Optimization of Mechanical Extraction by Response Surface Methodology and Oil Yield Characterization from Single‐Grain Castor Seed

Abstract

AbstractThe application of castor oil in various fields and the development potential in renewable energy have attracted large attention, and castor oil extraction has become a research focus. The objective of this study is to determine the relationship between important parameters during single grain extrusion and to obtain the maximum oil yield. Mechanical compression tests are conducted by a texture analyzer. Response surface methodology is used to optimize the process variables. A four‐factor three‐level Box–Behnken design is used to generate factor combinations. A quadratic mathematical model between each variable and oil yield is developed. The significance of each interaction term is assessed by analysis of variance. The model's oil yield prediction accuracy is verified by tests. The highest oil yield of 30.18% is obtained under strain 94.97%, speed 0.1 mm s−1, dwell time 480 s, and drying at 20 °C. The determination coefficients (R2 = 0.9885; R2Adj = 0.8979; R2Pre = 0.8782) indicate the fitted quadratic mathematical model has a good correlation between experimental and predicted values. Lack of fit (0.9998) term is not significant and indicates the model reliability. Predicted oil yield value (30.18%) is close to experimental testing value (31.47%), which further demonstrates the mathematical model reliability.Practical applications: The extracted castor oil can be used for industrial research such as biodiesel. Castor seed is rich in oil content. Optimization of process parameters can improve oil recovery rate of castor seed. The developed model in this study provides a basic reference for improvement equipment related to the oil extraction.