&EPA United States Environmental Protection Agency EPA/540/SR-93/523 September 1993 SUPERFUND INNOVATIVE TECHNOLOGY EVALUATION Emerging Technology Summary Handbook for Constructed Wetlands Receiving Acid Mine Drainage A treatment technology based on constructed wetlands uses natural geochemical and biological processes inherent in the aqueous environment and designs a system to optimize pro- cesses best suited to removal of con- taminants specific to the site. Key features of this wastewater technology are that it is a passive treatment sys- tem, the cost of operation and mainte- nance is significantly lower than that for active treatment processes, and the removal methods try to mock rather than overcome natural processes. In this study, the contaminant waters were metal-mine drainages with low pH (<3.0) and high concentrations of metals (Al, Mn, Fe, Ni, Cu, Zn, and Pb). From studies done at constructed wetlands at the Big Five Tunnel near Idaho Springs, Colorado, the important process for raising the pH and remov- ing metals was found to be bacterial sulfate reduction followed by precipita- tion of metal sulfides. By optimizing the process and determining how to properly load the wetland with contami- nant drainage, the following results were achieved: • pH was raised from 2.9 to 6.5. • Dissolved AI, Cu, Zn, Cd, Ni, and Pb concentrations were reduced by 98 % or more. • Iron removal was seasonal with 99% reduction in the summer. • Mn reduction was relatively poor unless the pH of the effluent was raised above 7.0. • Biotoxicity to fathead minnows and Ceriodaphnia was reduced by factors of 4 to 20. Once it was found that microbial pro- cesses were primarily responsible for contaminant removal, a staged design process comparable to the design pro- cess used for other wastewater treat- ment technologies was devised. Laboratory studies determine whether in principle contaminants could be re- moved and the best substrate combi- nation for their removal. Bench scale studies determine the optimum loading capacity and treatment system configu- ration. From these studies design of a reasonably sized module that is spe- cific to the site can proceed with the expectation that it will successfully treat the contaminanted water. This summary was developed by EPA's Risk Reduction Engineering Laboratory, Cincinnati, OH, to announce key findings of the SITE Emerging Tech- nology Program that is documented in a separate report (see ordering infor- mation at back). OyD Printed on Recycled Paper ------- Introduction In response to the Superfund Amend- ments and Reauthorization Act of 1986, (SARA), the U. S. Environmental Protec- tion Agency's (EPA) Office of Research and Development (ORD) and the Office of Solid Waste and Emergency Response (OSWER) have established a formal pro- gram to accelerate the development, dem- onstration, and use of new or innovative technologies as alternatives to current con- tainment systems for hazardous wastes. This program is called Superfund Innova- tive Technology Evaluation or SITE. The SITE program is part of EPA's re- search into cleanup methods for hazard- ous waste sites throughout the nation. Through cooperative agreements with de- velopers, alternative or innovative tech- nologies are refined at the bench-scale and pilot-scale level and then demon- strated at actual sites. EPA collects and evaluates extensive performance data on each technology to use in remediation de- cision making for hazardous waste sites. The report summarized here documents the results of laboratory and pilot-scale field tests on the applicability of sulfate- reducing bacteria operating in the anaero- bic zone within a wetland constructed to remove contaminant metals associated with mine drainages. These metals can include Al, Mn, Fe, Co, Ni, Cu, Zn, As, Ag, Cd, Hg, and Pb. In the mine drainage Water used in this study, Mn, Fe, Cu, and Zn are the primary contaminants in the water. Also, most mine drainages have an actdic pH that causes concern and has to be Increased to effect treatment. In the water used in this study, the average pH is 3.0. The Concept of Constructed Wetlands Ecotogists have long understood that soils in wetlands are often foul because they naturally accumulate contaminants by: • filtering of suspended and colloidal material from the water; • uptake of contaminants into the roots and leaves of live plants; * adsorption or exchange of contami- nants onto inorganic soil constituents, organic solids, dead plant material, or algal material; • neutralization and precipitation of con- taminants through the generation of HCOj'and NH3 by bacterial