Strength analysis of marine biaxial warp-knitted glass fabrics as composite laminations for ship material

2.1 Material development

In this study, FRP composites were developed for ship hulls using marine-grade UPR. UPR has good mechanical properties, low in weight, non-corrosive, and less costly than epoxy resin [23], making it very suitable for FRP ship hull applications. The matrix has been formed from different marine-grade UPR brands, namely Yukalac 157 BQTN-EX (UPR brand A), which was supplied by PT. Justus Sakti Raya, Indonesia, and Everpol 324 AR-1 (UPR brand B) purchased from PT. Arindo Pacific Chemicals, Indonesia. Both brands are commonly used in FRP shipyards in Indonesia and have been approved by the Indonesian Classification Bureau (BKI) with material characteristics, as shown in Table 1. UPR A and B have similar specific gravity, tensile and flexural strength, but the viscosity is different, UPR A is 450–500 cps, and UPR B is 540 cps. Methyl ethyl ketone peroxide as a curing agent was obtained from PT. Kawaguchi Kimia Indonesia. Methyl ethyl ketone peroxide was added 5% by weight of the resin to help initiate the hardening process and control the pot life, gel time, and curing time in the fiber lamination process.

The types of fiberglass used in this study were Chopped Strand Mat 300 g/m² (CSM 300), Chopped Strand Mat 450 g/m² (CSM 450), Woven Roving 600 g/m² (WR 600), Woven Roving 800 g/m² (WR 800), biaxial warp-knitted fabrics 800 g/m² (Biaxial 800), double biaxial warp-knitted fabrics 800 g/m² (DB 800), and double biaxial warp-knitted fabrics 800 g/m² (DB 800/M225E). CSM and WR fiberglass fibers are supplied from PT. Justus Sakti Raya, Indonesia, while the Biaxial 800, DB 800, and DB 800/M225E fiberglass fibers were purchased from PT. Triaxis Composites, Indonesia. The Indonesian Classification Bureau (BKI) has approved all types of fiberglass fibers above. Warp- and weft-knitted structures [11], including stitch yarn on a biaxial warp-knitted fabric [24], are illustrated in Figure 1.

Figure 1

(a) Warp-knitted structures, (b) weft-knitted structures, and (c) biaxial warp-knitted fabric.

The 35 cm × 70 cm specimen sheet was designed using the hand lay-up technique and cured at room temperature, as illustrated in Figure 2(a), with a laminated arrangement as shown in Table 2. Specimen I was a laminated arrangement according to Indonesian yard practices, while specimens II, III, and IV were developmental lamination arrangements for ship hulls with biaxial wrap-knitted fabric fibers. The combination of CSM/WR and biaxial laminates was also applied to yacht shipbuilding [21]. A hand roller was used to press the resin onto the fiberglass and to remove air bubbles. The hand lay-up technique was applied in several studies [12,25] (Figure 2(b)). In contrast, the hand lay-up technique was often used in FRP shipyards, including yacht shipbuilding [21] and fishing boats [7].

Figure 2

(a) Schematic of hand lay-up technique and (b) specimen sheet fabrication process.

The fiber volume fraction was calculated using formula 1 [12,26] with a fiberglass density of 2.56 g/cm³ [18]. Thus, there were four lamination arrangements, namely I, II, III, and IV, with resins A and B and fiber orientation 0° (specimen code 0) and 90° (specimen code 90)

where B is the weight of one layer of fiberglass (g), ρ f is the density of fiberglass (gr/cm³), A is the surface area of the FRP laminate (cm²), and h is the thickness of the FRP laminate (cm).

The specimen sheet was cut according to the tensile test specimen’s and flexure test’s size and adjusted to the orientation of the fibers 0° and 90°. The cutting method was also carried out in previous studies [10,16,17] with the same type of material as this study, namely biaxial warp-knitted composite. These specimen variations were designed to determine the effect of resin type, fiberglass type configuration, number of layers/laminations, and fiber orientation on FRP composite ships’ tensile and flexural strength.

2.2 Experimental method

This study investigated the strength of FRP ship materials through a uniaxial tensile and three-point flexure test using a calibrated Universal Testing Machine (RME 100 Schenck Trebel) in a certified laboratory (KAN No. 077-IDN and KNAPP No. PL 007-INA, P68-01 a). A uniaxial tensile test was carried out referring to ISO 527-3 with a crosshead speed of 5 mm/min [27]. Figure 3 shows the tensile test specimen type 3, which had dimensions of overall length (L3) 250 mm, width (b1) 25 mm, and thickness (h) 8 mm, with a total of five specimens for each type. The reported data from the uniaxial tensile test data represented an average value of five specimens for each type with standard errors. The tensile loading can be seen in Figure 4, where the strain was measured using a strain gauge sensor, as was done in the study of Demircan et al. [12].

