The Delamination Behaviour of Basalt Fibre-Reinforced In Situ-Polymerisable Acrylic and Epoxy Composites: A Sustainable Solution for Marine Applications

Abstract

This research paper employed novel sustainable alternative materials to reduce the environmental impact of thermoset/synthetic fibre composites. The effect of seawater hydrothermal ageing at 45 °C for 45 and 90 days on the physical and interlaminar fracture toughness (mode I and mode II) of a semi-unidirectional non-crimp basalt fibre (BF)-reinforced acrylic matrix and epoxy matrix composites was investigated. Optical and scanning electron microscopes were used to describe the fracture and interfacial failure mechanisms. The results show that the BF/Elium composite exhibited higher fracture toughness properties compared to the BF/Epoxy composite. The results of the mode I and mode II interlaminar fracture toughness values for the BF/Elium composite were 1280 J/m² and 2100 J/m², which are 14% and 56% higher, respectively, than those of the BF/Epoxy composite. The result values for both composites were normalised with respect to the density of each composite laminate. The saturated moisture content and diffusion coefficient values of seawater-aged samples at 45 °C and room temperature for the BF/Elium and BF/Epoxy composites were analysed. Both composites exhibited signs of polymer matrix decomposition and fibre surface degradation under the influence of seawater hydrothermal ageing, resulting in a reduction in the mode II interlaminar fracture toughness values. Enhancement was observed in mode I fracture toughness under hydrothermal ageing, particularly for the BF/Epoxy composite, due to matrix plasticisation and fibre bridging.

1. Introduction

Over the past decade, the circular economy has emerged as a central theme in sustainability discussions, driven by the goal of achieving full circularity and net-zero greenhouse gas emissions by 2050. The European Commission specified in its European Plastics in a Circular Economy Strategy that 55% of all plastic and composite waste must be reused or recycled by 2030 [1]. The environmental impact of thermoset composites poses a considerable challenge for society and industry, prompting the exploration of sustainable alternatives to mitigate this impact. Recent studies have investigated the use of a novel thermoplastic liquid resin, Elium, as an alternative to various thermosets, including epoxy, unsaturated polyester, and vinyl ester resins, for lightweight structural engineering applications in the marine sector, such as boatbuilding, shipbuilding, offshore structures, submarines, underwater components, aquaculture, and marine farming [2,3,4,5]. For instance, Elium resin has recently been used with glass fabric to fabricate large boats and wind turbine blades (WTBs) [6,7,8]. Additionally, Smith et al. [9] investigated the exothermic polymerisation behaviour of two low-exotherm Elium^® grades in thick-section laminates for tidal turbine blade applications. Elium^® 188 XO showed a higher exothermic peak than Elium^® 191 XO/SA, though both remained below critical boiling points. The results offer practical insights into resin selection and process control for large composite structures.

Elium resin is a liquid thermoplastic methyl methacrylate (MMA) resin developed by Arkema in France. Based on polymethyl methacrylate (PMMA), it is uniquely infusible and offers enhanced ductility and toughness. Elium-based composites are fully recyclable through the depolymerisation of the thermoplastic matrix. In addition, the formulation of Elium resin does not include cobalt salts, which are commonly used as catalysts to initiate radical polymerisation and are often classified as carcinogenic in Europe (CMR substances) [10,11]. Additionally, it does not contain styrene, which is toxic to reproduction and recognised as hazardous to human health [12]. Both Elium resins 188 XO and 191 XO/SA are low-exotherm, low-viscosity, room-temperature-curing resins that have been used for structural thick-section infusion components, such as those found in WTBs and boats. Elium is an in situ polymerisable, infusible thermoplastic resin that can be recovered and recycled at the end of its life cycle, supporting the circular economy by reducing landfill waste. The 188 O and 188 XO resins have received certification from DNV-GL and have been applied in WTB manufacture following their use in the ZEBRA (Zero Waste Blade Research) project. However, similar extensive characterisation work has not been published on Elium^® 191 XO/SA. For this reason, Elium 191 XO/SA has been selected in this investigation to assess the resin’s potential for use in similar applications [12,13].

The most commonly used reinforcing fibres are synthetic materials, such as those made from glass and carbon, both of which have a high carbon footprint. The increase in glass fibre (GF) production can have environmental impacts, as one ton of GF results in approximately one ton of CO₂ emissions, along with other greenhouse gases, due to the significant energy demands associated with its production. Carbon fibre (CF) is produced by refining petroleum to extract acrylonitrile, which is then utilised to create carbon yarn. The production of one ton of CF emits approximately 20 tons of CO₂, mainly due to the high manufacturing temperature, which reaches around 1000 °C. In this context, producing one ton of basalt fibre (BF) generates approximately 398 kg of CO₂ emissions. BF, derived from natural volcanic rock, offers benefits, including high strength, stiffness, chemical resistance, and thermal stability. The chemical composition of BF used in this study is comparable to that of GF (

Table 1. Comparison of the chemical composition of BF in this study with that of GF [14].

Composition	SiO2	Al2O3	Fe2O3	MgO	Na2O	TiO2	K2O	B2O3	F
BF Weight (%)	57.5	16.9	9.5	3.7	2.5	1.1	0.8	-	-
GF Weight (%)	55	15	0.3	3	0.8	-	0.2	7	0.3

).

