Freeze-thaw durability of a partially bio-based unsaturated polyester composite reinforced with glass fibres

Abstract

The pursuit for more sustainable fibre-polymer composites has motivated the development of renewably sourced bio-based resins. Although these can provide short-term performance similar to that of conventional composites, uncertainties about their long-term durability are still an important limitation. This study investigates the freeze-thaw (FT) durability of a partially bio-based unsaturated polyester resin (UPR) composite, incorporating fumaric acid, 1,3-propanediol, and isosorbide, and reinforced with glass fibres, comparing its performance to that of an oil-based counterpart with identical reinforcement. The composites were preconditioned by immersion in water at 20°C for 30 days, followed by exposure to 300 FT cycles. Each FT cycle consisted of 3 h of dry freezing at −20°C, 8 h of thawing submerged in water at 23°C, and 2.5 h for temperature transitions. Water uptake at 20°C was monitored for up to 180 days. The materials were assessed through mechanical (tensile, compression, in-plane shear, interlaminar shear) and thermomechanical (dynamic mechanical analysis) testing. Most degradation occurred during preconditioning, with property reductions of 20%– 40%, with higher incidence in the bio-based composite. FT ageing led to further but slower degradation in both composites. In-plane shear strength retention of the bio- and oil-based composite was 59% and 93%, respectively, after the preconditioning, which further reduced to 38% and 75% after 300 FT cycles. In comparison, the tensile modulus of both composites remained almost unaffected (∼4-8% reductions). The findings highlight the need for further optimisation of bio-based resin formulations to improve their environmental durability and enable wider structural application.

Keywords

freeze-thaw durability bio-composite bio-based unsaturated polyester resin GFRP water uptake fumaric acid 1,3-propanediol isosorbide vacuum infusion

Introduction

Fibre-reinforced polymer (FRP) composites are increasingly being adopted across various sectors, including aerospace, automotive, and sports industries, due to their significant advantages over conventional materials, such as high specific strength, environmental durability and customisability.^1–4 In civil engineering, FRPs have also gained prominence, being utilised in both new construction projects and in the rehabilitation of ageing infrastructure.⁵

FRP composite materials are primarily composed of synthetic fibres, such as glass, carbon, and aramid, and polymer resins, including epoxy, vinyl ester, and unsaturated polyester (UP), with the resin typically constituting 30%–70% of the total FRP volume.⁶ Among these, glass fibre and unsaturated polyester resin (UPR) are the most widely used combinations for FRP fabrication, particularly when cost-effectiveness is a key consideration.⁷ UPRs are synthesised via a polycondensation reaction involving a mixture of saturated and unsaturated dicarboxylic acids or anhydrides with glycols, producing water as a by-product. The resulting UP polymer is subsequently diluted with a reactive diluent, typically styrene, which facilitates cross-linking during the curing process.⁸

Currently, the chemical compounds used in the synthesis of UPRs and other polymer resins are predominantly derived from petrochemical sources. These oil-based sources are significant contributors to global warming and various other environmental challenges. Additionally, the depletion of fossil fuel reserves and increasing concerns over climate change are driving the FRP industry toward the adoption of more sustainable and environmentally friendly materials.⁸ Of particular concern is the reactive diluent styrene, widely used in UPR formulations, which is classified as a carcinogen and poses health risks to personnel involved in manufacturing processes.⁹ These factors underscore the urgent need for the development and implementation of bio-based and less hazardous alternatives within the composite materials industry.

Environmentally sustainable alternatives for the development of FRP composites primarily include the use of bio-based polymer resins and plant-based fibres. However, plant fibres have demonstrated significant limitations in structural applications, particularly due to their vulnerability to moisture, which adversely affects their mechanical stability and long-term performance.¹⁰ Consequently, the more viable sustainable option at present is the production of FRP composites using bio-based polymer resins in combination with synthetic fibre reinforcements. This approach enables partial substitution of petrochemical components while maintaining the mechanical integrity required for demanding applications.

Fully or partially bio-based UPRs have been developed by several researchers,^11–17 wherein renewable chemical constituents have been incorporated into the resin formulation to substitute conventional petroleum-based components. In parallel, significant research efforts have also focused on identifying renewable and non-renewable alternatives to styrene, the commonly used reactive diluent in UPRs, due to its carcinogenic nature and to environmental concerns.^9,13,18–21 While many bio-based alternatives to styrene have demonstrated potential, they often exhibit drawbacks, such as low glass transition temperatures (T_g) and high viscosity,^18,19 limiting their applicability in structural composites. Conversely, oil-derived acrylates have emerged as more compatible and viable replacements for styrene, offering a balance between performance and processability.^13,20

Although bio-based UPRs have been widely discussed in the literature, their application in actual product development remains limited. A notable exception is presented by Hofmann et al. (2023)¹³ and Akbari et al. (2025),²² who demonstrated the successful synthesis and processing of partially bio-based UPRs. In 13, several resin formulations were developed using bio-derived raw materials, including fumaric acid, furan dicarboxylic acid, dimer fatty acid, isosorbide, and 1,3-propanediol, as part of the UP polymer backbone. Moreover, up to 50% of styrene was substituted with 2-hydroxyethyl methacrylate (HEMA), a less hazardous alternative diluent. This approach yielded UPR formulations that showed compatibility with a variety of reinforcing fibres, such as glass,²³ basalt,²⁴ and carbon,²⁵ highlighting the versatility of bio-based UPRs in composite applications. Furthermore, improved environmental sustainability associated with these bio-based UPR systems was attested by life cycle assessments reported by Shahid et al. (2024; 2025).^17,26

Despite the promising potential of bio-based resins combined with synthetic fibres to enable more environmentally sustainable fibre-reinforced polymer (FRP) composites, concerns regarding their long-term durability persist. This uncertainty presents a critical barrier to the broader adoption of bio-based alternatives in structural applications. FRP materials are typically exposed to a variety of environmental agents during their service life, including moisture (in some cases, immersion), thermal variations, ultraviolet radiation, and synergistic effects among these factors. One particularly demanding condition is freeze-thaw (FT) cycles encountered in cold regions, where repeated thermal cycling in the presence of moisture or water can induce significant damage. Such degradation may occur through both reversible (e.g., plasticisation) and irreversible (e.g., microcracking, hydrolysis) mechanisms.²⁷ Therefore, assessing the durability of bio-based FRP composites under these environmental conditions is essential to determine their suitability for real-world applications.

According to literature reports, the damage caused by isothermal freezing conditions on FRP materials is significantly lower than that induced by FT cycling.^28,29 Similarly, exposure to thermal cycling (TC) alone has been shown to cause minimal environmental degradation, as the internal stresses generated by TC generally remain substantially below the strength thresholds of the fibre-matrix interface and the polymer matrix itself.³⁰ However, when FT cycles occur in the presence of moisture or water, the internal stress accumulation is greatly magnified. For instance, Aniskevich et al. (2012)³¹ reported a ∼13% reduction in the flexural strength of a glass/polyester composite following FT exposure, in contrast to a 6% increase under dry thermal cycling conditions.

Water ingress into FRP composites initiates degradation through mechanisms such as plasticization and hydrolysis of the polymer network.^4,32 In polyester-based FRPs, water molecules react with the ester bonds in the polymer backbone, causing chain scission, leaching of degradation products, and irreversible damage, ultimately reducing mechanical integrity.³³ Moreover, at sub-zero temperatures, absorbed water expands upon freezing, generating internal stresses that may lead to matrix microcracking, which further accelerates moisture ingress.²⁹

Additionally, the mismatch in the coefficient of thermal expansion (CTE) between the fibre reinforcement and polymer matrix introduces localised differential stresses during FT cycling. These stresses promote fibre-matrix debonding and transverse cracking.^34–36 Overall, the combination of moisture exposure and FT cycling results in synergistic damage mechanisms, plasticization, hydrolysis, leaching, microcracking, and interfacial debonding, leading to a cumulative and often irreversible reduction in the performance of FRP composites.

