The combination of data yielded by laser-induced breakdown spectroscopy (LIBS) and laser-induced plasma acoustics (LIPAc) is a topic of many prospective applications as these coexisting phenomena can cover different sample traits. Among the most interesting features that LIPAc could add to the expanded target picture is information concerning structure and geophysical characteristics elusive to LIBS. In the present work, frequency spectra of minerals were explored to discriminate between chemically similar mineralogical phases. Several replicas of four different Fe-based minerals were analyzed to identify spectral traits linked to their chemistry in the frequency domain. First, the similarity between replicas of the same mineral family was verified and then, the cosine and Euclidian distances to minerals of different species were calculated to evaluate the discrimination capabilities of frequency spectra with results being compared to those obtained by LIBS. A partial least-squares one-vs-all model is described seeking to demonstrate sample classification by frequency means exclusively. As the use of LIBS-LIPAc for in-field mineral sorting has sparked interest, experiments reported were performed in stand-off within a thermal vacuum chamber (TVC). The TVC allowed data acquisition under Earth and Mars-like conditions, with the latter serving as a test of high relevance to assess the general applicability of the conclusions reached in Earth environment. Thorough discussion of data treatment is included with a focus on the impact of interference patterns arising from the laser-induced shockwave interaction with the medium surrounding the sample to avoid non-sample related information in the data processing schemes.
Keywords: Frequency domain; LIBS; Laser-induced acoustics; Mars; Minerals; PLS.
Copyright © 2022 The Authors. Published by Elsevier B.V. All rights reserved.