Introduction

Reinforced concrete structures are widely used in water-related engineering projects. However, long-term exposure to aquatic environments frequently leads to corrosion of the embedded steel reinforcement, which in turn induces concrete cover deterioration and spalling1,2,3. This corrosion leads to structural damage, significantly impacting durability and resulting in considerable economic loss and safety risks. To address this issue, engineers frequently apply anti-corrosion measures to rebars, concrete, or structures, including epoxy coating or galvanization of rebars, adding corrosion inhibitors to concrete, or applying protective coatings to the water-facing surfaces of structures4,5. However, these methods often suffer from drawbacks such as complexity, poor effectiveness, and high costs. To fundamentally resolve corrosion issues in wet environments, researchers are increasingly exploring the use of corrosion-resistant reinforcements and concrete to replace traditional rebar and ordinary concrete6,7.

Glass fiber reinforced polymer (GFRP) bars, composed of continuous fibers for load-bearing and resin as the matrix, provide benefits including corrosion resistance, high tensile strength, lightweight characteristics, insulation, and electromagnetic wave transparency8,9,10. Polymer concrete (PC) offers advantages over conventional concrete, including increased strength, reduced curing time, and improved corrosion resistance, leading to its widespread use in engineering projects11,12. There are various types of resins available, among which unsaturated polyester resin is particularly cost-effective, especially the unsaturated polyester resin concrete (UPC)13. Although FRP bars exhibit excellent durability, their mechanical properties and bond strength with concrete can deteriorate over time when exposed to aqueous environments in combination with ordinary concrete. This deterioration is primarily caused by the alkaline environment of the pore solution in ordinary concrete, which negatively impacts the performance of FRP bars14,15,16.Therefore, in water-related projects, the use of FRP bars and UPC as substitutes for steel reinforcement and conventional concrete, respectively, represents a viable and effective solution.

Researchers have extensively analyzed the bond performance of fiber reinforced polymer (FRP) bars with various concrete types, including ordinary17, recycled18,19,20,21, lightweight self-compacting22, air-entrained23, coral24,25, ultra-high performance26,27,28, and seawater sea-sand concrete29,30. On this basis, constitutive models describing the bond relationship between FRP bars and ordinary concrete have been established31. The bond behavior between FRP bars and concrete in humid aquatic environments requires considerable attention when using FRP-reinforced concrete structures32,33,34,35,36. For instance, Lu et al.37 investigated the bond performance between basalt fiber reinforced polymer (BFRP) bars and concrete with fly ash, concluding that the bond strength retention was about 48.0% after 50 years of seawater immersion. Saqib et al.38 examined the bond performance of BFRP bars with high-strength concrete in erosive environments by submerging pull-out specimens in alkaline and seawater solutions for three months. The study found that bond strength decreased post-immersion, with retention rates of 86% in alkaline solution and 82% in seawater. Altalmas et al.39 found that after 90 days of water exposure, the bond strength decreased by 25% for BFRP bars and 17% for GFRP bars when embedded in concrete. Discrepancies in bond durability results among different studies may be attributed to variations in bar properties used in each investigation. Dong et al.40 projected that over a 50-year design service life, the bond strength retention between BFRP bars and seawater sea-sand concrete would vary from 47 to 83% under various environmental conditions. Additionally, Belarbi et al.Research by41 demonstrated that environmental conditions, including freeze–thaw cycles, elevated temperatures (60 °C), and deicing salt solutions, notably diminished the bond strength between FRP bars and concrete.

Although FRP bars and different concrete types have been extensively researched, investigations into the bond performance of FRP bars with UPC are still scarce. Prior studies on the bond performance of steel rebars with UPC have shown notably greater bond strength than that of conventional concrete. For instance, Orsolya et al.42 found that polymer concrete significantly enhances the bond strength with steel bars compared to conventional concrete. Smooth steel bars in polymer concrete exhibit over ten times the bond strength of those in ordinary concrete, while ribbed steel bars need only a 40-mm bonding length. Douba et al.43 proposed that the incorporation of aluminum nanoparticles can enhance the bond strength between polymer concrete and steel by influencing the epoxy resin curing process. Li et al.44 examined the influence of bar diameter, type, surface morphology, and concrete cover thickness on the bond performance between FRP bars and UPC.

This study examines the bonding characteristics between GFRP bars and UPC under different aging conditions (25℃, 40℃, and 60℃) in a water environment. GFRP bars and UPC center drawing specimens were subjected to immersion in an aging water environment for different durations to analyze their bonding properties.

