Influence of Thermal Gradients and Arctic Temperatures on the Mechanical Properties and Fracture Behavior of Woven Carbon and Woven Kevlar^® Composites

Composite materials are extensively used across various industries, including railway, automotive, aerospace, space exploration, wind energy, medical, and military applications [1–3]. They have superior strength-to-weight ratio and stiffness-to-weight ratio, along with minimal thermal expansion [4–6]. By carefully choosing the right constituents (fiber and matrix) and fiber orientations, these materials can be engineered to exhibit exceptional thermal and mechanical properties, surpassing conventional structural materials like metals. Nonetheless, ensuring their durability and understanding of their mechanical behavior under harsh environmental conditions remains a challenge that must be addressed.

The increase in oil exploration activities in the Arctic Ocean and various Arctic and polar expeditions have necessitated materials capable of enduring extreme weather conditions, including subzero temperatures, heavy snow and ice, and high winds [7]. Fiber-reinforced polymer (FRP) composite materials have become an attractive solution for these challenging environments, finding applications in ship hulls, undersea vehicles, Arctic drilling, high-altitude aircraft or unmanned aerial vehicles, drones, and scientific instruments [8]. Several researchers have reported that exposing FRP composites to low temperatures can degrade their mechanical properties [9]. As the temperature decreases, the matrix undergoes thermal contraction, resulting in its shrinkage. This creates interlaminar shear stresses due to the difference in coefficient of thermal expansion (CTE) between the fiber and matrix [10]. If these stresses exceed the matrix's shear strength, microcracks can occur. Microcracks have the potential to escalate into fiber/matrix debonding and delamination [11,12]. These failure mechanisms can decrease the overall stiffness and strength of the FRP composites, which could lead to catastrophic failures of the structures.

Woven textile FRP composites are preferred over unidirectional FRP composites because of their superior structural and thermal properties [13–16]. In addition, the interlocking nature of the fibers in these composites contributes to their increased resistance to crack propagation [17]. Despite these benefits, the behavior of FRP composites at low temperatures has primarily concentrated on unidirectional composites and their performance under cryogenic conditions. Kim et al. [18] evaluated the tensile properties of T700/epoxy unidirectional laminates under thermo-mechanical cyclic loading between room temperature and various low temperatures (−50 °C, −100 °C, and −150 °C). They found out that the tensile stiffness increased significantly with decreasing temperature, but the thermo-mechanical cycling had minimal impact on stiffness. In contrast, tensile strength decreased as the temperature decreased. Kim et al. [19] studied the tensile properties of carbon FRP composites with different resin compositions subjected to thermo-mechanical cyclic loading from room temperature to −150 °C. They found that resin compositions containing bisphenol-A epoxy and CTBN-modified rubber exhibited strong mechanical performance at −150 °C. Avila et al. [20] studied the short-beam strength of hybrid textile FRP composites made of carbon and Kevlar^®, subjected to temperatures of 77 K. Their findings indicated that incorporating Kevlar^® layers reduced the short-beam strength of the composite. FRP composites made of only carbon layers, without adding Kevlar^® layers, demonstrated superior short-beam strength properties. Dutta and Hui [21] evaluated the flexural properties of unidirectional commercial E-glass/polyester resin and S2-glass/polyester resin FRP composites subjected to thermal cycling between 50 °C and −60 °C. They determined that at low temperatures, FRP composites become stiffer; therefore, there will be an increase in their stiffness. There has been limited research reported on the behavior of woven FRP composites when subjected to low temperatures. Garcia et al. [22] studied the flexural properties of woven carbon vinyl ester composites when exposed to seawater and Arctic temperatures (−60 °C). They concluded that the flexural strength of the composites increased when exposed to seawater and arctic temperatures. Moreover, they noticed that the composite materials failed in a brittle manner. LeBlanc et al. [23] investigated the behavior of plain weave carbon/epoxy and E-glass/epoxy laminates under quasi-static and dynamic conditions, spanning temperatures from 20 °C to −2 °C. They found that the laminates' tensile properties and short-beam strength increased at lower temperatures. On the other hand, the impact response of the laminates did not exhibit temperature influence. Islam et al. [24] examined the tensile, flexural, and interlaminar shear strength (ILSS) behavior of woven carbon and woven Kevlar^® FRP composites at cryogenic temperatures. They determined that cryogenic exposure did not have significant influence in the tensile, flexural, and ILSS properties. They highlighted that a crucial consideration in designing woven FRP composites for low-temperature environments is understanding the impact of thermal gradients across the materials. In all prior research involving both unidirectional and woven composites tested at Arctic and cryogenic temperatures, the samples underwent uniform exposure to the temperature conditions from all sides. However, this approach does not accurately reflect real-world conditions. Typically, one side of an FRP composite will be exposed to a specific temperature while the opposite side experiences a different one, creating a thermal gradient across the sample. For example, drones used for Arctic exploration are designed to maintain an internal temperature of 15 °C, essential for optimal battery performance, while facing external temperatures down to −46 °C [8]. According to the authors' knowledge, the impact of such thermal gradients at Arctic (−60 °C) temperatures on the tensile and flexural properties of woven and Kevlar^® FRP composites has yet to be explored.