Moisture exposure, combined with the volumetric fluctuations associated with FT cycling, may exert a significant detrimental impact on both the fibre reinforcement and the fibre–matrix interface of composite materials.^4,36 Moreover, synthetic fibre reinforcements, particularly glass fibres, are susceptible to chemical degradation upon contact with moisture that diffuses into the composite structure. When immersed in polar aqueous environments, surface cations on glass fibres interact with hydroxyl ions, initiating the dissolution of these cations into the surrounding medium.³⁷ This interaction facilitates the formation of a corrosion shell composed primarily of silicate compounds, which progressively dissolves, elevating the local alkalinity. The resulting alkaline microenvironment promotes fibre surface degradation through mechanisms such as pitting and leaching, thereby compromising the intrinsic mechanical properties of the fibres.³⁷ Simultaneously, the degradation of both the polymer matrix and the fibre surface undermine the fibre–matrix interfacial integrity. This degradation forms preferential pathways for moisture ingress and ion transport, further accelerating the composite’s deterioration under FT conditions.⁴ In light of these complex and interrelated degradation mechanisms, addressing the FT durability of FRP composites in moisture-laden environments is essential for ensuring their long-term performance and structural reliability.

Table 1 summarises selected FT durability studies on different FRP composites, including two notable examples involving bio-based resin systems: a previous investigation on a bio-UPR-based basalt-FRP composite²⁴ and another on a bio-epoxy-based glass-FRP (GFRP) composite.³⁸ To the best of our knowledge, these represent the only FT durability assessments of FRP composites developed with bio-based resins currently available in the literature. Relative to the study of Shahid et al. (2023),²⁴ it is important to note that the current study concerns composites with a different type of fibre reinforcement (E-glass fibres) and a different formulation of bio-based UPR. Figure 1 presents the retention of mechanical properties under FT cycling, benchmarked against the 50-year conversion factor η_cm (0.6) recommended in CEN/TS 19101:2022³⁹ for FRP composites continuously exposed to moisture (class III environments).

Table 1.

Summary of studies on freeze-thaw cycling for FRP composite materials.

References	Fibre	Resin	Production method	Thickness [mm]	Fibre fraction [%]	Precondition	Thermal profile and FT medium	Full cycle time [h]	No of cycles	Assessed properties
⁴⁰	Glass	UPRⁱ, VE	Pultruded	9.5	NA	NA	−17.8°C to 4.4°C, saline water (2% NaCl)	3	100, 200, 300	σ_f; E_f
⁴¹	Glass, Carbon	Epoxy	Wet layup	11.5	NA	Saline water at RT for 35 days	−17.8°C to 4.4°C, water	48	180, 360	σ_t; E_t; σ_ILS; T_{g (tanδ)}
⁴²	Glass	UPR	Pultruded	4.7, 3.2	0.6 (v_f)	Water at 70°C for 192 days	−17.8°C to 4.4°C, water	3	310, 564	σ_t; E_t
⁴³	Basalt	VE	Vacuum infused	2.2	0.4 (v_f)	NA	−8.5°C to 20°C, saline water	4.8	199	σ_ILS
³⁰	Glass	VE	Vacuum infused	3.5	0.5 (v_f)	Distilled water at RT for 14 days	−17°C to 4.4°C, distilled water	2, 5	250	σ_f
³¹	Glass	UPRⁱ	Pultruded	10.0	0.70 (v_f)	Water at RT for 208 days	−30°C to 20°C, water	24	15, 54, 125	T_{g (E’ decay)}; σ_f; E_f
³⁵	Glass	UPRⁱ	Pultruded	6.4	NA	Water at 25°C for 224 days	−10°C to 20°C, water	2	100, 200, 300	σ_t; E_t; τ_xy; G_xy; T_{g (E’ decay)}; T_{g (tanδ)}
³⁸	Glass	Bio-Epoxy	Wet layup	1.3	NA	NA	−18°C to 4°C, tap water	17	300	σ_t; E_t
⁴⁴	Glass	VE	Pultruded	3.0	0.7 (v_f)	NA	−18°C to 4°C	4	150, 450, 600	σ_t; E_t; σ_f; E_f
⁴⁵	Glass^{u, w}	Epoxy	Vacuum infused	2.0, 5.0	0.7 (v_f)^u, 0.65 (v_f)^w	NA	−20°C to 20°C, tap water	24	30, 50, 80	σ_t
⁴⁶	Glass	Epoxy	Vacuum infused	NA	NA	NA	−20°C to 20°C, saline water	24	45, 90	σ_t; σ_f; σ_c; σ_ILS
²⁴	Basalt	Bio-UPRⁱ, UPRⁱ	Vacuum infusion	4.0	0.65 (m_f)	Water at 20°C for 30 days	−20°C to 23°C, tap water	13.5	100, 200, 300	σ_t; E_t; σ_c; τ_xy; G_xy; σ_ILS; T_{g (E’ decay)}; T_{g (tanδ)}
⁴⁷	Glass	VE	Pultruded	6.0, 8.0	0.70 (v_f)	NA	−18°C to 4°C, tap water, sea water	24	270, 540	σ_t
³⁶	Glass	UPR, VE	Vacuum infusion	6.5	0.57 (v_f)	Water at 20°C for 30 days	−20°C to 23°C, tap water	13.5	100, 200, 300	σ_t; E_t; σ_c; τ_xy; G_xy; σ_ILS; σ_f; T_{g (E’ decay)}; T_{g (tanδ)}

σ_t: tensile strength; E_t: tensile modulus; ε_t: tensile strain; σ_f: flexural strength; E_f: flexural modulus; σ_c: compressive strength; E_c: compressive modulus; τ_xy: in-plane shear strength; G_xy: in-plane shear modulus; σ_ILS: interlaminar shear strength; T_{g (E’ decay)}: glass transition temperature based on onset of E’ decay; T_{g (tanδ)}: glass transition temperature based on the peak of tanδ curve; i: isophthalic; m_f: mass fraction; v_f: volume fraction; UPR: unsaturated polyester resin; VE: vinyl ester resin; u: unidirectional fibres; w: woven fibres; and NA: not available.

Figure 1.

Effects of freeze-thaw cycling on the mechanical properties of FRP composites.

Despite the availability of several FT-related studies on conventional FRP composites, as listed in Table 1, a critical gap remains in the literature regarding preconditioning composites in water prior to FT exposure. This preconditioning step, explicitly outlined in ASTM D7792/D7792 M,⁴⁸ is essential for accurately simulating realistic FT degradation mechanisms. Immersing the composites in water prior to FT cycling enables moisture to occupy the microstructural voids and pores, thereby facilitating more representative freeze-induced damage, such as internal cracking, delamination, and fibre–matrix debonding. Overall, the number of studies that rigorously assess FT durability of FRP composites, particularly those incorporating bio-based polymer matrices, remains extremely limited and thus warrants further investigation.

Literature on FT durability consistently reports that matrix-dominated properties are more susceptible to degradation than fibre-dominated properties under environmental exposure.^36,49 Furthermore, the temperature range of the FT cycling plays a critical role in governing the extent of damage mechanisms. FT cycles with a lower upper temperature limit induce greater degradation in composite properties^50,51 compared to those with a higher upper bound.^30,52 This discrepancy is likely due to post-curing effects at elevated temperatures, which counterbalance degradation of the properties of FRP composites.

Furthermore, the available FT studies listed in Table 1 reveal significant discrepancies. Most notably, the absence of standardised testing procedures results in considerable variation in the number of cycles and exposure durations across different studies. This issue is further complicated by the diversity of FRP material variables, including fibre and resin types, material thickness, manufacturing methods, and component geometry. These factors collectively hinder direct comparison of degradation behaviour in FRP materials under FT conditions. Moreover, the limited number of FT studies specifically addressing bio-resin-based GFRP composites intensifies this challenge. Consequently, there is a pressing need to follow the standardised testing protocols and more comprehensive investigations, particularly focusing on bio-resin GFRP composites, to enable meaningful comparisons and robust conclusions.

