LAUSR

Abstract The present study aims to maximize the conversion of microalgae oil to fatty acid methyl ester (FAME) using supercritical methanol (SCM) transesterification by sequential hybrid optimization using response surface methodology (RSM), artificial neural network (ANN) and genetic algorithm (GA). The three process parameters selected for the optimization of SCM transesterification were temperature (240 to 300 °C), time (15 to 45 min) and MeOH: oil molar ratio (15:1 to 45:1). Initial experiments performed according to the central composite design (CCD) generated matrix of RSM and further validated by ANN. The 1H-NMR analysis confirms the formation of methyl esters. Moreover, the corresponding regression coefficient (R2) for the model were 0.97 and 0.99 for RSM and ANN, respectively indicated excellent fit of the model to the experimental data. Furthermore, the final optimized condition for FAME conversion efficiency of RSM and ANN predicted models were 98.01% and 98.15%, respectively. The fitness function for GA was obtained from ANN predicted model equations and presented as globally optimized (GA conditions) reaction conditions for SCM: temp - 285.21 °C, time - 26.57 min and MeOH: oil molar ratio – 23.47. The predicted percent conversion efficiency of GA optimized conditions was 99.16% whereas, the experimental optimum FAME conversion reached to 98.12%. Additionally, the gas chromatography-mass spectroscopy (GCMS) analysis revealed the presence of palmitic (28%), oleic (33%), linoleic (8%) and other saturated and unsaturated fatty acids. The other biodiesel properties such as acid value, iodine value, cetane number, calorific value, etc. were also analyzed and exhibited an analogous trend with standard ASTM D6571 standards.