This study experimentally investigates the effect of a thermal gradient (−60/23 °C) on the tensile and flexural properties of woven carbon and Kevlar^® FRP composite materials. This study utilizes a custom-built environmental chamber capable of exposing one side of the samples to a specific temperature, while the opposite side is subjected to a different temperature. Furthermore, to enhance the understanding of how thermal gradients affect the performance of woven carbon and Kevlar^® FRP composites, the tensile and flexural properties of these materials under Arctic (−60 °C) and room (23 °C) temperatures will be evaluated.

The structure of the article is outlined as follows: Sec. 2 provides a comprehensive overview of the manufacturing and testing procedures for the woven carbon and Kevlar^® FRP composite materials. Additionally, it includes a detailed description of the construction of a custom-built environmental chamber designed to expose a sample to two distinct temperatures simultaneously. This is followed by Sec. 3, which presents the results of the mechanical tests. Next, Sec. 3.4 thoroughly examines the failure mechanisms of the composite materials exposed to different temperatures: −60 °C, 23 °C, and −60/23 °C. This article concludes by summarizing the key findings and suggesting potential avenues for future research.

The material system is presented first, followed by sample manufacturing, and finally experimental setup.

The selected reinforcement materials consisted of 3 K woven plain weave carbon fabric and Kevlar^® plain weave fabric, obtained from FibreGlast² [25]. The two part thermosetting matrix selected was EPON™ resin 828 and EPIKURE™ curing agent 3015, which were purchased from Hexion.³ The mechanical properties of the constituent materials are given in Table 1. The hardener was mixed with the resin at a ratio (weight) of 100:50 as recommended by the manufacturer.

Woven fabric was selected over unidirectional fabric due to its lower permeability and resistance to crack propagation at low temperatures [26]. EPON™ 828 resin was selected due to its balance of stiffness, strength, and ductility and use at low temperatures [27,28]. Epikure™ 3015 was selected due to its good adhesion, chemical resistance, and low viscosity, which makes it ideal for out-of-autoclave manufacturing processes [29]. Kevlar^® fiber was selected due to its strength, modulus, toughness and exceptional elasticity [2,30]. Carbon fibers were selected due to their high strength and high modulus [31].

