Cheryl L. Dahlin, Connie A. Williamson, W. Keith Collins, and David. C. Dahlin, US Department of Energy, Albany Research Center

Speciation and distribution of heavy metals in soils controls the degree to which metals and their compounds are mobile, extractable, and plant-available. Consequently, speciation impacts the success of remediation efforts both by defining the relationship of the contaminants with their environment and by guiding development and evaluation of workable remediation strategies. The U.S. Department of Energy, Albany Research Center (Albany, OR), under a two-year interagency project with the U.S. Environmental Protection Agency (EPA), examined the suitability of sequential extraction as a definitive means to determine species of heavy metals in soil samples. Representative soil samples, contaminated with lead, arsenic, and/or chromium, were collected by EPA personnel from two Superfund sites, the National Lead Company site in Pedricktown, NJ, and the Roebling Steel, Inc., site in Florence, NJ. Data derived from Tessier=s standard three-stage sequential-extraction procedure were compared to data from a comprehensive characterization study that combined optical- and scanning-electron microscopy (with energy-dispersive x-ray and wavelength-dispersive x-ray analyses), x-ray diffraction, and chemical analyses. The results show that standard sequential-extraction procedures that were developed for characterizing species of contaminants in river sediments may be unsuitable for sole evaluation of contaminant species in industrial-site materials (particularly those that contain larger particles of the contaminants, encapsulated contaminants, and/or man-made materials such as slags, metals, and plastics). However, each sequential extraction or comprehensive characterization procedure has it=s own strengths and weaknesses. Findings of this study indicate that the use of both approaches, during the early stages of site studies, would be a best practice. The investigation also highlights the fact that an effective speciation study does not simply identify metal contaminants as specific chemical compounds but also defines their variations in chemistry, size, associations, concentrations, porosity, adsorption, liberation, fracturing, alteration (products and potential), and host materials.

Paul Carroll, M.S., USEPA, H. Patricia Hynes, M.S., Boston University School of Public Health, and Rob Maxfield, M.S., USEPA.

In-situ measurements were made at more than 100 properties in low-income neighborhoods around Greater Boston to identify and delineate lead (Pb) contamination in these urban residential soils as part of the Lead Safe Yard Program (LSYP). The LSYP is an EPA/EMPACT funded program designed as a pilot study to develop on-site soil lead monitoring, low-cost remedial landscape techniques, and educational and instructive materials for reducing children's exposure to lead in soil in a high risk urban neighborhood. The program uses innovative field portable X-ray fluorescence spectrometry (FPXRF) methods to collect these in-situ measurements and low cost/no cost landscape-based mitigation methods to reduce the risk of human exposure to elevated levels of Pb in these residential soils. We present field data (n=2784) focusing on typical findings at an urban residential property and background levels of lead in soil for this urban setting. QC data, focusing on sensitivity, and precision and accuracy, will also be presented to demonstrate that FPXRF is an excellent tool to delineate lead contamination in urban soils at concentrations suitable for evaluation against current health based standards.

Michael V. Ruby, Yvette W. Lowney, Exponent; Stephen M. Roberts, University of Florida; Rosalind A. Schoof, Gradient Corporation

This presentation will discuss the state of the science with regard to measuring the oral bioavailability of arsenic in soil. Currently, there are two animal models available for measuring relative arsenic bioavailability from soil: the young swine model developed by U.S. EPA Region VIII at the University of Missouri-Columbia, and a Cebus monkey model developed at the University of Florida on behalf of the Florida Department of Environmental Protection. The design characteristics of these two models will be reviewed, with particular attention to their environmental relevance and the manner in which they mimic incidental human exposures to arsenic in soil. Arsenic bioavailability estimates obtained with these two models will be discussed.

This talk will also review a study that is underway to measure relative arsenic bioavailability from 7–10 additional soils in the Cebus monkey model. These soils derive from mining, milling, smelting, refinery, pesticide manufacturing, and orchard sites, and were selected to obtain a range of arsenic forms and soil types. In addition, the two in vivo models will be compared by testing in monkeys, or swine, a minimum of three soils previously tested in the other in vivo model. The ultimate goal of this research effort is to develop a robust in vivo database against which an in vitro extraction test can be validated. Progress towards achieving this goal will also be discussed.

