Best Western Winrock Inn
Albuquerque, New Mexico
October 26-27, 2004

Welcome and Introductions
Bob Puls, U.S. Environmental Protection Agency (EPA)

Bob Puls opened by welcoming attendees to the meeting and thanking EnviroMetal Technologies, Inc., and ARS Technologies, Inc., for providing refreshments and lunch for the meeting attendees. (A list of attendees is included as Attachment A (PDF, 3 pp., 72 KB).) For years, the Permeable Reactive Barriers (PRB) Action Team has focused on zero-valent iron (ZVI) PRBs that are designed to remediate the dissolved phases of chlorinated-solvent plumes. Over the last year, however, the Action Team officially expanded its scope to encompass more innovative applications, such as using iron (in some cases nanoscale iron) to address contaminated source zones, using alternative reactive media in PRB walls, and using PRBs to treat a broader suite of contaminants. Puls said that many of the presentations that would be delivered during this meeting would focus on these newer, more innovative applications.

Session Chair: Tom Krug, GeoSyntec Consultants

Treatability of CFC 11 and CFC 113 with ZVI
John Vidumsky, DuPont

John Vidumsky talked about column studies that have been performed to determine whether ZVI has a remedial effect on chlorofluorocarbon (CFC) 11 and CFC 113. Both of these chemicals were manufactured by DuPont at one time; they have since surfaced as environmental contaminants. Vidumsky said that DuPont would like to use a ZVI PRB to address a carbon tetrachloride plume that is commingled with CFC 11 and CFC 113. Before going forth, however, DuPont must determine whether ZVI is even capable of defluorinating CFC 11 or CFC 113, and if it is, whether it creates long-lived (and perhaps environmentally harmful) daughter products.

Vidumsky provided an overview of the column study design. Three 24-inch stainless steel columns were used: one column served as a control and the other two columns (one with influent CFC11 concentrations of 100 parts per million, or ppm, and the other with influent CFC 113 concentrations of 20 ppm) were both treated with Connelly-GPM iron. About 60 pore volumes ran through the columns before any samples were collected. Initial flow rates were about 0.5 milliliters per minute (ml/min), but the rate was lowered to 0.1 ml/min later in the experiment to increase residence times. Vidumsky summarized the fate of the contaminants as follows:

Vidumsky said that these findings suggest that ZVI degrades CFC 11 and CFC 113 through single electron reactions and double electron reactions. When the former occurs, the contaminant is dechlorinated but not defluorinated and stable long-lived daughter products form. With the double electron reactions, however, complete dechlorination and defluorination of CFCs occurs and acetate, formate, fluoride, and chloride emerge as end products.

Vidumsky said that work is continuing on this project; in fact microcosm studies are currently underway to determine whether the abovementioned daughter products are influenced by biodegradative pathways. In addition, researchers are striving to determine whether using a different type of iron will induce more of the contaminant to go down the complete degradation pathway rather than the routes that yield daughter products.

Vidumsky fielded questions from the audience. Paul Tratnyek asked whether any studies have been performed to examine the effect ZVI has on CFC 11 and CFC 113 when they are commingled. Vidumsky said that such tests have been performed and that Michelle Thompson would describe the results in her presentation (see below). Richard Steimle asked Vidumsky what types of technologies are currently being used to address CFCs. Vidumsky responded that pump-and-treat is commonly chosen as an option and that some groups have started examining bioremediation as an in situ solution.

Degradation of Explosives at Cornhusker Army Ammunition Plant Using a ZVI PRB
Tom Krug, GeoSyntec Consultants

Tom Krug, whose presentation is included as Attachment B (PDF, 43 pp., 771 KB), talked about research that is being performed to determine whether ZVI barriers have the ability to remediate ground water contaminated with hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) and trinitrotoluene (TNT). These contaminants, which pose a problem at many Department of Defense (DoD) sites, are often addressed using ex situ pump-and-treat technology—an approach that is costly over the long term because it requires intensive operation and maintenance (O&M) activities. Therefore, interest has been expressed in identifying an in situ remedial strategy to address these contaminants. Laboratory studies suggest that ZVI degrades RDX and TNT rapidly. Based on this finding, the Environmental Security Technology Certification Program (ESTCP) provided funding to install a pilot-scale ZVI PRB at the Cornhusker Army Ammunition Plant (CHAAP), a site that has a TNT and RDX plume emanating from its manufacturing area. GeoSyntec Consultants and Oregon Health and Science University have joined forces on this project, with the former taking the lead on PRB design and installation and the latter leading the effort to assess the wall's performance.

Before discussing the pilot PRB in detail, Krug talked briefly about the degradation pathways that RDX and TNT undergo in the presence of ZVI. Although both disappear rapidly, it is unclear what degradation products are produced and how long they persist. The following is known, however: the degradation products are more amenable to biodegradation than the parent compounds.

The 50-foot-long PRB at CHAAP was installed in November 2003 using trenching equipment and a biopolymer slurry that held the walls open long enough for the installation team to emplace a sand/iron mixture into the trench. Krug said that this installation technique—once regarded as novel—is now a well-accepted PRB emplacement technique. Installation was completed in one day; contractors worked into the night because they did not want to delay the backfilling process and risk encountering problems with trench stability. Monitoring wells and access tubes were installed during the PRB construction process. (The latter, which are intended to guide field workers to the front face of the PRB, will be used to assist with future soil coring efforts.) Noting that the PRB does not extend all the way through the saturated zone, Krug reminded attendees that the PRB is a research project and that limited funding prevented the research team from extending the wall to deeper depths.

Krug said that many ground-water samples were collected between November 2003 and August 2004, and that the research team is still reviewing the results reported for TNT, sulfate, conductance, alkalinity, ORP, and pH. (RDX, which was present in very low concentrations, was not included in the analytical suite because the analytical methods available do not have low enough detection limits.) Krug also noted that a bromide tracer test was performed at the site to analyze ground-water flow patterns. Attachment B provides details about the data collected to date. In summary, Krug said, the results suggest the following: (1) the PRB is effectively degrading TNT but it is unclear what is happening to the degradation products, and (2) the vast majority of the ground water is flowing through the PRB although some evidence suggests that a little ground water is flowing under it. In closing, Krug said that ESTCP is also supporting a project that involves using a biomulch barrier to treat TNT and RDX. He thought it would be interesting to compare the results obtained with the biomulch barrier against those obtained with the ZVI PRB.

One attendee asked whether the degradation products are more toxic than the parent compounds. Krug said that he did not have any information to share about the toxicity of the daughter products, but that evidence does suggest that these products biodegrade more readily than their parent compounds do. In response to another question, Krug said that the sulfate data suggest that interesting biological reactions are occurring at the site. Fielding other questions from the audience, Krug noted the following: (1) the PRB was designed to handle a ground-water flow rate of 1 foot per day, (2) no monitoring wells were installed at the wings of the wall to determine whether ground water is going around the wall, and (3) the site is characterized by a clean beach sand and has no clay lenses acting as contaminant repositories.

In Situ Remediation with EHC™
David Hill, Adventus Remediation Technologies, Inc.

David Hill, whose presentation is included as Attachment C (PDF, 22 pp., 1.48 MB), provided information about EHC™, a solid material consisting of microscale ZVI and a controlled-release fibrous organic carbon, as well as major, minor, and micro-nutrients. Hill opened by explaining that EHC™ degrades contaminants via two pathways—ZVI-induced reduction processes and biological processes—and that the product has been shown (at least in the laboratory) to be highly effective in treating a broad range of contaminants, including carbon tetrachloride, chloroform, methylene chloride, chloroethanes, chloroethenes, perchlorate, pentachlorophenol, chlorinated pesticides, and organic explosive compounds. Laboratory studies and some field work suggest that the controlled-release carbon supports bioremedial processes for at least 3 to 5 years. Hill also noted another favorable characteristic associated with EHC™: it does not have to be in direct contact with contaminants to promote remedial effects.

Hill said that EHC™ will be deployed at an active grain storage facility in the near future. He provided some background information on the site. Carbon tetrachloride has been detected at concentrations as high as 4,000 parts per billion (ppb) in the source zone, and has impacted a creek located 2,000 feet downgradient of the site. In addition, some evidence suggests that carbon tetrachloride (as well as chloroform, one of its daughter products) has impacted nearby irrigation wells. In an effort to identify possible remedial solutions, site managers have agreed to pilot the EHC™ technology as a source zone treatment strategy in 2005.

