Reactive separation of malic acid from aqueous solutions and modeling by artificial neural network (ANN) and response surface methodology (RSM)

Malic acid is a commercially valuable chemical which used in food, pharmaceutical and personal care products, and its biotechnological production is of great attention. The biotechnological manufacture of carboxylic acids by fermentation processes produces multi-constituent aqueous solutions with the acid concentration less than 10% (w/w). Therefore, carboxylic acid separation from the fermentation environment or waste stream is economically important. In this study, the reactive separation of malic acid with tributylamine (TBA) in octyl acetate was carried out. The effects of initial malic acid concentration, initial amine concentration in organic phase, and phase ratio (organic phase volume/aqueous phase volume) on the separation were determined experimentally. Experimental results were analyzed, optimized and modeled by both artificial neural network (ANN) and response surface methodology (RSM). It was obtained that ANN had a slightly better coefficient of determination. The modeling results showed that TBA concentration in the organic phase among the parameters examined was the most effective parameter for the reactive separation of malic acid from aqueous solutions.