples. samples. Figure 9. Light microscopy images captured from the tension side of all the flexurally broken samples. (a,a1)( acommercial,a1) commercial glass glass (G’G’); (G’G’); (b (,b,1b) 1glass) glass (GG); (GG); ( (cc,,cc11)) onlyonly flaxflax (F);(F); ( d(d,d,d1)1 )hybrid hybrid flax flax and and glass glass (FG); (FG); (a,a1) commercial glass (G’G’); (b,b1) glass (GG); (c,c1) only flax (F); (d,d1) hybrid flax and glass (FG); (e,e1)( eonly,e1) only hemp hemp (H); (H); and and (f, f(1f),f hybrid1) hybrid hemp hemp and and glass glass (HG).(HG). (e,e1) only hemp (H); and (f,f1) hybrid hemp and glass (HG). 3.5. Compressive3.5. Compressive Properties Properties 3.5. Compressive Properties The Thehighest highest specific specific compressive compressive modulu moduluss value value observed was was 1382.82 1382.82 MPa/(g·cm MPa/(g·cm3) for3) for the the G’G’ G’G’ The highest specific compressive modulus value observed was 1382.82 MPa/(g¨ cm3) for the G’G’ sample.sample. On Onthe theother other hand, hand, the the specific specific compress compressiveive modulusmodulus forfor GG GG samples samples was was not not statistically statistically sample.different On the(p > other0.05) from hand, FG, the HG, specific and F compressivesamples (Figure modulus 10). The for glass GG fiber samples in the wasFG sample not statistically has a different (p > 0.05) from FG, HG, and F samples (Figure 10). The glass fiber in the FG sample has a differenthigher ( pmodulus> 0.05) fromand strength FG, HG, than and the F samplesflax fiber, (Figurebut when 10 ).the The tension glass force fiber goes in thebeyond FG sample a certain has a higher modulus and strength than the flax fiber, but when the tension force goes beyond a certain higherpoint modulus the glass and fiber strength breaks thanfirst since the flax it has fiber, lesser but extensibility when the tension than the force natural goes fibers. beyond This a leads certain to pointa point the glass fiber breaks first since it has lesser extensibility than the natural fibers. This leads to a the glassstress fiberconcentration breaks first resulting since in it hasa failure lesser of extensibility stress transfer than to the the flax natural fiber causing fibers. Thisthe composites leads to a to stress stress concentration resulting in a failure of stress transfer to the flax fiber causing the composites to concentrationfail. Figure 11 resulting shows the in results a failure from ofthe stress specific transfer compressive to the strength flax fiber of all causing the composite the composites samples tested to fail. fail. Figure 11 shows the results from the specific compressive strength of all the composite samples tested3 Figurein the 11 study.shows The the highest results compressive from the specific strength compressive was observed strength for G’G’ of allsamples the composite at 24.43 MPa/(g·cm samples tested). in the study. The highest compressive strength was observed for G’G’ samples at 24.43 MPa/(g·cm3). in theThere study. was The no significant highest compressive difference (p strength > 0.05) observed was observed between for natura G’G’l samplesfiber reinforced at 24.43 samples. MPa/(g ¨ cm3). There was no significant difference (p > 0.05) observed between natural fiber reinforced samples. There was no significant difference (p > 0.05) observed between natural fiber reinforced samples. 1800 A 18001600 A 16001400 1200 1400)] B,C 3 1000 1200 B B )] B,C B,C 3 800 1000 600 B B C 800 B,C 400 600[MPa/(g·cm C 200 400 [MPa/(g·cm 0 200 G'G' GG FG HG F H Specific Compressive Stiffness Specific Compressive 0 Samples G'G' GG FG HG F H Specific Compressive Stiffness Specific Compressive Figure 10. Specific compressive stiffness for compositeSamples samples containing commercial Glass fiber alone (G’G’), Glass fiber alone (GG), Flax and Glass fiber (FG), Hemp-agave and Glass fibers (HG), FigureFlax 10. fiberSpecific alone (F),compressive and Hemp fiber stiffness alone for (H). composite samples containing commercial Glass fiber Figure 10. Specific compressive stiffness for composite samples containing commercial Glass fiber alone (G’G’), Glass fiber alone (GG), Flax and Glass fiber (FG), Hemp-agave and Glass fibers (HG), Flax alonefiber alone(G’G’), (F), Glass and fiber Hemp alone fiber (GG), alone Flax (H). and Glass fiber (FG), Hemp-agave and Glass fibers (HG), Flax fiber alone (F), and Hemp fiber alone (H).

