A fast and reliable approach for asbestos recognition in complex matrices adopting an hyperspectral imaging based approach

In this paper the possibility to implement fast, low cost and reliable strategies addressed to recognize the presence of asbestos inside cement matrices of different characteristics was investigated. The use of innovative sensing techniques could represent an important step forward to prevent an “uncontrolled” diffusion of the asbestos in the environment due to the lack of knowledge of the “real” composition of a product, that has to be removed, demolished or handled in some way. HyperSpectral Imaging (HSI) and micro X-Ray Fluorescence (micro-XRF) were applied in order to reach the objective. Selected samples of reference asbestos minerals (chrysotile, crocidolite, tremolite and amosite) and cement-asbestos of different origin were analysed by HSI in the short wave infrared range (1000-2500 nm). The acquired data were processed applying chemometric techniques, such as Principal Component Analysis (PCA), Partial Least Square Discriminant Analysis (PLS-DA) and Soft Independent Modeling of Class Analogies (SIMCA) classification. Micro-XRF maps, on the same cement-asbestos materials, were also acquired and analyzed using a BRUKER Tornado M4, in order to validate the classification results obtained by HSI. The proposed approach, combining the potentialities of the HSI sensing devices and chemometric processing tools, allows to perform a direct detection of asbestos inside different matrices, without any physical sampling action and without following a human knowledge based characterization, not requiring the presence of an expert and the related “classical” analytical devices. The development and utilization of in situ detection techniques could thus represent a useful solution for the identification of materials and/or products containing asbestos.