Intelligent approaches to process optimization of biodiesel synthesis from Ricinus communis seed using a fusion of chicken and duck eggshells doped with KBr catalyst

Abstract

This study focused on modelling and optimization of castor biodiesel synthesis utilizing a fusion of calcined chicken and duck eggshells doped with potassium bromide (CaO-KBr) catalyst. The eggshells were calcined at a temperature of (800−1000 °C) for 3 h and doped with KBr at a mixing ratio of (4:1–2:1 wt%) and activated at 500−700 °C for 2 h. Characterization of catalysts showed that CaO-KBr catalysts have smaller grains and a greater specific surface area as compared to CaO catalysts. Response surface methodology (RSM), artificial neural network coupled with genetic algorithm (ANN-GA), and adaptive neuro-fuzzy inference system-genetic algorithm (ANFIS-GA) were utilized for the optimization of process parameters. Results showed that the performance of all the models exhibited adequate prediction accuracy with a coefficient of determination (R²) and root mean squared error (RMSE) of ANFIS (0.999, 0.012), ANN (0.925, 0.111) and RSM (0.928, 0.104). Under optimal conditions, maximum biodiesel yield of 97.83 ± 0.49 % was achieved using ANFIS-GA which was higher than ANN-GA (97.28 ± 0.57 %) and RSM (97.04 ± 0.43 %). The reusability study of the CaO-KBr showed improved recyclability up to the 7^th cycle (>80 % yield) compared to the CaO catalyst (4^th cycle >80 % yield). The properties of the synthesized biodiesel also meet EN 14214 and ASTM D6751 standards. Utilization of CaO-KBr catalyst resulted in cheap and eco-friendly method of castor biodiesel production.