Non-Destructive Assay of Gold and Other Metals in Bulk Mineral Samples

"Gold is conventionally analyzed using fire-assay or other destructive chemical methods that impose significant sample preparation, OH&S, and skilled labour requirements. Developed over the past 15 years by the CSIRO, PhotonAssay provides a new approach. A high intensity, high-energy x-ray source is used to activate atoms of gold and other elements in bulk mineral samples. The resulting activity is counted to automatically, accurately and rapidly determine metal grade. Samples weighing 0.5–1.0 kg can be measured in a few minutes with minimal preparation. As the method is matrix-agnostic, traditionally hard-to-assay materials such as highly refractory ores, carbon pulps and copper concentrates require no special treatment. We present the status of commercial PhotonAssay systems being developed by Chrysos Corporation. The first unit with a throughput of up to 80 samples per hour is to be deployed in an Australian assay laboratory in late 2017; smaller systems suitable for providing real-time assay directly at the mine-site will be progressively rolled out from mid-2018. Performance results from extensive testing using both mine-sourced and standard reference materials are presented, demonstrating the sensitivity and accuracy of the method. Implications for replacing conventional gold assay in exploration, mining and processing operations are discussed.INTRODUCTION Accurate methods for assaying ore grade are essential throughout the exploration, mining and mineral processing value chain. The discovery and evaluation of new ore deposits, reserve and resource definition, mine operations planning, ore processing, metal accounting and product quality evaluation all rely on accurate measurement of the valuable metal content of bulk materials. The analysis of gold presents particular challenges, stemming from the very low concentrations at which commercial operations run (often part-per-million (ppm), or equivalently one gram-per-tonne levels or below), and the highly heterogeneous nature of many gold deposits. These challenges mean that any analytical method must have high sensitivity and be capable of measuring large-mass samples. The most widely used quantitative analytical method for gold is fire-assay (Hoffman, Clark and Yeager 1998). An initial ore sample typically weighing several kilograms is dried, crushed and pulverized to allow a small sub-sample (10–50 g is common) to be representatively drawn. The sub-sample is mixed with flux, litharge (lead oxide), a reducing compound and other reagents. The mixture is fired at a temperature of 1000–1200°C until molten. After cooling, the slag is removed and the lead button containing any gold present in the sample is remelted in a porous cupel. The lead is absorbed into the cupel, leaving a gold prill which is either weighed directly, or dissolved in acid and analysed via atomic absorption spectroscopy (AAS) or inductively coupled plasma spectroscopy (ICP or ICP/MS)."