Pre-concentration at crushing sizes for low-grade ores processing – Ore macro texture characterization and liberation assessment

• Published in: Minerals engineering Vol. 147; p. 106156
• Main Authors: Sousa, Rui, Futuro, Aurora, Fiúza, António, Leite, Mário Machado
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.03.2020
• Subjects: Mineral liberation; Ore sorting; Pre-concentration; Separation efficiency; Mineral liberation; Separation efficiency; Pre-concentration; Ore sorting
• ISSN: 0892-6875; 1872-9444
• DOI: 10.1016/j.mineng.2019.106156

•Mineral liberation analysis at coarse crushing sizes using image analysis procedures.•Assessment of Pre-concentration feasibility using Ultimate Upgrading Curves.•Technical evaluation of an Optical Sorter applied to a Li-mica lepidolite ore. Mineral processing of low-grade ores requires high processing flow rates. Pre-concentration is a clever technological solution for this challenge because the rejection of barren rock, in the early stages of the processing flowsheet, leads to increasing the head grade of the concentration plant and reducing the corresponding input flow rate. As pre-concentration at crushing sizes depends on the ore texture and consequently, it relies on the achieved mineral liberation degree, a straightforward methodology to assess quantitative mineralogical data of the ore macro-texture, based on image analysis, was developed. For the purpose, photos of hand samples collected at the mine site were converted into a digital image representative of the ore macro-texture. Then, a random comminution algorithm simulating a coarse crushing stage, to generate particles in size range 19.3/6.7 mm was applied. The surface grade of each particle was calculated by pixel counting, allowing for the computation of the Grade Histograms (wt%) and then for the construction of ultimate upgrading descriptors, which has been proved to be a useful tool to assess pre-concentration feasibility and the separation efficiency. This methodology was applied to a Li-Mica lepidolite ore from Alvarrões deposit (Portugal) by simulating several separation scenarios. The separation efficiency of an optical sorting separation was also assessed using the proposed methodology. Furthermore, the influence of mineral liberation at different size ranges on the global separation efficiency was predicted. This paper intends to underline the importance of quantitative mineralogical data in the study of the pre-concentration process. Data acquisition must be improved, taking advantage of the more accurate and faster systems, such as the modern multispectral analytical devices and ore sorting technologies.