RESPONSE SURFACE OPTIMIZATION OF DESIGNED PARAMETERS FOR ENHANCED STARCH YIELDS FROM CASSAVA, SWEET POTATO AND COCOYAM PEELS

Abstract

The purpose of this study was to extract starch and optimize its yield from cassava, sweet potato and cocoyam peels. Response Surface Methodology (RSM), adopting the Box-Behnken design of experiment, was used to determine the optimal parameters such as weight of starch (W), concentration of reagent (C_R) and treatment time (T_t), and the optimum starch yields (y), which had high coefficient of determination (R²) of 1.000 along with low probability value of P < 0.0001 for the three sources of starches, indicating the reliable predictability of the models. The optimal parameters for the extraction of starches from the peels were W = 33.20 g, C_R = 1.00 g/100 mL and 2.00 h for cassava peel flour, W = 25.00 g, C_R = 1.00 g/100 mL and 2.00 h for sweet potato peel flour and W = 31.53 g, C_R = 1.25 g/100 mL and 2.48 h for cocoyam peel flour, which resulted in an optimum yield of 30.6 g of starch per 50 g (61.2%) of cassava peel flour, 28.1 g of starch per 50 g (56.2%) of sweet potato peel flour and 30.1 g of starch per 50 g (60.2%) of cocoyam peel flour. The results obtained indicated that RSM can be used to find the relationships among process variables and response using minimum number of experiments, and hence enhancing the starch yields.