NC State
Mohd Radzi, M. H., Abdan, K., Zainal Abidin, Z., Md Deros, M. A., and Zin, M. H. (2020). "Effect of blend composition on characteristics and performance of jatropha bio-epoxy/epoxy matrix in composites with carbon fiber reinforcement," BioRes. 15(2), 4464-4501.


Characteristics and performances of a blended jatropha bio-epoxy/epoxy as a matrix in carbon fiber reinforcement was studied. The amount of bio-epoxy was arranged from 0 wt%, 25 wt%, 30 wt%, 40 wt%, and 50 wt% of the total matrix. Several analyses were performed to characterize and observe their performance. Fourier transform infrared spectroscopy, thermal analysis, physical characteristics, flammability, and soil burial were conducted, as well as mechanical tests. The results showed that introducing bio-epoxy in the matrix changed characteristics and increased or decreased their performance. Blending more than 25 wt% of bio-epoxy led to improved thermal stability between 280 °C to 350 °C and better biodegradability. However, tensile and flexural strength as well as modulus of elasticity decreased once the proportion of bio-epoxy was greater than 25 wt%. The paper proposed an optimal amount of jatropha bio-epoxy so that an alternative biocomposite application could be introduced to minimize carbon footprint in the environment.

Download PDF

Full Article

Effect of Blend Composition on Characteristics and Performance of Jatropha Bio-epoxy/Epoxy Matrix in Composites with Carbon Fiber Reinforcement

Mohd Hafiezal Mohd Radzi,a,e,* Khalina Abdan,a,c,* Zurina Zainal Abidin,b,c Mohd Azaman Md Deros,d and Mohd Hanafee Zin f

Characteristics and performances of a blended jatropha bio-epoxy/epoxy as a matrix in carbon fiber reinforcement was studied. The amount of bio-epoxy was arranged from 0 wt%, 25 wt%, 30 wt%, 40 wt%, and 50 wt% of the total matrix. Several analyses were performed to characterize and observe their performance. Fourier transform infrared spectroscopy, thermal analysis, physical characteristics, flammability, and soil burial were conducted, as well as mechanical tests. The results showed that introducing bio-epoxy in the matrix changed characteristics and increased or decreased their performance. Blending more than 25 wt% of bio-epoxy led to improved thermal stability between 280 °C to 350 °C and better biodegradability. However, tensile and flexural strength as well as modulus of elasticity decreased once the proportion of bio-epoxy was greater than 25 wt%. The paper proposed an optimal amount of jatropha bio-epoxy so that an alternative biocomposite application could be introduced to minimize carbon footprint in the environment.

Keywords: Bio-epoxy; Jatropha oil, Biocomposite; Fiber carbon; Thermal properties; Mechanical properties

Contact information: a: Department of Biological and Agricultural Engineering, Faculty of Engineering, Universiti Putra Malaysia, 43400 Serdang, Selangor Malaysia; b: Department of Chemical Engineering, Faculty of Engineering, Universiti Putra Malaysia, 43400 Serdang, Selangor Malaysia; c: Laboratory of Biocomposite Technology, Institute of Tropical Forestry and Forest Product (INTROP), Universiti Putra Malaysia, 43400 Serdang, Selangor Malaysia; d: School of Manufacturing Engineering, Universiti Malaysia Perlis, 02600 Arau, Perlis Malaysia; e: Faculty of Engineering Technology, Universiti Malaysia Perlis, 02600 Padang Besar, Perlis Malaysia; f: Aerospace Malaysian Innovation Centre (AMIC), No. 3-6-01, GMI, 43000 Kajang, Selangor Malaysia;

* Corresponding authors:;


Usage of polymer is vast, and all applications currently utilize plastics as one of its main materials. Such plastics are usually made of non-renewable resources, such as petroleum-based polymer. The biodegradation rate of such plastic is long, which causes harm to the environment, especially humans and wildlife in the ocean after the materials are disposed of. A life cycle assessment made to compare synthetic epoxy and bio-based epoxy has revealed favourable results of the bio-based version in terms of environment impact and energy use (La Rosa et al. 2014).

