British Columbia Mine Reclamation Symposium

Britannia Mine water treatment plant : the first year of operations Madsen, C. 2007

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


59367-04 Madsen final.pdf [ 118.88kB ]
JSON: 59367-1.0042527.json
JSON-LD: 59367-1.0042527-ld.json
RDF/XML (Pretty): 59367-1.0042527-rdf.xml
RDF/JSON: 59367-1.0042527-rdf.json
Turtle: 59367-1.0042527-turtle.txt
N-Triples: 59367-1.0042527-rdf-ntriples.txt
Original Record: 59367-1.0042527-source.json
Full Text

Full Text

BRITANNIA MINE WATER TREATMENT PLANT: THE FIRST YEAR OF OPERATIONS   C. Madsen, M.Sc.  EPCOR Suite 215 10451 Shellbridge Way Richmond, BC, V6X 2W8  INTRODUCTION  EPCOR began construction of the Britannia Mine high density lime sludge plant in February 2005, and the plant was treating Acid Rock Drainage (ARD) by October 21, 2005, beginning a twenty-year design build finance operate contract for the B.C. Ministry of Agriculture and Lands.  2006 was a milestone year for the Britannia Mine site with the first complete year of operation of the water treatment plant (WTP) and the cleanest water being discharged from the mine into Howe Sound in decades.  Over the first year, water quality coming into the plant from the mine workings was significantly different from what was predicted in the Request for Proposal (RFP) stage of the project, with lower metals concentrations and higher pH than expected. Flow through the mine was also only 80% of the typical predicted flow from the RFP.  Even with this unexpected water quality and flows the plant performance was consistent throughout the year and the through the first freshet  PLANT PROCESS DESCRIPTION  The primary source of contaminated water to the WTP is ARD from the Britannia Mine workings.  With the installation of a plug at the 4100 Adit, near the top level of the Mine Concentrator building, a reservoir with a volume of approximately 485,000 m3 was created.  ARD can be stored in the mine and the mine level managed from a low of 20 m up to 260 m of head.  The water is delivered to the plant through either a penstock or gravity line with the control valves located at the plug, 400m into the Adit. Typically, the water is delivered to the plant via the penstock which feeds two Micro Hydro turbines which can generate up to 110 Kw each.  The secondary source of water to the plant comes from pumping the seven groundwater wells in the alluvial fan area on the seaward side of the railway tracks from the Mine Concentrator.  These wells capture mine tailings contaminated groundwater.  The groundwater pumps are variable frequency drive (VFD) controlled using feedback from a conductivity meter on each well.  The conductivity has been correlated with chloride concentration and limited to a maximum of 1,000 mg/L of chloride or 4,400 υs/cm.  The ARD from these wells is pumped up the hill to the treatment plant at flows of up to 100 m3/hr.  The contaminated water from the mine workings and groundwater system enters the treatment plant where it is treated with slaked lime bringing the pH to 9.3–9.5 in the reactor tanks.  Blowers add air to the reactors to aid in the oxidation of iron and manganese and mixers ensure that the air and lime are evenly mixed with the influent.  Potassium permanganate is also added to reactor #1 to complete the oxidation of manganese.  From the reactors, the treated water flows into a 30 m diameter clarifier where polymer is added just prior to a static mixer to assist settling.  The treated effluent is discharged to Howe Sound through the new outfall.  The sludge settles in the clarifier and 15 – 20 % of the flow, by volume, is recycled back to #1 reactor via the sludge/lime mix tank.  This ensures a liquor of 2 – 5% solids in the reactors which assists in the rapid reduction of dissolved metals.  The underflow solids can reach up to 35% density and still be easily pumped.  Waste sludge is pumped to an onsite holding tank and dewatered in batches.  The dewatering takes place in a 94 plate, 1200 mm filter press which holds 3.5 m3 of solids dried at up to 55%.  The solids are unloaded from the bottom of the press directly into a truck by gravity for temporary on site storage. Twice a year the stored sludge is trucked for authorized permanent disposal to one of the mine’s “Glory Holes” at Jane Basin.  