Comminution Characterization of Lithium Bearing –Mica Ores from Different Devices

The present paper highlights on the comminution and liberation properties of two greisen-type lithium bearing-mica ore blends (L1, L2) subjected to different breakage devices; cone crusher (CC), roller crusher (RC), rotor beater mill (RBM) and a screen mill (SM). The particle size distribution (PSD) of the products from each device were evaluated to search for an appropriate PSD model using Gates-Gaudin-Schuhmann (GGS) and Rosin-Rammler (RR) functions. In order to determine an appropriate function that could better describe the PSDs from the devices used, the coefficient of determination (R2) was used as a criterion. In terms of mineral liberation studies, due to costs, only products from rotor beater mill (RBM) were examined through an automated scanning electron microscope (SEM) technique. It was found that RBM, RC and SM products were better described by the RR model than the GGS model with higher R2 values (0.97-1.0). However, cone crusher products for L1 were better described by GSS model, while that for L2 were better described by RR model. In terms of the spread of size distribution as indicated by RR model parameters, RC products were more uniformly distributed compared to products from other devices, for both ore blends. Also the RBM products were more scattered than products from other devices. The results indicate that the composition of individual ore blends affected the comminution products PSDs since different PSD model parameters were obtained for samples comminuted by same devices. The modal mineralogy of the two ore blends indicate both to be rich in quartz, topaz, zinnwaldite and muscovite. Further, the result indicate that, for both ore blends, zinnwaldite phase is more enriched in the fraction < 250 µm. Moreover, better liberation of zinnwaldite is observed for L1 compared to L2. This could be explained by differences of the two ore blends in three aspects; the nature of mineral association, reduction ratio of the fractions analysed and the nature of spread of the size distribution.