Groundwater Movement at the Portsmouth Gaseous Diffusion Plant
Beat Hintermann, Marilynn dela Merced and Marvin Resnikoff
Radioactive Waste Management Associates
PRESS and the Uranium Enrichment Project
PORTS Report Summary
The Portsmouth Gaseous Diffusion Plant (PORTS), located in Pike County, in the south central region of Ohio, covers an area of about 3,700 acres. There are two groundwater layers below the site; there is also a creek on-site that eventually flows to the Scioto River, serving as a discharge for liquid effluents produced by the plant. In addition, a total of 120 acres of land has been converted into landfills. Several residences, including two nursing homes, are directly adjacent to the plant, and the total population of the area within a 50-mile radius is approximately 600,000.
The plant's basic function, like that of all gaseous diffusion plants (also Paducah and Oak Ridge), was to enrich uranium (U) in the fissionable isotope U-235. As a result of the enrichment process, significant amounts of toxic and radioactive materials were released into the air, water, and soil as byproducts.
PORTS began operations at the end of 1954 in order to increase the government's capacity to produce highly enriched uranium for naval reactors and for the burgeoning nuclear power industry. In 2001, the United States Enrichment Corporation, the present operator, placed the Portsmouth plant on cold standby; by that time, PORTS had processed more than 320,000 metric tons of uranium. Meanwhile, the Department of Energy (DOE) is presently engaged in a major remediation effort to control and remove contamination caused by past operations.
DOE's and local residents' opinions greatly differ on the environmental impact caused by the Portsmouth plant. Believing the local cancer is excessive, local residents continue to be concerned about releases from the plant and the subsequent effects on their health. The DOE states that the plant has had only a minor impact on the environment. The department presents current soil and water concentrations as evidence. However, current soil and water radionuclide concentrations would not explain an increased rate for cancers with a latency period of 10 to 30 years. In many cases, DOE did not take samples of these concentrations in the early years of operation (1954 until late 1970's), and we may never know how much contamination was released.
Two public interest groups, PRESS and the Uranium Enrichment Project, sponsored this work. It represents a different perspective from that seen in reports by federal and state agencies and their contractors, even though this report has been limited by the available data, especially for the earlier years.
Contamination On- and Off-site
The gaseous diffusion process involves passing uranium in gaseous form, uranium hexafluoride (UF6), through thousands of stages that favor the passing of the lighter isotope U-235 and restrain that of the heavier isotope U-238. This series of stages is called a cascade, which is essentially a uranium enrichment machine, housed in three enormous buildings. First, UF6 is fed to the cascade in large cylinders. Then, the enriched U (product) is retrieved at the top, whereas the depleted uranium (tails) is collected at the bottom of the cascade. The product and the tails are stored and transported in the same cylinders as the feed.
As a result of introducing recycled uranium into PORTS (a total of about 1,100 metric tons between 1955-98), miles of pipes were coated with radionuclides, primarily technetium (Tc), but also plutonium (Pu) and neptunium (Np). These radionuclides are produced in nuclear reactors through fission (Tc) and capture of neutrons by uranium (Pu, Np) and are not present in natural uranium. Tc-99 is present in the cascade in large quantities, because it forms a hexavalent form similar to uranium. By 1976, a total of 60 - 90 kg of Tc has been fed to the cascade.
A purge gas stream extracts the light gases (everything lighter than UF6) from the cascade. Before being discharged to the atmosphere, the radionuclides are partially filtered out. Considerable amounts of U and Tc were not captured by the filters and were vented through the cascade stacks. Again, the exact amounts are unknown due to the lack of sampling in early years. The filters and other capture media would be buried in an unlined landfill.
During periods when the cascade was upgraded, or when pumps and other parts had to be replaced, PORTS removed contaminated pipes and equipment. In the decontamination process, radionuclides (especially Tc) entered the air and liquid waste streams and were discharged to a holding pond on the east side of the site. Clearly plant workers were the group most heavily exposed to hazardous chemicals and radionuclides in the air. For wastes in liquid form, contamination flowed from the east holding pond to Little Beaver Creek and eventually the Scioto River and the Ohio River, but liquids also seeped into the groundwater below the site and formed a plume extending east past Little Beaver Creek. In the mid 1970's, PORTS began to monitor and capture Tc with ion exchange resins before discharging the waste solutions. The contaminated resins were buried in unlined landfills, along with spent filter material from the cascade, other contaminated solid material from the plant, and ashes from the feed material plant, the oxide conversion facility and the incinerator. From these landfills, contaminants inevitably leaked into the groundwater. Groundwater beneath the site locally flows towards Little Beaver Creek in the northeast, Big Run Creek in the southeast, and to a drainage ditch in the west. Based on groundwater flow, the site is divided into four Quadrants (Figure A). However, the aquifers are much larger than the site and extend past these local discharge areas.