Materials 2016, 9, 390 13 of 15 Materials 2016, 9, 390 13 of 15 Materials 2016, 9, 390 13 of 15 35 A 35 A 30 30 25 )] 3 25 )] A,B 3 20 20 A,B B,C B,C 15 C C C 15 C C C [MPa/(g·cm 10

[MPa/(g·cm 10 5 5 Specific Compressive Strength Specific Compressive 0 Specific Compressive Strength Specific Compressive 0 G'G' GG FG HG F H G'G' GG FGSamples HG F H Samples

FigureFigure 11. 11. MeasureMeasure of of specific specific compressive compressive strength strength obtained obtained for for composite composite samples samples containing containing commercialFigurecommercial 11. MeasureGlass Glass fiber fiber of alone alonespecific (G’G’), (G’G’), compressive Glass Glass fiber fiber strength al aloneone (GG), (GG),obtained Flax Flax forand and composite Glass Glass fiber fiber samples (FG), (FG), Hemp-agave Hemp-agave containing andcommercialand Glass Glass fibers fibers Glass (HG), (HG), fiber Flax Flaxalone fiber fiber (G’G’), alone alone Glass (F), (F), and andfiber Hemp Hemp alone fiber fiber(GG), alone Flax (H). and Glass fiber (FG), Hemp-agave and Glass fibers (HG), Flax fiber alone (F), and Hemp fiber alone (H). 3.6.3.6. Hardness Hardness Testing Testing 3.6. Hardness Testing TheThe commercial commercial glass glass based based samples samples G’G’ G’G’ showed showed the the highest highest specific specific ha hardnessrdness value value of of 6875.46 6875.46 The3 commercial glass based samples G’G’ showed the highest specific hardness value of 6875.46 N/(g·cmN/(g¨ cm),3 ),followed followed next next by by GG GG samples samples (Figure (Figure 12). 12). The The unhybridized unhybridized flax flax and and hemp-agave hemp-agave samples samples N/(g·cm3), followed next by GG samples (Figure 12). The unhybridized flax and hemp-agave samples showedshowed no no significant significant ( (pp >> 0.05) 0.05) difference difference between between each each other other with with the the lowest lowest value value for for hardness hardness showed no significant (p > 0.05) difference between each other with the lowest value for hardness amongamong all all the the samples tested. When When hybridized, hybridized, both both flax flax and and hemp-agave hemp-agave samples samples showed showed a a among all the samples tested. When hybridized, both flax and hemp-agave samples showed a significantsignificant increase increase ( (pp << 0.05) 0.05) in in the the hardness hardness property property due due to to the the better better resistance resistance provided provided by by the the significant increase (p < 0.05) in the hardness property due to the better resistance provided by the glassglass reinforcement. reinforcement. glass reinforcement. 8000 8000 A 7000 A B 7000 B )]

3 6000 )]

