Modeling of the nanocrystalline-sized mesoporous zinc oxide catalyst using an artificial neural network for efficient biodiesel production

Abstract

The sulfonated mesoporous zinc oxide catalyst (SO₃H–ZnO) was hydrothermally fabricated and functionalized by sulfonation to catalyze the palm fatty acid distillate (PFAD) to esters. The effect of different reaction parameters including reaction time, reaction temperature, metal ratio, and calcination temperature was modeled by artificial neural networks (ANNs) to find out the possible relative optimum conditions of the synthesized mesoporous SO₃H–ZnO catalyst for the prediction of the nanocrystalline size. Under the optimized conditions of calcine temperature 700 °C, 18 min reaction time, 160 °C reaction temperature, and 4 mmol of Zn concentration predicted a 56.41 nm size of the mesoporous SO₃H–ZnO catalyst. The acquired model was statistically verified for its utility. The quick propagation model with four nodes in the input layer, six nodes in the hidden layer and one node in the output layer (QP-4-6-1) was chosen as the final model due to its optimum statistical characteristics. Furthermore, the most effective parameter was found to be the zinc concentration whilst the reaction time demonstrated the least influence. The optimized mesoporous SO₃H–ZnO catalyst was further utilized for esterification of PFAD, depicting a high fatty acid methyl ester yield (96.11%). It shows a valuable application for the conversion of discarded oils/fats containing high free fatty acids for the production of renewable green biodiesel.

Keywords:

• Esterification
• Genetic algorithm
• Optimization
• Sulfonated mesoporous catalyst

Additional information

Funding

The authors would like to extend their sincere appreciation to King Abdulaziz City for Science and Technology (KACST) for funding the project number LGP-36-21.