Abstract
Two different thermoset biocomposite systems are experimented in this study with the hope to improve their mechanical properties. Fiberglass and hemp, in form of fabrics, are used to reinforce the thermoset polymer matrix, which includes a traditional epoxy resin and a linseed oil-based bioresin (UVL). The fiber/polymer matrix interface is modified using two different approaches: adding a plant-based oil (pine or linseed) to the polymer matrix or coating the fibers with 3-(aminopropyl)triethoxysilane (APTES) prior to integrating them into the polymer matrix. Epoxy resin is cured using an amine-based initiator, whereas UVL resin is cured under ultraviolet light. Results show that hemp fibers with APTES prime coat used in either epoxy or UVL matrix exhibit some potential improvements in the composite’s mechanical properties including tensile strength, modulus of elasticity, and ductility. It is also found that adding oil to the epoxy matrix reinforced with fiberglass mostly improves the material’s modulus of elasticity while maintaining its tensile strength and ductility. However, adding oil to the epoxy matrix reinforced with hemp doubles the material’s ductility while slightly reducing its tensile strength and modulus of elasticity.
1. Introduction
Biocomposites have the potential to replace traditional, petroleum-based composites, which present serious environmental concerns due to their long-lasting existence in landfills. However, the mechanical properties of biocomposites are often inferior compared to their conventional, fully synthetic counterparts [1–7]. To improve the mechanical properties, several studies have focused on altering the surface energy of natural fibers in order to modify the adhesiveness between the fiber and thermoset matrix. There have been numerous examples where the particle fillers in composites are treated with silane before incorporation into polymer matrices [8–15]. The dispersion aids, modifiers, or other additives were shown to influence the processing and the final mechanical properties of the materials. Reported treatments included the physical alteration of the fibers such as stretching, surface oxidation, corona treatment, plasma treatment, chemically coating the fibers, and adding an intermediate layer (in either fiber or powder form) between the hydrophilic fibers and hydrophobic polymer matrix. Silane treatments were tested by Cui and Kessler in 2012 [16]. Specifically, glass fibers were treated with two different types of silanes in separate trials, norbornenylethyldimethylchlorosilane and norbornenylethyltrichlorosilane, and then used to reinforce the linseed oil-based polymer matrix. It was found that both chemical treatments were able to increase the flexural strength of the materials.
Only a few published studies on biomaterials have recently explored the silanes and related treatments for biocomposite fillers. Williams et al. reported an alkali treatment of the composites, which showed the amount of strengthening achieved by coated kenaf fibers depended on the matrix being reinforced [17]. This fiber treatment reduced the strength of the composite to less than that of the neat unsaturated polyester matrix. However, the fiber treatment resulted in an increase in mechanical properties when reinforcing a biopolyester resin. In another study reported by Cantero et al., flax fibers were chemically treated with vinyl trimethoxysilane, which were then used to reinforce a polypropylene matrix [18]. This chemical treatment resulted in no change in mechanical properties compared to untreated fibers.
This study aims to improve the mechanical strength and ductility of several fiber-reinforced thermoset composites through modifications of either the fibers or the polymer matrix. Epoxy and a linseed oil-based (UVL) resin were used for the polymer matrix, while fiberglass and hemp were used for the reinforcing agent in a fabric form. Epoxy is a common resin that is widely used in industrial applications, such as sports equipment, wind turbine blades, and automotive parts. UVL is also a thermoset resin with potential to replace traditional resin, like epoxy, in applications that require high impact resistant surfaces [4]. Material modifications were accomplished by two different approaches. The first method involved adding plant-based oils, specifically pine oil and linseed oil, to the polymer resin mix before laminating the fabric. These oils were selected due to their ability to reduce the microhardness of the composite material surface, improve ductility, and increase the compostability of the material [5, 19]. In the second method, the fabric was coated with a silane coating prior to integrating them into the polymer matrix. Tensile strength, modulus of elasticity, and ductility of the modified composites were then characterized and compared to the untreated composites.
