Adsorptive removal of palm oil free fatty acids onto silica-smectite composite: a statistical study using Box–Behnken design in response surface methodology

High content of free fatty acids in palm oil is undesirable and is unhealthy for direct consumption and limits its quality for the production of some industrial products such as oleochemicals and biofuels. In this study, response surface methodology (RSM) was used to optimize the free fatty acids adsorption onto a silica-smectite composite for the first time. The regression model was found to be able to predict more than 99% of the targeted response variation in the studied range (R2 = 99.31%). ANOVA test was used to evaluate the accuracy of the regression model, and it was found to predict more than 99% of the response variation. The Pareto charts of standardized effects showed that almost all the explanatory factors were influential. The factorial plots evidenced that time/temperature interaction was most effective and the corresponding contour and surface plots were drawn. The second-order kinetic model was better followed with the highest value of the determination coefficient (R2 = 0.996) and the lowest value of root mean square deviation of 0.555. Response optimizer function showed an optimum for FFAs removal of more than 90% (27.18 mg of KOH/g) at the adsorbent dosage of 2%, contact time of 60 min, and temperature of 90 °C. A good correlation between the predicted and experimental values was observed within reasonable limits. RSM provided a facile mean to optimize the process compared to some classical methods of optimization. Compared with other silica-based adsorbents, the silica-smectite composite is more suitable and efficient for the removal of FFAs from palm oil.