Modelagem da difusão intrapartícula para predição da cinética da esterificação catalítica heterogênea de ácido acético e isopropanol empregando Nb2O5

Abstract

The esterification reaction is one of the most important processes of organic synthesis in liquid phase, since esters have many applications in the chemical industry. Isopropyl acetate, for example, is used as a solvent in synthetic resins, fragrances and printing inks. Esters can be obtained in various chemical reactions; however, they are often synthesized via the Fischer esterification reaction. Niobium oxide has been reported in the literature as an efficient catalyst for esterification. Its physical-chemical properties, such as surface area and acidity, are highly dependent on heat treatment conditions, due to changes that occur in the oxide structure at high temperatures and consequently affect the catalytic activity of niobium oxide. There are few studies in the literature that use diffusion models in heterogeneous catalysis. An intraparticle diffusion model was developed to explain the kinetics of the esterification of acetic acid with isopropanol, catalyzed by niobium oxide. The kinetic equations were defined by the pseudo-homogeneous (PH), Eley-Rideal (ER) and Langmuir-Hinshelwood-Hougen-Watson (LHHW) models. The kinetic and adsorption parameters were estimated using particle swarm optimization and nonlinear least squares. Finite difference approximations were applied to solve the partial differential equations present in the model. The increase in the calcination temperature caused changes in the structure of Nb2O5, as well as a reduction in the specific surface area, average pore volume and acidity. The oxides calcined at 150 and 350 °C showed similar catalytic activity, however it was negligible for the oxide calcined at 550 °C. The results showed that the conversion increased with increasing temperature and mass of catalyst. No difference in conversion was observed with increasing stirring speed. The values of kinetic and mass transfer parameters were similar to those reported in the literature. The LHHW model predicted the reaction kinetics better than the PH and ER models. Through an experimental design, an empirical model was developed correlating the conversion with the operational variables. The graphs of the empirical model as a function of the operational variables indicated that temperature, molar ratio and the interaction between these two variables are the most significant factors for the process.