The paper my company recently presented at Physical Separation '15 addressed this very issue. Titled "Bridging the gap: Understanding the economic impact of ore sorting on a mineral processing circuit," this work was a follow up study to the work short-listed by the CEEC. In this study we studied the cost, revenue and CAPEX associated with installing ore sorters at a Cu operation in the southwest USA. We had some very interesting findings that demonstrated the real economic feasibility of the technology. This work has been submitted for publication in Minerals Engineering.

I feel that method of mining (open cast/underground) and ore formation will give a preliminary indication whether ore sorting would be feasible or not.

I'll look for the paper in upcoming Minerals Eng.

Avoiding dilution in mining is a form of gross level sorting.

I agree. Now can we identify whether we have any success stories where we avoided dilute ion in mining but the ore is lean.

At IMPC 2006 in Cape town I had the opportunity to give a plenary speak with the title: "Sensor sorting technology – is the minerals industry missing a chance? ". That time many mineral processing colleagues did not even know that such technology exists. I remember that I had said that sensor based sorting should not be considered as last possibility if all other technologies fail but it should be considered in the early stage of a project- not only in the design of the mineral processing plant, but of the whole integrated mining and processing system. Now, nine years later the situation begins to change. SBS turns from a niche application to -hopefully- a key technology.

Btw, "ore sorting" in my opinion this is not the correct term, as we also sort industrial minerals, gemstones, salt, coal, etc. 

I fully agree that rejection by any technique before treating the ore in a conventional beneficiation plant is a "must and highly profitable".

Just for my information, I wanted to know where it is used for low grade (say <2% cu in the r.o.m.) ore deposits.

I agree with you, it's been a long road to get the base and precious metal miners to be aware of SBS technology. It will be an even longer road to gain the investors trust that it is not a waste of funds to adopt SBS to build a smaller Mill, reduce the milling power costs, water consumption, size of the tailings and waste water facilities. What we need is a company like Phelps-Dodge which went with SX-EW technology for copper in the 1980's instead of bigger mines and mills, and survived. This time it could be SBS! What we also need is lower operating costs for base metal SBS. The value per ton of ore is generally too low to allow the cost of sorting plus separate discard of sorter reject. ORETOME Limited has been doing some test work along these lines, google "Van Weert, humberetc" for (scanty) details. On the mine strategy side, the industry has to get rid of the tonnage bonus for Mine Superintendents, institute a bonus instead for (metal) value delivered to the Mill. The system should reward quality, not quantity!

Interesting post, but I'll argue with you on a point. The word "ore" can only be used if it applies to a mineral assembly that can pay for all of the costs of mining, processing, transport, environmental concerns, administration, and profit to investors, and anything else to which a currency value can be applied to the operation. Miners do not mine metals, or minerals, dimensions stone, or any other geologic material; we mine cash! If any separation device sorts individual rocks into "keepers" or "waste" then it is sorting ore from waste. Whether the keepers are coal or diamonds (most diamondstoday are sorted) is irrelevant.

A year ago I did an analysis that showed that current ore sorting technology should be able to improve the economics of a 0.3% Cu porphyry down to about 0.10% benches. It specifically reduced the grinding circuit feed by 50% for major savings in grinding, roughing, water used, and tailings storage. I also noted that by crushing the pit waste as I have see done in Bulgaria, the entire dry waste stream could be disposed of by conveyor and stacker. While I did not estimate those savings they would be significant vs. trucking. Then there is the carbon footprint saving for the tails. Yet another thought, sorting could be used to either separate out the PAG waste or include it with the ore for grinding and wet disposal, these being site-specific choices for base deficient deposits.

I agree in principle but the giants will not lead the charge. SX-EW, for example was first used by Bagdad Mining, a one-mine company then. The giants want to say proudly that they were the third or fourth company to embrace the change. There are thousands of sorters in use today but a couple of years ago only a couple were in non-diamond mineral service. It's ironic because hand sorting, something I've seen used commercially twice, was probably the original "mineral process" technology tens of thousands of years ago.

It looks that it has potential and one has to look at this before jumping into developing a beneficiation flow sheet based on conventional practices.

