Caracterização geográfica e de características de qualidade da Jatropha mollissima (POHL) BAILL utilizando espectroscopia no infravermelho e quimiometria

• Autor(a) principal: Fernandes, Caroline Lins
• Data de Publicação: 2022
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UEPB
• Texto Completo: http://tede.bc.uepb.edu.br/jspui/handle/tede/4589
• topic: Plantas medicinais; Pinhão-bravo; MID; NIR; PCA; Quimiometria; Medicinal plants; QUIMICA ANALITICA::INSTRUMENTACAO ANALITICA

The use of medicinal plants for medicinal purposes is an ancient custom used by different peoples and continues to gain notoriety. It is within this context that more in-depth studies on Jatropha mollissima (Pohl) Baill, popularly known as Pinhão-bravo. This plant, as well as others in the same family, may have antiviral properties and improve other respiratory diseases. In addition, several studies have already shown that the sap of J. mollissima has an antiophidic, antioxidant and healing action. Due to its proven potential, studies are being carried out to identify the differentiation of this species present in some areas of Paraíba, highlighting three main regions: Boa Vista, Campina Grande and Cacimba de Areia. The plants are analyzed to verify differences in their majority composition, taking into account their location, seasonality, among other aspects. In this work, the sap of J. mollissima was analyzed using Near Infrared (NIR) and Medium Infrared (MID) spectroscopy associated with chemometrics. For the exploratory analysis, principal component analysis (PCA) was used, in which the tendency of separation between the samples collected in different locations was visualized. Next, we set out to build supervised classification models, using partial least squares regression with discriminant analysis (PLS-DA). The constructed models presented satisfactory results, indicating classification error rates, for the MID spectra of the liquid sap of the three locations, of 0% for the training sets and 1.4%, 0% and 0% for the test set for regions A, B and C, respectively. For the powdered sap NIR spectra, the error rates were 2% for the A and B region saps and 0% for the C region, when used as a training set. And 3%, 4% and 0% for regions A, B and C, respectively, when used as a test set. It is noteworthy that the analyzes carried out do not destroy the samples, do not use chemical reagents or standards and require little or no handling of the samples.