Tailings Dam Failures, ARD, and Reclamation Activities John A Meech Professor of Mining Engineering The University of British Columbia Email: [email protected] Outline • Tailings Dam Construction Methods • Tailings Dam Failures • Reclamation of Dams, Waste Piles, and Sites • Britannia Beach and the Millennium Plug Project • Atmospheric Risks at the Sullivan Mine • Acid Rock Drainage – what is it? • ARD Control Methods • Microbiology of ARD Issues Stability of dam structures a. Use borrowed coarse material b. Cyclone tailings to extract coarse fraction c. Control pond water level so ground water does not enter the structure (phreatic surface) - Use barge/pump system - Use a tunnel/overflow tower system Water-Retention Type Dam Steven G. Vick, 1983. Planning, Design, and Analysis of Tailings Dams, John Wiley & Sons, New York, pp. 369, ISBN 0-471-89829-5 [The textbook on the subject! A reprint was published in 1990 by BiTech Publishers Ltd., Richmond B.C., Canada (ISBN 0-921095-12-0) Sequentially-built Tailings Dams Each lift requires more material – 1,3,5,7, etc.) Each lift requires more material – 1,2,3,4, etc.) Sequentially-built Tailings Dams Sequencing of Up-steam Tailing Dam Lifts Phreatic Surface in Upstream Dams kL = permeability at the edge of the pond water at the slimes zone k0 = permeability at the spigot point (dam crest) kF = permeability of foundation kh / kv = anisotropy ratio (horizontal vs. vertical) Ring Dike Construction - Kalgoorlie Valley Deposit - HVC Cross-Valley Plan View CROSS VALLEY IMPOUNDMENT - SINGLE AND MULTIPLE (Extracted from Vick, 1983. Planning, Design, and Analysis of Tailings Dams) Side-Hill and Valley-Bottom Plan Views SIDE-HILL and VALLEY-BOTTOM IMPOUNDMENT - SINGLE AND MULTIPLE (Extracted from Vick, 1983. Planning, Design, and Analysis of Tailings Dams) In-Pit Storage Underground Storage • Hydraulic sand – Cycloned tailings sand (coarse fraction) • Cemented fill – Required to fill void space and create strength • Paste backfill – All tailings dewatered to 60-65% solids • Dry rock fill – With and without cement Paste Backfill - Lisheen Mine, Ireland • Backfill plant with deep cone thickener Hazards for Tailings Dam Stability Two Major Hazards: • Excessive increase in level of pond water on impoundment • • • • Operational error during filling Natural events (thunderstorms and/or flood inflow) Beach width between the water and dam crest becomes too small Phreatic surface rises in the dam and leads to collapse • Liquefaction during an earthquake • Tailings may change physical properties under seismic stress • Cyclic stresses can lead to liquefaction • Highly susceptible due to low bulk density and high saturation • Hazards are not theoretical • Many tailings dam failures prove the theories over and over again. • Recent example - Harmony gold mine tailings dam in South Africa (Feb. 1994) after heavy rainstorm - village completely buried - 17 people killed Water Balance in a Tailings Dam Up-steam Tailing Dam Typical Failure Up-steam Tailing Dam Piping Failure Up-steam Tailing Dam Failure too rapid rise - must be < 15 m/year Up-steam Tailing Dam Failure over-topping Up-steam Tailing Dam Failure liquifaction Up-steam Tailing Dam Failure slope stability Comparison of Surface Impoundment Types Water Retention Mill Tailings Requirements Discharge Requirements Water Storage Suitability Upstream 40-60% sand in tailings. Suitable for any Low feed pulp density type of tailings to enhance size segregation Peripheral discharge Any discharge and well-controlled procedure suitable beach necessary Good Not suitable for significant water storage Downstream Centerline Suitable for any Sands or low-plasticity type of tailings slimes Varies according to design details Good Peripheral discharge and nominal beach necessary Not so good for permanent storage. Temporary flood storage adequate with proper design Comparison of Surface Impoundment Types Water Retention Seismic Resistance Good Raising Rate Restrictions Entire embankment constructed initially Embankment Fill Requirements Natural soil borrow Relative Embankment Cost High Upstream Poor in high seismic areas 4.5 - 9 m/yr desirable. > 15 m/yr is hazardous. Downstream Centerline Good Acceptable None Height restrictions for individual lifts may apply Sand tailings or mine waste if Natural