BF production does not involve harmful chemicals or additives, unlike GF, making it a safer material for workers and the environment. Therefore, BF can be used as a sustainable alternative to traditional E-glass, S-glass, and CF, serving as a cost-effective, eco-friendly natural material to reinforce various lightweight structural applications [15,16,17].

Table 2. BF’s physical and mechanical properties compared to other synthetic fibres [14].

Property	E-GF	S2-GF	AF *	CF	BF
Density (gr/cm3)	2.55–2.58	2.45	1.45	1.74–1.80	2.67
Modulus (GPa)	78–80	91	70–140	200–250	85–89
Strength (MPa)	2000–2500	2000–2200	2900–3600	2700–3750	2900–3100
Moisture (%)	0.1	0.1	3.5	0.1	0.008

* AF is Aramid fibre.

presents BF’s physical and mechanical properties compared with other synthetic fibres. Additionally, BF is non-toxic and non-combustible, with no additives required during manufacturing. BF exhibits high strength and modulus, good resistance to thermal corrosion and moisture, and outstanding chemical stability, and is considered a sustainable alternative to E-glass [18,19].

Recently, investigating the mechanical properties of Elium-based composites has attracted substantial attention to gain a deeper understanding of their behaviour and enhance their effectiveness under various loading conditions. For instance, Alsaadi et al. [20] investigated the mechanical performance of basalt fibre-reinforced Elium^® 191 XO/SA and Epoxy Prime™ 37 composites under hydrothermal seawater ageing at 45 °C. The BF/Elium composite showed significantly higher tensile and flexural strength (up to 33% and 71% greater, respectively) than the BF/Epoxy composite. Ageing effects led to resin degradation and fibre surface damage, with flexural properties being more affected than tensile properties. However, a few researchers have examined the interlaminar fracture toughness (IFT) characteristics of Elium-based composites. Ning et al. [21] conducted mode I-IFT and mode II-IFT tests on UNI-GF with an areal density of 750 g/m², reinforced with Elium 150 resin and found that the processing temperature significantly influences the IFT. The authors reported that the maximum mode I-IFT and mode II-IFT values were 1.56 and 2.14, respectively. Obande et al. [22] investigated the mode I-IFT of GF–Elium and GF–epoxy composites with a GF areal density of 646 g/m², using Elium 188 O resin as the acrylic-based polymer. The findings revealed that the GF–Elium composite exhibited a 15% higher mode I-IFT than the GF–epoxy composite. Bhudolia et al. [23] reported that the hybrid carbon–ultrahigh molecular weight polyethylene (UHMWPP)–Elium composite demonstrated a 22.81% higher mode I-IFT and a 22.2% higher mode II-IFT compared to the carbon–UHMWPP–epoxy composite. The study utilised Elium 150; both fibres were woven hybrid fibres (50:50) with an areal density of 200 g/m². Bandaru et al. [24] employed non-crimp fibre (NCF) quadriaxial (QA) GF with an areal density of 1184 g/m² reinforced with Elium 150 using the VARTI process. The authors reported that the mode I-IFT and mode II-IFT values were 0.75 kJ/m² and 4.33 kJ/m², respectively.

The hydrothermal ageing of fibre-reinforced polymer composites (FRP) can lead to moisture absorption, as water molecules infiltrate the material’s layers. This penetration initiates hydrolysis, a process that breaks covalent bonds within the polymer chain, resulting in swelling, debonding, and a weakening of interlaminar bonds. These changes degrade the fibre surface and decompose the resin, ultimately causing interfacial failure [25,26,27]. Furthermore, moisture absorption causes matrix swelling, resulting in a decrease in the glass transition temperature (T_g) and an increase in the coefficient of thermal expansion. At the same time, shear strength and fracture toughness were deteriorated [28,29]. Understanding the impact of hydrothermal ageing on the physical and IFT properties of FRP is therefore essential to reveal the underlying ageing mechanisms. Bandaru et al. [24] employed a quadriaxial NCF BF/Elium composite and found that hydrothermal ageing degraded the mode II-IFT by 37%, while the mode I-IFT of aged samples was 90% higher than that of the unaged composites, indicating significant fibre bridging induced by ageing. González et al. [30] examined the effect of seawater ageing at 70 °C on the mode I-IFT and mode II-IFT behaviour of GF–Epoxy and found that seawater ageing increases the mode I-IFT by 50% higher than that of the control composites due to matrix plasticisation and fibre bridging. Dhakal et al. [31] investigated the influence of water absorption on the mode I-IFT of flax fibre and BF–vinyl ester hybrid composites. The results showed that water immersion improves the mode I-IFT, highlighting the positive effects of basalt fibre hybridisation on the durability and moisture resistance of natural fibre composites.

