Quantitative analysis of P2O5 in solid raw phosphate matrices using laser-induced breakdown spectroscopy (LIBS) for online mining industry monitoring
Year: 2026
Type: article
Abstract: Laser-Induced Breakdown Spectroscopy (LIBS) is increasingly considered for real-time compositional monitoring in the mining industry, offering minimal sample preparation, multi-element capability, and sub-second analysis times. The growing global demand for fertilizers, combined with the progressive depletion of high-grade phosphate ore, highlights the need for fast, accurate, and reliable analytical methods for upstream process control. This work develops and validates a complete LIBS calibration protocol for the quantitative determination of \(\text {P}_{2}\hbox {O}_{5}\) in solid phosphate rocks. A set of 84 calibration samples, prepared by gravimetric dilution of NIST SRM 694, spanning 6.4–24.8 wt.% \(\text {P}_{2}\hbox {O}_{5}\) and characterized by ICP-OES was used to build quantitative models by systematically benchmarking 30 calibration configurations crossing six regression families, three variable-selection strategies, and two replicate-handling schemes, on spectra preprocessed by Savitzky–Golay smoothing combined with standard normal variate (SNV) normalization. The optimal configuration, based on partial least squares regression achieved a coefficient of determination \(R^{2}_{CV} = 0.949\) and a root-mean-square error \(RMSE_{CV} = 1.16\:\text{wt.}{\%}\:\) \(\text {P}_{2}\hbox {O}_{5}\) . These results establish a laboratory validated calibration baseline and suggest that LIBS, coupled with chemometric modeling, constitutes a credible candidate for real-time, upstream \(\text {P}_{2}\hbox {O}_{5}\) monitoring in phosphate-mining operations, pending validation on site under actual production conditions.
Source: Scientific Reports
Language: English
Subfield: Mechanics of Materials
Field: Engineering
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,690
OpenAlex ID: W7215917454
FWCI:
Cites:
Cited by:
Related to:
