Comparison of the solid sampling techniques laser ablation-ICP-MS, glow discharge-MS and spark-OES for the determination of platinum group metals in Pb buttons obtained by fire assay of platiniferous ores

Abstract

This work reports on the performance of a new Pb fire assay procedure for the extraction of Au, Cu, Ir, Ni, Pd, Pt, Rh and Ru from platiniferous ores, followed by the direct analysis of the Pb buttons thus obtained by means of three solid sampling techniques (spark-OES, LA-ICP-DRC-MS and GD-sector field-MS), therefore avoiding the process of cupellation. It can be concluded that this working scheme allows the reliable analysis of different ore materials, except for Ni and Ir, for which no quantitative extraction was achieved. The preparation of the Pb buttons using large masses of the ores permitted us to obtain representative and homogeneous buttons, resulting in acceptable precision values (typically RSD values ranged between 2–5% for spark-OES and GD-MS, and between 4–10% for LA-ICP-MS), while the simple composition of these buttons (≥99% Pb) enabled the preparation of matrix-matched standards for calibration. In general terms, all these techniques perform in a satisfactory way, although they showed different strengths and weaknesses. While spark-OES provides the highest sample throughput, its detection limits are not always sufficiently low for some analytes (Au, Ir, Pt). It is demonstrated that these detection limits can be further decreased by using techniques based on mass spectrometry (LA-ICP-MS and, particularly, GD-MS), although the use of an instrumental configuration capable of dealing with spectral overlap is required, namely a dynamic reaction cell pressurized with NH₃ in the case of LA-quadrupole ICP-MS and a double-focusing sector field mass analyser in the case of GD-MS.