Studies in the literature have demonstrated that Elium-based composites can offer excellent recyclability and durability in FRP, highlighting their strong suitability for various engineering applications, such as the marine and tidal turbine blade sectors. These Elium-based composites must endure long-term seawater immersion without significant mechanical degradation. To the best of the authors’ knowledge, no previous studies have examined the mode I-IFT and mode II-IFT properties of semi-unidirectional non-crimp basalt fibre (NCF BF) composites reinforced with novel Elium^® 191 XO/SA resin and Prime^™ 37 epoxy resin under the influence of hydrothermal seawater ageing. The physical, mode I-IFT, and mode II-IFT behaviours of semi-UNI NCF BF-reinforced Elium matrix and epoxy matrix composites have been investigated. The deterioration of the mode I and mode II interlaminar fracture toughness properties of the BF/Elium and BF/Epoxy composites under the effect of seawater ageing at 45 °C for 45 and 90 days was analysed. The modes of failure and fracture were explored using an optical microscope (OM) and a scanning electron microscope (SEM) to demonstrate the toughening mechanism of each composite laminate.

2. Materials and Methods

2.1. Materials and Manufacturing Process

Two resins were used to conduct the infusion and manufacture the composite laminates: a recent novel recyclable thermoplastic resin (Elium^® 191 XO/SA) from Arekma, Colombes, France, and a thermoset epoxy resin (Prime^TM 37) from Gurit, Newport, UK, both of which are suitable for large composite components due to their low exotherm and high gel time.

Table 3. Typical physical and mechanical properties of the Elium and epoxy resins.

Property	Elium® 191 XO/SA	Epoxy PrimeTM 37
Viscosity (cP) at 25 °C	100	181
Density (g/cm3) at RT	1.01	1.10
Gel time (min) (MEKP) at 25 °C	210–260	240
Curing Process at RT	24 h infusion and cure	24 h infusion and cure
Post-curing	2 h at 80 °C	16 h at 50 °C
Tensile Strength (MPa) (ISO527)	47.10	72.50
Tensile Modulus (GPa) (ISO527)	2.68	3.21
Flexural Strength (MPa) (ISO178)	80.51	113
Flexural Modulus (GPa) (ISO178)	2.77	3.01

presents the typical physical and mechanical properties of the Elium and epoxy resins, as provided by the manufacturer’s datasheets.

Two semi-UNI NCF (untwisted roving fabric arranged in parallel at 0° and 90°) reinforcements of BF, namely, BAS-UNI 550 and BAS-UNI 350 (Figure 1), featuring multi-compatible sizing, were used as received from Basaltex, Wevelgem, Belgium.

Table 4. Constituents and structure of the composite samples.

Composite	Basalt Fibre Structure	Polymer Matrix
BF/Elium	BAS-UNI 550: - 0° 520 g/m2 - 90° 50 g/m2 - Stitching 8 g/m2	Elium® 191 XO/SA, three parts: 191 XO acrylic resin (50 wt%), 191 SA accelerator (50 wt%), and methyl ethyl ketone peroxide (MEKP) initiator (2 wt%)
BF/Epoxy	BAS-UNI 350: - 0° 357 g/m2 - 90° 50 g/m2 - Stitching 9 g/m2	Epoxy PrimeTM 37, Two parts: Epoxy resin (100 wt%) and slow hardener Ampreg 3X (29%w)

shows the constituents of the composite laminates. The BAS-UNI 550 (578 g/m²) consists of 90% of the total areal weight of the fabrics aligned with the 0° direction, 8.6% aligned with the 90° direction, and 1.4% is a polyester stitching that offers stability to the fabric, helping to prevent the misalignment of the 0° fibres. On the other hand, BAS-UNI 350 (416 g/m²) consists of 86% of the total areal weight of the fabrics aligned with the 0° direction, 12% aligned with the 90° direction, and 2% is polyester stitching.

The manufacturing process began by laying the dry BF plies on a flat steel plate to impregnate them with liquid thermoset or thermoplastic resin at room temperature (RT) using the VARTI technique. The VARTI process (Figure 2a,b) requires a vacuum pump, vacuum bag, a flow mesh, peel ply, and a resin delivery system (pot, spiral wire, silicone tubes, valves, and connectors). Prior to infusion, a mandatory leak/vacuum test must be performed, ensuring that the vacuum drop does not exceed 5 mbar within a 10 min period. Both resin compositions were degassed prior to infusion. Specifically, the resins were degassed for 5 min at 900 mbar using a vacuum chamber to remove entrapped air and minimise void formation during infusion. The composite panels produced for this study measured 800 mm × 200 mm. For the vacuum infusion process, a single spiral inlet tube was positioned along one edge of each panel, while the outlet was placed on the opposite edge to promote uniform, linear resin flow across the fibre preform. Figure 2b illustrates the vacuum infusion layout. A slow resin infusion strategy was employed to ensure the complete impregnation of all fabric layers. The infusion time for the BF/Elium^® panels was approximately 60 min, while the BF/Epoxy panels required around 50 min. This slight variation in infusion duration is attributed to the different viscosities of the two resin systems. The recorded infusion times were consistent across repeated fabrication runs for each material. The vacuum draws the liquid resin from the pot into the vacuum bag, allowing the resin to impregnate and consolidate with the BF plies. Due to Elium’s low density, a low vacuum infusion pressure of 700 mbar was selected until the reactivity of Elium (gel time) was achieved, compared to a vacuum infusion pressure of 900 mbar. Table 3 and Table 4 show the mixing ratios, gel time, curing, and post-curing process.

2.2. Testing Procedures

2.2.1. Testing of Density, Voids, and Fibre Volume Fraction

The average density (ρ_c), void volume (V_v), fibre volume fraction (V_f), and glass transition temperature (T_g) of the unaged composites are shown in

Table 5. Physical properties for each composite type.