soil, sand production rates tailings, or mine waste are sufficient, or Natural soil Low High Sand tailings or mine waste if production rates are sufficient, or Natural soil Moderate Tailings Dam Failures • From 1968 to August 2009 - 149 documented failures worldwide • 3,500 tailings dams exist around the world 25,000 to 48,000 large water storage dams exist around the world. • Tailings dam failures closely match water storage dam failures So, failure frequency is far higher (an order of magnitude). • Since 2001, the failure rate is roughly one every 8 months. • 85% of incidents were Active tailings dams / 15% Abandoned dams • 76% of incidents were Upstream construction methods • 56% of incidents were dams greater than 30 m in height M. Rico, G. Benito, A.R. Salgueiro, A. Díez-Herrero, H.G. Pereira, 2010. Reported tailings dam failures. A review of the European incidents in a worldwide context. 20th Century Tailings Dam Failures Ten Causes of Failure ________________________________________________ Type of Failure Number % ________________________________________________ Unusual Rainfall 36 24.5 Seismic Liquefaction 21 14.3 Poor Management Operation 15 10.2 Structural Failure 13 8.8 Piping/Seepage 10 6.8 Foundation Failure 9 6.1 Overtopping 9 6.1 Slope Instability 7 4.8 Mine Subsidence 3 2.0 Snow melt 2 1.4 Unknown 22 15.0 _________________________________________________ TOTAL 147 100.0 _________________________________________________ Dam Failures due to Management Issues • Poor beach management • Faulty maintenance of drainage structures • Inappropriate dam procedures – rapid dam growth • Heavy machinery on top of unstable dam Real-Time Monitoring of Tailings Dams • • • • • Piezo-electric gauges Pore pressures at depth Both horizontal and vertical directions Control of barge pumps Controllable CCD cameras – On top of dam structure – Along all diversion ditches • Water levels in all collection ditches/drains Piezo-electric Gauges Basis of piezoelectric effect: - crystals under compressive loading generate an electric charge directly proportional to force applied. Piezo-electric Gauges Strain gauge transducer with bridge circuit Charge is amplified into a proportional output voltage Piezo-electric Gauges • Piezoelectric sensors are small in construction • Their high natural frequency is ideal for dynamic measurements. • Virtually no displacement, as quartz gives mechatronic component with an electrical output signal. • Sensitivity doesn't depend on size of quartz crystal Spigot Discharge Other Methods Submarine Tailings Disposal • Alpine lake disposal – High alpine regions (no fish) • Riverine disposal – Banned except in Indonesia • Deep Ocean disposal – Kitsault and Island Copper Sub-aqueous Tailings Disposal Options • • • • Impoundment Covered Dam Pit Filling Submarine Factors affecting Submarine Disposal Island Copper Site Reclamation After 20 years of operation, the Island Copper Mine began reclaiming its waste dumps in 1996. Tailings were discharged deep into the adjacent fjord known as Rupert Inlet. Island Copper Pit Flooding Pit was flooded with sea water to create a Meromictic lake – 3 layers: Top – clean water; Middle – a reactor for surface ARD; Bottom – retain precipitated solids. Island Copper Pit Flooding Pit was flooded with sea water to create a 3-layer meromictic lake: Top – clean water; Middle – a reactor for surface ARD; Bottom – retain precipitated solids. Deep Sea Disposal of Tailings MillMill EZD – Euphotic depth UWD – – Upwelling depth EZD Euphotic depth MLD – – Mixed Layer depth UWD Upwelling depth MLD – Mixed Layer depth Thickened Discharge • Water drainage management is key http://technology.infomine.com/articles/1/1507/tailings.paste.thickened/paste.and.thickened.aspx Dry Stack Tailings • Anglo-American's La Coipa Mine in Chile http://www.tailings.info/index.htm Dry Stack Tailings • Anglo-American's La Coipa Mine in Chile Dry Stack Tailings • Deposition by trucking Dry Stack Tailings • Anglo-American's La Coipa Mine in Chile – Dewatering tailings to a filtered wet (saturated) or dry (unsaturated) cake – Must be transported by conveyor or truck – Material is deposited, spread and compacted as unsaturated tailings pile – Produces a stable deposit requiring no retention dam – Typical moisture content is below 20% - several percent below