Parameter screening, optimization and artificial neural network modeling of cadmium extraction from aqueous solution using green emulsion liquid membrane

Abstract

This research was aimed to investigate the extraction of cadmium using waste cooking oil (WCO) based emulsion liquid membrane. The green emulsion liquid membrane (GELM) phase consists of waste cooking oil, a non-toxic solvent as diluent, Span 80 as surfactant, D2EHPA (Bis (2-ethylhexyl) phosphoric acid) as carrier, and hydrochloric acid solution as stripping phase. The influence of the operating parameters namely surfactant concentration (1–5v/v%), carrier concentration (2–6v/v%), agitation speed (200–600 rpm), agitation time (10–15​ min), pH of the feed solution (4–6), treat ratio (1:10–1:15), internal stripping agent (HCl) concentration (0.5–1.5 N), initial cadmium ion concentration (100–500 mg/L) and phase ratio (1:1–1:3) were studied to identify the key variables to be screened using Placket Burman design based on ‘P’ value. The significant variables were then optimized using Box–Behnken design. The stripping efficiency of cadmium was verified by varying the HCl concentration. The recycling and reuse of membrane phase was studied for 8 cycles. The mechanism of cadmium extraction by GELM was investigated. The feed forward neural network (FFNN) model having input layer composed of 5 neurons and output layer with 1 (cadmium extraction) neuron was employed to model the extraction data.