British Columbia Mine Reclamation Symposium

A duplicate column study of arsenic, cadmium and zinc treatment in an anaerobic bioreactor based on a… Kawaja, Jonathan D. E.; Morin, Katy; Gould, W. Douglas 2005

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

Item Metadata


59367-18 Kawaja Final.pdf [ 175.68kB ]
JSON: 59367-1.0042488.json
JSON-LD: 59367-1.0042488-ld.json
RDF/XML (Pretty): 59367-1.0042488-rdf.xml
RDF/JSON: 59367-1.0042488-rdf.json
Turtle: 59367-1.0042488-turtle.txt
N-Triples: 59367-1.0042488-rdf-ntriples.txt
Original Record: 59367-1.0042488-source.json
Full Text

Full Text

A DUPLICATE COLUMN STUDY OF ARSENIC, CADMIUM AND ZINC TREATMENT IN AN ANAEROBIC BIOREACTOR BASED ON A SYSTEM OPERATED BY TECK COMINCO IN TRAIL, BRITISH COLUMBIA Jonathan D.E. Kawaja, Katy Morin, and Dr. William D. Gould Natural Resources Canada (CANMET) Mining and Mineral Science Laboratories, Mine Effluents Program, 555 Booth St., Ottawa, Ontario, Canada K1A 0G1 ABSTRACT The study’s objective was to identify and profile the treatment mechanisms within an anaerobic bioreactor (ABR) designed to remove high concentrations of arsenic, cadmium and zinc from contaminated drainage. The experiment used duplicate 15.6 L ABR columns, containing limestone and an organic substrate. The design and operation parameters were based on a larger field scale system operated by Nature Works Remediation for Teck Cominco in Trail, British Columbia. After an acclimatization period, the experiment was conducted for six months at various hydraulic loadings intended to evaluate optimal and stressed conditions, as well as, the ABR’s ability to re-establish optimal conditions. The study suggested that cadmium and zinc were removed as metal sulphides after 20 hours residence time. The behaviour of arsenic was independent of cadmium and zinc, and the majority was removed within 7 hours residence time. This was attributed to its adsorption to iron. Correlations of arsenic and iron concentrations throughout the organic substrate demonstrated that adsorption was inconsistent and unreliable as a treatment mechanism without the subsequent oxic conditions at the column’s headwater. Considerations for the field were identified for treatment, management of hydraulic loadings, and maintenance of an ABR. INTRODUCTION Background Over the past several years, the Mining and Mineral Science Laboratories (MMSL) has taken an interest in the feasibility of passive systems to mitigate mine drainage in Canada. The recognition that natural systems can play a role in the attenuation of acidity, sulphate and metals has led to various designs for passive systems. In general, these systems utilize material to generate alkalinity (e.g., limestone) and provide a medium for microbial and chemical processes to precipitate metals in various forms, such as carbonates, hydroxides, or sulphides (e.g., biosolids generated from a pulp and paper mill). At MMSL, some investigations have been ongoing related to a passive system at Teck Cominco in Trail, British Columbia (i.e., referred to as the Teck Cominco System (TCS)). At this site a combination of anaerobic bioreactors (ABRs) and vegetated cells are used to treat metal contaminated leachate containing arsenic, cadmium and zinc (Duncan et al., 2004). These systems are often treated as “black boxes”, and although to some extent successful, there are times when effluent quality does not meet expectations. Nature Works Remediation has operated the TCS for eight years, not without challenges, however practical “on the job” experience and research networks have helped overcome many challenges to create an overall story of progress. MMSL’s interest in the TCS continues due to several key factors: at this point in time it is one of only a few working examples in Canada; it concentrates a significant amount of arsenic with little to no iron present in the influent; it is believed that the  technology is being applied appropriately (i.e., it concentrates contaminants into a small volume from a relatively low flow that originates over a large land mass); each year of operation provides more insight into its maintenance and longevity; and when its capacity is reached the choices made to decommission or rebuild it will be an