The pilot project will involve injecting EHC™ (at an application rate of 1 percent) into a 100-foot by 50-foot by 10-foot section of the source zone. A final decision has not yet been made regarding what type of installation method will be used; hydraulic fracturing, direct injection, pneumatic injections, and high-pressure water jetting are all being considered. At this point, Hill said, the project team is leaning toward hydraulic fracturing, an installation method that worked well at another EHC™ pilot site.

Hill provided background information on how the project team arrived at the decision to use EHC™ at this site. He noted that column studies were performed on site soils, and that some of the columns were treated with EHC™ and others were treated with ZVI. The results revealed that, while EHC™ did not exhibit a significant advantage over ZVI when it came to degrading carbon tetrachloride, it was better at degrading the contaminant's breakdown products (e.g., chloromethane and methylene chloride). Hill also noted that different EHC™ configurations were tested in the laboratory; the results indicated that it would be better to install the EHC™ in a dispersed fashion rather than in a traditional trench-style PRB configuration. Detailed information about the laboratory tests is included in Attachment C.

Jackie Quinn asked whether the project team has checked whether guar, a product used to assist with hydraulic fracturing, could impact the efficacy of EHC™. Hill said that this issue has not been evaluated yet. Another participant asked whether microscopic studies have been performed to determine whether trauma introduced during the installation process could cause the iron and the carbon source to separate from one another. Hill said that such analyses have not been performed.

Using an Apatite II PRB To Remediate Ground Water Contaminated With Zinc, Lead, and Cadmium
James Conca, New Mexico University

James Conca, whose presentation is included as Attachment D (PDF, 35 pp., 1.50 MB), talked about using PRBs filled with processed fish bones to remediate metal-contaminated ground water. Before discussing field applications, he provided background information on phosphate-induced metal stabilization processes. He said that fish bones—also referred to as Apatite II—are composed of a reactive form of the phosphate mineral group, called apatite, that can immobilize metals in metal-contaminated water through three different pathways: precipitation, co-precipitation, and sorption. Once a metal enters the apatite phase, Conca said, the contaminant becomes very insoluble. For example, when lead enters the apatite phase, it forms lead pyromorphite—a material that has the lowest solubility of any known solid. Conca noted that several types of apatite are available, including fish bones, phosphorite rock, cow bones, and cannery waste. After analyzing the different forms, Conca's research team decided that fish bones are the best at remediating metals. Attachment D provides a detailed explanation of why this is the case. In summary, Conca noted that Apatite II can sequester more than 20 percent of its weight in metals and that it remediates metals via four general non-mutually-exclusive processes, one of which is biological stimulation. He also noted that Apatite II costs about $350 per ton.

Conca said that fish bones have been proposed for use as a remedial agent in three types of field applications: treatment tanks, direct soil mixing, and PRBs. Concentrating on the latter, he said that PRBs have been installed at the Success Mine and Mill Site and the Nevada Stewart Mine Site, both of which are in Idaho. He focused his discussion on the Success Mine and Mill Site, which operated between 1886 and 1956 and dumped about 500,000 tons of tailings adjacent to Ninemile Creek. Site soils are contaminated with lead, zinc, and cadmium; concentrations are in the range of 1,000 to 4,000 ppm. Ground water and surface seeps are also impacted by contamination, with concentrations ranging from 8 to 1,250 ppb for cadmium, 70 to 1,440 ppb for lead, and 4,850 to 177,000 ppb for zinc. To address the site, Conca said, the Idaho Department of Environmental Quality decided to do two things: (1) install a 425-foot grouted containment wall down to the bedrock layer along the edge of Ninemile Creek and (2) establish a PRB (15.4 meters long, 4.2 meters high, and 4.6 meters wide) between the tailings pile and Ninemile Creek. The PRB was completed in January 2001. It is filled with 100 tons of fish bones and is segregated into five chambers. Water flows through the five chambers, is discharged to a rock apron, and is then discharged into Ninemile Creek. Conca said that 3.5 years of monitoring data have been collected and that the PRB appears to be performing better than anticipated, with effluent concentrations meeting drinking water criteria with respect to metals. To make his point, Conca noted the following about the water that is exiting the PRB: lead and cadmium concentrations are below detection limits (5 ppb and 2 ppb, respectively), zinc concentrations are below background levels (i.e., less than 100 ppb), and the pH is neutral. (Attachment D provides additional data on influent and effluent concentrations.) Since the PRB has been in place, about 150 pounds of lead, 100 pounds of cadmium, and 10,000 pounds of zinc have been sequestered. Investigations have been performed to examine the longevity of the Apatite II. The results suggest that about 60 percent of the barrier remains reactive. Conca said that he thinks zinc will be the first contaminant to break through since it is present at the highest concentrations. Once breakthrough occurs, regulators will have to determine what additional steps should be taken to address this site. One option would be to install a new Apatite II PRB just upgradient of the first PRB so that the old PRB is only expected to serve as a polishing step.

Before concluding, Conca said a few words about the Apatite II PRB installed at the Nevada Stewart Mine Site. He said that toxicity studies have been performed to assess how the PRB has impacted ground-water quality. The results (summarized in Attachment D) indicate that the PRB has induced environmentally beneficial effects.

Chris Clausen asked whether Apatite II can be synthesized. Conca said that it can but doing so would be cost-prohibitive. Commenting on data that Conca presented for the Success Mine and Mill Site, one attendee noted that ammonia was detected in the PRB effluent at concentrations exceeding regulatory standards. Conca said that the ammonia disappeared as the effluent passed through the rock apron. As a result, ammonia was not discharged to the Ninemile Creek at high concentrations. Responding to another question from the audience, Conca noted that the residence time for water in the Success Mine PRB was about 10 hours.

Electrically Induced Redox Barriers for Treatment of Ground Water
Tom Sale, Colorado State University

Tom Sale provided information about e-barriers, which remediate contaminated ground water as the water passes through a series of electrodes. (His presentation is included as Attachment E (PDF, 26 pp., 853 KB).) Sale explained the theory behind e-barriers, presented some laboratory and field data, and discussed future plans. He opened by saying that e-barriers exert remedial effects by inducing localized shifts in redox conditions, a conclusion that researchers came to when performing column studies with funds provided by the Solvents-in-Groundwater Research Consortium. Describing these column studies, Sale said that two titanium electrodes were set up in a column, low voltages were applied to the electrodes, and contaminated water was allowed to pass through the electrodes. Strong oxidizing conditions were apparent at the positive electrode and strong reducing conditions were apparent at the downstream negative electrode. These dramatic shifts in thermodynamic conditions induced contamination degradation. In fact, in the laboratory, e-barriers were shown to achieve removal rates of 90 to 99 percent for the following contaminants: trichloroethane (TCA), trichloroethene (TCE), RDX, and TNT. Although e-barriers are highly effective at treating TCE, Sale said, other technologies (some of which are more cost-effective) have already been developed to treat this contaminant. Thus, in the future, the e-barrier development team plans to shift its focus to other contaminants, such as energetics, TCA, and chemical mixtures.

Sale said that e-barriers have been piloted in the field at two sites: CFB Borden and Warren Air Force Base. At the former, a 4-foot-by-4-foot e-barrier prototype was installed downgradient of a PCE/TCE source zone. The e-barrier consisted of three electrodes that were surrounded by an HDPE Geonet mesh and encased within an HDPE geotextile fabric. Although much was learned at CFB Borden, the project team suffered two disappointments: they were unable to install the barrier as deeply as they had hoped, and they did not witness dramatic reductions in downgradient contaminant concentrations. ESTCP/SERDP provided funds for a larger demonstration project (involving a 20-foot-by-20-foot e-barrier prototype) at the Warren Air Force Base. Sale said that this e-barrier, which has been operating for 18 months, is being used to treat TCE. Data suggest that the titanium electrodes are holding up well and that 90 to 95 percent of the TCE is being removed as ground water flows through the e-barrier. Elevated TCE concentrations persist downstream of the barrier, however, a finding that leads Sale to believe that some ground water is bypassing the barrier. Sale provided cost data from the two field studies: it cost about $200 to $400 per square foot to construct the e-barriers, and O&M costs have been approximately $10 per square foot. Sale hopes, though, that if e-barriers were deployed full scale, economies of scale would reduce the unit cost by 25 to 50 percent. He also thinks the e-barrier development team could drive costs down by adopting a simpler design, switching to photovoltaics to supply power, using conventional HDPE barrier walls for framing, and improving upon remote instrumentation and monitoring capabilities. The team hopes to implement these improvements and test them in the field at the Pueblo Chemical Depot, a former ammunition and material storage center that is contaminated with RDX, 2,4-dinitrotoluene (2,4-DNT), and nitrate. The proposal for this project is currently being reviewed by ESTCP and SERDP. If it is approved, the project team will also explore the possibility of adding a fourth electrode to the e-barrier in an effort to improve upon contaminant removal rates.