Several studies have been conducted by researchers in the past using renewable resources for composite application, such as sugar palm starch (Jumaidin et al. 2017), hemp oil (Manthey et al. 2012), tannins from trees (Lagel et al. 2016), soybean oil (Takahashi et al. 2008; Roudsari et al. 2014; Wang 2014), polylactic acid (Yusoff et al. 2016; Morales et al. 2017), karanja oil (Kadam et al. 2015), etc. All of these studies have the same purpose, which is to introduce and replace existing non-renewable polymer with bio-based polymer and apply it in composite application.

These renewable polymers are inexpensive, have a low processing cost, are easily available, and are abundant compared to a petroleum-based polymer. However, they have some drawbacks, mainly performance-wise, overall due to their biodegradability characteristics. Thus, its application is specific and limited to some areas, such as interior parts of a dry area, or in conditions in which they make use of their main advantage. Otherwise, the biopolymer needs to be modified with fillers or blended with other polymers or biopolymers so that its total characteristics will be enhanced.

Jatropha seed pricing ranges from 0.1 to 0.34 $/kg with yield up to 6.5 tons per year (Kalam et al. 2012). System of generating jatropha oil is straightforward, and technology involved in the oil production is readily available (Yee et al. 2011). The overall jatropha oil manufacturing process are: Sowing ⇒ Cultivation ⇒ Harvest ⇒ Seed Dehulling and Cleaning ⇒ Oil Extraction ⇒ Oil Filtration and Purification ⇒ Oil Refining. There are several advantages of the plant that make it a useful plant with a number of potential applications, as suggested by Ariza-Montobbio and Lele (2010). Unlike soybeans, canola and many other agricultural sources of biodiesel, jatropha can be cultivated on marginal and non-agricultural land. This implies that growing jatropha will not result in forgoing food crop cultivation. In addition, it starts producing seeds as early as 12 months after being planted, but the effective yield is obtained only after 2 to 3 years. The jatropha plant remains useful for 35 to 50 years, and its seeds can produce around 35% oil content. Based on per acre production, jatropha’s yield is the second highest, next to that of palm oil. It is quite simple to produce biodiesel from the jatropha, while its waste (deoiled jatropha seed extract) can be used to produce organic fertilizer because it has high contents of protein, nitrogen compounds, and some anti-pest compounds. After 4 or 5 years of treatment with this cake, the soil of this original non-agricultural land will become suitable for planting food crops or trees for reforestation.

A number of studies about jatropha oil have been conducted, such as in bio-diesel improvement (Silitonga et al. 2014; Dharma et al. 2016), coating applications (Gogoi et al. 2015), tribology aspects (Satheesh Kumar et al. 2010), adhesive applications (Aung et al. 2014), etc. A study about jatropha based alkyd was performed on a composite application. It was blended with epoxy resin with expanded graphite as a filler to boost its total performances. Via mechanical tensile test, it was found that blended matrix of jatropha based alkyd resin led to improved distribution and interfacial adhesion between matrix and filler to provide reinforcement, therefore offering an improvement in tensile strength (Gogoi et al. 2014). Aung et al. (2014) had synthesized jatropha-based polyurethane adhesive for wood application and compared with palm oil-based adhesive which possessed poor performance due to multiple saturated acids. Jatropha oil-based adhesive was fond to have better shear strength and better reaction to water, acid, and alkali compared to palm oil adhesive (Aung et al. 2014).

Sammaiah et al. (2014) synthesized jatropha bio-epoxy for the purpose of lubrication application. Jatropha bio-epoxy and soybean bio-epoxy were compared, and both showed comparable characteristics in a high temperature environment (Sammaiah et al. 2014). In a study carried out by Gogoi et al. (2015), jatropha oil was converted into bio-based alkyd resin. It was blended with different wt% of an aqueous citric acid solution for the purpose of improving the performances of the bio-based resin. It was observed that the properties such as curing time, chemical resistance, scratch hardness, thermal stability, and tensile strength of the alkyd resins improved significantly on blending. With 50 wt% jatropha bio-epoxy content, the thermal stability and tensile strength of the blends were improved by 42 °C and 3.18 MPa, respectively. The results suggest a strong influence of the amount of jatropha bio-epoxy and citric acid on the alkyd resins output. The thermal and mechanical properties of the cured films can be further improved by undergoing a postcuring process at 160 °C, and the study indicated suitability of these blends in surface coating applications (Gogoi et al. 2015).

