Emprego de imagens hiperspectrais e espectrometria de emissão em plasma introduzido por laser na análise direta de amostras de lixo eletrônico

Abstract

In this study, laser induced breakdown spectroscopy was combined with chemometric tools to study the chemical composition of WEEE discarded. For this, three samples of printed circuit board were chosen for analysis: PCBs samples of mobile phone, mouse and memory stick. The first sample with an area of 30 x 40 mm was forwarded to the LIBS, laser pulses were made in 1200 points generating a total of 12,000 spectra. Normalizing the data, initial exploratory analysis was conducted in which 18 emission lines were separated corresponding to Al, Au, Ba, Ca, Co, Cu, Fe, K, Li, Mg, Mn, Na, Ni, Sb, Si, Sn Ti and Zn. Semi quantitative information were acquired with scanning electron microscopy with X-ray microanalysis. In a second step, the data were normalized by the relative intensity and self-scaled, where principal component analysis was performed. Scores maps were generated with PC1 containing 19 and 16% of variance explained by the pulses 1 and 10. In the mouse sample, a small area containing 5 x 15 mm was analyzed, and investigated the presence of Pb, the five more intense emission lines of Pb were separated. Data were normalized by the relative intensity and self scaled, PCA was used and scores of maps were generated, with PC1 containing 88% of the explained variance. The Pb element was quantified by Flame Atomic Absorption Spectrometry finding a con-centration of 25% (w/w). Small areas of 10 x 10 mm were analyzed in the memory stick using laser ablation inductively coupled plasma mass spectrometer. Data were normalized, scores were generated and maps the elements Cr and Pb were identified and quantified by Inductively coupled plasma optical emission spectrometry, with concentrations of 7 and 70 mg/kg, respectively