In response to questions from the audience, Sale said that the titanium mesh used in the e-barrier electrodes is expensive (about $30 per square foot) and that ESTCP is providing funds to test alternate materials. Rich Landis suggested testing a product that consists of a less dense titanium mesh to determine whether it could be used as a substitute. Sale thanked him for the suggestion, noting that the material Landis referred to is indeed much more affordable (about $3 per square foot). Clausen asked whether e-barrier-induced reactions are restricted to the area immediately surrounding the electrodes. Sale said that this was indeed the case at the Warren Air Force Base site, but that there was some evidence that e-barriers indirectly spurred downgradient bioremedial processes at the CFB Borden site. Quinn said that Lynntech Inc., a company that has performed some work at a NASA site, has evaluated the effect that electrokinetics have on biological reactions. She thought that Lynntech's data would be of interest to Sale and she agreed to forward him a copy of these data.

In Situ Removal of Heavy Metal Contaminants Using Emulsified Nanoscale or Microscale Metal Particles
Kristen Milum, University of Central Florida

Kristen Milum, whose presentation is included as Attachment F (PDF, 23 pp., 1.17 MB), described efforts that are underway to examine whether heavy metal contaminants can be remediated using a technology that combines emulsified liquid membranes with zero-valent metals, such as iron or magnesium. Her ultimate goal is to determine whether this technology can be used to remove heavy metals from contaminated sediments. Such an application would involve injecting metal-filled emulsion droplets into site sediments, allowing the droplets to sit for a couple of weeks, and then pumping the droplets back out of the formation so that any heavy metals trapped inside them could be recovered and potentially reused. Milum said that this type of in situ treatment strategy might be less costly and more environmentally friendly than some of the conventional remedial strategies (such as ex situ treatment of dredged materials) that are usually used to address sediments.

Milum said that emulsified zero-valent iron (EZVI) was field-tested at Cape Canaveral's Launch Complex 34 (LC-34) site in an effort to remediate chlorinated compounds. Encouraging results were observed during the field test, with reduction rates of 86 percent being recorded for TCE. Given this success rate, much interest has been expressed in determining whether emulsified zero-valent metals have the potential to remediate other contaminants, such as heavy metals. Toward this end, the University of Central Florida is performing a series of laboratory studies to examine what happens to heavy metal contaminants when treated with emulsified zero-valent metals. The University is testing emulsions filled with nanoscale iron, emulsions filled with microscale iron, and emulsions filled with magnesium. The goal is to determine whether metal-filled emulsions remove heavy metals from solution (and/or soils) and whether heavy metals are transported to the interiors of the emulsion droplets or simply trapped in the oil phase. If the latter, Milum said, this technology could have a deleterious effect on the environment because it could end up making the contaminants more mobile.

A series of vial studies and partitioning studies were performed to explore the abovementioned research questions. The results (see Attachment F for details) showed that the metal-filled emulsions do indeed extract metal ions from a variety of solutions and from soils. The metal ions are transported to the interiors of the emulsion droplets, where they are treated by zero-valent metals. More laboratory studies will be conducted in the near future to determine how the technology works in more complex environments and to determine exactly what happens to metal ions once they enter the droplets. If all goes well, Milum said, the research team will initiate a small-scale field test in the near future.

Breakdown and Fate of Biostat in Biopolymer Trenching Fluids
Lloyd Marsden, Rantec Corporation

Lloyd Marsden talked about a research project that is being performed to evaluate the persistence and fate of dazomet, a preservative that is used to inhibit biopolymer degradation. He opened by saying that biopolymers are an important tool in the PRB construction tool box: when remediators install a PRB, they pump a biopolymer slurry into the trench to keep it open, then pour reactive materials into the trench. A portion of the biopolymer is displaced in the process, and the remediators break the remainder down by introducing a liquid enzyme breaker. In order for the biopolymer to serve its intended role, a preservative must be added to prevent the biopolymer from breaking down before construction activities are completed. When choosing a preservative, one must consider three important factors: (1) the preservative must effectively reinforce the soil-stabilizing properties of the biopolymer, (2) the preservative must be economical to use, and (3) the preservative must be labeled and registered by EPA as a pesticide that is acceptable to use for construction purposes. Marsden said that dazomet fits these three criteria and is therefore commonly used as a preservative to assist with liquid shoring construction activities. Some concern has been expressed, however, about injecting this chemical into the subsurface because it is a skin and eye irritant, it exhibits some oral and aquatic toxicity, and it is listed as a reportable substance under SARA 313. (When used as part of a PRB construction project, however, dazomet would probably not be present in high enough quantities to trigger reporting requirements.) Given these concerns, some regulators refuse to allow dazomet to be used as a preservative in the field.

To determine whether such concern is justified, Marsden said, a research project has been initiated to learn more about how dazomet behaves when used in trenching applications. Although evidence suggests that the preservative degrades rapidly in water, more research is needed to determine what daughter products are produced, how long they persist, and whether dazomet and its daughter products are more persistent in the deep, cold, anaerobic conditions that characterize a trench environment. Toward this end, ETI and Buckman Laboratories are performing column studies to determine what happens to dazomet during the PRB construction process. The experiments involve placing a standard biopolymer (consisting of guar gum, soda ash, and dazomet) in a column, allowing it to age for 4 hours before adding site soil, waiting an additional 2 days before adding ZVI, introducing a liquid enzyme breaker, closing the column, and then letting it sit at room temperature. Marsden said that dazomet and its daughter products (e.g., formaldehyde, monomethyldithiocarbonate ion, monomethyl amine, methyl isothiocyanate) are being monitored to determine how quickly they degrade to inert substances, such as carbon dioxide, ammonia, and water. If time allows, the column studies might be repeated to determine whether degradation rates are impacted by temperature and ground-water flow conditions. Marsden said that the results of the column studies will be available in January 2005.

John Vogan asked whether efforts have been made to identify a preservative with fewer toxicity-related concerns. Marsden said that alternatives have been sought, but that none have met all three of the abovementioned criteria for a preservative. For example, while less-toxic preservatives have been identified that are equally as effective as dazomet, they are not viable alternatives because they are not labeled and registered by EPA for use in liquid shoring construction applications. Manny Saenz, who works with the Interstate Technology and Regulatory Council (ITRC), said that ITRC has performed some research on alternatives and he recommended continuing efforts to identify potential alternatives. In response to a question posed by Tratnyek, Marsden said that the project team is not currently examining the impact that light could have on degradation rates. Tratnyek strongly advised evaluating this parameter. Vidumsky asked Marsden whether he thought biological or physical/chemical processes are responsible for breaking down dazomet and its daughter products. Marsden believed that the latter are responsible, but admitted that he could not say for sure.

Session Chair: John Vidumsky, DuPont

Materials Science Perspectives of Injectable Zero-Valent Metals and Alloys
Clint Bickmore, OnMaterials LLC

Clint Bickmore opened by noting that interest has steadily grown over the years in using iron as an in situ remedial agent. In response, iron manufacturers are focusing efforts on developing injectable irons, products that are colloidally stable and resist flocculating when pumped into the subsurface. When designing a material, Bickmore said, the goal is to create an iron product that is highly reactive, highly mobile, and cost-effective. Striking a balance between these three goals can be challenging. Bickmore focused his presentation, which is included as Attachment G (PDF, 29 pp., 1.90 MB), on issues that OnMaterials LLC considers when designing an injectable iron product.

