Espectroscopia na região do infravermelho próximo (NIR) e calibração multivariada: desenvolvimento de modelos PLS para a determinação da atividade antioxidante em amostras de própolis

• Ano de defesa: 2018
• Autor(a) principal: Calegari, Matheus Augusto
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Tecnológica Federal do Paraná; Pato Branco; Brasil; Programa de Pós-Graduação em Tecnologia de Processos Químicos e Bioquímicos; UTFPR
• Palavras-chave em Português: Antioxidantes; Espectroscopia de infravermelho; Quimiometria; Análise espectral; Antioxidants; Infrared spectroscopy; Chemometrics; Spectrum analysis; CNPQ::ENGENHARIAS::ENGENHARIA QUIMICA; Engenharia Química
• Link de acesso: http://repositorio.utfpr.edu.br/jspui/handle/1/3199

Propolis began to gain popularity due to its use in complementary and alternative medicine and has now stood out among researchers due to its biological and pharmacological properties such as antitumor, antimicrobial, anti-inflammatory antiseptic, healing antioxidant and anesthetic. The chemical composition of propolis varies according to the characteristics of the vegetation and the location where the hives are inserted, usually consisting of resin, vegetable balsam, wax, essential oils, pollen and other substances such as flavonoids, phenolic acids, esters and the ketones. The objective of this work was to determine the antioxidant activity of propolis extracts by the free radical capture methods ABTS•+ (2,2'-azino-bis (3-ethylbenzothiazolin) -6-sulphonic acid), DPPH (2,2-diphenyl-1-picrylhydrazine) and a reduction of Fe+3 to Fe+2 (FRAP), to determine the total phenolic and flavonoid content of propolis extracts by means of spectrophotometric analyzes and to obtain the NIR spectra for to develop multivariate calibration models using PLS and spectral pre-treatments that have the capacity to predict the antioxidant activity and chemical composition of propolis samples. As a final step, an analytical methodology by HPLC-PDA for the identification of phenolic acids and flavonoids in propolis extracts was optimized. It was possible to develop multivariate calibration models using PLS regression associated with spectral pre-treatments. These models were efficient for predicting the antioxidant activity and chemical composition of propolis samples, based on R², RMSEC, RMSECV, RMSEP, RMSEE, RPD, RER and number of latent variables. From the CLAE-DAD technique it was possible to optimize a chromatographic methodology with good values for the quality parameters. The condition of the developed method was: flow 1 mL min-1, column temperature 30 ° C, mobile phase composition: (A) H2O:H3PO4 (99,8:0,2 v v-1); (B) CH3OH (100%) with gradient starting with 30% B, in 15 min 64% B, 11 min 75% B, 2 min 95% B, 1 min 95% B, 3 min 30% of B and finally 10 min in 30% of B, totaling 42 min of analysis, to quantify caffeic acid, ferulic acid, p-coumaric acid, trans cinnamic acid, rutin, quercetin, pinocembrina, chrysin, canferol, mangiferin and galangin. It was possible to verify that the samples that presented higher values for the antioxidant activities were those that presented richer chromatographic profiles with more intense signals.