Utilization of factorial design methodology to optimize Pr Red Hegxl dye uptake and prediction of removal efficiency via artificial neural network: comparison of linear vs non-linear sorption isotherm and kinetic parameters

Abstract

The evacuation of highly colored effluents in the ecosystem by textile industries generates extreme depredation to the environment and all living creatures. Sorption is preferred because of the wideness of biomass, and cost-efficient, minimized sludge in proportion to conventional treatment methods. A factorial experimental design and ANOVA techniques were utilized to examine the sorption of reactive Pr Red Hegxl dye by Daphne seed-based sorbents, and to optimize the operating conditions. The effects of main variables and their interaction effects on dye removal efficiency (%) were determined; pH was designated as significant at 95% confidence level for all types of sorbents. The maximum removal efficiencies (%) of Daphne seed and char were obtained as 70.8% and 83.2% when pH = 2, sorbent dosage = 0.4 g/50 mL, initial concentration = 50 mg/L, and temperature = 40 degrees C, respectively. 90.2% and 85.4% removal efficiencies were also attained for KOH- and K2CO3-activated carbons, respectively. Besides, artificial neural network models based on several back-propagation training algorithms and transfer functions were used to predict removal efficiency (%). The findings demonstrated that the proposed models had reasonable capabilities of predicting the removal efficiency (%). Daphne seed and carbonaceous products could be effectively used for the dye removal from aqueous solutions as affordable and abundant sorbent materials.