Refining Gold Processing

E&MJ looks at designs and technologies that are making gold flowsheets more efficient and economic

By Carly Leonida, European Editor

Gold processing is one area of mining where the old adage ‘if it ain’t broke, don’t fix it’ really rings true.

If you took a gold miner from the 1990s and put them into an operation today, they would probably be familiar with every piece of equipment and circuit in front of them of them. Sure, there would be incremental changes, mainly technological, in the control systems, there’s a greater focus on carbon management now, energy efficiency, the addition of pre-leach thickeners etc. And there are more plants dealing with refractory ores. But, on the whole, not a lot has changed.

“We’re still using the same basic processes and equipment that we were 20 or 30 years ago,” Ben Murphy, Global Key Account Manager and Key Industry Director for Gold at FLSmidth, told E&MJ.

“But people are tweaking certain areas…for example, there has been a lot of work into alternative lixiviants for gold leaching, but to date there hasn’t been widespread implementation. When it comes down to it, the old cyanide leach plan is pretty efficient at getting gold out.”

“We’ve now got the international cyanide code. People are a lot more conscious of cyanide and the environment than they were 30 years ago. That is a good thing. There are various strategies… Different separation methods, preconcentration to minimize material that comes into contact with the cyanide, and detox to reduce the concentration in tailings.

“It is a lot more common now to see a flotation circuit in gold plants, because they need to capture the gold that may be associated with sulphides. “Interest in fine-grinding is also increasing because, for some refractories, they simply require grinding down to a size that allows the gold to be exposed for leaching. Once again we are still using the same basic technologies, albeit it more efficiently. I don’t see that changing any time soon.”

Gradual Progress

Jan Van Niekerk, Director of Gold Process Solutions at Metso Outotec, agreed: “20 or 30 years ago, cyanide was already established, even the use of carbon had been established. And, yes, the process is fairly much still the same.

“But there have definitely been incremental improvements. For example, ore sorting; that’s certainly been gaining more market acceptance. A lot of that is to do with the technology and the instrumentation that’s available.

“But something like the gravity circuits, if you think of 20 years ago, for the most part the gravity circuit was an afterthought or add-on, that wasn’t really well designed and engineered for most plants. Whereas now it’s an essential part of the plant design.

“The principles of milling are still very much the same, but the controls have been fine tuned. Things like mill optimizers that create benefits in power saving and managing the throughput of your mill…

“Even on flotation, the instrumentation has advanced significantly. It does give you that step change in ability to control the plant and get better recoveries. Again, we are getting more variable ores, and you can’t just set a plant and leave it to run for a long time; you have to constantly monitor and adjust.”

Process modeling programs have also contributed to the optimization of plant designs.

“Modeling has certainly made a difference. And, in gold, it’s been under-utilized for the most part,” Van Niekerk said. “But we’ve seen, especially with more complex orebodies where you’re starting to get copper in and there’s a copper circuit attached, or those with high variability, then a process modeling program is a really powerful tool to have.

“Mining models are normally built around the gold and sometimes a sulphur model or maybe an arsenic model. But, with geometallurgy and variable orebodies, you can almost get to the point where you’re building an operating envelope model to determine what the recovery is in certain areas of the plant and for certain areas of the orebody. That’s going to change more and more over the next few years.”

Geometallurgy and process modeling coupled with instrumentation and process control will become even more important in the coming years. Especially as we see mines and concentrators developed in more remote locations; it’s not easy to get highly skilled people permanently stationed on remote sites, and there is a growing requirement to control plants remotely.

We’re also starting to see a shift in design philosophy, partly driven by higher gold prices and partly by mining companies’ attitudes toward risk.

“A few years ago, the focus was really on building the cheapest plant possible because the orebodies were fairly simple,” explained Van Niekerk. “Whereas, with more complex orebodies in more remote locations, there’s a requirement to build a slightly more expensive plant, but you get better reliability, lower power consumption and better water management.

“There’s more a philosophy of investing in and designing a plant for life, rather than just seeing it as an upfront capital cost. That’s a big change that I hope it’s going to continue.”

Orebody Complexity

Increased ore complexity is one of two over-arching themes that are driving innovation in gold processing.

