Maximization of biodiesel production from sunflower and soybean oils and prediction of diesel engine performance and emission characteristics through response surface methodology

Abstract

Production of methyl ester from sunflower and soybean oil mixture is performed throughout a catalyzed transesterification procedure. The significance of the four reaction parameters such as methanol to oil ratio, catalyst concentration, mixing speed, and reaction time and their combined effect on biodiesel yield is investigated through twenty-nine of the pre-designed and performed experiments. Box-Behnken design (BBD) based on response surface methodology (RSM) was applied for process optimization. A quadratic regression model was established for biodiesel yield prediction with a coefficient of determination R2 of 0.9861. An maximum biodiesel yield of 93.38% is accomplished at 203.5:1 ml:l methanol to oil ratio, 0.57 wt% catalyst concentration, 52 min reaction time and 530 rpm mixing. Obtained results show that there is a superior compatibility among the calculated yield of 93.38% and the experimental data of 93.2%. The estimated biodiesel fuel properties met with the American society for testing and materials (ASTM) D6751 standards. Engine operating parameters optimization have been executed using central composite design method (CCD) to achieve an optimum break thermal efficiency of a lone cylinder DI-engine fueled by biodiesel/diesel mixtures. Engine input parameters were considered as engine load and blends percentage for the optimization of engine response represented in break thermal efficiency (BTE), unburned hydrocarbon (UHC), and Nitrogen oxide (NOx) emissions. Examination of inconsistency (analysis of variance) ANOVA indicated that the quadratic representation were statistically important. RSM optimizer results indicated that the best possible values of BTE, UHC, and NOx were 13.656%, 120.7748 ppm, and 234.8926 ppm, respectively, at the maximum value of biodiesel mixture of 70% and break power of 2.05 kW. A validation test was performed and the error percentage is found to be within the range of 5%. The error percentage for BTE, UHC, and NOx was found to be 3.34%, 1.35%, and 2.31%, respectively.