Figure 3

Dimensions of the uniaxial tensile test specimen [27].

Figure 4

Uniaxial tensile loading using UTM.

Meanwhile, the three-point flexure test refers to ISO 14125 with a strain rate of 0.01 (i.e., 1% per minute) [28]. The flexure test results were based on the average values of six specimen measurements for each type with standard errors. The specimen and scheme of three-point flexure referred to Class II for plastics reinforced with mats, continuous matting, and fabrics with specimen length (l) 80 mm, outer span (L) 66 mm, width (b) 15 mm, thickness (h) 4 mm, the radius of central loading (R1) 5 mm, and radius of supports members (R2) 5 mm, as shown in Figure 5. Normalized strength was calculated using formula (2) [18,29] so that the tensile and flexural strength had a value comparable to the same fiber volume fraction, namely 30%. Formula (2) was also used to analyze other studies’ tensile and flexural strength to compare it to this study

where E _normal (30%) (MPa) is the normalized strength at 30% volume fraction, E(V _f) is the tensile/flexural strength (MPa), and V _f is the volume fraction (%).

Figure 5

(a) Flexural test schematic [28] and (b) flexural loading using UTM.

3.1 Tensile properties

The results of the tensile test with standard errors on resins A and B are shown in Figure 6. All variations of the specimens showed adjacent volume fractions, which were between 27.06 and 31.89%. In other references, the difference in volume fraction variation of 10% indicates that the higher the volume fraction, the higher the tensile strength [30,31,32]. Figure 6 and Table 3 show something similar; the higher the fiber volume fraction, the higher the tensile strength of the same resin type. However, this trend did not apply to laminates IIA and IIB, which had pretty high tensile strength, even with a fiber volume fraction of 27.06%. The volume fraction of 29.69% in CSM–biaxial resin A had an average tensile strength of 135.2 MPa. The increase in the volume fraction of 0.76–30.45% increased the tensile strength to 145.1 MPa. A slight increase in volume fraction also impacted the tensile strength of CSM–biaxial resin B, where a rise of 0.39% increased the tensile strength from 148.9 to 152.3 MPa. So, the FRP with the CSM–biaxial fiber combination in this study, especially the lamination types I, III, and IV, had a tensile strength that was sensitive enough to be affected by volume fraction, which did not occur in vinyl ester resin/biaxial weft-knitted fiberglass materials using hand lay-up specimen fabrication method [12] and polyester resin/CSM-continuous fiberglass material using the injection molding fabrication method [33].

Figure 6

Effect of volume fraction on tensile strength: (a) Resin A and (b) Resin B.

In contrast, all specimens in this study met the minimum tensile strength criteria of 121.23 MPa as required by the Indonesian Classification Bureau [22], so the hand lay-up technique, material type, and laminate I–IV arrangement types in this study can be applied to FRP shipyard. The hand lay-up technique is the most applicable FRP shipbuilding technique for shipyards in Indonesia. Even yacht manufacturing in Italy still uses hand lay-up and not standardized procedures [21].

Normalized strength at a 30% volume fraction was used so that tensile strength had a comparative value to several tensile strengths between variations in this study and other studies, as can be seen in Figure 7 and Table 4. The average tensile strength of resin A was 160.7 MPa, while resin B of 160.5 MPa. Meanwhile, after normalization using formula (2), the average normalized tensile strength of all resin A specimens was 164.1 MPa, and resin B was 163.1 MPa, so that differences in resin brands did not affect the tensile strength. FRP yards could choose brand A or B based on price or ease of purchase for parts of the ship that were subject to frequent tensile loads.

Figure 7

Normalized tensile strength: (a) Resin A and (b) Resin B.

More specifically, specimens of the IIA and IIB laminate types had the highest tensile strength on average, 186.1 MPa. Specimens IIIA and IIIB had an average tensile strength of 176.7 MPa, specimens IA and IB had 147.8 MPa, and specimens IVA and IVB had 144 MPa. Specimens II and III, a combination of CSM and biaxial fiber laminations, had higher tensile strength, although the number of laminations was less than specimen I (CSM and WR fibers). Even specimen III (CSM–biaxial), which consisted of 7 layers, had a 19.5% higher tensile strength than specimen I (CSM–WR), which consisted of 12 layers. Specimen IV (CSM–biaxial) also consisted of seven layers and had almost the same tensile strength as specimen I (CSM–WR). Thus, the CSM–biaxial lamination type had better tensile strength than CSM–WR.

