Modelagem e simulação da síntese enzimática de oligossacarídeos catalisada por dextrana-sacarase usando sacarose como substrato e maltose como aceptor

• Ano de defesa: 2019
• Autor(a) principal: Magalhães, Madson Linhares
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Palavras-chave em Português: Engenharia química; Modelagem matemática; Oligossacarídeos; Polimerização; Mathematical modeling; Oligosaccharides; Polymerization
• Link de acesso: http://www.repositorio.ufc.br/handle/riufc/42204

Food has great importance in the maintenance of health, due to this, foods like oligosaccharides, have been gaining space in the world market. Oligosaccharides are non-digestible carbohydrates, with degree of polymerization (DP) between 3 and 10 monomers, which have prebiotic action, beneficially altering the intestinal microbiota. Recent studies show large applications of oligosaccharides in the food, cosmetic and pharmaceutical industries. Dextransucrase is an enzyme that can produce different oligosaccharides depending on the acceptor used. As there are not many studies on the enzymatic kinetics of these compounds, this work aims at the development of a mathematical model for the synthesis of oligosaccharides catalyzed by dextransucrase using sucrose as substrate and maltose as acceptor. Three models were developed: 1) synthesis of oligosaccharides only; 2) synthesis of oligosaccharides and dextran by the same active site; 3) synthesis of oligosaccharides and dextrans by different sites. All models were developed approaching techniques of polymerization kinetics such as: initiation / formation, propagation and termination steps. The model parameters were adjusted by the Levenberg-Marquardt method. The implementation of the models and simulations were performed using the Python programming environment. The third model was the one that best represented the experimental data, presenting a standard residual deviation of 7.06% for the total oligosaccharides in the experiments used to adjust the parameters. These deviations were 1.63% and 8.49% with the two data sets used for validation. Model 3 predicted the molecular weight distribution for dextran, being a differential, since most models found in the literature do not capture this information. After validation of model 3, simulations were performed to evaluate the best form of large-scale operation. The following modes of operation were simulated: batch, batch, semi-batch, with fixed feed, half-batch with feed in pulses equally spaced in time and semi-batch with feed varying with time. The best strategy was the semi-batch with fixed feed, presenting a higher production of oligosaccharides with DP5 and DP6. This operational strategy was optimized using the particle swarm (PSO) method to maximize the production of oligosaccharides. Optimized conditions were obtained with an initial reaction volume of 1 m3 with concentrations of 40 and 45 g / L of sucrose and maltose and the fixed feeds of 109.1 and 235.1 L / h of sucrose and maltose, respectively.