Laser-Induced Breakdown Spectroscopy Assisted by Machine Learning for Plastics/Polymers Identification

Abstract

In the present work, Laser-Induced Breakdown Spectroscopy (LIBS) is used for the discrimination/identification of different plastic/polymeric samples having the same polymeric matrix but containing different additives (as e.g., fillers, flame retardants, etc.). For the classification of the different plastic samples, some machine learning algorithms were employed for the analysis of the LIBS spectroscopic data, such as the Principal Component Analysis (PCA) and the Linear Discriminant Analysis (LDA). The combination of LIBS technique with these machine learning algorithmic approaches, in particular the latter, provided excellent classification results, achieving identification accuracies as high as 100%. It seems that machine learning paves the way towards the application of LIBS technique for identification/discrimination issues of plastics and polymers and eventually of other classes of organic materials. Machine learning assisted LIBS can be a simple to use, efficient and powerful tool for sorting and recycling purposes.