Coal Preparation » Dewatering
The processing, transport and management of ultrafine coal tailings often represents major financial and logistical concerns for owners and operators of collieries. The problems almost invariably relate to the volume of water that has to be managed when dealing with the ultrafine tailings. There are also increasing concerns about the cost and availability of water, with pressure on mining companies to reduce their water consumption rates likely to grow. Managing surface deposits of material having a very wet consistency, and thus low shear strength, also represents a potential hazard to workers and to adjacent communities, plus making final closure and rehabilitation more difficult.
There are a number of established technologies for reducing the water content of coal slurries before impounding the slurry. This report deals with a technology that is not well established in the mining industry, although it has been around for about two centuries. It is the process of electrokinetic dewatering, whereby water is extracted from a slurry through the application of direct current voltage using two or more electrodes. Water flows towards the negatively charged electrode, the cathode, where it can be collected. Although the concepts are old, implementation of the technology for large scale dewatering has not occurred to date. Typical problems with implementation of the technology using traditional metallic electrodes include very rapid corrosion of the anode, difficulties with collecting the water flowing to the cathode, and loss of contact with the electrodes as the treated soil or tailings dries out.
The development of conductive polymers has resulted in the availability of electrodes that can address the above problems. By forming the polymer into tubular wells, around which a filter cloth is wrapped, the problem of water collection is solved, and improved contact is assured because of the greater surface area of the polymer electrodes. Corrosion is also significantly reduced as there is no longer any exposed metal. This ACARP project sought to verify that the findings of work carried out about two decades ago on coal tailings was valid when conductive polymer electrodes (commonly known as EKGs - ElectroKinetic Geosynthetics) are used. This earlier work had illustrated that dewatering of ultrafine coal tailings and coal product was technically feasible, although the problems with metal electrodes (discussed above) were a concern. Specific objectives of this current ACARP project were to:
- Verify the applicability of electrokinetic dewatering to a range of ultrafine coal products and coal tailings.
- Characterise the dewatering efficiency and effectiveness and consequent potential economic benefits for different electrokinetic treatment regimes for both in-situ and in-process dewatering methods using EKGs.
Fine and ultrafine coal tailings were obtained from six collieries, namely Bengalla, Bulga, Dawson, Griffin, Peak Downs and Wambo. Samples were generally supplied in 200 litre drums, and represented a range of material, with some of the tailings having fines contents (finer than 75μm) of 70% and in excess of 50% finer than 2μm. A range of characterisation tests were carried out on all samples, including particle size distribution tests, Atterberg Limits, zeta potential and X-ray diffraction.
Two different testing procedures were used to determine the viability of dewatering these materials electrokinetically. These were bench top tub-tests, and the use of a purpose built electroosmotic (EO) cell. The tub tests required about 25 litres of material and utilised actual EKG electrodes. A DC voltage was applied using a power supply, and the water collected in the cathode drain monitored over time. In addition, control tests were carried out in which no voltage was applied, in order to ascertain how much of the dewatering was due to evaporation. The tubs were placed on scales throughout the test duration, so the total mass loss could be monitored. The total mass loss consists of three components, which are the volume extracted from the cathode, the evaporation loss, and the volume volatised at the electrodes as a result of the electrolytic reactions that occur. The water content before and after each test was measured, and the current drawn by the system was also measured throughout the tests. This enabled the power consumption rates and efficiencies to be calculated. In the majority of the tests, a constant voltage of 12 V was applied throughout, although in two of the tests, the technique of intermittent power application was used, where the voltage was switched on and off, typically at 15 minute intervals. International experience has shown that this reduces the power consumption, while usually achieving the same rate and extent of dewatering. This was confirmed in this study.
The second test used the EO cell, in which both stress or voltage (or both) can be applied to a specimen. In these tests, the sample was initially consolidated in the cell under a vertical stress of 20 kPa and the drainage outflow measured. A relatively small (0.5 V/cm) voltage gradient was then applied, and the drainage measured again. The ratio between these two flow volumes was used as an indicator of the relevant benefits of electrodewatering versus applied stress (which generates hydraulic dewatering, much like what would occur in a filter press).
All of the tailings responded favourably to electrokinetic treatment, although the response of the Dawson tailings was much poorer than the other five materials. The Griffin tailings provided a surprisingly good result, given the relatively coarse nature of the material. The remaining four materials (Bengalla, Bulga, Peak Downs and Wambo) all responded extremely well, and very substantial dewatering was possible. For example, four tests on Wambo tailings were completed in the tubs, and the longest running of these (35 days) improved the solids content from an initial value of 30% to 47.6%. The four tests on Wambo tailings all started from different initial solids content values, ranging from 30 to 45%, and effective electrokinetic dewatering was achieved in all four cases.