Composite	Vf (%)	Vv (%)	ρc (g/cm3)	Tg °C
BF/Elium	53.5	1.3	1.96	106
BF/Epoxy	47.8	3.5	1.84	71

. The panels were manufactured from eight plies of BAS-UNI 550 for the BF/Elium composites and ten plies of BAS-UNI 350 for the BF/Epoxy composites to control and reduce the thickness difference between the two composite panels. The thickness values of the BF/Elium and BF/Epoxy samples were 3.39 ± 0.04 mm and 3.33 ± 0.02 mm, respectively. The densities of the BF/Elium and BF/Epoxy composites were measured via a displacement method (ASTM D792 standard [32]) using a digital solid density meter. Six specimens were taken from different panels, and the average values of V_f, ρ_c, and V_v were determined. The V_f and V_v of the composites were determined using a matrix burn-off method [33] in accordance with ASTM D3171 standard [34]. The mass of the dry sample was weighed to the nearest 0.0001 g before and after placing the sample in a ceramic crucible in an oven at 600 °C for 2.5 h. This process removed the Elium and epoxy matrices, leaving only the BF. The ρ_c, V_v, and V_f values of the BF/Elium were 1.96 g/cm³, 1.3%, and 53%, while those of the BF/Epoxy composite were 1.84 g/cm³, 3.5%, and 48%, respectively. Hence, the ρ_c and V_f values of the BF/Elium composite were higher than those of the BF/Epoxy composite by 7% and 12%. These increments were primarily due to the areal density of the BAS-UNI 550 fabric in the BF/Elium being higher than that of the BAS-UNI 350 fabric in the BF/Elium laminate by 39%, in addition to the effect of the density of both matrices. The T_g values for BF/Elium and BF/Epoxy composites were 71 °C and 106 °C, respectively, as determined by dynamic mechanical analysis (DMA).

2.2.2. Testing of Moisture Absorption

The samples of each composite were dried in an oven for one hour at 40 °C. Then, the samples were immersed in synthetic seawater at RT and 45 °C in a Memmert water bath machine for 0, 45, and 90 days until complete saturation in order to assess the hydrothermal accelerated ageing performance. According to estimates often made using Arrhenius-based models, accelerated ageing for 90 days at 45 °C in seawater is roughly equivalent to about 5.4 years of natural ageing at 12 °C (the typical ocean temperature in Northern Europe), assuming an activation energy of 70 kJ/mol as mid-range ageing for acrylic-based and epoxy-based composites. Moisture ageing and exposure conditions were conducted in accordance with the ASTM D5229 standard [35]. This method involved removing samples from the water and weighing them at regular intervals using a digital balance with a precision of 0.0001 g. This process measured the mass of moisture absorbed or the weight gained over a specified duration. Based on the ASTM D5229 standard and Fickian diffusion theory, Equations (1) and (2) were used to calculate the equilibrium moisture content (M_c) and the diffusion coefficient (D_z) for the BF/Elium and BF/Epoxy composites.

$$M_{\mathrm{c}}(\%)={\displaystyle \frac{{\mathrm{m}}_i-m_o}{m_o}}\times 100$$

(1)

$$M_{\mathrm{\infty }}({\mathrm{m}\mathrm{m}}^2/\mathrm{s})=\mathrm{\pi }{\left({\displaystyle \frac{\mathrm{h}}{{4M}_m}}\right)}^2{\left({\displaystyle \frac{{\mathrm{M}}_2-{\mathrm{M}}_1}{{\sqrt{\mathrm{t}}}_2-{\sqrt{\mathrm{t}}}_1}}\right)}^2$$

(2)

where m_o represents the mass of the dry sample, m_i is the mass of the wet sample, and M₂ − M₁ is the slope of moisture absorption, corresponding to times t₂ and t₁.

2.2.3. Testing of DCB Samples

The DCB and ENF tests were conducted at RT using the Zwick Roell Universal Testing Machine, equipped with a 45 kN load cell (GmbH & Co. KG, Baden-Württemberg, Germany). The IFT is the energy release rate required to resist delamination growth during crack propagation in a pre-cracked body. The mode I-IFT (G_IC) of the BF/Elium and BF/Epoxy composites was determined via DCB test, as per the ASTM D 5528 standard [36]. The DCB specimens were cut to dimensions of 150 mm × 20 mm. A Permabond TA4610 adhesive was used to adhere aluminium loading blocks measuring 20 mm × 25 mm × 12 mm to the front and back sides of the DCB specimens. The Teflon film was inserted to generate a crack length a of 50 mm. Figure 3a shows the configuration of the DCB sample.

The crosshead speed was 5 mm/min, as specified in ASTM D5528. The crosshead movement was considered the specimen’s crack opening displacement (COD). A digital camera was employed to record the crack propagation length. The data were taken in terms of P-COD values and corresponding P-a values. P represents the applied load at which crack propagation arises, while a denotes the length of the crack extension. The G_IC is calculated using the following equation derived from linear elastic fracture mechanics [36,37,38]:

$$G_{IC}={\displaystyle \frac{3P\delta }{2b(a+\left|∆\right|)}}$$

(3)

where $\delta$ is the COD, b is the sample width; the crack length correction (Δ) was determined by plotting the cube root of the compliance, C^1/3 (where C is equal to δ/P), as a function of a.

