While a commodities boom is welcome news for the industry, it also poses significant challenges - including how to handle much higher volumes of mine waste. We are all familiar with the operational risks associated with tailings storage facilities, highlighted by the collapse of the Brumadinho tailings dam in Brazil in 2019 which killed 270 people and destroyed more than one million square metres of forest.

More recently, a tailings dam near the historic Jagersfontein diamond mine in South Africa's Free State province burst in September 2020, killing one person, injuring many others and causing damage to houses and cars.

Other challenges associated with tailings facilities include the need to mitigate potential environmental impacts, such as impacts to biodiversity from water pollution, and the difficulty of managing the massive quantities of water consumed in the tailings process. This raises a multitude of questions, including one that was the focus of a recent Mining Journal panel discussion: how important is it that mining companies embrace new and smarter technologies to help them deal with the risks associated with tailings?

Embracing new technologies… and building on the old

Our panellists agreed on the importance of developing new technologies to manage risks associated with tailings, but also stressed the need to build on technologies already being used in mining and other sectors.

Dan Leybourne, Product Owner from Shell at OREN, a partnership between Shell and IBM offering an Intelligent Tailings Storage Facility (TSF) management solution alongside other digital decarbonisation products for mining companies, said the sector can use the opportunity of having to address risk in tailings to build on existing technology and develop understanding from what is already known.

"One of the challenges we have is that tailing properties change spatially over time, and they require constant review against targets to ensure we have meaningful metrics that are linked to preventative controls. When we talk about risk management, how do we move from looking at understanding historical performance to proactively managing and predicting performance in the future?" Leybourne said.

"We can build on what we know already but the criticality is leveraging that information in a different way. When we're leveraging it, when we talk about being smarter, it's not necessarily about using new data and digital technologies; the technologies are already there. It's just about applying it quicker and accelerating its implementation… rather than historically a slower approach to changing the way we do things because of the risks associated with tailings."

Ken Rahal, Director of Tailings Solutions at multinational engineering firm FLSmidth, agreed that much of the tailings equipment currently in use has stood the test of time. But, he said, "over time, orebodies have gotten to finer grinds, some of them [have required] higher tonnages to deal with those lower grades, mineralogies have become more complex, and it's led to a lot more variability. That's where these smarter technologies are taking and building upon existing technologies. Linking them together, you have more inputs on what's actually going on in the process to better control the equipment."

Dr. Priscilla Nelson, E&O Director at the Tailings Center and Professor of Civil, Geological and Mining Engineering at Colorado School of Mines, noted that work is being done to enable existing technologies such as filtration to be used in larger-scale operations. She also expressed a wish to see electrokinetics, another technology that has "been around for decades," used in tailings.

Echoing Leybourne's comments about spatial variability over time, she said the sector doesn't do a good enough job of characterising the material coming into tailings facilities. "If we did have real-time characterisation, we might be able to make different decisions about where to place or how to manage. We need that kind of spatial and temporal variability in order to understand how to optimise existing thickening or filtration systems," she said.

"I think filtration's coming on. I call it a new technology, although it's [really an old technology that is] getting bigger; it's like the dinosaurs at the end of the Cretaceous, they're just getting bigger and bigger and bigger. But it's not the silver bullet; it works really well in some cases and in other cases probably shouldn't be tried. I don't think we know the limits of it." Ultimately, the big challenge is to get the water out of the tailings, in Nelson's opinion.

"If you do that the tailings behave more strongly, better behaved and the water can be returned into the water balance management on the site, so it's extremely important," she said. "I think it's time to move to an electrokinetic gradient that will drive the flow. This is not new technology, it's been around for decades but it's never had this largescale pilot operation. I'd be interested in combining electrokinetics with some of the thickening operations, and even to look at the material that's in the impoundment, the tailings slurries, to be able to thicken it in-situ in the ponds."

Treating tailings as a whole system

At the Tailings Center, Nelson and fellow academics from the Colorado School of Mines, Colorado State University and the University of Arizona guide industry stakeholders on best-available practices for the design and operation of tailings management systems.

Nelson spoke during the panel discussion on the importance of thinking about tailings as a "whole system", beginning with the flow of material from the orebody and continuing all the way through to its arrival at the tailings facility.

"I think we can go all the way upstream to reduce the volume of tailings," Nelson said, listing selective and in-situ mining as well as selective comminution and processing as methods enabling operators to extract ore but leave waste in the ground.

Mineral processing should also play a role, she said. "We have [seen] a drive towards coarser grinding, reducing water and recycling and reuse of agents. Lots can be done in the processing, including microwave technologies which are being used in some cases to very selectively break down the material… without going down to the extremely fine grade sizes. But with stoping and low-grade mining, we're always going to be down in the extremely fine tailings that are very hard to manage - unless we can go upstream and look at exactly how those low-grade ores need to be processed."

