Paste: A Maturing Technology
Mining companies worldwide are gaining the benefits from optimizing their tailings-disposal systems
By Simon Walker, European Editor
Paste technology covers a range of applications, all of which are founded on one basic concept: reducing the moisture content of a processing plant's tailings stream to produce a material that remains sufficiently fluid to pump yet sufficiently stiff to gain early stability. Drivers for the use of paste technology include the need to conserve water resources, reducing the environmental impact of tailings disposal, hastening tailings restoration, and producing a material that has better properties for use as backfill underground.
Key to the more widespread use of paste technology, whatever the reason for its application, has been the development of the deep-cone (or deep-bed) type of thickener. As Fred Schoenbrunn from the U.S. operations of FLSmidth noted in a keynote presentation at the Paste 2011 conference earlier this year, John Dorr patented the thickener concept in 1906, with the concepts for deep-cone technology dating from the 1960s. Organized by the Australian Center for Geomechanics (ACG), Paste 2011 (the 14th International Seminar on Paste and Thickened Tailings) took place on April 5–7 in Perth, Western Australia.
There is a lot more, of course, to operating a successful paste tailings system than merely reducing the water content of the material before placement. The technology involved includes pumps that can handle slurries with a high solids content, pipelines that can withstand both the abrasion and the pressures involved, and the design of appropriate discharge facilities, whether the tailings are being sent for storage or for backfill. Add to these the need for a much greater understanding of the rheology of the material, and of the various reagents and admixtures that can be used to optimize settling on the one hand, and long-term stability on the other, and it is easy to see how complex this can become. The fairly lengthy list of projects that were less successful than anticipated, certainly in the early days of paste technology, bears witness to the need for careful planning, engineering and implementation.
So, what are the benefits? Lower operating costs is one key advantage, particularly in relation to long-term tailings management. Although the concept of "walk-away" storage has yet to be realized, there is no question that tailings deposited as paste can be restored earlier than would be the case in a conventional dam scenario, with lower long-term costs. Waste rock dumps capped with paste can be less amenable to internal weathering and sulphide-based acid formation, since tailings paste housed within voids in the dump can reduce the potential for sulphide oxidation.
Saving process water is another important driver, especially with the more widespread development of new mines in arid parts of the world. As an example, Antofagasta and Marubeni's new Esperanza mine in Chile, where capacity is currently being ramped up after commissioning began last November, uses paste tailings technology for this very reason. Part of the $2.6-billion capital cost of developing Esperanza has been spent on building a 145-km-long pipeline from the coast to the mine site at an altitude of 2,250 m, to carry sea water for the process plant.
Saving water wastage is critical, and as Antofagasta explains in its 2010 Annual Report: "Esperanza will achieve efficient waste management through a high-density thickened tailings deposit. This will be a global first for a large-scale copper project. The main benefit is that it will recover more water than a comparable conventional tailings system. Once deposited, the thickened tailings are more stable because they contain less water, and the desert conditions also allow them to solidify more quickly. The saline quality of the water in the tailings facilitates the formation of a stable crust that substantially reduces dust, further reducing the environmental impact.
"In recognition of this," the company said, "Esperanza was awarded both the Avonni prize as one of the most innovative mining companies in Chile, and also received the Benjamín Teplizky prize in Chile, again for its innovative technology."
Used as underground fill, paste offers advantages in terms of better drainage, quicker support integrity and better long-term stability. There is also the further advantage that—in common with other fill-based mining methods—the quantity of tailings for long-term storage is reduced significantly, again providing an environmental benefit.
As Schoenbrunn went on to explain, at low solids concentrations mineral slurries behave much like Newtonian fluids, in which the viscosity remains constant for different rates of shear and does not change with time. Once the solids concentration in the slurry is increased, he expanded, inter-particle reactions become greater to the extent that the material begins to show resistance to deformation. Pastes are non-Newtonian fluids, in that there is a non-linear relationship between shear stress and shear rate.
Compared with conventional slurries, in which the solids concentration is often no more than 35%, paste can contain up to 80% solids. The relationship between the coarse and fine particle constituents of the material holds the key to whether pumping and pipeline transport can be achieved successfully.
Inappropriate design in terms of the solids concentration, the pumping capacity and the pipeline parameters can lead to problems. High-solids pastes often move along pipelines in laminar, rather than turbulent, flow characteristics, and any segregation during the transport process can lead to the formation of a settled layer along the invert of the pipe. When this happens, there is no way to get the settled particles back into suspension, so the effect is to reduce the effective diameter of the pipeline and increase the pressure demand on the pumps to keep the material flowing at the required capacity. Hence there has to be a sufficient proportion of fine material within the tailings to support and bind coarser particles, preventing them from settling out.
