Plastic Identification and Comparison by Multivariate Techniques with Laser-Induced Breakdown Spectroscopy

The classification of plastics is very important in the recycling industry. A quick online classification allows the installation of the equipment in this line of work. Whether qualitative or quantitative analysis, the basic component of any laser-induced breakdown spectroscopy (LIBS) measurement is the emission spectrum recorder from single plasma. Each fire of the laser atomizes a portion of the sample in the pulse focal volume and produces plasma that excites and re-excites the atoms to emit light. The plasma light is collected and recorded in an ensuing measurement. In this sense, the LIBS technique offers all possible advantages: speed, the possibility of online analysis, nondestructive analysis, and so on. In this article, we discuss details related to the analysis of the emission spectrum. The plastics used in this study were low-density polyethylene (PE), high-density PE, polypropylene, polystyrene, and poly(ethylene terephthalate). Hierarchical cluster analysis was proven to be the best method because the four polymers could be divided into two clusters, which allowed their identification and classification in a fast and easy way that could be carried out with commercial software and could be implanted online in a recycling factory, as conventional data analysis techniques are limited to the qualitative identification and calculation of elemental abundances. Principal component analysis on LIBS spectra can be used to better describe the chemical variations in the samples and to extract a greater understanding of the chemical structure. (C) 2011 Wiley Periodicals, Inc. J Appl Polym Sci 121: 2710-2716, 2011