Hard kinetic modeling of the industrial reaction of hydrogenation of soybean oil optimized by heuristic problem-solving techniques

Abstract

Partial hydrogenation of vegetable oils is an important process for generating fats with particular characteristics. In this study, the hard kinetic modeling of the nickel-catalyzed soybean oil hydrogenation reaction was optimized by three algorithms considering five reaction mechanisms. In three mechanisms, it was considered to have a formation of half-hydrogenated intermediate compounds, and two did not consider this possibility. The Particle Swarm Optimization (PSO) algorithm allowed for achieving lower errors between the experimental and modeled values compared to the Simulated Annealing (SA) and Sequential Simplex (SP) algorithms for the five evaluated reaction mechanisms. In addition, the optimized apparent rate constants values indicated that in these reactional conditions, the industrial reaction mechanism of hydrogenation follows some preferred reaction pathways, such as a preliminary isomerization of cis to trans fatty acids to be then hydrogenated to other fatty acids less unsaturated. These optimization algorithms proved to be efficient tools for investigating the kinetics of the partial hydrogenation reaction of soybean oil and offer promising perspectives for other industrial processes.