Direct chemical inspection of eye shadow and lipstick solid samples using laser-induced breakdown spectroscopy (LIBS) and chemometrics: proposition of classification models

This study proposes classification models focusing on the identification of contaminated (high concentration of Cd, Co, Cr and Ni) and non-contaminated eye shadow and lipstick samples. The analytical procedure was developed by combining emission spectral information obtained by laser-induced breakdown spectroscopy (LIBS) and reference concentration values from inductively coupled plasma optical emission spectrometry (ICP OES). Chemometric tools, such as principal component analysis (PCA), soft independent modelling of class analogies (SIMCA), K-nearest neighbor and partial least squares for discriminant analysis (PLS-DA), were used to perform exploratory analyses (PCA) and to calculate classification models (KNN, SIMCA and PLS-DA). Through this study, it was possible to classify the samples according to the values of Cd, Co, Cr and Ni below and above the tolerance levels indicated by several legislations and the literature (an average and consensus value of 5 mg kg(-1) was used). A total of 46 samples intended to be used by adults and children from three different countries were inspected by LIBS and analyzed by ICP OES (after acid digestion), and 16 samples had at least one of the elements with a concentration above 5 mg kg(-1). The most informative spectral range was from 186 to 580 nm, and the proposed classification models presented good results with 98% correct predictions for KNN. No chemical differences were observed between the samples intended to be used by adults and children. (AU)