Parametric optimization of hexavalent chromium removal by electrocoagulation technology with vertical rotating cylindrical aluminum electrodes using Taguchi and ANN model

• Published in: Journal of environmental health science and engineering Vol. 21; no. 1; pp. 255 - 275
• Main Authors: Kumar, Amit, Basu, D.
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.06.2023; BioMed Central Ltd; Nature Publishing Group
• Subjects: Aluminum; Analysis; Artificial neural networks; Cadmium; Chromium; Earth and Environmental Science; Electrocoagulation; Electrodes; Energy demand; Environment; Environmental Economics; Environmental Engineering/Biotechnology; Environmental Health; Environmental Law/Policy/Ecojustice; Error functions; Hexavalent chromium; Hydroxides; Neural networks; Optimization; Quality of Life Research; Research Article; Root-mean-square errors; Rotation; Sludge; Tanning; Textile industry wastewaters; Waste Management/Waste Technology; Working conditions; India; Hexavalent chromium; ANN; Electrocoagulation; Vertical rotating electrodes; Taguchi method
• ISSN: 2052-336X; 2052-336X
• DOI: 10.1007/s40201-023-00859-w

This study aims to evaluate the performance of rotating aluminum electrodes in the electrocoagulation reactor for removing hexavalent chromium (Cr 6+) from synthetic tannery wastewater. Taguchi and Artificial Neural Network (ANN) based models were developed to obtain the optimum condition for maximum Cr 6+ removal. The optimum working condition obtained by Taguchi approach for the maximum Cr 6+ removal (94%) was: Initial Cr 6+ concentration (Cr 6+ i ) = 15 mg/L; Current Density (CD) = 14.25 mA/cm 2 ; Intial pH = 5; Rotational Speed of Electrode (RSE) = 70 rpm. In contrast, the optimal condition for maximum Cr 6+ ions removal (98.83%) obtained from the BR-ANN model was: Cr 6+ i  = 15 mg/L; CD = 14.36 mA/cm 2 ; pHi = 5.2; RSE = 73 rpm. Compared to the Taguchi model, the BR-ANN model outperformed in terms of providing higher Cr 6+ removal (+ 4.83%); reduced energy demand (-0.035 KWh/gm Cr 6+ remove); lower error function value (χ 2  = -7.9674 and RMSE = -3.5414); and highest R 2 value (0.9991). The data for the conditions 91,007 < Re < 227,517 and Sc = 102.834 were found to fit the equation for the initial Cr 6+ concentration of 15 mg/l; Sh = 3.143Re 0.125 Sc 0.33 . The Cr 6+ removal kinetics was best described by Pseudo 2 nd Order model, as validated by high R 2 and lower error functions value. The SEM and XRF analysis confirmed that Cr 6+ was adsorbed and precipitated along with metal hydroxide sludge. The rotating electrode led to lower SEEC (10.25 kWh/m 3 ), as well as maximum Cr 6+ removal (98.83%), compared to EC process with stationary electrodes.