Despite many explorations and investigations on jatropha oil that have been done, usage of epoxidized jatropha oil in composite applications is scarce. In this article, jatropha oil was epoxidized and blended with synthetic epoxy. Then, it was used as matrix in laminated fiber carbon composites. Early studies have been conducted, and they showed promising results for further experimental works (Hafiezal et al. 2017, 2018, 2019). Thus, the aim of this article was to reveal the characteristics and performance trends of blended epoxidized jatropha oil with synthetic epoxy in certain ranges of blended ratios and its limitations when used as a matrix in laminated composite.



Sample preparation

Fabrics of woven carbon fiber and glass fiber type S, both of 200 gsm with 2000 tex, were used as the reinforcements. Plain weave type was chosen because it is versatile and possesses the most balanced characteristics amongst other weave types. Synthetic epoxy brand Epoxamite 100 with a slow hardener named 103 was used as matrix in the composite panel (Mechasolve Engineering Sdn. Bhd. Company, Selangor, Malaysia). Composite panel was fabricated using a vacuum infusion process, which used the strategic resin transfer moulding method in an environment condition of 25 ± 1 °C and 50 ± 5% humidity. After that, it was left to cure for 24 h prior to use. Jatropha bio-epoxy was fabricated by an in-house laboratory of University Putra Malaysia – Aerospace Malaysia Innovation Center (UPM-AMIC) collaboration, and jatropha oil was supplied by Bionas Sdn. Bhd. Company (Kuala Lumpur, Malaysia). Ten layers of fiber carbon fabric with dimensions of 300 mm × 300 mm were stacked and wetted by bio-epoxy/epoxy according to experimental design. Samples were cut according to each testing specification described by the testing standards such as ASTM, ISO, and manufacturers.

Experimental sample composition design.

In Table 1, total experimental design of various composition between carbon, glass fiber and different wt% of blended bio-epoxy/epoxy were shown. For carbon fiber laminated composites


Fourier transform infrared spectroscopy – attenuation total reflection (FTIR-ATR)

The FTIR spectroscopy with ATR technique was used to detect functional groups in both the fully synthetic and the blended jatropha bio-epoxy in uncured and cured forms. The spectra of the material were obtained using an IR spectrometer (Pelkin Elmer Frontier GladiATR; Pike Technology, Madison, WI, USA) that scanned over the sample surface in the range of 400 cm-1 to 4000 cm-1.

Table 1. Composition Design of Laminated Composites with Various Fibers and wt% of Blended Bio-epoxy/epoxy

Tensile and flexural test

A universal testing machine (Model AG-X 250kN; Shimadzu, Kyoto, Japan) was used for testing tensile and flexural properties. Five specimens of 250 mm × 25 mm were cut from the composite panel according to ASTM D3039/D3039M (2017) for the tensile test. Meanwhile, specimens for flexural test were prepared by taking into consideration the span-to-depth ratio of 40:1 as the minimum according to ASTM D790-17 (2017). This was to ensure the elimination of shear effects during the test. The loading speed for tensile test and flexural test were 4 mm/min and 2 mm/min, respectively. The room conditions for the testing were set at a temperature of 23 ± 1 °C and relative humidity of 50 ± 5%.

Impact test

Charpy impact tests were conducted according to ASTM D6110-18 (2018) at a temperature of 23 ± 1 °C and relative humidity of 50 ± 5%. The samples were cut into dimensions of 50 mm (L) × 13 mm (W) × 3 mm (T) with a notch in the middle. The tests were performed on three replications by using a mechanical pendulum impact tester, model QC639C (Plasmost Enterprise Sdn. Bhd., Penang, Malaysia). The impact strength was calculated based on the impact energy and cross-section area of the specimen.

