| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Colorado School of Mines, Division of Environmental Science and Engineering, Coolbaugh Hall, Golden, CO 80401
* Corresponding author (rcohen{at}mines.edu)
Received for publication January 13, 2002.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: SRB, sulfate-reducing bacteria
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Infiltration of snowmelt and rainfall into mine workings, waste rock, and tailings perpetuates the contamination of downstream aquatic environments. High concentrations of sulfate and ferrous iron and additional metals such as lead, zinc, copper, cadmium, and manganese may result. This aquatic impairment not only poses an immediate threat to the fauna and flora of the region, but, left unchecked, often has deleterious effects on human activities such as recreation, irrigation, industry, and livestock watering for the duration of acid-rock drainage generation.
Typical acid-rock drainage wastewater treatment processes employed by active and inactive mines are (i) lime and/or caustic soda chemical precipitation and (ii) sulfide addition, although sulfide addition is infrequently used. These treatment processes can be expensive and inefficient, and generate large quantities of wet sludge.
One innovative technique for the treatment of acid-rock drainage has been the use of natural and artificial wetlands as a biological pollution abatement process (Dvorak et al., 1992). The self-regenerative properties of biological treatment systems have been thought to reduce the need for continuous maintenance, offering an attractive, alternative abatement technology to the conventional systems. The biological systems also avoid the production of copious amounts of wet sludge associated with oxidative and hydrolytic processes. These passive biological systems can also operate at a fraction of the production and maintenance costs of the conventional chemical and physical treatment methods.
Cohen and Staub (1992), Reynolds (1991), and Machemer (1992) examined the chemical and biological processes in wetland treatment systems receiving acidic drainage and found that the rate of sulfate reduction was the most crucial process involved.
The metal-removing characteristics of the natural wetland have been improved through ongoing research and engineering development. Factors such as hydraulic conductivity, hydraulic detention time, and loading rates can now be manipulated through a passive treatment system to provide greater removal efficiency and throughput. The generic wetland substrates have been replaced with organic materials selected to provide an optimum microenvironment for sulfate-reducing and other heterotrophic bacteria. However, these passive systems continue to be limited by the low throughput and long contact and hydraulic detention times required to achieve greater than 95% removal efficiency of metals.
In order for the sulfate-reducing bacteria to thrive, they require a strict anaerobic environment (they are obligate anaerobes) with a pH in the range of 5 to 8 (Brown et al., 1973). When pH and/or redox conditions are not optimum, the rate of microbial sulfate reduction declines. This, in turn, reduces metal removal capacity. The rapid influx of acidic, aerobic waters appears to drive the pH of the treatment system down and redox up, thus inhibiting bacterial sulfate reduction. The metal removal efficiency and loading capacity of the treatment system then becomes a function of not only size and hydraulic conductivity, but of the acidity and oxygen content of the influent water.
As obligate anaerobes, it is important that their microenvironment remain anaerobic, and that an reduction–oxidation (redox) potential of less than -100 mV be maintained (Brown et al., 1973; Reynolds, 1991; Tuttle et al., 1969b). Since redox potential is dependent, among other parameters, on pH, it becomes necessary to maintain a pH of 5 to 8 as a condition of sulfate reduction. The sulfate reduction reaction is capable of contributing HCO-3 alkalinity to the system to maintain an "optimal" pH range. Lower pH inhibits sulfate reduction and increases the chemical instability of any metal sulfides that may have been formed.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
|
The effluent was recovered from a point 2.5 cm below the standardized bed level for all of the reactors (identified as Port 6 on all reactors; Fig. 2). The two primary reactors, A1 and B1, had additional sample ports attached at regular intervals along their height; Port 1 at 25 cm from the bottom of the reactor, Port 2 at 48 cm, Port 3 at 71 cm, Port 4 at 95 cm, and Port 5 at 118 cm, with the effluent Port 6 at 141 cm. These sampling ports consisted of perforated HDPE tubing, which spanned the inside diameter of the reactor and were covered with water-permeable geomembrane to prevent passage of solids from the substrate. The HDPE tubing was connected to a hose barb mounted in the reactor casing, to which an external tube was attached and from which liquid samples could be drawn. The total internal volume of each sample port was approximately 10 mL. Therefore, 10 mL was purged from each port before sample collection to ensure a fresh sample.
