Modelling of Complex Flow Sheets

Complex multi-element ores, like the skarn ores from the AFK project and the Tellerhäuser pilot plant, are difficult to process due to their complex and fine-grained mineralogy. The valuable elements are enriched in various scales. The deposit hosts a very large, high-grade tin orebody, which is challenging to process because of the high energy consumption during grinding and the very fine-grained tin mineralisation. Traditional way of processing such ores would consist of a milling to desired liberation size, followed by a separation of cassiterite based on density or floatability properties. As it has been done historically due to the specific properties of the ore this approach however requires a selection between very intensive grinding with lots of fines, or less grinding with insufficient liberation, both approaches essentially leading to low final concentrate grades.
Unlike many other deposits, the Tellerhäuser skarn ore has a much more complex structure, which can be exploited with more advanced flow sheets. Additionally to the tin mineralization there is a substantial enrichment in multiple potential by-products. The fine-grained tin mineralization itself is locatedin lithologies (units of consistent geometallurgical, mineralogical and physical characteristics) dominated by tin free minerals. Despite a background concentration of unrecoverable tin as a trace element in the whole skarn body, the cassiterite mineralisation is localized within the ore body at mining block scale. It is thus possible to reject gangue material at various scales: Exploration based on mineralogical information allows distinguishing processable ore from gangue. A separation based on mineral groups allows to reject unmineralized skarns at various scales and to enrich preconcentrates for various by-products before cassiterite is liberated.
Applied in the right way, automated mineralogy data allows characterizing the spatial and mineralogical structure and physical properties of particles at various scales from mining blocks (selective processing), through cm-scale (sensor sorting), sub mm-scale (physical processing) down to µm scale (ultra-fine processing). This allows the prediction of potential separation behaviour of complex processing chains and thus to infer optimal separation criteria, separation thresholds, milling targets, mass streams, savings potential and environmental properties of all processing steps from mine to final concentrate.
In this way, the detailed understanding of the deposit and ore structure allows to model the different processing steps in an optimal scale and detail, combined to one flowsheet. The effect of imperfect processing behaviour can be quantified and understood in particle-level detail and used to determine suitable processing equipment.