However, success or failure can also depend on the application, as Dr. Angus Paterson of South African consultants, Paterson & Cooke, pointed out in his Paste 2011 presentation. Early successes achieved in pumping high-solids paste backfill underground "led many to believe that the same technology would easily be adaptable to surface disposal," he said. "However, the reasons for the success of paste backfill turned out to be the hurdles for surface pumping of the same material. Pumping viscous materials in small diameters when energy and power is not a concern, as in a backfill operation, turned out to be very different when attempting to pump the same material in larger diameter pipelines at high pressures."
According to Paterson, Preussag Metall was among the first mining companies to use high-density backfill—the forerunner of today's paste—at its Bad Grund mine in the Harz district of Germany, in 1979. In North America, Inco was a technology leader with paste backfill being introduced at its Garson mine in 1994. In each case, however, the concept was reliant on achieving a high enough solids concentration for the system to be viable, and this depended on the use of deep-cone and high-density thickeners.
The aluminum company, Alcan, was instrumental in this respect, albeit with a focus on handling red mud from its alumina refineries, while the coal industry was also an early developer for handling washery waste. High-rate thickening technology was also introduced in other industries, such as sugar processing, and was later transferred to mining.
Alcan licensed its technology to Eimco Process Equipment which, together with Dorr-Oliver, now constitutes part of the FLSmidth group. Outokumpu also took the concept forward, with its former technology company, Outotec, now a leading supplier of paste thickeners.Thickeners are the Key
Jason Palmer and Venkatesh Viswanathan from Outotec's Australian branch put some figures to the problem in their Paste 2011 presentation. A typical flotation circuit operates on a pulp density of 25%–35% solids, they noted, with subsequent sedimentation in a conventional thickener capable of increasing that to 60%. Using flocculants and a deeper thickener structure can increase this further to around 70% solids in the thickener underflow, at which point the question of pumpability becomes critical.
"Paste thickening is simply the art of driving the thickening process closer to the theoretical minimum limit of water in the thickened underflow," they said. "The process itself is not new; many existing thickeners have been known to produce paste. Unfortunately, this generally occurs right before the thickener bogs and the rakes stop.
"The challenge is not the formation of paste, but the design of thickeners capable of generating paste in a controlled and continuous manner."
The challenge here is to balance the release of water from the sedimented solids at the bottom of the thickener with the addition of solid material from new pulp being added from above, while simultaneously withdrawing paste-consistency underflow from beneath. "The time solids spend in the total bed does not govern the density achieved, only the time available for release of the water from the bottom of the bed," they said. In consequence: "The thickener must be designed to combine the effects of bed compression and solids flux rate to optimize the underflow density."
Schoenbrunn explained what this means in practice. "Deep cone thickeners are currently the best technology for achieving maximum underflow densities utilizing sedimentation equipment alone," he said. "These units typically utilize very deep mud beds in order to take maximum advantage of mud compressive forces for dewatering and provide sufficient time for the mud to dewater to a paste consistency. The tank height to diameter ratio is frequently 1:1 or higher."Which Pump to Use?
One specific type of positive displacement pump, the piston diaphragm pump, formed the focus for a Paste 2011 paper by Jack Kuenen from GEHO Pumps, the Weir Group company in The Netherlands. GEHO has developed and patented its GLORES system (Geho LOad REduction System), he said, which is a load-balancing system for single-acting piston diaphragm pumps. The upshot is that pumping stations for paste materials can be made more efficient, with higher capacities being handled by fewer pump units.
The hydraulic power available per pump has increased in steps over the past 30 years or more, Kuenen added, from around 500 kW (670 hp) in the early 1980s to more than 2 MW (2,700 hp) today. This has led to this type of pump being used on increasingly large-volume, high-pressure applications, such as iron-ore slurries and mine tailings.
Citing an earlier paper that had looked at the potential for transporting high-density tailings at rates of more than 100,000 t/d, Kuenen noted the design proposed there would require between 20 and 25, 1.3-MW pump units. By adopting the GLORES approach, the number of pumps could be more than halved, making such a system much more feasible economically.
And then there is the question of pipeline sizes and materials of construction. Clearly, there are differences here, depending on the precise application (underground backfill or surface tailings disposal) and the distances involved. Higher pumping pressures and more abrasive materials dictate stronger pipes, although lower transport velocities can offset some of the internal abrasion effects. Overriding all of the technical aspects of thickening, pumping and pipeline design, a good understanding of the rheology of the paste is essential. As Fiona Sofrà and David Boger noted in their Paste 2011 paper, even 10 years ago, tailings rheology hardly featured on the agenda, yet today it is recognized as being critical to the success or failure of a paste system.
