2002 Progress Report: Metals Removal by Constructed Wetlands

EPA Grant Number: R828770C005
Subproject: this is subproject number 005 , established and managed by the Center Director under grant R828770
(EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).

Center: HSRC (2001) - Midwest Hazardous Substance Research Center
Center Director: Banks, M. Katherine
Title: Metals Removal by Constructed Wetlands
Investigators: Fitch, Mark W. , Burken, Joel
Institution: Missouri University of Science and Technology
EPA Project Officer: Lasat, Mitch
Project Period: October 1, 2001 through September 30, 2004
Project Period Covered by this Report: October 1, 2001 through September 30, 2002
Project Amount: Refer to main center abstract for funding details.
RFA: Hazardous Substance Research Centers - HSRC (2001) RFA Text |  Recipients Lists
Research Category: Hazardous Waste/Remediation , Land and Waste Management


This research project focuses on the use of constructed wetlands for the remediation of mine drainage. The primary focus is lead mine drainage, because it is a significant problem in Missouri, and many industrial effluents are essentially neutral, as is lead mine drainage. The objectives of this research project are to: (1) determine the chemistry of metals removal in constructed wetlands; (2) determine the failure mode of constructed wetlands at low hydraulic residence times; (3) determine the bioavailability of sequestered metals in constructed wetlands; and (4) determine the effect of operational disturbances on constructed wetlands.

Progress Summary:

The first objective is nearing completion; analyses by scanning electron microscope are underway. Results reveal that three mechanisms dominate metals removal in the wetlands. The three removal mechanisms are: (1) adsorption, primarily to organic substances in the wetlands; (2) co-precipitation with iron oxyhydroxides; and (3) precipitation as metal sulfides. The relative importance of and removal by these three mechanisms will vary from wetland to wetland, based on: media selection, influent water composition, and biological activity in the wetland. Adsorption characteristics for different wetland substrates can vary greatly, given the wide range of materials that have and can be used. We determined individual Langmuir isotherms for the materials used in the laboratory-scale wetlands (see Figure 1). From these isotherms, the sorption capacity and breakthrough times for lead and zinc were predicted. Breakthrough times for the laboratory-scale wetland ranged roughly from 3 to 74 years. We developed predictions, assuming 100 percent sorption efficiency and no competing removal mechanisms. Second-order kinetics relationships also were developed, allowing for approximation of the rate of removal by sorption. To better ascertain what the relative importance of adsorption is in the laboratory wetland, we passed a 1M MgCl2 solution through a wetland to remove zinc and lead sorbed after 52 months of operation. Effluent sampling showed that roughly 13 percent of the lead and 5 percent of the zinc in the wetland were desorbed during the desorption experiment. Based on the specific materials used, this approach can be applied to constructed wetland designs, while predicting the sorption capacity for an individual design.

Figure 1. Langmuir Isotherms for Individual Wetland Substrates

The amount of lead and zinc in the amorphous iron oxy/hydroxide (called iron (hydr)oxide), precipitated form was substantial. Initial testing revealed that the iron (hydr)oxide formed in the influent area of the wetlands, which is before contact with the wetland substrate, was up to 1.4 percent lead by mass and 2.5 percent zinc by mass. To further investigate this removal mechanism, wetland cells of only gravel (allowing for iron (hydr)oxide precipitation) reached a removal efficiency of 38 percent for zinc and 46 percent for lead. Coprecipitation with iron (hydr)oxide was a substantial removal mechanism. The coprecipitation of lead and zinc with the iron precipitates is a water-chemistry specific removal mechanism. Coprecipitation can be investigated for a specific wetland design and estimation can be conducted based on the amount of iron (or manganese) present in the water and the pH of the water in question. Engineering design can optimize this process by ensuring adequate aeration of water containing ferrous iron and allowing precipitation, such as including a cascade aeration inlet and a brief sedimentation period. Removal of heavy metals as iron (hydr)oxide coprecipitate cannot be expected in the anaerobic (i.e., sulfate reducing) portions of the wetlands.

We accomplished the detection of lead and zinc sulfides in the wetland media using scanning electron microscopy (SEM). Quantification is difficult because the sulfide form is not delineated by the Tessier sequential extraction test, and SEM is not quantitative. The presence of sulfate-reducing bacteria and hydrogen sulfide in the wetlands is a strong indication that metal sulfides will be formed, given their low Ksp values (such as 2 x 10-15 for PbS). To better demonstrate the removal capacity via sulfide precipitation, 3 grams (dry weight) of wetland media with 0.2 mg sulfate added was isolated in dialysis tubing and placed in a 50 mL vial containing 3 ppm lead in aqueous solution. We selected the dialysis tubing to allow HS- cross-membrane transport, but not lead. Lead was rapidly removed from the solution in roughly 48 hours, and five repeat aqueous lead additions (3 ppm) were removed within 60 to 80 hours after addition. Faster reaction rates are possible and expected; this reaction appeared to be diffusion-limited, as HS- must diffuse through the dialysis membrane. SEM analysis of the precipitate formed in the experiment revealed that the precipitate formed was PbS. The removal capacity of the media as added with the sulfate was determined to be 32.7 µg Pb/g dry media. Removal of metals sulfides is a desirable form, given the low solubility. The potential for forming sulfides is directly related to the sulfate levels in the water, which can be used to predict sulfide generation potential. In the engineering design of wetlands, removal as sulfides can be optimized by the addition of sulfate-rich substrates if sulfate is low in the influent water and by ensuring an adequate electron donor supply, the organic substrates.

