The construction of transportation infrastructure in plateau regions, such as the Sichuan-Tibet Railway and the Xinjiang-Tibet Railway, China, is progressing rapidly under the national development strategies. The annual minimum temperature in Tibet can reach -45 ℃, and the number of days with alternating positive and negative temperatures accounts for more than half of the year, forming an extreme low-temperature, high-frequency freeze-thaw environment. Furthermore, structures such as bridges, culverts, and tunnels in plateau areas are often subjected to erosion from de-icing salts or chlorides in the soil. The combined action of freeze-thaw and chloride attack poses a serious threat to the durability of concrete structures. Fiber-reinforced polymer composites, such as carbon fiber reinforced polymer (CFRP), basalt fiber reinforced polymer (BFRP), and glass fiber reinforced polymer (GFRP) bars, are widely used in corrosive environments like coastal and saline areas due to their high strength and corrosion resistance. The bond performance of the steel/FRP bar-concrete interface significantly affects the mechanical behavior of concrete structures. However, the existing research mostly focuses on the room-temperature or conventional freeze-thaw environments, and the understanding of the degradation mechanism of the bar-concrete bond performance under the combined action of low-temperature freeze-thaw and chloride attack specific to plateau regions remains insufficient. It is thus necessary to conduct a targeted research on the bond performance of the steel/FRP bar-concrete interface under low-temperature freeze-thaw and chloride attack environments in the Tibetan plateau.

A total of 144 pull-out specimens were fabricated. The test parameters included bar type (i.e., steel, BFRP, CFRP, and GFRP bars), bar diameter (i.e., 10, 16 mm, and 20 mm), and the number of freeze-thaw cycles (i.e., 0, 50, 100, and 150). After 28 d standard curing, the specimens were first subjected to a 4 d water saturation pretreatment to ensure that they reached a saturated state before the freeze-thaw cycle test. A programmable high-low temperature test chamber was used to simulate a freeze-thaw cycling process. Each freeze-thaw cycle lasted for 260 min (i.e., consisting of 180 min of freezing and 80 min of thawing). After every 5 freeze-thaw cycles, the specimens were placed at 20 ℃ and sprayed with a NaCl solution to maintain a stable moisture content. Following the designated freeze-thaw cycles, the specimens were immersed in a 5% NaCl solution for 15 d to accelerate chloride ingress. Finally, the central pull-out tests were conducted by a 300 kN MTS testing machine at a loading rate of 0.5 mm/min, with the free-end slip measured by a displacement transducer at a data acquisition frequency of 10 Hz.

The failure modes of the specimens are primarily governed by bar type and diameter, with freeze-thaw cycles showing negligible influence. All the steel bar specimens exhibit a pull-out failure accompanied by longitudinal cracks, with crack width increasing alongside bar diameter. The GFRP bar specimens also fail due to pull-out without concrete cracking. In contrast, for the BFRP and CFRP bars, the failure mode transitions from pull-out to splitting failure when the diameter increases from 10 mm to 16 mm. The bond strength decreases with increasing bar diameter for all bar types. When the diameter increases from 10 mm to 20 mm, the bond strength reduction for steel bars (i.e., 51.7%-56.3%) is significantly greater than that for FRP bars (i.e., below 18.2%). After 150 freeze-thaw cycles, the bond strength of steel bars is decreased by more than 31.4%, while the reduction for BFRP bars does not exceed 22.1%. The degradation mechanisms differ between steel and FRP bars. For steel bars, the bond degradation is mainly attributed to concrete deterioration and steel corrosion induced by the coupled environment. For FRP bars, in addition to concrete deterioration, the degradation involves resin matrix damage from freeze-thaw cycles and the formation of interfacial gaps due to the mismatch in thermal expansion coefficients between FRP and concrete. For peak slip, steel bar specimens exhibit the values between 1.6 mm and 2.5 mm, which are considerably lower than those of FRP bar specimens (i.e., ranging from 2.7 mm to 4.1 mm). As the diameter increases from 10 mm to 20 mm, the peak slip is increased by 5.0%-13.3% for all the specimens. Conversely, as freeze-thaw cycles increase from 0 to 150, the peak slip is decreased by 13.5%-23.8%. Note that no significant interaction effect between bar diameter and freeze-thaw cycles on peak slip occurs. Based on the experimental data, the prediction models for bond strength and peak slip are proposed through normalization, incorporating the effects of freeze-thaw cycles and bar diameter. The data by the models are in reasonable agreement with the experimental results, with coefficients of determination (R²) of exceeding 0.86.

This study investigated the bond performance of steel and FRP bars with concrete under low-temperature freeze-thaw and chloride attack. The failure mode was primarily determined by bar type and diameter. The bond strength decreased with increasing diameter, and steel bars had a more pronounced reduction than FRP bars. The freeze-thaw cycles significantly reduced the bond strength for all bar types, the underlying degradation mechanisms differed between steel and FRP bars. The peak slip increased with diameter but decreased with more freeze-thaw cycles, with no significant interaction between these two factors. The proposed prediction models, which could account for freeze-thaw cycles and normalized bar diameter, effectively captured the degradation of bond strength and peak slip, providing a valuable tool for durability design and assessment of concrete structures in harsh plateau environments.