2.2.4. Testing of ENF Samples

The mode II-IFT (G_IIC) was calculated using a three-point bend test on the ENF test samples as per the ASTM D7905/D7905M-19 standard [39]. The ENF test sample was prepared in a 140 mm × 20 mm size with a span length and a/L of 100 mm and 0.5, respectively (Figure 3b). Where L represents the half-span length and a is the pre-crack length. The loading rate under displacement control was 0.8 mm/min. During the ENF test, the specimen created shear stress at the crack tip, and the load dropped to its maximum when crack propagation began. The direct beam theory, as per ASTM D7905/D7905M-19 standard, was employed for calculating G_IIC via the following equation [39,40,41]:

$$G_{IIC}={\displaystyle \frac{9P\delta a^2}{2b(2L^3+3a^3)}}$$

(4)

3. Results

3.1. Physical and Moisture Absorption Performance

The physical properties of BF/Elium and BF/Epoxy for 0-, 45-, and 90-day ageing at 45 °C are presented in

Table 6. Physical properties of BF/Elium and BF/Epoxy for 0-, 45-, and 90-day ageing at 45 °C.

Composite	Vf (%)	h (mm)	Increment (%)	ρc (g/cm3)	Increment (%)
BF/Elium-0D	53.5 ± 0.01	3.39 ± 0.04	-	1.96 ± 0.02	-
BF/Elium-45D	-	3.41 ± 0.01	0.51	1.99 ± 0.03	1.6
BF/Elium-90D	-	3.43 ± 0.03	1.03	2.01 ± 0.02	2.4
BF/Epoxy-0D	47.8 ± 0.01	3.33 ± 0.02	-	1.89 ± 0.01	-
BF/Epoxy-45D	-	3.37 ± 0.03	1.12	1.87 ± 0.03	1.75
BF/Epoxy-90D	-	3.38 ± 0.04	1.65	1.90 ± 0.04	3.2

. Hydrothermal ageing resulted in a thickness increase of 1.03% and a density increase of 2.4% in the BF/Elium composite. In comparison, the BF/Epoxy composite exhibited increases in thickness and density of 1.65% and 3.2%, respectively, after 90 days of exposure to seawater at 45 °C. These changes are primarily attributed to moisture absorption and swelling, which cause an increase in thickness, and to post-curing (In thermosets) and crystallisation effects (In thermoplastics), which contribute to the increase in density. Similar trends have been reported in the literature; for instance, hygrothermal ageing led to a 3.13% increase in thickness and a 4.14% increase in density in the GF–Elium composite [24].

On the other hand, the moisture absorption performance of the acrylic- and epoxy-based composites at RT and 45 °C ageing temperature is shown in Figure 4. All the samples reached saturation after about 84 days (2016 h) of immersion. A gradual increase in moisture uptake (as weight gain) occurred during the 64 days, with absorption plotted against the square root of time in Figure 4. The absorption rate slowed progressively until it reached a state of full saturation. This trend aligns with Fick’s law, indicating that moisture diffusion in the composites follows Fickian behaviour. As moisture penetrated the Elium and epoxy matrix, it caused matrix swelling and an apparent increase in specimen weight. This swelling can induce internal stresses that lead to micro-cracking, providing additional pathways for moisture ingress. Over time, as the composites approached saturation, the absorption rate decreased and eventually plateaued, indicating the limit of moisture uptake. The saturated moisture content and diffusion coefficient values of seawater-aged samples at RT and 45 °C for the BF/Elium and BF/Epoxy composites are presented in

Table 7. Moisture content and diffusion coefficient values of aged composite samples at RT and 45 °C compared with other studies in the literature that used the VARTI manufacturing process and seawater ageing.

Composite	Mc (%)	Increment (%)	Dz (mm2/s)	Increment (%)	Source
BF/Elium-RT	0.91 ± 0.02	-	0.19 × 10−6	-	Current study
BF/Elium-45 °C	1.12 ± 0.05	22.9	0.20 × 10−6	9.1	Current study
BF/Epoxy-RT	1.09 ± 0.03	-	0.18 × 10−6	-	Current study
BF/Epoxy-45 °C	1.48 ± 0.07	35.9	0.19 × 10−6	6.3	Current study
BF/Epoxy-35 °C	1.97	-	0.07 × 10−6	-	[42]
BF/Epoxy-40 °C	1.5	-	19 × 10−6	-
GF–Elium-50 °C	0.81	-	1.81× 10−6	-	[44]
GF–Epoxy-50 °C	0.63	-	0.15 × 10−6	-	[44]
Bulk Elium-60 °C	1.90	-	4.23 × 10−6	-	[45]

. The BF/Elium and BF/Epoxy composites exhibited similar absorption trends in both immersion temperature environments. However, the BF/Epoxy composite displayed a comparatively higher weight gain of 19% and 31% than the BF/Elium composite at RT and 45 °C, respectively, which can be attributed to the lower V_f and higher V_v values of the BF/Epoxy composite [42]. Similar results were found in the literature, indicating that GF–epoxy composites absorbed 24% more water compared to GF–Elium composites [43].

