Ricerc@Sapienza

Significant amount of valuable materials are contained in waste from electrical and electronic equipment (WEEE), however they are not easily recovered due to the complexity and heterogeneity of such typology of products. For this reason, a careful recycling-oriented characterization must be the starting point for gathering information about the different elements and materials and their distribution and concentration.
The main target of this study was to investigate the use of hyperspectral imaging and micro X-ray fluorescence in order to characterize printed circuit boards (PCBs) with particular reference to copper content and distribution. The studied samples come from small and medium appliances by-products, obtained from a Magnetic Density Separation (MDS) process.
Samples are mainly constituted by PCBs, plastics and glass, with a size Hyperspectral images in the SWIR spectral range (1000-2500 nm), and micro-XRF maps of the analyzed bulk materials were acquired. A Partial Least Square – Discriminant Analysis (PLS-DA) was carried out on hyperspectral images to classify PCBs and the other materials (i.e. black plastics, white plastics, plastic wires, wood, and glass). PLS-DA was also applied on micro-XRF maps to classify copper on the PCBs surfaces. Promising results were obtained for the automatic classification of the different materials detected on the investigated samples, and in particular for the evaluation of the copper concentration and its topological assessment on the PCBs surfaces.