The ideal orebody would be one where the gold is located at shallow depth, is well liberated at a coarse size and can be recovered using simple gravity techniques. Next would be a free milling oxide orebody where a cyanidation and gold recovery circuit would be required. Transition material with a mixture of oxides with increasing amounts of sulphides would then be the next hardest to treat with the flowsheet having to possibly incorporate ultrafine grinding and partial oxidation technology.

Finally, there are fully refractory orebodies where the gold is locked within sulphides, and the flowsheet becomes more complex with the addition of an oxidative treatment method such as pressure oxidation (POX), roasting or bacterial oxidation (bio-oxidation).

The presence of carbon in the ore can result in the orebody being classified as double refractory and this may require additional process steps again.

To clarify: refractory gold ores are generally defined as those ores that do not give economic recoveries in conventional cyanide circuits where the ore has been ground to around 53-75 microns. There are several reasons as to why an ore may be refractory. The two most common are:

1) The gold is locked in refractory sulphide minerals such as pyrite, arsenopyrite or pyrrhotite and occurs in both the chemically bonded state and as micro or nano-size grains of metallic gold.

2) Or the gold is locked in silica.

Other reasons for an ore not responding to conventional cyanidation include:

1) Preg-robbing due to the presence of carbonaceous material;

2) Other minerals reacting with cyanide;

3) The presence of lead, copper, or antimony minerals; and

4) Interference and passivation from the decomposition products of pyritic and other minerals.

The rise in the processing of refractory ores is mainly because most of the shallow, free milling, oxide orebodies across the globe have been exhausted. Elevated gold prices mean that attentions are now turning to mixed orebodies or refractory ones where the gold is harder to liberate and recover, and this of course puts additional pressure on the project economics.

“Evaluation of refractories has become a lot more common,” Murphy said. “We’ve seen a massive uptick. Last year FLSmidth acquired Barrick’s technical research arm, AuTec. Not only for its expertise in complex gold processing, but for their expertise in POX piloting.

“We brought the pilot plant down from Vancouver to our lab in Salt Lake City. It was commissioned in September 2019 and it’s been running, pretty much, 24/7 since, doing test work on different orebodies.”

The Environment and Economics

The second major theme revolves around environmental responsibility. Over the past 10 years, every mining company, large or small, has turned their attention to more sustainable mining practices.

In gold processing, this translates into using technologies and processes that are more energy efficient, for example, next-generation ultrafine grinding mills or high-pressure grinding rolls (HPGRs). Technologies with smaller footprints, or those that consume less water and cyanide are also growing in popularity and, there is a growing focus on total cost of ownership.

“One thing that’s becoming more and more important is water balance and the management of water around the mine,” said Van Niekerk. “There is increasing legislation around the use of water and discharge requirements. And that links to the environmental permitting which, in itself, is becoming more complex. Community relations and all those aspects that need to be taken into consideration too.

“Also, sites in more remote locations aren’t always easy to access, and that carries certain requirements from a design and selection of technologies point of view.”

Underpinning these two themes are gold prices and the commodity super cycle. Prices are, as of July 30, at an all-time high. It’s been a bullish couple of years and, during July, geopolitical uncertainty and global financial instability fueled by COVID-19 saw gold prices surge toward $2,000/oz. As a result, mines, engineers and OEMs alike are mostly working flat out right now.

“Everyone is trying to make hay while the sun shines,” Murphy said. “That provides the opportunity to go back to basics. There are a lot of old gold plants out there that haven’t seen love for a number of years because of the price.

“So, yes, we’re seeing a lot of activity, and a lot of enquiries.”

Van Niekerk reported similar: “In general, we are seeing increased interest, particularly in plant audits and optimization services. A lot of that has to do with the gold price – there’s pressure to maximize recovery. But also, with a higher gold price there’s actually some cash available to spend on R&D and process optimization. Whereas when the gold price is low, that money isn’t always there.”

Test, Test and Test Again

Every expert interviewed for this article agreed that, the more complex the mineralogy, the more important it is to get test work and characterization right at the start, and to continue sampling throughout the life of the facility.

“With traditional geometallurgical test work, you do your test, you get some results, A plus B equals C… Great. But if you have some unseen complexity there, you may be in trouble,” Murphy said.