The industry is worse than this example suggests. Give a lab a set of drill core for evaluation and they will start with a composite head sample for elements and mineralogy then go to a cook book for the standard flowsheet for those results. They do not, unless asked, look at how alternatives might benefit the specific deposit. Here in British Columbia we have a history of copper flotation plants with several percent magnetites in the tailings. The thinking was copper, not total cash flow. Did you know that the original Virginia City gold mines were plagued by blue clay that kept squeezing into the drifts and was removed at great cost to be dumped where the annual rains could carry it away. Two years later someone analysed it and discovered 50% silver. From then on Virginia City was a silver camp, the first in the USA to be exact.

Back to sorting, it stands to reason that a 12 meter column of rock, a standard bench height, is unlikely to be entirely one grade, or that the boundaries of bedding planes, intrusions, etc. are exactly planar and either vertical or horizontal. Our current approach is to use the average for decision making as to if it is ore or waste. How primitive!

I can feel the passion you have for the industry and good practices; I am sure all will get the message you are trying to convey.

Eventually but in which decade?

I am basically a Teacher and so an optimist of the future of the profession.I still go around and when I meet young professionals, fresh from Institutes , and working in operating plants, what I gather is that they are taught subjects of the outer orbit(lots of simulation, fluid mechanics, surface chemistry etc, which are no doubt important) but were not taught basic concepts. I hope companies spend some money so that Teachers interact with industries and get to know the basic requirements of industry which is nothing more than" how to look at data and come to good conclusions and action plans.

You are correct; it's individuals that lead the charge. But it is the successes of big companies that inspire the bankers and investors to risk their money on new methods of processing.

In my quest for lowering ore sorting costs, I decided to sort drill core. I have been looking for a company or R&D institute to re-drill an already "known" block of ore, with, say three holes around the original hole, cut the new drill core in, say, 1.5 inch pieces, assay one or both cut ends by XRF and determine from these assays and sorting test work how sorting might affect project economics.

Such a procedure would also allow a check on the value/validity of the present method of averaging grades for decision making. And that's the problem! Everybody realizes that averaging is primitive, but nobody dares to say so for fear of the consequences. It would destroy the money raising system! How could one put out N43 -101's? Simpler to Ignore the elephant in the room.

I strongly advised against cutting up lengths of drill core. The resulting cylinders are not representative of crushed rock so a test could be very misleading. It would be better to use large drill core, 200 or 300 mm if possible, then gently crush and screen it over 150 mm. You could limit the test to those zones that had nearby values. Forget using three smaller holes near one known one; put the resources into three locations.

The objective is to prove that scanning works and confirming DD holes is not that important IMHO. Get a geo to go through the rocks, presumable kept in batches by intersection, and grade them into waste or ascending value by sight. 1-200 rocks are an initial study; it is unclear how much work you want to do. Each would be numbered with the geo's comments recorded, then put through a scanner or scanners. That complete, each would be crushed, pulverised, and analysed. Those results would tell the tale.

Good advice from you. When considering air blast sorting. I want to eliminate air blasting - it represents approx. 70% of the operating costs of "standard" sorters. Separation is achieved by passing sulphide rock through a microwave heating "zone". Rocks containing sulphides heat up, gangue/waste does not. MW'ed rocks are discharged on a belt coated with a thermo sensitive adhesive, sulphidic material sticks, waste does not, resulting in different discharge trajectories. No sensor, no air blast! I have a short video demonstrating the principle with white and coloured glass beads, approx. 1/2" diameter. The coloured beads were heated in a toaster oven. The belt was made of tailor's tape, same material as the cloth patches used on jeans, etc.These patches use vinyl acetate glue.

I want to use cut drill core because I have to evaluate glues. For that I need a fixed surface area to contact the glue layer and an approx. constant weight per piece.Cutting up drill core into true cylinders will give me that. Sorting broken rocks is a later problem. For that, we don't need to use a belt. I have a domestic MW oven fitted with a small rotating drum inside. Tumbling small rocks with powdered glue present in a MW field might make high sulphide agglomerates. Haven't got around looking at that. Or, we may borrow from the diamond industry, use some equivalent of their grease tables based on temperature differences of the rocks.I agree with you, a little good luck would help!