To address the aforementioned concerns, a study was conducted about the FT durability of a sustainable FRP composite utilizing a bio-based UPR matrix reinforced with glass fibres, manufactured via vacuum infusion.¹^,53). The testing followed the ASTM D7792/D7792 M standard,⁴⁸ including a preconditioning step where the specimens were submerged in water for 30 days at a constant temperature of 20°C. The resulting damage from both preconditioning and FT cycling was evaluated through a comprehensive suite of mechanical and thermomechanical characterisations. These analyses aimed to assess the FT durability of the bio-UPR-based GFRP composite, with particular attention to the preservation of fibre reinforcement and overall structural stability. For comparative purposes, an identical study was performed on a GFRP composite made with an oil-based UPR, with the same glass fibre architecture. This enabled a direct comparison of the performance and durability between the bio-based and the conventional oil-based UPR composites.

Materials

Polymer resins

In this study, a partially bio-based UPR (UPR-Bio) was used to develop the GFRP composite material. The UP polymer chain was synthesised using renewably sourced chemical platforms, namely fumaric diacid, isosorbide, and 1,3-propanediol, alongside oil-based phthalic anhydride. Fumaric acid fully replaced conventional oil-based unsaturated diacids, such as maleic acid, while isosorbide and 1,3-propanediol completely substituted conventional diols. The molar ratio of diacids to diols was maintained at 48:52, resulting in a bio-derived content of approximately 84 wt.% in the UP polymer chain according to the bio-based mass content determination method outlined in ISO 16620-4.⁵⁴

To crosslink the UP polymer chain, styrene was used in combination with 2-hydroxyethyl methacrylate (HEMA), replacing styrene partially (50%). The final composition of the bio-based UPR consisted of 60 wt.% UP polymer chain, 20 wt.% styrene, and 20 wt.% HEMA, yielding an overall bio-based content of around 53 wt.%. The UPR-Bio was cured using a radical initiator system comprising methyl ethyl ketone peroxide (MEKP, 2 wt.% of neat resin) and cobalt octoate (1 wt.% of neat resin) as an accelerator. Further details on the resin synthesis and formulation can be found in 13,23.

For comparison, a conventional fossil/crude oil-based UPR (UPR-Oil), branded as Crystic U 904LVKTM and supplied by Scott Bader (Wollaston, UK), was also studied. Since UPR-Oil was pre-accelerated, curing was performed by mixing with only MEKP (1 wt.% of neat resin).

Table 2 presents the initial short-term mechanical and thermomechanical properties of UPR-Bio and its oil-based counterpart. The tensile strength (σ_t) and in-plane shear strength (τ_xy) of UPR-Bio were found to be 30% and 20% higher than those of UPR-Oil, respectively, while the moduli of both resins were comparable. However, UPR-Bio exhibited a lower glass transition temperature (T_g) by 9–13°C, compared to the conventional UPR-Oil.

Table 2.

Mechanical and thermomechanical properties of bio-based and oil-based UPRs.

Properties	UPR-Bio	UPR-Oil	Standard
σ_t [MPa]	58.8 ± 7.0	45.3 ± 3.3	ISO 527-2:2012 ⁵⁵
E_t [GPa]	3.3 ± 0.1	3.3 ± 0.1	ISO 527-2:2012 ⁵⁵
τ [MPa]	64.9 ± 4.4	54.1 ± 2.5	ASTM D5379/D5379M-19 ⁵⁶
G [GPa]	1.5 ± 0.1	1.3 ± 0.1	ASTM D5379/D5379M-19 ⁵⁶
T_{g (E’ decay)} [°C]	62.2	75.3	ISO 6721-11:2019 ⁵⁷
T_{g (tanδ)} [°C]	93.6	102.4	ISO 6721-11:2019 ⁵⁷

σ_t: Tensile strength; E_t: Tensile modulus; τ: In-plane shear strength; G: In-plane shear modulus;

T_{g (E’ decay)}: Glass transition temperature (based on onset of E’ decay); and T_{g (tanδ)}: Glass transition temperature (based on the peak of tanδ curve).

Composite production

The GFRP composites were fabricated by stacking ten layers of glass fibre mats, comprising two types: unidirectional (U) and plain weave bidirectional (B), arranged in the sequence [U/B/U/B/U/U/B/U/B/U]. The unidirectional and bidirectional fibre mats were sourced from S&P Clever Portugal and Castro Composites (Spain), respectively, with areal weights of 440 g/m² and 500 g/m². The fibre mats were impregnated with either UPR-Bio or UPR-Oil resin via a vacuum infusion process to produce the bio-based (GFRP-Bio) and oil-based (GFRP-Oil) composites.

Prior to infusion, peroxide was added to the UPRs, followed by degassing for 15 min to remove entrapped air bubbles. The resin was then infused into the fibre stack. The composites were cured at room temperature for 48 h and subsequently post-cured at 100°C for 4 hours. The nominal thickness of the composites was 4.6 mm, with a glass fibre mass fraction of 65%. Finally, the composite plates were machined to dimensions of 350 mm × 170 mm using a Computer Numerical Control (CNC) milling machine.

Methodology

Freeze-thaw environment

In this study, the FT resistance of the bio-based GFRP composite materials was evaluated following the ASTM D7792/D7792 M⁴⁸ (ASTM, 2015) standard. The environmental exposure was conducted using a FT chamber (Aralab FITOCLIMA 500 EPC45). The composite plates, with the aforementioned dimensions, were first preconditioned by immersion in tap water at 20°C for 30 days to ensure water diffusion into the GFRP composites and to promote (later) the effect of repeated thermal cycling in the presence of moisture.

Following preconditioning, the wet GFRP plates were subjected to FT cycling in the chamber for 100, 200, and 300 cycles. Each cycle lasted 13 h and 25 min, with temperatures cycling between-20°C and 23°C, as illustrated in Figure 2. During the thawing phase at 23°C, tap water was introduced into the chamber within 15 min, fully submerging the composite plates, which were then kept submerged for 8 h. At the end of the thawing period, water was drained over 40 min, followed by cooling the chamber to −20°C for 45 min. The composites then remained under dry freezing conditions for 3 h before the chamber was heated back to 23°C over 45 min, completing the cycle.

Figure 2.

Freeze-thaw cycling protocol during one complete cycle.

After preconditioning and each set FT cycles, the composite plates were removed from the chamber and kept submerged in water at 20°C. The plates were then cut using a CNC milling machine to dimensions specified by mechanical and thermomechanical testing standards defined ahead. Throughout this period, all test specimens were stored in water and tested within 2 days to maintain consistent moisture content.

Water uptake

In parallel with the FT cycling exposure, the water uptake behaviour of the GFRP composite materials was evaluated through immersion in water at 20°C for up to 180 days. This testing was based on guidelines from ISO 62,⁵⁸ ASTM D5229,⁵⁹ and ASTM D570⁶⁰ standards. Specimens measuring 100 mm × 100 mm were prepared, with three replicates tested for each composite type. Prior to immersion, specimens were cleaned on their edges and surfaces, then dried in a thermal chamber at 35°C and 30% relative humidity (RH) for 24 h, followed by conditioning at 25°C and 30% RH for 60 days until constant mass was achieved (defined as mass variation within 0.1%).

Water uptake was recorded at intervals of 1, 2, 4, 8, 14, 21, 30, 60, 90, 124, and 180 days. At each interval, specimens were removed from the chamber, the surface water was carefully wiped off using lint-free cloths within 1 minute to ensure consistency and then weighed to determine the moisture gain.

Mechanical and thermomechanical characterisations

The damage caused by both preconditioning and FT cycling was evaluated through a series of mechanical tests, including tensile, compressive, in-plane shear, and interlaminar shear, alongside thermomechanical characterisation using dynamic mechanical analysis (DMA). All mechanical tests were performed on a universal testing machine (Instron 5982, Norwood, MA, USA) with 100 kN load capacity, under displacement control. For each material and property, a minimum of five specimens were tested, cut from the 350 × 170 mm plates (excluding material close to their edges).

