2026-05-05T22:57:19Zhttps://uvadoc.uva.es/oai/request

oai:uvadoc.uva.es:10324/352582025-03-26T19:10:04Zcom_10324_1158com_10324_931com_10324_894col_10324_1242

2019-03-26T17:42:49Z urn:hdl:10324/35258 Analytical methodology to elemental quantification of weathered terrestrial analogues to meteorites using a portable Laser-Induced Breakdown Spectroscopy (LIBS) instrument and Partial Least Squares (PLS) as multivariate calibration technique Gómez Nubla, Leticia Aramendia, Julene Fernández Ortiz de Vallejuelo, Silvia Madariaga Mota, Juan Manuel The present work is focused on the in situ quantitative analysis of Si, Al,Mg, Ca, Ba, Na, and Fe, present in weathered terrestrial analogues to meteorites (black steel slag and impact glasses), using a portable Laser Induced Breakdown Spectroscopy (LIBS) instrument. For that purpose, several standards pellets of known elemental concentrations were manufactured. The elemental and molecular homogeneity of the pellets was studied by means of Scanning Electron Microscopy coupled to Energy Dispersive X-ray spectroscopy (SEM-EDS) and Raman spectroscopy. This checkwas always made before the LIBS analysis. Univariate andmultivariate (Partial Least Squares (PLS) regression) calibration approaches on LIBS spectra were selected as initial calibration models. After a comparison between both approaches, the former was discarded due to the poor linearity of the calibration curves, and PLS regressionwas chosen to treat the LIBS spectra as themultivariate calibration approach (in the ultraviolet (UV) and infrared (IR) spectral ranges). Predictive capabilities of each calibration model were evaluated by calculating regression coefficient (r), number of PLS factors (rank), relative errors of cross validation (RMSECV), residual predictive deviation (RPD) and the Bias value. At the end, the simultaneous use of both ranges of wavelengths was demonstrated to be more fruitful rather than using the individual ones, probably due to the higher number of emission lines, number of spectral variables and the PLS latent variables for each element. Moreover, a Reference Material was used as external validation, obtaining satisfactory results in the determination of elements. The predictive ability of the PLSmodelswas evaluated on samples of Darwin Glasses (Australia), Libyan Desert Glasses (Western Desert of Egypt) and black steel slag residues (steelworks of Basque Country). The obtained results were in concordance with the range of composition measured also by X-ray Fluorescence Spectrometer (ED-XRF). Our methodology is a good, rapid, simple and cost-effective alternative for in situ analysis of these terrestrial analogues over other techniques. 2019-03-26T17:42:49Z 2019-03-26T17:42:49Z 2018 info:eu-repo/semantics/article Microchemical Journal, 2018, vol. 137. p. 392–401 0026-265X http://uvadoc.uva.es/handle/10324/35258 10.1016/j.microc.2017.11.019 eng https://www.sciencedirect.com/science/article/pii/S0026265X17310524?via%3Dihub info:eu-repo/semantics/openAccess © 2018 Elsevier Elsevier