important example for others. Several experimental approaches were taken by MMSL; this report details the results of the duplicate column study. Objective The objective was to gain an understanding of the removal of arsenic, cadmium and zinc, in an ABR such as the one operated by Teck Cominco. To meet the objective: the physical zones at which optimal treatment parameters were maximized were identified; the treatment rates and the residence times required were measured; the likely mechanisms in which the contaminants were precipitated were identified; the activities of relevant microbes were profiled; and some implications for a field operation were considered. Experimental - Conceptual Design and Construction The columns were built with Plexiglas, while the piping was rigid PVC and the tubing was Nalgene. A Masterflex peristaltic pump was used to pump the synthetic contaminated leachate (i.e., the feed), which was directed vertically from the bottom-up. The total volume of each column was 15.6 L: the initial layer contained ~2 L of 3 mm CaCO3; the next layer contained ~10.8 L of compost and gravel (i.e., organic substrate); and on top there was 2.8 L of headwater. The organic substrate was obtained from the Celgar “Kraft” Pulp and Paper Mill in Trail, British Columbia. Headwaters were maintained over each column as a buffer between it and the ambient air. An important component of the column design was the addition of ports, to sample pore water in various regions. Table 1 summarizes physical aspects of the columns. Table 1: Column Study Material Components Cell  Materials/Components  Physical Parameters  Anaerobic Bioreactor (Total Volume 15.6L)  Bottom section: - 6 mm Limestone (CaCO3)  Volume (V)= 2 L Cross Sectional Area (A) = 184 cm2 Height (h) = 11cm Porosity (n) = 0.255 Voidage (Vn) = 0.5 L Hydraulic Conductivity (k)= 7.17 cm/s Volume (V)= 10.8 L Cross Sectional Area (A) = 184 cm2 Height (h) = 59 cm Porosity (n) = 0.286 Voidage (Vn) = 3.1 L Hydraulic Conductivity (k)= 1.51 x 10-3 cm/s Volume (V) = 2.8L  Middle section: - Composted, Celgar Pulp mill sludge - Industrial grade sand Top section: -Standing Effluent  The expectation was to create a medium for limestone dissolution, according to equations 1-4 (Skousen, 1991) as well as for the fermentative (i.e., acid-producing bacteria (APB)) and sulphate-reducing bacteria (SRB). Limestone dissolution produces alkalinity (i.e., Zipper, 2001, equation 5). SRB also contribute to alkalinity when reducing sulphate (e.g., equation 6). APB metabolize complex and simple organics in the organic substrate to produce compounds such as ethanol and acetate that can be further metabolized by SRB. The microbial production of sulphides allows for cadmium and zinc sulphide precipitation (Ying et al., 2001), and possibly arsenic. Equation 7 shows the reaction for divalent metals (Drury, 1999).  CaCO3 + 2H+ → Ca2+ + H2O + CO2 CaCO3 + H+ → Ca2+ + HCO3CaCO3 + H2CO3 → Ca2+ + 2HCO3CaCO3 → Ca2+ + CO32HCO3- + H+ → H2O + CO2 (aq)  (1) (2) (3) (4) (5)  SO42- + 2CH2O → S2- + 2HCO3- + 2H+ H2S + Me2+ → MeS + 2H+  (6) (7)  Acclimatization took place in the first month following construction. The acclimatization period involved the establishment of SRB by loading the columns with the synthetic feed. This feed contained ~126 mg/L [Zn2+], ~25 mg/L [As5+ as AsO43-], ~2 mg/L [Cd2+], and ~218 mg/L [SO42-] at a pH ~5.5. At a flow of 77 L/m2/day, the feed met the criteria of 0.3 mol/m3/day used by Gusek et al. (1997). The proportion of feed contaminants was based on averages observed entering the TCS. After acclimatization, operation at a temperature ~25°C commenced and was divided into three phases: (1) For a two month period the columns were operated at a hydraulic loading of 77 L/m2/day; (2) The hydraulic loading was doubled to stress the system for a two month period; (3) The hydraulic loading was returned to 77 L/m2/day to evaluate whether the system could re-establish its key operational conditions observed in Phase 1. Monitoring and Analyses Once a week the input and outputs of each column, as well ports 1 through 4 were monitored with a WTW Multiline P4 Universal Pocket Meter for temperature, pH, redox potential (ORP), dissolved oxygen (DO), and conductivity. As well, samples