Tensile properties of the GFRP composites were determined according to ISO 527-4:2023,⁶¹ using specimens sized 300 mm × 25 mm and tested at a crosshead speed of 2 mm/min. Compressive strength testing followed ISO 14126:2023,⁶² with specimens measuring 150 mm × 25 mm, tested at 1.3 mm/min. In-plane shear strength, perpendicular to the composite longitudinal axis, was evaluated following ASTM D5379/D5379M-19⁵⁶ using V-notched specimens (76 mm × 20 mm), loaded at 2 mm/min. Interlaminar shear strength was measured according to ISO 14130:1997,⁶³ with specimens sized 46 mm × 23 mm and tested at 1 mm/min.

Maximum failure load was used to calculate strength for all tests except in-plane shear, where strength was determined using the 5% strain offset method. Tensile and in-plane shear strains were measured using a video extensometer system consisting of a SONY FLIR BFS-U3-51S5M-C camera equipped with a Fujinon HF25SA-1 lens, operated through MatchID software (v2022.2). The tensile modulus was calculated according to the respective standard.

Thermomechanical properties were assessed by dynamic mechanical analysis (DMA) in accordance with ISO 6721-11:2019,⁵⁷ using specimens of 60 mm × 10 mm. The DMA employed a two-point cantilever setup with an amplitude of 15 μm and frequency of 1 Hz, over a temperature range from -30°C to 150°C with a heating rate of 2°C/min. The glass transition temperature (T_g) was determined using two criteria: (a) onset of the storage modulus decay, T_{g (E’ decay)}, and (b) peak of the tan δ curve, T_{g (tanδ)}.

Results and discussion

Water absorption

Figure 3 illustrates the water diffusion behaviour of both bio-based and oil-based GFRP composites immersed in water at 20°C for up to 180 days. It is evident that the bio-based composite exhibited greater hydrophilicity compared to the oil-based counterpart, with water absorption after 30 days of immersion reaching approximately 2.1% in GFRP-Bio, whereas GFRP-Oil absorbed only around 0.4%. The elevated water uptake in the bio-based GFRP is primarily attributed to the bio-based UPR composition, which contains hydrophilic components, such as HEMA, isosorbide, and 1,3-propanediol, all of which have a strong affinity for water.⁶⁴ Furthermore, the potentially poorer fibre-matrix adhesion in the bio-based composite may contribute to increased water absorption, as micro-cracks caused by insufficient fibre encapsulation can serve as pathways for water ingress. Additionally, water facilitates hydrolysis of ester bonds within the polymer chain, promoting further moisture accumulation.

Figure 3.

Water uptake profile of bio-based and oil-based GFRP (error bars = 1 SD).

When compared to water uptake studies by Sousa et al. (2021)⁶⁵ and Hasan et al. (2024),³⁶ which examined conventional UPR-based GFRP composites with thicknesses of 5 mm (pultruded) and 6.5 mm (vacuum infused), respectively, the GFRP-Oil composite in this study exhibited 1.5 to 2 times higher water ingress. This difference is likely due to the thinner specimens and uncoated edges used in the current water uptake tests. Moreover, GFRP-Bio showed approximately 20% higher water absorption compared to the bio-based basalt-FRP (BFRP) composite studied by Shahid et al. (2023).²⁴ This increase is mainly attributed to differences in resin composition: the bio-based UPR in GFRP-Bio contains about 10% more UP polymer chain and 10% less styrene than the bio-based UPR used in the BFRP composite.

Dynamic mechanical analysis

Figure 4 presents the DMA test results, showing the impact of preconditioning and FT exposure on bio-based (Figure 4(a)) and oil-based (Figure 4(b)) GFRP composites, compared to unaged specimens. The evaluated properties include storage modulus and damping behaviour. Additionally, the variations in glass transition temperature (T_g) based on two criteria, T_{g (E’ decay)} and T_{g (tanδ)}, are plotted in Figures 5a and 5(b), respectively. The respective T_g estimates are detailed in Table 3.

Figure 4.

Effect of preconditioning and FT cycling on storage modulus (E’) and tanδ curves of (a) bio-based and (b) oil-based GFRP composites.

Figure 5.

Variation of T_g of bio-based and oil-based GFRP composites based on (a) T_{g (E’ decay)}, and (b) T_{g (tanδ)}.

Table 3.

Average ( $\bar{x}$ ), standard deviation (SD), coefficient of variation (CV) and retention (R) of mechanical properties and T_g of bio-based and oil-based GFRP composites in unaged, preconditioned, and freeze-thaw cycling exposure.

Property	Parameter	Unaged	Bio-Composite						Oil-Composite
Property	Parameter	Unaged	Precondition	100 Cycle	200 Cycle	300 Cycle	Unaged	Precondition	100 Cycle	200 Cycle	300 Cycle
σ_t	$\bar{x}$ [MPa]	538	408	335	325	310	533	433	439	413	399
	SD	24	19	7	10	7	18	14	6	12	7
	CV [%]	4.4	4.6	2.2	3.1	2.1	3.5	3.3	1.3	2.9	1.8
	R [%]	-	76	62	60	58	-	81	82	78	75
E_t	$\bar{x}$ [GPa]	21.6	23.8	21.4	20.0	20.7	22.3	23.0	24.3	22.7	23.9
	SD	0.8	0.3	0.4	0.4	0.4	0.9	0.3	0.5	0.5	0.5
	CV [%]	3.8	1.2	1.7	2.2	1.9	3.9	1.4	2.2	2.0	2.1
	R [%]	-	110	99	92	96	-	103	109	101	107
τ_xy	$\bar{x}$ [MPa]	33.9	26.7	16.7	13.8	17.4	47.1	43.8	41.0	37.3	35.1
	SD	3.1	1.0	0.6	0.3	1.0	0.8	1.1	2.3	1.2	1.2
	CV [%]	9.3	3.6	3.8	2.3	6.0	1.6	2.4	5.6	3.1	3.4
	R [%]	-	79	49	41	51	-	93	87	79	75
σ_c	$\bar{x}$ [MPa]	152	114	83.2	58.0	78.7	203	171	187	180	157
	SD	24	8	7.5	10.9	7.0	18	10	8	21	6
	CV [%]	16.0	6.6	9.0	18.7	8.9	9.0	6.1	4.4	11.5	3.8
	R [%]	-	75	55	38	52	-	84	92	89	77
σ_ILS	$\bar{x}$ [MPa]	21.0	15.4	14.4	8.8	9.7	21.0	18.6	21.4	20.9	16.4
	SD	2.5	0.7	0.4	0.1	0.3	0.7	1.2	0.4	0.5	0.9
	CV [%]	11.7	4.5	2.9	1.4	3.6	3.4	6.6	1.8	2.6	5.6
	R [%]	-	73	69	42	47	-	88	102	99	78
T_{g (E’ decay)}	[°C]	55.8	27.8	40.5	35.9	34.7	72.1	64.3	62.8	64.9	62.9
T_{g (E’ decay)}	R [%]	-	50	73	64	62	-	89	87	90	87
T_{g (tanδ)}	[°C]	76.8	53.2	60.6	56.6	54.7	94.2	87.3	87.6	88.7	88.0
T_{g (tanδ)}	R [%]	-	69	79	74	71	-	93	93	94	94

σ_t: tensile strength; E_t: tensile modulus; σ_c: compressive strength; τ_xy: in-plane shear strength; σ_ILS: interlaminar shear strength; T_{g (E’ decay)}: glass transition temperature based on onset of E’ decay; T_g (tanδ): glass transition temperature based on the peak of tanδ curve.

As seen in Figure 4(a), the storage modulus curves of both preconditioned and FT-aged GFRP-Bio composites shifted left relative to the unaged curve, indicating that the transition from the glassy to rubbery state occurred at lower temperatures. A similar but less pronounced shift was observed for the GFRP-Oil composites (Figure 4(b)). Comparable trends were found in the damping curves (Figure 5).