decay of organic matter; « destruction or precipitation of contami- nants in the aerobic zone catalyzed by the activity of bacteria; and • destruction or precipitation of chemicals in the anaerobic zone catalyzed by the activity of bacteria. With so many possible removal pro- cesses, a wetland, such as depicted in Figure 1 , is the typical contaminant treat- ment system in a natural ecosystem. In addition, it operates in a passive mode requiring no additional reactants and no continuous maintenance. In the last decade, engineers began to use wetlands to remove contaminants from water. In some instances, natural wetlands were used. A natural system however, will accommodate all the above removal pro- cesses and probably will not operate to maximize a certain process. A constructed wetland, on the other hand, can be de- signed to maximize a specific process suit- able for the removing of certain contami- nants. Engineering and ecological reasons lead to using a constructed wetland for contaminant removal rather than using an existing natural ecosystem. As an example of constructing a wet- land to maximize specific removal pro- cesses, consider the bacterial processes that are items 6 and 7 in the above list. Typical microbially mediated reactions that are possible in the aerobic zone of a wet- land include: 4 Fe2* + O2 + 10 H2O — > 4 Fe(OH)3 + 8 H+ 2 02 + H2S — > S04- + 2 H* 2 H20 + 2 N2 + 5 Oz— > 4 NCy + 4 H* Typical microbially mediated reactions that are possible in the anaerobic zone of a wetland include: 4Fe2*+CO2-i-11 H2O 5 CHO + 4 NO- + 4 H* — > 2 N + 5 C0 7 H20 SO4- +2 CH2O — > H2S 2 HCO- In these reactions, "CH2O" is used to symbolize organic material in the substrate. It is apparent that the anaerobic reac- tions are approximately the reverse of the aerobic reactions. Both zones exist in a wetland. If removal involves aerobic pro- cesses, then the wetland should be con- structed so the water remains on the sur- face. If removal involves anaerobic pro- cesses, then the wetland should be con- structed so the water courses through the substrate. In a natural wetland, the water primarily remains on the surface. In the important area of microbially me- diated removal, the wetland must be con- structed to maximize removal reactions and minimize competing reactions. When removing contaminants from acid mine drainage, the removal processes should consume hydrogen ions and, conse- quently, anaerobic processes are empha- sized. The research and development at the Big Five Tunnel site in Idaho Springs, Colorado has concentrated on understand- ing the chemistry and ecology involved in removal and designing structures from readily available materials that maximize these processes. Although this appears to be "low tech- nology", an intense interdisciplinary effort and creative engineering skills are needed to design and perfect systems that maxi- mize natural processes. For more details on what should be considered, The full report cites recent references. The Big Five Pilot Wetland The research reported here has involved studying removal processes from a pilot constructed wetland designed to receive metal mine drainage from the Big Five Tunnel in Idaho Springs, CO. The chemis- try from the adit drainage is reasonably constant throughout the year and is sum- marized in Table 1. After a number of modifications of the pilot cells, removal results were excellent. Table 1. Contaminant Concentration, Big Five Tunnel Drainage, Averages. Constituent Concentration,mg/L Mn Fe Co Ni Cu Zn Cd Pb 31 38 0.10 0.15 0.73 9.4 0.03 0.03 Figure 2 shows the removal trends for a 2-year period as outflow concentrations over influent concentrations. Cu and Zn are completely removed; Fe removal changes with the seasons. During this 2-year period, analysis of chemical data accumulated at the site led to the conclusion that microbial reduction of sulfate to sulfide followed by precipita- tion of heavy metal sulfides is the pre- dominant process accounting for the re- moval of over 90 % of the Fe, Cu, and Zn and the rise in the pH from below 3 to ------- above 6. Procedures for construction of wetland ceils that emphasize anaerobic removal processes has been considered. Another consideration is how to employ knowledge of the biochemistry of sulfate reduction in the wetland design process. This has been done by using two ideas particularly suited to wetlands-ideas that employ anaerobic removal processes such as sulfate reduction: The limiting reagent concept, and staged design of wetlands. The Limiting Reagent Concept In the design of wetlands for wastewa- ter treatment, there is