Each reactor was packed with a composite of composted livestock manure and porous ceramic pellets. The pellets were used as a bulking agent to ensure adequate permeability.
A layer of pea gravel, approximately 4 cm thick, was loaded into the base of each reactor and was covered with permeable geomembrane. The pea gravel disperses the influent across the reactor. The manure–ceramic pellet substrate mixture was wetted with water in a 113-L vessel and then loaded as a slurry into the reactors. Washed fiberglass insulation was used as a head space fill material; it is light, porous, and does not contribute significantly to compaction of the substrate. Sand was then added for the remaining 5 cm of each reactor. Each alternating layer (substrate, fiberglass, and sand) was separated by a disc of permeable geomembrane to reduce migration of solid materials between layers.
Simulated Mine Drainage
An artificial solution, with a neutral pH and elevated metals concentrations, was mixed to simulate acid-rock drainage (ARD). Simulated ARD was used because the neutral pH conditions for the experiment would induce precipitation of most metals in actual ARD. Also, the simulated ARD permitted better experimental control. The simulated mine drainage was stored in a 1136-L (300-gallon) plastic vessel. The effluent from each reactor was collected in a similar vessel. The simulated mine drainage composition is presented in Table 1
.
|
All four reactors were inoculated with a sample of organic substrate from another, working reactor that contained a consortium of sulfate-reducing bacteria. The inoculum speeds the development of bacterial biomass in the bioreactors (Bolis et al., 1991). One gallon (3.78 L) of the SRB inoculate media was pumped into each reactor.
Experimental Design
Experiment 1, Series 1: Neutral Drainage pH
In Experiment 1, metal-leaden, simulated mine drainage was created with a near-neutral pH. Aluminum and copper were omitted from the simulated mine drainage due to their propensity to precipitate at neutral pH. Three metal ions were used: zinc, manganese, and cadmium. The simulated mine drainage was pumped through the reactors for 14 d.
The use of the four reactors of different dimensions (A1, A2, B1, and B2), and two different pump tubing sizes, without recirculation, permitted four hydraulic detention times to be tested concurrently. The flow rates were adjusted to produce approximate empty-bed hydraulic detention times of 30 h for A1 and A2 and 85 h for B1 and B2. The system flowed for an additional 7 d during which time periodic measurements were made on samples obtained from each effluent port (Port 6) and the influent tank. Data were collected on pH, redox potential (mV), and conductivity (S/m). These variables were used to monitor the approach to steady state conditions for the reactor and to determine an appropriate time at which to sample. Steady-state conditions were decided to exist when these variables were stable for three or more days.
After 21 d of flow of the original simulated mine drainage, 100-mL samples were drawn from each of the 14 separate sample ports and the influent tank. Physical parameters, including pH, redox potential (mV), and conductivity (S/m), were measured on all samples. The samples were filtered through 0.45-µm glass fiber filters. Anions were analyzed with a Dionex (Sunnyvale, CA) ion chromatograph. The remaining samples were acidified with 1 mL concentrated nitric acid and stored until cation analysis could be performed.
Experiment 1, Series 2: Neutral Drainage pH
Identical simulated drainage solution as that for Experiment 1, Series 1 was used for this experiment. Tubing sizes and pump speeds were varied to obtain four different flow rates to the four reactors, simultaneously. The influent flowed for 10 d at approximate hydraulic detention times of 20 h for A1, 55 h for A2, 15 h for B1, and 50 h for B2. Physical measurements were made on the effluents to confirm steady state conditions. After 10 d, samples were collected and analyzed.
Experiment 2: Feed with Low pH and Dissolved Oxygen in Equilibrium with the Atmosphere
For Experiment 2, all of the constituent metals were dissolved in the feed water and the pH was reduced to 2.7. The sulfate concentration was decreased to 500 mg/L. This concentration was still in excess of the earlier calculated minimum requirement of 170 mg/L SO2-4 needed by the SRB. The tank was allowed to equilibrate with atmospheric oxygen. The solution flowed for 17 d at reactor hydraulic detention times of 30 for A1, 95 h for A2, 30 h for B1, and 80 h for B2, after which samples were collected. Measurements on pH, redox potential, dissolved oxygen, and conductivity were made throughout the 17-d period. Once the pH, redox potential, and conductivity stabilized, 20- to 30-mL samples were drawn for analysis.