saturation – Combination of belt, drum, horizontal and vertical pressure plates and vacuum filtration systems Dry Stack Tailings • Advantages – Dewatering tailings to a filtered wet (saturated) or dry (unsaturated) cake – Must be transported by conveyor or truck – Material is deposited, spread and compacted as unsaturated tailings pile – Produces a stable deposit requiring no retention dam – Typical moisture content is below 20% - several percent below saturation – Combination of belt, drum, horizontal and vertical pressure plates and vacuum filtration systems Dry Stack Tailings • Disadvantages – – – – – – – – – – High capital and operating costs due to filtration Limited to low throughput operations (~20,000 tpd) Diversion systems to prevent inundation of stack Surface contour management to handle surface water Must prevent ponding and erosion of the stack No option to store water within a dry stack facility Sulfide oxidation creates high metal levels, low volumes Dust generation is problematic in arid climates Not suitable in high rainfall environment Seasonal fluctuations are important considerations Co-Disposal of Waste & Tailings • Co-mingling – Tailings and coarse waste rock material transported independently – Mixed together mechanically in storage facility or slurry-pumped – Mixing promotes voids filling (mingling) to maximise density • Co-placement – – – – Tailings and coarse waste rock material transported independently Not mixed to form a single discharge stream Waste rock end dumped into tailings facility Waste rock used to create internal berms or retaining walls (sometimes) • Co-deposition – Similar to co-placement, but waste streams placed in layers – Deposited tailings naturally enters voids in underlying rock – End-dumping waste rock with tailings deposition down face prior to further end dumping Dam Remediation Efforts By today's standards this dam is just too high for its design water flow and material properties. Built over many decades, a second dam was required to be built in the late 1990s to prevent water release (high As content). Main dam of the Helmsdorf uranium mill tailings deposit, Oberrothenbach (Saxony) Reparation Work Stava Fluorite Mine Dam Failure, Italy 1985 Before After Tailings dam consisted of two basins built on a slope. Failure started with collapse of the up-slope basin. Inflow of released material caused overtopping and collapse of the lower basin. The resulting slurry wave travelled to Stava at a speed of 30 km/h; later it reached 90 km/h. Lives lost = 268 Damages = $133 x 106 Failure at Aznalcollar, Spain - 1998 Failure at Aznalcollar, Spain - 1998 1. Slab of soil beneath the dam slid ~1m towards Río Agrio. 2. The dam cracked and broke; the wall collapsed sweeping out the separation dam. 3. Between 5 to 7 million m3 contaminated water and slurry spilled through the gap. 4. The Río Agrio rose 3m, changing its course and eroding bed rock. Los Frailes tailing dam failure, 1998 Los Frailes tailing dam failure, 1998 Reclamation and Revegetation Reclamation at Igarapé Bahia Mine in the Carajás Region, Amazon, Brazil 20 cm of organic soil over leached material Mines Operate in Sensitive Regions Waste Dump Reclamation, Igarape Bahia Mine, Carajas, Brazilian Amazon Mining Protects the Environment Installing a Heap Leach Liner in Chile Mining Repairs its Past Problems Rio Algom's Reclamation Operation at the Poirier Mine Tailing Dam in Northern Quebec Mine Site Reclamation and Closure BHP's Beenup Titanium Minerals mine at closure in early 1999 - W. Australia BHP's Beenup Titanium Minerals mine after final revegetation and reclamation The Britannia Mine Reclamation Project UBC at Britannia Beach Britannia Mine UBC at Britannia Beach Britannia Beach UBC-CERM3 has been involved at Britannia Beach since 2001 when we installed a plug inside the 2200 Level tunnel to create a research facility. This plug had the “spin-off” benefit of eliminating all pollution flowing into Britannia Creek and the surface waters of Howe Sound. Reclamation Issues in 2001 • Acid mine drainage from tunnels (620 m3/hr) • About 800 kg of Cu & Zn discharged per day • Over 10,000 tonnes of metal since closure • Groundwater contamination on the Fan • Potential impacts on aquatic life • Waste dumps and stockpiles • Tailings at bottom of Howe Sound • Sealing