Remediation of mine polluted soil with nano-enhanced materials: Development of extreme learning machine approaches

Abstract

Successful mining soil reclamation promotes ecosystem recovery, reduces negative environmental effects, adds more area for forestry or agricultural purposes, and increases carbon (C) sequestration. In order to increase soil erosion management, improve soil quality, reduce pollutants, and assure safe land application of traditional amendment materials, nanoparticles with exceptionally high deliverability and reactivity may be used as amendments. Iron oxide is being researched for the remediation of industrial soil that is co-contaminated with arsenic (AS) due to the absence of Nano enhanced materials for mine soil reclamation. In order to improve the soil quality of a mine waste that was heavily polluted with As (1807 mg/kg), the effects of iron oxide on the non-specifically and specifically-sorbed As were investigated. Iron oxide was added to the polluted soil at concentrations of 0.5 percent, 2 percent, and 5 percent (w/w). The goal of this work is to define the effect of iron oxide and Zero Valent I nanoparticles (nZVI) in reducing the contamination of soil by the use of soft computing models of extreme learning machines (ELM) with particle swarm optimization (PSO). In this case, the hybrid ELM-PSO has shown good performance as a trustworthy approach after the regression study of RMSE, R-square, and r. The addition of iron oxide dosages decreased the easily accessible As by 92.4 when compared to the untreated soil, with the 5 percent doses having a noticeably greater effect.

Keywords: Arsenic; Artificial intelligence; ELM-PSO-GA; Iron oxide; Mine; Nano-enhanced material; Soil.