Emprego de técnicas espectroanalíticas para a análise de amostras de solos e de coprólitos : análise de textura e exploratória

• Ano de defesa: 2015
• Autor(a) principal: Sperança, Marco Aurélio
• Orientador(a): Pereira Filho, Edenir Rodrigues
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de São Carlos; Câmpus São Carlos
• Programa de Pós-Graduação: Programa de Pós-Graduação em Química - PPGQ
• Palavras-chave em Português: Coprólitos; Solos; Quimiometria
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::QUIMICA
• Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/7723

In this study, samples of coprolites (fossilized feces) were analyzed in LIBS (Laser-Induced Breakdown Spectroscopy) to evaluate the chemical profile in these samples using hyperspectral images, which enable us to assess the different concentration of several elements in the same image. One of these samples was submitted to analysis by ICP OES (Inductively Coupled Plasma Optical Emission Spectroscopy) separately (rock and coprolites) to assess the actual concentration of each element in both of the arrays. Through the results of both techniques, it was possible to infer the diet of animals that generated the coprolites. The samples correspond to the Oligo-Miocene, Cretaceous, and Permian periods (28, 85 and 260 million years, respectively). Coprolites samples showed positive correlation between the Ca and P elements, indicating the presence of Ca3(PO4)2, compound generally present in carnivorous animal waste. In rock samples, the predominant elements were Al, K, Na and Si. Through a PCA analysis (Principal Component Analysis), it was found that the elements that characterized the samples changed as the age. Through scores graph analysis, it is possible to identify groups of samples in chronological order. The soil samples were analyzed by NIRS (Near-Infrared Spectroscopy) in an attempt to create classification models for soil texture, which is characterized by sand fractions, clay and silt soil. These are differentiated by particle size. Various strategies have been adopted in an attempt to improve the created models and several chemometric tools were utilized, such as SIMCA, KNN and PLS-DA for classification, giving satisfactory results as accuracy 83-100% for the calibration set. PLS to quantify the levels was also tested, but without success. The samples were also subjected to analysis by LIBS and soil encapsulation strategy with a polymer was tested, also without success, since the LIBS generates the spectral profile elements and does not distinguish particle size.