Morphology observation

The morphology of samples was observed under an optical microscope and a scanning electron microscope (SEM) (Hitachi S-3400N; Hitachi, Tokyo, Japan) with an acceleration voltage of 10 kV.

Thermogravimetric analysis (TGA)

Thermogravimetric analysis was conducted according to ASTM E1131-08 (2014) by using a TGA Q500 (TA Instruments, New Castle, DE, USA) instrument to characterize thermal degradation. The two samples weighed 8.35 mg and 6.38 mg for the fully synthetic composite structure and the blended jatropha bio-epoxy composite structure, respectively. They were heated from 30 °C to 600 °C with a constant heating rate of 10 °C/min in a nitrogen flow of 50 mL/min. The heat resistant index Ts was calculated according to Eq. 1, where Td5 is the temperature degradation at 5% and Td30 is the temperature degradation at 30%:

TS = 0.49(Td5 + 0.6(Td30 / Td5)) (1)

Differential scanning calorimetry (DSC)

Differential scanning calorimetry was conducted according to ASTM D3418-15 (2015) using a DSC Q20 device (TA Instruments, New Castle, DE, USA) to determine the thermal transition characteristics. The two samples weighed 6.06 mg and 5.23 mg for the fully synthetic composite structure and the blended bio-epoxy composite structure, respectively. They were heated from 30 °C to 300 °C with a constant heating rate of 10 °C/min in a nitrogen flow of 40 mL/min.

Dynamic mechanical analysis (DMA)

Dynamic mechanical analysis was conducted according to ASTM D4065-12 (2012) using a DMA Q800 (TA Instruments, New Castle, DE, USA). Samples of 35 mm × 12 mm for each composite structure were put in single cantilever mode from 30 °C to 140 °C with a constant heating rate of 5 °C/min in a nitrogen atmosphere at a frequency of 1 Hz and an amplitude of 15 μm.

Fiber volume fraction and density

Three specimens of 25 mm × 25 mm were cut from composite panel according to ASTM D3171-15 (2015). The mass of crucible and specimen were measured at 0.001 g precision, and specimen density was measured by a water displacement density calculator device (Densimeter MDS300; Qualitest, Plantation, FL, USA). Each sample in a crucible was burnt in an electrical muffle furnace at 550 °C for 2 h. Afterwards, the remaining mass of sample was measured, and volume fraction of fiber was calculated using Eq. 2,

Vf = 100(Mf /Mi)(ρc/ρf) (2)

where Mi is initial mass (g) of specimen, Mf is final mass (g) of specimen after combustion, ρc is density (g/cm3) of specimen, and ρf is density (g/cm3) of specimen in the fiber reinforcement.

Moisture content

Five samples (10 mm × 10 mm × 3 mm) were prepared for moisture content (MC) investigation. All samples were heated in an oven for 24 h at 60 °C. The weight of samples before (Mi) and after (Mf) heating was measured to calculate moisture content. Moisture content was determined using Eq. 3,

Mc = 100((Mi – Mf) / Mi) (3)

where Mc (%) is the moisture of composite, Mi is initial moisture (g), and Mf (g) is final moisture.

Flammability UL94 horizontal burning

Three samples with dimensions of 250 mm × 15 mm × 2 mm were used to test flammability performance according to the horizontal burning UL94 standard test. The samples were mounted horizontally with a stand and burned with a torch at around 1200 °C at a 45° angle of inclination. A flame was ignited at the end of the specimen. This was maintained for 30 s and then left to propagate until the flame reached the first mark. If the flame reached the first mark in less than 30 s, the sample was considered to have failed the test. Two marks were set on the specimens: the first mark was for the time to start recording, and the second mark was where the flame should stop (75 mm from the first mark). The time was recorded until either the flame self-extinguished or the flame reached the second mark. The specimens were left under pre-treatment for 48 h at 23 °C and 50% humidity. If any drops fell due to burning, this was taken into consideration. The burning rate was calculated using Eq. 4:

Burning rate (mm/s) = (Length burned (mm)) / (Burning duration (s)) (4)

Soil burial test

A biodegradability test (soil burial) was conducted. The laminated biocomposite and hybrid laminated biocomposite samples were weighed and buried in an individual container of soil in which soil nutrients were added. These tests were conducted at ambient temperature (27 °C to 34 °C) in an open-air area. The biodegradation was monitored for 120 d at intervals of 30 d. Each sample would be taken out of the container, cleaned, and then its wet mass was weighed. Then, the sample was washed and dried at 40 °C for 24 h, so that its dry mass could be recorded. The sample was observed under an optical microscope (Olympus Stream; OLYMPUS Corporation, Tokyo, Japan) to see if there were traces of microorganism activities.


Curing Behaviour

As this bio-epoxy was new, it needed to be characterized because the blended elements would provide different properties originating from jatropha bio-epoxy. Figure 1 shows the FTIR-ATR comparison data for both fully synthetic and blended jatropha bio-epoxy in uncured and cured forms. These spectra were obtained to identify the difference of functional groups in chemical bonding when jatropha bio-epoxy was added into the matrix.

The spectra could be separated into four regions: 4000 cm-1 to 2500 cm-1, which is for N–H, C–H, and O–H single bonds; 2500 cm-1 to 2000 cm-1, which is for triple bonds; 2000 cm-1 to 1500 cm-1, which is for C=O, C=N, and C=C double bonds; and 1500 cm-1 to 400 cm-1, which are the fingerprints of the compound that are C–O single bonds and C=C alkene double bonds.

In the first region, a strong and broad peak at 3452 cm-1 was –OH stretching, which was higher for uncured fully synthetic epoxy compared to bio-epoxy. This is normally an alcohol with intermolecular bonding. For cured matrices, both bio-epoxy and fully synthetic epoxy were almost at the same intensity. At 2923 cm-1 and 2853 cm-1, the peaks of both bands were medium and sharper, and percentage of transmittance was higher in uncured blended jatropha bio-epoxy compared to fully synthetic epoxy. These were the aliphatic sp3 C–H stretching groups of the alkane compound class, and they are commonly found in triglyceride oil (Ishak and Salimon 2013).

There were no spectrum changes in the second region, which was for the triple bonding spectrum. Meanwhile, in the third region, a broad and sharp peak at 1736 cm-1 was identified only when jatropha bio-epoxy was added. This was the C=O stretching of ester groups, which are commonly present in fatty acids (Ezzah 2016).

Then, in the fourth region, where the fingerprints of the materials were, the spectra became less sharp and a little broader in some bands for uncured blended bio-epoxy and fully synthetic epoxy. Both showed the same patterns of peaks but at different intensities. At 829 cm-1, C–O–C stretching of oxirane group was identified in the blended bio-epoxy as reported in several studies that used bio-based epoxy as well (Manthey et al. 2012; Gogoi et al. 2015; Kadam et al. 2015). There were O–H bending of phenol group and carbonyl C–O stretching of aromatic ester at 1361 cm-1 and 1295 cm-1, respectively.

For cured blended jatropha bio-epoxy, the spectra were almost identical with those of the cured synthetic epoxy. This was important, as the cross-linking process occurred successfully as the band at 829 cm-1 disappeared in both cured matrices, same as reported by several prior studies (Vlček and Petrović 2006; Chauhan and Chibber 2013; Hazmi et al. 2013; Mincheva et al. 2013). At the end, the fingerprints were different, as most of double bonds were opened out, and the molecules became cross-linked in three-dimensional networks. Crosslinking is a bond that links one polymer to another different polymer chain. Those disappeared bonds were where the crosslinks happened for covalent bonds, ionic bonds, etc., for the purpose of strengthening physical properties of the resin.