a strong emphasis on determining the loading factor which gives an indication of how large a wetland should be to remove the contaminants of concern. This can be stated as the amount of square feet of wetland per gallon per minute of water to be treated or as the grams of contaminant removed per day per square meter of wetland. In our expe- riences at the Big Five site, typical mea- sures of loading factor do not seem to explain the removal of metals even though heavy metals such as Cu and Zn are Porous Rock Dam Figure 1. Diagram of a typical wetland ecosystem that emphasizes subsurface flow. Cell E Removal Trends From Sept. 1989 1.00 MnE FeE CuE ZnE ._. SO4E 0.00 Months Figure 2. Two year removal trends for a subsurface wetland cell located at the Big Five Tunnel in Idaho Springs, Colorado. 3 ------- reduced by greater than 99 %. We have discovered that a key factor in sulfate reduction is to Insure that the optimum microenvironment for sulfate-reducers is maintained. The most important environ- mental conditions are reducing conditions and a pH of around 7. Since the wetland cell is receiving mine drainage with pH below 3 and Eh of above 700 mV, the water can easily overwhelm the micro- environment established by the anaerobic bacteria. This leads to the limiting reagent concept for determining how much water can be treated, as an alternative to the use of typical bading factors. Consider the following precipitation re- action: Fe^-t-S-- ->FeS At high flows of mine drainage through the substrate, sulfide will be the limiting reagent, the microbial environment will be under stress to produce more sulfide, the pH of the microenvironment will drop, and removal will be Inconsistent. At low flows of mine drainage through the substrate, iron will be the limiting reagent, the ex- cess sulfide will insure a reducing envi- ronment and a pH near 7, the microbial population will remain healthy, and removal of the metal contaminants will be consis- tent and complete. Using this idea, load- ing factors should be set to insure that the heavy metal contaminants are always the limiting reagents. The question then is how much sulfide can a colony of sulfate-re- ducing bacteria produce per cubic centi- meter of substrate per day? Studies by the U. S. Bureau of Mines wetlands group suggests that a reason- able figure for sulfide generation is 300 nanomole sutfide/cubic cm/day. This num- ber, the volume of the wetland cell, and the metals concentrations in the mine drainage are used to set the flow of mine drainage through the wetland cell. Using this concept in a subsurface wetland cell to determine the loading factor has re- sulted in year round complete removal of Cu and Zn (Rgure 2). Staged Design of Wetlands After determining that precipitation of metals by sulfide generated from sulfate- reducing bacteria is the important process, H was realized that establishing and main- taining the proper environment in the sub- strate is the key to success for removal. This means that processes operating on the surface of the wetland are not that Important. In particular, plants are not nec- essary In a wetland emphasizing subsur- face processes. If this is so, then a large pilot cell, such as was built at the Big Five site, is not necessary to determine whether a wetland that emphasizes anaerobic pro- cesses for removal will work. Conse- quently, the study of wetland processes and the design of optimum systems can proceed from laboratory experiments to bench scale studies to design and con- struction of actual cells. We call this "staged design of wetland systems". In current laboratory studies, culture bottle experiments are used for funda- mental studies on how to establish simple tests to determine the production of sul- fide by bacteria, and of what substrate will provide the best initial conditions for growth of sulfate-reducing bacteria. In these ex- periments, laboratory production of sulfide at 18 °C has been 1200 nanomole/gm of dry substrate/day. Culture bottle tests have shown that in the case of cyanide, sulfate reduction was retarded until the concentration of total cyanide was below 10 mg/L and that Cu concentrations above 100 mg/L would kill or retard sulfate-reducing bacteria. How- ever, other culture bottle tests have also shown that sulfate reduction was still vig- orous at Cu and Zn concentrations above 100 mg/L. For bench scale studies, plastic gar- bage cans are used to conduct experi- ments to provide answers necessary to the design of a subsurface cell, e.g. deter- mining the optimum loading factor, sub- strate, cell configuration, and substrate