Data fusion of laser-induced breakdown and Raman spectroscopies: Enhancing clay mineral identification

Abstract

A straight-forward, data-driven approach for the reliable identification of clay minerals based on spectroscopy and multivariate analyses is presented here. No other group of inorganic materials have so many species, exhibit such a range of physicochemical properties, or enjoy a greater diversity of practical applications as the clay minerals. The emerging role of clays in a variety of pioneering disciplines, such as the pharmaceutical and astrobiological sciences, highlights their broad-ranging significance in natural science and engineering efforts. However, the highly variable chemical compositions and defect-rich structures of the clay minerals pose difficulties in their classification and identification. We introduce a new methodological approach which uses Raman and laser-induced breakdown spectroscopies (LIBS) to discriminate geological specimens based on their dominant clay mineralogy. Raman and LIBS provide complementary information about the molecular structure and elemental composition of an interrogated target, and, when considered simultaneously, contribute to a more comprehensive characterization of the system under study. What distinguishes this work from previous spectral investigations of clay mineralogy is the way in which the spectra are pre-processed and combined before analysis. Raman and LIBS data were collected from various clay-rich specimens and subsequently concatenated into a single data matrix to serve as a unique identifier of specimen composition – an approach known as low-level data fusion. Multivariate statistical analyses were used to identify mineralogical groups and to discriminate specimens based on their compositional similarities and differences. We evaluated the discrimination achieved by the fused data sets compared to that obtained by standalone use of Raman and LIBS data. Our results show that the use of data fusion strategies improved the discrimination model and allowed correct classification of all the samples based on their dominant clay mineralogy.