“Whereas, if you do in-depth test work, using a lab like our one in Salt Lake City, Utah, we have the mineralogy team working side-by-sde with the process people. They do the test work, and they’ve got a more holistic view of what’s happening and why. They’re also, brilliant for troubleshooting, especially in the early stages.”

E&MJ has spoken to multiple mining companies over the years that have picked up projects rejected by others due to orebody complexity or geometallurgical complications that were missed at the pre-feasibility stage. This is just one reason for the cost overruns that seem to have become almost endemic in the gold industry.

Steve Flatman, General Manager at UK-based engineering specialist, Maelgwyn Mineral Services, told E&MJ: “Unfortunately the metallurgy is often an afterthought. There’s often little metallurgical representation on the boards of mining juniors. They start off with a conceptual study, and management just assumes that they’re going to recover 100% of whatever gold’s there.

“Only later on when they’ve actually done some metallurgical test work, do they realize that it’s not quite as simple as that. And it’s difficult to talk the share price up after that. So, they tend to sell the project on to somebody else. That’s why you see some of these more difficult deposits bandied about. Because yes, the capital required for technologies like POX is not insignificant.”

E&MJ asked if we should be doing more test work early on in the scoping process to help circumvent this issue?

“Yes,” said Flatman unequivocally. “The geometallurgy must be an integral part of the whole program. You don’t have to wait until the definitive feasibility study — the last stage of the project — to start doing the metallurgical and pilot test work, which we quite often see. At that stage you’re almost too far down the line in terms of environmental permitting. For instance, if it comes up that you need higher cyanide additions, then that could affect your environmental permitting and put the project back again. That’s why it’s important.”

Van Niekerk added: “The biggest issue when you’ve got a highly variable orebody is the testing requirements for various ore zones. It makes the whole process of developing the project so much more complex.

“It’s not as simple as selecting a sample and then doing work on it. Often the mining model is developed over time, so you’re always changing the blend that’s going into the plant. If you’ve got a uniform orebody it doesn’t matter that much. But when you get variability it becomes difficult. It means that you’ve tested one zone, but suddenly you’re not going to treat that on its own, you’re going to mix it with something else and you have to do another test program on that.

“Often some of the tests are done at one laboratory and the next set of tests at another and in between the origins of the samples get lost or diluted or there’s different test procedures. Most companies do a lot of test work, but often the management or the planning around that is not coherent over the whole development of the project and management of the data isn’t always that good.

“Again, that’s where modeling programs are coming to the fore; where you can have a set of standard tests on the different ore zones and the software can then integrate that into an overall model based on your feed parameters.”

“Getting back to the point about cost overruns; we’ve certainly seen underperformance on some projects because there was never a proper understanding of the orebody.”

Optimization Opportunities

E&MJ asked Murphy where he sees the greatest opportunities for optimization improvements in gold circuits at present.

“Preconcentration works really well on certain orebodies. I think it’s something that we need to look at more, but the economics are very orebody dependent,” he said. 

“In the gold sector, I’m really an advocate of going back to basics. We’ve got a fantastic gold price right now, but as we all know the industry is cyclic. Sooner, or later, that gold price is going to drop and tough times will return. Driving your operating costs down as low as possible will not only improve profitability for now but will secure your future when the downturn comes.

“There is some very low-hanging fruit, for example pump optimization. Getting the pump matched to the duty can save you in maintenance costs but also can reduce energy consumption. When you multiply this by the number of pumps that run in a plant, it adds up quite quickly. There are many pumps on plants that were installed when the plant started and have never been changed despite the duty changing. In the last 20 to 30 years there has been a lot of progress in wear materials and understanding of slurry pumps which can now be applied to the benefit of the operator. A quick review can lead to major savings.

“Another area is sampling. A lot of gold mines, especially older ones, don’t have the best sampling equipment. It’s so critical when you’re trying to control the process; if you’re striving to get the lowest cost per ounce, you really need a good understanding of that process.

“It sounds cliché, but it’s true: if you can’t measure it, you can’t manage it. And the first step in measuring that is getting a good sample. Yes, downstream, where you’re sampling liquids, it’s a lot easier, making sure you have really good slurry samples is also important. The samples need to be collected in the right places and ensuring they are representative of the process.

