Multivariate approach to the chemical mapping of uranium in sandstone-hosted uranium ores analyzed using double pulse Laser-Induced Breakdown Spectroscopy

Abstract

The goal of this work is to provide high resolution mapping of uranium in sandstone-hosted uranium ores using Laser-Induced Breakdown Spectroscopy (LIBS) technique. In order to obtain chemical image with highest possible spatial resolution, LIBS system in orthogonal double pulse (DP LIBS) arrangement was employed. Owing to this experimental arrangement the spot size of 50μm in diameter resulting in lateral resolution of 100μm was reached. Despite the increase in signal intensity in DP LIBS modification, the detection of uranium is challenging. The main cause is the high density of uranium spectral lines, which together with broadening of LIBS spectral lines overreaches the resolution of commonly used spectrometers. It results in increased overall background radiation with only few distinguishable uranium lines. Three different approaches in the LIBS data treatment for the uranium detection were utilized: i) spectral line intensity, ii) region of apparent background and iii) multivariate data analysis. By utilizing multivariate statistical methods, a specific specimen features (in our case uranium content) were revealed by processing complete spectral information obtained from broadband echelle spectrograph. Our results are in a good agreement with conventional approaches such as line fitting and show new possibilities of processing spectral data in mapping. As a reference technique to LIBS was employed X-ray Fluorescence (XRF). The XRF chemical images used in this paper have lower resolution (approximately 1–2mm per image point), nevertheless the elemental distribution is apparent and corresponds to presented LIBS experiments.