permeability. In a recent study, garbage cans filled with substrate were used to determine whether using the sulfide gen- eration figure of 300 nanomole sulfide/cm3 of substrate/day could be used to set the conditions for treating severely contami- nated drainage that flows from the Quartz Hill Tunnel in Central City, CO. Contami- nant concentrations are shown in Table 2. Using the limiting reagent concept de- scribed above and the amount of sub- strate contained in the garbage can, flow could not exceed 1 ml/min to ensure that sulfide would always be in excess. Con- taminant concentrations from the outputs of three different bench scale cells are shown in Table 2. For cell A the mine drainage was passed through the cell with no delay. For cell B the substrate was soaked with city water for one week be- fore mine drainage started passing through the cell. For cell C, the substrate was inoculated with an active culture of sul- fate-reducing bacteria and soaked with city water for one week before mine drainage started passing through the cell. Prepara- tions on cells B and C were done to en- sure that there would be a healthy popu- lation of sulfate-reducing bacteria before mine drainage flowed through the sub- strate. All cells were run in a downflow mode of the mine drainage through the substrate. In all three cells removal of Cu, Zn, Fe, as well as Mn is greater than 99%. The increase in pH is from about 2.5 to above 7. These results were con- sistently maintained for over ten weeks of operation. The substrate used was a mix of 3/4 cow manure and 1/4 planting soil. The results from cells B and C show that the cow manure has an indigenous popula- tion of sulfate-reducing bacteria that are quite active. Inoculation with an active cul- ture of bacteria is not necessary in this case. Also, since the results from cell A are comparable to those of cells B and C, the population of sulfate reducers can with- stand immediate exposure to severe mine drainage and still produce sufficient quan- tities of sulfide. The key to good initial activity is to ensure that the flow of mine drainage is low enough that its low pH does not disturb the micro-environment established by the bacteria. These bench scale systems also serve as permeameters and thus can provide important information for others aspect of wetland design. Determination of soil con- ductivity and how this physical property changes with time is found to be an im- portant geotechnical parameter for the de- sign of subsurface constructed wetlands. Conclusions Using constructed wetlands for waste- water treatment is still a developing tech- nology. The results from the Big Five Pilot Wetland study, however, show promising removal of heavy metals and increase of pH for acid mine drainage. Conclusions from the project include: • Toxic metals such as Cu and Zn can be removed and the pH of mine drainage can be increased on a long term basis. • The major removal process is sulfate reduction and subsequent precipitation of the metals as sulfides. Exchange of metals onto organic matter can be important during the initial period of operation. • A trickling filter type of configuration achieves the best contact of the water with the substrate. • Removal efficiency depends strongly on loading factors. In the Big Five wetland and in bench scale studies, flow of water should not exceed the 300 nanomoles/cmVday of sulfide that can be generated by the microbes in the substrate. ------- • Permeability of the substrate is a critical design variable for successful operation. Using laboratory and 5. bench-scale tests, a good indication of the soil permeability in a constructed wetland can be determined. • As with, any other wastewater removal technology, design of a constructed wetland or passive bioreactor is specific to the site and the water to be treated. 6. • A staged design and development sequence can be used where laboratory studies are used to determine the best conditions and substrate, bench scale experiments help to determine loading factors and substrate properties, and pilot modules test the performance of a typical field module. References 7. 1. Reed, S. C., Middlebrooks, E. J., and Crites, R. W. Natural Systems for Waste Management and Treat- ment. McGraw-Hill, New York, 1988. 308pp. 2. Hammer, D. A. Constructed Wet- lands for Wastewater Treatment. 8. Lewis Publishers, Chelsea, Michi- gan, 1989. 800 pp. 3. Kleinmann, R. L P. (ed.), Proceed- ings of a Conference on Mine Drain- age and Surface Mine Reclamation. Vol. 1. Mine Water and Mine Waste. U. S. Department of the Interior, Bureau of Mines Information Circu- lar 1C 9183, 1988, 413 pp. 9. 