An ancillary finding of the speciation experiments is the inaccuracies of the Tessier sequential extraction test. This test often is cited as a method of testing metals speciation in sediments. In repeated testing of wetland materials, little to no lead and zinc are detected in the nitric acid digestion step, which is referred to as the "sulfide fraction," whereas multiple analyses presented here have shown the undeniable presence of substantial lead and zinc sulfides.

The failure mode of wetlands remains under investigation. Although the removal mechanisms have been determined, the relative importance and reliability of each is not fully defined. For example, sorption is clearly a removal mechanism that plays a significant role; however, desorption released only 5 to 13 percent of the lead and zinc that had been sequestered. The level of desorption indicates that although adsorption is significant, it may be only a short-term form of removal for the metals that are subsequently reacted with to form sulfides. To operate the wetlands at higher loading rates, we have constructed vertical flow wetlands to characterize critically high flow rates (we ran our horizontal flow wetlands at maximal flow rates-surface flow occurs at too high a rate of flow). We will run these wetlands at hydraulic retention times, ranging from 3,000 minutes to a potential low of 10 minutes, corresponding to potential surface loading rates up to 26.5 L/m2/min and volumetric loading rates up to 53 L/m3/min. These high loading rates and short retention times should exceed the treatment capacity of the wetlands, and at these critical rates, the relative importance of the removal mechanisms can be determined. Once failure modes are understood, the critical design parameters can be determined and incorporated into engineering design.

Bioavailability is the main focus of the experiments currently being conducted. Three plant species that are dominant in the local wetland environment are being planted in controlled laboratory experiments. A survey of wetland areas in the lead belt areas of Missouri showed that cattails (Typha latifolia), bulrush (Scirpus validus), and duck potato (Sagitarria latifolia) are dominant. We ordered the plants and planted them in the new wetlands. We will operate the wetlands for 18 months. At the end of this period, plants will be harvested for analysis. One wetland will receive acid mine drainage (AMD) with a target pH of 3.5 and a total acidity of 130 mg/L. The metal constituents mimic AMD, as reported in the literature (see Table 1). One wetland will receive the neutral mine drainage that we have used for 5 years in the existing wetlands. The third wetland will receive tap water to serve as a control. In addition to the wetland media plantings, a hydroponic side channel was constructed to allow the study of metals and plant roots and the metals containing water, without interference from the wetland media. Dr. Marshall Porterfield in the Biological Sciences Department at the University of Missouri at Rolla (UMR) is developing self-referencing ion specific electrodes that allow for flux measurements over very short distances (i.e., sub-mm distances). These electrodes will allow the determination of metals fluxes to the roots if the roots serve as a sink or deposition site for the metals. Dr. Porterfield's work should contribute to his research project regarding lead impacts on plant roots and add benefit to our molecular-scale understanding of bioavailability issues. At 6-month intervals, plant tissues will be collected, ashed, digested, and analyzed via atomic absorption (AA) spectrophotometry, or inductively coupled plasma-mass spectrometry (ICP MS). Wetlands planted with cattails and bulrush from previous operations have been examined; the stems and shoots had negligible lead or zinc. An additional analysis is planned for the root tissues. Iron plaques that form on roots are hypothesized as a deposition site for lead and zinc. The plaque from the roots will be dissolved using a dithionite-citrate-bicarbonate (DCB) technique and analyzed. Overall, literature searches have revealed a wide variety of experimental results that allude to minimal metals uptake or bioavailability with regard to wetland plants.

Table 1. Acid Mine Drainage

Element mg/L M
SO4 1000 0.010
Fe 40 7.2 x 10-4
Cu 30 4.7 x 10-4
Mn 15 2.7 x 10-4
Ni 0.5 8.47 x 10-6
Zn 0.5 7.69 x 10-6
Cd 0.5 4.5 x 10-6
Pb 0.05 2.42 x 10-7

The final objective is related to operational disturbances and the potential impacts that may occur. Within the next 6 months, we will thoroughly sample a wetland that has been operating for 4 years, and then hand mix it to simulate a substantial physical disturbance, such as a vehicle being driven through the wetland. We will determine the effect on effluent metals concentrations. The same wetland will be drained, air dried, and flushed with water to determine the potential metals release and to investigate changes in metals chemistry under aerobic conditions. This experiment will have great meaning with regards to the theory that the sulfides sequestered in the wetlands have the potential to be re-released. Operational guidelines will result.

An overall goal of this research project is to disseminate all acquired information, as well as existing data that have not been widely utilized or distributed. Dr. Burken is now a member of the Interstate Technology and Regulatory Council (ITRC) Constructed Wetlands Group and is working on the current technical assistance document, including bioavailability issues and construction guidelines. The document began well before Dr. Burken's involvement, but his participation will have a great impact on the overall document, and the results of this research project shall have considerable bearing on the technical acceptance and understanding of constructed wetlands for the removal of heavy metals. In addition to Midwest Hazardous Substances Research Center Technology Transfer activities, the ITRC will be utilized to disseminate results and findings.

Future Activities:

Doe Run Mining Company currently is in the design stages for a new constructed wetland at their facilities for treating low-level lead and zinc. Barr Engineering will design the wetland. Drs. Fitch and Burken have been contacted to review the plans and have been invited to include the wetland in the current study. Planning is underway to establish a sampling plan, for both media samples and plants. This wetland will provide a unique opportunity to include an analytical sampling plan in the design, construction, and operations that will mimic the laboratory-scale wetlands that are part of this study. Sampling ports already are included in the design, and preliminary sampling of the wetland material will occur before the wetland becomes operational. The analysis of the wetland will address all parameters covered in the objectives of this research project. We will analyze speciation and sequestration. Plant tissues will be tested for bioavailability information. Any disturbances at the sites will be recorded and subsequent sampling will provide insight to the impacts of such disturbances.

Sampling at the Cominco wetlands in Trail, British Columbia, Canada is anticipated this spring. In situ samples will be gathered from preplaced media samples put into the wetland this summer. These samples consist of "virgin" media placed in porous fiberglass mesh, and inserted in flow-through samplers placed into the wetland media. Analysis of the media will provide enumeration of sulfate reducing bacteria, samples for SEM analysis, and samples for total and sequential extraction analysis. This wetland also treats a high arsenic load and will provide additional information regarding treatment of metals other than zinc and lead. Sampling at Steifel labs also is anticipated; however, the exact protocols have not yet been determined.

Journal Articles on this Report : 2 Displayed | Download in RIS Format

Other subproject views: All 9 publications 2 publications in selected types All 2 journal articles
Other center views: All 108 publications 22 publications in selected types All 14 journal articles
Type Citation Sub Project Document Sources
Journal Article Song Y, Fitch M, Burken J, Nass L, Chilukiri S, Gale N, Ross C. Lead and zinc removal by laboratory-scale constructed wetlands. Water Environment Research 2001;73(1):37-44. R828770 (2002)
R828770 (2003)
R828770 (Final)
R828770C005 (2002)
R828770C005 (2003)
  • Abstract from PubMed
  • Abstract: Water Environment Federation-Abstract
  • Journal Article Song Y, Fitch M, Burken J, Ross C. Adsorption of lead and zinc in the substrates of constructed wetlands. Water Environment Research. 2001;73(1):37-44. R828770 (2002)
    R828770C005 (2002)
    R828770C005 (2003)
    not available

    Supplemental Keywords:

    lead mine drainage, constructed wetlands, metal geochemistry, acid mine drainage, alternative technology, aqueous waste, aqueous waste stream, aqueous waste streams, constructed wetlands, effluents, hazardous waste treatment, heavy metal contamination, heavy metals, industrial wastewater, lead, lead compounds, metal removal, metal wastes, metals, mine drainage, mining wastes, wastewater remediation., RFA, Industry Sectors, Scientific Discipline, Toxics, Waste, Water, Hydrology, Remediation, Wastewater, Mining - NAIC 21, Hazardous Waste, Engineering, Hazardous, 33/50, Engineering, Chemistry, & Physics, Environmental Engineering, hazardous waste treatment, wastewater treatment, industrial wastewater, wastewater remediation, lead & lead compounds, acid mine drainage, lead, alternative technology, aqueous waste, constructed wetlands, metals removal, effluents, mine drainage, metal wastes, heavy metal contamination, heavy metals, metal removal, aqueous waste stream, mining wastes, aqueous waste streams

    Relevant Websites:

    http://bridge.ecn.purdue.edu/~mhsrc/ Exit

    Progress and Final Reports:

    Original Abstract
  • 2003 Progress Report
  • Final

  • Main Center Abstract and Reports:

    R828770    HSRC (2001) - Midwest Hazardous Substance Research Center

    Subprojects under this Center: (EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).
    R828770C001 Technical Outreach Services for Communities
    R828770C002 Technical Outreach Services for Native American Communities
    R828770C003 Sustainable Remediation
    R828770C004 Incorporating Natural Attenuation Into Design and Management Strategies For Contaminated Sites
    R828770C005 Metals Removal by Constructed Wetlands
    R828770C006 Adaptation of Subsurface Microbial Biofilm Communities in Response to Chemical Stressors
    R828770C007 Dewatering, Remediation, and Evaluation of Dredged Sediments
    R828770C008 Interaction of Various Plant Species with Microbial PCB-Degraders in Contaminated Soils
    R828770C009 Microbial Indicators of Bioremediation Potential and Success