Preconditioning through 30 days of water immersion caused notable water-related degradation, including hydrolysis and plasticization, particularly in the bio-based composite. In this respect, it is worth referring that, in parallel to the experiments described in this paper, the materials used herein were subjected to hygrothermal ageing, involving mass uptake measurements and various mechanical tests, which clearly indicated the occurrence of plasticization and hydrolysis⁵³ (Shahid, 2026). These degradation mechanisms resulted in a substantial decrease in T_{g (E’ decay)}, from 55.8°C (unaged) to 27.8°C (a 50% reduction), and T_{g (tanδ)}, from 76.8°C to 53.2°C (a 31% reduction). In contrast, the oil-based composite exhibited much smaller reductions during preconditioning, with T_{g (E’ decay)} dropping from 72.1°C to 64.3°C (11%) and T_{g (tanδ)} from 94.2°C to 87.3°C (7%), as shown in Table 3.

Interestingly, the bio-based GFRP demonstrated partial property recovery during FT cycling. After 100 FT cycles, T_{g (E’ decay)} rose by 50% (from 27.8°C to 40.5°C) relative to the preconditioned state and stabilised with some fluctuations upon further cycling. T_{g (tanδ)} showed a similar recovery trend. In contrast, the T_g values for GFRP-Oil remained largely unchanged throughout FT cycling. At the conclusion of 300 FT cycles, T_{g (E’ decay)} and T_{g (tanδ)} for GFRP-Bio were 34.7°C and 54.7°C, respectively, while GFRP-Oil maintained higher values of 62.9°C and 88.0°C.

The observed recovery in the bio-based composite is likely due to factors such as embrittlement and crystallisation of the polymer matrix during freezing. Additionally, water diffusing into the matrix freezes and induces stress, compounded by localised pressures arising from the mismatch in the CTE between the polymer and glass fibres. The higher hydrophilicity of the bio-based composite, as confirmed by the water uptake results in Figure 3, may have accelerated damage progression in GFRP-Bio. These findings align with previous FT cycling studies on oil-based GFRP composites by Grammatikos et al. (2016)³⁵ and Hasan et al. (2024),³⁶ which also reported negligible effects on T_g after FT exposure.

Tensile properties

Figures 6(a) and 6(b) show the tensile stress-strain curves of representative specimens for the bio-based and oil-based GFRP composites, respectively, illustrating the effects of FT exposure. The retention (%) of tensile strength (σ_t) and tensile modulus (E_t) after preconditioning and FT cycling are also presented accordingly in Figures 7(a) and 7(b), with the absolute values detailed in Table 3.

Figure 6.

Tensile stress-strain curves (representative specimens) of (a) GFRP-Bio, and (b) GFRP-Oil composites.

Figure 7.

Retention curve of (a) tensile strength (σ_t), and (b) tensile modulus (E_t) of the GFRP-Bio and GFRP-Oil composites.

From Figure 6 and 7, it is evident that both bio-based and oil-based composites exhibit an overall decline in σ_t due to preconditioning and FT exposure. A pronounced reduction in σ_t occurs during the preconditioning phase for both composites. This reduction continues through 100 FT cycles for the bio-based GFRP, after which the σ_t stabilises. Conversely, GFRP-Oil shows minimal changes during FT cycling. Following preconditioning, σ_t retention was approximately 76% for the bio-based and 81% for the oil-based composites, decreasing further to 58% and 75%, respectively, after 300 FT cycles.

In contrast, the tensile modulus (E_t) retention (Figure 7(b)) increased by 3%–10% for both composites after preconditioning compared to unaged specimens, with the bio-based composite showing a more pronounced increase. However, the E_t in the bio-based GFRP gradually declined during FT cycling, yet retention remained above 90% by the end of 300 cycles. Better performance in E_t compared to σ_t, mainly derived from the fact that is a more fibre-dominated property.

The degradation during the water-submerged preconditioning is primarily attributed to water-induced damage mechanisms, such as diffusion, plasticization, hydrolysis, and chain scission of the polyester matrix.^4,65 These processes weaken fibre protection by promoting fibre-matrix debonding and matrix deterioration. Furthermore, poorly bonded fibre-matrix interphases act as pathways for water ingress, exacerbating damage progression, as supported by the moisture uptake depicted in Figure 3. During FT cycling, simultaneous recovery and degradation of material properties occur, as reflected in the T_g behaviour of the bio-based composite. However, recovery within the polymer matrix does not sufficiently improve fibre-matrix bonding, leading to continued σ_t degradation up to 100 FT cycles, followed by stabilisation.

Previous studies reported that tensile properties of bio-based composites are generally unaffected by FT exposure.^24,38 In the earlier study,²⁴ SEM analysis did not reveal any changes in degradation mechanisms upon freeze-thaw cycling. In contrast, this study observed a comparatively higher degradation (10%–14%) in tensile strength for the GFRP-Bio composite used. This discrepancy mainly originates from the differences in the resin composition, as discussed earlier. Meanwhile, the oil-based composite results align with previous findings,^35,36 showing moderate FT effects.

Compressive strength

Figure 8 illustrates the retention (%) in compressive strength (σ_c) of GFRP composites following preconditioning and FT environmental exposure, comparing bio-based and oil-based materials. The corresponding absolute values are provided in Table 3. Overall, the bio-based GFRP exhibited greater degradation than its oil-based counterpart.

Figure 8.

Compressive strength retention of (a) GFRP-Bio and (b) GFRP-Oil composites.

During the preconditioning phase, GFRP-Bio experienced a substantial 25% loss in σ_c. Subsequent FT cycling led to further degradation, resulting in a final retention of 52% after 300 FT cycles. In contrast, GFRP-Oil showed good resistance to environmental ageing, with only a 10% reduction observed after 300 FT cycles (90% retention).

These findings indicate that GFRP-Bio is significantly more susceptible to moisture-induced degradation mechanisms and fibre-matrix debonding, processes that critically impact compressive strength due to its matrix-dominated nature. Furthermore, the hydrophilic nature of components within the bio-based resin, such as isosorbide, 1,3-propanediol, and HEMA, which contain hydroxyl (-OH) groups, likely contributes to the additional moisture-induced damage.

FT cycling introduces additional stress through thermal expansion and contraction. The mismatch in the CTE between the polymer matrix and the glass fibres contributes to micro-damage at the fibre-matrix interface. However, post-curing effects induced during low-temperature freezing cycles appear to offer partial property recovery. For GFRP-Bio, this phenomenon became noticeable after 200 FT cycles, resulting in a ∼10% recovery by the end of 300 cycles. In the case of GFRP-Oil, post-curing effects were evident earlier, with a ∼9% strength gain after 100 cycles. However, prolonged exposure shifted the dominant mechanisms back to water- and FT-related degradation, leading to a net 10% loss after 300 cycles.

Compared to previous findings by Hasan et al. (2024),³⁶ where oil-based GFRP showed 84% σ_c retention after 300 FT cycles, GFRP-Oil in this study performed almost similar with 90% retention. Conversely, GFRP-Bio displayed inferior FT durability when compared to the bio-based BFRP composites studied by Shahid et al. (2023),²⁴ which is consistent with the degradation trends and resin-related vulnerabilities discussed in earlier sections.

In-plane shear strength

Figure 9 presents the in-plane shear stress–strain curves of representative specimens for GFRP-Bio and GFRP-Oil composites under unaged, preconditioned, and FT-aged conditions, with Figures 9(a) and 9(b) showing the respective behaviours. The corresponding retention (%) of in-plane shear strength (τ_xy) at 5% strain is provided in Figure 10, and absolute values are detailed in Table 3.

Figure 9.

In-plane shear stress-strain curves (representative specimens) of (a) GFRP-Bio, and (b) GFRP-Oil composites.

Figure 10.

Retention of in-plane shear strength (τ_xy) at 5% strain of bio-based and oil-based GFRP composites.