Fig. 1. FTIR-ATR spectra comparison of uncured bio-epoxy and cured bio-epoxy

Figure 2 shows the IR spectra of cured bio-epoxy with incremental variation from 0 wt% to 50 wt%. At 1500 cm-1 and below, the fingerprint region, the similarity of the peaks between all curvatures was shown, but there was reduced intensity as the bio-epoxy amount increased. At 829 cm-1 the absorbance was attributed to epoxy groups in which the epoxy group became more intense as the wt% of bio-epoxy increased. This indicated that the amount of bio-epoxy should not exceed a certain ratio to avoid leftovers of epoxy group that were not crosslinked after curing. Up until 25 wt% bio-epoxy, the peak of epoxy group could be said to be almost non-existent.

From 1508 cm-1 until 1608 cm-1, aromatic stretching was identified, and it became higher when bio-epoxy wt% increased. From 1600 cm-1 until 1700 cm-1 was the C=O bond stretching of the ester, and the intensity became higher as bio-epoxy wt% increased compared to 0 wt%.

At 2855 cm-1 and 2923 cm-1, the C-H aliphatic carbon peak was identified. This indicated that the bio-epoxy had a long hydrocarbon chain in the backbone similar to epoxidized soybean oil (Subbiah 2016). Lastly, at 3452 cm-1, O–H stretching was identified, and it became higher as bio-epoxy wt% increased.

In general, 0 wt% bio-epoxy was the fully synthetic epoxy that exhibited almost complete crosslinking, with fewer C=O bonds, C–H bonds, and O–H bonding. Meanwhile, as bio-epoxy was blended up to 50 wt%, there was a noticeable presence of C=O, C–H, and O–H bonding. For 0 wt% bio-epoxy, most of the molecules were fully saturated, and not many double bonds were left, as can be seen from the IR spectra. As the amount of bio-epoxy increased, more unsaturated molecules were left even after curing process with the same hardener. Fortunately, as the amount of bio-epoxy increased, more aromatic groups could be detected. This led to better thermal stability of the compounds compared to fully synthetic epoxy.

Fig. 2. FTIR-ATR spectra with variation of wt% of cured bio-epoxy

Chemical reaction of the bio-epoxy can be understood from Fig. 3. Hardener used in this experiment was methylenebiscyclohexanamine,4,4’ as shown in Fig. 3(a). It contains two active sites of multiple amine functional groups. Figure 3(b) shows the synthetic epoxy (oxirane, 2,2′-((1-methylethylidene)bis(4,1-phenyleneoxymethylene))bis-, homopolymer), while Fig. 3(c) shows the jatropha bio-epoxy (triglyceride) compound, which consisted of at least two epoxide functional group, respectively. All these compounds were mixed for the curing process as presented in Fig. 3(d), where the reactive hardener would crosslink with epoxy and bio-epoxy at the epoxy ring. Note that as anticipated, the amine attacks the least substituted carbon of the epoxy ring which was not protonated. After proton transfer, the product of this reaction has a new N-C bond and an alcohol O-H functional group, resulting a three-dimensioning crosslinking network that promoted a very hard thermoset that could be reversed. In addition, the -OH groups help it to adhere to other structures such as reinforced fibers via hydrogen-bonding.

Fig. 3. Chemical structure of compounds in (a) hardener, (b) synthetic epoxy, (c) jatropha bio-epoxy, (d) curing reaction of blended bio-epoxy

Mechanical Performances

Figure 4 demonstrates the tensile properties, which are tensile strength, modulus of elasticity, and elongation at break, of laminated fiber carbon with varied wt% of bio-epoxy. It was expected that bio-epoxy would decrease the tensile properties of the laminated composites. Compared to 0 wt% bio-epoxy, as soon as bio-epoxy was added for 18 wt%, its tensile strength and modulus of elasticity decreased. Low modulus of elasticity revealed a lower rigidity attribute. The elongation at break was higher, which indicated that the specimen was more ductile when bio-epoxy was added.