Experiment 3: Low pH and Reduced Dissolved Oxygen Drainage
High concentrations of dissolved oxygen may reduce metal removal efficiencies of the reactors due to the increase of redox potential beyond the level at which SRB can optimally function. The examination of the effect of dissolved oxygen on treatment efficiencies was conducted using drainage containing all of the metal constituents, including aluminum and copper. The water then was purged of dissolved oxygen. The synthetic drainage had a pH of 2.7. The drainage was allowed to flow at >75-h hydraulic detention times for 5 d before deoxygenating the influent tank.
Traditional techniques for oxygen removal include simple methods, such as purging with an inert gas and vacuum degasification, to more elaborate catalytic, industrially applied methods. However, due to the high cost of most of these methods, we removed dissolved oxygen with sodium bisulfite.
Suitable oxygen scavengers are strong reducing agents that have an exothermic heat of reaction with O2 and a kinetic requirement of reasonable reactivity. Reagents that fulfill these requirements include chromous sulfate, vanadous sulfate, sodium sulfite(s), and hydrazine. Of these reagents, the sulfite compounds, which include sodium sulfite (Na2SO3), sodium bisulfite (NaHSO3), sodium metabisulfite (Na2S2O5), and ammonium sulfite [(NH4)2SO3·H2O], were the most suitable for this experiment, though the use of the latter was restricted due to the possible formation of ammonia gas. Sodium bisulfite reacts according to the following equation:
 |
Though this reaction contributes to the total dissolved solids of the solution, it also yields an additional sulfate source that could be utilized by the SRB. The reaction would also contribute protons to the solution.
The rate of the sulfite-oxygen reaction is dependent on both temperature and pH. An increase in temperature decreases reaction time. We allowed the drainage water to reach thermal equilibrium with the atmosphere and maintained a circumneutral pH until deoxygenating was complete.
Stochiometric calculations using a ratio of 2 mol NaHSO3 to 1 mol O2 and assuming a saturation of 8.0 mg/L O2 at a temperature of 28°C required the use of 49.3 g of sodium bisulfite to deoxygenate the remaining 946 L (250 gallons) of synthetic acid-mine drainage. To maintain low dissolved oxygen through the experiment, 147.7 g of sodium bisulfite (or three times the calculated value) was used. The concentration of dissolved oxygen dropped to 0.0 mg/L almost instantaneously after the addition of the sodium bisulfite. Plastic bubble wrap covered the water surface in the synthetic mine drainage feed tank to minimize reaeration.
The reactor flow rates were adjusted to establish hydraulic detention times of 30 h for A1, 90 h for A2, 25 h for B1, and 75 h for B2. The flow rates were maintained for 15 d. Physical measurements were taken of the effluent from Port 6. Dissolved oxygen was now included among the measurements. Samples were taken and analyzed according to the same procedures as in Experiments 1 and 2. Instead of 100-mL samples, two 30-mL samples were taken from each sample port. The second sample was sent to an outside laboratory for aluminum analysis.
Experiment 4: Bench-Scale Substrate Sorption Test
Because adsorption to the solid phase can remove a large mass of metal ions in wet-substrate bioreactors, it was important to quantify the sorptive capacity of the composted manure substrate. Two 1-L solutions, containing 75 mg/L copper, were created using 0.4 g cupric chloride, dihydrate (CuCl2·2H2O) dissolved in deionized water. The two beakers were stirred on magnetic stir plates. At time (t) = 0, a 2-mL sample was drawn from each beaker. Once this sample was taken, 10 g of dry manure used in the previous experiments was added to the stirring solution. The two samples taken before introduction of the manure were filtered through a 0.8 µm/0.2 µm syringe-tip filter. A 1-mL sample was drawn from this solution and added to 9 mL of deionized water to create a 1:10 dilution. Sample collection was repeated at time intervals (t) = 10, 20, 35, 60, 180, 1440, 2880, 4320, and 5760 min. The samples were analyzed for copper.