Bickmore said that injectable iron products can be synthesized using a "breakdown" or a "buildup" process. The former involves starting with something big and breaking it into smaller particles; the latter involves starting with an atomic- or molecular-scale precursor and building a larger material off it. Challenging cost management issues are associated with both approaches.

The reactivity of an iron product is enhanced whenever a potential difference occurs between neighboring areas. Thus, irons can be made more reactive by introducing any of the following: bimetallic corrosion, compositional variations, microstructural/phase variations, pits and crevices, stray currents, dislocations and strains, or morphological features (e.g., flaking and turning). Once a material has been created, bench-scale and column studies can be performed to assess its overall reactivity. (Bickmore presented information about one of the iron products that OnMaterials has developed. Data from bench-scale studies of that product are presented in Attachment G.) In addition, in order to obtain a fuller understanding of how effective the material might be in a field application, it is important to examine the following characteristics:

Bickmore concluded by saying that the best injectable irons are those consisting of small, low-density, discrete particles that have reactive "neat" surfaces. He said that OnMaterials is pleased with the most recent generation of injectable iron that it has created—a robust product that sells for about $20 per pound.

Nanoscale Iron Particles for Site Remediation: Potentials and Problems
Wei-xian Zhang, Lehigh University

Wei-xian Zhang said that nanoscale iron exhibits great promise as a remedial agent, noting that studies show that it induces rapid degradation of a wide variety of contaminants, including chlorinated solvents, carbon tetrachloride, perchlorate, lindane, and hexavalent chromium (CrVI). He said that he and his colleagues at Lehigh University have been working with nanoscale iron particles for several years, noting that most of the materials they have produced consist of particles in the 50 to 75 nanometer (nm) range. One of Lehigh's nanoscale iron products has already been tested in the field, Zhang said, noting that the results showed that injecting nanoscale iron into the subsurface induced favorable remediation conditions. Zhang said that he and his colleagues plan to continue experimenting with nanoscale iron and that they hope to create a product that exhibits even greater reactivity than existing nanoscale iron products. Toward this end, Zhang's group has initiated efforts to create porous nanoscale iron particles. Interest in such an approach, Zhang said, stems from the fact that the amount of surface area that is available to react with contaminants would be greatly enhanced in a porous particle.

Although nanoscale iron exhibits great promise, Zhang said, scientists and site managers should keep in mind that there could be problems associated with this technology. He said that very little work has been performed in the field at this point; as a result, much remains to be learned regarding the fate and transport of these materials in a field setting. He advised thinking about the following question: Could introducing nanoscale iron particles to the environment pose any risks?

Use of ZVI for Ground-Water Remediation: Three Case Studies
Manny Saenz, Naval Facilities Engineering Command

Manny Saenz, whose presentation is included as Attachment H (PDF, 21 pp., 1.32 MB), said that ZVI is being used to treat contaminated ground water at the following naval sites:

Saenz concluded by saying that the Navy plans to release a report about these three sites in January 2005.

Advantages of Nanoscale Iron for Remediation of Carbon Tetrachloride
Paul Tratnyek, Oregon Health and Science University

Paul Tratnyek opened by saying that a team of surface chemists, microscopists, spectroscopists, and modelers from Oregon Health and Science University, the Pacific Northwest National Laboratory, and the University of Minnesota have joined efforts to examine the reaction specificity of nanoscale iron in solution. Their efforts are being supported by two Department of Energy (DoE) programs: the Environmental Management Science Program (which focuses on practical applications) and the Basic Energy Science Program (which focuses on fundamental research). Tratnyek said that the project's scope is large and that he did not intend to present a complete description of all of the work that has been performed to date. Instead, he limited his discussion to a description of the laboratory work that has been performed to examine the structure, composition, and kinetics of different nanoscale iron sources. Although a variety of iron products were analyzed, Tratnyek focused his discussion on analyses performed on the following two: FeH2 (produced by Toda Americas, Inc.) and FeBH (produced by Wei-xian Zhang of Lehigh University). He presented information about the structure and composition of FeH2 and FeBH, noting that this information will soon be published in Environmental Science and Technology.

Tratnyek said that the research team is trying to characterize the reactivity of FeH2 and FeBH particles in solution-phase conditions. To date, the team has focused most of its efforts on analyzing how well these materials degrade carbon tetrachloride and its daughter products. (A little work has been performed on TNT, but information about this contaminant was not included in Tratnyek's presentation.) Tratnyek said that reactivity is being analyzed through two different modes: electrochemical cells and batch reactor work. He focused on the latter and described the results collected to date. The following represent the highlights:

Based on the positive reactivity results obtained so far, some interest has been expressed in using FeH2 in the field at DoE's Hanford site, which has a very large carbon tetrachloride plume. Tratnyek expressed enthusiasm for this suggestion but reminded attendees that reactivity is not the only factor that determines how well a remedial agent will work at a site. Mobility is also crucial, and little work has been performed to determine how mobile FeH2 is in porous media.

In response to questions posed by audience members, Tratnyek made the following points: (1) the research team is trying to identify the specific reactive sites on nanoscale iron particles, and (2) some members of the research team are trying to identify ways to engineer particles that will be more effective remedial agents.

Switching to a new topic, Tratnyek noted that Oregon Health and Science University has created a CD-ROM that provides an overview of PRB-related chemistry topics. He advised attendees to contact him soon if they would like a free copy.

Remediation of Chlorinated Solvent Source Zones Using ZVI-Clay in Conjunction with Soil Mixing
Tom Sale, Colorado State University

Tom Sale, whose presentation is included as Attachment I (PDF, 33 pp., 509 KB), talked about an innovative source zone treatment strategy that involves using a large-diameter auger to install a ZVI-clay mixture into the subsurface. Subsurface materials are mixed continuously throughout the emplacement process to help remove site heterogeneities. The technology was originally developed by DuPont, but the patent has been donated to Colorado State University. Sale said that he is excited about the technology, which he believes can effectively remediate basal dense nonaqueous-phase liquid (DNAPL) pools as well as contaminants trapped in stagnant areas. He described how each component of the technology works. First he focused on granular ZVI, noting that iron (which acts as an electron donor) dechlorinates chlorinated solvents (which act as electron acceptors) when the two come into contact with each other. Next, Sale talked about clay, which is important for two reasons. First, by reducing the permeability of source zones, it reduces the amount of contaminated ground water that discharges to downstream areas, increases the amount of time contaminated ground water resides in the reactive treatment zone, and reduces the inflow of oxygen and other competing electron acceptors into the treatment zone. Second, the clay facilitates uniform iron delivery during installation processes because it supports iron particles in a high-viscosity suspension. Finally, Sale talked about the importance of the mixing component: it serves to homogenize the source zone, and in so doing promotes uniform reagent and contaminant distribution, disperses DNAPL pools, and overcomes many of the mass transfer issues that so commonly stymie source zone treatment projects. Sale said that he believes that the deep-mixing ZVI-clay technology could find a niche at sites contaminated with chlorinated compounds, characterized by a mixable alluvium, free of complicated overhead and/or buried obstructions, and in need of a technology that delivers quick results.

The deep-mixing ZVI-clay technology has been simulated in the laboratory using column studies. Sale said that a small mixer was used to emplace a ZVI-clay mixture into soils that were spiked with a red-dyed TCE pool. Investigators found that the red TCE pool dispersed into multiple ganglia-shaped blobs, all of which vanished (at least to the visible eye) within 2 days. Subsequent coring analyses revealed that half of the TCE had disappeared within 14 days of treatment. Sale said that he thinks the TCE degraded so quickly because the mixing process facilitated direct contact between contaminants and the iron.

Deep-mixing ZVI-clay technology is being considered as a remedial strategy at a site in South Carolina. Sale presented data from the treatability studies that have been performed for this site. He noted that studies were performed to determine how 1,1,2,2-tetrachloroethane (1,1,2,2-TCA) and carbon tetrachloride respond to varying iron concentrations (1%, 2%, and 5%) and iron sources (i.e., Connelly, Peerless, and GMA). Degradation rates for chloroform and methylene chloride (both of which are degradation products of carbon tetrachloride) were also measured. Detailed results are presented in Attachment I. In summary, the results indicated that 1,1,2,2-TCA and carbon tetrachloride both degraded rapidly regardless of how much iron was used. As for daughter products, the data indicated that chloroform degrades rapidly, but methylene chloride degrades very slowly—an issue that Sale thought could be cause for concern.