The BF/Elium samples exhibited lower moisture uptake, with M_c values of 0.91% and 1.12% at room temperature (RT) and 45 °C, respectively, compared to those of BF/Epoxy, which had M_c values of 1.09% and 1.48% at RT and 45 °C, respectively. On the other hand, the M_c of the BF/Elium samples increased by 22.9% at the hydrothermal ageing temperature of 45 °C, while that of the BF/Epoxy samples increased by 35.9%, compared to those values taken at RT. On the other hand, the BF/Epoxy samples showed relatively lower moisture diffusion coefficients at both temperature conditions. Hence, the diffusion coefficients of the BF/Elium exhibited an increase of 9.1% at 45 °C, while that of the BF/Epoxy exhibited an increment of 6.3%, compared to their values at RT. Compared with other studies in the literature that used the VARTI manufacturing process and seawater ageing (Table 7), it was noted that there is a variation in the diffusion coefficient values of the BF/Epoxy in the literature, which may be attributed to several factors, including high void content, the type of matrix, and fibre sizing [15,42].

3.2. Mode I-IFT Performance

The DCB sample was subjected to mode I opening delamination loading that initiated and propagated a crack from the crack tip at the inserted film along the 50 mm delamination length. This DCB test resulted in load–COD curves (Figure 5), which were used to calculate the mode I-IFT values (G_Ic) and the normalised mode I-IFT values (G_Icn). As shown in this figure, both composites exhibit a linear load–COD performance up to the crack initiation point at the crack tip (the tip of the film insert). After that, these curves display a non-linear crack growth behaviour up to the maximum load point and beyond the crack propagation stage. This behaviour indicates steady-state crack growth, followed by a significant increase and subsequent decrease in load values (zigzag) [46,47,48].

Table 8. The mode I-IFT properties under the effect of seawater ageing.

Composite	Pmax (N)	COD (mm)	GICO (J/m2)	R* (%)	GIC (J/m2)	R* (%)
BF/Elium-0D	39.2 ± 2.3	13.1 ± 1.2	816	-	1242	-
BF/Elium-45D	41.1 ± 1.6	14.3 ± 2.1	862	5.6	1060	14.7
BF/Elium-90D	32.7 ± 3.4	11.9 ± 1.8	611	−25.1	831	−33.1
BF/Epoxy-0D	33.3 ± 3.8	14.5 ± 1.5	773	-	1079	-
BF/Epoxy-45D	47.9 ± 4.7	25.4 ± 2.4	1026	32.7	1182	9.5
BF/Epoxy-90D	55.1 ± 1.9	31.1 ± 2.7	1154	49.3	1471	36.3

R* is the increment or decrement.

presents the load–COD and mode I-IFT values. Additionally, it was observed that the load values required for BF/Elium at the same COD were higher than those for BF/Epoxy. The average crack initiation load obtained for the BF/Elium composite is 39.2 N, which is 11% greater than that of the BF/Epoxy composite. There was a slight difference in the COD between the two unaged composites. It is also noted from Figure 5 that the load for BF/Elium was suddenly reduced at the end of crack propagation, while it was gradually reduced for BF/Epoxy. The same zigzag behaviour was noticed after 45 days and 90 days of seawater ageing for BF/Elium composites, while it was a significant change for BF/Epoxy composites. The (load and COD) values of BF/Elium composites were both increased by (4.8% and 9.1%) after 45 days and decreased by (16.5% and −9.2%) after 90 days, respectively, compared to the unaged composite.

During delamination growth, the bridging fibres were fractured or peeled away from the Elium and epoxy matrixes, leading to an increased separation between the two sides of the specimens. This behaviour accounts for the higher propagation mode I-IFT (G_IC) compared to the initiation mode I-IFT (G_ICo). The G_Ico values followed the same trend as the maximum loading, increasing by 5.6% and then decreasing by 25.1% after 45 and 90 days, respectively. On the other hand, the (load and COD) values of BF/Epoxy composites both increased by 43.8% and 75.2% after 45 days and increased by 65.5% and 114.5% after 90 days, respectively, compared to the unaged BF/Epoxy composite. These values led to the G_Ico values following the same trend, as they increased by 32.7% and 49.3% after 45 days and 90 days, respectively. Several factors contribute to the improvement of interlaminar fracture toughness. One is the matrix plasticisation effect, as it lowers the T_g and makes the matrix more ductile [24]. Another reason is that the fibre pull-out and bridging during delamination increased with hydrothermal ageing [22,23,37]. Moreover, the interleaf (or interlayer) exhibits greater toughness than the resin-rich layer of the base laminates [49].