Experiment 6: Composite Substrate Adsorption Reactor Test
To measure the effective adsorption capacity of the larger reactors (A1, A2, B1, and B2), a smaller version was constructed. Table 2 contains the technical specifications for this reactor.
|
A single-channel, variable-speed peristaltic pump was used to deliver the influent drainage to the reactor. Hydraulic detention time was not constant at 40 h during the experiment but varied from a low of 30 to 36 h (measured on three occasions) to more than 40 h. The average hydraulic detention time, however, was maintained above the 40 h threshold.
The experiment ran for 60 d. Thirty-milliliter samples were collected approximately every 3 d. The pH, redox potential, and conductivity were measured. The samples were filtered (0.45-µm filters), acidified using 0.3 mL concentrated nitric acid to preserve metals in solution, and stored until the metals could be analyzed.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
There appeared to be a direct relationship between metal removal capacities and metal loading rates. Based on daily molar metal removal amounts, a three-times increase in flow yielded a three-times increase in moles of metals removed between the two sets of reactors (Fig. 3) .
|
The molar sulfate loss per day appeared independent of the loading rate of sulfate for all hydraulic detention times examined (Fig. 3).
At 85- and 86-h hydraulic detention times, Reactors B1/1 and B2/1 had 0.023 and 0.026 mol/d of sulfate removed, respectively. Even though Reactors A1/1 and A2/1 had flow rates three times greater than B1/1 and B2/1, and the sulfate loading rate was three times greater, the sulfate loss was similarly 0.023 mol/d for A1/1 and 0.016 mol/d for Reactor A2/1. The results indicate that, in these experiments, the SRB may be at zero-order reaction rates with respect to sulfate.
The removal rates of sulfate, if predominantly due to sulfate reduction, are higher than those reported by Hedin et al. (1989) of 300 nmol SO2-4 per cm3 of substrate per day, and Reynolds (1991) of 600 nmol SO2-4 per cm3 of substrate per day. These rates are near or higher than the rate of those previously reported and indicate that either there are larger numbers or higher activities of SRB per unit volume of reactor than in previously reported systems, resulting in higher sulfate reduction.
An alternative sink for sulfate is the formation of gypsum (CaSO4). Though no calcium was added directly, the manure could have provided a source of Ca2+. Gypsum is formed under oxidizing conditions and is manifested by white crystalline precipitates. All four reactors remained under reducing conditions, top to bottom, during both flow series. No white crystalline scale was observed. Thus, most of the sulfate loss can be presumed to be attributed to reduction to sulfide.
If the rate of sulfate removal is interpreted to be reduction to sulfide, then the moles of S2- per cm3 per day can be assumed to react with metal cations on a mole to mole basis. With a four-reactor average of 820 nmol per cm3 per day of sulfide produced during the first flow series, a loading rate of 820 nmol per cm3 per day of cationic metals could be precipitated as metal sulfides (or acid volatile sulfides [AVS]; Reynolds, 1991). During the second flow series, this average increased to 1241 nmol per cm3 per day of sulfide produced (Table 4) .
|
The results indicate that all measured metals, except for manganese, were consistently removed at near 100% removal efficiency at all hydraulic detention times. In reactors with detention times greater than 54 h, excess S2- was detected as H2S.
Figure 3 shows metal and sulfate removal rates as a function of flow rates (the higher the flow rate, the shorter the hydraulic detention time). The regressed sulfate lines are almost horizontal, demonstrating the independence of sulfate reduction rate to flow rate (and, therefore, sulfate loading rate) (Table 5) .
|
Metal Loading Rate Capacity under Acidic Conditions—Oxygen in Equilibrium with Atmosphere
The flow rates to each reactor were 14.5, 5, 14.9, and 5 mL/min for A1, A2, B1, and B2, respectively. The flow rates corresponded to empty bed hydraulic detention times of 34, 97, 28, and 84 h, respectively. The influent contained zinc, manganese, cadmium, aluminum, and copper.
There was removal of cadmium, aluminum, and copper at nearly 100% removal efficiency. Zinc removal efficiency was 98.6% for Reactor A2; 94 and 93% for Reactors A1 and B2, respectively; and only 69.2% for Reactor B1.
The removal of manganese was increasingly inefficient, suggesting perhaps that adsorption could no longer fully remove it from solution and that the reduced pH of the system was inhibiting the formation of MnCO3. Adsorption is as much a function of pH as it is of the species' affinity to sorb, for which manganese has perhaps the weakest affinity of all of the constituent metal cations.
