Laser ablation-inductively coupled plasma-dynamic reaction cell-mass spectrometry (LA-ICP-DRC-MS) for the determination of Pt, Pd and Rh in Pb buttons obtained by fire assay of platiniferous ores

Abstract

As a result of the low concentration in which the platinum group metals (PGMs) Pt, Pd and Rh occur in platiniferous ores and their heterogeneous distribution over the matrix, their determination is usually preceded by analyte pre-concentration and separation of the analyte from the matrix. The Pb fire assay technique is an elegant method for this purpose, whereby the elements of interest are collected from a relatively large sample mass in a Pb button. It was the aim of this work to evaluate the capabilities of laser ablation-ICP-mass spectrometry (LA-ICP-MS) for the analysis of such Pb buttons. Matrix-matched standards were produced by spiking lead ingots with an aqueous standard solution and subsequent re-melting of these ingots to obtain homogeneous solid standards with known PGM concentrations. These standards permitted quantitative results to be obtained. In standard (vented) mode, the determination of Rh by ICP-MS was hindered by spectral overlap of the 103Rh+ signal with that of 206Pb2+. Cool plasma conditions permitted reduction of the Pb2+ signal intensity, but since analyte signal intensities decreased as well, the improvement in the limit of detection (LOD) for Rh was insufficient to make this approach fit-for-purpose. Use of NH3 as a reaction gas (at a flow rate of 0.2 Ar-equivalent mL min−1) in a dynamic reaction cell (DRC) resulted in electron transfer to Pb2+, while Rh+ showed no reactivity towards this gas. The use of chemical resolution to overcome the Rh+/Pb2+ overlap did not interfere with the determination of the other PGMs of interest. LODs achievable with this approach were <100 ng g−1 for Pt, Pd and Rh. The capability of this approach was evaluated by analysing Pb buttons produced from actual flotation concentrate samples from platiniferous ores, originating from the Merensky reef and the UG-2 reef, both located in the South-African Bushveld Igneous Complex. The accuracy attainable using LA-ICP-DRC-MS was estimated by comparison of the results thus obtained with those calculated for the lead buttons based on NiS fire assay and subsequent ICP-OES analysis of the same flotation concentrate samples. No significant differences could be established and in the majority of cases, the deviation between the LA-ICP-DRC-MS result and the corresponding reference value was <10%. The precision was typically ∼10% relative standard deviation (RSD) for Pt and Rh. For Pd, however, the precision was observed to be worse (typically ∼15% RSD), probably as a result of a more heterogeneous distribution over the Pb matrix.