The plant has been designed to treat ARD to meet the discharge permit criteria at a flow of 1050 m3/hr; however, the hydraulic capacity of the plant is 1400 m3/hr.  In high flow periods, for example freshets where the snow melt occurs extremely rapidly, the plant can treat the incoming ARD at a slightly reduced quality.  In the event of mine inflows exceeding the hydraulic capacity of the WTP for an extended period of time, the capability to bypass the WTP was built into the plant infrastructure.  The bypass water is treated with lime and mixed with the WTP effluent and the combined flow is discharged into Howe Sound via the newly constructed outfall.  A sampling station on the outfall just prior to discharge allows for pH feedback for the bypass and plant combined effluent.  The total hydraulic capacity of the plant and bypass works including the outfall is 3600 m3/hr.  WATER QUALITY  With the start-up and continuing operation of the WTP, significantly more water quality testing has been performed on the mine water than had been previously done and there were some unexpected results that differed from the testing that had been done prior to the project implementation.  Table 1 shows a summary of the WTP influent including both the ARD from the mine workings and the water pumped to the water treatment plant from the groundwater pump station on the foreshore.  All values presented in the table are monthly mean values.  The only trend evident in the data is that the mine working total solids seems to have leveled off after starting quite high early in 2006.  Conductivity in the groundwater being pumped to the plant appears higher in the late summer months, which are typically quite dry, when it was likely that the wells were slightly over pumped.  It is likely that by optimizing the pumping rates and the rate of change that they respond to the conductivity further, these overall values can be reduced to the target 4400 us/cm, with improved pumping rates.  Table 1:  2006 Britannia Mine WTP Influent Water Quality Summary Month Mine Workings pH Mine Workings TSS (mg/L) Groundwater pH Groundwater TSS (mg/L) Groundwater Conductivity (υs/cm) January 4.0 19 3.7 6 3382 February 4.0 8 3.9 4 3678 March 3.8 3 3.8 N/A 2585 April 3.8 11 6.9 5 4935 May 4.8 11 6.0 5 4886 June 5.3 11 4.0 5 4929 July 3.8 10 3.8 6 5153 August 3.8 5 4.6 4 5567 September 3.7 4 3.8 3 5841 October 3.7 2 4.0 4 6264 November 3.7 3 5.1 2 4898 December 3.8 4 6.1 4 3943 Average 4.0 8 4.6 4 4672   Additional data on the mine workings water quality is shown below in Table 2.  This data is monthly mean dissolved metals levels for the parameters of concern (i.e., those that are regulated in the Discharge Permit).  The monthly average mine level is also shown.  It should be noted that the actual mine level can change quite dramatically over a single month based on plant operations, especially during freshet when there is rapid increase in the mine inflow.  The management of the mine reservoir level is done according to the procedures in the EPCOR Britannia Mine Reservoir Management Plan.  From the data in Table 2, there is no apparent seasonal impact on the dissolved metal levels from the mine workings coming into the water treatment plant.  A more detailed graphical analysis was done between all of the water quality parameters and the level of water in the mine.  It was found that of all of the parameters, including dissolved metals, total suspended solids (TSS), and sludge generation, there is surprisingly little relationship between the water level in the mine and the water quality parameters.  Only dissolved iron and manganese showed a very slight negative correlation with mine reservoir level, with all other measures being scattered quite randomly.  Table 2:  2006 Plant Influent Metals Water Quality Summary  Mine Workings Dissolved Metals (mg/L) Month Al Cd Cu Fe Mn Zn Mean Mine Level January 22 0.104 20 1.7 4.4 22 86 February 21 0.084 16 1.4 4.3 18 97 March 19 0.082 14 0.8 4.3 17 138 April 24 0.093 17 2.2 5.9 21 77 May 28 0.112 25 2.7 5.0 25 58 June 22 0.105 22 0.8 3.8 22 160 July 18 0.090 17 0.7 4.4 20 151 August 20 0.055 14 0.5 4.9 19 101 September 23 0.078 14 1.5 5.4 18 43 October 17 0.074 13 1.0 5.0 20 59 November 20 0.102 18 2.0 5.4 21 102 December 19 0.097 16 0.7 5.5 22 140 Average 21 0.090 17 1.3 4.9 20 101  As it is quite apparent that the metals concentrations are quite low, it will be interesting to see if this pattern continues with repetitive filling and emptying of the mine reservoir.  