Fig. 4. Tensile performances of laminated fiber carbon with wt% variation of bio-epoxy

As the wt% of bio-epoxy continued to vary in increment up to 50 wt%, the tensile strength continued to decrease up to 41% for 50 wt% bio-epoxy. Modulus of elasticity followed the tensile strength trend as well, at up to 42% reduction for 50 wt%. Surprisingly, 25 wt% and 18 wt% had approximately the same performances. Specimen 18 wt% exhibited lower tensile strength and lower modulus of elasticity than 25 wt%, most probably due to higher void content in 18 wt%. It had lower density, but its higher thickness explained the high void content that was formed in gasification process during fabrication. Meanwhile, for 30 wt% bio-epoxy, a remarkable reduction in tensile performances occurred. Therefore, 25 wt% bio-epoxy was a favourable blended ratio that proved to have high tensile strength, modulus of elasticity, and acceptable elongation at break.

Fig. 5. Tensile performances of laminated carbon fiber and fiber glass with wt% variation of bio-epoxy

Good tensile performances in 18 wt% and 25 wt% bio-epoxy were likely due to strong intermolecular hydrogen bonding in the laminated composite, as can be seen in the IR spectra shown in Fig. 2. This demonstrated that the performance came from a strong interfacial adhesion between fiber and bio-epoxy, which subsequently resulted in high efficiency of stress-transfer from bio-epoxy to the fibers obtained from micrography images, which is shown in Fig. 9.

Figure 5 shows a comparison of tensile strength, modulus of elasticity, and elongation at break between 10-layer laminated composites of fiber carbon (FC) and fiber glass (FG). Both were with 0 wt%, 25 wt%, and 50 wt% bio-epoxy. All three results for both FC and FG showed decreased trends when bio-epoxy content increased. This was a similar trend of low performance as well as performance degradation when a natural matrix or biopolymer was used as matrix of a composite, such as polysaccharides (Jumaidin et al. 2017), bio-based epoxy (Gogoi et al. 2015), soybean bio-epoxy (Tee et al. 2016), karanja oil (Kadam et al. 2015), etc. This was because bio-epoxy had farther crosslinking distance compared to synthetic epoxy, as found by Takahashi et al. (2008), in addition to low oxirane oxygen content (OOC), which was approximately 61% as found by Ezzah (2016).

Laminated composite FG of 0 wt% and 25 wt% bio-epoxy showed almost the same tensile performances. This might have been because the specimen with 0 wt% contained more voids compared to 25 wt% as its density was higher, but thickness and fiber volume fraction were almost the same as in Figs. 21, 22, and 23. Surprisingly, the laminated composite FC with 50 wt% showed better performance, such as higher tensile strength, higher modulus of elasticity, and lower elongation at break, than the laminated composite FG even with 0 wt% bio-epoxy. Thus, it was reasonable to replace fully synthetic fiber glass composites with fiber carbon composite with 50 wt% bio-epoxy in any related application.

Fig. 6. Flexural performances of laminated carbon fiber with wt% variation of bio-epoxy

Figure 6 shows flexural performances of laminated composites FC with variation of bio-epoxy of 0 wt%, 18 wt%, 25 wt%, 30 wt%, 40 wt%, and 50 wt%. As expected, laminated composite FC with 0 wt% showed the highest flexural stress and flexural modulus, followed by 18 wt%. However, once the wt% of bio-epoxy exceeded 18%, the maximum flexural stress decreased over 50%. For flexural modulus, 18 wt% and 25 wt% bio-epoxy were almost the same even though their deformation was at 31% difference. When bio-epoxy exceeded 25 wt%, their flexural modulus and deformation decreased. High deformation with high flexural stress meant that the composite was flexible, strong, and not easily ruptured during bending configuration.

Meanwhile, Fig. 7 shows flexural performances, such as maximum flexural stress, flexural modulus, and deformation, between laminated composite FC and FG with 0 wt%, 25 wt%, and 50 wt% bio-epoxy. Similar trends of decrement were seen for both reinforcements as the amount of bio-epoxy increased up to 50 wt%. Fully synthetic composite FC had the highest performances amongst them, and 25 wt% was the second highest in these comparisons. The 25 wt% bio-epoxy was even better than fully synthetic composite FG by 14% and 70% in terms of maximum stress and flexural modulus, respectively. In contrast, both laminated composites FC and FG with 50 wt% bio-epoxy exhibited the lowest performances with similar trends as seen in tensile results.