were collected for total arsenic, cadmium, zinc, iron, sulphate, and acidity/alkalinity. Samples for dissolved metals were filtered with a 0.45 μm filter before acidification with concentrated nitric acid. Total arsenic, cadmium, iron, and zinc in the mg/L range were determined using ICP-AES, and in the μg/L range were determined using ICP-MS. Sulphate was determined using a gradient elution HPLC, with a DIONEX. Total acidity was determined using a two-endpoint (pH 4 and 8.3) titration using the NaOH technique to determine the equivalence point. Total alkalinity was determined in water by titration of a sample aliquot with a standard solution of dilute acid. A two-endpoint (pH 4.5 and 4.2) technique determined the inflection point or equivalence point of the titration. Before and after each adjustment in the experiment hydraulic loading rate: SRB, iron reducing bacteria (IRB), and APB were enumerated using a most probable number (MPN) determination from the first, second, third and fourth pore water sample ports. For SRB determination, the bacteria were grown in modified Postgate medium C in 20 mL serum bottles (Postgate, 1984). The composition of the medium was as prescribed by Benner et al. (1999). IRB were completed as described by Gould et al. (2003). APB enumerations were completed as prescribed by Hulsof et al. (2003). RESULTS AND DISCUSSION Optimal Parameters – Hydraulic Loading, Retention Time, pH, Dissolved Oxygen, and Oxygen Reduction Potential Generally for Phases 1, 2 and 3, the last four weeks were accepted as representing a steady state; and therefore, served as the basis for most of the discussion. For Phases 1 and 3, residence times  of 7 hrs, 18 hrs, 31 hrs, 44 hrs, and 60 hrs, and half that for Phase 2, correspond to column heights of 9 cm, 20 cm, 36 cm, 51 cm, and 85 cm from the base. Table 2 shows the hydraulic loadings and the total retention time for the columns at various phases. Table 2: Phase 1, Hydraulic Loading rates Days 1-161 Phase  Column  #1  #1 #2 #1 #2 #1 #2  #2 #3  Influent (L/Day) 1.42 ± 0.05 1.42 ± 0.05 2.68 ± 0.08 2.81 ± 0.13 1.44 ± 0.12 1.32 ± 0.06  Retention Time (Days) 2.5 ± 0.1 2.5 ± 0.1 1.3 ± 0.04 1.3 ± 0.06 2.5 ± 0.21 2.7 ± 0.11  As expressed by Gusek et al. (m2min/L)* 19 19 9.9 9.5 19 20  Hydraulic Loading (L/m2/day)* 77 77 146 153 78 72  * The hydraulic loading value does not account for porosity; rather it is based on the total theoretical cross-sectional area.  Monitoring of the system parameters indicated that the total retention time was adequate to provide the conditions for sulphate-reduction and stability of metal sulphide precipitates. However, the physical point at which the mechanisms within the columns attained each condition was dependent on the hydraulic loading. Figures 1, 2 and 3 illustrate the rate at which pH, DO, and ORP were acclimatized within the columns for each phase. The limestone basin was the largest contributor to alkalinity (i.e., 3.5 to 7 hrs). The microbial contribution to alkalinity was clearly reflected only in the pH of Phase 1 after 20 hrs. Data from Phase 3 indicated some increase, but the increase was slight, suggesting that the population did not rebound after the setback in Phase 2. Adjusting the pH to neutral helped facilitate the proper environment for the APB and SRB. An optimal pH of 7.8 for SRB was never obtained in either phase (Postgate, 1984). The rate of DO removal accelerated in the organic substrate. Overall the longer retention times in Phases 1 optimized the decrease in DO concentrations. The maximum removal rate observed in Phase 2 after 15 hrs was consistent with the other phases, but this point was physically shifted upwards in the columns compared to Phases 1 and 3 (i.e., 36 cm as opposed to 20 cm from the base). In contrast to trends in pH, the DO reduction performance was responsive to a correction of flows in Phase 3, and show that DO concentrations in different regions of the columns were directly dependent on the hydraulic loadings. pH DO ORP  Arsenic, Cadmium and Zinc Treatment and Retention The graphs in Figures 4, 5 and 6 show the removal rates for contaminant concentrations for each phase. The trends were consistent with adjustments in redox conditions discussed in the previous section. The best performance was  8.0  0.3  Limestone  7.0  0.25  6.0  0.2  5.0  0.15  4.0 0.1  3.0  0.05  2.0  0  1.0  -0.05  0.0 0  5  10  15  20  25  30  35  40  45  Time (Hours)  Figure 1: Phase 1, pH, DO and ORP for each Column as a Function of Residence Time  ORP (v)  pH and DO  The ability to reduce ORP became poorer after each consecutive phase. However, in Phase 2 the rate of reduction was actually inhibited between 9 cm and 20 cm of the columns length compared to Phase 1 and 3.  