3 6000 5000 5000 C C 4000 C C 4000 D D 3000 D D 3000 2000 Hardness [N/(g·cm 2000 Hardness [N/(g·cm 1000 1000 0 0 G'G' GG FG HG F H G'G' GG FGSamples HG F H Samples Figure 12. Specific hardness testing results obtained for composite samples containing commercial Figure 12. Specific hardness testing results obtained for composite samples containing commercial Glass fiber alone (G’G’), Glass fiber alone (GG), Flax and Glass fiber (FG), Hemp-agave and Glass fibers GlassFigure fiber 12. Specificalone (G’G’), hardness Glass testing fiber resultsalone (GG), obtained Flax forand composite Glass fiber samples (FG), Hemp-agave containing commercial and Glass (HG), Flax fiber alone (F), and Hemp fiber alone (H) using Janka ball. fibersGlass (HG),fiber alone Flax fiber(G’G’), alone Glass (F), fiberand Hempalone (GG),fiber alone Flax and(H) usingGlass Jankafiber (FG)ball. , Hemp-agave and Glass 4. Conclusionsfibers (HG), Flax fiber alone (F), and Hemp fiber alone (H) using Janka ball. 4. Conclusions 4. ConclusionsThis study explored the viability of using natural fiber in combination with synthetic glass fiber in a hybridThis PU study composite. explored The the hybridization viability of ofusing flax natural and hemp-agave fiber in combination fibers with glasswith fibersynthetic in polyurethane glass fiber in a Thishybrid study PU exploredcomposite. the Theviability hybridization of using natural of flax fiber and in hemp-agave combination fibers with syntheticwith glass glass fiber fiber in in a hybrid PU composite. The hybridization of flax and hemp-agave fibers with glass fiber in

Materials 2016, 9, 390 14 of 15 composite was performed to investigate their physical and mechanical properties. The void content of 7.9% was observed for flax2 mat reinforced samples, showing its incompatibility with the PU resin. The hybrid samples from flax and hemp-agave fibers showed considerable improvement in water absorption properties to the unhybridized flax and hemp composite samples by 37% and 43%, respectively. The specific flexural properties showed that the hybridized flax and hemp-agave composite samples behaved similarly while the flax alone composite showed better specific flexural stiffness than all other combinations of flax and hemp-agave composite samples. The specific flexural strength results showed no significant difference between the flax and hemp-agave hybridized composite samples. The compressive tests showed no improvements among all the natural fiber reinforced samples, even after hybridizing with the glass fiber. Hybridization of flax and hemp-agave fiber composites improved their hardness properties because of the addition of the high strength glass fibers.

Acknowledgments: The research described herein was funded by the North Dakota Agricultural Products Utilization Commission (APUC), for which the authors are thankful to them. The authors acknowledge SpaceAge Synthetics, Fargo, ND for providing some of the essential raw materials needed for this project. Author Contributions: Sreekala Bajwa proposed the research topic and supervised the entire project along with Dilpreet Bajwa. Chad Ulven provided the natural fiber mat used in the study. Pankaj Pandey designed the experiments, performed all the mechanical and physical tests, analyzed the data, captured images, and wrote the manuscript. All the authors contributed to the writing and approved the final manuscript. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Muhammad, P.; Mohini, M.S. Carbon storage potential in natural fiber composites. Resour. Conserv. Recycl. 2003, 39, 325–340. 2. Wambua, P.; Ivens, J.; Verpoest, I. Natural fibres: Can they replace glass in fibre reinforced plastics? Compos. Sci. Technol. 2003, 63, 1259–1264. [CrossRef] 3. Monteiro, S.N.; Terrones, L.A.H.; D’Almeida, J.R.M. Mechanical performance of coir fiber/polyester composites. Polym. Test. 2004, 27, 591–595. [CrossRef] 4. Bos, H.L.; Van Den Oever, M.J.A.; Peters, O.C.J.J. Tensile and compressive properties of flax fibres for natural fibre reinforced composites. J. Mater. Sci. 2002, 37, 1683–1692. [CrossRef] 5. Niu, P.F.; Liu, B.Y.; Wei, X.M.; Wang, X.J.; Yang, J. Study on mechanical properties and thermal stability of polypropylene/hemp fiber composites. J. Reinf. Plast. Compos. 2011, 30, 36–44. 6. Kazuya, O.; Toru, F.; Yuzo, Y. Development of bamboo-based polymer composites and their mechanical properties. Compos. Part A 2004, 35, 377–383. 7. Davoodi, M.M.; Sapuan, S.M.; Ahmad, D.; Ali, A.; Khalina, A.; Jonoobi, M. Mechanical properties of hybrid kenaf/glass reinforced epoxy composite for passenger car bumper beam. Mater. Des. 2010, 31, 4927–4932. [CrossRef] 8. Naozumi, T.; Kohei, U.; Koichi, O.; Mitsuhiro, S. Biodegradation of aliphatic polyester composites reinforced by abaca fiber. Polym. Degrad. Stab. 2004, 86, 401–409. 9. Jawaid, M.; Khalil, H.P.S.A.; Hassan, A.; Dungani, R.; Hadiyane, A. Effect of jute fibre loading on tensile and dynamic mechanical properties of oil palm epoxy composites. Compos. Part B 2013, 45, 619–624. [CrossRef] 10. Saunders, J.H.; Frisch, K.C. Polyurethanes: Chemistry and Technology; Parts I and II; Interscience Publishers: New York, NY, USA, 1962. 11. Dwan’isa, J.P.L.; Mohanty, A.K.; Misra, M.; Drzal, L.T. Biobased polyurethane and its composite with glass fiber. J. Mater. Sci. 2004, 39, 2081–2087. 12. Saheb, D.N.; Jog, J.P. Natural fiber polymer composites: A review. Adv. Polym. Technol. 1999, 18, 351–363. [CrossRef] 13. John, M.J.; Thomas, S. Biofibres and biocomposites. Carbohydr. Polym. 2008, 71, 343–364. [CrossRef] 14. Koradiya, S.B.; Patel, J.P.; Parsania, P.H. The preparation and physicochemical study of glass, jute and hybrid glass-jute bisphenol-c-based epoxy resin. Polym. Plast. Technol. 2010, 49, 1445–1449. [CrossRef] Materials 2016, 9, 390 15 of 15