2. Materials and Methods
2.1. Materials
Two different thermoset resins were used in this study: a traditional epoxy resin and a linseed oil-based bioresin (referred to as UVL resin). The epoxy resin was a marine grade A side 314 (viscosity of 400–500 mPas at 298 K) consisting of >80% bisphenol A reaction product, <10% 2-ethylhexyl glycidyl ether, and <10% proprietary ingredients. This epoxy resin was cured with the aid of a B side 109 medium hardener (triethylenetetramine-based, viscosity of 550–650 mPas at 298 K). Both resin and hardener were purchased from TAP Plastics in California. The Eco Comp UVL resin, purchased from Sustainable Composites Ltd. in the United Kingdom, was used as a light brown fluid that contained approximately 95% linseed oil, 2.5% photoinitiator, and 2.5% propylene carbonate. It had a density of 1.03 kg/L and a viscosity of approximately 1500 mPas at 293 K.
Two different types of fibers were used to reinforce the thermoset polymer matrix: fiberglass (containing 100% E-glass, purchased from TAP Plastics) and hemp (containing 100% organic hemp, purchased from NearSea Naturals). The reinforcements used were in the form of fabrics with plain weave structure. Fiberglass had a thickness of 0.30 mm and a surface weight of 0.2796 kg/m2, while hemp had a thickness of 0.38 mm and a surface weight of 0.1970 kg/m2.
Pine oil and linseed oil were used as modifiers to the epoxy and UVL resins. Both were purchased from Sigma-Aldrich in pure liquid form. Linseed oil (density of 0.93 kg/m3, boiling point of >588 K) contained mostly linoleic acid, whereas pine oil (density of 0.86 kg/m3, boiling point of 426–448 K) contained mostly alpha- and beta-pinene. For the silane treatments of the fabrics, hydrochloric acid and ammonium hydroxide (both purchased from Fisher) were used to catalyze the sol-gel reactions of 3-(aminopropyl)triethoxysilane (purchased from Aldrich, CAS number 919-30-2). This particular silane (known as APTES) was selected for this study due to its common use in the composites industry [20].
2.2. Sample Preparation
Sample preparations were divided into three different categories: unmodified resin with plain fibers, modified resin with plain fibers, and unmodified resin with coated fibers. When no modification was done to the resin or the fibers, the composite samples were prepared using a modified version of the hand lay-up method similar to those reported in previous studies [4, 19].
For the group of samples whose fibers were modified with the silane coating, the fiberglass and hemp fabrics were first cut into 0.229 m × 0.305 m sheets. The fabric samples were rinsed briefly with 95% ethanol and then submerged into a premixed solution of 3-(aminopropyl)triethoxysilane (16 mL) in ethanol (800 mL) with 1 mL of either 1 M hydrochloric acid (to produce A-APTES coatings) or 1 M ammonium hydroxide (to produce B-APTES coatings) added as catalyst and water source. The fabrics were agitated gently in this sol-gel solution for 5 minutes with a glass stirring rod, then removed from solution, briefly rinsed with toluene, and spread to air dry for 0.5 h in a fume hood. The fabric sections were subsequently placed in a forced air drying oven (set to 353 K) for ten minutes.
For the group of epoxy composite samples in which the epoxy resin was modified with plant oil, the resin was first mixed with the hardener at 4 : 1 volume ratio. The oil (either linseed or pine oil) was next blended in with the resin/hardener mixture at 2% or 10% by weight. 2% was found to be the minimum oil content for making reproducible samples, and going above 10% oil content resulted in phase separation between the resin and the oil. The modified resin was then used to laminate with precut fabric sheets (either fiberglass or hemp) to form the composites.
