Laser-induced breakdown spectroscopy (LIBS), wavelength dispersive x-ray fluorescence (WDXRF) and chemometrics: possibilities for analytical applications in food analysis

• Autor(a) principal: Gamela, Raimundo Rafael
• Data de Publicação: 2020
• Tipo de documento: Tese
• Idioma: eng
• Título da fonte: Repositório Institucional da UFSCAR
• Texto Completo: https://repositorio.ufscar.br/handle/ufscar/13148
• topic: Amostras de alimentos; Análise direta de sólidos; Ferramentas quimiométricas; Planejamento de experimentos; Estratégias de calibração; Food samples; Direct solid analysis; Chemometric tools; Experimental design; Calibration strategies; CIENCIAS EXATAS E DA TERRA::QUIMICA::QUIMICA ANALITICA::METODOS OTICOS DE ANALISE; CIENCIAS EXATAS E DA TERRA::QUIMICA::QUIMICA ANALITICA

This PhD thesis proposes evaluation of the possibilities and analytical applications of the laser-induced breakdown spectroscopy (LIBS) and wavelength dispersive x-ray fluorescence (WDXRF) techniques for the determination of chemical elements in edible seeds as bean seeds, pumpkin, pea, lentil, and cocoa bean. These seeds are important for diet of the Brazilian population because, besides dietetic fibers, amino acids, vitamins phenolic compounds, it contains chemical elements that play different rules in human organism. The use of the LIBS and WDXRF allow the direct solid samples without or with a minimum sample treatment, high analytical frequency, lower reagent consumption, which become attractive when compared with conventional techniques. However, the disadvantage of the LIBS is its lower sensitivity, high limit of detection and absence of certified reference material for calibration. Moreover, the elemental determination by direct solid analysis using LIBS and WDXRF present challenges due to matrix effects caused by spectral interferences, physic characteristic of the samples and self-absorption phenomena, which can compromise the accuracy of the methods. In this sense, different calibration strategies were evaluated for direct solid sample analysis by LIBS. Moreover, was evaluated the possibility of the use of hyperspectral images to characterize the chemical profile of the distribution of the elements on the seeds. In all cases, chemometrics tools were used for data treatment.