7.0  DO  ORP 0.3  Limestone  6.0  0.25 0.2  4.0 0.15 3.0  ORP (v)  pH and DO  5.0  0.1  2.0  0.05  1.0 0.0  0 0  5  10  15  20  25  Time (Hours)  Figure 2: Phase 2, pH, DO and ORP for each Column as a Function of Residence Time pH 7.0  DO  ORP 0.3  Limestone  6.0  0.25  5.0  0.2  4.0 0.15  ORP (v)  In Phase 1 the total cadmium and zinc concentrations for each column were treated to below 0.06 mg/L and 0.003 mg/L, in Phase 2 between 0.22 - 4.3 mg/L and 0.003 mg/L, and in Phase 3 between 1.7 – 14 mg/L and 0.001 mg/L respectively (Figures 4, 5 and 6). The rate at which removal took place within the columns, correlated with the physical substrate passed, suggested that there were two different mechanisms sequentially taking place. Initially as the feed passed through the limestone basin a first order rate constant of 0.018 applied; however, upon entering the organic substrate the rate accelerated, and a rate constant of 0.19 applied. In Phase 2, the rate constant of 0.018 was applicable throughout most of the column. With the limestone/organic interface, the total treatment ranged from 20 - 30% for zinc and 10 - 20% for cadmium. Within the organic substrate the initialization of an accelerated removal rate for cadmium and zinc were likely the result of metal sulphide precipitation, as shown in equations 6 and 7.  pH  pH and DO  observed in Phase 1. The trends in cadmium and zinc removal were very similar in many respects; however, the treatment of cadmium occurred preferentially. The bulk of the treatment appeared to be tied to sulphate removal (Table 3) and the efficiency was negatively impacted by the increased hydraulic/DO loadings. In all three phases the trends in total cadmium and zinc removal depicted a different treatment mechanism than that of the arsenic.  3.0 0.1  2.0  0.05  1.0 0.0  0 0  5  10  15  20  25  30  35  40  45  Time (Hours)  Figure 3: Phase 3, pH, DO and ORP for each Column as a Function of Residence Time  In Phase 1, the reactions approached completion over a residence time of 44 hrs signifying that the substrate volume was adequate for the flowrate. In Phase 2 the removal rates were maintained; however, flowrate did not provide enough residence time for the complete removal of zinc. In Phase 3, zinc removal did not respond to the re-establishment of an optimal hydraulic loading. As well, deviations in the zinc concentrations for each column in Phase 2 persisted in Phase 3. The higher loading of DO in the upper regions of the columns provoked in Phase 2 can account for this. The DO reduced the SRB population/activity in the upper regions of the column. The poor performance in Phase 3 compared to Phase 1 after 44 hrs was likely a result of the population’s inability to rebound under a flow regime. Microbial counts discussed in a later section support this interpretation. The implications with respect to hydraulic loadings are important because the northern Rockies are subject to higher flows during the spring freshet; therefore, without controls the probability of loading an ABR beyond its design criteria is high.  Concentration (mg/L)  1000  As  Cd  Max DO Reduction  Max pH, and ORP Reduction  100 10.0 1.00 0.10 0.01 0.00 0  10  20  40  30  50  60  Time (Hours)  Figure 4: Phase 1 Mean Arsenic, Cadmium and Zinc (mg/L) for each Column as a Function of Residence Time Limestone/ Organic Interface  Concentration (mg/L)  As indicated in Figures 4, 5 and 6, arsenic removal in all three phases took place between the columns input and the limestone/organic substrate interface, and appeared to be unaffected by an increased hydraulic loading. Some arsenic persisted in the pore waters of the organic substrate. The concentrations increased and decreased in relation to total iron concentrations (i.e., solid