Although the unaged τ_xy of GFRP-Bio was comparable to that of GFRP-Oil, the bio-based composite experienced a substantial decline in performance due to environmental exposure. Preconditioning alone led to a 21% reduction in τ_xy of GFRP-Bio. This trend continued throughout FT cycling, culminating in a 49% degradation after 300 cycles. In contrast, the oil-based GFRP exhibited a more gradual decline, with only a 25% reduction at the end of 300 cycles.

Both composites also showed signs of matrix softening and plasticisation, consistent with moisture-induced degradation mechanisms. The deterioration in τ_xy can be attributed to several factors discussed previously, particularly water diffusion, plasticisation, hydrolytic scission of ester bonds, and the weakening of the fibre-matrix interphase. These mechanisms are especially detrimental in matrix-dominated properties, like τ_xy.

Thermal stresses from repeated FT cycles further exacerbated the damage, particularly in GFRP-Bio, due to the mismatch in thermal expansion between the matrix and fibre, and its inherently higher water uptake. The combined effect of mechanical softening, internal stress build-up, and moisture damage contributed to the severe reduction in shear performance.

Compared to the literature, the τ_xy of oil-based GFRP studied by Hasan et al. (2024)³⁶ remained largely unaffected by FT exposure, while Grammatikos et al. (2016)³⁵ reported a 70% retention after 300 FT cycles. In this study, GFRP-Oil showed slightly better retention (+5%) than the former study. On the other hand, GFRP-Bio performed significantly worse than the bio-based BFRP examined by Shahid et al. (2023),²⁴ reflecting the higher vulnerability of the GFRP-Bio resin system to environmental ageing.

Interlaminar shear strength

The effect of preconditioning and FT cycling on the retention of interlaminar shear strength (σ_ILS) for the GFRP composites is illustrated in Figure 11, while the absolute values are detailed in Table 3.

Figure 11.

Retention of interlaminar shear strength of bio-based and oil-based GFRP composites.

The unaged σ_ILS of both bio- and oil-based composites were found to be comparable. However, consistent with the trends observed in other matrix-dominated properties, such as σ_c and τ_xy, GFRP-Bio showed greater susceptibility to degradation under FT environmental exposure. Following the 30-day water immersion preconditioning, the σ_ILS retention dropped to 73% in GFRP-Bio and to 88% in GFRP-Oil.

In GFRP-Bio, σ_ILS remained relatively stable up to 100 FT cycling, with retention of 69%, but a marked degradation occurred after 200 cycles, reducing the retention to 42%. This was followed by slight fluctuations and eventual stabilisation at 47% retention after 300 FT cycles. In contrast, GFRP-Oil exhibited a mild improvement in σ_ILS after 100 FT cycles – reaching the value of the unaged material – likely due to post-curing effects. However, the strength began to decline in the later exposure stages, ending with a retention of 78% after 300 FT cycles.

The σ_ILS degradation observed is primarily attributed to moisture-induced damage mechanisms, namely, plasticisation, hydrolysis, and chain scission of the polyester matrix, as discussed before. These mechanisms compromise the interfacial bonding and promote matrix softening. In the case of GFRP-Bio, the higher hydrophilicity and inherently weaker fibre-matrix interfacial adhesion led to more permanent damage during FT cycling, particularly beyond 100 cycles. While some matrix property recovery was evident (as seen in T_g results), it did not effectively restore interfacial strength, which is crucial for σ_ILS. In contrast, GFRP-Oil benefited from early-stage post-curing during FT exposure, helping to maintain its interfacial integrity for longer, though degradation mechanisms eventually dominated in later stages due to matrix cracking and fibre-matrix debonding.

Compared to previous studies, Hasan et al. (2024)³⁶ reported negligible σ_ILS degradation (−3%) in oil-based GFRP composites after 300 FT cycles. In this study, the GFRP-Oil showed moderately higher degradation (−22%), possibly due to differences in material configuration, thickness, or edge protection. On the other hand, the bio-based composite demonstrated inferior performance compared to the bio-based BFRP reported by Shahid et al. (2023),²⁴ which retained approximately 70% σ_ILS (−30% degradation) after 300 FT cycles, while the GFRP-Bio, studied herein, retained only 47% under similar conditions.

Summary and comparison with other studies

Figure 12 plots the retention of the mechanical properties of various fibre-polymer composite materials exposed to FT cycling environment, including the studies summarized in Table 1 and the present study: Figure 12(a) shows the σ_t, E_t, and E_c, while Figure 12(b) shows the remaining mechanical properties, together with the 50-year moisture conversion factor for continuous water exposure (class III), η_cm (0.60), set in CEN/TS 19101:2022.³⁹ This differentiation was made based on the properties investigated in this study, where σ_t and E_t suffered less (property retention generally varying between 80% and 120% based on literature) compared to the other mechanical properties, namely σ_c, τ_xy, and σ_ILS (relative variations between 60% and 120% based on literature).

Figure 12.

Comparison of data from the current study and literature regarding retention (%) of properties: (a) tensile strength (σ_t) and modulus (E_t), and compressive modulus (E_c), and (b) remaining mechanical properties.

Although all these studies have reported the FT durability of fibre-polymer composite materials, it is difficult to compare them directly, as they involve significant differences, such as resin and fibre types, laminate thickness, duration of ageing, and assessment methods. In general, the σ_t, E_t, and E_c were found to be less significantly affected by the environmental exposure, as illustrated in Figure 12(a). A similar trend was observed in this study, with property retention values generally remaining above the threshold of η_cm = 0.60 specified in CEN/TS 19101:2022.³⁹ Notably, the bio-based GFRP composites approached, but only marginally touched this minimum retention criterion. Regarding the other mechanical properties of the FRP composites shown in Figure 12(b), these show moderate degradation due to FT exposure. Nevertheless, all of the test data exhibited property retention above 60%, i.e., the conversion factor defined in 39, except for the bio-based GFRP composite studied herein, which presents much higher degradation compared to that threshold. This means that such design guidance, developed for fibre-polymer composites produced with (oil-based) thermoset resins, is not applicable to the bio-based composite investigated herein, especially for matrix-dominated properties.

Conclusion

This study investigated the wet freeze-thaw (FT) durability of glass fibre-reinforced polymer (GFRP) composites developed using a bio-based unsaturated polyester resin (UPR) and compared it with that of an oil-based GFRP composite with identical fibre architecture.

The initial performance of the unaged bio-based GFRP was generally comparable to that of the oil-based counterpart; notably, the bio-based GFRP exhibited a 16°C lower T_g. However, water uptake measurements at 20°C revealed that the bio-based GFRP has significantly greater hydrophilicity, absorbing nearly five times more water than the oil-based GFRP. This increased water uptake led to pronounced degradation in both mechanical and thermomechanical properties, particularly following preconditioning.

The most substantial reductions occurred during the water immersion preconditioning stage. Matrix-dominated properties were especially affected, with the in-plane shear strength of the bio-based GFRP decreasing by 20%, and that of the oil-based GFRP reducing by only 7%. In contrast, some enhancements were also observed, such as up to 10% increase in tensile modulus of both composites.

During FT cycling, the composites exhibited further degradation, accompanied by partial recovery in some properties. This recovery was primarily seen in matrix-dominated parameters, including compressive strength, T_g, and in-plane shear strength. In the case of the bio-based GFRP, the extent of recovery was insufficient to offset the accumulated damage, particularly after prolonged FT exposure.

The bio-based GFRP demonstrated greater vulnerability throughout FT ageing, primarily due to the chemical composition of the bio-UPR. The resin includes bio-based 1,3-propanediol and isosorbide, along with oil-derived hydroxyethyl methacrylate (HEMA), all of which contain hydrophilic hydroxyl (-OH) functional groups. These groups increased the resin’s affinity for moisture, thereby exacerbating water-induced degradation mechanisms, such as plasticisation, chain scission, hydrolysis, and fibre-matrix debonding. Overall, the FT cycling itself had comparatively less influence on property degradation. The oil-based GFRP demonstrated superior stability, with minimal losses even after prolonged FT exposure.