The carbon and Kevlar^® laminates were manufactured using the vacuum-assisted resin transfer molding (VARTM) process. Each laminate was manufactured with 12 layers that were cut to dimensions of 305 mm × 305 mm. These layers were then sandwiched between two layers of flow media, two layers of breather and four layers of nylon peel plies. This complete arrangement of fabrics was placed on top of an aluminum mold, which was then wrapped with Stretchlon 800 bagging film and sealed with vacuum-sealant tape, ensuring spaces for both inlet and outlet connectors, as seen in Fig. 1. The first vacuum bag had an inlet for the resin to be infused through and an outlet connector to collect the excess resin. The second vacuum bag had a connector that applied continuous pressure during and after the resin infiltration process at a constant pressure of 80 kPa. EPON™ 828 and Epikure™ 3015 were mixed with a weight ratio of 100:50 as recommended by the manufacturer. The mixture was first placed in a desiccator to eliminate any air bubbles. After that, the outlet was connected to a vacuum pump, maintaining the vacuum bag at a pressure of approximately 80 kPa. The vacuum bag's inlet was then immersed in the resin/hardener mixture, initiating the resin infusion through the laminate. After the resin transfer process was completed, the laminate was left to cure at room temperature for 16 h under a constant pressure of 80 kPa, maintained by the second vacuum bag. A total of four laminates were manufactured (two for carbon fabric and two for Kevlar^® fabric). The approximate fiber weight fraction for carbon and Kevlar^® fiber laminates was 41% and 44%, respectively. Once the laminates were cured, they were water jet cut.

Resin samples were manufactured to evaluate the effect of temperature on the matrix. The specimens were manufactured using 3D-printed molds (Fig. 2) following the dimensions of ASTM D638-22 [32] and ASTM D790-17 [33] for tension and flexural tests of plastics, respectively. EPON™ 828 and Epikure™ 3015 were mixed with a weight ratio of 100:50, as recommended by the manufacturer. The mixture was initially placed in a desiccator to eliminate any air bubbles. Following this, it was carefully poured into 3D-printed molds. These were then left to cure at room temperature for a duration of 16 h. A total of 10 samples were manufactured for tensile and 10 for flexural testing.

An environmental chamber was designed to expose the FRP composite specimens' front and back faces to two different temperatures, 23 °C and −60 °C, and produce a thermal gradient of −60/23 °C across the sample. This design effectively induced a temperature gradient across the specimens. To facilitate this temperature variation, an insulated shipping container with a removable lid was used. The container had dimensions of 330 mm in width, 330 mm in length, 317.5 mm in height, and a wall thickness of 38.1 mm. To maintain the inside of the shipping insulated container to 23 °C, polyurethane foam insulation sheets and a BN-Link digital thermostat controller heat Mat (temperature accuracy of 23 ± 2 °C) were added inside, as seen in Fig. 3(a). To establish an external temperature of −60 °C, the container was placed inside a Thermo Fisher Scientific TSU refrigerator, precisely calibrated to this temperature. Cutouts slightly larger than the samples and evenly spaced were made in the lid, ensuring that one side of the samples was exposed to 23 °C and the other to −60 °C, as seen in Fig. 3(b). The samples, wrapped with double-sided tape around their edges, were snugly fitted into these lid perforations. The lid was sealed using double-sided and thermal tape. Throughout the test, thermocouples were placed inside the chamber, attached to the front and back surfaces of all the samples, on top of the lid and the refrigerator, facilitating comprehensive temperature monitoring through the conditioning of the samples. Figure 3(c) displays the final configuration of the insulated container, heating pad, and samples. Drones used for Arctic exploration usually operate between 13 and 32 min [8]. They are designed to maintain an internal temperature of 15 °C, essential for optimal battery performance, while enduring external temperatures as low as −46 °C. The heating pad was limited to a maximum operating time of 5 h to guarantee precise readings and to avoid potential malfunctions. A leak test spanning 4 h and 40 min was used to validate the formation of a thermal gradient within the FRP composite specimens. Thermocouples recorded the sample surface (top and bottom) temperatures and the interior and exterior temperatures of the environmental chamber, which were held at 23 °C and −60 °C, respectively. Figure 4 illustrates the temperature graphs recorded by the thermocouples versus the test duration. After 40 min, the sample's top surface (exposed to −60 °C) attained a temperature of −23.7 °C, while the bottom surface (exposed to 23 °C) reached −15.6 °C. Following this period, the temperatures across the top and bottom surfaces of the sample remained relatively stable, resulting in a consistent temperature gradient of approximately 8 °C across the sample.

