Metodologia automática em fluxo-batelada baseada em análise de imagens digitais para determinação de N-acetil-L-cisteína em formulações farmacêuticas

• Ano de defesa: 2018
• Autor(a) principal: Lima, Rosinaldo Telis de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal da Paraíba; Brasil; Química; Programa de Pós-Graduação em Química; UFPB
• Palavras-chave em Português: Analisador automático em fluxo-batelada; Imagens digitais; N-acetil-Lcisteína; Formulações farmacêuticas; Automatic flow-batch methodology; Digital images; N-acetyl-L-cysteine; Pharmaceutical formulations; CNPQ::CIENCIAS EXATAS E DA TERRA::QUIMICA
• Link de acesso: https://repositorio.ufpb.br/jspui/handle/123456789/14843

It is of paramount importance to carry out quality control in the pharmaceutical industry in order to give quality, safety, efficacy and credibility to the consumer market. One of the most important quality parameters in the routine analysis of a pharmaceutical industry is the quantitative analysis of the active principle found in the medicine, since these industries must comply with the Good Manufacturing Practices of Medicines established by the National Sanitary Surveillance Agency. Therefore, it is proposed an automatic flow-batch methodology based on digital image analysis for the determination of N-acetyl-L-cysteine in pharmaceutical formulations. The proposed automatic system uses a high resolution webcam to capture the RGB data of the digital images of the colorimetric reaction product. This reaction involves two steps: 1) reaction of the analyte with Fe(III) to generate Fe(II) and then 2) complexation of Fe(II) with 1,10-phenanthroline. All analytical solutions were prepared online and all analytical processes were performed simply by changing the operational parameters of the control application. The optimum color component of the study was a B, with a linear response in the working range of 8.08×10-5 to 1.85×10-4 mol L-1. The calibration model was validated through analyzes of variance for lack of fit and regression significance, performed after a graphical analysis of the residuals left by the model. The limits of detection and quantification and the analytical frequency were estimated to be, respectively; 2.18×10-6 mol L-1; 7.26×10-6 mol L-1 and 30 h-1. The results obtained through the proposed method were compared with those of the reference method. When applying the t-paired test at a 95% confidence level, it was found that both methods did not present statistically significant differences in terms of accuracy. Therefore, it can be stated that the proposed methodology is reliable, low cost, fast and presents low consumption of reagents and samples.