Influence of stacking sequence on mechanical properties and moisture absorption of epoxy-based woven flax and basalt fabric hybrid composites

: Hybrid composite laminates (HCL) were prepared by lay-up molding using a hot-press bed for reinforcing epoxy-based woven flax and basalt fabric composites. Mechanical properties and moisture absorption of HCL were measured, and the fracture surface was examined by scanning electron microscopy (SEM). The present results indicated that the mechanical properties of HCL are strongly dependent on the sequence of fiber reinforcement. HCL (S2, S3, S4, S5) with symmetric stacking sequences that increase with basalt fibers (wt %) showed a positive hybridization effect on mechanical properties and curve characteristics. However, the mechanical properties of S7 (asymmetric stacking) were lower than S3, S4, S5 (symmetric stacking), which indicated that symmetric stacking sequences of HCL had superior mechanical performance compared with the cross arrangement of HCL. The moisture absorption of HCL samples immersed in water at 26 °C showed the Fickian behaviour up to 42 days and were not affected by altering the stacking sequence on HCL. The SEM of the fracture surface, fiber-matrix bonding, and interfacial bonding of flax-basalt fabric HCL were also presented.


Introduction
Composite materials have gained growing interest in structural applications due to their improved performance as compared with single materials in the last few decades [1,2]. Typically, hybrid composites consist of at least two fibers bonded together in a common matrix (Fig. 1). A hybrid composite can be created from an artificial-artificial, natural-artificial or natural-natural fiber combination. Consequently, new composite materials are developed with new and enhanced properties [3]. One of the most effective ways to achieve desirable properties in laminated composites is hybridization [4]. Unlike single-type fiber-reinforced composites, hybrid composites combine two or more types of fibers to provide superior properties [5]. Through hybridization, one can benefit from the advantages of various fibers and eliminate their weaknesses, and this can improve the fracture toughness, fatigue resistance, as well as reduce the overall cost or weight of composite laminates [6].
Because composite materials are low-cost, have a high strength-to-weight ratio, and are easy to manufacture, they have wide application in engineering. Natural fibers as reinforcement materials for polymeric composites has gained increased interest in recent years [7,8] due to their low costs, good mechanical properties, high strengths, non-abrasive properties, eco-friendliness and biodegradability [9][10][11][12]. Natural fibers, on the other hand, have lower mechanical properties than synthetic fibers, including carbon fibers, glass fibers, aramid fibers, and so on. Moreover, the disadvantage of natural fiber composites is the poor resistance to moisture absorption [11][12][13][14], hence, a stronger and stiffer fiber is introduced into hybrid composite laminates (HCL) to enhance mechanical properties and reduce moisture absorption. Basalt fiber is obtained from basaltic volcanic rocks. It is of research importance due to its excellent chemical stability, superior temperature resistance, higher strength, modulus, and better strain to failure than carbon fiber as an inorganic fiber [11]. It is non-toxicity, resistant to weather, and has non-combustibility properties [15][16][17]. Basalt fiber is environment-friendly, with good mechanical properties and easy processibility [11,18]. Basalt fiber has been used to reinforce military-grade composites, structural retrofitting and seismic reinforcement [19]. It is an excellent reinforcement material in composites when combined with other fibers [20][21][22].
As regards hybridization and stacking sequences, previous studies suggested that degradation patterns are most favorable with an asymmetric configuration including a low strength material (glass) in the core of a high strength material (basalt) [23]. The fibers of the basalt fabric as the core and the glass fabric as the outer layer were reported to have high tensile strength reaching 356.39 MPa in composite reinforced with glass fiber fabric and basalt fabric with sandwich stacking sequence configuration [24]. One study suggested that external basalt fiber layers result in the greatest enhancement of flexural and tensile properties among various stacking sequences of glass/basalt [25]. In composites fabricated with alternate layers of basalt fiber and jute fiber, different layering patterns influence the tensile and flexural properties [26]. The stacking order provided higher strength in basalt/glass fiber intercalated hybrid laminates composite [24,26]. Basalt and jute fibers in an epoxy bio-based hybrid composite showed that inter-ply hybridization resulted in better mechanical properties from their stacking sequences [24,27].
In addition to basalt fibers, some researchers have used jute fibers, nylon, woven fabrics, and aluminum fibers as reinforcements in HCL structures [28,29], etc. Flax fiber is becoming increasingly promising as co-reinforcement in polymer composite due to its commercial availability, low cost and specific mechanical properties [30]. Many studies have reported on flax fiber-reinforced composites. For instance, it has been used as co-reinforcement in carbon and flax fabric reinforced hybrid composites [31], pineapple leaf-flax reinforced hybrid composite [32], as reinforcement in polyamide 11 resin [33], and Flax/Jute epoxy-based composites [14]. The effect of stacking sequence and hybridization on the properties of the hybrid composite material has only been assessed in a limited amount of studies involving flax fibers and basalt fibers. Hybridizing flax fibers with basalt fibers would reduce costs and increase the application scope of composite materials [34].
As a result of the above considerations, flax and basalt fabrics were used as reinforcements in epoxy resin matrix for HCL. This study examines how different stacking sequences affect woven flax and basalt fabrics' mechanical properties and water absorption.
The objective is to investigate the effect of different stacking sequences on the mechanical properties and water absorption of the woven flax and basalt fabric HCL.   The HCL series were fabricated manually using a hot press. The flowchart is shown in Fig.2(a) and Fig.2(b). Polytetrafluoroethylene (PTFE) film was applied to the surface of the mold. Woven basalt and flax fabric, cut to 220 mm × 220 mm, modified with silane coupling agent KH570 (G-570, γ-methacryloxypropyltrimethoxysilane) in 80/10 volume ethanol/water. The modified fiber was then stacked on a stainless plate mold (250 mm × 250 mm) in different sequences. The epoxy resin was brushed on to the flax and basalt fabrics (1:1 epoxy/hardener) and pressed in the mold for 6 minutes (5 MPa) before being removed. The mold was designed to allow hot gases to escape. We used spacers of the desired thickness to achieve uniform thickness on the mold plates. After 6 min, the HCL was removed from cold press and then cured for 40 min at 110 °C and 8 MPa, respectively. Then the HCL was removed from the hot press and cooled with the pressure at 5 MPa for 60 min at room temperature. Finally, the HCL was removed from the mold for characterization. -woven flax fabric, basalt flax fabric. For the present investigation, woven flax and basalt fabrics were stacked between ten layers in each composite laminate configuration consisting of flax, basalt and epoxy. Based on the arrangements in Fig. 3, we designed different woven flax, and basalt fabric layers to study the effect of stacking sequences. In each panel of HCL, flax and basalt fabrics made up 33.3 wt% of the weight as shown in Table 3. The total fiber volume fraction is calculated using Eq. (1): [35] f