The bio-based UPR studied herein, despite its promising mechanical properties in the unaged state, exhibited poor durability in moisture-rich environments due to its hydrophilic nature. To enable its practical application in structural composites exposed to harsh environmental conditions, future work should focus on modifying the bio-UPR formulation to reduce water uptake and improve resistance to hydrothermal ageing and FT-induced damage.

Footnotes

Acknowledgement

The first author gratefully acknowledges the financial support from Fundação para a Ciência e a Tecnologia (FCT), through PhD scholarship 2022.14149.BD. Additionally, the financial support from FCT through research project PTDC/ECI-EGC/29597/2017 (EcoComposite) is also gratefully acknowledged. Moreover, the authors are also grateful for FCT’s funding through the research unit CERIS (project UIDB/04625/2025).

ORCID iD

Md Abu Toyob Shahid

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Fundação para a Ciência e a Tecnologia (2022.14149.BD, PTDC/ECI-EGC/29597/2017, UIDB/04625/2025).

Declaration of conflicting interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: A.T. Shahid, M. Garrido, and J.R. Correia are co-authors of the Portuguese patent no.117321 granted to Instituto Superior Técnico.

Data Availability Statement

All data that support the findings of this study are available within the manuscript. Further information can be provided upon request.*

Note

References

Mahdavi

Gupta

Hojjati

. Thermal cycling of composite laminates made of out-of-autoclave materials. Sci Eng Compos Mater 2018; 25(6): 1145–1156. https://doi.org/10.1515/secm-2017-0132

Muhammad

Rahman

Baini

, et al. Applications of sustainable polymer composites in automobile and aerospace industry. In: Advances in sustainable polymer composites. Woodhead Publishing, 2021, pp. 185–207.

Liu

, et al. A review on basalt fiber composites and their applications in clean energy sector and power grids. Polymers 2022; 14: 2376. https://doi.org/10.3390/polym14122376

Hamzat

Murad

Adediran

, et al. Fiber-reinforced composites for aerospace, energy, and marine applications: an insight into failure mechanisms under chemical, thermal, oxidative, and mechanical load conditions. Adv Compos Hybrid Mater 2025; 8(1): 1–57. https://doi.org/10.1007/s42114-024-01192-y

Correia

. Pultrusion of advanced composites. In: Advanced fiber-reinforced polymer (FRP) composites for structural applications. Woodhead Publishing, 2023, pp. 137–177.

Hollaway

. A review of the present and future utilisation of FRP composites in the civil infrastructure with reference to their important in-service properties. Constr Build Mater 2010; 24(12): 2419–2445. https://doi.org/10.1016/j.conbuildmat.2010.04.062

Dholakiya

. Unsaturated polyester resin for specialty applications. In: Polyester. IntechOpen, 2012, pp. 167–202.

Raquez

Deléglise

Lacrampe

, et al. Thermosetting (bio) materials derived from renewable resources: a critical review. Prog Polym Sci 2010; 35(4): 487–509. https://doi.org/10.1016/j.progpolymsci.2010.01.001

Suriano

Gonzalez

MNG

Turri

. Environmental profile and technological validation of new high-Tg unsaturated polyesters from fully bio-based monomers and reactive diluents. J Polym Environ 2021; 29: 1122–1133. https://doi.org/10.1007/s10924-020-01928-z

10.

Elfaleh

Abbassi

Habibi

, et al. A comprehensive review of natural fibers and their composites: an eco-friendly alternative to conventional materials. Results Eng 2023; 19: 101271. https://doi.org/10.1016/j.rineng.2023.101271

11.

Hofmann

Garrido

Machado

, et al. Development of high‐performance partially biobased thermoset polyester using renewable building blocks from isosorbide, 1,3‐propanediol, and fumaric acid. J Appl Polym Sci 2022; 139(42): e53029. https://doi.org/10.1002/app.53029

12.

Hofmann

Shahid

Garrido

, et al. Biobased thermosetting polyester resin for high-performance applications. ACS Sustainable Chem Eng 2022; 10(11): 3442–3454. https://doi.org/10.1021/acssuschemeng.1c06969

13.

Hofmann

Bordado

Garrido

, et al. (2023). High performance unsaturated polyester resins based on renewable resources . WO 2023/277718 A1.

14.

Akbari

Root

Skrifvars

, et al. Novel bio-based branched unsaturated polyester resins for high-temperature applications. J Polym Environ 2024; 32(5): 2031–2044. https://doi.org/10.1007/s10924-023-03112-5

15.

Rubeš

Vinklárek

Podzimek

, et al. Bio-based unsaturated polyester resin from post-consumer PET. RSC Adv 2024; 14(12): 8536–8547. https://doi.org/10.1039/d3ra08500g

16.

Zhang

Lei

Zhang

, et al. Biobased unsaturated polyester thermosets from castor oil and isosorbide with life cycle assessment. ACS Sustainable Chem Eng 2024; 13(1): 374–385. https://doi.org/10.1021/acssuschemeng.4c07624

17.

Shahid

Silvestre

Hofmann

, et al. Life cycle assessment of an innovative bio-based unsaturated polyester resin and its use in glass fibre reinforced bio-composites produced by vacuum infusion. J Clean Prod 2024; 441: 140906. https://doi.org/10.1016/j.jclepro.2024.140906

18.

Campanella

Scala

JJL

Wool

. Fatty acid‐based comonomers as styrene replacements in soybean and castor oil‐based thermosetting polymers. J Appl Polym Sci 2011; 119(2): 1000–1010. https://doi.org/10.1002/app.32810

19.

Cousinet

Ghadban

Fleury

, et al. Toward replacement of styrene by bio-based methacrylates in unsaturated polyester resins. Eur Polym J 2015; 67: 539–550. https://doi.org/10.1016/j.eurpolymj.2015.02.016

20.

Sousa

Fonseca

Serra

, et al. New unsaturated copolyesters based on 2, 5-furandicarboxylic acid and their crosslinked derivatives. Polym Chem 2016; 7(5): 1049–1058. https://doi.org/10.1039/c5py01702e

21.

Spasojevic

Seslija

Markovic

, et al. Optimization of reactive diluent for bio-based unsaturated polyester resin: a rheological and thermomechanical study. Polymers 2021; 13(16): 2667. https://doi.org/10.3390/polym13162667

22.

Akbari

Joseph

Skrifvars

, et al. Glass fiber reinforced composite produced with a novel Matrix of bio-based unsaturated polyester resin made from 2, 5-Furan dicarboxylic acid and isosorbide. J Polym Environ 2025; 33: 2798–2812. https://doi.org/10.1007/s10924-025-03539-y

23.

Hofmann

Shahid

Machado

, et al. GFRP biocomposites produced with a novel high-performance bio-based unsaturated polyester resin. Compos Appl Sci Manuf 2022c; 161: 107098. https://doi.org/10.1016/j.compositesa.2022.107098

24.

Shahid

Hofmann

Garrido

, et al. Freeze–Thaw durability of Basalt Fibre reinforced bio-based unsaturated polyester composite. Materials 2023; 16(15): 5411. https://doi.org/10.3390/ma16155411

25.

Hofmann

Machado

Shahid

, et al. Pultruded carbon fibre reinforced polymer strips produced with a novel bio-based thermoset polyester for structural strengthening. Compos Sci Technol. 2023; 234: 109936. https://doi.org/10.1016/j.compscitech.2023.109936

26.

Shahid

Hofmann

Silvestre

, et al. Life cycle assessment of novel partially bio-based unsaturated polyester resins. J Polym Environ 2025; 33: 1–19. https://doi.org/10.1007/s10924-025-03596-3

27.

Karbhari

. Durability of composite for civil structural applications. 1st ed. Woodhead Publishing, 2007.

28.

Zaman

Gutub

Wafa

. A review on FRP composites applications and durability concerns in the construction sector. J Reinforc Plast Compos 2013; 32(24): 1966–1988. https://doi.org/10.1177/0731684413492868

29.

Karbhari

. E-glass/vinylester composites in aqueous environments: effects on short-beam shear strength. J Compos Construct 2004; 8(2): 148–156. https://doi.org/10.1061/(asce)1090-0268(2004)8:2(148)

30.

