Chemical inspection and elemental analysis of electronic waste using data fusion - Application of complementary spectroanalytical techniques

Daniel Fernandes Andrade; Eduardo de Almeida; Hudson Wallace Pereira de Carvalho; Edenir Rodrigues Pereira-Filho; Dulasiri Amarasiriwardena

doi:10.1016/j.talanta.2020.122025

Chemical inspection and elemental analysis of electronic waste using data fusion - Application of complementary spectroanalytical techniques

Talanta. 2021 Apr 1:225:122025. doi: 10.1016/j.talanta.2020.122025. Epub 2020 Dec 17.

Authors

Daniel Fernandes Andrade¹, Eduardo de Almeida², Hudson Wallace Pereira de Carvalho², Edenir Rodrigues Pereira-Filho³, Dulasiri Amarasiriwardena⁴

Affiliations

¹ School of Natural Sciences, Hampshire College, Amherst, MA 01002-3359, USA; Group of Applied Instrumental Analysis, Department of Chemistry, Federal University of São Carlos, Rod Washington Luiz, Km 235, BR-13565905, São Carlos, SP, Brazil.
² Laboratory of Nuclear Instrumentation, Center for Nuclear Energy in Agriculture, University of São Paulo, Av. Centenário, 303, BR-13416000, Piracicaba, SP, Brazil.
³ Group of Applied Instrumental Analysis, Department of Chemistry, Federal University of São Carlos, Rod Washington Luiz, Km 235, BR-13565905, São Carlos, SP, Brazil.
⁴ School of Natural Sciences, Hampshire College, Amherst, MA 01002-3359, USA. Electronic address: daNS@hampshire.edu.

PMID: 33592755
DOI: 10.1016/j.talanta.2020.122025

Abstract

This study is focused on the development of analytical methods for characterization of printed circuit boards (PCBs) from mobile phones by direct analysis using three complementary spectroanalytical techniques: laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS), laser-induced breakdown spectroscopy (LIBS), and micro X-ray fluorescence spectroscopy (micro-XRF). These techniques were combined with principal component analysis (PCA) to investigate the chemical composition on the surface and depth profiling of PCB samples. The spatial distribution of important base metals (e.g. Al, Au, Ba, Cu, Fe, Mg, Ni, Zn), toxic elements (e.g. Cd, Cr, Pb) as well as the non-metallic fraction (e.g. P, S and Si) from conductive tracks, solder mask and integrated components were detected within the PCB samples. Univariate and multivariate approaches were also performed to obtain calibration models for Cu determination. The results were compared to reference concentrations obtained by inductively coupled plasma-optical emission spectrometry (ICP-OES) after microwave-assisted acid leaching using aqua regia. To this end, two PCB samples (50 × 34 mm²) were cut into small parts of 40 subsamples (10 × 8.5 mm²) and analyzed by ICP-OES and the Cu concentrations ranged from 13 to 45% m m^-1. Partial least squares (PLS) regression was used to data fusion of analytical information from LIBS and micro-XRF analysis. The proposed calibration methods for LIBS and micro-XRF were tested for the 40 PCB subsamples, in which the best results were obtained combining both data sources though a low-level data fusion. Root mean square error of cross validation (RMSEC) and recoveries were 3.23% m m^-1 and 81-119% using leave-one-out cross validation.

Keywords: Atomic spectrometry; Chemometrics; Data fusion; E-waste; Elemental mapping.