Biaxial warp-knitted fabric fiberglass had a significant effect on tensile properties. The average tensile strength of specimens II and III was better than that of specimen IV because specimens II and III had four layers of biaxial fiberglass. In comparison, specimen IV consisted of three layers of biaxial fiberglass. The CSM–biaxial fiber combination can speed up the production process of FRP ship materials because to get the same volume fraction, a smaller number of laminations is required compared to the CSM–WR fiber combination, which so far has been more frequently used by FRP shipyards, especially Indonesia. Thus, the CSM–biaxial fiber combination has the potential to be further applied to FRP vessels because it can accelerate production and has high tensile strength.

This result confirmed other studies that biaxial warp-knitted fabric fiberglass has high strength and stiffness [8], low production cost, and high production and efficiency [9]. These fibers have superior strength due to straight fiber bundles and have been widely applied in various applications [13]. However, these fibers have yet to be widely used to produce FRP vessels, especially in Indonesia.

Furthermore, the combined lamination arrangement, commonly called rovimat/combimat [7] as in Table 2, has been used in other references, namely the CSM–biaxial combination [21,25] and the CSM–WR combination [7]. Several CSM layers in this study and other studies [7,21,25] have a function to increase the resin content [7]. So that, this rovimat/combimat technique is widely applied to the marine industry, such as pleasure boats and fishing boats [7]. It was proven in this study that the specimens with relatively low resin content, specimens IVA and IVB, had not as good a tensile strength performance as specimens IIA, IIIA, IIB, and IIIB. This result showed that the biaxial warp-knitted fabric fiber should be combined with CSM fiber layers to increase the resin content and improve tensile properties.

In addition, Table 4 shows that specimens with 90° fiber orientation had greater tensile strength than 0° fiber orientation, both for resin A and B resin specimens. Specimens with 0° fiber orientation had an average tensile strength of 154 MPa, while fiber orientation 90° of 173.3 MPa. Stitch yarn provides more support for tensile loads in specimens with a fiber orientation of 90° than 0°, where the effect of stitch yarn is shown in the study of Zhao et al. [17]. Fibers with a fiber orientation of 90° in the outer layers of the laminate are in the direction of loading so that fibers in that direction will bear the tensile load and increase the load-carrying capacity of the composite. Warp fibers in that direction are also distributed more evenly [16].

For various FRP composite studies, normalized tensile strength was also used as a comparative value. The average normalized tensile strength of CSM–biaxial resin A (laminations II, III, and IV) is 171.2 MPa, while CSM–biaxial resin B is 166.6 MPa, as can be seen in Table 5. In Table 5, the CSM–biaxial composite in this study had better normalized tensile strength compared to several other studies, which were also made using the hand lay-up technique [30,32,34]. Compared to other studies using a matrix in the form of polyester resin, the CSM–biaxial composite in this study was still better [31,32,35]. Furthermore, the composites in this study had higher normalized tensile strength than other reference composites using more advanced fabrication techniques, such as spray forming [31] and contact molding [35]. However, the composites in this study were not better than those of previous studies [7,12,18,29,36–38], so that the CSM–biaxial composite can be classified as having normalized tensile strength at an intermediate level.

3.2 Flexural properties

Flexural strength with standard errors on resins A and B is shown in Table 6 and Figure 8. Figure 8 shows flexural strength was not correlated with variations in volume fraction with adjacent ranges. In general, the higher the volume fraction, the higher the flexural strength if the difference in fractional volume variation is around 10%, as in the previous studies [30,32,39]. The discrepancy in the correlation between the variation in the volume fraction range below 2.3% was also confirmed in other studies using better fabrication methods, namely the injection molding technique [33] and a hot-press machine [15]. Few references have examined the effect of flexural strength on FRP ships with low volume fraction range variations due to the difficulty of controlling fiber volume fraction over the overall volume. In addition, experimental variability has an influence that cannot be ignored [16], especially at low-volume fraction variations.

Figure 8

Effect of volume fraction on flexural strength: (a) Resin A and (b) Resin B.

