OPTIMIZATION OF MICROWAVE ASSISTED H2SO4 HYDROLYSIS OF COCOA POD SHELLS: COMPARISON BETWEEN RESPONSE SURFACE METHODOLOGY AND ARTIFICIAL NEURAL NETWORK AND PRODUCTION OF BIOETHANOL THEREOF

Abstract

To release reducing sugars from cocoa pod shell, sulphuric acid pre-treatment was adopted where the variables affecting H2SO4 pre-treatment i.e., cocoa pod shell weight, H2SO4 concentration, microwave irradiation time and power were screened using one factor at a time approach. The weight of cocoa pod shell, concentration of H2SO4 showed a significant effect on the pre-treatment process and the levels of these factors were further optimized by central composite design using response surface methodology .The optimized conditions were found to be 15.65g ofcocoa pod shell, 6% v/v H2SO4 at 8 min of irradiation, released maximum reducing sugar of 9.10 g/L. A second order model was generated and validated, which was found to be a good fit with R2value of 0.89. Artificial neural network modelling proved validation R2 of 0.94 comparatively better than Response surface methodology R2 of 0.89. The reducing sugars released after acid hydrolysis at optimized conditions were subjected to fermentation by Pichia stipitis to produce bioethanol. The bioethanol produced was 3.2g/L at 2% (v/v) inoculum concentration after 72h of fermentation.