The tests in the EO cell confirmed the data obtained from the tub tests, with the Dawson tailings again performing poorly relative to the other five materials. Although the EO cell should provide useful fundamental property information because the applied electrical field is uniform (one-dimensional), the drawback with the cell is that it uses metallic (brass) electrode disks, whereas this study was concerned with the viability of using EKG electrodes. Therefore although the tub tests appear to be less sophisticated than the EO cell tests, they at least utilise the correct electrode material, and also effectively represent a scaled-down version of a possible field installation. For these reasons, the tub tests were used to calculate power consumption and efficiency rates for the different materials, and the cell tests were used to verify these findings.
The literature on electrokinetic dewatering can be misleading when it comes to power consumption rates. This is because the unit that is usually used is kWhour/tonne. This unit of measure does not distinguish between material that is initially very wet (low density) and that which is very dry. In the case of initially low density material (i.e. low solids content), dewatering might remove a significant volume of water, but result in relatively little increase in solids content. The opposite is true for a material with a high initial solids content - removal of a relatively small volume of water produces a big change in solids content. The unit of measure adopted in this study was therefore kWhour/dry tonne, which removes the misleading effect of initial solids content, and focuses on the material that is being dewatered - the mass of solid material. For the tests conducted in this study, power consumption rates of between 5.9 and 37.8 kWhour/dry tonne were achieved. The high value (37.8) was for the test on Wambo tailings that was run for the longest time (35 days) and achieved the greatest volume of water reduction. Thus even the alternative method used in this report is not infallible, as the relative effect of initial and final solids content and duration of test are not adequately captured. Alternative suggestions for dealing with this issue are discussed in the report, but the more usual definitions are retained in this summary. The efficiency of the process is sometimes defined in terms of millilitres per milliamp-hour (ml/mA-hr). The best result achieved was 7.5 ml/mA-hr (Wambo tailings), with most values being around 2 to 6 ml/mA-hr.
There are two different procedures through which electrokinetic dewatering could be implemented using EKGs. The first is in-situ dewatering of existing coal tailings (or coal product) ponds or lagoons, and the second is in-plant dewatering using filter belt machines or filter presses that have been retrofitted with EKG belts.
The first option, in-situ dewatering, has been successfully trialled at a small scale (3 m diameter, 1.2 m deep tank) on mineral sand slimes. What is needed now is to conduct a relatively large (say one hectare) field evaluation of the technology, where practical issues with the implementation of the system can be identified and dealt with. In such a trial it is envisaged that the EKG electrodes would be installed at about 2 m centres, potentially using conventional vertical drain installation techniques. If a one hectare pond, 20 m deep, were to be dewatered (hypothetically) from 45% to 55% solids, this would require the extraction of about 46 000 m3 of water. Using the relatively conservative value of 12.5 kWhr/dry tonne from this study, this means the required dewatering would be achieved at around 1450 kWhr. In addition, the limited amount of work carried out in this project on optimisation of power consumption indicates that the figure of 12.5 kWhr/dry tonne could be significantly reduced if the system were to be fully optimised.
The second option, which is for in-plant dewatering, would require the retrofitting of a conventional filter belt machine with an EKG belt. This has already been done in the UK, and trials have been underway for about a year in South Africa, where kimberlite tailings (from diamond mining) are being tested using the retrofitted machine. Kimberlite tailings are notoriously difficult to dewater, consisting primarily of smectitic clays. The results published to date have been extremely positive, with solids contents of as much as 85% achieved. Corresponding tests have not been carried out to date on coal tailings, but using the experience gained from the current work on diamond tailings, together with the information on coal tailings from this ACARP project, it could be possible to dewater an amount of 50 dry tonnes per hour using approximately 1000 kW. This estimate is necessarily speculative, as the required tests on coal tailings have not been done, and the current trial in South Africa is only treating about 25 tonnes per hour. The machine being used is relatively small, and scale-up is of course possible. In addition, work is currently taking place on retrofitting an existing frame filter machine with EKG filters, which could improve throughput rates further. To facilitate implementation of the technology, the next step would be a site trial using a retrofitted filter belt press (and perhaps a frame filter) to determine viable throughput rates and achievable solids contents.
In summary, the current project has again proved the technical viability of dewatering ultrafine coal tailings electrokinetically. Before the technology can be adopted at full scale, however, it would be important to conduct field trials. Depending on interest from potential sponsors, these trials could be either in-situ dewatering, or in-plant dewatering, or both.