A total of 30 samples were water jet cut for tensile tests (15 for carbon and 15 for Kevlar^®) of FRP composite materials. Five samples were selected for each environmental condition for each test: −60 °C, 23 °C, and −60/23 °C. The tensile specimens had dimensions of 250 mm in length and 25 mm in width. The average measured thickness was 2.1 mm for carbon and 2.5 mm for Kevlar^® laminates. The dimensions were in accordance with ASTM D3039/D3039-17 [34] standards for tensile properties of FRP composites. Tension tests were performed in an Instron 8801 servo hydraulic fatigue testing system (100-kN load cell) with a 2 mm/min loading rate until failure. The tensile stress–strain responses for each sample were calculated using the cross-sectional area, and the strain was recorded by a 2620-601 Instron extensometer with a gauge length of 25.4 mm. The tensile modulus, strength, and failure strain were calculated according to standard D3039.

A total of 10 samples were manufactured for resin tensile tests, 5 samples were tested at −60 °C and 5 samples at 23 °C. It is important to note that temperature gradient conditioning was not applied to these samples, considering the relatively small section of resin within the FRP composites, typically around 0.2 ± 0.01 mm for woven composites. Testing the resin in its pristine state provided significant insights into how temperature variations impact the overall mechanical behavior of FRP composites. Figure 5 shows the dimensions of the tensile resin specimens. The samples had an average thickness of 3.2 mm. The dimensions were in accordance with ASTM D638-22 [32] standard for tensile properties of plastics. Tension tests were performed in an ADMET eXpert 1000 machine (2.2-kN load cell) with a 5 mm/min loading rate until failure. The tensile stress–strain responses for each sample were calculated using the cross-sectional area, and the strain was recorded by an Epsilon axial extensometer with a gauge length of 25.4 mm. The tensile modulus, strength, and failure strain were calculated according to standard D638.

A total of 10 samples were manufactured for resin flexural tests, 5 samples were tested at −60 °C and 5 samples at 23 °C. The flexural specimens maintain a span-to-thickness ratio of 16:1. Figure 6 shows the dimensions of the flexural resin specimens. The samples had an average thickness of 3.3 mm. The dimensions were in accordance with ASTM D790-17 [33] standard for flexural properties of plastics. Flexural tests were performed in an ADMET eXpert 1000 machine (2.2-kN load cell) with a 1 mm/min loading rate until failure. The flexural stress and strain were calculated using Eqs. (1) and (2), respectively. Flexural modulus, strength, and failure strain were calculated according to standard D790.

The samples were immediately subjected to tensile and flexural testing after conditioning. To preserve their conditioned state, we carefully extracted them from the environmental chamber or fridge using tweezers, ensuring that direct hand contact, which could alter their conditioning, was avoided. Three different environmental temperatures were used for carbon and Kevlar^® FRP composites: 23 °C, −60 °C, and temperature gradient of −60 °C/23 °C. To ensure uniformity with the operating limits of the designed environmental chamber, which had a maximum runtime of 4 h and 40 min, samples designated for full exposure to −60 °C were placed in a Thermo Fisher Scientific TSU refrigerator set at −60 °C for the same duration. The samples exposed to the temperature gradient of −60 °C/23 °C were conditioned for 4 h and 40 min in the designed environmental chamber discussed in Sec. 2.3.1. Previous authors have reported that the FRP composites can reach uniform temperature after 20 min [34–37]. Two different environmental temperatures were used for the resin samples: −60 °C and 23 °C. The samples were conditioned at −60 °C and exposed to this temperature for 4 h and 40 min.

Tension and flexural tests were performed on both carbon and Kevlar^® FRP composites, as well as on pristine resin, to understand the effect of Arctic temperature (−60 °C) and thermal gradient (−60/23 °C) on FRP composites. The findings from these experimental investigations are explored and discussed in detail in this section. A detailed explanation of how temperature affects the composites and resin samples is provided in Sec. 3.3.

Table 2 presents the average tensile stiffness, tensile strength, and failure strain for five samples at each test temperature for carbon and Kevlar^® FRP composites and pristine resin. The carbon FRP composites exhibited greater tensile modulus and strength compared to the Kevlar^® FRP composites. However, the failure strains between the carbon and Kevlar^® FRP composites were very similar.

