Abstract

Many industrial processes, such as mining, metal plating, or pigment and battery manufacturing, result in the release of heavy metals to aquatic ecosystems. Heavy metals are toxic pollutants, which can accumulate in living tissues causing various diseases and disorders. Removal of toxic contaminants from wastewaters is one of the most important environmental issues. Since all heavy metals are non-biodegradable, they must be removed from the polluted streams for the environmental quality standards to be met. Development of efficient and low-cost separation processes is therefore of utmost importance. Biosorption is an efficient method that can be used for removal of heavy metals [1]. Recent studies showed that common agricultural waste products or natural polymers can be used as potential biosorbents for the removal of heavy metals [2]. Biosorption can be used for the treatment of wastewater with low heavy metal concentration as an inexpensive, simple and effective alternative to conventional methods. The present study used the highly available and cost effective waste biomass of cotton residue for the removal of chromium ions from wastewater solutions. The main objective of this study was to obtain a model on the basis of batch adsorption experiments performed with four different process variables including pH, initial concentration of adsorbate, contact time, and dose of adsorbent that could make reliable prediction of adsorption quantity of Cr(III) in wastewater using the cotton residue. In this way, wavelet neural network and support vector machine were used for model building. These models were validated using the leave one out crossvalidation method. Performance of the selected models was evaluated using criteria such as RMSEC, RMSECV, R2 for calibration, R2 for cross-validation, and RMSE of prediction. The results showed that support vector machine predicted adsorption quantity of Cr(III) satisfactorily and the model can be used as a powerful tool for modeling of Cr(III) removal using cotton residues. Additionally, the influence of several parameters such as initial heavy metal concentration, pH of the solution, contact time and dose of the adsorbent on the adsorption capacity of the cotton residue was also investigated. The results demonstrated that initial metal concentration and pH of the solution are two most important parameters in removal of Cr(III) ions from wastewater solutions using cotton residues.