"Both shear stress–shear rate and yield stress data are required as a first step for pipeline design and start-up, for the design of the suction side of the pump and for the design of the thickener; in particular for the torque requirements for the rake and rake-drive mechanism," they said.
In a paper presented at Paste 2009 in Viña del Mar, Chile, Steve Slottee and Jerold Johnson from U.S.-based PasteThick Associates looked at the situation from the other end of the pipeline. "Downstream requirements determine the underflow target and therefore the type of paste thickener," they said. "Installed paste projects repeatedly demonstrate the correct approach of a team of the thickener supplier, pumping/pipeline and geotechnical consultants.
"Sharing rheology data between the team is essential," they said. "Rheology is greatly influenced by the particle size—all parties need the same awareness of the particle size range. Design to a target yield stress, not wt% solids. Establish the rheology of the application, then the pump and pipeline required, and then the rheology requirements of the thickener."Applications Extend
Outotec has reported on its involvement in the design and implementation of a paste backfill plant at Boliden's Garpenberg lead-zinc mine in Sweden, which has at its heart one of the company's Supaflo high-compression thickeners. Cycloned fine tailings from the concentrator are thickened from 38% solids to 70% paste, while the coarse fraction is filtered to 90% solids, before the backfill is prepared from the recombined streams. A positive displacement pump is used to transport the fill to the mine delivery system.
Outotec cites advantages of this system as including the release of ore reserves that had previously been needed as hydraulic fill pillars, shorter fill cycle times, and increased safety, as well as improved water management.
However, the real advance in paste applications has come not from underground, but in surface tailings storage. Major advantages here include greater long-term stability of the material, shorter consolidation times, the need for smaller storage areas and, critically, better use of water resources.
In a conventional tailings dam, run-of-mill slurry is discharged from pipelines and finds its own level within the saturated pond. Coarse particles segregate quickly to form a beach, while fines are washed into deeper water where they slowly settle.
In a paste tailings application, the material extruded from the pipe end is viscous and is unlikely to segregate. It will not flow away from its emplacement point, so forms a steeper beach that consists of both coarse and fine fractions. Less moisture has to be lost before the surface has consolidated, while any cracking that occurs as the surface dries out forms a good key for subsequent layers as they are deposited.
At Paste 2011, Shiu Kam and a number of colleagues from Golder Associates and Goldcorp outlined the paste tailings disposal system now in operation at the Musselwhite gold mine in Ontario. The challenge here, they said, was that at 13.7 million mt, the original tailings storage area was too small to contain the expected 32-million-mt long-term output from the mine. Increasing the depth of the existing tailings dam was not considered to be feasible, so thickened tailings deposition offered a way ahead.
The system was commissioned last year, with thickened tailings being transported through two 150-mm-diameter feeder pipes that in turn supply a series of discharge spigots. The thickening plant itself is situated close to the tailings storage area, reducing the pumping duty for the thickened material. Kam and his colleagues concluded that the option of adding storage capacity by stacking tailings within the existing tailings management area was "cost effective, more flexible and environmentally superior to other options considered."
Without doubt, the use of paste technology for handling tailings will become more commonplace as time goes on. From niche beginnings, the concept has found a far greater range of applications, for a variety of reasons. Better water-resource management and having a smaller environmental footprint are just two of these. Improved economics have already been demonstrated, and this, if nothing else, will surely provide a spur for further future development.The Expert Viewpoint
Lee explained paste backfill and surface disposal are now reasonably mature technologies compared with even 10 years ago when they were not widely used and there were few practitioners who had any experience with them. "As such, a lot of the progress that has been made has been in the fine-tuning of designs and perfecting operating methodologies," he said.
"In addition, the envelope has been extended somewhat to service more difficult paste applications—higher tonnages, tricky rheological properties such as those due to flash setting or phase changes in the tailings material, and applications such as pumping up large vertical heights from the bottom of a mine rather down from the top."
"In terms of above-ground storage, the key technology developments have been in the design (and physical size) of the thickeners, feedwell efficiency and in the capability of centrifugal pumping systems to handle higher pressures. Filter systems also appear to be gaining credibility in terms of efficiency and cost-effectiveness, for certain applications, throughputs and objectives," Jewell said.
Looking at the reasons why paste technologies have not been adopted more rapidly, Lee said it is generally more difficult to change a mine from an existing backfill or disposal system, than to install the concept at a greenfield site where no prior system exists. "Frequently, this is because the capital investment has already been made for the alternate technology, and any change must be driven entirely by lower operating costs that override the capital costs that are not otherwise needed, or by technical advantages or regulatory requirements where the mining companies are required to switch technologies, despite the fact that it may not be justifiable on a purely economic basis."