abandoned adits, demolition of derelict buildings (public safety issues) Groundwater discharge < 5% of the flow 2-3% of the copper 3-4% of the zinc 4100 Level effluent 50-80% of the flow 30-55% of the copper 60-75% of the zinc 2200 Level effluent 20-50% of the flow 45-70% of the copper 25-40% of the zinc Plug the 2200 Adit Build a Treatment Plant Reclaim pits and waste dumps Cutaway View of the Mine Workings Cutaway View of the Mine Workings Jane Creek after confluence of 2200 level effluent Britannia Mine – October 2000 Millennium Plug Research Project Pollution Plume – pre 2001 Millennium Plug Research Project Pollution Plume – pre 2001 Outcome – September 2011 Return of Adult Pink Salmon to Britannia Creek Numerous Media Reports The Sullivan Mine Reclamation Failure Reclamation Activities at Sullivan Mine 1998 - mine closed after 92 years 2000 – site reclamation on waste dumps (Number 1 Shaft and North dumps) 2004 - ditch was partially covered when the dump toe was extended 70m 2005 - 1m of glacial till was placed over the dump surface and the ditch • Reduce water percolation • Restrict air infiltration • Slow rate of oxidation Monthly sampling to monitor flowrate and contaminant levels Sampling Shed Sullivan Mine Accident – May 15-17, 2006 • Four people lost consciousness and died after entering the sampling shed Douglas Erickson, 48, a contractor Robert Newcombe, 49, Teck employee Kim Weitzel, 44, a paramedic Shawn Currier, 21, a paramedic • Reason: lack of oxygen • Immediately after the accident, O2 level in sump was ~2% & CO2 was ~7% • Shed used regularly with no problem and effluent flow was previously open channel • Reasonable to conclude shed was not a confined space at that time • Shed was used 1 week before tragedy • Oct. 2006, accident was identified as being • Other mines were warned immediately by B.C. Chief Inspector of Mines to treat all sampling sheds as confined spaces "unprecedented in the history of mining" Contributing Factors to the Accident • During Summer of 2005 • Dump & drainage ditch were covered to limit air/water • • infiltration and prevent human exposure to ARD O2-depleted effluent now isolated from the atmosphere Air in shed now directly connected to "bad" air in dump • Prior use showed no problem (1 week before) • False sense of security (9 years without any problem) • Shed was safe before the ditch became a drain • Design change created dangerous hazard • Atmospheric conditions play a major role • Temperature & pressure affect gas flowrate and direction Contributing Factors to the Accident • Before covering, ARD effluent was not O2-depleted • O2-depleted out of dump, but contact with air restores O2 level • After covering, ARD effluent was O2-depleted • O2-depleted out of dump, and no contact with air until shed • Possible mechanism • O2 removal from static air in the shed by O2-depleted effluent Before O2 transfer In ditch After O2 transfer In shed Breathing Waste Dump • August 2006 - dump was instrumented • Measure air velocity and gas composition in shed and pipe • Temperatures below ~10°C- the dump "inhales“ (positive flow) • Temperature above ~10°C- the dump "exhales“ (negative flow) • May 13-17, 2006 - Increase in temperature / decrease in pressure DANGEROUS SAFE DANGEROUS Temperature during week of the accident 25 Temperature (oC) 20 15 10 5 0 5/1/2006 5/6/2006 5/11/2006 5/16/2006 5/21/2006 5/26/2006 Daily average air temperature at Cranbrook airport in May 2006. Monitoring station was entered safely on May 8, 2006. 5/31/2006 Gas Velocity vs. Outside Temperature Cyclical Changes in Risk For a Confined Structure near dump toe Seasonal Variations • • • • • Safe in winter / Dangerous in summer In Summer, minimum night temperature may lie above maximum dump temperature Dump blows toxic gas all the time - deadly. In Winter, maximum day temperature may lie below maximum dump temperature Dump will suck in air all the time - safe Cyclical Changes in Risk For a Confined Structure near dump toe Diurnal Variations • • • • • Safe at night / Dangerous in day time Outside temperature cycles from hot to cool Dump may transition from blowing to sucking if maximum dump temperature lies between maximum day and minimum night temperature In Spring – transition from Safe all the time to Dangerous in day In Fall – transition from Dangerous