As seen in Figure 6a, the fracture resistance curves (R-curves) exhibit the variation in delamination resistance values (zigzag shape) versus delamination length values. The average G_IC of the BF/Elium composite was 1242 J/m², which is 15% greater than that of the BF/Epoxy composite. In marine and tidal blade industries, component density plays a key role in material selection. After normalising the IFT values by dividing each G_IC value by its composite density, the increase in the G_IC for the BF/Elium composite was only 7% (Figure 6b). The hydrothermal ageing weakens the matrix interface through matrix swelling and debonding of BF/Elium after 90 days of seawater saturation, resulting in a reduction in G_Ic of approximately 33.1%. Interestingly, the BF/Epoxy composite demonstrated G_Ic values of 1182 J/m² and 1471 J/m², which are 9.5% and 36.3% higher than those of the unaged BF/Epoxy composite. This behaviour was due to more pronounced fibre bridges, resulting in rising R-curves (Figure 6a) and a noticeable increase in the mode I critical energy release rate. This was also attributed to post-curing in thermosets and crystallisation effects in thermoplastics. [24]. As presented in Table 3 (Section 2.1), the bulk epoxy matrix shows tensile and flexural properties higher than those of the bulk Elium matrix. As shown in Figure 7, the increase in the G_IC of the aged BF/Epoxy samples resulted in compression failure in the DCB test sample due to polymer matrix decomposition and fibre surface degradation under the influence of seawater hydrothermal ageing. Figure 6b displays the normalised G_Ic values (G_Icn) of both composites compared to the G_Ic values. The BF/Elium composite demonstrated higher normalised delamination resistance energy by 8%, compared to the BF/Epoxy composite. These normalised G_Ic values followed the same trend as the G_Ic values, even after dividing them by the varied densities of each composite due to hydrothermal ageing.

3.3. Mechanisms of Mode I-IFT

The fracture mechanism of the mode I-IFT is illustrated in the images of Figure 8 and Figure 9. The OM images of the failed DCB specimens were taken at the crack tip and the crack propagation for the 0-day, 45-day, and 90-day seawater ageing (Figure 8).

As seen in these images, the interlayer crack was straight at the crack tip for the BF/Elium, indicating that crack initiation occurred earlier. Furthermore, as presented in Figure 8a–c, some fibre bundles were pulled and broken during the propagation of the crack. Thus, the interlayer crack was not straight (i.e., it exhibited a kinked crack path) and had a rougher surface, resulting in an increased fracture area. Accordingly, crack initiation in the BF/Epoxy sample began later than in the BF/Elium. The delamination resistance was improved after 45-day hydrothermal ageing, as shown in the images of Figure 8b, where no degradation was seen in the interlayer surface of both thermoset and thermoplastic composites. On the one hand, deterioration signs were seen in the interlayer surface of the BF/Elium sample after 90-day ageing, while there was a dense pull-out of fibres bridging the crack in the interlayer surface of the BF/Epoxy sample (Figure 8c). Additionally, compression failure in the BF/Epoxy sample occurred due to polymer matrix decomposition and fibre surface degradation caused by seawater hydrothermal ageing.

As shown in Figure 9, the SEM image of the fracture surfaces in the crack propagation locations in both composites revealed pull-out fibres from the Elium and the epoxy matrix, providing bridging reinforcement during the delamination process of the DCB test. Consequently, the mode I-IFT energy was dissipated in the fibre-bridged zone during crack initiation and propagation. As seen from these images, the pull-out fibre density of the BF/Elium-45D, BF/Epoxy-45D, and BF/Epoxy-90D increased after hydrothermal seawater ageing, which reflects the increasing delamination energy of these composites.

3.4. Mode II-IFT Performance

The ENF sample was subjected to mode II shear loading to initiate and propagate a crack from the crack tip at the inserted film toward the specimen’s midspan. The load–displacement curves of the ENF tests (Figure 10a) were used to calculate the mode II-IFT values (G_IIc) and the normalised mode II-IFT values (G_IIcn). It was observed that the response curves of both composites consist of two distinct phases. The first phase (I) exhibited a peak load value, referred to as crack initiation at the crack tip of the film insert, followed by a rapid load reduction; however, the sample remained intact at this point. At this stage, both composites displayed linear load–displacement behaviour up to the maximum load point, which corresponds to the initiation of the starting crack. Then, the load suddenly decreased, resulting in a sharp reduction with constant displacement for BF/Elium, whereas there was a slight reduction in load with small displacement for the BF/Epoxy due to unstable crack propagation. The second phase (II) demonstrated a non-linear increase in the load due to unstable crack propagation up to the midspan of the specimen. Liu et al. [50] presented similar behaviour, indicating two stages of failure during the ENF test. The maximum load point in the first region was utilised to determine the mode II-IFT using Equation (4). It is noted that the maximum load of BF/Elium-0D in the first stage was 493 N higher than that of BF/Epoxy, representing a 66% increase due to its superior shear loading resistance.

The same two-phase behaviour was observed after 45 days and 90 days of seawater ageing for both thermoset and thermoplastic composites, whereas a significant reduction in the maximum load values was noticed for both composites after 90 days of ageing. Figure 10b displays the normalised G_IIc values (G_IIcn) of both composites compared to the G_IIc values. The mode II delamination resistance shear energy (G_IIc) of the BF/Elium-0D composite was 2110 J/m², which is 56% higher than that of the BF/Epoxy-0D composite (Figure 10b). After normalising, this increment ratio becomes 45%. The G_IIc values of BF/Elium and BF/Epoxy composites were reduced by 9.6% and 21.1% after 45 days and by 44.3% and 32.4% after 90 days of seawater ageing, compared to their unaged composites. These G_IIcn values followed the same trend as the G_IIc values, even after dividing them by the varied densities of each composite due to hydrothermal ageing.