The mining industry is inherently conservative," said Jewell. "This will remain an issue until at least one large operation proves to be successful (rather like the drawn-out introduction of CIP/CIL technology into the gold industry). The physical aspects of retrofitting an operation from the production and discharge of conventional low-density slurries to high-density slurry is not great, but the economics will be best determined on a case-by-case basis. However, I believe conservatism is endemic in both operating and corporate thinking on this issue," he said.
Responding to a question about the key issues that companies need to understand in relation to adopting paste technology, Lee provided three guidance points. First, he said, paste may not be the best solution. "Clients should always do a conceptual study to look at all the options. Because paste is currently in vogue doesn't mean that it is the best solution for every case.
"Second, the comparisons between various technologies should be looked at in a broader perspective—for example, the cost of backfill may be more expensive with paste than with rockfill. When the difference in dilution costs is looked at between the two types of fill then the equation is frequently changed. The whole picture needs to be looked at.
"Clients should make sure the consultants or in-house resources they use have the requisite experience," he said. "Since paste is relatively new, the amount of expertise in the area is small, and it is recommended clients employ designers who have personally executed a number of paste plants from design to construction to commissioning. In other words, a corporate resumé is not enough, and clients should demand individual design-team members who have a track record in building paste plants," he said.
"The benefits relate very much to the long-term license to mine, and costs must be considered on a 'whole life-of-mine' basis. Achieving the highest density through thickening (or the maximum recovery of water) is not necessarily the most efficient and economical process," he said.Investment Implications
Turning to the challenges involved in using paste technology in underground fill systems, he cited access to the stopes, with lots of pumping sometimes being required; having too much head available—which must be dissipated; pipeline wear; interference between piping installation and current mining operations; minimizing cement usage (and therefore minimizing cost); and quality control of the paste mix, among other issues.
"Additives such as superplasticizers and water-reducing agents are effective and can reduce the amount of cement required," he said, pointing out the cement savings may not be enough to pay for the cost of the additives. "This needs to be evaluated on a case-by-case basis, and sometimes the use of additives does make economic sense."
Additives such as retarders can be used to delay the paste curing, and are effective in that sense, he explained. "The time required for curing depends on how much cement is added. If you want a shorter cycle time, then you need to add more cement to reach the target strength within the allowable amount of time. Paste is beneficial compared to hydraulic fill since there is no percolation time required, and curing starts as soon as the cemented paste enters the stope."
As a final point, E&MJ asked both experts for their predictions for paste applications over the next five to 10-years, and the potential hurdles to be crossed. Lee said, "I think we will see more really big surface tailings-disposal systems go into production and will be able to see how very large paste-deposition systems perform. I think we will see more use of advanced materials such as ceramic-lined piping, and advanced equipment such as electrokinetic dewatering. We will also see a lot more regulation-driven paste projects, where public pressure forces mining companies to adopt best practice for the environment rather than the most cost-effective practice."
Jewell said, "The potential for polymer injection at discharge to enhance water release from the thickened tailings could be promising (as described in papers presented at Paste 2011), but this is still in its early stages and will need to be first proven in practice at field scale. As long as miners grind to smaller and smaller sizes to increase mineral return, and this finer particle size distribution retains greater volumes of water, the practical challenges of thickening the resulting slurries will only increase, while filtering—which is only efficient for the coarser fraction—will become less viable.
"I believe, the major challenge is to convince one major company to 'take the leap' to adopt the technology to a large-throughput operation and for the operation to be successful. Part of this relates to a number of suppliers promising water recoveries and beach slopes that are unlikely to be achieved in the field because of practical issues. The worst thing that can happen is for a large-scale installation to be installed and then fail to meet the design criteria."Acknowledgements
The next meeting in the annual series, Paste 2012, will take place at the Sun City Resort in South Africa on April 16–20, 2012. Information on the meeting can be obtained from Jacqui van der Westhuizen, head of conferencing, Southern African Institute of Mining & Metallurgy, at +27 11 834 1273 or by e-mail at Jacqui@saimm.co.za.
