Desenvolvimento de método de análise direta de folhas de soja e milho empregando a Laser-Induced Breakdown Spectroscopy (LIBS)

• Autor(a) principal: Assis, João Victor Borges
• Data de Publicação: 2023
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UFSCAR
• Texto Completo: https://repositorio.ufscar.br/handle/ufscar/18254
• topic: LIBS; Direct Analysis; Leaves; Análise direta; Folhas; CIENCIAS EXATAS E DA TERRA::QUIMICA; CIENCIAS EXATAS E DA TERRA::QUIMICA::QUIMICA ANALITICA::INSTRUMENTACAO ANALITICA; CIENCIAS EXATAS E DA TERRA::QUIMICA::QUIMICA ANALITICA::METODOS OTICOS DE ANALISE; CIENCIAS EXATAS E DA TERRA::QUIMICA::QUIMICA ANALITICA

This academic master's dissertation was devoted to the development of analytical methods for the determination of macronutrients (P, K, Ca, Mg) and micronutrients (B, Cu, Fe, Mn, S and Zn) in corn, soy (wihout and with petiole) leaves. The main purpose of this study was to present Laser-induced breakdown spectroscopy (LIBS) as a viable alternative for the direct analysis of leaf samples using chemometric tools to interpret the data obtained. The usage condition chosen for LIBS was 70 mJ of energy, 1.0 µs of delay time and 100 µm of spot size, which were applied to 952 samples, 81 of corn, 305 of soy and 566 of soy with petiole. The reference values of the analytes for the construction of calibration models were obtained using the Inductively Coupled Plasma Optical Emission spectroscopy (ICP OES). To minimize signal variations and sample matrix differences, twelve normalization modes and two calibration strategies were tested. The following were studied: multivariate calibration using Partial Least Squares (PLS) and nivariate calibration using area and height of several emission lines. Thus, sought to identify the best normalization mode, emission line and calibration strategy for each analyte. The noteworthy normalization mode for most models was the Euclidean norm. No analyte showed good results for univariate calibrations. Multivariate models for P, S and micronutrients did not show satisfactory results. The models obtained for Ca, K and Mg showed good results, the standard error of calibration (SEC) varied from 0.5 g/kg, for Mg in corn leaves with 2 latent variables, up to 4.9 g/kg, for K in soy with petiole leaves with 2 latent variables. As a conclusion of the work, the proposed calibration models were tested on validation and prediction datasets with promising results for direct routine analysis of the plant for Ca, K and Mg.