The composite samples were prepared between two 0.216 m × 0.279 m sheets of transparency film (3 M PP2950 type, 0.1 mm thick) in order to provide a smooth surface and uniformity throughout the samples. Each prepared fabric sheet (coated or uncoated) was first laid down on a transparency film which was placed on top of a flat glass mold surface. The prepared liquid resin or resin mixture was then poured onto the fabric surface and spread evenly with a brush. Once the resin had effectively soaked through the fabric, a second transparency film was placed on top of the wet fabric. Next, a roller was applied to remove any air bubbles in the sample and produce a consistent surface thickness throughout the sample. The samples were subsequently cured. Epoxy composite samples were cured for 5 to 5.5 hours at room temperature, whereas UVL composite samples were cured in a UVP CL-1000 UV cross-linker (made in the United Kingdom, 8-W, 365 nm UV) for five minutes on each side.
Once the composites had been cured, they were prepared for mechanical testing. Only samples with higher thickness uniformity and no significant air bubbles were included in the mechanical testing and data analyses. The samples were cut into individual test coupons of size 0.254 m × 0.0254 m using a Speedy 300 Trotec laser cutter. Sample thickness was measured at three distinct points along the test coupon length using a micrometer. There were between 4 and 13 coupons tested for each of the composite types. Thickness averages and sample-to-sample standard deviations are reported in Table 1 for all of the tested composites. The reinforcing fiber contents were also calculated for each sample type usingwhere is the surface weight of either fiberglass or hemp fabric (0.2796 kg/m2 for fiberglass, and 0.1970 kg/m2 for hemp), is the surface area of the sample (0.254 m × 0.0254 m), and is the mass of each sample (kg). The reported fiber contents in Table 1 are average values determined from a minimum of four samples per each type.
2.3. Materials Testing
For tensile testing, a United Tensile Tester (made in the United States of America, model STM-50KN) with 5-kN maximum load Instron grips and 10-kN load cell was used. Selection of geometry for test specimens and tensile testing methods were based on the American Society for Testing and Materials (ASTM) standards D3039/D3039 M-08 (for polymer matrix composite materials) [22] and D882-10 (for thin plastic sheeting) [23], with 0.1524 m gauge length. The tension test employed an initial 45-N preload period at a head displacement rate of 0.0127 m/min, followed by constant head displacement rate of 0.00127 m/min for the remainder of the test. Tensile strength and percent elongation data, measured at break, were obtained directly from the tensile tests for all composite samples. Modulus of elasticity was determined based on the initial slope of the stress versus strain curve for each sample (using approximately 20 data points). The microhardness of epoxy composites with no oil addition was measured using a HVS-5 low-load Vickers hardness tester (Laizhou Huayin Testing Instrument Company, Ltd., China) with 0.5-kg indentation load and a 5-s time duration.
3. Results and Discussion
3.1. Modification of the Polymer Matrix
UVL polymer surface is inherently softer and more flexible than epoxy. A previous study reported a Vickers hardness number of for UVL reinforced with fiberglass and for UVL reinforced with hemp [4]. In the currently reported study, the Vickers hardness numbers were determined to be for fiberglass-reinforced epoxy and for hemp-reinforced epoxy. Because epoxy is more brittle than UVL, modification to the polymer matrix was only performed with epoxy, with the intent to improve its mechanical properties.
Figures 1 and 2 show the effects of the modification of epoxy polymer matrix with oil addition on the material’s mechanical properties. The reported values are based on the averages and standard deviations calculated from the replicates tested for each sample type. Data show that fiber reinforcement was able to improve both material tensile strength and modulus of elasticity as expected. When the epoxy matrix was modified with oil addition prior to reinforcement, the oil seemed to slightly improve the fiberglass-reinforced composites’ tensile strength further, with 3% to 14% increase compared to the fiberglass-reinforced epoxy baseline, with the exception of 2% linseed oil/fiberglass/epoxy combination. The effect was more significant on the modulus of elasticity. For example, 10% linseed oil addition to the epoxy matrix increased the modulus of the composite by 50%, whereas 2% pine oil addition to epoxy resulted in 74% increase of the modulus of the composite compared with fiberglass-reinforced epoxy with no oil presence. This improvement in mechanical properties was not observed for 2% linseed oil addition to epoxy, however. It is suspected that carboxylic acid groups in linoleic molecules (main ingredient of linseed oil) interacted with the epoxy cross-linked network, thus weakening the material. However, at higher linseed oil content (e.g., 10%) the chemical interactions among the linoleic acid molecules themselves became much stronger, overtaking their interference with the epoxy matrix. As a result, a secondary linoleic acid network formed within the epoxy matrix helped strengthen the polymer, improving the composite’s tensile and modulus properties.
