Biodiesel yield optimization from ternary (animal fat-cotton seed and rice bran) oils using response surface methodology and grey wolf optimizer

Abstract

The attributes of constant convergence speed, capacity for reconnoitering the search space and absence of technical drawbacks viz. trapped in local minima and unswerving solutions have led to wider adoption of Metaheuristic Algorithms (MAs) compared to conventional modelling tool. Grey Wolf Optimizer (GWO) surpassed others such as Whale Optimizer and Anti-lion Algorithms among the MAs. For the first time, GWO was utilized to build novel composite biodiesel from ternary oils (TSOs), specifically animal waste fat, cottonseed, and crude rice bran. As a result, in this study, (1) Response Surface Methodology (RSM) and GWO were used to model the novel composite biodiesel (NCB) from the CTOs, (2) the efficacy in terms of biodiesel yield (BY), and (3) the commercial viability of the produced NCB were investigated. The 94.2% of yield for NCB was obtained for RSM determined optimal conditions set at the methanol/CTOs molar ratio of 9, catalyst amount of 0.45 wt%, reaction temperature of 50 oC, time of 105 min. A quadratic model was developed to predict the biodiesel yield, with an R2 value of 0.995, indicating high accuracy. Grey wolf optimization technique optimized the biodiesel yield to 98.6%, corresponding to catalyst concentration, methanol to TSO molar ratio, reaction temperature, and time set at 0.53 wt%, 9.32, 60 °C, and 105 min. The predicted biodiesel yield was 99.18%, closely matching the experimental yield of 98.60%. 1 H NMR and FTIR provide insights into biodiesel's composition and chemical structure, ensuring its compliance with quality standards. The high-quality NCB obtained from the optimized condition ensures TSO can be employed for higher-scale production in industries and to run in diesel engines.