“I see a lot of shortcuts in existing plants. Having good sampling makes life easier from a control perspective, and also from an accounting perspective. If you’ve got better sampling, you’ve got a better idea of the inventories in the process. So, at month-end, you’re not searching around for ‘missing ounces’ which, I’m sure, keeps a number of CEOs up at night.

“It’s important to remember that there are two ways of changing the mining cost per ounce: you can either produce more ounces or you can reduce costs. Never rule out the first one. If you can produce more efficiently by having better process control that is a big win for the site.”

Gravity Concentration

“As I said, for existing plants, it’s very much getting the basics right,” added Murphy. “For design, going forward, we’re likely to see a more holistic approach with greater focus on total cost of ownership. But there are also some new, exciting technologies coming.”

Gravity concentration has been an area of focus for the team at FLSmidth of late. Murphy explained their progress to E&MJ…

“A number of years ago, we acquired the Knelson batch centrifugal concentrator (BCC), which was and still is the industry leader in gravity technology,” he said. “The Knelson is typically installed in the grinding circuit and pulls out free gold. It captures the gravity gold upfront which reduces the chance of that gold getting lost to tails and gives you a quicker return.

“Most gold mines, in this day and age, will do testing at the start of the project to see if the gold is gravity amenable and is able to be recovered by a BCC. 

“What we’ve been developing over the last few years is a Continuous Knelson. It uses a similar concept to the BCC but has some important differences. A batch Knelson has a low mass pull. It can pick up gold associated with other minerals, such as sulphides but generally, it is most efficient at targeting free gold due to the high density difference.”

As the name suggest the Continuous Knelson is designed to be run in continuous mode and doesn’t need to be installed in a circulating load to maximize recovery efficiency. In addition, because of the way it’s set up, it can target a wider density range for efficient separation.

“We’ve been doing a lot of lab work,” said Murphy. “And we’ve got a few clients who have bought pilot units from us and done their own test work on concentrating gold that is associated with sulphide minerals. This work has culminated in a number of Continuous Knelson applications, typically installed on the cyclone overflow stream.

“I can’t mention who or where yet, but let’s just say these machines are very different beasts. They’re separating out the sulphides with gold, in much the same way that a float circuit would.

“In one case, the client didn’t want to put in a float circuit because of logistical challenges. They’ve done all their test work and due diligence, which resulted in them installing production units. Feedback from the site is that they are working really well. ”

“A key limitation of the batch Knelson is that it only has a mass pull of around 0.1% of the feed. The Continuous Knelson can pull 5% to 30% quite easily, depending on the ore characteristics. So, it’s a much more flexible machine. These units can handle up to about 300 nominal tons per hour. Obviously, it depends on the exact site and the feed conditions. It’s a different way of doing things, there are no reagents and a very small footprint.”

“We also have some exciting developments that will improve our batch Knelson coming out soon. We have done a lot of optimization and computational fluid dynamics (CFD) work. Based on these results and improved understanding we have changed a few things around the design. Early 2021, we will be releasing a new modification that can be retrofitted to existing models. Early results with this redesigned unit show a marked improvement in gold recovery in certain size ranges which we are getting pretty excited about.”

Harnessing Screening

Another incredibly efficient and time-tested way to separate and classify materials in gold concentrators — one that is often overlooked as an optimization tool — is screening. New designs and capabilities mean that screens can now be used to replace some of the more traditional elements of flowsheets.

“Anywhere that a particle separation takes place, a screen is the most efficient device,” explained David E. Perkins, Mining & Industrial Commercial Manager and Director of Precious Metal Applications at US-based manufacturer Derrick. “Often in gold circuits, screens are used to capture very fine carbon and boost gold recovery. Also, models like the Derrick Stack Sizer are capable of replacing hydrocyclones in the grinding circuit, resulting in increased recovery as well as reducing the power consumption per ton and increasing the capacity of the grinding circuit.”

Derrick’s approach is to tackle the applications that are the most difficult. Its line of machines with high flux rates and non-blinding panel technology allows the company to provide a machine that is a fraction of the size of traditional screen units and capable of consistent capacities through the life cycle.

