Determinação de elementos traços cobre e estrôncio em piroxenitos por espectrofotometria de absorção atômica com atomização electrotérmica

Abstract

The present work summarizes the efforts of the Department of Chemistry, Faculty of Sciences, University Eduardo Mondlane, in introducing adequate methods for the analysis of geological materials. This work presents the optimization of instrumental parameters for the determination of Cu and Sr in pyroxenite samples by atomic absorption spectrophotometry with electrothermal atomization using the standard addition method. The ten samples analyzed were selected from a group consisting of various types of rocks collected in the province of Niassa, specifically in the Ngoo region. The study is of great importance since the results achieved will serve as a basis for the analysis of pyroxenites and other types of materials using similar equipment recently acquired at the National Directorate of Geology where the final year student is a worker. The optimization of the instrumental parameters was done using the reference material SARM 5 and the results obtained were statistically treated using the t test for comparison of an experimental mean with a known value at a confidence level of 95%, with agreement between the found contents and the known values. The contents found in the samples are between 5 and 84µg/g with detection limits between 1.0 and 8.6µg/l for Cu. For Sr, they ranged from 32 to 517µg/g with detection limits between 1.2 and 12.4µg/l. The accuracy of the method was controlled by comparing the results obtained with the known values ​​of the SARM 5 reference material, and it was found to be of good accuracy. The relative error of determination was 5.6% for Cu and 9.4 for Mr. In general, the method is very accurate for the determination of Cu. The relative standard deviations (%RSD) for the determination of this element were between 1.0 and 8.1. In general, for the determination of Sr the method presents low precision, with %RSD between 6.4 and 22.3 for ten samples (including the reference material). Of these samples, only three provided the %RSD below the maximum limit (10.0) for determinations with acceptable precision. (TRADUÇÃO NOSSA)