Fabrication of HCL
Wf, Wb and Wr are the weights of the flax, basalt and resin, respectively. ρb, ρf and ρr are the densities of basalt, flax and density of resin, respectively. The density of epoxy resin is 1.18 g/cm 3 , as provided by the manufacturer. The density of basalt and flax are about 200 g/m 3 and 168 g/m 3 , respectively. Both of them were taken from the supplier's datasheet.

Characterization
Tests for tensile strength, flexural strength, impact strength, and water absorption were carried out according to ISO 527-4 (test conditions for fiber-reinforced plastic composites), ISO 14125 (plastic composites-determination of flexural properties), and ISO 179-1 (plastics-determination of Charpy impact properties, and ASTMD570, respectively).

Tensile test
A diamond wheel saw was used to carefully cut out pieces of laminate for the tensile test before finishing to the correct size with emery paper. The dimensions of the test specimen are 180 mm by 20 mm by 3 mm. The tests were conducted on an MTS Systems CMT6104 microcomputer controlled electronic universal testing machine (SHENZHEN, China). The load applies was 50 kN and a gauge length of 50 mm. A rate of loading of 10 mm/min was used for testing. The tensile strength (σt) of the HCL was determined according to Eq. (2).
Where F, b and d represent the fracture load, specimen width, and specimen depth. Five replicate specimens were tested for each stacking sequence, and the average result was used for the report.