“The Derrick solution might be a 4×8, whereas the alternative is 8×16,” said Perkins. “Being capable of processing the same volume of material in a much smaller plant footprint can be a big advantage to gold operations.”

Screens are used in a variety of applications in gold concentrators. Removing preg-robbing trash, separating slurry from loaded carbon, sizing of carbon, dewatering of carbon and, as a safety device to prevent loaded carbon from going to tails, are just a few examples.

“I have experienced operations that have large screens that are shaking the foundations and, after replacing them with a Derrick, operators have difficulty in knowing if the machine is operating due to the very low transmitted forces and low sound levels,” said Perkins. “I had a client that was losing productivity by having to wait several hours to decant the feed hopper to their kiln and, after replacing the dewatering screen with a Derrick, there was no longer a wait required.”

By replacing 50+ year-old cyclone technology with high-frequency vibrating screens like the Derrick Stack Sizer or SuperStack, some mines can increase their plant capacity by 20%-50% and boost recovery by up to 10%. Aside from reducing the plant footprint, this also has the added benefit of reducing overall water and cyanide consumption, resulting in a much ‘greener’ approach to recovering gold and other minerals as well.

In early 2020, Derrick launched the G-Vault urethane-surfaced interstage screen for use in carbon-in-pulp and carbon-in-leach (CIP/CIL) circuits; an alternative to traditional stainless-steel wedge wire screens. The enhanced durability and anti-blinding technology mean that the G-Vault has shown substantial reductions in maintenance and downtime. Compared to the alternative of multiple, weekly shutdowns with existing wedge wire screens, the G-Vault can run beyond eighteen months with no cleaning required; a boon in remote locations where access to site is tricky.

The G-Vault features independent screen sections retained in a stainless-steel cage. Their interchangeable nature reduces maintenance costs by permitting replacement of only a heavily worn section, rather than an entire screen.

Alex Nicosia, Global Precious Metals Manager for Derrick’s New Business Development Division, told E&MJ that the company has seen strong interest in the G-Vault from its client base, and field testing has now been conducted in six countries across North and South America, as well as Asia.

Bio-oxidation

June 2020 also saw the launch of Outotec’s MesoTHERM BIOX process, an enhancement to the well-known mesophile BIOX process that combines bio-oxidation technology with a higher-temperature thermophile oxidative stage to enable more effective overall sulfide oxidation.

“The MesoTHERM process has been in development for quite a while,” Van Niekerk said. “We’re getting significant savings in cyanide — up to 50%. The higher temperature thermophile bacteria give a more benign product following bio-oxidation that then consumes less cyanide in the downstream processing.

“BIOX or the mesophile technology has been around for over 30 years. And even the thermophile technology has been around for a long time. It’s just nobody has really commercialized it. And we’ve gone through a very long-gated process to develop this, but it’s now being commercially operated at the Fairview BIOX plant in South Africa.”

Van Niekerk said the company has also had multiple enquiries from mines looking to upgrade the BIOX process they currently have. Upgrading to MesoTHERM is a relatively simple process that involves reconfiguring the circuit with the addition of Outotec’s High Rate Thickeners for interstage thickening and OKTOP Atmospheric
Reactors for the thermophile step.

“Most of the newer projects immediately ask if we’re testing MesoTHERM as part of their project development,” he said. “The Fosterville mine has already expressed an interest to test this too.

“We’re primarily targeting mines with higher-grade sulfide orebodies. That’s something that normally results in high cyanide consumption when you’re treating the ore using the normal mesophiles. So that’s where this will really add benefit.”

Partial Oxidation

In terms of flowsheet design, refractory sulphides are normally treated through a flotation route. The gold-rich sulfide concentrate is then treated further using roasting, POX or bio-oxidation. These are all aimed at breaking down the sulphide matrix to liberate gold. Ultra-fine grinding performs the same function particularly where gold is locked in silicates or other minerals. Many of these processes are well developed and can yield very high gold recoveries — more than 90% or even higher for POX — but all tend to have inherent issues.

Roasting although historically used is an environmentally unfriendly process and presents permitting issues in many countries. POX requires a fairly high degree of operator skill and control, not to mention often exotic materials of construction. Bacteria used in bio-oxidation, whether heap or tank leaching, are susceptible to changes in environmental conditions and require careful control.

