Rare-metal granites as a potential source of critical metals: A geometallurgical case study

This is the author accepted manuscript. The final version is available from Elsevier via the DOI in this recordBecause of their low grades in critical metals such as Light Rare Earth Elements (LREE) or Sn, rare-metal granites are not considered as economic for metal recovery but, when altered, they are often exploited for their industrial minerals. The St Austell rare-metal granite is well known for its world-class kaolin deposits which formed as a result of the extensive weathering and alteration of the underlying granite. The St Austell granite body is composed of several granite components, each having its own accessory minerals assemblage. As a result of the kaolinisation process, some metal-bearing accessory minerals of the granite, such as monazite (LREE) or cassiterite (Sn), are partially liberated from the gangue which allow their pre-concentration in the micaceous residue which is considered as a potential source for critical metals recovery. Similarities with other similar rare-metal granites suggest that topaz granite is the most prospective for disseminated magmatic Sn-Nb-Ta-REE mineralization. However, comparison of the potentiality of 3 granite types i.e., biotite, topaz and tourmaline granites suggest that biotite granites is actually the most prospective due to higher degree of kaolinisation of the biotite granite which favour pre-concentration of its accessory mineral in the micaceous residue. In order to develop a geometallurgical framework for extraction of kaolin and metals from the selected granite component, a field sampling campaign is performed. Core samples are processed in the laboratory using a characterisation program that mimics the full-scale kaolin refining route. Two main products are recovered through this program, viz. MR180 (−180 +53 µm) and P5 (−5 µm), which correspond to a fine micaceous residue and a fine kaolin product respectively. These products are both analysed routinely for major and minor trace elements by XRF and yields are recorded to indicate process performance. A selected number of MR180 samples are also being characterised in terms of particle size by laser light scattering, geochemistry by ICP-MS, and mineralogy by QEMSCAN®. Comparison of characterisation results of MR180 samples and corresponding industrial residue samples shows a good correlation, suggesting that sample analyses are representative for the in-situ deposit and the processing behaviour. Monazite is found to be either fully liberated or fully locked from one sample to the other. Next, pilot-scale gravity concentration tests are performed on micaceous residue samples. Characterisation of the processing products shows that monazite lost in the tailings is mostly locked within tourmaline or micas and is fine grained. Then, predictive regression models for spiral separation performance in terms of recovery, product grade and enrichment as a function of the feed grade are developed for MR180 LREE grade data. Finally, kaolin resources can be classified using quantitative indicators such as yield of the P5 product and the iron oxides content which provides insight into the kaolin quality in terms of whiteness. This geometallurgical classification can be used to delineate zones of interest within the deposit. Although kaolin quality and recovery primarily inform extraction planning, zones which are also of interest for metal recovery can be identified. The proposed model predicts whether the expected LREE grade and recovery satisfy the by-product requirements.European CommissionNatural Environment Research Council (NERC)French National Research Agenc