Gibson

, et al. Durability of FRP composite bridge deck materials under freeze-thaw and low temperature conditions. J Bridge Eng 2006; 11(4): 443–451. https://doi.org/10.1061/(asce)1084-0702(2006)11:4(443)

31.

Aniskevich

Korkhov

Faitelsone

, et al. Mechanical properties of pultruded glass fiber reinforced plastic after freeze–thaw cycling. J Reinforc Plast Compos 2012; 31(22): 1554–1563. https://doi.org/10.1177/0731684412462755

32.

Hussnain

Shah

SZH

Megat-Yusoff

PSM

, et al. Degradation and mechanical performance of fibre-reinforced polymer composites under marine environments: a review of recent advancements. Polym Degrad Stabil 2023; 215: 110452. https://doi.org/10.1016/j.polymdegradstab.2023.110452

33.

Apicella

Migliaresi

Nicolais

, et al. The water ageing of unsaturated polyester-based composites: influence of resin chemical structure. Composites 1983; 14(4): 387–392. https://doi.org/10.1016/0010-4361(83)90160-x

34.

Sousa

Correia

Cabral-Fonseca

, et al. Effects of thermal cycles on the mechanical response of pultruded GFRP profiles used in civil engineering applications. Compos Struct 2014; 116: 720–731. https://doi.org/10.1016/j.compstruct.2014.06.008

35.

Grammatikos

Zafari

Evernden

, et al. Moisture uptake characteristics of a pultruded fibre reinforced polymer flat sheet subjected to hot/wet aging. Polym Degrad Stabil 2015; 121: 407–419. https://doi.org/10.1016/j.polymdegradstab.2015.10.001

36.

Hasan

Correia

Garrido

, et al. Freeze-thaw durability of vacuum infused glass fibre composites with unsaturated polyester and vinyl ester matrices. Constr Build Mater 2024; 455: 139037. https://doi.org/10.1016/j.conbuildmat.2024.139037

37.

Wang

Sze

, et al. Effects of relative humidity, alkaline concentration and temperature on the degradation of plain/coated glass fibers: experimental investigation and empirical degradation model. Constr Build Mater 2023; 391: 131757. https://doi.org/10.1016/j.conbuildmat.2023.131757

38.

McIsaac

Fam

. Durability under freeze–thaw cycles of concrete beams retrofitted with externally bonded FRPs using bio-based resins. Constr Build Mater 2018; 168: 244–256. https://doi.org/10.1016/j.conbuildmat.2018.02.062

39.

CEN . CEN/TS 19101:2022 – design of fibre-polymer composite structure. CEN (European Committee for Standardization), 2022.

40.

Gomez

Casto

. Freeze-thaw durability of composite materials (No. VTRC 96-R25). Virginia Transportation Research Council, 1996. (VTRC).

41.

Zhang

Karbhari

Reynaud

. NOL-ring based evaluation of freeze and freeze–thaw exposure effects on FRP composite column wrap systems. Compos B Eng 2001; 32(7): 589–598. https://doi.org/10.1016/s1359-8368(01)00035-x

42.

Shao

Kouadio

. Durability of fiberglass composite sheet piles in water. J Compos Construct 2002; 6(4): 280–287. https://doi.org/10.1061/(asce)1090-0268(2002)6:4(280)

43.

Liu

Shaw

Parnas

, et al. Investigation of basalt fiber composite aging behavior for applications in transportation. Polym Compos 2006; 27(5): 475–483. https://doi.org/10.1002/pc.20215

44.

Zhang

. Study on long-term performance of pultruded FRP for ocean applications. Doctoral dissertation Thesis. Tsinghua University, 2018.

45.

Bazli

Ashrafi

Jafari

, et al. Effect of fibers configuration and thickness on tensile behavior of GFRP laminates exposed to harsh environment. Polymers 2019; 11(9): 1401. https://doi.org/10.3390/polym11091401

46.

Demircan

Ozen

Kisa

, et al. The effect of nano-gelcoat on freeze-thaw resistance of glass fiber-reinforced polymer composite for marine applications. Ocean Eng 2023; 269: 113589. https://doi.org/10.1016/j.oceaneng.2022.113589

47.

Shakiba

Hajmoosa

Mahmoudi

, et al. Short-term durability of GFRP stirrups under wet-dry and freeze–thaw cycles. Constr Build Mater 2023; 398: 132533. https://doi.org/10.1016/j.conbuildmat.2023.132533

48.

ASTM . ASTM D7792/D7792M-15 - standard practice for Freeze/Thaw conditioning of pultruded Fiber Reinforced Polymer (FRP) composites used in structural designs. ASTM International, 2015.

49.

Karbhari

Rivera

Zhang

. Low‐temperature hygrothermal degradation of ambient cured E‐glass/vinylester composites. J Appl Polym Sci 2002; 86(9): 2255–2260. https://doi.org/10.1002/app.11205

50.

Kim

Park

You

, et al. Short-term durability test for GFRP rods under various environmental conditions. Compos Struct 2008; 83(1): 37–47. https://doi.org/10.1016/j.compstruct.2007.03.005

51.

Xian

Lin

, et al. Freeze–thaw resistance of unidirectional‐fiber‐reinforced epoxy composites. J Appl Polym Sci 2012; 123(6): 3781–3788. https://doi.org/10.1002/app.34870

52.

Di Ludovico

Piscitelli

Prota

, et al. Improved mechanical properties of CFRP laminates at elevated temperatures and freeze–thaw cycling. Constr Build Mater 2012; 31: 273–283. https://doi.org/10.1016/j.conbuildmat.2011.12.105

53.

Shahid

. Durability and sustainability of fibre-reinforced bio-based polymer composite materials. PhD Thesis in Civil Engineering. Instituto Superior Técnico, University of Lisbon, 2026.

54.

ISO . ISO 16620-4:2024 - Plastics - biobased content - part 4: determination of biobased mass content. International Organization for Standardization, 2024.

55.

ISO . ISO 527-2:2012 - plastics - determination of tensile properties - part 2: test conditions for moulding and extrusion plastics. International Organization for Standardization, 2012.

56.

ASTM . ASTM D5379/D5379M-19 - standard test method for shear properties of composite materials by the V-notched beam method. ASTM International, 2021.

57.

ISO . ISO 6721-11:2019 - Plastics - Determination of dynamic mechanical properties - part 11: glass transition temperature. International Organization for Standardization, 2019.

58.

ISO . ISO 62:2008 – plastics – determination of water absorption. International Organization for Standardization, 2008.

59.

ASTM . ASTM D5229/D5229M-20 - standard test method for water absorption properties and equilibrium conditioning of polymer matrix composite materials. ASTM International, 2020.

60.

ASTM . ASTM D570-22 - standard test method for water absorption of plastics. ASTM International, 2022.

61.

ISO . ISO 527-4:2023 - plastics - determination of tensile properties - part 4: test conditions for isotropic and orthotropic fibre-reinforced plastic composites. International Organization for Standardization, 2023.

62.

ISO . ISO 14126:2023 - Fibre-reinforced plastic composites - determination of compressive properties in the in-plane direction. International Organization for Standardization, 2023.

63.

ISO . ISO 14130:1997 - Fibre-reinforced plastic composites - determination of apparent interlaminar shear strength by short-beam method. International Organization for Standardization, 1997.

64.

Fonseca

Lopes

Coelho

, et al. Synthesis of unsaturated polyesters based on renewable monomers: structure/properties relationship and crosslinking with 2-hydroxyethyl methacrylate. React Funct Polym 2015; 97: 1–11. https://doi.org/10.1016/j.reactfunctpolym.2015.10.002

65.

Sousa

Garrido

Correia

, et al. Hygrothermal ageing of pultruded GFRP profiles: comparative study of unsaturated polyester and vinyl ester resin matrices. Compos Appl Sci Manuf 2021; 140: 106193. https://doi.org/10.1016/j.compositesa.2020.106193