Sale also presented information about DuPont's Martinsville plant, a site with a large carbon tetrachloride source area. The deep-mixing ZVI-clay technology was deployed at this site in late 2002 in an effort to address the source zone. (About 2 to 6 pounds of iron were emplaced per cubic foot of soil.) Sale said the remedial technology appears to be effective, noting that sampling results suggested that 99.99 percent of the carbon tetrachloride and 99 percent of the total chlorinated compounds had been removed after 1 year of treatment. (Sale did note, however, that he is slightly concerned that there might be some potential for sampling bias at this site.) A second round of soil samples were recently collected to determine whether additional remedial effects were realized during the second year of treatment. The results from the second soil sampling event are not yet available. Sale said that he is particularly interested in finding out whether the iron is still present in the subsurface at this site.

Sale concluded by listing some of the key questions that require further exploration: How long will the iron last when used in this type of application? How will the amount of iron used and the size of the particles impact longevity? How much iron will be lost to potentially unproductive reactions? What are the rate-limiting steps associated with this technology? Do some of the monitoring techniques introduce sampling bias?

Clausen noted that some people have expressed interest in combining soil mixing with steam stripping and then following up with iron injections as a polishing step. Such an approach, he said, might reduce the amount of iron that must be injected into the subsurface. He asked Sale to comment on this idea. Sale responded that his group has not experimented with steam, but that it has performed column studies to assess the benefit of hot air stripping soils before adding iron. He did not think that the benefits obtained from such an approach were large enough to warrant the extra health and safety risks associated with field-scale applications of hot air stripping. Quinn said that the Air Force Center for Environmental Excellence (AFCEE) has performed side-by-side comparisons of steam only, iron only, steam plus iron, and EZVI; the results will be released in the near future.

EZVI Treatment of Chlorinated Solvent DNAPL Source Areas
Suzanne O'Hara, GeoSyntec Consultants

Suzanne O'Hara presented information about EZVI, a remedial agent that has been developed by the University of Central Florida, patented by NASA, and licensed to GeoSyntec Consultants. EZVI consists of iron-filled emulsions that are specifically designed to degrade DNAPLs in source zones. O'Hara noted that ZVI must be in the presence of water in order to promote reductive dehalogenation. Thus, when ZVI is injected into DNAPL source zones, it can only remediate dissolved-phase contaminants at the edges of the DNAPL. EZVI gets around this limitation and enhances contact between DNAPL and ZVI particles.

EZVI consists of nanoscale (or microscale) iron particles, food-grade surfactant, biodegradable vegetable oil, and water. The iron particles are suspended in water and encapsulated within a liquid oil membrane droplet which is miscible with DNAPLs. Once inside the aqueous interior of the droplet, the DNAPL contacts the ZVI and is quickly degraded. In addition to EZVI's abiotic remedial potential, the droplet's vegetable oil and surfactant serve as long-term electron donors and promote anaerobic degradation.

O'Hara said that EZVI was injected into the subsurface using Pressure Pulse Technology (PPT) during a field demonstration project that was conducted to address TCE contamination at NASA's LC-34 site. Attachment J (PDF, 34 pp., 1.40 MB) presents details about the injection scheme, the monitoring setup, and the data collected at the demonstration site thus far. In summary, soil cores revealed that although the EZVI induced significant contaminant mass removal in the areas it reached, the PPT installation methodology failed to get the EZVI into all portions of the target zone. (More fingering was observed than the research team had hoped for. Also, emulsion was detected in shallower locations than the team desired.) O'Hara said that the ground-water sampling data suggest that EZVI can induce significant TCE reductions and support some long-term biodegradation activities. Based on these findings, the research team concluded that EZVI exhibited great promise but that more work was needed in two areas:

O'Hara also said that the following is planned for the future: (1) deploying EZVI in two pilot areas within a DNAPL source zone using the two most promising EZVI injection technologies and (2) performing activity assays to evaluate the efficacy of EZVI produced using different nanoscale iron products (e.g., Toda America's product and OnMaterials Zloy product).

Use of Foodgrade Microscale Iron for Source Zone Treatment
John Liskowitz, ARS Technologies, Inc.

John Liskowitz said that ARS Technologies, Inc., has been injecting ZVI into source zones since the late 1990s. He stated that ZVI source zone treatment is not a standalone remedial strategy: rather, it is meant to achieve significant mass reduction so that downgradient remediation systems (such as PRBs, natural attenuation, or enhanced bioremediation) can adequately address the dissolved phase of a plume without being overburdened with continuous contaminant loading. Liskowitz said that the following are critically important to the success of ZVI source zone treatment projects:

Liskowitz presented a table summarizing results of ZVI source zone treatment at six DoD sites. In summary, one site reported contaminant reduction rates of about 85 percent, and the other five have experienced reductions in the 90 to 95 percent range. Liskowitz concluded by saying that source zones can be treated using reactive iron powders coupled with effective subsurface emplacement techniques, and the success rates and costs associated with this type of strategy might become even more appealing in the future if researchers succeed in developing more reactive and more mobile ZVI materials. Expanding on the latter, Liskowitz said that he thinks the following research avenues offer promise: developing iron particles that exhibit greater reactivity, using surfactants to create "slippery" particles, and using iron-filled emulsions to enhance contact between contaminants and reactive media.

Responding to a question from the audience, Liskowitz said that it costs about $4 to $8 to inject 1 pound of iron into the subsurface using the liquid atomized injection technique. Clausen asked whether DNAPL might be pushed out of the source zone (rather than treated) during the ZVI injection process. Liskowitz said that monitoring wells have been installed around some of the field sites to confirm that this is not happening. He also noted that there are strategies, such as using a bottom-up and outward-inward pumping approach, that can minimize the risk of mobilizing the DNAPL out of the source zone.

Session Chair: Bob Puls (EPA) and Bob Gillham (University of Waterloo)

Update on the ITRC's PRB Team Activities
Matt Turner, New Jersey Department of Environmental Protection

Matt Turner, whose presentation is included as Attachment K (PDF, 8 pp., 203 KB), said that the fall 2004 ITRC meeting was being held across the street and that RTDF PRB Action Team members were welcome to attend. The ITRC—a state-led organization—operates under a branch of the Environmental Council of the States, enjoys the support of two other state organizations (the Western Governors' Association and the Southern States Energy Board), and attracts participation from state regulatory agencies, industry, academia, federal agencies (i.e., DoD, EPA, and DoE), and public stakeholders. The ITRC has two objectives: to improve state permitting processes and to facilitate the acceptance of new environmental technologies. Toward this end, ITRC develops regulatory and technical guidelines, technology overviews, case studies, classroom training courses, and Internet-based training sessions. Turner encouraged attendees to visit for additional information about ITRC's products and services.

Several teams have formed under the ITRC, each focusing either on a specific technical topic or a problem area. Turner indicated that he is the leader of ITRC's PRB team, a group which formed in 1996 and is scheduled to sunset in 2006. To date, he said, the PRB team has written two guidance documents, partnered with EPA and the RTDF to deliver classroom training sessions at 14 locations across the country, and developed and delivered basic and advanced Internet-based training courses. The team is currently in the process of developing an additional guidance document, which will focus on lessons learned at PRB sites and new applications that are being explored. The text will provide information about ZVI walls as well as barriers that use alternative reactive media. A wide variety of topics will be covered, including PRB design and implementation, site characterization, regulatory issues, stakeholder issues, and concerns related to DNAPL treatment. Turner said that a draft has been completed and distributed for comments. Once the guidance document is completed, an Internet-based training program will be released to summarize the highlights of that document.

Steimle asked whether information about some of the innovative projects presented at this RTDF meeting could be incorporated into the ITRC PRB Team's guidance document. Turner said that innovative applications will be discussed under the section of the document that addresses alternative reactive material. He agreed to ask his team members whether they want to add a detailed case study that discusses an innovative application as an appendix.

