Extraction of oil from selected plants using Response Surface Methodology [RSM]

Abstract

Abstract This study involves the extraction of oil from three sources: orange peel, guava leaves, and cassia fistula leaves using Soxhlet apparatus. The variables considered in this study were time of extraction and sample weight. Minitab statistical software was used to randomize the runs. The combination of operating parameters that gave the optimum yield for the three sources were identified. The regression equation for each source was reported. The coefficient of determination (R2) value for orange, guava leaves and cassia fistula extract were 99.51%, 99.90%, and 99.77% respectively. This shows that the model is a good prediction tool for extraction of oil from these sources. Based on the R2 values guava leaves (99.90%) gave the highest prediction accuracy followed by Cassia fistula (99.90%), with orange leaves having the lowest R2 value (99.77%) among the three sources considered.