Dynamic Characteristics of Basalt Fiber–Reinforced Pisha Sandstone Cement Soil

Abstract Pisha sandstone is a weakly cemented, erosion-prone sedimentary rock widely distributed in arid and semiarid regions such as Inner Mongolia, China. Its poor mechanical properties necessitate improvement for engineering applications. To study the effect of basalt fiber (BF) dosage on Pisha sandstone cement-soil (PSC), the dynamic characteristics of basalt fiber Pisha sandstone cement-soil (BF-PSC) with 7% cement and varying BF dosages (0%, 0.15%, 0.30%, 0.45%, and 0.60%) were analyzed. Graded cyclic loading vibration triaxial tests were used to investigate the hysteresis curve area, dynamic elastic modulus (DEM), and damping ratio. Scanning electron microscopy test (SEM), ultradepth-of-field optical microscopy, and nuclear magnetic resonance (NMR) tests were performed to analyze the BF strengthening and toughening mechanism on PSC. Gray theory was used to explore the relationship between BF-PSC microstructure and its max DEM, leading to the establishment of a gray theory prediction model for max DEM. Results showed that appropriate BF dosages (0.15%–0.30%) improved PSC unconfined compressive strength (UCS). DEM and damping ratio of BF-PSC varied abruptly with stress amplitude, DEM decreased initially then stabilized with increasing vibration times, while damping ratio decreased rapidly then increased slowly. Under the 4th and 5th cyclic loading, DEM increased then declined with BF dosage, and increased with confining pressure (30–90 kPa). Damping ratio increased with BF dosage and decreased with confining pressure. Higher BF dosages led to larger hysteresis curve areas and greater energy dissipation capacity. Microscopically, BF enhances PSC’s mechanical properties through compatibility with cement hydration products and friction reinforcement, but excessive BF reduces PSC spatial structure continuity. Gray theory analysis found high correlation between bound fluid saturation, small pore proportion, and max DEM (values: 0.721–0.834 and 0.763–0.780, respectively). A microstructure-based max DEM prediction model was created with an average relative error less of than 8% between estimated and experimental values. These findings provide a basis for further application of Pisha sandstone in engineering.