Search

Water-salt migration is an important reason for the development of basal erosion which seriously threatens the conservation of earthen sites. Studies in cutting-off or redirection of the water-salt migration path become vital. Polyvinyl alcohol (PVA) composite soil with extreme hydrophobicity shows the potential as the cushion layer with the damp-proof function, but the feasibility of preventing basal erosion based on the PVA cushion layer still lacks verification. In this study, for the embedded and non-embedded PVA cushion layers, water-salt distribution and morphology of earthen wall after repeated injection of salt solution were compared. The results showed that PVA cushion layer changed the water-salt migration path and promoted the accumulation of soluble salt on the ground outside the wall rather than the bottom of cushion layer, which effectively prevented basal erosion. A numerical model based on HYDRUS-2D confirmed that a partial laying of hydrophobic cushion layer under the wall cannot effectively prevent the water-salt migration, and a full laying of cushion layer should be adopted to prevent basal erosion. The cushion layer was recommended to extend towards the direction of the earthen wall instead of towards the sides, which also resisted the destructive effects of sand-driving wind and runoff erosion. • Basal erosion caused by water-salt migration is directly confirmed. • Water-salt distribution for models with/without PVA cushion layer is explored. • PVA cushion layer can inhibit basal erosion. • Reasonable laying modes of the hydrophobic cushion layer is presented. [ABSTRACT FROM AUTHOR]

• PVA-treated soil is confirmed to prevent water-salt capillary rise. • Effectiveness of 1.0% PVA-treated soil is best. • Interface effects are proposed to explain experimental phenomena. • Capillary rise of layered soil is described by modified Green-Ampt model. In the loess region of northwest China, the accumulation of soluble salt in the soil due to the capillary rise of water often damages earthen buildings. Thus, studies on the cutting off of the capillary-rise path are vital. As a hydrophobicity inducer, polyvinyl alcohol (PVA) solution with water as the solvent has the potential to function as a barrier. This study mainly aims to evaluate the feasibility of using PVA-treated soil as a waterproof layer to prevent water-salt capillary rise through soil column experiments, and the effects of the waterproof layer thickness, waterproof layer position, PVA content, ambient temperature, and type of salt solution on the capillary action. The results indicate that a 1.0% PVA-treated soil with extreme water repellency (i.e. dried at 20 °C) can perfectly prevent the water-salt from migrating to the top of the sample, regardless of the waterproof layer position, waterproof layer thickness, and type of soluble salt. For a 0.5% PVA-treated soil dried at 20 °C, the lower waterproof layer prevents the water-salt capillary rise, while the middle and upper layers exhibit less effectiveness. However, the PVA-treated soil that loses its hydrophobicity after a treatment at 100 °C cannot prevent the water-salt capillary rise, confirming that the prevention of capillary rise is due to the soil hydrophobicity. The waterproof layer thickness and type of soluble salt are less significant than the waterproof layer position, PVA content, and ambient temperature. This study proposes a modified Green-Ampt model to describe the capillary rise, and the hydraulic conductivity and pore size are obtained using the modified model. These parameters and the interface effect are used to explain the experimental results. Finally, the application of PVA-treated soil as a waterproof layer to inhibit water-salt capillary rise is illustrated. [ABSTRACT FROM AUTHOR]