The mass loss of sulfate still remained constant over the four reactors. Anion samples in the acidic, elevated dissolved oxygen experiment were stored for 60 d. There was evidence of evaporation from the stored samples. Sulfate was added to the influent to a concentration of 500 mg/L. The analysis of the influent yielded a concentration of 922 mg/L. Accuracy may have been poor for the sulfate measurement, but precision may have been preserved. The total moles of sulfate lost per day appears to have remained consistent from experiment to experiment and reactor to reactor (Fig. 4 ; Table 6) .
|
|
The redox potential and pH data indicate that short-circuiting may have occurred in Reactor A1. The deviation from port-to-port trend of concentration of zinc, manganese, and cadmium at Port 5 supports this conclusion.
Metal Loading Rates and Hydraulic Detention Times under Acidic pH and Reduced Dissolved Oxygen Conditions
The pH of the influent drainage was decreased from 6.2 to 2.7. To examine the effect of dissolved oxygen and elevated redox potential on the activity of the SRB and metal removal efficiency, the dissolved oxygen was driven to near 0.0 mg/L using sodium bisulfite. We expected low pH to inhibit sulfate-reducing activities and low oxygen to enhance activity (Postgate, 1984).
The flow rates to each reactor were 17, 5.4, 16.9, and 5.5 mL/min for A1, A2, B1, and B2, respectively. The flow rates yielded empty bed hydraulic detention times of 29, 90, 25, and 76 h, respectively.
The first important phenomenon observed was that sulfate reduction was lower than the levels observed in the circumneutral pH drainage experiment, even though the hydraulic detention times were longer (Table 7) .
|
|
|
The "optimum" flow rates were lower, and hydraulic detention times were higher compared with the results in the neutral pH experiments (Fig. 5).
Metal removal efficiency decreased with the lower pH influent. Cadmium was removed to below detection limits (except in Reactor A1 where efficiency dropped from 100 to 96.9%). Zinc removal dropped to levels of 96.9% for A1, 80.3% for A2, 90.1% for B1, and 93.4% for B2. These results were not unexpected for the two higher flow reactors, A1 and B1, because their metal loading rates exceed sulfide production rates. The removal efficiency reduction observed in Reactors A2 and B2, at 90.1- and 75.7-h hydraulic detention times, was unexpected. The decrease in metal removal efficiency was observed in spite of the fact that sulfate reduction rates exceeded metal loading rates (Table 8).
Lower pH may have increased the formation of H2S, causing the loss of S2- that otherwise would be used for metals precipitation. The lower pH could have mobilized sulfides that were created and precipitated during the previous two experiments.
Of all of the metals, manganese removal efficiency declined the most dramatically with the application of low-pH influent. The manganese removal efficiencies for Reactors A1 and B1 were 36.3 and 2.8%, and 72.4 and 71.3% for Reactors A2 and B2, respectively. Both A2 and B2 attained system pH values of 7.1 and 7.0, which were more favorable to the formation of rhodochrosite than those of A1 and B1, at 6.7 and 6.3, respectively. Because of the assumption that manganese would be removed in the form of rhodochrosite (MnCO3), it follows that by reducing the pH of the system, the formation of MnCO3 was inhibited. The manganese passed unimpeded through the system, except for adsorption, because manganese sulfides are not stable under anaerobic conditions experienced in the reactors.
Considering that dissolved oxygen fluctuated around 1.0 mg/L in the circumneutral pH drainage experiment, while sulfate reduction remained high, and was approximately 0.0 mg/L in the acidic-reduced dissolved oxygen experiment, it is reasonable to assume that dissolved oxygen did not significantly affect treatment efficiency of these systems. The pH of the influent drainage was the predominant controlling variable.