As the operation of the mine reservoir at varying levels is important for power generation using the micro hydro turbines at the WTP influent, it is important to balance the reservoir level management with the level that results in the lowest operating cost along with highest effluent water quality.  From the data available to date, there is no impact on water quality and therefore on operating costs with varying the mine reservoir level.  Table 3 shows the actual influent and effluent water quality over the year compared with what was predicted for the project in the RFP stage.  The values in the table for the mine workings, groundwater, and water treatment plant effluent are annual average values.  The water coming from the mine workings had significantly lower metals values for all metals and a higher pH than was expected based on the data available at the time of the RFP.  Many of the metals coming into the process were below the tenth percentile of what was expected including aluminum, cadmium, copper, iron, and zinc.  Manganese was the only metal above the tenth percentile and averaged 0.3 mg/L above the tenth percentile value of 4.6 mg/L over the year.  The annual average of the pH of the mine workings at 4.0 was above the ninetieth percentile value of 3.7. Table 3:  2006 Plant Influent Water Quality Summary Actual vs. Predicted  Dissolved metals (mg/L)  Al Cd Cu Fe Mn Zn pH  Mine Workings (Annual Mean) Actual 21 0.090 17 1.3 4.9 20 4.0  RFP ‘Typical’ Value 39 0.120 32 14 6.4 24 3.5 RFP 10th Percentile Value 26 0.103 25 2.6 4.6 22 3.2 RFP 90th Percentile Value 73 0.126 55 63 12.0 27 3.7  Groundwater (Mean) Actual 27 0.100 9 16 1.5 9 4.6 Groundwater RFP Typical      (predicted) Value 50 N/A 10 50 N/A 20 4  WTP Effluent (Annual Mean) 0.5 0.002 0.01 0.007 0.3 0.02 9.1  Discharge Permit Limit < 1 < 0.01 < 0.1 < 0.1 < 0.4 <0.2 6.5 - 9.5 Provincial Guideline Limit < 0.5 < 0.01 < 0.02 < 0.01 < 0.2 <0.03 6.5 - 9.5  Groundwater chemistry for all metals were also significantly below the assumed concentration values with Aluminum and Zinc about half of what was expected and Iron less than a third of what was expected. Only the copper in the groundwater was close to the value predicted in 2005.  The average groundwater pH of 4.6 was higher than the predicted value of 4.0.  Overall, the water treatment plant performed very well for its first year of operation.  Although the details of the plant performance will be expanded in the next section, it can be seen in the plant effluent annual mean data in the table above that the average values for the metals are either below or at guideline limits for Aluminum, Cadmium, Copper, Iron, Zinc, and pH.  Only manganese is above the guidelines on average.  Although the dissolved manganese was below permit limits for the majority of the year, the most significant operational issues for the plant related to the manganese discharge levels.  To resolve the manganese removal issues, an additional chemical feed system had to be engineered, tested and installed at the plant.  Initially additional baffling was installed between reactor one and two in order to increase the detention time for the oxidation reaction to occur; however, at high flows, this was not quite enough and bench testing was done with hydrogen peroxide, potassium permanganate, and sodium hypochlorite.  A variety of injection points to the process were tested as well, and the most effective operational solution was to add potassium permanganate to consistently achieve the desired Mn removal.  OPERATIONS AND MAINTENANCE SUMMARY  The summary data for the WTP volumes treated over 2006 are shown below in Table 4.  The total ARD that is treated coming from both the mine workings and the groundwater pump station is shown.  The bypass flows shown for March and July were undertaken as part of the required hydraulic flow tests to complete the commissioning of the water treatment plant and outfall.  