Figure 7 shows representative flexural stress–strain curves of woven carbon FRP composites subjected to three different temperatures: −60 °C, 23 °C, and −60/23 °C. The samples exposed to −60 °C demonstrated a 6% increase in tensile stiffness and a 10% rise in tensile strength compared to those tested at 23 °C. In addition, these samples showed an 8% reduction in strain at failure as compared to the samples tested at 23 °C. The samples subjected to −60/23 °C exhibited a 3.5% increase in flexural stiffness, a 7.93% increase in tensile strength, and a 4.4% reduction in ductility, compared to those tested at 23 °C, as seen in Table 2. The minimal influence of low temperatures on the tensile mechanical properties is attributed to their fiber-dominated nature. While the resin does become more rigid and brittle at lower temperatures, this change has a negligible impact on tensile behavior, again owing to the dominant role of the fibers in tensile properties. The samples exposed to the thermal gradient demonstrated tensile properties intermediate to those tested at −60 °C and room temperature (23 °C). This is because, during exposure to thermal gradients, the samples achieved an equilibrium temperature of −23.7 °C on the top surface (exposed to −60 °C) and −15.6 °C on the bottom surface (exposed to 23 °C). Consequently, these samples were stiffer than those kept at room temperature but not as stiff as those tested at −60 °C. Figure 8 shows representative tensile stress–strain curves of woven Kevlar^® FRP composites subjected to three different temperatures: −60 °C, 23 °C, and −60/23 °C. The samples exposed to −60 °C demonstrated a 15.1% increase in flexural stiffness and an 8.13% rise in flexural strength compared to those tested at 23 °C. In addition, these samples showed a 16% reduction in strain at failure compared to the samples tested at 23 °C. The samples exposed to −60/23 °C exhibited a 15% increase in tensile stiffness, 9.45% increase in tensile strength, and a 6.6% reduction in ductility, compared to those tested at 23 °C, as seen in Table 2. As in the case of carbon fiber samples, the samples exposed to the thermal gradient demonstrated tensile properties intermediate to those tested at −60 °C and 23 °C. Across the range of temperatures tested (−60 °C, 23 °C, and −60/23 °C), carbon FRP composites showed a significant improvement over Kevlar^® FRP composites, with an average increase of 60% in tensile modulus and 68% in tensile strength. Nonetheless, they experienced an average 10% decrease in strain at failure across all temperatures when compared to the Kevlar^® FRP composites.

Figure 9 shows representative tensile stress–strain curves of resin samples subjected to −60 °C and 23 °C. The samples exposed to −60 °C demonstrated a 34.1% increase in tensile stiffness and a 13% rise in tensile strength compared to those tested at 23 °C. In addition, these samples showed a −44% reduction in strain at failure compared to the samples tested at 23 °C. The matrix becomes more rigid yet brittle at lower temperatures, increasing the stiffness and strength but reducing ductility.

Table 3 presents the average flexural stiffness, flexural strength, and failure strain for five samples at each test temperature for carbon and Kevlar^® FRP composites and pristine resin. Resin samples were only exposed to −60 °C and 23 °C.

