Rapid classification and quantification of gold–silver alloys for identification from e-waste using chemometrics-assisted laser-induced breakdown spectroscopy

A global surge in the use of electronic devices has caused a substantial accumulation of electronic waste (e-waste). Revitalizing the circular economy through the exploration of e-waste recycling approaches is imperative in this technological era. Exploration of methods for the assessment of metallic concentration in e-waste has high demand. In this work, a pathway for extracting the analytical information of alloys from calibration-free laser-induced breakdown spectroscopy (CF-LIBS) and with the assistance of chemometric methods has been presented to show the potential of these methods for the fast classification and quantification of alloys. Six types of gold–silver-based samples were taken in order to estimate their concentration by calibration-free technique and calibration curve-based chemometric methods using their LIBS spectra. The distinctive emission lines in these LIBS spectra were observed by a Q-switched Nd:YAG laser working at 532 nm with 5 ns pulse width. The plasma parameters such as temperature and electron number density were assessed by the Boltzmann plot technique and Stark broadening method. It was verified that the plasma was optically thin and in compliance with the local thermodynamic equilibrium (LTE) condition for the CF-LIBS method. The chemometric approaches principal component analysis (PCA) and partial least square analysis (PLS) were applied for building the classification models of data. Area normalization preprocessing of the data was performed. The statistical measures obtained from the PLSR model having R 2 and RMSE for gold are 0.98 and 5.06, while these factors corresponding to silver are 0.98 and 4.56, respectively. The results attained from the PLSR model were compared with that of CF-LIBS. The PLSR technique has shown high accuracy in classifying and predicting the composition of gold–silver alloys based on the spectral data. Overall, the results indicate that LIBS in combination with chemometric methods can be a precise and reliable technique for the discrimination and quantification of alloys and hence it has high potential as a portable technique in the process of evaluating alloys for sorting and recycling e-waste.