Hemp-reinforced epoxy composites exhibited different behavior compared to fiberglass-reinforced composites. Adding oil to epoxy matrix mostly reduced the tensile strength and modulus of elasticity of the composites, with the exception of 2% linseed oil/hemp/epoxy combination. Both linseed oil and pine oil are hydrophobic materials. Nonetheless, hemp fibers are very hydrophilic due to their cellulosic nature. As a result, the oil presence prevented the hemp fibers from fully absorbing the epoxy resin during composite formation and possibly not allowing full resin penetration through the hemp fabric when the composite was cured. Insufficient interaction between the hemp fibers and the epoxy matrix resulted in a weaker composite. Mechanical properties of the sample with 2% linseed oil were unchanged compared to no oil addition, most likely due to similar molecular interactions phenomenon explained above between the oil molecules and the epoxy cross-linked network.
Figure 3 shows the effects of linseed oil and pine oil on the composite’s ductility, expressed as the percent elongation of the materials until failure. Percent elongation of epoxy polymer alone (with no reinforcing fibers) is also included in Figure 3 for baseline comparison. Data show that the reinforcing fibers alone decreased the epoxy composites’ ductility by approximately 39% in the case of fiberglass and as much as 50% in the case of hemp. This could be due to the insufficient interaction between the epoxy matrix and the fibers themselves. Adding linseed oil to the fiberglass-reinforced composites seemed to have negligible effects on the material’s ductility, with the exception for 2% pine oil/fiberglass/epoxy combination. While 2% pine oil addition to the epoxy matrix decreased percent elongation of the material by approximately 28%, it was able to increase the material’s modulus of elasticity by as much as 74% (in the case of fiberglass). This observation seems to be consistent to past studies where tensile and modulus property improvements could sometimes result in more brittle and less ductile material [4, 9]. For hemp-reinforced epoxy composites, the addition of oil to the epoxy matrix mostly improved the material’s ductility (as much as 50% improvement in percent elongation for 2% pine oil/hemp/epoxy combination). In fact, adding 2% pine oil to the epoxy matrix before reinforcing it with hemp fibers resulted in an increase of 50% improvement in percent elongation compared to hemp-epoxy composite without the oil presence, which is comparable to the ductility level of plain epoxy with no fiber reinforcement. This observation could suggest that pine oil addition might have resulted in a higher flexibility at the fiber/matrix interface, allowing the fibers to stretch more freely before breaking.
3.2. Modification of the Fibers
Both fiberglass and hemp fibers were modified with APTES prime coat, prepared with an acid catalyst (A-APTES) or a base catalyst (B-APTES), prior to lamination with either epoxy or UVL resin. The sol-gel condensation reaction used to prepare the APTES coating typically leads to porous ceramic-like materials and in the conditions utilized here likely forms nanoparticle coatings on the filler fabrics, as previously reported in other similar studies [24, 25]. The short reaction times (five minutes) used in the coating process were intended to minimize potential damage to the natural fibers due to acid exposure and to encourage small nanoparticle growth. The acid and base catalyst loading levels might have had a profound effect on the growth and porosity of the nanoparticles (typically grown into fully densified but porous films) that determined final quality (optical, mechanical, surface area, and catalyst support quality). The sol-gel growth in this case should provide small, amino functionalized silica nanoparticles on the surface of the fabrics, regardless of whether acid or base catalyst is used.