all the time to Safe at night Summer Conditions Temperature 30 20 10 Daily Atmospheric Temperatures Maximum Internal Dump Temperature 0 -10 -20 Time of Day Fall Conditions Temperature 30 20 10 Daily Atmospheric Temperatures Maximum Internal Dump Temperature 0 -10 -20 Time of Day Winter Conditions Temperature 30 20 10 Daily Atmospheric Temperatures Maximum Internal Dump Temperature 0 -10 -20 Time of Day Spring Conditions Temperature 30 20 10 Daily Atmospheric Temperatures Maximum Internal Dump Temperature 0 -10 -20 Time of Day Cyclical Changes in Risk For a Confined Structure near dump toe Decadal Variations • • Safe(r) when maximum dump temperature has reached its long-term maximum value Dangerous when transitioning up or down Conceptual Period Boundaries: 0 - 10 years Initial period with rising danger 10 - 60 years Maximum danger - extremely hazardous 60 - 80 years Danger transitions from hazardous to problem 80 - 150 years Constant reduced danger – dump temp > max. outside temp. 150 - 170 years Rapid increase in risk - internal temp goes below max. outside temp. 170 - 180 years Maximum danger returns - extremely hazardous 180 - 190 years Danger transitions from hazardous to safe (pore gas O2 levels rise) 190 – onward Site is now safe - no O2-depleted gas generated or emitted Decadal Variation in Risk Assessment Risk of a Confined Space Accident Estimated Maximum Dump Temperature Maximum Outside Temperature Summer Conditions – transition to safe Dump reaches maximum temperature after 60-80 years Perhaps sooner with highly reactive dumps Temperature 30 20 10 Daily Atmospheric Temperatures Maximum Internal Dump Temperature 0 Time of Day -10 -20 Reference Dumps 1. White’s Dump at the Rum Jungle mine (U) in Australia (Harries and Ritchie, 1980, 1983, 1986, 1987; Ritchie, 2003) 2. Sugar Shack South Dump at Questa Mine (Mo) in New Mexico (Wels et al. 2003; Lefebvre et al., 2001a, 2001b & 2002; Shaw et al., 2002 Robertson GeoConsultants Inc., 2001) 3. South Waste Dump at the Doyon Mine (Au) in Quebec (Wels et al. 2003) 4. Nordhalde Dump at the Ronnenburg Mine (U) in Germany (Wels et al. 2003; Smolensky et al. 1999) 5. Aitik Mine dump (Cu) in Sweden (Stromberg and Bawart, 1999; Stromberg & Bawart, 1994; Ritchie, 2003; Takala et al., 2001) 6. Number One Shaft Waste Dump at the Sullivan mine (Pb/Zn) (Lahmira et al., 2009) Test Dumps 1. Main Waste Dump at Equity Silver Mine (Au/Cu/Ag) in British Columbia (Aziz and Ferguson, 1997; Lin, 2010) 2. West Lyell Dump at Mt. Lyell Mine (Cu) in Tasmania (Garvie et al. 1997) 3. North Dump at the Sullivan mine (Pb/Zn) (Lahmira et al., 2009; Dawson et al., 2009) Validation of the Model Dump Site Estimated Internal Temperature Reported Internal Temperature Nordhalde 10-15 14 Doyon 40 45 Sugar Shack South > 40 40 Aitik Mine 2-6 3 White’s Dump (1 year after cover) > 40 44 Number One Shaft 10 -15 12 Equity Silver Main > 40 52 West Lyell 35-40 38 (Max) Sullivan North 30-35 33 Nordhalde, Doyon, Sugar Shack S., Aitik, White’s, and Number One Shaft dumps are reference input cases North Dump, West Lyell, and Equity Silver Main are test cases Note: confined structure on top of the dump Sampling Shed ARD Dealing with Reactive Tailings • Two major types each creating a third issue – Acid Rock Drainage (ARD) – Cyanide • ARD leads to dissolution of Heavy Metals • Cyanide forms complex metallic ions • Metallic pollution (Al, Cu, Cd, Co, Fe, Mn, Pb, Zn) • Arsenic and/or selenium What is ARD and how do we deal with it? • Impact first reported in 1556 by Agricola in De Re Metallica • Yet the term Acid Rock Drainage wasn’t coined until 1970 • Significant work by NRCan (MEND Program) and Canadian companies developed innovative techniques to handle this ubiquitous problem • ARD requires sulphides, water, and air (and bacteria) – – – – Minerals are the source of sulphur and iron Air is the source of oxygen Water is the transfer medium for oxygen from air to rock Bacteria catalyze the reaction of Fe+2 to Fe+3 How long does ARD last? ROCK Generation of ARD from pyrite • ARD from surface coal mine in Missouri • Iron hydroxide (yellow boy) precipitates as pH rises from downstream dilution • Problem can last for decades Photo Credit: D. Hardesty, USGS Columbia Environmental Research Center The Colours of ARD How long does ARD last? - Forever! Corta Atalaya, Rio