Abstract
Plastic recycling is mainly limited by their sorting as their natures, forms and formulation are very numerous and most of them are strongly incompatible, leading to poor mechanical properties. Several industrial sorting technologies exist, and others are in development. However, each of them has drawbacks. Especially, NIR-HSI (Near-Infrared Hyperspectral Imagery) is limited by the use of carbon black, mainly as a pigment and UV agent in the case of thermoplastics. MIR-HSI (Mid-Infrared) could be a suitable and viable alternative to resolve this issue. Hence, this work, based on laboratory FTIR-ATR (Fourier-Transform Infrared Attenuated Total Reflection), focuses on possible sources of spectral alteration, which could impair identification of usual polymers using industrial MIR-HSI. It aims to help simple and rapid laboratory characterization and give tools to avoid misidentification or enable specific segregation during industrial sorting. First, acquisition parameters were degraded to simulate those imposed by industrial conditions: short acquisition time, diminished resolution and blank defaults. Then, impact on formulations of usual WEEE (Waste of Electric and Electrical Equipment) plastics were evaluated, with PE, PP, ABS and HIPS as matrices, and carbon black (at different concentrations), calcite, talc, titanium oxide and some flame retardants as additives. Several patterns found in homemade standard samples were recognized within a stock of about one hundred of real waste samples.