analysis shows the pre-existence of iron in the ABRs to be 17.4 mg/g). Bar graphs (Figure 7) illustrate the correlation between arsenic and iron for the headwaters. The trends suggested that arsenic was adsorbed onto iron associated with the substrates matrix (McGeehan, 1996; Smedley et al. 2002). As these zones became reducing some iron reduced to ferrous (Johnson et al. 2004) and released the associated arsenic. Once mobile, the elements could be transported through the columns. Upon reaching the columns final headwaters, an oxygen abundant zone, ferrous oxidized to a ferri(oxy)hydroxide, and co-precipitated the arsenic (Ball et al. undated; Tingzing et al., 1997). In a review of arsenic disposal  Zn  Limestone/ Organic Interface  1000  Zn  As  Max pH  Cd Max DO Reduction  Max ORP Reduction  100 10.0 1.00 0.10 0.01 0.00 0  5  10  15  20  25  30  Time (Hours)  Figure 5: Phase 2 Mean Arsenic, Cadmium and Zinc (mg/L) for each Column as a Function of Residence Time  Concentration (mg/L)  The removal of zinc and its retention as a metal sulphide was further quantified by conducting a mass balance of zinc and sulphur. The cadmium loadings were too small for a mass balance to provide supportive evidence for cadmium sulphide production. The mass balances for each phase, column, limestone basin, and organic substrate are tabulated in Table 3. The calculations assumed that the sulphur associated with the organic substrate became available for the metal sulphide precipitation reaction and zinc associated with the gravel did not (i.e., chemical analyses conducted but not included). If the ratios were compared on a stage-by-stage basis, the results were confounding; however, when the total zinc versus sulphur was compared for an entire column, the molar ratio was 1:1. The balance does not conflict with the possibility of other less significant removal mechanisms (i.e., carbonate or hydroxide precipitation), because it is unlikely that all the sulphur previously associated with the organic substrate reacted to form a zinc sulphide.  1000  Zn  Limestone/ Organic Interface  As  Cd Max pH and ORP Reduction  100 Max DO Reduction  10.0 1.00 0.10 0.01 0.00 0  10  20  30  40  50  60  Time (Hours)  Figure 6: Phase 3 Mean Arsenic, Cadmium and Zinc (mg/L) for each Column as a Function of Residence Time  practices, Riveros (2001) describes the precipitate as arsenate “tenaciously adsorbed on ferrihydrate by complexation (chemisorption), forming a stable inner-sphere complex”. Total and dissolved analyses of arsenic, cadmium, zinc and iron were performed on the columns headwater. The results tabulated in Table 4 show that most of the cadmium and zinc were dissolved; however, the majority of iron and arsenic were not. This re-enforced that the behaviour of cadmium and zinc in the pore waters were independent of iron and arsenic. Table 3: Total Zinc and Sulphur Balance in the Organic Substrate for each Column Phase  Column #1 Column #2 Zn (mmoles) S (mmoles) Zn (mmoles) S (mmoles) Limestone Organic Matrix Limestone Organic Limestone Organic Matrix Limestone Organic  1 2 3 Zn:S  48 99 30  122 231 103  9 43 0.32 631  374  633  101 82 22  47 103 30  124 237 60  13 36 3  374  601  92 91 10  619  The column labelled Matrix refers to sulphur associated with the solid matrix before the feed addition, while Limestone and Organic refer to zinc and sulphur removed from the feed in the limestone basin and the organic substrate respectively. TFe  12.00  30  1  12.00  10.00  10.00 8.00 8.00 6.00  6.00  4.00  4.00 2.00  2.00  0.00  0.00 15  22  29  34  43  0.8  20 0.6 15 0.4  10 5  0.2  0  0  58  64  71  78  Time (Days)  TFe  14.00  10.00  10.00  8.00  8.00  6.00  6.00  4.00  4.00  2.00  2.00  0.00  0.00 22  29  34  43  Time (Days)  Column 2(Phase 1)  58  TFe Concentration (mg/L)  12.00  15  112  119  1 0.8  15  0.6  10  0.4 5  0.2 0 134  141  TFe  50  2  40 1.5 30 1 20 0.5  10 0  0 71  78  85  92  155  161  Column 1(Phase 3)  TAs 2.5  64  148 Time (Days)  60  TAs Concentration (mg/L)  12.00  8  99  Column 1(Phase 2)  TAs  14.00  0  92  1.2  Time (Days)  Column 1(Phase 1) TFe  85  1.4  20  0  99  112  Time (Days)  Column 2(Phase 2)  119  TA s  25 TAs Concentration (mg/L)  8  25  1.6  25  TAs Concentration (mg/L)  14.00  TAs  30 