Along with radionuclides, a broad range of chemicals used for the operation and maintenance of the plant entered the environment, including (but not limited to) volatile organic compounds (VOC), oils, acids, heavy metals, nitrates and fluorides. Of all contaminants, the degreaser trichloroethene (TCE) migrates fastest in water, and therefore the TCE plume defines the extent of groundwater contamination.
Plumes of contaminated groundwater exist in different parts of the plant. The most extensive plumes are located below the holding pond (Figure B) and the buildings X-705, X-700, X-120, as well as under the X-749, X-749A, X-735, X-231B and PK landfills (Figure C). Concentrations of TCE in these plumes were measured as high as 820,000 mg/L (holding pond), whereas Tc was detected at levels of 565,836 pCi/L (X-705). In comparison, the drinking water standards for TCE and Tc are 5 mg/L (US EPA) and 50 pCi/L (Ohio EPA), respectively.
Low levels of Pu, Np and mercury were found in soil samples from a 340-acre area that is within the outer plant boundary, but outside the actual plant site. This area acted as a buffer zone between the plant and residents and is in the process of being transferred to the Southern Ohio Diversification Initiative (SODI) for industrial use. The 340 acres are to the north of the X-705 oxide conversion facility and an incinerator. The fact that low levels of contamination were found there suggests that there are likely to be higher levels of contamination downwind and to the east of these facilities. The soils to the east should be thoroughly investigated.
In addition to the contamination on-site, contaminants have been found off-site, though DOE continues to dispute this. The plume below the holding pond on the east side extends to Little Beaver Creek, which has to be considered off-site, since it flows out of the site boundary. In 1975, average levels of Tc in surface water in the creek were about 6,800 pCi/L. As recently as 1996, TCE and Tc were measured at levels of 110 mg/L and 68 pCi/L. Tc has been repeatedly found in local residential monitoring wells, even though it is below the allowable limit. Also, as recently as 1992, sediments in Little Beaver Creek, Big Beaver Creek and Scioto River were found to have gross beta concentrations that were 5 times above background. In addition, elevated levels of gross beta were found in fish tissue from all three rivers.
DOE is making progress in remediating contaminated facilities and soil on-site. A total of 4.36 million pounds of soil has been removed and shipped to a storage facility in Utah, and it appears that more will have to be shipped as more contaminated soil is found. Landfills have been capped with an impermeable layer to prevent rainwater from dissolving more contaminants. Underground slurry walls and interceptor trenches have been built to contain the groundwater plumes as much as possible.
Progress on groundwater contamination, and the ultimate status and use of the site, however, remains slow. Because of the large size of PORTS, it may be years before some of the toxic or radioactive materials actually reach the plant boundary through groundwater, though this has already happened on the eastern side of the plant.
To prevent contaminants from migrating further off-site, PORTS continuously extracts groundwater with sump pumps, partially treating the contaminants at 5 groundwater treatment facilities, and discharging the water directly to the Scioto River, thereby changing the direction of the groundwater flow. Radionuclides are not removed from the water during the treatment process. DOE claims that no Pu and Np were found in samples from the discharged water; however, no measurements for Tc were given, the only contaminant present in significant amounts in the groundwater plumes. It is possible that the Tc in the extracted water is dumped directly into the river. In 1999, a total of 24.6 million gallons of water was extracted. While we believe that this is an effective method to hold the groundwater plumes on-site, active maintenance will be required for the indefinite future, because some of the contaminants, especially radionuclides, have half-lives of hundreds of thousands of years.
The key problem of the site and the reason for the lack of environmental data is that DOE made up the rules as the situations arose. Radionuclides and hazardous chemicals were initially released to the environment with no containment. Over time, monitoring and containment was added, but considerable damage had already been done. Now it is a problem of costly catch-up and remediation.
An independent and continuing review of the contamination, risk analysis and cleanup process would ensure that all necessary measurements are carried out, that data is interpreted in a responsible way, and that the resulting cleanup will indeed be effective.
Because of significant contamination of the site, residential use of the premises is not being considered for the foreseeable future. Industrial use of the site is possible only because industrial occupants are not expected to live in subsistence on the site and therefore do not ingest groundwater and locally grown crops. However, it is impossible to predict where people will live in several hundreds or thousands of years, long after DOE loses institutional control of the site. For this reason, any risk assessment should consider the residential farmer scenario.
As it stands, environmental management at Portsmouth requires annual funding from Congress that waxes and wanes as a function of budget pressures and the Administration and Congress in power. This is an entirely unsatisfactory situation for long-lived waste materials at Portsmouth. We recommend that a trust fund, not subject to the political winds, be established for perpetual maintenance of these waste materials and the Portsmouth site.
This report was supported by
grants from the Citizens' Monitoring and Technical Assessment Fund and
from The John Merck Fund.
Copyright © 2002 by
|Please note: The complete report is available from Yggdrasil
(address below) at a cost $18 postpaid.
comments, suggestions, &
technical matters contact: Janet Powell