Bench-Scale Substrate Sorption Test
Keeping in mind that sodium azide was added to the solution to inhibit the activity of the SRB, sorption was probably the predominant copper removal process. The concentrations of copper (mg/L) were plotted against time to create Fig. 6
.
|
Composite Substrate Adsorption Reactor Test
To examine the substrate under conditions resembling those of the previous experiments, a solution of 106.5 mg/L zinc, 63.9 mg/L manganese, and 0.37 mg/L cadmium was pumped continuously to a 2.7-L reactor filled with 2.2 L of the organic substrate mixture. The SRB inhibitor, sodium azide, was added to ensure that adsorption of the metals was the predominant removal mechanism. Analysis of the data obtained from this adsorption reactor test is presented in Fig. 7
.
|
These two experiments support the premise that adsorption is a transient metal removal mechanism. Without active sulfide precipitation, the treatment capacity of the systems is severely limited.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
As the experiments progressed, we observed a decrease in metal removal capacities of the short detention time reactors. At high flow rates and short hydraulic detention times, metal loading rates may exceed the combination of sulfate reduction rates and sorption. As sorption sites approach capacity, strongly sorbing metals like copper will displace those that do not have as strong an affinity for surface bonding.
Regardless of the loading rate of sulfate to each of the reactors, the data support the conclusion that the mass removal rate of sulfate was constant during each experiment and from reactor to reactor. The zero-order removal rate in relation to sulfate concentration delineates the maximum level at which sulfide precipitation can function as the principal metal removal mechanism. The metal loading rates would need to be adjusted to match this sulfate reduction level to avoid contaminant breakthrough once the transient effects of adsorption removal processes were exhausted.
The experiments have shown that pH is more critical to reactor efficiency than dissolved oxygen. The metal loading rate capacity of a wet-substrate bioreactor can be enhanced, and hydraulic detention times reduced, by modifying the pH of the influent to near neutral. The neutral pH permits enhanced activity of SRB, the production of sulfide, and the removal of metals as metal sulfide precipitates. The dissolved oxygen was almost completely removed on entering the reactors, while the pH required a greater proportion of the substrate to reach suitable levels. Since the SRB are not only obligate anaerobes, but also require a narrow pH range near neutral for "optimal" sulfate reduction, it is reasonable to assume that pH is the predominant controlling factor. Optimization and enhancement of the reactors may be accomplished through the modification of mine drainage to a pH between 5 and 8 (Connell and Patrick, 1968). Perhaps the use of anoxic limestone drains (ALDs) before the SRB bioreactor would raise the pH to levels more acceptable to the bacteria. The ALDs could simultaneously decrease the dissolved oxygen content of the influent as well.
It appears that a significant proportion of the short-term removal of metal ions was due to adsorption onto the organic manure substrate. Experimental results indicate that adsorption was limited by the sorptive capacity of the substrate. Different substrate materials will have different sorptive capacities. Combined with the fact that adsorption in these systems is also a function of pH and the particular metal species being removed, each system will be unique in character. A universal sorption capacity cannot be assigned to passive mine drainage systems as a whole, but the experimental results are similar to those of Machemer (1992), that is, adsorptive capacity lasts for 30 to 60 d. Reactor life-span is 4 to 6 yr. Thus, most attention should be focused on optimizing conditions for sulfate reduction.
The experiments have shown that pH is critical to reactor efficiency. The metal loading rate capacity of a wet-substrate bioreactor can be enhanced, and hydraulic detention times reduced, by modifying the pH of the influent to near neutral. The neutral pH permits enhanced activity of SRB, the production of sulfide, and the removal of metals as metal sulfide precipitates. Since the SRB are not only obligate anaerobes, but also require a narrow pH range near neutral for "optimal" sulfate reduction, it is reasonable to assume that pH is the predominant controlling factor. Optimization and enhancement of the reactors may be accomplished through the modification of mine drainage to near-neutral pH. Perhaps the use of anoxic limestone drains before the SRB bioreactor would raise the pH to levels more acceptable to the bacteria.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  L. E. Kanagy, B. M. Johnson, J. W. Castle, and J. H. Rodgers Jr. Hydrosoil conditions in a pilot-scale constructed wetland treatment system for natural gas storage produced waters Environmental Geosciences, September 1, 2008; 15(3): 105 - 113. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.-M. Neculita, G. J. Zagury, and B. Bussiere Passive Treatment of Acid Mine Drainage in Bioreactors using Sulfate-Reducing Bacteria: Critical Review and Research Needs J. Environ. Qual., January 9, 2007; 36(1): 1 - 16. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Soil Science Society of America Journal | Journal of Plant Registrations | The Plant Genome | |||