Experimental overview

Experimental materials

The raw materials and mix proportions for the UPC in this experiment matched those detailed in the authors’ prior study44. The GFRP bars are composed of vinyl resin-based reinforcement manufactured by Shandong Sford Industrial Co., LTD. The reinforcement has a diameter of 10 mm and exhibits a measured tensile strength of 979.76 MPa, along with an elastic modulus of 55.54GPa. Additionally, the surface of the reinforcement is ribbed with ribs spaced at intervals of 9 mm, featuring a rib width of 1 mm and a depth of 0.5 mm. The specific reinforcement utilized is illustrated in Fig. 1.

Fig. 1
figure 1

Photo and diagram of GFRP bars for test.

Full size image

Specimen preparation

The specimens, intended for aging in a constant-temperature water bath, were sized at 100 mm × 100 mm × 100 mm to fit the experimental equipment, as illustrated in Fig. 2. To ensure precise analysis of bond strength degradation between GFRP bars and UPC, the specimen ends were fully encased in PVC pipes and sealed with hot-melt adhesive to prevent water ingress, which could influence the results. The bond length was established as 50 mm, equivalent to five times the bar diameter. Thirty pull-out specimens were prepared.

Fig. 2
figure 2

GFRP bar-UPC drawing test specimen.

Full size image

Aging conditions

This experiment utilized a constant-temperature water bath to simulate the aging environment, as shown in Fig. 3. Tap water was added to the water bath box, and temperatures were adjusted to 25 °C, 40 °C, and 60 °C. The specimens were immersed in the water, and pull-out tests were conducted at aging durations of 60, 120, and 180 days, comparing the results with those of unaged control specimens. During aging, the water bath box was sealed with plastic film to minimize water vapor and temperature loss.

Fig. 3
figure 3

Aging environment of the specimen.

Full size image

Test procedure

The pull-out tests utilized a WAW-1000D electro-hydraulic servo universal testing machine. Two displacement meters, shown in Fig. 4, were used to measure the relative slip between the GFRP bars and UPC. Following preparation, tests were performed at a loading rate of 1 mm/min, as outlined in CSA S807 19.

Fig. 4
figure 4

Special counterforce frame.

Full size image

Results and analysis

The failure mode observed in the central tensile pull-out test for GFRP bars with UPC was characterized by reinforcement dislodgement.

In this study, it is assumed that the adhesive stress is uniformly distributed throughout the bonding section. Therefore, the bond strength was calculated using formula 1 and the corresponding experimental results are presented in Table 1. The nomenclature for specimens in the table follows a format of temperature combined with aging period, such as T25-60d indicating an aging duration of 60 days at a temperature of 25℃ in a water bath. The control group represents specimens without undergoing accelerated aging.

$$\tau = \frac{P}{\pi DL}$$
(1)
Table 1 Summary of GFRP bar-UPC pull-out bond test results.
Full size table

The bond strength (MPa) is τ, P determined by the load magnitude (N), D is the bar diameter (mm), and L is the bond length (mm).

Effect of aging temperature on bond strength

The bond strength retention rate between GFRP bars and UPC at different temperatures is illustrated in Fig. 5. The graph indicates that bond strength consistently diminishes with rising temperatures. Following a 60-day aging period, the specimens demonstrated bond strength retention rates of 102.3%, 100.4%, and 98.6% at temperatures of 25℃, 40℃, and 60℃ respectively compared to non-aged specimens. The aged specimens demonstrated greater bond strength compared to the non-aged ones. The increased bond strength between GFRP bars and UPC is due to the greater expansion deformation of GFRP bars compared to UPC during temperature changes and water absorption in certain aging conditions, leading to a tighter bond. Similar experimental findings were also reported by Mohamed et al.45. After 120 days of aging in water at 25℃, 40℃, and 60℃, the bond strength retention rates are approximately 98.1%, 96.3%, and 94.7%, respectively. Subsequently, after an aging period of 180 days under similar conditions, the retention rates decrease to values around 96.4%, 94.2%, and 92.2% for temperatures of 25℃,40℃, and 60℃ respectively. As temperature rises, the softening and degradation of the resin in both GFRP bars and UPC are accelerated, resulting in reduced strength and stiffness. This deterioration negatively impacts the interfacial bond between the two materials, thereby further weakening the overall bond strength.

Fig. 5
figure 5

The retention rates of bond strength for GFRP bars and UPC across various temperatures.