Design of a PRB for Chloroethenes and Hexavalent Chromium Treatment in Ground Water at the Former Kelly Air Force Base
Jerry Jin, Science Applications International Corporation

Jerry Jin, whose presentation is included as Attachment L (PDF, 23 pp., 649 KB), provided information about a PRB that is being constructed to address PCE, TCE, and CrVI contamination in Zone 2 of the former Kelly Air Force Base. Recovery wells were installed at this site in the 1990s in an effort to achieve hydraulic control and prevent contaminants from migrating to Leon Creek. Although 13 wells were originally installed, the current (and more optimized) recovery system consists of six wells, all located along Citrus Road. Although the recovery wells appear to be meeting their objectives, Jin said, the Air Force is interested in using a more passive remedial system. Therefore, efforts are currently underway to construct a PRB adjacent to the Citrus Road area. The recovery wells will remain in place, but they will only be brought back into service if the PRB fails to meet its objectives.

Jin then talked about PRB design issues. He concentrated on the methods used to estimate the following three parameters, all of which provided insight on how deep and thick the PRB should be:

Based on the information collected on the abovementioned parameters, the design team concluded that the PRB needed to be 15 feet thick in order to reduce contaminants to maximum contaminant levels (MCLs). After the team made these calculations, the client announced that it had already bought the iron for the PRB and that the quantity purchased was not enough to fill a 15-foot-thick wall. Thus, the team was forced to split the PRB into five sections (sections A through E) and to ration the iron across the five sections based on need. More iron was used in the sections subjected to higher ground-water velocities and/or higher influent contaminant concentrations. Jin said that monitoring wells are being installed around the PRB; well locations have been selected based on which areas have the highest potential for contaminant breakthrough.

Jin said that there are several different DNAPL pools present at the Kelly Air Force Base, noting that he has samples of many of them. He said that he would be willing to send samples to anyone who is interested in studying the materials. He asked those who are interested to contact him by email.

Responding to a question from the audience, Jin noted that CrVI concentrations as high as 1,500 ppb were detected at the site at one point. More recent ground-water sampling data indicate, however, that the concentrations are more in the 400 to 500 ppb range.

Geochemical Characterization of a Multiple PRB, Mortandad Canyon, New Mexico
Patrick Longmire, Los Alamos National Laboratory

Patrick Longmire, whose presentation is included as Attachment M (PDF, 19 pp., 488 KB), presented information about a PRB that is being used to address ground water contaminated with actinides, plutonium, uranium, strontium-90, nitrate, and perchlorate. This PRB (constructed in early 2004) is located in Mortandad Canyon, a site impacted by rainwater runoff and discharge from Los Alamos National Laboratory. The site is characterized by a saturated alluvium and it has perched intermediate ground water zones that vary in depth from 500 to 700 feet bgs. The regional aquifer, which serves as a drinking water source, is located about 1,000 feet below the canyon. Perchlorate (a suspected thyroid inhibitor) is the primary contaminant of concern at this site, Longmire said; it is extremely mobile, and has already been detected in the regional aquifer at concentrations that could be of concern if EPA decides to establish a low MCL for this contaminant. Therefore, one of the PRB's main goals is to enhance reducing conditions because such conditions foster perchlorate degradation.

Water that enters the PRB flows through four cells before exiting. The first cell consists of a gravel basalt scoria that is designed to trap colloids. The second cell consists of apatite in the form of phosphate rock. This cell serves two purposes—not only does it adsorb actinides and strontium-90, but the solid organic matter (SOM) associated with the apatite fosters reductive conditions. The third cell—referred to as the bio cell—consists of pecan shells (which serve as growth media for microbes) and cottonseed meal (which was chosen because it was included as a component of a PRB in Texas that exhibited effective perchlorate remediation). The fourth (and final) cell consists of limestone and provides some additional adsorption capacity and serves to buffer ground-water pH.

Longmire said that a series of monitoring wells have been installed upgradient of the PRB, within the different compartments of the PRB, and downgradient of the PRB. He presented detailed data for several field parameters, including pH, dissolved oxygen, ammonium, nitrate, and sulfate. He also presented data showing how the PRB is impacting perchlorate and strontium-90 concentrations. Detailed data are included in Attachment M. In summary, Longmire made the following points:

Inquiring about the apatite cell, Puls asked Longmire to explain why he used phosphate rock rather than fish bones. Longmire said that the research team tested both materials in the laboratory. Although the fish bones proved to be effective at removing contaminants, the material broke down quickly (developing an oatmeal consistency) when exposed to this particular site's ground water.

Ground-Water Contamination in Chile and the Use of Zeolites as a Reactive Medium
Juan Ramón Candia, Fundación Chile

Juan Ramón Candia provided information about environmental activities being conducted under the auspices of Fundación Chile. This organization—created in 1976 by the Chilean government and the U.S.-based ITT Corporation—is a non-profit organization that promotes and develops technologies for five sectors: agribusiness; marine resources; sustainable forestry, industry, and tourism; environmental and chemical meteorology; and human resources and information technology. With strong market-based philosophies guiding its activities, Fundación Chile promotes the development of new companies, sells and licenses new technologies, identifies markets for new innovations, and supports applied research to assist with the development of innovative products and services. Additional details about Fundación Chile's activities and success stories are provided in Attachment N (PDF, 45 pp., 1.76 MB).

Candia devoted the rest of his presentation on the Environmental and Chemical Meteorology Technology Center—one of five technology centers that exist under Fundación Chile. This Technology Center supports five groups: the Center for Clean Production, Environmental Technologies, Sustainable Development, the Chemical Metrology Center, and Environmental Remediation. Candia is involved with the last of these. He said that the Environmental Remediation group's main goal is to adapt and develop methodologies, tools, and technologies to address contaminated sites. Before doing so, however, the group is trying to obtain a better grasp of the extent of contamination that exists across Chile. Where are the contaminated sites? How can they be identified? How does one measure the risks associated with particular sites? How should the sites be prioritized? What should be done to address them? Candia presented several anecdotal descriptions of contaminated sites in an effort to describe the range of challenges Chile faces. (Attachment N provides details.) He noted that several tools are being used to assist with site identification and site characterization efforts. For example, Fundación Chile is using GIS software to prioritize problem areas and is developing (in collaboration with the Swedish IVL Institute) a support tool called TORESA (a tool kit for remote sensing risk assessment) to evaluate environmental risks associated with mining activities (using geophysics as a basis).

Candia said that the Environmental Remediation group is searching for technologies that offer cost-effective solutions to environmental problems. As part of this effort, Fundación Chile (in cooperation with several international partners) has initiated an 18-month project to evaluate the possibility of using PRBs to address metal contamination at Chilean mining sites. The project team is currently in the first phase of the project, which involves site selection, site characterization, and laboratory studies. The latter will be performed to evaluate the remedial properties of a variety of reactive media, including iron and modified zeolites. (Attachment N provides detailed information about the reasons why modified zeolites are being considered as a potential remedial agent.)

In response to a question from the audience, Candia said that much of Chile's interest in remedial tool development is market-driven. Some evidence suggests, however, that the government might start focusing more attention on environmental remediation issues in the future. Another attendee asked whether olives, grapes, and other export crops have been evaluated to determine whether they have high metal content. Candia said that a proposal has been written to perform such evaluations.

Trenchless PRB Installation Projects Installed in San Antonio for Kelly Air Force Base
Jim Ortman, GeoSierra

Jim Ortman provided information about GeoSierra's trenchless PRB construction method, a technique that has been used to install PRBs in New Jersey, Iowa, California, Virginia, Oklahoma, and Texas. He said that GeoSierra's PRB construction approach offers several advantages, the most important being that it is capable of installing PRBs to much greater depths than would be possible using a more conventional installation approach. He also noted that GeoSierra's approach, unlike some other competing installation techniques, is well suited for use at sites with overhead obstructions and underground utilities. He presented a six-minute video that described GeoSierra's methodology, an approach that consists of the following steps: boreholes are drilled into the ground, frac casings are emplaced within the boreholes, the alignment of the casings is checked with a downhole camera and compass, the frac casings are grouted into place, packers are installed, iron filings are mixed aboveground in a viscous gel material, and the iron/gel mixture is injected into the casing at 120 psi. When the casing opens, an artificial fracture is created and the iron/gel mixture follows it. The surrounding soils, which are displaced in the process, push back on the iron/gel mixture and act like a vice that holds the reactive materials in place. An enzyme breaker, which is introduced at the same time that the iron/gel mixture is injected, breaks the gel down into water and sugar within a couple of hours. Ortman said that GeoSierra's construction method can be used to create walls up to 9 inches thick.

