- To participate in the 911Metallurgist Forums, be sure to JOIN & LOGIN
- Use Add New Topic to ask a New Question/Discussion about Geology and Geometallurgy.
- OR Select a Topic that Interests you.
- Use Add Reply = to Reply/Participate in a Topic/Discussion (most frequent).
Using Add Reply allows you to Attach Images or PDF files and provide a more complete input. - Use Add Comment = to comment on someone else’s Reply in an already active Topic/Discussion.
Implementing Geometallurgy (6 replies)
Agreed that some are making a meal out of geometallurgy. I thought all metallurgists were trained in recognizing that Lithology/geochemisrty/mineralization/alteration etc affected metallurgical response of process plants eg
Lithology has a big say in grindability and throughput, geochemistry affects pulp chemistry and tailings behavior/ARD, alteration affects geochemistry and mineralization etc.
Most of us are aware that the same minerals from different ore deposits can exhibit non-uniform response to similar metallurgical treatment. What progress is geometallurgy making in predicting the intrinsically variable response of same minerals from different ore-bodies to standard metallurgical processes?
Your requirements will dictate your needs. If you want to determine the hardness variability of your deposit to better forecast thoughput, mine plan, blending pattern, etc, then you must do grindability testing on discrete samples (no composites), as mineralogy alone wont help you.
To me any geometallurgy project implies both data integration and automation.
So I would suggest that at the very outset it is worthwhile ensuring data are stored in a relational database.
To ensure data can be analysed quickly various calculation methods should be automated.
While I see geomet. as a linkage of various disciplines( if I name them I am likely to miss one - so I am leaving it fairly generally).
My main interest is mineral processing; and simulation. However there is generally poor integration between lab. data, plant design and plant operation.
I do offer various services, but relevant to this discussion are 3 courses:
Advanced computational methods: Databases , Excel VBA, flowsheeting, automation
Basic simulation (via an integrated Excel/Visio general simulation), including mass balancing.
Advanced simulation. Advanced simulation covers issues not available through either Uni. courses or publications and focuses on a number of topics including information theory, using ore variability to advantage, mineral processing optimisation, mineral processing datastructures, advanced analysis of lab. data, identifying operational variables, customised simulation.
The three courses combined are mainly targeted at sophisticated Mining Companies who want to develop a highly skilled professional workforce, and focusing on increasing profitability.
Best to drop the "geo" label and refer to the long running practice of "ore body variability" assessment and analysis. Metallurgists and mineralogists have received the benefit of better tools over the past 20 years, specifically quantitative process mineralogy in the form of specialized scanning electron microscopy (ie: TIMA, QEMSCAN, MLA). The automation of running "scans" to large sample populations of mineralized zones, ore bodies, and the process stream was a step-change for the industry.
Some here are long experienced and note that the larger and more diverse the sample set, from ore body and process stream, the more reliable will be the results of a variability study.
I guess that presenting in-situ variability is what this subject is about. Maybe the Geomet label helps direct attention to describing variability in terms of ore processing parameters.
As pointed out, Mets would be trained to know that lithology and mineralisation have an effect on ore processing but I reckon that they may not have a decisive input in the planning and collection of resource data.
For me the Geo prefix represents the link to geophysics which can deliver in-situ measurements directly related to the separation forces used in metallurgical plant.
The problem of statistical representivity becomes apparent in geometallurgy programmes when test results, say on comminution are placed into the resource model.
The use of geophysical down-hole data can overcome this problem due to their output being continuous, of high resolution and known spatial position. Measures of density, porosity and magnetic susceptibility can then be expressed in terms of plant yield after calibration using suitable proxy tests. There are also strong indications that rock hardness can be described using this approach.
Francois touched on the subject of using discrete intervals as opposed to composite samples in tests to understand ore body variability. This pure form approach can present the opportunity to model process parameters throughout the deposit rather than be constrained by placing single outcomes from testing composites.
The outcome of this is a basis for a model that can adapt to the inevitable changes in the mining plan, the design basis for most composites.
So geophysics, to me offers so many solutions when applied to the in-mine situation. Although the inclusion of downhole surveys incurs only a modest increment to drilling costs, its application in Geomet is not often considered. I think we’re missing out here.
There are a couple of flavours of geometallurgy floating around the industry. The first type I would classify as "process mineralogy" where you associate plant performance to the minerals in the plant feed. The "geometallurgy" definition I use is "interpolating process parameters into mine block models". Examples are specific energy consumptions for grinding and R_max values for leaching or flotation. Oddly enough, this second variety is more valuable to the mining engineers than the metallurgists – the biggest value is in the mine and not the plant.
The "mine model" type of geometallurgy is used to optimize the pit shape so, for example, hard marginal grade blocks become waste and soft marginal grade blocks become ore. Building throughput and recovery estimates into a block model optimizing algorithm almost always drives the pit towards the softest material first putting off harder ore until later. Somewhat pointless to do this after a pit has been established; so it is most valuable to do the geomet sampling & interpolation five years ahead of the expected mining and let the process plant throughput and recovery help drive the pit evolution.
Extra reading for those interested, a talk I gave in Cajamarca on the subject: https://www.sagmilling.com/articles/11/view/?s=4
Am working on implementing a Geometallurgy program in a mining operation. We know that it will not be an easy task but it is a risk we need to address.
Understanding your mineralization and the distribution of metal carriers throughout your deposit is the first most important step. I was running geometallurgical studies for the company Noranda at a time when nobody knew about this field. I have been running geometallurgy programs on many different exploration and advance development projects. For the past two years I have organized and chair the technical session on Geometallurgy. Many people were doing 'geometallurgy' on their ore bodies long before some bright spark thought I can collect the data and fiddle with it on my computer and make a science of it, and I don't have to leave my office and go in to the dirty old plant. Mineralogical based variability will fundamentally drive your recovery. It as an important tool in optimising your metallurgy.