Tinto, Spain - abandoned pyritic open pit Rio Tinto in Spain – 2 millennium after mining Is it only Mining that causes ARD? Blood Falls at Taylor Glacier, Antarctica Acid Rock Drainage – Metal Leaching • ARD – Formed by atmospheric oxidation (i.e., water, oxygen, and carbon dioxide) of the common Fe-S minerals pyrite and pyrrhotite in the presence of bacteria Thiobacillus ferrooxidans, T. acidophilus, and T. thiooxidans • ML – Acid (H2SO4) leads to dissolution of metals and subsequent pollution of aquatic environments Basic Chemistry of ARD (from FeS2) Basic Issues behind the Chemistry: - Equilibrium of Ferrous-Ferric Ions Presence of Bacteria (Thiobacillus ferrooxidans) Must have an initial source of oxygen (i.e., air) Must have a way to transfer electrons (i.e., water) ARD Reactions Ferrous Sulphate formed by Abiotic Oxidation (slow): 2FeS2 + 2H2O + 7O2 = 2FeSO4 + 2H2SO4 Bacterial Oxidation of Ferrous Sulphate (T. ferrooxidans): 4FeSO4 + O2 + 2H2SO4 = 2Fe2(SO4) 3 + 2H2O Ferric Sulphate is Reduced and Pyrite Oxidized by these reactions: Fe2(SO4)3 + FeS2 = 3FeSO4 + 2S 2S + 6Fe2(SO4)3 + 8H2O = 12FeSO4 + 8H2SO4 Elemental Sulphur Oxidation (T. thiooxidans): 2S + 3O2 + 2H2O = 2H2SO4 Acid dissolves metals into solution meaning ARD is virtually always accompanied by high metal levels discharged into the environment. Bacteria are Essential Thiobacilli from bacterial generator (no flagella) Thiobacilli grown on ferrous iron (flagella) - left (x 5,000) - centre (x 20,000) - right (x 5,000) Formation of Bio-films can lead to long delay in onset of ARD (7-10 years) from Le Roux, N.W., et al., 1974. Bacterial Oxidation of Pyrite, Proc. 10th International Mineral Processing Congress, Institution of Mining and Metallurgy, London, 1051-1066.) Role of Bacteria • T. ferrooxidans acts to oxidize ferrous to ferric iron (Fe+2 to Fe+3) • The ionic reaction is: 4Fe+2 + O2 + 4H+ = Fe+3 + 2H2O • Fe+3 is a very powerful oxidizing agent • With Fe+3:Fe+2 ratio of only 1:106, ORP (Eh) > +0.4v * • General reaction of Fe+3 with base metal sulphides is: MS + nFe+3 = M+n + S + nFe+2 • Base metal sulphides react slowly with H2SO4 alone * ORP = Oxidation Reduction Potential (REDOX) Metal Leaching – Influence of ORP (Eh or REDOX) and Bacteria Malouf, E.E. and Prater, J.D. (1961), Role of Bacteria in the Alteration of Sulphide, J. Metals, NY, 13, p353-356. Garrels, R.M. and Christ, C.L. (1965), Solutions, Minerals and Equilibria, Harper & Row, New York, 216-222. Stages in ARD Generation (note the lag time) Control of ARD Removal of one essential component (sulfide, air, or water): 1. Waste Segregation and Blending – Blend-in neutralizing potential (NP) rock to yield pH 7.0 2. Base additives – Add limestone to buffer acid reactions 3. Liners, Covers, and Caps – Water covers are the most effective 4. Soil, clay, and synthetic covers (geomembranes) – minimize water and air infiltration Control of ARD 5. Bactericides – Chemicals that reduce/kill bacteria (T. ferrooxidans) – Effective, but costly, and “bugs” mutate 6. Collection and treatment of contaminants • Active or Passive treatment – Active treatment - high-density lime sludge – Passive treatment in constructed wetlands 7. Bioremediation (micro-organisms) – Remove metals directly – Introduce viruses against the bacteria Active Treatment • • • • Most effective Most expensive All effluent processed in a treatment plant May require processing for decades High-Density Sludge Water Treatment Plant WTP at Britannia Mine Site Howe Sound, British Columbia Capital Cost Operating Costs = ~ $12.0M = ~ $ 1.5M/year HDS Plant – Process Flow Diagram Recycle Water Flocculants Lime Paste Acidic Feed Water Flocculants Tanks Lime Tank Lime Reactor Sludge/Lime Mix Tank Sludge Recycle Clarifier Air Effluent Overflow Sludge disposal Sludge Disposal • Sludge Disposal by truck – cost = ~$40/tonne • Other options – Manufacture bricks by blending sludge with clay or pumice – Use low-temperature process with organic resins – Use high-temperature process to harden into a ceramic • Examine opportunities to recover Cu and Zn – From the effluent prior to HDS – From the sludge by leaching So Reduction Process Schematic Sulphur Nutrients Electron donor BIOREACTOR (So Reduction) H2S Soda Ash or Lime Contaminated Drainage Cu Precip Zn Precip CuS ZnS Metals, SO4 BioteQ After R.W. Lawrence, BioteQ Treated Water Production Summary Flow Feed Water Discharge Water Cu Concentrate 14,880 m3/d – average over 12 months [mg/L] 18.0 Cu, 20.0 Zn, 0.1 Cd [mg/L] 0.05 Cu, 0.01 Zn, 0.01 Cd 187.0 tonnes per year contained copper 51.1% Cu, 2.1% Zn, 0.24% Fe, 33.1% S Zn Concentrate 185.5 tonnes per year contained zinc 52.4% Zn, 1.5% Cu, 0.3% Cd, 0.8% Fe, 27.1% S Additional Benefits Lime Savings Sludge Reduction $64,000 per year (32%) 340-450 tonnes per year (15-20%) Commercial Scale Plants Pumice rock – extremely light Brick Veneer Cladding - examples NRC Process Evaluation - http://www.nrc-cnrc.gc.ca/eng/ibp/irc/ci/volume-4-n4-7.html Canyon Stone - http://www.canyonstonecanada.com Passive Treatment Technologies Name Description Function Shallow wetlands Fe and Mn oxidation, Co-precipitation • Limited Emergent to low effluent flowrates vegetation of Metals, Sorption on Biomass Aerobic wetlands Selected References Eger and Wagner, 2003 USDA and EPA, 2000 Open limestone channels Acidic water flows over limestone, or other alkali Alkalinity addition Al, Fe, Mn oxide precipitation Ziemkewicz et al., 1997 Anoxic limestone drains Water flows through limestone channel under anoxic conditions Alkali addition; Fe Precipitation; Limestone Armouring Prevention Watzlaf et al., 2000 Anaerobic wetlands Subsurface wetland, isolated from air by water or material Alkali addition; Sulphate Reduction; Precipitation of metal sulfides; Sorption on Vegetation Brenner, 2001 USDA and EPA, 2000 Successive Alkalinity Producing Systems Vertical flow systems drain through limestone layers & anaerobic organic matter Alkalinity addition; Sulphate Reduction Metal Precipitation Kepler and McCleary,1994 Zipper and Jage, 2001 Sulfate-Reducing Bioreactors Collected water in anoxic chamber containing organic matter and SRBs Alkalinity addition; Sulphate reduction; Metal Precipitation Gusek, 2002 Permeable Reactive Barriers Intercepted groundwater flows through permeable barrier containing reactive material Alkalinity addition; Sulphate reduction; Metal Precipitation and Sorption Benner et al., 1997 US DOE, 1998 Amendments Materials added to ARD sources or holding areas Alkalinity addition; Sulfate reduction; Metal Precipitation; Sorption; Chelation; Revegetation Chaney et al., 2000 Covers Liners, Covers, and Caps • • • • • • Liners used to prevent seepage form the dam Covers used to inhibit influx of water and air Caps used to seal dam entirely Expensive materials and installation Must be installed with great care Biggest issue – degradation over time Factors affecting Soil Cover Performance International Network for Acid Prevention, 2003. Evaluation of Long-term Performance of Dry Cover Systems, Final Report. O’Kane Consultants Inc., (Eds.), Report No. 684-02. Geomembranes 1. 2. 3. 4. 5. 6. 7. 8. Plastics (polyethylene (PE) High density poly. (HDPE) Chlorinated poly. (CPE) Chloro-sulphonated poly. (DuPont HYPALON) polyvinyl chloride (PVC) Low-density poly. (LLDPE) Geosynthetic clay liners (GCL) Geomembranes impregnated with bitumen After Meer, S.R. and C.H. Benson, 2007. Hydraulic conductivity of geosynthetic clay liners exhumed from landfill final covers. J. Geotech. and Geoenviron. Eng., 133(5):550-563. Geo-Membranes: Benefits and Disadvantages Benefits Disadvantages Low permeability High cost Relatively easy to install Cost depends on distance from supplier to site Resistant to chemical and bacterial attack Limited design life - 50 to 100 years Requires proper bedding and protective cover Geotechnical instability on steep slopes Vulnerabilities include: - Sun light (UV breakdown) - Puncture by surface traffic - Cracking and creasing - Seam difficulties - Uplift pressure from fluid or gases - Degradation by cation exchange with GCLs - Settlement of underlaid materials - Thermal expansion and contraction After Meer, S.R. and C.H. Benson, 2007. Hydraulic conductivity of geosynthetic clay liners exhumed from landfill final covers. J. Geotech. and Geoenviron. Eng., 133(5):550-563. Field Monitoring of a Waste Pile Cover MEND, 2004. Design, construction and performance monitoring of cover systems for waste rock and tailings. Report 2.21.4, O’Kane Consultants Inc., (Eds.), Natural Resources Canada. Sub-aqueous Tailings Disposal Options • • • • Impoundment Covered Dam Pit Filling