MODELLING OF PROCESSING PARAMETERS OF “PUPURU” - A FERMENTED CASSAVA (Manihot esculenta) PRODUCT

Abstract

Qualities of cassava processing into fermented products have been discovered to be different from one processor to another and also from one batch processing to another by the same processor. This study was therefore designed to establish models for the processing parameters for the production of “pupuru” (a cassava product). Cassava clone 07/0593 was used for the study. A defined three factor, three level central composite designs (CCD) of Response Surface Methodology was adopted to determine the optimum condition of fermentation by studying the effect of independent variables mass of root (2 to 6 Kg), fermentation time (72 to120 hrs) and fermentation temperature (28 to 32 oC) on the dependent variables pH and Titratable Acidity (TTA). The optimized parameters were used to produce fermented ‘pupuru’ mash for the toasting process. A three factor, three level central composite designs (CCD) was also adopted for the toasting process. The independent variables were the weight of fermented mash (500-900 g), toasting time (10-30 mins) and toasting temperature (65-85 oC) while the dependent variables were proximate composition, functional properties and pasting characteristics. The best runs from the optimized ‘pupuru’ flour were analyzed for mineral composition, anti-nutrient factor and sensory properties. ‘Pupuru’ sample prepared by traditional method was taken as control. The result showed that pH of the flour ranged from 3.83 to 4.60 while Titratable Acidity (TTA) ranges from 0.09 to 0.27. The data was fitted to a second order quadratic model. Based on the desirability level of 89.9% the optimum condition of fermentation temperature of 28.57 oC fermentation time of 105.30 h was established. While the optimum parameter for toasting process; weight of cassava wet meal 500 g, toasting time 20 min and toasting temperature 85 oC. The moisture contents ranged between 8.19 to 13.11%, crude protein content ranged between 0.14 to 0.84%, crude fibre 1.26 to 5.51%, crude fat 0.12 to 0.89%, ash content 0.14 to 1.26%, carbohydrate 82.17 to 89.08%, and energy 333.02 to 357.92 kcal. The functional properties result for water absorption capacity was between 2.00 to 4.40 ml/g, swelling index varies between 2.25 to 5.40 ml/ml, while bulk density varies between 0.42 to 1.67 g/cm3. R2 values were obtained as 0.5859, 0.7727 and 0.9866 for water absorption capacity, swelling index and bulk density respectively indicating significant model (p<0.05). The pasting characteristics shows that the model was insignificantly influence (p>0.05). The mineral results, calcium (Ca) content ranges between 10.25 to 12.20 mg/100g; Sodium (Na) content between 2.58 to 3.18 mg/100g, Potassium content 2.86 to 4.18 mg/100g, Magnesium (Mg) values 1.26 to 1.38 mg/100g, Iron (Fe) content 1.05 to 1.83 mg/100g. Zinc (Zn) values ranged between 0.84 to 0.99 mg/100g. The Ca/P ratio were less than the recommended value of one while Na/K ratio were less than value one. The anti-nutrient factor phytate content ranges from 150.88 mg/100g to 297.53 mg/100g while the tannin content was very low, the cyanide content ranged from 2.76 to 3.62 mg/kg. Sensory evaluation shows that there was significance different between the optimized samples and the control. All the optimized samples were rated higher than the control. Responses variables predicted with model equations under optimum conditions were in general agreement with experimental data. It can be concluded that the data obtained could be used for control of product quality.