Shifting his focus to a specific site, Ortman provided information (see Attachment O (PDF, 23 pp., 456 KB) for details) about work that GeoSierra is performing at Kelly Air Force Base. He said that the U.S. Air Force has asked GeoSierra to construct five PRBs at this site, a project that involves emplacing about 1,000 tons of iron at average depths of 20 to 40 feet bgs across 3,650 linear feet. Of this total linear distance, about 2,100 feet will be constructed underneath city streets and 1,550 feet will be constructed within an active railroad right-of-way—both of which, Ortman said, represent scenarios in which it would be nearly impossible to dig a trench. Ortman said that three of the five PRBs have been completed at Kelly Air Force Base. He presented a series of pictures (see Attachment O) to give attendees an idea of the steps that were involved with the construction process. He also described the intensive QA/QC that GeoSierra performs to validate its warranties that its PRBs are constructed in the correct location, meet design specifications, and are permeable. For example, he said, hydraulic pulse interference tests are performed before and after construction activities to make sure that GeoSierra's PRBs are indeed permeable. In addition, active resistivity imaging is used to ensure that all segments of the PRB are coalesced and that a continuous wall of iron filings (with no voids) has been established. Also, inclined profile restivity probes are used to verify the precise thickness of each PRB.

One audience member asked about the costs associated with GeoSierra's construction technique. Ortman said that it will cost about $8.5 million to install the five PRBs at Kelly Air Force Base. Providing another example, he noted that it cost about $2.6 million to install a 500-foot-long, 6-inch-thick PRB at a depth of 70 to 95 feet bgs at the Tinker Air Force Base. Although this cost might sound high, he said, it was much cheaper than the proposed alternative, which would have been a pump-and-treat system that was projected to cost $4 million in capital costs and $300 per year in O&M costs. Vidumsky asked whether GeoSierra has experimented with the idea of injecting anything other than ZVI into the subsurface. Ortman said that some work has been initiated in this area and that GeoSierra is currently evaluating remedial agents that could be injected to address perchlorate contamination. Referring to a presentation delivered earlier in the meeting, Tom Simpkin said that Jerry Jin's group is also in the process of installing a PRB at Kelly Air Force Base and that they have concluded that their PRB needs to be 15 feet thick—a thickness that is much greater than that which is being used in GeoSierra's PRBs. Simpkin asked Ortman and Jin to comment on this. Jin said that he and Ortman are working on different parts of the site and that contaminants and ground-water velocities differ between the two areas. Ortman acknowledged this but said that he still thinks that Jin's group might have overestimated the thickness needed for their PRB.

Laboratory Screening of Alternate Reactive Materials for PRB Construction
Michelle Thomson, URS Corporation

Michelle Thomson provided information about laboratory studies that are underway to compare the performance of different types of iron. This project, being performed through the collaborative efforts of DuPont and ETI, was inspired by the fact that conventional granular iron products (such as Connelly iron and Peerless iron) have experienced a dramatic price increase over the last 18 months as the global demand for scrap iron has risen. (The price is expected to continue on its upward path, Thomson said, noting that some estimates suggest that the cost per ton will increase by an additional $30 each month.) DuPont plans to install 1,500 tons of iron in a PRB at a site that has a carbon tetrachloride plume commingled with CFC 11 and CFC 113. (John Vidumsky spoke about this site earlier in the meeting; see page 1 of this summary for details.) Given the large amount of iron called for in the PRB design, Thomson said, DuPont would achieve significant cost savings if it identified a less expensive iron product that could be used as an alternative to the conventional granular iron products. Toward this end, a series of column studies have been performed to compare the reactivity and degradation potentials of the conventional granular iron products (Connelly and Peerless) against an atomized iron product (QMP iron) and a sponge iron (ISPAT iron). Each column received a feed solution that contained distilled water, 60 ppm of carbon tetrachloride, 50 ppm of CFC-11, 10 ppm of CFC-113, and 300 ppm of calcium carbonate. (The latter was included to simulate field conditions.) The results reported are all preliminary at this point, but they do suggest the following:

Thomson said that the research team also recently introduced some TCE into the columns to see how this contaminant responds to different irons. This study has not been running long enough, however, to allow definitive statements about performance. Additional tests are scheduled to begin soon to examine how other chlorinated solvents respond to the different irons. Thomson also said that the project team recently received another iron product—produced by Hoeganaes Corporation—and that column studies have been initiated for this material as well. Results are not yet available. She also said that the project team plans to conduct microscopic analyses on the iron products and to run column tests with the alternate iron sources to determine whether their performance could be impacted by biopolymer guar, crosslinkers, or enzyme breakers, all of which will play a role in constructing DuPont's PRB.

Developments in Nanoscale Iron Technology for the Treatment of CVOC Source Areas
David Vance, ARCADIS G&M, Inc.

David Vance provided information about studies that are being conducted to further the development of nanoscale iron products as remedial agents. His presentation is included as Attachment P (PDF, 36 pp., 541 KB). He opened by talking about a nanoscale iron pilot project at a fractured bedrock site in New Jersey. (A poster was presented about this site at Battelle's Monterey conference; Vance will send a copy of the poster's text to anyone who desires additional information.) Vance said that 30 months of molasses injections failed to adequately address the trace DNAPL that is present at this site. Thus, in an effort to address the problem, the site team injected 100 pounds of Crane Polyfon Polymetallix™ precipitated iron colloid into the subsurface. Although the iron exerted a significant initial remedial effect, contaminant concentrations rebounded over time. Vance said that he thinks rebound occurred because the amount of iron used in the pilot project was suboptimal. Although quantities of 1,000 pounds might have been justified from a design point of view, budget constraints prevented the use of such quantities.

Vance noted that several factors play a role in determining a nanoscale iron product's reactivity, including particle size, the amount of reactive surface area, the presence or absence of hydrogenation catalysts (e.g., palladium), the method of manufacture, the morphology of the particle, the crystalline structure of the particle, impurities and coating, and whether or not particles have been exposed to acid washing. Vance presented pictures of three nanoscale irons. One of the products—a ball milled material produced by OnMaterials using a top-down approach—has particles in the 400 nm range. The other two materials—one produced by Arcadis and the other by Crane Polyfon PolyMetallix—were created using a bottom-up approach. Vance said that batch reactor studies have been performed to compare the reactivity and kinetics associated with these three different nanoscale iron products. Results are presented in Attachment P.

Vance devoted part of his presentation to delivery issues, pointing out that nanoscale iron particles must reach target zones in order for a remediation project to be successful in the field. He talked about Brownian motion and the effects associated with gravitational settling, and he presented a slide that shows how points of zero charge differ across minerals. He also emphasized the importance of obtaining a firm grasp on site hydrodynamics. Vance said that column studies have been performed to evaluate nanoscale iron distribution; the results suggest that better ROIs are achieved when pressure pulse techniques are employed. Before leaving the topic of delivery, Vance said that he thinks it would be interesting to design a reagent that emulates the transport properties of NAPL. If this could be accomplished, he said, the reagent would likely be able to migrate to NAPL-impacted areas.

Vance concluded by discussing some of the key issues that require further attention. First, he said, efforts need to be made to reduce the price of nanoscale iron so that this material is considered a legitimate option for field application. Also, efforts need to be made to determine how well nanoscale irons treat DNAPLs and how long nanoscale iron is likely to remain reactive in the field. In fact, Vance said, ESTCP has asked Arcadis to start examining these questions in upcoming studies. He presented detailed information (see Attachment P) about the parameters that will be evaluated as part of this effort and the irons that will be examined. Before concluding, he also mentioned that more work should be performed to determine how palladium coatings affect reactivity—some regulators resist the idea of using palladium-coated iron particles because palladium is a carcinogen.