The small bypass that occurred in June was authorized to complete work on the flow control valves at the plug in the 4100 Adit.  It is interesting to note that the total flow through the plant of 3,907 ML is significantly lower than the annual flow assumed in the RFP of 4,977 ML despite the relatively high snow pack over the 2005/2006 winter. It was estimated that the return period of this snow pack was 3.7 years.  Table 4:  2006 Britannia Mine WTP Annual Flow Summary Month Mine Workings Flow (m3) Groundwater Flow (m3) Plant Flow (m3) Bypass Flow (m3) January         333,771             13,562          347,333             - February         254,903             12,994          267,897             - March           59,270                  784            34,909       25,204 April         359,055               9,968          368,046             - May         497,589               9,816          505,405             - June         754,874               9,120          763,994             10 July         534,554               8,849          556,183       13,618 August         307,741               7,451          315,192             - September         185,638               4,769          190,407             - October           75,960               3,309            79,269             - November         154,313               9,352          163,665             - December         301,732             13,639          315,371             - Total       3,819,400            103,613  3,907,671        38,832  The operation of the plant and the volume of water treated from the mine workings are tied to the Reservoir Operations Plan where the plant flow is essentially matched to the inflow to the mine reservoir. However, the mine level may be managed at a higher level during low inflow periods to allow for the efficient generation of power with the onsite micro hydro turbines.  Following the Reservoir Operations Plan, the mine reservoir was brought to a low level to allow capacity to buffer the 2006 freshet inflows into the mine.  The plant flow was brought up to the design capacity of 1,050 m3/hr and left there for the duration of freshet and the mine reservoir level reached a peak of 190 m.  The micro hydro turbines were commissioned in August; however, this is a low inflow period and power was not being generated regularly until December.  A total of 13,000 Kw were generated in 2006 to offset operational demand from BC Hydro.  In the future it is expected that approximately half of the running electrical load of the plant will be offset with onsite power generation using the incoming flow of ARD from the mine workings.  As operational experience is fine tuned and the reservoir operations plan is consistently implemented, it is expected that the prediction of mine inflows will improve as the local weather station data is collected under regular operating conditions.  The weather station, measuring precipitation and temperature, was relocated from the BC museum of Mining to the top of the lime silo at the WTP in November; however, due to equipment failures, it was not recording data through December.  The weather station at Jane Basin had a new data logger installed in November and records precipitation, including snow pack, and temperature.  The Jane Basin weather station communicates with the Britannia Mine WTP data collection and storage system via a radio link.  Over the next several years, it is expected that local weather conditions including temperature and snow pack level in Jane Basin will be useful in predicting the impact of freshet on mine level and implementation of the reservoir operating plan.  As was expected with the incoming ARD quality, chemical usage was proportional.  The table below shows the total chemical usage and effective chemical doses that were used at the WTP in 2006.  These totals are compiled on a mass balance basis and the average values are indicative of the relative amounts of lime and polymer added to the process on a dose basis.  Table 5:  Britannia Mine WTP 2006 Chemical Usage Month Lime Kg Polymer Kg Average Lime Dose (mg/L) Average Polymer Dose (mg/L) January 145,410 1,200 419 3.5 February 39,140 300 146 1.1 March 17,250 0 494 0.0 April 98,500 850 268 2.3 May 150,060 1,250 297 2.5 June 203,990 1,250 267 1.6 July 132,130 1,025 238 1.8 August 59,570 625 189 2.0 September 62,810 350 330 1.8 October 15,000 200 189 2.5 November 41,520 350 254 2.1 December 75,270 475 239 1.5 Total 1,040,650 7,875 266 2.0  The hydraulic flow testing that was done in March resulted in increased lime usage for treating the plant bypass ARD.  