FLSmidth's MissionZero similarly focuses on the entire flowsheet, taking a holistic approach to reduce water usage in tailings. As Rahal said, every mine site has its own unique orebody characteristics, and "once you understand that, you can treat the system from the beginning to the end." He elaborated: "Sometimes with tailings you need to start at the end and work backwards. Once you have the target that you need in your tailings area, you [ask] can you achieve it with just cyclones and pumps, or do you need to filter it or thicken it? If it's a large tonnage mine you need to look at larger filter presses to understand their cost, at how you're going to handle the material after it's filtered, whether for belt or pipe conveyors, the type of stacking system you're going to use, and how that is affected by the terrain."

FLSmidth has looked at "every nook and cranny" of the mechanical side of its filtration systems in an attempt to bring down costs, such as introducing more-effective plates, changing wear parts and extending media life. But he said there was also a place for control systems to take in large quantities of data and "make sure we're not just producing the concentrate we want for profitability, but that we are producing tailings that are going to meet the storage area's requirements."

For its part, OREN offers an end-to-end tailings solution which utilises data and advanced analytics to address pain points across the tailings process.

Leybourne commented that digital solutions providers had moved into real-time monitoring and were now starting to leverage the predictive space. This was enabling operators to look into potential scenarios, the risks, and how these risks evolve over time as variables change, he said. "What we're seeing in this space is optimising strategies that are site specific. Those physical and chemical characteristics of tailings change again with the process methods, topology, climate conditions, and even the regulations between countries. We are starting to see a pull to help acquire and manage data, to visualise information not only improve understanding of our current performance but predict and manage future performance ."

Dawn of the digital tailings era

In the final part of the panel, the experts spoke about the areas of tailings where tailings operators and technology developers should be focusing their efforts. The consensus was that digitalisation would be critical to addressing a range of challenges associated with tailings facilities - including the need to characterise tailings over time and space, as well as the need for transparency. The Global Industry Standard on Tailings Management, which was published in 2020 in the wake of the Brumadinho disaster, requires operators to commit to transparency. Furthermore, it requires operators to participate in global initiatives to create standardised, independent, industry-wide and publicly accessible information about tailings facilities. "The push to better understand and monitor and record and have all the sensors in place feeding into a larger control system will enable that transparency without it being a nightmare for the sites to manage," Rahal commented.

"If you look at the past, one of the biggest problems has been publicly available data about the industry. It's very difficult. We've done research projects with universities, and students have spent weeks trying to find information and it's not available. A lot of that information is difficult to get without proper sensors. The hope for me is we move towards these smarter systems where data is more readily available to be more transparent publicly, so that the industry can pull that knowledge together to learn."

Leybourne said that transparency begins with understanding through our data, adding that the next step is to leverage predictive technologies to ensure we have the performance we want and need as well as revealing new ways of looking into managing the tailings process.

"When we look ahead, how are mining companies partnering and leveraging their data - what does that look like? There will be more collaboration that comes in the future around tailings, particularly [looking at] what can we do to get the right information to understand that performance. I think when you specifically look at digital, what we're seeing is a combination of digital technologies into a single solution - [including] sensors, analytics, visualisation, automation and predictive capabilities - rather than isolated solutions," Leybourne said.

"As we move in that [digital] space, we're going to be leveraging understanding from different businesses to accelerate mining. When we look forward, we need to look at how we work better together with those outside mining, because they also have a lot of creativity and innovation at applying solutions that we have no idea about that could solve some of our problems. If we get the right datasets to them, we can excite them to work and play in this space - because we need it to be fun and we need to start with how can we be at the cutting edge of that type of innovation."

Rahal and Nelson echoed the call to collaborate with those currently outside the mining space to develop smarter solutions, but warned that it would not be easy.

"In the past it's been difficult to get young people interested in the industry - but as we move into the digital age kicking and screaming at times, the hope is we will get the younger generations brings new ideas and allow a positive effect not just on the labour pipeline but the idea pipeline. They'll bring new ideas that help us become a better industry and solve our problems and where we're going in the future," Rahal said.

Nelson said the mining industry was "largely disengaged" from academia, adding that there were almost no faculty members doing research on tailings and almost no universities in the United States teaching anything about tailings at the present moment. As a result, "the students are not coming out. [If] there is a connection - and this data and transparency has a potential to be a connection - where faculty can look at this data and say I'm going to spend part of my career messing around with data on tailings and see what kind of contribution I can make, then they will teach courses, produce students and we'll have a workforce," she said.

"We have to connect the pipeline because it's broken right now. This whole discussion of new technologies - if we do it the right way we can recreate a pipeline that the industry needs for the future."

i For recent long-term demand forecasts for critical metals, see Benchmark Minerals, S&P Global and McKinsey.