Fig. 7. Flexural performances of laminated carbon fiber and glass fiber with wt% variation of bio-epoxy

Impact energy is the measure of a force acting over distance to fracture the specimen. As the striker hits the specimen, the energy is absorbed until it yields. At this point, the specimen will begin to undergo plastic deformation starting from the notch created. It continues to absorb energy and hardens the plastic zone at the notch. Fracture occurs subsequently after the specimen cannot absorb more energy. The Charpy test can determine the relative toughness of bio-epoxy with different wt%. Low toughness materials tend to be more brittle and thus easier to break.

Figure 8 shows the impact energy versus the amount of bio-epoxy in the specimens, as indicated by the Charpy impact test. For laminated composite LC, it is normal that LC had higher impact energy compared to the pure matrix specimen as the fiber loading had improved the impact energy capability of a composite body. Because the fiber loading was constant for all specimens, only the bio-epoxy amount had varied its wt% from 0%, 25%, 30%, 40%, and 50%. For pure epoxy specimens, the highest energy absorbed during impact was 50 wt% and the lowest was 0 wt%, with 1% of difference. This indicated that higher bio-epoxy content led to slightly better energy absorption and thus more toughness in the matrix. This was due to its low brittleness compared to 0 wt% bio-epoxy. The tensile elongation at break results showed that higher amounts of bio-epoxy led to higher ductility. However, an acceptable amount of bio-epoxy needed to be chosen to avoid performance loss.

For specimens with carbon reinforcement, a similar trend of energy absorption was obtained from the impact tests, with lowest for 0 wt% and highest for 50 wt% bio-epoxy with 2.3% of difference. Moreover, the values of energy measured were a bit higher due to the presence of reinforcement inside the resin. The reinforcement played an important role in distributing impact load, which resulted in better energy absorption amounts. Furthermore, woven reinforcements also contributed to lessen the impact energy by damping the impact received throughout the continuity of its fiber length.

Because all specimens had the same woven reinforcement, nothing would differ except the amount of bio-epoxy inside the matrix. For pure resin, the failure mode of 0 wt%, 25 wt%, 30 wt%, and 40 wt% bio-epoxy involved complete breakage into two pieces, but 50 wt% broke but still held as one piece. A similar report on neat epoxy was obtained by Barbosa et al. (2014). The 50 wt% bio-epoxy was ductile and soft, which explained its ability to absorb high impact energy compared to others. Failure modes were the same for all specimens with reinforcement agent, where it started from the notch, followed by vertical interfacial fiber/matrix bonding failure, fiber breakage at the horizontal continuous fibers, matrix rupture, and then less fiber was pulled out. Furthermore, all of them were broken into two pieces. However, compared to pure matrix that failed to hold as one piece, for 50 wt% bio-epoxy with fiber carbon as reinforcement, the specimen was broken into two pieces.

The reinforcement acted as a barrier of the crack propagation initiated from the notch. The weave of 0°/90° provided a balanced resistance in horizontal and vertical breakage compared to uniform woven reinforcement even though it absorbed less energy during impact. Bio-epoxy improved the energy absorbed during impact and reduced brittleness with or without reinforcement.

Fig. 8. Impact energy of laminated carbon fiber composite and pure matrix, both with different wt% of bio-epoxy

Surface Morphology of Specimens with Ruptures

Micrograph images have a good advantage in giving a detailed observation on the morphology or the physical surface of a material at a small scale that is invisible to the naked eye. Images in Fig. 9 are 0 wt%, 18 wt%, 25 wt%, 30 wt%, 40 wt%, and 50 wt% bio-epoxy, with carbon fiber as reinforcement. They are images of tensile failure modes for several specimens at micrometer scale from SEM.