Abstract: Continuous degradation of the shear properties of joint rock masses caused by aqueous solution erosion is a critical factor affecting slope stability. Therefore, we proposed a dynamic dissolution testing method based on gas-liquid circulation under gas-liquid-solid three-phase coupling conditions. Dynamic dissolution tests and direct shear tests were conducted on joint samples in CO₂ solution environment. The deterioration law of the shear mechanical parameters of the joint samples was characterized. Meanwhile, by combining three-dimensional morphology and microstructure scanning technology, the deterioration mechanism of the joint samples under the dynamic dissolution effect of CO₂ solution was revealed. Results show that the shear-displacement curves of the joint samples can be divided into three stages: initial locking, intermediate failure, and late-stage shear-friction-resistance sliding. As the number of dissolution cycles increased, the shear hardening characteristics and stress levels of the samples decreased. After 30 dynamic dissolution cycles, the internal friction angle and cohesion decreased by 37.78% and 29.73%, respectively. Concurrently, progressive microstructural damage and pore development reduced the joint surface roughness and the compressive strength of joint rock masses, weakened frictional interlocking between joint surfaces, and thereby degraded shear mechanical performance. Finally, a numerical stability model incorporating dissolution-induced degradation of shear parameters was established. Analyses indicate that the decline in the safety factor of jointed slopes is primarily governed by the deterioration of joint shear parameters. Owing to spatial variations in stress states, the potential slip path dynamically migrates from shallow to deeper joints. The methods and findings provide a theoretical basis for long-term stability assessment of joint slopes.

Key words: rock slope, joint surface, dynamic dissolution, shear mechanical characteristics, deterioration mechanism, time-varying stability