Figures 4, 5, and 6 show the effects of each prime coat on the tensile strength, modulus of elasticity, and percent elongation of UVL and epoxy composites reinforced with either hemp fibers or fiberglass, respectively. Overall, there seemed to be no significant differences between the two coatings considering their effects on the composites’ mechanical properties. This was expected due to the short reaction times for sol-gel transition and minimized deposition times during the coating preparation. The prime coat was able to improve both tensile strength and modulus of elasticity for hemp-epoxy composites (6% to 10% improvement in tensile strength, and 16% to 18% improvement in modulus), while maintaining the materials’ ductility. For fiberglass-epoxy, hemp-UVL, and fiberglass-UVL composites, the prime coats had slightly negative impact on both tensile strength and modulus of elasticity of the composite materials. The hemp cellulose structure (with abundant reactive hydroxyl groups) may allow more internal pore access and reactions compared to solid fiberglass. This may help stiffen the fibers themselves before incorporation of the resins. The increase in tensile strength in the case of epoxy (Figure 4) and not UVL may indicate an additional reaction with the epoxy groups. The additional amine content of APTES may serve as the catalyst for ring opening and ensure a greater extent of the epoxidation reaction, which are often considered only partial.
Regarding ductility, the prime coats appeared to be more compatible with hemp fiber-reinforced UVL than any other combinations. For example, APTES coating of hemp fibers resulted in 84% improvement in percent elongation of the hemp-UVL composite, while reducing its tensile strength and modulus of elasticity by 13% to 42%. Conversely, prime coating of the fiberglass fibers inside the UVL matrix had slightly negative effects on both modulus of elasticity and percent elongation of the material. For epoxy composites, whether reinforcing with fiberglass or hemp fibers, the prime coating of the fibers had negligible effects on the percent elongation of the composites. The application of this coating may provide a nanoparticle coating both on the outside (glass and hemp) and on the inside of the porous (hemp) fibers. This may influence the ductility and provide a more graceful failure mechanism and allow greater energy absorbing ability before part failure.
The significant enhancement in ductility property of prime coated hemp-UVL composite was also evident in its matrix cracking behavior during tensile testing. As seen in Figure 7(d), many major cracks formed along the UVL matrix, but the whole composite structure was still able to withhold tensile stress because of the very ductile coated hemp fibers. This was true regardless of the type of APTES coating used. These coated hemp-UVL samples took more than 20 minutes before reaching the failure point compared to a less than 5-minute time duration to failure for other composite samples. For other coated composite samples (different than hemp-UVL combination), matrix cracking behaviors were very similar to that of a fiber-reinforced thermoset composite as reported in previous studies [4, 5], as shown in Figures 7(a)–7(c). These observations suggest that APTES prime coats are more compatible with biocomposites, in which both fiber reinforcement and polymer matrix phases are made of biobased materials, and suitable for low-strength applications where highly ductile materials are required.
4. Conclusions
Modifications of both fibers and polymer matrices were tested for several different types of fiber-reinforced thermoset composites, with the ultimate goal of improving the mechanical properties of the materials. Two approaches for material modification were investigated: adding a plant-based oil to the polymer matrix (UVL and epoxy) or coating of the fibers (hemp and fiberglass) prior to integrating them into the polymer matrix. Results show that adding oil to the polymer matrix of the fiberglass-reinforced epoxy composites mostly improved the material’s modulus of elasticity while maintaining its tensile strength and ductility. However, adding oil, either pine or linseed oil, to the polymer matrix of the hemp-reinforced epoxy composite could double the material’s ductility while only slightly reducing its tensile strength and modulus of elasticity. Also, the modification of hemp fibers using an APTES prime coat prior to integrating with either epoxy or UVL matrix showed some potential improvements in the material’s mechanical properties including tensile strength, modulus of elasticity, and ductility.
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
Acknowledgments
This research was financially supported by the University of San Diego through the Summer Undergraduate Research Experience program, the Sustainability Environmental Education Development grant, and the Engineering Faculty Research grant. The authors are also grateful for the assistance of Mr. Sam Burt and Mr. Devyn Bryant on sample preparation and tensile testing.