The study of diffusion characteristics of soil bentonite to control contaminant transport

Abstract

Purpose The purpose of this paper is to assess sand-bentonite liners (SBL) which could be used as hydraulic barriers with a controllable quality, relatively low cost and easy operation in solid waste landfills. Design/methodology/approach These barriers have been used successfully in various applications and have attracted much attention in a short period of time. The only precautionary use of SBLs is related to the change of their hydraulic properties in high alkaline chemical environments. The main reason for this phenomenon is the presence of high ion exchange minerals in bentonite. By exposure to these environments, it is also laid open to degradation of the montmorillonite microstructure leads to change in hydraulic behavior. Three different compounds were used for laboratory-scale SBL, and diffusion was considered as the dominant mechanism of contamination transmission in these liners. Chlorine ion has been used as pollutant, and its diffusion coefficient was determined in the tested SBLs. Findings The sample’s diffusion coefficient for the first experiment containing 3% bentonite and 97% Semnan sand were 2.5 × 10^(−9) (m^2/s) and 2.44 × 10 ^(−9) (m^2/s), respectively. Similarly, for two samples with 6% bentonite and 94% Semnan sand, this parameter was equal to 2.17 × 10 ^(−9) (m^2/s) and 2.22 × 10 ^(−9) (m^2/s) and for two samples with 3% agglacial clay, 12% bentonite and 85% Semnan sand was 5.55 × 10 ^(−10) (m^2/s) and 6.11 × 10 ^(−10) (m^2/s). These values correspond to the range reported in previous studies. Also, it was observed that with comparing the diffusion coefficients of test, it was concluded that with increasing bentonite, the molecular diffusion decreases significantly. Originality/value In this study, three laboratory samples with different percentages of bentonite, clay and sand were considered and the results obtained from the laboratory were compared with the results obtained from numerical modeling.