This study addressed the use of chloride as a chelating agent for the selective removal of lead from a calcareous soil. The test soil employed in this study was contaminated by lead-acid battery waste. A component of the study focussed upon the use of precipitating agents for treatment of the soil extracts thereby allowing recycling of the leaching solution. In the soil under study, the Pb concentration in the particle sizes smaller than 4.75 mm was found to be 3,200 mg/kg, and the calcium concentration was found to be 85,700 mg/kg. A finer fraction of the soil (£ 2.0 mm) contained 13,200 and 155,200 mg/kg of Pb and Ca respectively. With a NaCl concentration of 8 M the removal of Pb was found to be 83% while 9% of the Ca was removed. Multiple extractions of Pb with 8 M NaCl were performed and the removal of Pb was found to 93% after six successive extractions. The removal of Pb increased with time in a batch test and approached 80% after 90 h. Increasing removal of Pb was found as the water/soil ratio was increased at a constant concentration of NaCl. The addition of 25 g/L of lime, 4.5 g/L of sodium carbonate, and 8.2 g/L of calcium carbonate to the soil extract resulted in Pb removals of 90, 99.7, and 35% respectively. In the case of sodium carbonate, the removal efficiency of Pb was further improved when the pH was adjusted to 8.2. The recycling of free chloride generated from the extract in a remediation process revealed that 91% of Pb was removed from the soil (particle size £ 4.75 mm) after six cycles.

Jiasheng Cao and Wei-xian Zhang, Lehigh University

Dimitris Dermatas and Nektaria Menounou, Stevens Institute of Technology

Lead soil contamination is an environmental problem in army firing ranges. In particular, training or shooting range areas are an important component of the military activities but on the same time they adversely impact the environment. There is inevitable buildup of bullets and residues of other related activities in impact areas and therefore ranges are potential source areas for lead and other heavy metals and other contaminants. If left unattended, the metals may be transported into the environment via various pathways. This study was performed to evaluate the extent of the problem and the fate of lead in selected firing range soils. Both total digestion and toxicity characteristics leaching procedure (TCLP) experiments were conducted to examine lead presence and mobility. These tests indicated that high concentrations of lead were generally present and that the soil leached more lead than what is acceptable under EPA regulations. In some cases TCLP-Pb was as high as 300 mg/L. Lead was also measured in selected particle size ranges in the soil. Different quantities and particle size distribution of lead will likely be encountered depending up on weapons fired and projectile backstop/ impact surface. In this study, higher quantities of lead were found at the –#10+40 group for most of the samples tested. This is critical in assessing potential remedial alternatives.

Edward J Zillioux, Florida Power & Light Company, Donald B. Porcella, Environmental Science & Management, Thomas M. Grieb, Tetra Tech, James R. Newman, Pandion Systems, Geoffrey B. West, Florida Power & Light Company

An historic perspective on mercury in biota is important to identify factors that control its accumulation. Raccoons are an important food source for certain mammals, especially including the endangered species, the Florida Panther, when preferred upland prey is scarce. We selected five areas in Florida, mostly south of US Highway 41 (south of latitude 26 ^oN), as appropriate sources of historic (museum – most from 1939-1957, six from 1990) and modern (1999-2000) samples: the Shark Valley Slough (SVS), the Ten Thousand Islands, (Chokolosksee Bay area (CB) area, the Flamingo area (FLM), Long Pine Key (LPK), and North Key Largo (NKL). Sites were based in part on where the museum samples had been collected. Modern collections were designed to provide nearly equal numbers of museum and modern specimens, totaling respectively, 14 and 19 (SVS), 25 and 25 (CB), 20 and 21 (FLM), 9 and 11 (LPK), and 15 and 17 (NKL) in North Key Largo. A total of 209 specimens (91 museum and 118 modern) were analyzed for a variety of chemicals in addition to monomethylmercury (MMHg). Hair provides a good archive for MMHg, which constitutes more than 95 percent of the total Hg in hair. Also, hair MMHg correlates well with concentrations in many tissues (heart, blood, brain, kidney, and muscle), and therefore can act asrepresent the principal variable to understand the role of raccoons in transferring mercury. Liver MMHg is relatively low and invariable and does not correlate well with other tissues. Substantial quality control analyses show the results are high quality. Collection of modern samples from comparable locations and habitats as historic samples increases confidence that temporal differences can be seen. For example, a preliminary inspection of comparable SVS data (between latitudes 25 ^o 38’ to 25 ^o 46’N) suggests that mean levels of hair mercury occurring in animals from 1990 (n=6) are c. about 6-fold higher than in specimens from the 1940s (n=8) and c.about 3-fold higher than in the 2000 animals (n=8). We will draw present conclusions about raccoon historic and modern Hg levels at the conference.