The WTP was shut down for the month of March to allow the mine reservoir level to rise to be able to complete the necessary high flow testing.  For the six days that the plant was run, the only polymer used was that which had already been prepared in the batch feed tank, showing no polymer use over the month.  The monthly effluent water quality data for the WTP are shown in Tables 6 and 7.  As can be seen in Table 6, effluent TSS and turbidity was higher in the early part of the year then leveled off after May. This was a direct result of repairing clarifier equipment failures; optimizing polymer dosing, optimizing sludge wasting, and optimizing recycle rates.  The optimization of WTP operations kept the physical removal values far below permit levels in the last half of 2006.  All toxicity tests for the WTP effluent came back non-toxic.  The permitted toxicity test is the rainbow trout 96 hour LC 50.  Through the entire year, the toxicity results were consistently 100% survival in 100% concentration effluent.  It is worth noting that even during the hydraulic flow testing, where approximately 75% of the flow was bypass flow that had been pH adjusted to around 8.0 with lime, the combined effluent was non-toxic.  Table 6:  Britannia Mine 2006 Monthly Mean WTP Effluent Quality Month WTP Effluent TSS (mg/L) WTP Effluent Turbidity (NTU) WTP Effluent pH 96 Hr LC50 Toxicity Test January 14 18 9.2 100% survival in 100% concentration February 27 21 9.2 100% survival in 100% concentration March N/A 16 8.9 100% survival in 100% concentration April 27 23 8.8 100% survival in 100% concentration May 5 6 9.1 100% survival in 100% concentration June 4 6 9.0 100% survival in100% concentration July 6 7 9.0 100% survival in 100% concentration August 10 7 9.2 100% survival in 100% concentration September 3 6 9.3 100% survival in 100% concentration October 3 5 9.1 100% survival in 100% concentration November 4 5 9.1 100% survival in 100% concentration December 2 5 9.0 100% survival in 100% concentration Average 10 10 9.1  Although turbidity is not a regulated parameter in the plant discharge permit, it is constantly monitored and recorded at the WTP effluent.  Turbidity is used because it is a water industry standard method for measuring particles in water, is easily maintained, and can produce reliable, consistent results which can be used in an online process environment to notify operations staff of process upset that require operator attention.  The figure below shows the relationship between turbidity and TSS, the regulated parameter, which has been developed for the Britannia Mine WTP. Online Turbidity Vs. TSS y = 0.4354x + 4.1509 R2 = 0.8163 0 5 10 15 20 25 30 35 0 10 20 30 40 50 60 7 Lab TSS (mg/L) O nl in e Tu rb id ity  (N TU ) 0 Figure 1:  Turbidity and Total suspended solids correlation  The relationship between turbidity measurement for the WTP clarifier effluent and Total Suspended Solids as measured by an external laboratory is generally linear.  This relationship was used to establish an automated alarm system using a turbidity meter as TSS measurement cannot be done reliably with online instrumentation.  Lab values less than the limits of detection of the test (i.e., either 4 mg/L or 2 mg/L, depending on the test report), were not used in the development of a correlation formula between the two parameters.  The permit limit for TSS of 30 mg/L, correlates to an online turbidity measurement of approximately 17 NTU.  It should be noted, however, that there is limited data at the higher values and the relationship, though generally linear, is not particularly strong; there are significant numbers of outliers.  Turbidity is measured continuously for the plant effluent and is impacted by flocculent dosing, recycle rates, and sludge wasting.  With respect to the treatment of metals, a summary of the monthly averages of dissolved metals in the WTP effluent is shown below in Table 7.  