“For large high-grade deposits, POX is generally used as these deposits can justify the high CAPEX and OPEX,” explained Maelgwyn’s Flatman. “For smaller deposits, partial oxidation processes such Maelgwyn’s Leachox can be used. Although the recovery is lower at around 80%-90%, the costs can be an order of magnitude lower.”

Central to Leachox is the Maelgwyn Aachen Shear Reactor — a proprietary low-pressure high shear mass transfer device — which has been available for many years now.

“The majority of our Aachen Reactor applications are in preoxygenation, and Aachen assisted leaching which improves the kinetics of gold leaching, increases gold recovery and reduces cyanide consumption,” Flatman explained. “We use them on oxide ores and also on mixed orebodies and refractories. As you start to get more sulfides present and cyanide consumption starts to increase, then we use the Aachen aerators to add additional oxygen into the leach, but with the shear to keep the mineral surface clean so you’re continually chasing that mineral reaction. Then, as you move further into refractory type of materials, the best way to process them is normally through flotation and that’s because you can concentrate the gold into a smaller mass, so your process plant size is smaller.

“Again, we use the Aachen Reactors there, but in a much more intensive manner. Typically for pre-oxygenation roles, we would pass the concentrate once or twice through the Aachen Reactor, but in the Leachox role, we’re doing partial oxidation so we may
pass the slurry 30 to 40 times through the Reactor.”

Leachox uses simple ultra-fine grinding equipment combined with the Aachen Reactor to partially oxidize
the sulfides. When combined with
the liberation of gold by the ultra-fine grinding of concentrates it results in acceptable but, more importantly economic gold recoveries. The use of the Reactor accelerates the leach kinetics allowing for higher throughputs through the CIL/CIP plant and reduces cyanide consumption by initiating efficient oxidation of cyanicides in the pulp that react with cyanide in preference to gold and reduces the leaching effect. In parallel, the Aachen Reactor removes passivating layers on the mineral surface that can otherwise impair the leach reaction.

Where graphitic carbon is present leading to preg-robbing the Leachox circuit can be enhanced by using Maelgwyn’ s patented G-Cell to remove the carbon.

Aachen reactors can also be used for cyanide destruction at the end of the process as well, so they’re a pretty flexible piece of kit.

“We’ve got around 70 reactors running at different sites now,” said Flatman. “We commissioned a large facility with one of the eight reactors in South Africa last year at Evander Mines’ Elikhulu tailings operation.”

Barrick’s flagship Kibali mine in the DRC is another fan. The Kibali process circuit consists of eight ultrafine grinding mills which mill the flotation concentrate to approximately 80% minus 18 microns. It is then subjected to two-stage pre-oxidation at pH 10 utilizing Aachen Shear Reactors before cyanidation.

Flatman spoke honestly: “If you’ve got a multi-million-dollar project — over a million-ounce gold project — and it’s high grade, I would put in POX rather than our leach operation. And the reason for that is that POX gives a very high recovery, often as high as 98%.

“The problem is it’s the most expensive option in terms of CAPEX and OPEX. Straight away you’re talking high pressures, titanium vessels, you need skilled operators and instrument technicians. And if you’re in a remote location, that can be expensive.

“Up until 30 years ago, roasting would have been used, but that’s fallen out of favor due to environmental considerations. In roasting you can create gaseous emissions of sulphur dioxide, sulphur trioxide and if there’s any arsenopyrite present, there can be arsenic emissions. Which can obviously cause major problems.

“Now, you can trap that through gas cleaning equipment. But again, it adds to the cost. So, POX has become the go-to process, like cyanide.

“Bio-oxidation became popular approximately 20 years ago. Although it’s cheaper than POX, it’s only really suitable for smaller deposits because of the time involved. Bacteria by nature are not instantaneous, you can be talking residence times of four to five days in some of the vessels, rather than four to five hours for the cyanide leaching. And if you don’t have the technical people on site to understand how the bacteria work, you can kill them.

“So, that leaves a gap in the market for partial oxidation processes such as Leachox. And in our process, we don’t get as high a recovery as POX or bio-oxidation, but the process is an order of magnitude cheaper. For smaller deposits, where you don’t want to put in huge amounts of infrastructure, then Leachox is a good option.”