3.5. Mechanisms of Mode II-IFT

The OM and SEM images (Figure 11 and Figure 12) were taken to clarify the failure mechanisms of the ENF samples and to aid in the interpretation of the mode II-IFT results. The side view of the specimens was inspected using OM for both composites (Figure 11a) near the crack tip. This figure shows that the BFs are exposed from the matrix due to crack initiation and during crack propagation. Hence, the plastically deformed zones generated a high shear load during crack propagation due to good adhesion at the fibre/matrix interface [30,40]. As seen in Figure 11b,c, the fibre being out and broken, with delamination near the film inserted into a crack, reflects the effect of seawater ageing on the degradation of the matrix and basalt fibres.

The SEM images (Figure 12) demonstrated the fracture surface of the ENF specimens near the crack tip. As shown in this figure, the BFs were pulled out from the matrix and broken. On the other hand, these pulled-out fibres can bridge the crack in the matrix–fibre interface, delaying crack initiation and propagation along the crack growth path to provide more interlaminar toughness and, thus, higher mode II shear delamination energy [51]. It is clear that dense fibre bridging covered the matrix, and matrix hackles occurred for the BF/Elium thermoplastic composite, while thin fibre bridging and matrix hackles occurred for the thermoset BF/Epoxy composites, resulting in the high shear loading resistance of the BF/Elium composites. The density of the fibre pull-out was increased, and matrix hackles were reduced due to the degradation of the matrix and basalt fibres under the effect of seawater hydrothermal ageing.

4. Conclusions

The BF/Elium and BF/Epoxy composites were manufactured using the VARTI process and cured at RT. The physical, DCB, and ENF behaviours were investigated under hydrothermal seawater ageing at 45 °C for 45 and 90 days. The main conclusions from this study are summarised as follows:

• The ρ_c, V_v, and V_f values for the BF/Elium composite were 1.96 g/cm³, 1.3%, and 53.5%, while those of the BF/Epoxy composite were 1.84 g/cm³, 3.5%, and 47.8%, respectively. Hydrothermal ageing resulted in a slight increase in the thickness and density of both composites due to moisture absorption and matrix swelling.
• All the samples reached saturation after approximately 84 days (2016 h) of immersion in the seawater at 45 °C.
• The BF/Epoxy composite displayed a comparatively higher weight gain of 19% and 31% than the BF/Elium composites after reaching seawater saturation at RT and 45 °C, respectively.
• The diffusion coefficients of BF/Elium exhibited an increase of 9.1% at 45 °C, while that of the BF/Epoxy exhibited an increment of 6.3%, compared to its values at RT.
• The load–COD curves and the R-curves behaviour of DCB test specimens were different between the two composites. Hence, the crack propagation for BF/Elium was more unstable (zigzag) and greater than for BF/Epoxy. As a result, the average GIC value of the BF/Elium composite was 1242 J/m² higher than that of the BF/Epoxy composite, representing a 15% increase.
• The hydrothermal ageing weakens the matrix interface through matrix swelling and debonding of BF/Elium after 90 days of seawater saturation, resulting in a reduction of G_Ic of approximately 33.1%. Interestingly, the BF/Epoxy composite demonstrated increments in the G_Ic values by 9.5% and 36.3% higher than those of the unaged BF/Epoxy composite.
• The OM and SEM images showed that deterioration signs were seen in the interlayer surface of the BF/Elium sample after 90-day ageing, while there was a dense pull-out of fibres bridging the crack in the interlayer surface of the BF/Epoxy sample.
• The load–displacement curves of the ENF test showed a two-stage behaviour. The G_IIC value of the BF/Elium composite was 2110 J/m², higher than that of the BF/Epoxy composite, by 56%, due to the high shear loading resistance of the BF/Elium composite.
• The G_IIc values of BF/Elium and BF/Epoxy composites were reduced by 44.3% and 32.4% after 90 days of seawater ageing, compared to their unaged composites.
• The OM and SEM images of both composites show fibre pull-out, fibre breakage, and matrix hackles formed during the ENF test, generating resistance to shear loading.
• The G_IC and G_IIC values for both composites were normalised with respect to the density of each aged and unaged composite. Even after normalising, the strength, stiffness, mode I-IFT, and mode II-IFT values of the BF/Elium composite were higher than those of the BF/Epoxy composite.
• Despite the reduction in the mode I-IFT and mode II-IFT values of the BF/Elium composite, which were higher than those of the BF/Epoxy composite after both composites reached seawater saturation, these values of the BF/Elium composite were still higher than those of the BF/Epoxy composite.

This experimental study analysed new sustainable alternatives to mitigate the environmental impact of thermosets and synthetic fibre composites. The presented results of this work underscore the importance of utilising sustainable materials in various advanced applications, including marine structures and tidal energy. Finally, the recyclability of the novel multi-functional BF/Elium composite in this work is a key aspect that enables a significant reduction in environmental impacts, such as CO₂ emissions and landfill waste, compared to the end-of-life scenarios of thermoset-based composites, thereby achieving the goal of net zero.

Future research can include investigating the long-term durability of BF/Elium composites under cyclic loading and varying environmental conditions, optimising various parameters such as resin formulation, curing temperature, and vacuum pressure for improved seawater resistance, and exploring hybrid and nanoparticle reinforcement strategies to enhance fracture toughness. The revised conclusions now reflect these future research avenues.