The plant performed relatively consistently through the year and this data provides a solid baseline for measuring future performance and process improvements to the plant.  Table 7:  Britannia Mine WTP 2006 Monthly Effluent Quality (metals)  WTP Effluent Dissolved Metals (mg/L) Month Al Cd Cu Fe Mn Zn January 0.6 0.002 0.03 0.013 0.3 0.04 February 0.6 0.002 0.01 0.005 0.3 0.01 March N/A N/A N/A N/A N/A N/A April 0.6 0.002 0.01 0.014 0.4 0.04 May 0.7 0.002 0.01 0.005 0.4 0.01 June 0.6 0.002 0.01 0.005 0.3 0.01 July 0.6 0.002 0.01 0.010 0.4 0.02 August 0.5 0.001 0.01 0.005 0.3 0.01 September 0.6 0.001 0.02 0.005 0.2 0.02 October 0.4 0.001 0.01 0.005 0.2 0.02 November 0.5 0.001 0.02 0.005 0.4 0.03 December 0.3 0.003 0.01 0.005 0.3 0.02 Average 0.5 0.002 0.01 0.01 0.3 0.02  Discharge Permit Limit < 1 < 0.01 < 0.1 < 0.1 < 0.4 <0.2 Provincial Guideline Limit < 0.5 < 0.01 < 0.02 < 0.01 < 0.2 <0.03  Sampling of the WTP effluent is done with a composite sampler and both daily and weekly composite samples are taken. The permit limits for metals are for a weekly sample, thus most values are the mean of 4 samples. If the daily composite sample for the weekly test showed a test result at or near the permit values, the daily composite samples for the days surrounding the sample are tested to determine where the process upset occurred. Referring back to the discharge and permit limits shown in Table 3, all metals, including Manganese are at or within permit limits on a monthly basis.  Aluminum is slightly above the guideline limit of 0.5 mg/L through most of the year.  Keeping the aluminum below the guideline limit is challenging due to the relatively low precipitation pH of aluminum relative to copper and zinc. Other than the first month of the year, both copper and zinc were within guideline limits. Cadmium and Iron were consistently at or below guideline levels.  RESIDUALS HANDLING  There were 897 filter presses done at the WTP in 2006 using the plate and frame press, producing approximately 3,000 m3 of sludge. Additional plates have been purchased to increase the volume of sludge dewatered in each press. These plates will be installed in early 2007.  A polymer line has been added to the filter press feed line and tests with polymer addition will be done in 2007 to test if improvements can be made on the percent solids of the filter cake.  A summary of the filter presses and dewatering performance by month is shown in the Table below.  Table 8:  Britannia Mine WTP 2006 Sludge Filter Press Summary  Month Filter Presses Average % Solids January 8 38 February 3 38 March 1 38 April 78 42 May 150 40 June 154 44 July 156 46 August 104 45 September 202 45 October 0 N/A November 18 42 December 23 43 Total 897 Average  44                   The sludge is stored on site in an existing concrete basin located beside the WTP.  The structure is covered with a tent cover set up over lock block walls.  Runoff from the unloading area and the exposed surfaces is captured and the slurry is returned to the WTP reactors.  In 2006, the sludge was hauled in a single campaign to the Jane Basin site; however, it is expected that hauling will typically be done twice a year to be able to contain all of the sludge manageably in the concrete basin.  SUMMARY  Over the first year of operation, there were significant challenges in the operation of the plant.  Other than the typical start-up and commissioning issues that would be expected in any Greenfield plant start-up, the unexpected water chemistry provided some process challenges.  In terms of compliance with the discharge permit, the most significant issue has been with the removal of manganese to below 0.4 mg/L in the plant effluent.  Through bench and pilot testing with various oxidants, however, this issue was quickly resolved within the first few months of operation.  Most importantly, the plant effluent has consistently proven that it is completely non-toxic, even when mixed with pH adjusted bypass water, and the water quality in Howe Sound is no longer subjected to the consistent and large volume of ARD from the mine workings.  Upcoming optimization of the plant and process will focus on improvements in residuals dewatering and consistently achieving Provincial Guideline limits on water quality parameters.


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items