Pump Supports Backfill Operation at Mexican Mine
A batch plant at Agnico-Eagle’s Pinos Altos gold and silver mine in Mexico delivers
a steady 126 tons per hour of paste backfill material to the mine’s three current main
stopes. (Photos courtesy of Schwing-Bioset
Paste backfilling, the process by which a combination of tailings, water and cementacious materials are blended and used to fill voids in underground mining operations, provides a broad range of benefits. These can include an improvement in safety, the ability to make subsequent surface development possible, a reasonable solution to the ever-present problem of what to do with tailings, and more. It’s not surprising, then, that Canadian miner Agnico- Eagle chose paste backfilling as the key tailings disposal method for one of its newest developments, the Chihuahua, Mexico-based Pinos Altos mine. But while creating the paste is one issue, get- ting it to the mine stopes—which in this case can be anywhere from 1.5 to 2 km away—is another one entirely. To make that hap- pen, Agnico-Eagle turned to a pump model from Schwing Bioset (Somerset, Wisconsin, USA), and is reporting impressive perform- ance delivering material in this around-the-clock operation.
Back into the Ground
“The tailings operation was designed with this procedure in mind,” said Palma. “Currently, we take waste material by convey- or from the milling plant, and run it up an inclined belt into a batch plant. There, the tailings make their way through a process in which they are combined with cement and water to create a paste. The cement—roughly 5% of the overall mixture—is added to make the paste suitable for filling and supporting the existing underground cavities that have already been mined. Once the cor- rect mixture is achieved, it is dropped down into the hopper for the paste pump and ready for delivery back to the mine.”
Mine backfilling is hardly a new concept. What has changed over the years, however, is the characteristic of the backfill materi- al itself. In the past, to accommodate the limited pumping tech- nology available in most mines, tailings had to be turned into a slur- ry with a water content of 70% or more. By comparison, the paste material generated at Pinos Altos is stout, with a water content of roughly 30% and a slump generally in the 8-in. (203-mm) range.
Technology Changes the Game
A Schwing Bioset KSP 140 H(HD)XL piston pump, driven by an 800-hp power
pack, moves paste from the batch plant to the stopes about 1.5 km away.
The difference between what could once be pumped and what can be moved today has been mining companies’ preference for pis- ton pumps over centrifugal pumps in tailings operations. Where other previous technology offered limited pumping pressure (often as low as 10 bar), the Schwing KSP 140 H(HD)XL pump in place at Pinos Altos provides a maximum conveying pressure of 130 bar with paste outputs up to 85 m3/hr—ideal, according to Palma, for moving the high-solids material long distances. “The Schwing pump we use here at Pinos Altos is perfect for what we do,” he said. “This is a challenging application, given the thickness of the paste, the distance it has to be pumped—cur- rently about 1.5 km—and the pump’s almost continuous opera- tion. Yet, we are getting a steady rate of about 126 tons of mate- rial pumped every hour—roughly 2,500 tons per day. We have been extremely pleased with that kind of performance.” Based on the nature of the paste and the desired production rates, engineers from Schwing-Bioset recommended its XL pump option for the Pinos Altos operation. Poppets in the XL
series are designed to reduce material velocity through the pop- pet housing. Doing so minimizes the pressure drop through the valve housing, increases the filling efficiency of the pumping cylinders and reduces wear on the poppet discs, seats and pumping rams—all of which lead to better performance and less downtime. To keep the pump running and the paste flowing, Pinos Altos relies on an air-cooled Model PP2400 drive unit that is powered by twin 400-hp (300-kW) motors, also supplied by Schwing Bioset. This is mounted adjacent to the pump at the base of the paste plant, and includes PLC-based controls. While the initial plan was for the pump to operate up to 10 strokes per minute and deliver 126 t/h, operations are running so smoothly that mine offi- cials are looking to raise production another 20%.
Safety Leads the Way
“We are pumping into three main stopes and doing so in three- foot lifts with three days between pours,” he said. “Not only does that allow enough time for the paste mixture to harden, it also avoids placing the massive stress load on that cell and adjacent areas that would be present if we filled it all. This first cell alone will be taking on more than 13,000 tons of material, which is over 8,300 m3 of paste.”
To put that figure into context, 8,300 m3 ( 10,855 yd3) is enough material to bury an entire basketball court over 13 ft (4 m) deep. Though the other two stopes currently being filled are small- er—one will hold 7,100 tons and the other 6,600 tons—they still represent a sizable movement of material from the paste plant to the mine. Discharge into the cells is monitored via closed-circuit camera to ensure the process runs smoothly and any problems can be quickly spotted.
“This has been a very successful approach to tailings disposal for us,” said Palma. “As a result of what we do here, the mining operation is safer, there is almost no impact on the environment and, over time, it is actually a lower cost alternative to other meth- ods. Agnico-Eagle has already started talks to boost production by almost twice what we are doing now. When that happens, we will add another pump from Schwing Bioset which will allow us to han- dle between 5,000 and 6,000 tons per day. We know they will be up to the task and we are excited for what lies ahead.”