One audience member asked about the cost of nanoscale iron products. Vance said that prices vary across different products. In general, he said, the "precipitated" materials are selling for $80 to $100 per pound and the ball-milled material is selling for about $5 to $20 per pound.

Treatment of Arsenic and Metals in Ground Water Using a Compost/ZVI PRB
Ralph Ludwig, EPA

Ralph Ludwig, whose presentation is included as Attachment Q (PDF, 50 pp., 1.87 MB), described a pilot-scale PRB that has been installed at the Columbia Nitrogen site. This site was formerly used to manufacture phosphate fertilizer via a phosphate rock acidification process. The sulfuric acid that was used in the manufacturing process was produced on site and stored in lead-lined concrete vats. Pyrite was used to generate the sulfuric acid at first, but elemental sulfur was used in later years. Many materials, such as pyrite cinders, bricks, and other debris, were dumped on site over the years. In a nutshell, Ludwig said, the site poses an acid rock drainage problem. The ground water has been impacted by arsenic, ferrous iron, lead, and cadmium and is characterized by pH values in the 2 to 4 range. It discharges to a tidal marsh; additional acid forms in the marsh when ferrous iron daylights and is transformed (through oxidative processes) to ferric hydroxide. In an effort to identify a remedial solution, a pilot PRB was installed at this site about 2 years ago. Ludwig said that the objectives for the PRB are fourfold: (1) raise ground-water pH; (2) promote sulfate reduction to remove heavy metals, arsenic, and ferrous iron; (3) prevent additional downgradient ecosystem impacts by preventing additional acid production that results from ferrous iron oxidation; and (4) contribute to ecosystem restoration by converting ground water from an acid-producing system to an acid-consuming system.

Ludwig said that the reactive material chosen for the PRB consist of the following: ZVI (20 percent), compost (30 percent), pea gravel (45 percent), and limestone (5 percent). Before going into the field, laboratory studies were performed to confirm that this mixture would exhibit strong reactivity and that it would be more permeable than the surrounding aquifer. Ludwig said that the project team would have preferred to install the PRB at the edge of the tidal marsh, but that they were unable to do so due to cost and permitting constraints. As an alternative, they were forced to install the PRB in the source zone, which means that there are contaminated materials located downgradient and upgradient of the PRB. Ludwig said that the PRB is 27 feet long and 6 feet wide. In terms of depth, the PRB extends to 12 feet bgs, but the reactive materials are only located in the bottom 7.5 feet. (The top 4.5 feet consist of 0.5 feet of limestone, a plastic sheet, and 4 feet of gravel.) Attachment Q presents a series of pictures that depict the PRB installation process. In summary, the installation effort involved digging a trench, using guar gum to support the trench's side walls, and backfilling with reactive materials. The reactive materials got stuck to the sides of the mixing bucket when they got wet. As a result, the contractor had to shake the bucket hard to loosen the materials. There is some concern that he banged the trench in the process and caused some minor cave-in.

Ludwig presented information about the monitoring system that was installed. He also listed the chemical parameters (lead, arsenic, cadmium, iron, pH, Eh, alkalinity, cations, anions, TOC/DOC, sulfide) and microbial parameters (MPN sulfate reducer measurements) that are being evaluated, as well as the hydraulic monitoring activities that are being performed. Data are available for the first 18 months of the PRB's operation; Attachment Q provides a detailed breakdown of the findings. In summary, Ludwig said that the following can be concluded thus far: (1) the PRB has been effective in removing lead, cadmium, and arsenic; (2) the pH of influent ground water has increased to values greater than 6; (3) the ground water is being converted from acid-producing conditions to acid-consuming conditions; (4) a significant number of sulfate-reducing microbes are present in the PRB; and (5) sulfur 34 enrichment has been observed in the PRB. Although these results are very encouraging, Ludwig said, some unexpected problems have emerged. For example, the project team has found that ground-water flow through much of the site is very slow and that the PRB is less permeable than anticipated in some areas. Ludwig said that more work remains to be performed at this site. For example, the research team plans to continue monitoring the PRB on a semi-annual basis to assess performance and longevity issues. In addition, the team hopes to perform a tracer test in the near future to obtain a better grasp on site hydraulics. If a decision is made to go full-scale at this site, Ludwig said, the project team might consider making some modifications to the PRB design.

Audience members asked whether the research team has formulated any theories to explain why the permeability of the PRB is lower than anticipated. Ludwig said that part of the problem might be pyrite cinders migrating into the PRB materials. Clausen advised performing some research to determine whether biomass buildup could be contributing to the problem, noting that including compost into the reactive media mixture could place the PRB at risk of developing such problems. Puls asked whether the project team saw any visual evidence of biomass buildup when they collected soil cores. Ludwig said they had seen none in the cores but that some black mass was pulled from the ground when the team was developing wells for a slug test. Another audience member suggested looking into whether the low pH conditions could have interfered with the degradation of the guar gum. Switching to a new topic, Clausen asked whether the project team knows what mechanisms are responsible for the metal-removal activities that have been recorded within the PRB. Ludwig said that he did not know the answer to this question yet, but that work is currently ongoing to shed light on these issues.


Puls closed the meeting by reiterating his gratitude to ETI and ARS Technologies, Inc., for sponsoring the meeting refreshments and lunch break. He also thanked the speakers for their presentations and the audience members for their questions. He also thanked those who played a role in planning the meeting.

Attachments A through Q

Attachments A through Q are available on the Internet. To view these attachments, visit the RTDF home page at, click on the “PRB Action Team” button, then click on the “Team Meetings” button. The attachments will be available as part of the October 2004 meeting summary.

Attachment A: Final Attendee List (PDF, 3 pp., 72 KB)

Attachment B: Degradation of Explosives at Cornhusker Army Ammunition Plant Using a ZVI PRB (Tom Krug) (PDF, 43 pp., 771 KB)

Attachment C: In Situ Remediation with EHC™ (David Hill) (PDF, 22 pp., 1.48 MB)

Attachment D: Using an Apatite II PRB To Remediate Ground Water Contaminated With Zinc, Lead, and Cadmium (James Conca) (PDF, 35 pp., 1.50 MB)

Attachment E: Electrically Induced Redox Barriers for Treatment of Ground Water (Tom Sale) (PDF, 26 pp., 853 KB)

Attachment F: In Situ Removal of Heavy Metal Contaminants Using Emulsified Nanoscale or Microscale Metal Particles (Kristen Milum) (PDF, 23 pp., 1.17 MB)

Attachment G: Materials Science Perspectives of Injectable Zero-Valent Metals and Alloys (Clint Bickmore) (PDF, 29 pp., 1.90 MB)

Attachment H: Use of ZVI for Ground-Water Remediation: Three Case Studies (Manny Saenz) (PDF, 21 pp., 1.32 MB)

Attachment I: Remediation of Chlorinated Solvent Source Zones Using ZVI-Clay in Conjunction with Soil Mixing (Tom Sale) (PDF, 33 pp., 509 KB)

Attachment J: EZVI Treatment of Chlorinated Solvent DNAPL Source Areas (Suzanne O'Hara) (PDF, 34 pp., 1.40 MB)

Attachment K: Update on the ITRC's PRB Team Activities (Matt Turner) (PDF, 8 pp., 203 KB)

Attachment L: Design of a PRB for Chloroethenes and Hexavalent Chromium Treatment in Ground Water at the Former Kelly Air Force Base (Jerry Jin) (PDF, 23 pp., 649 KB)

Attachment M: Geochemical Characterization of a Multiple PRB, Mortandad Canyon, New Mexico (Patrick Longmire) (PDF, 19 pp., 488 KB)

Attachment N: Ground-Water Contamination in Chile and the Use of Zeolites as a Reactive Medium (Juan Ramón Candia) (PDF, 45 pp., 1.76 MB)

Attachment O: Trenchless PRB Installation Projects Installed in San Antonio for Kelly Air Force Base (Jim Ortman) (PDF, 23 pp., 456 KB)

Attachment P: Developments in Nanoscale Iron Technology for the Treatment of CVOC Source Areas (David Vance) (PDF, 36 pp., 541 KB)

Attachment Q: Treatment of Arsenic and Metals in Ground Water Using a Compost/ZVI PRB (Ralph Ludwig) (PDF, 50 pp., 1.87 MB)