Determinação direta e simultânea de sódio, potássio e cálcio em biodiesel por espectrometria de emissão atômica em chama

• Ano de defesa: 2013
• Autor(a) principal: Conny Cerai Ferreira
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: análise direta; FAES; metais; Biodiesel; Química analítica; Espectroscopia de emissão atômica; Metais; Espectroscopia de chama
• Link de acesso: http://hdl.handle.net/1843/SFSA-96PTJX

We have developed an analytical procedure for the quantification of Na, K and Ca in biodiesel samples by direct analysis by Flame Atomic Emission Spectrometry (FAES) using a continuous emission detector. Were shown two possible methods of calibration were avaliable, the first is external calibration and the second external calibration using an internal standard to correct possible effects of transport of solutions. The calibration solutions were prepared by solubilizing organometallic standards in methyl oleate, wich was used as mock matrix. Chemical and spectral interferences were investigated in the development of the procedure. In addition, an extensive work to optimize and improve the operation of the equipment and procedure. The procedure was partially validated according to the instructions of the Validation Manual MAPA, following a rigorous statistical treatment of the data. Spreadsheets were developed for processing the data of the calibration curve by Method of Weighted Least Squares (MWLS) and to estimate the limits of detection and quantification through Prediction Limit Curves. The procedure showed good linearity (R > 0.99), high sensitivity, low limits of detection and quantification for all analytes. A small matrix effect was observaed. It may result from chemical interference due to the presence of other elements in organometallic multielement standard used in the calibration curves. The accuracy was assessed by comparing the results of quantitation of analytes in samples and the procedure proposed by the standard procedure ABNT NBR 15556 through hypothesis tests at 95% confidence level, shawed that results obtained by both procedures are statistically equal. The repeatability was adequate and reproducibility was impaired due to the instability of the equipment.