Flexural test
Samples length and width of 70 and 15 mm were cut from laminate for the three-point bending with a span to depth ratio of 16:1. The test was conducted at a loading rate of 5 mm/min and load of 10 kN. The flexural strength (σF) and flexural modulus (EF) of the HCL were determined using Eq. (3) and Eq. (4).
Where L, b, d, P and m are the support span, width, depth, load, and the initial slope of the load-displacement curve, respectively. For each stacking sequence, five parallel measurements were performed.

Impact test
The unnotched specimens were used for the Impact tests with a length, width, and thickness of 80 mm, 10 mm and 4 mm respectively, and tested on a ZBC-25B Charpy Impact Tester (MTS System, Guangdong, China). The impact strength (acU) of the HCL was determined according to Eq. (5).
Where Ec, h and b represent, the test specimen's absorbed energy, thickness, and width. Five samples were tested for each case of stacking sequence.

Moisture absorption test
Samples of 76.2 mm (3 in.) long and 25.4 mm (1 in.) wide were used for the moisture absorption. After preconditioning at 70 °C for 24 hours, all samples were weighed accurately to 0.0001 g in an electronic balance with a single precision, cooled in desiccators, and weighed immediately (m1), and the samples were placed back in water. The samples were immersed in a water bath at 26 °C for each group of five samples. During the next 24 h, the samples were removed, the surface water was wiped off and the samples were weighed (m2), after that, the samples were immersed again in water bath at 26 °C. This procedure was repeated until the samples reached saturated point. The weight gained by the samples was carefully monitored. During the first week, the weight was checked everyday (24 h), then once a week for one month, and then once every two weeks until saturation was reached. This helped to closelt monitor the sample's weight gain. During this period, the water was changed regularly until saturation. The moisture absorption of the HCL was determined according to Eq. (6). Where m0, m1 and m2 represent the moisture absorption, conditioned weight, and wet weight.

Morphology of fracture surfaces
The fracture surface of the specimens was investigated using a scanning electron microscope (Nova Nanosem 230). All samples were sputtered with gold before SEM analysis.

Data analysis
Statistical analysis of mechanical properties was performed with SPSS 19 (IBM Corp., USA) using a one-way ANOVA at a 95% confidence interval.

Tensile properties
The changes in tensile strength and modulus for various laminates are shown in Fig. 4(a). The tensile strength and modulus of S6 (woven basalt fabric) are 212% and 483%, respectively, greater than S1 (woven flax fabric). The inclusion of basalt fiber as extreme basalt plies increased the tensile strength of the composite (S2, S3, S4, S5 and S6). Munikenche et al. [36] reported that the strength and modulus fibers has significant effect on the tensile strength of composites. According to the Rule of Mixture (ROM), Eq. (7) is given: Where Ec, Ef and Em are the moduli of composites, fibers and matrix, respectively. As a result, the hybrid composite had greater tensile and modulus strength due to basalt fibers having a higher strength and stiffness than flax fibers [37]. Compared to S1, the tensile strength and modulus of S5 (Flax fibers accounted for 19% of the total fibers) increased 154% and 417% respectively. Fig. 4(a) revealed that the HCL increases in tensile strength and modulus with the increase in basalt fiber content from 23% to 81% of the total fibers. The addition of glass fibers to jute fiber composites in isophthalic polyester in a previous study improved the tensile strength and tensile modulus of the composites [38]. At the same time, different stacking sequences on HCL can affect the mechanical properties. We observed that the tensile strength and tensile modulus of S7 were lower than those of S3. Interestingly, the proportion of basalt fibers in S7 (51%) is higher than in S3 (45%). The reason may be ascribed to the uneven loading of asymmetric stacking sequence of S7 to speed up the failure process of tensile fracture.

Flexural properties
The averaged flexural strength and modulus of the HCL are shown in Fig. 4(c). The flexural strength and modulus of S6 rank at the top, 111% and 222% higher than S1. The flexural strength and modulus of S5 are 173 MPa and 10.7 GPa, which is 57% and 49% higher than S2, respectively. Flexural strength and modulus of HCL are increased with the basalt fibers content from 23% to 81% of the total fiber weight. The initial slopes of S1, S2, S3, S4, S5, and S6 show a steep increase in Fig.  4(d). The corresponding two-order polynomial equation (y=0.0004x 2 +0.585x+97.27) and R 2 (0.9895) were obtained. No improvement in flexural properties was observed by altering the arrangement sequence, as was the case for S7. Among the HCL (S3, S4, S5), S7 has the lowest flexural strength (128 MPa) and modulus (7.8 GPa), similar to that of S3. These findings are in line with the findings of [29,39] that the modulus and flexural strength of HCL were strongly affected by the sequence of fiber reinforcement.
Note: S1 and S6 represent all flax and basalt laminates, respectively. The rest means HCL with woven flax and basalt fabric.