Figure 10 shows representative flexural stress–strain curves of woven carbon FRP composites subjected to three different temperatures: −60 °C, 23 °C, and −60/23 °C. The samples exposed to −60 °C demonstrated a 9% increase in flexural stiffness and a 10.6% rise in flexural strength, compared to those tested at 23 °C. In addition, these samples showed a 10.8% reduction in strain at failure as compared to the samples tested at 23 °C. The samples subjected to −60/23 °C exhibited a 4.7% increase in flexural stiffness and a 10.1% increase in flexural strength, along with a 10.35% reduction in ductility, compared to those tested at 23 °C, as seen in Table 3. Figure 11 shows representative flexural stress–strain curves of woven Kevlar^® FRP composites subjected to three different temperatures: −60 °C, 23 °C, and −60/23 °C. The samples exposed to −60 °C demonstrated a 33.52% increase in flexural stiffness and a 26.6% rise in flexural strength, compared to those tested at 23 °C. In addition, these samples showed a 7.2% reduction in strain at failure compared to the samples tested at 23 °C. The samples exposed to −60/23 °C exhibited a 29.9% increase in flexural stiffness, 23.5% increase in flexural strength, and a 4.15% reduction in ductility, compared to those tested at 23 °C, as seen in Table 3. Flexural mechanical properties are predominantly influenced by the matrix. The increase in rigidity and decrease in ductility of the matrix at low temperatures has a more significant effect on the flexural properties compared to the tensile properties of composite materials. As in the case of tensile samples, the samples exposed to the thermal gradient demonstrated flexural properties intermediate to those tested at −60 °C and 23 °C. Across the range of temperatures tested (−60 °C, 23 °C, and −60/23 °C), carbon FRP composites showed a significant improvement over Kevlar^® FRP composites, with an average increase of 67% in flexural modulus and 51% in flexural strength. Nonetheless, they experienced an average 74% decrease in strain at failure across all temperatures when compared to the Kevlar^® FRP composites.

Figure 12 shows representative flexural stress–strain curves of resin samples subjected to −60 °C and 23 °C. The samples exposed to −60 °C demonstrated a 32.8% increase in flexural stiffness and a 40% rise in flexural strength compared to those tested at 23 °C. In addition, these samples showed a −60.4% reduction in strain at failure compared to the samples tested at 23 °C. The matrix becomes more rigid yet brittle at lower temperatures, increasing the stiffness and strength but reducing ductility.

As temperature decreases from 23 °C to −60 °C, the difference in the CTE between the fiber and matrix leads to the generation of thermal stresses [10]. This results in compressive forces at the fiber–matrix interface, increasing tensile and flexural stiffness and strength in CFRP and Kevlar^® FRP composites. The observed reduction in strain at failure can be attributed to decreased mobility of the matrix's polymer chains at lower temperatures, leading to increased rigidity and brittleness [36]. This observation is reflected in the tensile and flexural properties of the resin, as detailed in Secs. 3.1 and 3.2. The tensile mechanical properties are fiber-dominated, while the flexural mechanical properties are matrix-dominated. For this reason, low temperatures have a more significant effect on flexural than tensile mechanical properties.

Low temperatures have been shown to enhance composites' tensile and compressive properties, specifically by increasing stiffness and strength while decreasing the strain at failure [9,21,37]. This improvement is due to the reduced mobility of the resin's polymer chains at lower temperatures, which results in increased rigidity and brittleness [38]. Consequently, the Poisson's ratio is expected to decrease as the resin becomes stiffer [39]. Woven Kevlar^® fibers exhibit lower interlaminar shear strength (ILSS) than carbon fibers with the same resin at −100 °C, attributed to the poor interfacial properties between Kevlar^® fibers and resins [40]. The CTE is also temperature-dependent; it decreases for resins, glass fiber and Kevlar^® fibers with decreasing temperature [41], while it increases for carbon fibers [42]. Additionally, the mechanical fatigue strength of composites generally improves at lower temperatures compared to room temperature [43–46].

Following tensile and flexural testing, an optical microscope was used to examine and identify the failure mechanisms in the FRP composite materials.

The tensile mechanical properties are fiber-dominated, as discussed in Sec. 3.3. The woven carbon FRP composite samples, subjected to tensile testing, exhibited fiber fracture across all tested temperatures, as illustrated in Fig. 13. Figure 14 shows the front and side views, respectively, of the woven Kevlar^® FRP composites. The front view exhibited some fiber breakage and matrix damage, while the side view showed more fiber fracture and delamination across all tested temperatures. Unlike the woven carbon FRP composites, these samples did not undergo complete ruptures. This can be attributed to the high toughness inherent of the Kevlar^® fibers [39,40].