15. Wong, K.J.; Nirmal, U.; Lim, B.K. Impact behavior of short and continuous fiber-reinforced polyester composites. J. Reinf. Plast. Comp. 2010, 29, 3463–3474. [CrossRef] 16. De Carvalho, L.H.; Moraes, G.S.; D’Almeida, J.R.M. Influence of water absorption and pre-drying conditions on the tensile mechanical properties of hybrid lignocelluloses fiber/polyester composites. J. Reinf. Plast. Comp. 2009, 28, 1921–1932. [CrossRef] 17. ASTM D2734. Standard Test Method for Void Content of Reinforced Plastics; ASTM International: West Conshohocken, PA, USA, 2009. 18. ASTM D570. Standard Test Method for Water Absorption of Plastics; ASTM International: West Conshohocken, PA, USA, 2010. 19. ASTM D6341. Standard Test Method for Determination of the Linear Coefficient of Thermal Expansion of Plastic Lumber and Plastic Lumber Shapes Between ´30 and 140 ˝F(´34.4 and 60 ˝C); ASTM International: West Conshohocken, PA, USA, 2014. 20. ASTM D790. Standard Test Methods for Flexural Properties of Unreinforced and Reinforced Plastics and Electrical Insulating Materials; ASTM International: West Conshohocken, PA, USA, 2015. 21. ASTM D695. Standard Test Method for Compressive Properties of Rigid Plastics; ASTM International: West Conshohocken, PA, USA, 2010. 22. ASTM D1037. Standard Test Methods for Evaluating Properties of Wood-base Fiber and Particle Panel Materials; ASTM International: West Conshohocken, PA, USA, 2012. 23. Kishore, M.R. Jute-glass sandwich composites. J. Reinf. Plast. Comp. 1985, 4, 186–194. 24. Bismarck, A.; Askargorta, I.A.; Springer, J.; Lampke, T.; Wielage, B.; Stamboulis, A. Surface characterization of flax, hemp and cellulose fibres: Surface properties and the water uptake behaviour. Polym. Compos. 2002, 23, 872–894. [CrossRef] 25. Tserki, V.; Matzinos, P.; Zafeiropoulos, N.E.; Panayiotou, C. Development of biodegradable composites with treated and compatibilized lignocellulosic fibers. J. Appl. Polym. Sci. 2006, 100, 4703–4710. [CrossRef] 26. Agrawal, R.K.; Drzal, L.T. Adhesion mechanisms of polyurethanes to glass surfaces. Part II. Phase separation in polyurethanes and its effects on adhesion to glass. J. Adhes. Sci. Technol. 1995, 9, 1381–1400. [CrossRef]

© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).