Full size image

Effect of aging duration on bond strength

Figure 6 illustrates the variation in bond strength retention rate of GFRP bar -UPC over different aging periods. The graph demonstrates a slight increase in bond strength within an aging period of 0 to 60 days when exposed to temperatures of both 25℃ and 40℃, while under other aging conditions, the bond strength gradually decreases with increasing duration. At 25℃, the bond strength retention rates between GFRP bars and UPC after 60, 120, and 180 days are approximately 102.3%, 98.1%, and 96.4% of the unaged specimens, respectively. In a water environment with an aging temperature of 40℃, the retention rates of bond strength after 60d, 120d, and 180d are recorded as 100.4%, 96.3%, and 94.2% respectively. In a 60℃ water environment, the bond strength between GFRP bars and UPC progressively decreased as aging time increased. The retention rates of bond strength after accelerated aging for 60d, 120d, and 180d are determined to be 98.6%, 94.7%, and 92.2% respectively due to progressive moisture accumulation in specimens over time resulting in increased erosion on both materials and their interface leading to diminished bond strength.

Fig. 6
figure 6

Effect of aging age on bond strength retention of GFRP bars and UPC.

Full size image

Effect of humid-heat aging on bond-slip behaviour

Figure 7 illustrates the bond-slip curves for representative specimens subjected to different aging environments. The bond-slip curves maintain their overall shape before and after aging, displaying three distinct stages: ascending, descending, and residual. During the initial loading stage, the bond stress increases rapidly while slip remains minimal. With increasing pull-out force, the bond-slip curve’s slope progressively diminishes. Upon attaining its maximum value, the bond-slip curve begins to decline during the pre-peak stage, referred to as the ascending segment. In this descending segment, there is a gradual decrease in bond stress accompanied by rapid relative slip incrementation. When reaching a certain extent of descent, slight fluctuations occur within a narrow range of bond stress values resulting in a sinusoidal decay shape resembling that of a sine wave-referred to as residual segment.

Fig. 7
figure 7

Bond slip curves of GFRP bars and UPC under different aging environments.

Full size image

Figure 7 shows that aging leads to a decrease in both the relative slip associated with bond strength and the residual bond stress. To analyze these changes, the average relative slip values at peak bond strength and the average residual bond stresses for each test group are plotted in Figs. 8 and 9, respectively.

Fig. 8
figure 8

Average relative slip of bond strength under different aging conditions (mm).

Full size image
Fig. 9
figure 9

Adhesive residual stress in different aging environments.

Full size image

From Fig. 8, it is evident that the relative slip values at peak bond strength show considerable variation but generally exhibit a decline post-aging compared to unaged specimens. After 180 days of aging at 25 °C, 40 °C, and 60 °C, the average relative slip values decreased to 1.38 mm, 1.34 mm, and 1.21 mm, respectively, compared to the unaged specimen value of 1.62 mm. These findings are consistent with those reported by Ahmad et al.39 and Alaa et al.33. The aging-induced decrease in ductility of GFRP bars and UPC results in specimen failure at lower relative slip values.

Figure 9 demonstrates that aged specimens exhibit lower residual bond stress than unaged specimens. The erosive effects on the surface of GFRP bars and the GFRP-UPC interface during aging reduce frictional resistance, thereby decreasing residual bond stress.

Bond-slip constitutive model of GFRP bars-UPC interface after aging

Based on extensive experimental research and theoretical analysis, numerous scholars have developed bond constitutive relationships between reinforcing bars and conventional concrete. Prominent examples encompass the BPE model46, Malvar model47, MBPE model48, CMR model49, and continuous curve model50. The BPE model is designed to characterize the bond behavior between steel bars and concrete, but it is not applicable to FRP bars. The Malvar model, characterized by a higher number of fitting parameters and a more intricate formulation, is less frequently utilized. The CMR model fails to adequately address the constitutive relationship in the descending and residual sections, which restricts its practicality. The MBPE and continuous curve models are more commonly used to characterize the bond constitutive relationship between FRP bars and concrete. Key features of these two models are summarized below.

1. MBPE model

Cosenza et al.48 modified the BPE model and introduced a curve model that accurately represents the bond-slip characteristics of FRP bars and concrete. This model consists of three distinct sections: the ascending section, the descending section, and the residual section, as illustrated in Fig. 10. The mathematical expression for this model is as follows:

$${\text{Ascending section }},\,\,\, \frac{\uptau }{{\uptau _{1} }} = \left( {\frac{s}{{s_{1} }}} \right)^{\upalpha } \quad s \le s_{1}$$
$${\text{Descending section }},\,\,\, \frac{\uptau }{{\uptau _{1} }} = 1 - p(s/s_{1} - 1) \quad s_{1} \le s \le s_{3}$$
(2)
$${\text{Residual section}}, \,\,\, \uptau =\uptau _{3} \quad s \ge s_{3}$$
Fig. 10
figure 10

MBPE model.