4. Wildeman, T. R., and Laudon, L. S. The use of wetlands for treatment of environmental problems in min- ing: Non-coal mining applications. In: D. A. Hammer (ed.), Constructed Wetlands for Wastewater Treat- ment. Lewis Publishers, Chelsea, Michigan, 1989. p. 221. Howard, E. A., Emerick, J. C., and Wildeman, T. R. The design, con- struction and initial operation of a research site for passive mine drain- age treatment in Idaho Springs, Colorado. In: D. A. Hammer (ed.), Constructed Wetlands for Waste- water Treatment. Lewis Publishers, Chelsea, Michigan, 1989. p. 761. Machemer, S. D., Lemke, P. R., Wildeman, T. R., Cohen R. R., Klusman, R. W., Emerick, J. C., and E. R. Bates. Passive Treat- ment of Metals Mine Drainage through use of a Constructed Wet- land. In: Proceedings of the 16th Annual Hazardous Waste Research Symposium, U. S. EPA, Cincinnati, OH, 1990. EPA Document No. EPA/ 600/9-90-037, pp. 104-114, 1990. Machemer, S. D., and Wildeman, T. R., Organic Complexation Com- pared with Sulfide Precipitation as Metal Removal Processes from Acid Mine Drainage in a Constructed Wetland". Jour. Contaminant Hydrology,vol. 9, pp. 115-131,1992. Hedin, R. S., and R. W. Nairn. Siz- ing and Performance of Constructed Wetlands: Case Studies. In: Pro- ceedings of the 1990 Mining and Reclamation Conference, J. Skousen, J. Scencindiver, and D. Samuel eds., West Virginia Univ. Publications, Morgantown, WV, 1990. pp. 385-392 Wildeman, T. R., Machemer, S. D., Klusman, R. W., Cohen, R. R., and P. Lemke. Metal Removal Efficien- cies from Acid Mine Drainage in the Big five Constructed Wetland. In: Proceedings of the 1990 Mining and Reclamation Conference, J. Skousen, J. Scencindiver, and D. Samuel eds., West Virginia Univ. Publications, Morgantown, WV, 1990. pp. 417-424. 10. Mclntire, P. E., and H. M. Edenborn. The use of Bacterial Sulfate Re- duction in the Treatment of Drain- age from Coal Mines. In: Proceedings of the 1990 Mining and Reclamation Conference, J. Skousen, J. Scencindiver, and D. Samuel eds., West Virginia Univ. Publications, Morgantown, WV, 1990. pp. 409-415. 11. Reynolds, J. S., Machemer, S. D., Wildeman, T. R., Updegraff, D. M., and R. R. Cohen, 'Determination of the Rate of Sulfide Production in a Constructed Wetland Receiving Acid Mine Drainage". Proceedings of the 1991 National Meeting of the American Society of Surface Min- ing and Reclamation, ASSMR, Princeton, WV, 1991, pp 175 -182 12. Filas, B., and T. R. Wildeman, The Use of Wetlands for Improving Wa- ter Quality to Meet Established Standards", Nevada Mining Assoc. Annual Reclamation Conference, Sparks, NV, May, 1992. 13. Bolis, J. L., Wildeman, T. R., and R. R. Cohen, ' The Use of Bench Scale Permeameters for Preliminary Analysis of Metal Removal from Acid Mine Drainage by Wetlands". Proceedings of the 1991 National Meeting of the American Society of Surface Mining and Reclamation, ASSMR, Princeton, WV, 1991, pp 123-136. 14. Wildeman, T. R., Brodie, G. A., and J. J. Gusek, Wetland Design for Mining Operations, Bitech Publish- ing Co. Vancouver, BC, Canada, 1992, 300 pp Table 2. Constituent concentrations in mg/L in the Quartz Hill Tunnel mine drainage and in effluents from the bench scale tests Sample Mine Drainage Cell A CellB CellC Mine Drainage CellA CellB CellC Mine Drainage CellB Cell C Days Operated 24 24 24 24 43 43 43 43 71 71 71 Mn 80.0 0.94 0.91 0.99 80.0 0.97 0.64 1.6 70.0 0.48 1.6 Fe 630.0 1.6 1.9 1.0 640.0 0.87 0.96 0.46 820.0 0.40 0.40 Cu 48.0 0.06 <0.05 <0.05 50.0 <0.05 <0.05 <0.05 70.0 <0.05 <0.05 Zn 133.0 0.27 0.17 0.16 135.0 0.18 0.24 0.14 101.0 0.21 0.25 so4 4240 450 70 412 4300 1080 660 1180 NA NA NA pH 2.4 7.4 7.5 7.4 2.5 7.2 7.4 7.2 2.6 8.0 7.9 •fru.8. GOVERNMENT PRINTING OFFICE: IM3 - 7SO-071/80M* ------- ------- ------- Thomas Wildeman is with the Department of Chemistry and Geochemistry, Colorado School of Mines, Golden, CO 80401 Edward R. Bates is the EPA Project Officer (see below). The complete report, entitled "Handbook For Constructed Wetlands Receiving Acid Mine Drainages," (Order No. PB93-233914AS; Cost: $36.50, subject to change) will be available only from: National Technical Information Service 5285 Port Royal Road Springfield, VA 22161 Telephone: 703-487-4650 The EPA Project Officer can be contacted at: Risk Reduction Engineering Laboratory U.S. Environmental Protection Agency Cincinnati, OH 45268 United States Environmental Protection Agency Center for Environmental Research Information Cincinnati, OH 45268 Official Business Penalty for Private Use $300 BULK RATE POSTAGE & FEES PAID EPA PERMIT No. G-35 EPA/540/SR-93/523 ------- |