ABSTRACT

This study was on modeling and optimization of bioethanol production process from a lignocellulosic material. The process entailed separate hydrolysis and fermentation (SJ{F) of corn stover to produce bioethanol. Corn stover is the left over portion of the maize or non-grain portion of maize, usually above the ground, after harvest. It is a lignocellulosic material because it has a tough lignin covering over the cellulose and hemicellulose layers. The corn stover used in this study was from the yellow variety of sweet corn; it was analyzed in the Central Laboratory, Nigerian Institute for Oil Palm Research (NIFOR). The corn stover was cleaned, chopped, oven-dried at 60 °C for 48 hours (to moisture content of 10 %), milled, sieved to produce uniform particle size range of 0.18 — 0.25 mm and analyzed for its composition. The corn stover was then pretreated with 2% (w/w) Sodium Hydroxide (NaOH) solution followed by hydrolysis with dilute Sulfuric acid (H2SO4;v/v) of 1, 2.5 and 4 % respectively. Yeast, Saccharomyces cerevisiae was used for the fermentation and at fermentation time of 12, 30 and 48 hrs. Fermented samples were withdrawn for bioethanol content analysis using gas chromatography. Box-Behnken design (BBD), a statistical tool and feature of the Design Expert software (7.0.0 Trial version) was used in the design of the experiments and a total of 54 experimental runs was obtained. The response variable, Bioethanol yield (mg/i) was represented with Y, while the six independent variables; Sulfuric acid concentration, Hydrolysis time, Fermentation time, Concentration of yeast, Fermentation temperature and pH of Hydrolysate samples were represented by A, B, C, D, E & F respectively. The results obtained from the experiments were inputed into the Design Expert software, for model equation development, using the analysis of variance (ANOVA) feature of the Design Expert software. From the ANOVA, only A, F, AF, CD, A2were significant (having P-value S 0.05), resulting in a reduced Model Equation (i.e adjusted), in terms of actual faôtors. The optimum predicted process parameters for the optimum bioethariol yield of 149.41 mg/l were 1.08 %, 3.32 hrs, 14.32 hrs, 6.43 gIl, 39.34 °C and 7.64 for A, B, C, D, E and F respectively, while an average bioethanol yield of 143.15 mg/i was obtained experimentally in three replicates using the optimum process parameters. The experimental bioethanol yield obtained was 95.8 1% close to the predicted optimum bioethanol yield. The values of R, R2, adjusted R2and Adeq Precision were 0.8747, 0.7651, 0.5212 and 8.309 respectively. The value of R2indicates a good degree of correlation between the experimental and predicted bioethanol yields, being close to one. Since the "Adeq Precision" was greater than 4, it implies that the model was adequate.