Optimization of Biodiesel Production from Mango Seed Kernel Oil Using Taguchi and Weighted Sum Method
DOI:
https://doi.org/10.5935/jetia.v12i60.4006Abstract
The increasing need for second generation feedstocks for biodiesel production due to the drawbacks associated with edible oil-based biodiesel has led to the use of non-edible feed stocks like mango seed kernel oil (MSKO). The objectives of this research are to improve the biodiesel production from MSKO through transesterification process and to maximize the biodiesel yield as well as the calorific value of the biodiesel. Four factors, namely methanol-to-oil ratio (4:1, 6:1, 8:1 and 10:1), reaction temperature (45°C, 50°C, 55°C and 60°C), reaction time (30 min, 60 min, 90 min and 120 min) and catalyst concentration (0.5%, 1.0%, 1.5% and 2.0%) were chosen and studied at four levels using Taguchi’s L16 orthogonal array. A multi-objective optimization approach was used in this study by applying the WSM on the response variables that were normalized and weighted (yield 0.6 and calorific value 0.4). The findings indicated that the best conditions for transesterification were methanol to oil ratio of 8:1, temperature of 55°C, time of 120 minutes and catalyst concentration of 1.5% to yield biodiesel of 96.15% and calorific value of 39.67 MJ/kg. Under these optimal conditions, confirmation tests were conducted which gave nearly similar results (96.16 ± 0.07% yield and 39.70 ± 0.08 MJ/kg calorific value) thereby confirming the effectiveness of the optimization process. Analysis of variance (ANOVA) revealed that reaction temperature has an effect on calorific value at p < 0.05 while other parameters have a combined effect on yield. This study shows that MSKO is a feasible feedstock for biodiesel production that is not suitable for human consumption and provides a statistically significant model for improving the quality of the fuel and the efficiency of the conversion process.
Downloads
Downloads
Published
Issue
Section
License
Copyright (c) 2026 ITEGAM-JETIA

This work is licensed under a Creative Commons Attribution 4.0 International License.