Submarine Factors affecting Submarine Disposal Microbiology of ARD Microbiological Aspects of ARD • Bacteria form films on sulfide surfaces • Reaction rate accelerates up to 108 times • T. ferrooxidans/L. ferrooxidans considered responsible for catalytic behaviour • Microbial makeup is controlled by site environment • Microbes not well-studied or understood Microbiological Aspects of ARD Thiobacillus ferrooxidans Leptospirillum ferrooxidans http://www.mines.edu/fs_home/jhoran/ch126/thiobaci.htm http://microbewiki.kenyon.edu/index.php/Leptospirillum Microbiological Aspects of ARD • Key organisms (T. & L. ferrooxidans) >> global significance • Physiological and genetic aspects well studied • Microbial diversity specific to site environment • Recent advances on structural dynamics of communities • Biofilms on sulfide surfaces play a key role • Bacteriophage impact negatively on bacterial populations ARD Mitigation with Bacteriophage • Novel approach to ARD control • Isolate phage that infect ARD-generating “bugs” • Create deadly mixture of viruses to control microbial ARD communities with biology • New and unexplored cross-disciplinary field Microbiological Aspects of ARD Structure of microbial communities Microbiological Aspects of ARD Biogeographic distribution of key microbes Microbiological Aspects of ARD Diversity revealed by molecular methods Bacteriophage • Viruses that infect bacterial cells • Intracellular parasites – do not generate energy or synthesize proteins by themselves • Infection results in death, if phage is virulent • Temperate phages kill only a small fraction of cells and cause infected host to mutate Bacteriophage – friend or foe • Many shapes and sizes • Phage are very small, (20-200 nm in diameter) • Some phage break down biofilm matrix to infect "protected" cells Photo credits: ICTV Database www.ncbi.nlm.nih.gov/ICTVdb Bacteriophage – friend or foe T4 bacteriophage attacking an E. Coli bacterium Bacteriophage – an ARD solution? Like lunar landers, bacteriophage attach to the microbial cell wall and inject their DNA for replication Cell Wall Bacteriophage - an ARD Solution? Photomicrographs of T4 bacteriophage for E. Coli Bacteriophage - an ARD Solution? The Lytic Cycle leads to the death of the host The Lysogenic Cycle leads to mutation of the host ARD Mitigation with Phage • Inject phage into a dump/dam • Coat surfaces with phage-containing biofilm • Phage will control microbial population, not eliminate it • Phage for ARD-microbes do exist – why do they become dormant at low pH? Biofilms and Quorum Sensing • Complex association of cells with an exopolysaccharide matrix • Adhere strongly to sulfide surface or grow deep within pores and cracks • Play integral role in composition & stability of microbial communities Progression of Biofilms Evolution from Planktonic Behaviour to a Biofilm REVERSIBLE ADSORPTION OF BACTERIA (seconds) IRREVERSIBLE ATTACHMENT OF BACTERIA (sec - min) GROWTH & DIVISION OF BACTERIA (hours - days) EXOPOLYMER PRODUCTION & BIOFILM FORMATION (hours - days) ATTACHMENT OF OTHER ORGANISMS TO BIOFILM (days - months) Benefits of Biofilms for Microbes Common Example of a Biofilm Dental Plaque Stained with Gram's Iodine Bio-Films and Quorum-Sensing • Gene expression regulated by cell density changes • Q-S bacteria release signal molecules (auto-inducers) • Auto-inducer concentration increases with cell density • At threshold concentration, gene expression changes • Q-S communication regulates many physiologies: - symbiosis - competence - antibiotic production - sporulation - virulence - conjugation - Programmed Cell Death (PCD) - biofilm formation Bio-Films and Quorum-Sensing • Big Question? Can we use Q-S knowledge to get microbes in a bio-film to “commit suicide” without creating new side-effect problems? Conclusions • Tailings Dam Construction must be done with care and knowledge about the tailing material, about the foundation material – both physical and chemical factors are important • Every 8 months, a tailings dam fails somewhere in the world • Downstream methods are safest • Reactive Tailings require additional care and concern for ARD and Metal Leaching • Cyanide Tailings also generate metal pollution • Confined Space issues may exist with ARD wastes • Microbiology is a new approach receiving attention Questions?