Impact properties
S5 has the highest impact strength of 167.4 kJ/m 2 , (Fig. 5(a)), 71% and 86% higher than S1 and S2, respectively. The impact strength of laminates gradually increased from S1 to S6, and this phenomenon agrees with the results of tensile and flexural properties presented in Fig. 5(b). The polynomial (y=-0.045x 2 +1.33595x+83.1747) and R 2 (0.90584) were obtained by curve-fitting impact strength versus basalt fiber content. According to the studies of Dehkordi et al. [3], which examined the impact properties of intra-ply hybrid composites based on nylon and basalt fibers, the number of basalt fibers in these composites significantly affected the impact performance. Although S7 contains more basalt fiber (wt%) than S3, its impact properties is much lower than S3. Higher impact strength is due to the basalt fiber arrangement on both sides of the HCL. Hence it is difficult for cracks to propagate. Fiber stacking sequence is highly important to HCL, whose layers have different fiber types [23].
(a) (b)  The moisture absorption curves of HCL are shown in Fig. 6(a). Every point is the average of five data sets. There was a linear portion to the curves initially, but all composites reached saturation after 42 days of immersion and showed a typical Fickian behavior. During the early phase of immersion, the water absorption of natural fiber composites increased linearly according to Fickian law [40], the composites in this study follow this law. S1 has the highest moisture absorption in all the composites. The optimum moisture absorption of S1, S2, S3, S4, S5 and S7 were 13.72%, 5.72%, 4.87%, 4.68%, 4.11% and 3.44%. The moisture absorption of S6 HCL was lower than all the HCL. The reason is that basalt fiber is an inorganic and hydrophobic material consisting of metallic oxides, such as SiO2, Al2O3, FeO, Fe2O3, MgO, K2O, Na2O and so on [15]. In this way, basalt fiber acts as a barrier and prevents direct contact between flax and water. Fig. 6(b) shows that adding woven basalt fabrics to extreme surfaces reduces moisture absorption considerably for various immersion times. Flax is highly hydrophilic nature and is accountable for the high moisture absorption in all composites. By altering the sequence of arrangement in S7, no change in the water absorption was found, similar to the case of S3. It was revealed that the water absorption of woven flax and basalt fabric HCL was associated with the type and proportion of the fiber rather than the stacking sequence.

Morphology of fracture surfaces
Each fiber within a composite laminate has a unique fracture characteristic, such as fiber pull-out, breakage, debonding, etc. Failure of a laminate depends on the maximum bending moment the fibers in the laminate can handle. To provide insight into how the samples are damaged following flexural testing, optical and scanning electron microscope images were captured of the front and fractured surfaces of the samples. Flax fibers were pulled out and fractured from Fig. 7 (A and A1). S1 shows poor interfacial adhesion between fiber and matrix. When basalt fiber layers are sandwiched between the top and bottom of the flax fiber layers, for S5, no basalt fiber was pulled out, and the matrix was not destroyed (Fig. 7(B)). This may be due to basalt fiber's higher strength and modulus than flax fiber [33]. A closer look in Fig. 7 (B1 and B2) reveals a few defects, such as debonding and cracks between flax fibers and matrix. The reason is that basalt fibers of the outer layer have absorbed most of the applied load, resulting in the protection of the flax layer. For the basalt/epoxy composites, fiber fracture and matrix fragment were observed in Fig. 7(B1), indicating that the crack of the matrix under the external force resulted in the collapse of the composites. S6 shows that the interfacial adhesion between basalt fiber and matrix is better than between flax fiber and matrix, as shown in Fig.  7(C). So the S6 has the highest mechanical performance.

Conclusions
Based on the results from this study, it can be concluded that: (1) Incorporation of basalt fabric in flax fabric HCL improved the mechanical properties and resistance to moisture absorption of the epoxy-based woven flax and basalt fabric composites.