Multi-element analysis of iron ore pellets by laser-induced breakdown spectroscopy and principal components regression

Laser-induced Breakdown Spectroscopy (LIBS) in combination with Principal Components Regression (PCR) has been applied to determine the elemental composition of a series of run-of-mine (ROM) iron ore samples. The samples were presented for measurement both as compressed pellets and as loose chipped material. The present paper details the results of the measurements of the compressed pellets. Results from ore chips will be reported separately. LIBS spectral data was recorded in three separate spectral regions to measure major, minor and trace components of the iron ore sample pellets. Background stripping, normalization and spectral cleaning were applied to minimize the relative standard deviations of the LIBS data. PCR analysis was then applied to produce calibration models for iron, aluminum, silicon, manganese, potassium and phosphorous. These calibration models were then validated using independent LIBS measurements. Robust calibration models were determined for iron, aluminum, silicon and potassium, whilst the results for manganese were encouraging. Phosphorous, present at low levels in the ores measured, remained the most difficult element to determine accurately. The combination of LIES and PCR shows potential for in-situ on-line determination of ore composition. Crown Copyright (C) 2008 Published by Elsevier B.V. All rights reserved.