TFe Concentration (mg/L)  1.2  TFe Concentration (mg/L)  35  0  TFe Concentration (mg/L)  TAs  14.00  TAs Concentration (mg/L)  TFe  TAs  TAs Concentration (mg/L)  TFe Concentration (mg/L)  TFe 16.00  1 0.9 0.8 0.7  20 15  0.6 0.5 0.4 0.3 0.2 0.1 0  10 5 0 134  141  148  155  161  T ime ( D ays)  Column 2(Phase 3)  Figure 7: Total Arsenic (Red) and Iron (Blue) Concentration Trends (mg/L) in the Compost/Headwater of each Column There was further evidence to suggest arsenic and iron were mobile under reducing conditions. Immediately following the initial acclimatization, the one time the limestone basin approached anaerobic conditions (i.e., average DO 1.5 mg/L), arsenic was not removed from the regions pore water. On day 0 in Column 1, the concentration of arsenic in the limestone basin was 32.5 mg/L, and in Column 2 it was 22.2 mg/L. On day 0, arsenic was absent only in the columns oxygenated headwater. These observations do not incorporate the formation of arsenic sulphides, which have been the focus of some discussion and observations made for the TCS (Duncan et al, 2004; Gould et al, 2002). Cutler (2004) determined the oxidation states of arsenic in work related to the TCS,  using x-ray absorption near edge structure (XANES), at the National Synchrotron Light Source, Brookhaven National Laboratories. The analysis showed the likelihood that arsenic sulphides were present, but once exposed to oxygen, oxidation was rapid. The formation of iron(oxy)hydroxides in the columns was consistent with field observations. It has been observed during several field visits to the TCS that iron(oxy)hydroxides deposit around the input of the vegetated cell following the ABR. Table 4, Total and Dissolved Arsenic, Cadmium, Iron, and Zinc in the Final Headwater/Effluent of each Column in Phase 3 Day  134 141 148 155 161 Day 134 141 148 155 161  Column 1 Tot. Cd Dis. Cd (mg/L) (mg/L) 0.0026 0.0002 0.0014 <0.0007 <0.088 <0.088 <0.032 n.d. <0.036 <0.036 Tot. Zn Dis. Zn (mg/L) (mg/L) 2.36 1.75 1.52 <1.38 1.65 n.d. 2.15 n.d. 1.63 1.52  Column 2 Tot. Cd Dis. Cd (mg/L) (mg/L) <0.0022 <0.0022 0.0008 <0.0007 <0.088 <0.088 <0.032 n.d. <0.036 <0.036 Tot. Zn Dis. Zn (mg/L) (mg/L) 12.2 11.2 16.2 14.8 16.8 16.0 20.2 n.d. 18.9 18.7  Column 1 Tot. Fe Dis. Fe (mg/L) (mg/L) 26.46 0.993 6.27 0.229 15.6 <0.38 7.2 n.d. 6.86 <0.0887 Tot. As Dis. As (mg/L) (mg/L) 1.34 0.074 0.39 0.047 1.03 <0.3 <0.5 n.d. 0.35 <0.19  Column 2 Tot. Fe Dis. Fe (mg/L) (mg/L) 22.8 0.18 17.1 0.21 19.9 <0.38 13.5 n.d. 10.6 0.13 Tot. As Dis. As (mg/L) (mg/L) 0.93 0.057 0.64 0.026 0.93 <0.3 0.59 n.d. 0.38 <0.19  Microbial Population Profiles – APB, IRB and SRB After the initial acclimatization of the columns (Day 0), an APB, IRB and SRB profile was completed at each sample port; another profile was conducted after Phase 1 (Day 58); and a final profile was completed after Phase 2 (Day 119). Figures 8, 9 and 10 illustrate the shift in each population’s magnitude with respect to the columns length. In general, following acclimatization, each population was established at its peak level. APB remained fairly well distributed throughout the columns, and there wasn’t any significant change between Phase 1 and 2 with the exception of Phase 2 (Column 2) that illustrated an unexplained die-off. It is worth noting the zinc treatment performance deviated over time in Column 2. As expected, IRB were not robust at the limestone/organic interface, but within the organic substrate, populations were slightly elevated, and Phases 1 and 2 were comparable. Column 1, Day 0  Column 2, Day 0  Column 1, Day 58  Column 2, Day 58  Column 1, Day 119  Column 2, Day 119  1.00E+06 1.00E+05 MPN (cells/mL)  The IRB populations play a significant role in the mobilization of iron (Sorensen, 1981) and as a consequence arsenic, as discussed in the previous section. Shifts in SRB populations dramatically responded to adjustments in hydraulic/oxygen loadings. Following acclimatization, populations were elevated in all areas of the columns other than the limestone/organic interface, which was still relatively good. Oxygen loadings during Phase 1 operations significantly hindered the SRB populations for the first 20 hrs of residence time, but after 31 hrs the ABRs maintained anaerobic conditions. Phase 2 further impacted the populations upward in the columns and  1.00E+04 1.00E+03 1.00E+02 1.00E+01 1.00E+00 9  23  37  Sample Port (cm)  Figure 8, APB as a Function of the Columns Height within the Pore Water of each Column Column  51  beyond the final sample port. The response was a reflection of the SRB sensitivity to oxygen, and the trends were consistent with the treatment efficiencies for zinc.  The study placed emphasis on the limestone basin to establish pH conditions more conducive to microbial growth. With respect to its role in treatment, removal rates for cadmium and zinc were illustrated at room temperature. However, comparing substrates, sulphate reduction, and redox conditions in the columns, it was unlikely that the metals were precipitated as sulphides. Arsenic removal in the pore water of the limestone/organic interface was significant. In the absence of SRB and sulphate reduction, the best explanation was its adsorption to iron preexisting in the organic substrate.  Column 1, Day 58  Column 1, Day 119  Column 2, Day 119  1.00E+05 MPN (cells/mL)  Establishment of Anaerobic Conditions, Treatment, and Metal Retention  Column 2, Day 0  Column 2, Day 58  1.00E+06  1.00E+04 1.00E+03 1.00E+02 1.00E+01 1.00E+00 9  23  37  51  Sample Port (cm)  Figure 9, IRB as a Function of the Columns Height within the Pore Water of each Column Column 1, Day 0  Column 2, Day 0  Column 1, Day 58  Column 2, Day 58  Column 1, Day 119  Column 2, Day 119  1.00E+06 1.00E+05 MPN (cells/mL)  CONCLUSIONS  Column 1, Day 0  1.00E+04 1.00E+03 1.00E+02 1.00E+01 1.00E+00  The most significant removal of cadmium and 9 23 37 51 zinc occurred in the upper regions of the Sample Port (cm) organic substrate and was related to metal Figure 10, SRB as a Function of the Columns sulphide production; however, the rates Height within the Pore Water of each Column suggested that as much as 31 hrs residence time was required to achieve maximum depletion of DO and 44 hrs was required to maximize a reduction of the redox conditions for its stability. Arsenic retention in the organic substrate was not consistent and appeared to be correlated with the iron, which again suggested arsenic retention was the result of adsorption to iron and faltered when reducing conditions converted ferric to ferrous. The relationship could potentially act as a transport mechanism to move low concentrations of arsenic through the ABRs, where it finally co-precipitates with iron(oxy)hydroxides in the oxic headwaters, and forms a deposit. Solid samples were not removed from the system for analyses in order to maintain the integrity of the system for further work. Cold temperature work is currently being pursued. Field System Considerations (1) With respect to arsenic, a design should consider the placement of an oxic polishing step after an ABR to precipitate iron, thereby co-precipitating/adsorbing arsenic. The mobility of arsenic at the limestone/organic interface during acclimatization would suggest that precipitated arsenic could possibly remobilize in anaerobic conditions (e.g., hypothetically, in the spring after a winter shutdown, or start-up after a prolonged work stoppage due to maintenance or repairs.)  (2) With respect to hydraulic loadings, the DO loading appears to have a direct role in staging the mechanisms that lead to cadmium and zinc precipitation as a sulphide. Good controls (e.g., use of a feed reservoir and pump) can help manage periods of high regional flow due to spring freshet or heavy periods of rain. It is also suggested that the hydraulic loadings not be actively increased beyond its original design criteria due to treatment success. This could potentially reduce the active volume that facilitates metal sulphide precipitation. It may not be apparent in the short term, but excessive loadings may significantly shorten treatment-life-expectancy. Increasing the size or the addition of cells should be given preference. (3) With respect to system maintenance and recovery, the inability of the system to recover within 1½ months under a flow regime should be considered against the operation’s optimal season, and emphasis placed on the worse case scenario, especially in Canada where summer operation may be as short as 2-months. The