It’s also fairly straightforward to operate; basically, a pump sends material through the Aachen Reactor, which contains no moving parts.

“The slurry is accelerated up to 10-12 liters per second, deliberately to create the shear,” explained Flatman. “In creating that shear, a significant amount of abrasion takes place and the Reactors have to be maintained; you can’t just put them in leave them forever. So, we usually lease them. That has a two-fold advantage. Firstly, we ensure that it works — it helps to de-risk the project for the client — because the maintenance is included as part of the lease contract. Secondly, the operators themselves have the opportunity to believe in the process. Because if it doesn’t work, then they can just stop it. It’s very easy to prove it’s working.”

Ultrafine Grinding

To date, the primary focus for Aachen Reactors has been gold ores. However, Maelgwyn recently commissioned its first application on a silver orebody at Gumustas Mining’s Nigde-Bolkar gold-silver operation in Turkey with excellent results, including a recovery increase of 8% and a 30% reduction in cyanide consumption.

“We do see potential for silver,” said Flatman. “One of the problems with silver is that the leaching is much slower, you tend to get a lot more passivation taking place. So, if the gold ore leaches in 24 hours it might take 76 hours for a silver ore to leach. The Aachen Reactor thins the boundary layer and speeds up the leach process. We are talking to a couple of silver operations at the moment, so watch this space.”

To complement the Aachen Reactor in the Leachox process, Maelgwyn is also developing its own ultra-fine grinding mill — the Ro-Star mill.

“We’ve been developing it for the last two years,” Flatman told E&MJ. “To date, we’ve had to rely on buying in mills from elsewhere. Although the mills have improved in quality, they tend to be very expensive. So, that increases the overall capital cost of the Leachox process. Our aim with the Ro-Star is basically to produce a better ‘mouse trap’ for want of a better term and drive down the cost of Leachox even further.”

Pressure Oxidation

Which leads us nicely to the gold standard for refractory processing: POX.

Although POX does carry significant CAPEX and OPEX costs, it can be harnessed to create significant value. One such example is Russian gold producer, Petropavlovsk. The company’s POX Hub in Russia’s Far Eastern Federal District is a world-class facility and a strategic asset.

Approximately 15%-30% of Russian gold reserves are classed as refractory, while nearly 6 million oz of the company’s own reserves are also refractory.

The POX Hub is a centralized facility that was commissioned late in 2018 and began ramping up in early 2019, treating both Petropavlovsk’s refractory gold reserves as well as those of third parties in Russia and the surrounding regions. The facility has enabled the company to increase its own gold production and ultimately reduce its cash costs. It also means it can leverage stranded refractory ore deposits, which cannot be mined and processed in an economical and environmentally sound manner without a POX plant.

Petropavlovsk has three active gold mines: Pioneer, Albyn and Malomir which cover license areas of more than 3,200 km². The mines are a mixture of open pit and underground, with 15 million metric ton per year (mt/y) ore processing capacity in addition to the 500,000 mt/y potential of the POX Hub.

The POX Hub is located at the site of the company’s first and now depleted mine, Pokrovskiy, where it benefits from existing milling facilities, road and rail infrastructure, low-cost renewable power (hydro-electric) as well as a nearby limestone deposit — a key ingredient in the POX process.

The processing of refractory gold ores begins with crushing and grinding, followed by flotation at the mine site which reduces the volume of material down to between 2.8% and 4.2% of the original mass by selectively enriching gold mineralization. The resulting concentrate is then transported to the POX Hub by road and/or rail for further processing and gold recovery.

The POX Hub is designed to operate at a pressure of 3,500 kPa and a temperature of 225°C which is higher than most other POX plants and enables the efficient processing of a variety of refractory feeds, including double-refractory concentrates.

What makes this POX plant unique is that there are four separate autoclave vessels that provide flexibility, allowing concentrates from different sources, and with varying characteristics and grades to be treated optimally, either individually or as blends, in separate autoclaves. It also means that scheduled maintenance can be undertaken by shutting down only one autoclave at a time. Each individual autoclave is expected to operate for 7,500 hours per year once at full capacity. The layout has been designed to accommodate a fifth and sixth autoclave in the future, something which is being considered and would increase capacity by around 50% compared to the existing level.

Throughput rates are determined by the specific characteristic of the concentrate and particularly the level of sulphur, with high sulfur concentrates taking longer to process given the additional time required to oxidize sulphur in the autoclaves. As a result, the capacity of the plant varies on an annual basis according to the material it is processing.

The group’s first 3.6 million mt/y flotation plant was successfully commissioned at Malomir in 2018. Concentrate grades vary between 21 and 32 grams per mt (g/mt) of gold with a sulphur content ranging from 21% to 29%. The ore at Malomir is double refractory and requires special measures to neutralize the carbon so that gold recoveries are not negatively impacted. These include strict control over chlorine ion levels and the relatively high pressure and temperature of the autoclaves. As a result, the average recovery of gold in the POX circuit is expected to be between 93% to 96%.

The construction of a second 3.6 million mt/y flotation plant began in H2 2019 at Pioneer. Once operating at full capacity, Pioneer’s flotation plant will double Petropavlovsk’s flotation capacity of its own ore to 7.2 million mt/y. Construction is well under way and the new plant is expected to be fully operational later this year. Once commissioned, Pioneer concentrate grades are anticipated to vary between 20 and 33 g/mt of gold, with an average gold grade of 24.2 g/mtand average sulfur grade of 21.0%. Gold recovery in the POX circuit is expected to be around 93%, although the company’s engineers believe 98% is achievable.

Subject to board approval, Petropavlovsk is also planning to expand Malomir’s flotation plant, potentially beginning construction in late 2020. This would increase the capacity from 3.6 to 5.4 million mt/y of concentrate. Longer-term, a flotation plant at Albyn is envisaged for 2028 to process refractory ores at the Elginskoye and Unglichikanskoye deposits. The grade of these concentrates is expected to vary between 20 and 40 g/mt, and the average sulfur grade will be around 15%.

“The smooth ramp-up of the POX Hub stands as a record for the industry. It reflects more than a decade of R&D and is testament to the strength of the company’s scientific and engineering capabilities. At the onset, POX processed our own refractory ore but since July 2019 we have been treating third-party concentrates from several sources, quickly achieving the design recovery rate,” said Pavel Maslovskiy, co-founder of Petropavlovsk.

Technological improvements are constantly being made to the POX Hub. Current projects include: working to isolate and recover antimony as a by-product; reducing the amount of organic carbon in concentrates; and suppressing the sorption activity of a carbonaceous substance through heat treatment of concentrates which have passed through the autoclaves and low- and high-temperature POX.

In the future, Petropavlovsk said it might be able to process some very complex refractory materials, at the Hub including:

• Pyrrhotite-bearing concentrates;

• Cuprous and antimony gold-bearing concentrates;

• Triple-carbon concentrates;

• Concentrates from the tailings of existing mines; and

• Bio-cake.

---

Alternative Lixiviants

While cyanide is still the go-to reagent for gold leaching, there are operations where, for environmental, social or logistical reasons, it cannot be used. And, in these instances, it’s good to have alternative options.

Clean Mining, part of the Clean Earth Technologies Group, offers a non-toxic, non-flammable, reusable and water-soluble reagent based on an inorganic compound that can be used as an alternative to cyanide.

The technology, originally developed by CSIRO, is particularly aimed junior producers that don’t want the compliance and rehabilitation costs associated with cyanide. Clean Mining offers a mobile/transportable and scalable plant design that it said is suited to both large and small high-grade deposits.

The front-end milling circuit is the same as a cyanide-based plant — so the new technology applies only from the leaching to gold recovery phases. This makes for an easier transition, and there is also an option for a dewatering process to produce dry tailings.

“The plant itself is very similar to a cyanide plant with our technology fitting into the process specific to leaching and dewatering,” Kevin Fell , Group CEO, told E&MJ. “The solution is scalable from artisanal operations all the way through to major operations in gold processing.

“We have been overwhelmed with enquiries on a global basis. Our test plant in Menzies, Western Australia, validated the process, and a variation of our solution was provided by CSIRO, to Barrick Gold in Nevada in 2014 prior to us obtaining the IP. We currently have in testing several opportunities globally and will be rolling out several plants in the not too distant future.”