Hence, to get better flexural strength, the fiber volume fraction should be increased by at least 10%. However, all variations in this study fully met the minimum flexural strength criteria of 116.15 MPa as required by the Indonesian Classification Bureau [22]. So the hand lay-up technique, material type, and laminate I–IV arrangement types in this study can be applied to the FRP shipyard. Even the lowest flexural strength obtained at the IA0o variation of 200.3 MPa is 72% above the minimum criteria.

Normalized flexural strength analysis was carried out to obtain comparative values between variations in this study and other studies, as shown in Table 7. After normalized flexural strength analysis, specimens using B resin showed a higher average normalized flexural strength of 288.8 MPa than resin A specimens, with an average normalized flexural strength of 269.3 MPa. Alternatively, in other words, resin B has 7.2% higher flexural strength compared to resin A. Thus, the flexural properties analysis shows that resin B was more recommended for FRP ship materials to have higher flexural strength.

Furthermore, flexural properties analysis further proved that the CSM–biaxial fiber combination was better than CSM–WR. Specimen I, a CSM–WR combination, had an average flexural strength of 214.3 MPa, while all CSM–biaxial specimens had a higher average flexural strength. The CSM–biaxial lamination type (specimens II, III, and IV) indicated that the greater the number of laminations, the higher the flexural strength. Specimen II (10 layers) had the highest average flexural strength of 319.2 MPa, followed by specimens III and IV (7 layers), which were 302.7 and 280 MPa, respectively.

Specimen IV, which only had seven layers, has 31% higher flexural strength than specimen I, which had 12 layers. Even the flexural strength of specimen III, which also consisted of 7 layers, was 41% higher than that of specimen I. Thus, the flexural strength of CSM–biaxial was much higher than that of CSM–WR, even though CSM–WR (specimen I) had more lamination. In addition, type IV biaxial composites had flexural strength, which was not as good as type II and III biaxial composites because specimen IV had low resin content due to a minimal CSM layer. In type II and III laminations, a layer of CSM was applied between the biaxial fiber layers to increase the resin content [7]. Thus, marine composites with biaxial fibers should be combined with CSM fibers to increase the resin content and result in better flexural properties.

In this study, flexural strength was more influenced by lamination quality. The biaxial warp layer of knitted fabric fiberglass significantly affected the flexural properties, which also happened to affect the tensile properties. Biaxial warp-knitted fabric fiberglass contributed significantly to flexural strength because this layer had high strength, high yarn utilization, high orientation, and stable size. It has high resistance to delamination [8], where delamination is often a reducing factor for flexural properties. The effect of through-layer binding from a knitted tricot provides excellent delamination resistance compared to conventional laminated fibers [9]. Fewer layers with better tensile and flexural properties allow CSM–biaxial fiber (lamination types II, III, and IV) to be applied in the FRP shipping industry. These advantages provide lower production time and labor costs. Not surprisingly, composites reinforced with biaxial warp-knitted fabric fiberglass are reported to have low production costs and high production efficiency [9].

Furthermore, the average flexural strength in the 90° fiber orientation was higher than the 0° fiber orientation for both resins A and B. Specimens with 90° fiber orientation had an average flexural strength of 287.8 MPa. In comparison, 0° fiber orientation was 270.3 MPa. The flexural strength of 90° is higher than 0° because, in the orientation of 90°, the biaxial fiber has stitch yarns that better support the increase in bending strength [10,14].

In Figure 9, the CSM–biaxial resin A composite (laminations II, III, and IV) had an average normalized flexural strength of 290.7 MPa. CSM–biaxial resin B had an average normalized flexural strength of 310.5 MPa. This value indicated that the flexural strength was relatively high compared to other references, as summarized in Table 8. Compared to other references that also use specimen fabrication techniques in hand lay-up, the composites in this study have higher normalized flexural strength [12,14,30,32]. This study also had better normalized flexural strength than [32,33,39], which used a polyester matrix. Even the CSM–biaxial composite (made by hand lay-up technique) had higher normalized flexural strength compared to composites fabricated using more advanced techniques, such as vacuum-assisted resin transfer molding [10], contact molding [39], and injection molding [33].

Figure 9

Normalized flexural strength: (a) Resin A and (b) Resin B.

In the future, the mechanical properties of the CSM–biaxial composite obtained in this study can be applied to FRP ship structure models to determine stress and structural deformation using the finite-element method, as shown in previous studies [40,41]. Structural dynamic analysis of the CSM–biaxial composite is also interesting to be developed in the future both experimentally [42] and numerically [43,44,45,46]. Furthermore, structural analysis of sandwich materials with core materials in the form of CSM–biaxial composite or CSM–WR needs to be studied, as in previous studies [47,48].