Flexural mechanical properties are matrix-dominated, as discussed in Sec. 3.3. In Fig. 15, a square enclosed the specific section from which the optical microscope images were taken for all samples. Figure 16 shows the representative woven carbon FRP composite samples after three-point bending testing across all tested temperatures. All the samples tested at 23 °C exhibited fiber fracture as their main failure mechanism, as seen in Fig. 16. The samples tested at −60 °C and −60/23 °C experienced fiber fracture, delamination, and debonding. This behavior is attributed to the difference in the CTE between the fiber and matrix. As a result, increased debonding and delamination were observed throughout the thickness of the samples when exposed to low temperatures. Figure 17 shows the representative woven Kevlar^® FRP composite samples after three-point bending testing across all tested temperatures. The failure modes observed for the samples tested at −60 °C and at −60/23 °C were similar, characterized by fiber fracture at the top, under the loading point. In the case of the samples tested at 23 °C, their failure mechanism was the formation of a shear band under the load point, which other researchers have reported [24]. The samples exposed to thermal gradients displayed brittle behavior, similar to those tested at −60 °C, because, despite achieving a thermal gradient of 8 °C across them, both sides of the samples registered temperatures below 0 °C. Ultimately, the top surface, exposed to −60 °C, reached an equilibrium temperature of −23.7 °C, and the bottom surface, exposed to 23 °C, settled at −15.6 °C, resulting in increased brittleness of the samples.

This study conducted a thorough analysis of the effects that thermal gradients (−60/23 °C), Arctic temperature (−60 °C), and room temperature (23 °C) have on the tensile and flexural properties of woven carbon and Kevlar^® FRP composites. To gain deeper insights into how temperature influences these FRP composites, tensile and flexural tests were carried out on pure resin samples. Based on the experiments, the following results are obtained:

• The carbon and Kevlar^® FRP composites experienced an increase of their tensile and flexural properties (stiffness and strength) when exposed to −60 °C and −60/23 °C and a reduction in the strain at failure, compared to the samples tested at 23 °C.

• The carbon FRP composites exhibited higher tensile and flexural properties (stiffness and strength) across all temperatures and reduced strain at failure compared to the Kevlar^® FRP composites.

• The pure resin samples exposed to −60 °C experienced higher tensile and flexural properties (stiffness and strength) and reduced strain at failure compared to those tested at 23 °C. The increase in rigidity and decrease in ductility at low temperatures have a greater impact on the flexural properties of FRP composites compared to the tensile properties. This is because flexural properties are primarily determined by the matrix, whereas tensile properties are influenced mainly by the fibers in woven composites.

• The samples subjected to thermal gradients, whether under tensile or flexural loading conditions, exhibited brittle behavior similar to those tested at −60 °C. Their tensile and flexural properties (stiffness, strength, and strain at failure) were intermediate, falling between those observed in samples tested at −60 °C and at 23 °C.

• Fiber fracture was the main failure mechanism for the woven carbon FRP composite samples subjected to tensile testing across all tested temperatures. In contrast, the woven Kevlar^® FRP composites exhibited a combination of fiber breakage, matrix damage, and delamination under the same conditions but without complete ruptures.

• Fiber fracture was the main failure mechanism for the woven carbon FRP composite samples subjected to flexural testing at 23 °C. However, when tested at −60 °C and under the thermal gradient of −60/23 °C, the samples exhibited fiber fracture, delamination, and debonding. These behaviors are attributed to the difference in CTEs between the fiber and matrix, which produced microcracks that could evolve into debonding and delamination once a load is applied. In contrast, the woven Kevlar^® FRP composites subjected to flexural testing at −60 °C and at −60/23 °C experienced fiber fracture at the top, under the loading point. In the case of the samples tested at 23 °C, their failure mechanism was a shear band under the load point.

While this study has offered significant insights into how thermal gradients affect the mechanical properties of woven carbon and Kevlar^® FRP composites, future research could explore these materials' long-term durability and fatigue behavior under cyclic thermal loading or cryogenic conditions.

The authors express their sincere gratitude to The University of Texas at El Paso for their invaluable support via the University Research Initiative (URI) and startup grants, which were instrumental in the research presented in this article.

There are no conflicts of interest. This article does not include research in which human participants were involved. Informed consent is not applicable. This article does not include any research in which animal participants were involved.

The datasets generated and supporting the findings of this article are obtainable from the corresponding author upon reasonable request.