Machine learning-optimized methyl oleate production using synthesized [Lys][CF ₃ SO ₃ ] ₂ ionic liquid as catalyst

An amino acid ionic liquid, with lysine as the cation and CF3SO3- as the anion, was synthesized by ion‐exchange between lysine and trifluoromethanesulfonic acid. This ionic liquid was then explored as a catalyst for the esterification of oleic acid with methanol to produce methyl oleate, commonly known as biodiesel. Using this catalytic ionic liquid helps to mitigate common issues in conventional biodiesel synthesis—such as difficult product separation, catalyst recovery, and waste liquid disposal. This study investigates the effect of the catalyst on the conversion of oleic acid and identifies the optimal reaction conditions, including temperature, methanol-to-oleic-acid molar ratio, catalyst loading, and reaction time. The process was first optimized using traditional single-factor experiments to evaluate the main operating variables. Subsequently, data-driven methods, including Response Surface Methodology (RSM) based on a Box–Behnken design and Random Forest regression modeling, were employed to determine the optimal conditions for maximizing methyl oleate yield. The computational approach served to significantly reduce the required number of experimental trials, and allows to systematically identify optimal reaction conditions. The verified computational results had only 0.17% deviation from the predicted values.