Full size image

Here, τ and \(s\) denote the bond stress (MPa) and relative slip (mm), respectively; τ 1 and s1 signify the bond strength (MPa) and its corresponding slip (mm); τ 3 and s3 represent the residual section stress (MPa) and its associated slip (mm); α and p are test fitting parameters, determined by equating the areas under the ascending and descending sections of both the test and theoretical curves.

The MBPE model is notable for its straightforward design and limited fitting parameters. Despite the linear depiction of the model’s descending and residual segments not aligning with the test curve, its simplicity has enabled widespread use.

2. Continuous curve model

Gao et al.50 introduced a continuous curve model to overcome the limitations of discontinuous curves and inadequate fit with experimental data in the constitutive model of FRP bar and concrete. The model, depicted in Fig. 11, is defined by the following equation.

Fig. 11
figure 11

Continuous curved model.

Full size image

Ascending section,

$$\frac{\tau }{{\tau_{u} }} = 2\sqrt {\frac{s}{{s_{u} }}} - \frac{s}{{s_{u} }} \quad s \le s_{u}$$

Descending section ,

$$\tau = \tau_{u} \frac{{(s_{r} - s)^{2} (2s + s_{r} - 3s_{u} )}}{{(s_{r} - s_{u} )^{3} }} + \tau_{r} \frac{{(s - s_{u} )^{2} (3s_{r} - 2s - s_{u} )}}{{(s_{r} - s_{u} )^{3} }} \quad s_{u} \le s \le s_{r}$$
(3)

Residual section ,

$$\tau = \tau_{r} \quad s { \ge }s_{r}$$

There, τu is the bond strength (MPa); τr is the residual bond stress (MPa); su is the slip value corresponding to the bond strength (mm); sr represents the slip when the residual bond strength is just reached (mm).

The continuous curve model exhibits the following characteristics: a) It contains no fitting parameters; b) The constitutive relationship curve is continuous; c) The slope at the initial point of the ascending segment approaches infinity; d) The descending segment is a curve that closely aligns with the experimental curve, although the expression is complex; e) The residual segment is a horizontal straight line, which does not match the experimental curve.

The bond-slip relationship between FRP bars and ultra-performance concrete is typically developed by refining existing models using experimental data. To date, there has been a notable absence of comprehensive studies on the bond-slip constitutive relationship of FRP bars-UPC. As depicted in Fig. 7, the bond-slip curve of GFRP bars-UPC exhibits minimal changes before and after aging. This study evaluates the MBPE and continuous curve models against experimental curves obtained after aging at three different temperatures for 180 days, to determine their effectiveness in characterizing the bond-slip behavior of GFRP bars-UPC post-aging. The comparison of the curves generated by both models with the experimental data is illustrated in Fig. 12.

Fig. 12
figure 12

Presents a comparison between the theoretical and experimental curves.

Full size image

Figure 12 shows that the ascending segments of the test curve align closely with those of the two model curves. However, the linear descending segment of the MBPE model exhibits a marginally lower degree of congruence with the test curve when compared to the continuous curve model. Nevertheless, the simplicity of the MBPE model facilitates the derivation of an analytical solution. The residual section reveals notable differences between the curves of the two models and the experimental curves.

Conclusions

This study conducted pull-out tests on GFRP bar-UPC specimens after aging treatment. The primary research findings are summarized as follows:

1. Increasing the aging temperature accelerates the reduction in strength of GFRP bars and UPC, which directly impacts their interface bonding and subsequently decreases bond strength. Aged specimens exhibit greater bond strength compared to unaged ones. This is primarily due to the fact that, in certain aging environments, GFRP bars exhibit greater expansion deformation compared to UPC under conditions of temperature changes and water molecule absorption, leading to a tighter bond and thus higher bond strength.

2. The bond strength generally decreases with the increase in aging duration (with the exception of specimens aged at 25 °C and 40 °C for up to 60 days). This phenomenon results from the gradual water accumulation in the specimens over time, which enhances the erosion of the materials and their interface. As a result, the mechanical interlocking between the two materials is weakened, leading to decreased bond strength.

3. Prior to and following the aging process, the bond-slip curve of GFRP bars-UPC demonstrated minimal alterations, preserving its distinct ascending, descending, and residual segments. The MBPE model and the continuous curve model accurately represent both the ascending and descending segments of the bond-slip curve for aged GFRP bars-UPC.

4. After aging, the bond strength slip values between GFRP bars and UPC show greater variability, with aged specimens typically having lower values than unaged ones. This phenomenon is due to the reduced plasticity of GFRP bars and UPC following aging. Consequently, the specimens are more prone to damage with a smaller relative slip. The residual bond stress in most aged specimens decreases due to surface erosion of the GFRP bars and at their interface with UPC, resulting in diminished frictional forces.