Applicable analytical methods for the determination of solid impurities content in raw sugarcane

• Autor(a) principal: Guedes, Wesley Nascimento [UNESP]
• Data de Publicação: 2020
• Tipo de documento: Tese
• Idioma: eng
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://hdl.handle.net/11449/193319
• topic: Sugarcane; Straw; Digital images; Laser-induced breakdown spectroscopy; Chemometrics; Cana-de-açúcar; Quimiometria; Imagens digitais; Palha

The introduction of sugarcane stalks free of solid impurities in the sugar and ethanol manufacturing system is the ideal situation for sugar and alcohol mills. The solid impurities are mainly related to the presence of green and dry leaves and, also of soil in the raw material, sugar cane However, this type of material is inevitably introduced into the process during the harvest. The presence of solid impurities in the manufacturing system of the sugarcane mills increases the costs and steps in the production of sugar and ethanol. Thus, the aim of this thesis study was to identify which chemical elements could be considered characteristic, such as a fingerprint of solid impurities, as well as to develop models to classify and estimate different levels of impurities in sugarcane. In chapter 1, an analytical method was developed to identify the chemical elements in samples with levels of impurities between 0 and 10% (g/100g) using laser-induced breakdown spectroscopy (LIBS) and the principal component analysis (PCA). The chemical elements Ca, Mg and K were those most related to variations in the levels of solid impurities. In chapter 2, an analytical method was developed using digital images and chemometric techniques to classify the content of sugarcane stalks in the presence of solid impurities. The models had success rates above 97%. In chapter 3, the same data set from the digital images used in chapter 2 was applied, the artificial neural network (ANN) tool. For this case, the sugarcane content was estimated in the presence of solid impurities. The three methods developed showed good rates of precision and accuracy. Besides the applicability present itself possible in view of the absence of analytical methods related of the sugarcane mills process, it must also be considered that the character of simplicity in terms of instrumentation and sample preparation has been achieved.