Aspectos cinéticos e termodinâmicos da adesão microbiana e modelagem e otimização da higienização Clean in Place utilizando fluxo pulsado

• Ano de defesa: 2023
• Autor(a) principal: Silva, Lucas Donizete
• Tipo de documento: Tese
• Tipo de acesso: Acesso embargado
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Engenharia Química
• Palavras-chave em Português: Adesão; Higienização; Fluidodinâmica; Otimização; Inteligência Artificial; Bactérias psicrotróficas; Leite; Segurança microbiológica; CNPQ::ENGENHARIAS
• Link de acesso: https://repositorio.ufu.br/handle/123456789/37782; http://doi.org/10.14393/ufu.te.2023.8034

Clean in place (CIP) is one of the most used techniques for cleaning and sanitizing surfaces in the food and pharmaceutical industry. However, some points in the pipes are complicated due to the geometry and fluid dynamics of the flow. The search for models and optimization of the CIP process is essential for the quality and microbiological safety of products. In this sense, the objective of Chapter 1 was to evaluate the multiplication of Pseudomonas fluorescens (ATCC 13525) and Listeria innocua (ATCC 33090), estimate kinetic parameters of growth, measure the hydrophobicity of cells and the surface of stainless steel and verify the presence of exopolysaccharides (EPS). In chapter 2, the geometry of five geometries was evaluated: straight cylindrical section, tee branching, 90º curve, expansion and diameter reduction, contaminated with the two microorganisms using a constant flow of sanitizer. In chapter 3, sanitization was applied with pulsed flow to evaluate the effect of time, ripple, pulse period and chemical agent concentration. The behavior of the decimal cell reduction was modeled using the Response Surface (RMS) methodology and the optimization was performed using the desirability technique. In addition, the fluid dynamic behavior within these geometries was investigated. In chapter 4, a model based on artificial neural networks (ANN) was developed for the sanitization of L. innocua with pulsed flow and the optimal point of operation was determined. 304 stainless steel tubes and tubes, #4 polish, were used. To determine cell growth, milk and surface priming were periodically performed using colony forming units (CFU) counts on plates. Kinetic models were made by parameter compliance. The hydrophobicity and thermodynamics of adhesion were determined by means of the contact angle. EPS production was investigated by FTIR infrared spectroscopy. In hygiene, a circulation line was contaminated with microorganisms. Then, the geometries were sanitized and the volunteer performed by the swab technique and the stabilization performance was evaluated based on the decimal reductions of cells. The fluid dynamic study was carried out with the FLUENT software Student version and the energy consumption with an electric current sensor It was observed that under refrigeration L. innocua presented a lag phase and then growth in the number of cells. P. fluorescens, on the other hand, did not show growth and there was no improvement in adherence. At optimal temperature, L. innocua did not show a lag phase and P. fluorescens showed a lag phase. Adherence of L. innocua was significantly higher than P. fluorescens. L. innocua showed a higher specific growth rate than P. fluorescens. Furthermore, the Luedeking-Piret equation parameter showed that adhesion is associated with cell growth in both microorganisms. FTIR analysis revealed that L. innocua did not synthesize EPS within the evaluation period, while P. fluorescens produced EPS from 12 h. Both species were hydrophilic, with L. innocua being less hydrophilic and energetically more favorable to adhesion. The decimal reduction in the traditional sanitization showed that the straight cylindrical section and the reduction presented the best levels of sanitization in detriment of the tee branching, for both microorganisms, due to the formation of stagnation zones, observed in the fluid dynamic simulation. Pulsed flow variables were significant in sanitization. The optimal condition required less operating time, saving energy and sanitizing agent. Lastly, the ANN model was able to accurately fit the experimental data and was comparatively more accurate than the predictions indicated by the RMS technique.