use of carbon that can be added to a feed (e.g., acetate, ethanol, and sugar) for unforeseen events should be anticipated in the initial planning. ACKNOWLEDGEMENTS The authors wish to acknowledge: CANMET’s Analytical Services Group for the chemical analysis; Dr. David Koren for his editing, and conceptual input; as well as, Al Mattes of Nature Works Remediation his editing of the report. Finally, Al Mattes and Bill Duncan of Teck Cominco for their conceptual support, for their time in the field, and making the field system accessible, without which much perspective and relevance would be lost. REFERENCES Ball, Brandon R., P.E. and Kevin V. Brix. Undated. Passive Treatment of Metalloids Associated with Acid Rock Drainage. Parametrix, Inc. Sumner, Washington. Benner, S. G., D.W. Blowes, W.D. Gould. R.B. Herbert, JR., and C.J. Ptacek. 1999. Geochemistry of Permeable Reactive Barrier for Metals and Acid Mine Drainage. Environmental Science and Technology, 1999, 33, 2793-2799. Cutler, Jeffery. 2004. Determination of the Oxidation States of Arsenic found in an Anaerobic Bioreactor. Canadian Light Source, preformed for Nature Works Remediation Corp. (unpublished). Drury William J. 1999. Treatment of Acid Mine Drainage with Anaerobic Solid-Substrate Reactors. Water Environment Research, Volume 71, Number 6, pg 1244-1250. Duncan, William F., Al Mattes, and W. Doug Gould. 2004. Multi-stage Biological System for Metal Removal-Anaerobic Cell Deconstruction. Teck Cominco, Trail British Columbia, Nature Works, Trail, British Columbia, and CANMET, Ottawa, Ontario. Gould, W.D. and M. Skaff. 2002. Microbial Transformations of Arsenic in Anaerobic Cells. Natural Resources Canada (CANMET), Mining and Mineral Science Laboratories, Ottawa, Ontario, Canada (unpublished). Gould, W.D., M. Stichbury, M. Francis, L. Lortie, and D.W. Blowes. 2003. An MPN Method for the Enumeration of Iron-Reducing Bacteria. Sudbury 2003 Proceedings.  Gusek, Jim and Wildeman, Tom. 1997. Short Course #6 Treatment of AMD. Fourth International Conference on Acid Rock Drainage Vancouver, BC, Canada, May 31-June 6. Hulsof, A. H. M., David W. Blowes, Carol J. Ptacek, and W. Douglas Gould. 2003. Microbial and Nutrient Investigations into the Use of in Situ Layers for Treatment of Tailings Effluent. Environmental Science and Technology, 2003, 37, 5027-5033. Johnson, D. Barrie and Kevin B. Hallberg. 2004. Biogeochemistry of the Compost Bioreactor Components of a Composite Acid Mine Drainage Passive Remediation System. School of Biological Sciences Memorial Building, Deiniol Road Bangor, LL57 2UW, UK. McGeehan, Steven L. 1996. Arsenic Sorption and Redox Reactions: Relevance to Transport and Remediation. Journal of Environmental Science and Health, A31(9), 2319-2336. Postgate, J.R. 1984. The Sulphate-Reducing Bacteria. Cambridge University Press: Cambridge, England. Riveros, P.A. and J.E. Dutrizac. 2001. A Review of Arsenic Disposal Practices. Proceedings of EMC 2001. CANMET, Natural Resources Canada, 555 Booth St., Ottawa, Canada K1A 0G1. Skousen, J.G. 1991. Anoxic Limestone Drains for Acid Mine Drainage Treatment. Green Lands 21(4): 30-35. Smedley, P.L., D.G. Kinniburgh. 2002. A Review of the Source, Behaviour and Distribution of Arsenic in Natural Waters. Applied Geochemistry 17 (2002) 517-568. Sorensen, Jan. 1981. Reduction of ferric Iron in Anaerobic, Marine Sediment and Interaction with Reduction of Nitrate and Sulfate. Applied and Environmental Microbiology, Feb 1982, p. 319324. Tingzong Guo, R.D. Delaune, and W.H. Patrick, Jr. 1997. The Influence of Sediment Redox Chemistry on Chemically Active Forms of Arsenic, Cadmium, Chromium, and Zinc in Estuarine Sediment. Environment International, Vol. 23, No. 3, pp. 305-316. Ying Song, Mark Fitch, Joel Burken, Linda Nass, Somnath Chilukiri, Nord Gale, Cade Ross. 2001. Lead and Zinc Removal by Laboratory-Scale Constructed Wetlands. Water Environmental Research, Volume 73, Number 1. Zipper C. and C. Jage. 2001. Passive Treatment of Acid-Mine Drainage with Vertical-Flow Systems. Reclamation Guidelines for Surface Mined Land in Southwest Virginia. Powell River Project. Virginia Cooperative Extension Publication 460-133, Virginia State University.  


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