Research on the water-sensitivity characteristics and consolidation mechanism of biopolymer-modified dispersive soils

J X Liu, Yao Meng, Gang Li, Siwen Ren, Qinchen Zhu, J. Qin, Jingxuan Guan

Dispersive soil is a water-sensitive geomaterial predominantly found in arid and semi-arid regions. Characterized by low erosion resistance, it is prone to piping, cavity formation, and gully erosion, posing significant threats to infrastructure safety. This study investigates the modification of dispersive soil using guar gum (GG) and xanthan gum (XG). Through a comprehensive suite of tests, including identification, permeability, disintegration, X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), and scanning electron microscopy (SEM) analyses, the evolution of dispersion characteristics, water sensitivity, and microstructure was examined to elucidate the underlying modification mechanisms. Furthermore, a macro-micro correlation model was established by integrating macroscopic performance with microscopic parameters. The results demonstrate that the soil experiences a clear transition from a dispersive to a non-dispersive state, with 0.2% and 0.4% serving as the critical dosage thresholds that define these distinct behavioral boundaries for both XG and GG. The permeability coefficient ( k ) of the stabilized soil decreased progressively with increasing dry density, curing time, and modifier dosage. Notably, XG-modified soil exhibited significantly lower permeability compared to GG-modified soil. Regarding slaking behavior, the disintegration rate increased with duration but decreased with higher dry density and modifier dosages, while the time to complete disintegration was extended by the additives. Microstructural analysis revealed that the modifiers did not alter the fundamental crystal structure of the clay minerals. Instead, GG enhanced interparticle bonding via surface adsorption driven by its long-chain molecular structure. In contrast, XG acted primarily through cementation, filling pore voids with gel-like substances, thereby reducing porosity and permeability. Grey relational analysis (GRA) identified the correlation between the permeability coefficient and pore parameters in the following order: shape factor = pore number > fractal dimension > porosity, indicating a strong dependence on pore shape and quantity. Consequently, a prediction model for the permeability coefficient was constructed using shape factor and pore number as independent variables. This model demonstrated high accuracy, offering a reliable tool for evaluating the permeability of XG-modified dispersive soils. These findings establish a theoretical basis and practical guidance for the effective stabilization of dispersive soils.