A length of drill core with a smooth surface would tend to lie on the belt on that surface, and this would be an unfair advantage in a thermal system. You might be able to test glues that way, but not commercial rock. Heating a tonne of rock by 20 C degrees amounts to 5 kwh/t. Assuming a yield of 20% to concentrate that 1 kwh/t of feed. If the rejection of gangue is 20% that is 4 kwh/t. According to one supplier that I have used, an optical sorter would require 0.32 kwh/tonne of feed for either yield. These numbers suggest that a glue system would only be more economic for the range of <7% yield to concentrate, and that assume that there are no middling particles.

To the best of my knowledge no one has demonstrated the glue process commercially. It requires, my guess, 10% sulphides in any particle to work. At that level I'd expect that EM methods would also work or a commercial anti-magnet. My efforts involved <5% sulphides in a typical ore particle, a range where thermal probably cannot be used.

Something I've noticed is that the drop distance in a separator is about a meter. Surely if they were located, say, ten meters above the splitter then much weaker and cheaper forces could be used, the "throw" distance being three times that of the present machines. This would cut the energy required by 1/3rd but let’s say by 1/2. The 0.32 kwh/r becomes 0.16, trivial compared to the other energy requirements that total close to 2 kwh/t.

Again you are right on the money, we don't want to use more energy than the standard system with air blasts. Another common objection to my proposal is the dirt falling of the bigger rocks and coating the glue, making it ineffective. My proposal (already submitted for funding) for the third stage or R&D towards low cost sulphide ore sorting is to not use MW heating, but use a regular sensing system to identify the valuable rocks, with a bank of jet printers perpendicular across the belt dispensing the appropriate glue to the top of the sulphide rocks on the belt. The rocks meet a second belt immediately after the glue dispensing, parallel to the main belt. The rocks become the ham in the sandwich, so to speak. And don't tell me that for this to work all rocks have to have the same height on the belt, I have thought about that as well. All this to say, that if we don't go in the workshop and experiment, we will be stuck with what we have. The old Cornwall fellows did not give up easily, our university professors and R&D directors of mineral processing need a dose of that spirit...

P.S. The Humber College machine positions the MW heated drill core cylinders vertically upright on the glue coated belt, on the cut surface which has been analyzed by XRF.

P.P.S. I would be careful using such precise figures as 0.32 kWh/tonne from a supplier. Power consumption is a function of rock size and the valuable / waste rock ratio.

I rounded the manufacturer's number down one decimal! When I checked another manufacturer, the number was 0.4 kwh/t (rounded) for the same mass split. Operating experience, spelt "headaches", suggest that adding glue to a feed stream is a dangerous direction but by all means try it. Heck, water is a deadly foe in an ore chute. How does your glue stick to a wet rock? Would your device target the highest part of the rock? I agree about not worrying about the absolute heights of the target rocks as long as the top belt is made of flexible, floppy, material. There is one material that is great for picking up rocks and dirt; it's called "children's clothing". A pet's fur works well too.

You're getting the drift! (Sounds like an old miner's expression). Once you start mulling over the variants, there is no end to what can / should? Be tried. I have stayed away from sorting wet rocks to date; I believe that processing dry rocks will become dominant in view of the water supply worldwide. But again, MW'ing offers opportunities for a wet system. Wash the rocks with water carrying some dissolved sugar. Heat on belt with MW's. The sulphide rocks dry and will stick to the belt. Easily explored in your kitchen MW oven. Don't try this with a rubber belt, it heats up as well. Use a piece of Teflon. Sugar (saccharose) is surprisingly cheap, environmentally friendly and, likely does not interfere with subsequent flotation.

We have successfully commissioned and XRT ore sorter at the plant I work. The performance from the unit has exceeded all of our expectations. Application is sorting of ball mill pebbles, previously a waste stream, upgrading gold from 0.045 opt to 0.250+ opt. Recovery is right where testing had indicated. Currently we have processed 130,000 tons of pebble through the sorter in 10 months. We have hosted several miners who want to see the unit in action, they have all been thoroughly impressed. We will now look to incorporate the sorter into the crushing circuit to upgrade ROM material once pebble stockpile is completed.