Skip directly to search Skip directly to A to Z list Skip directly to site content





PGDP operations and waste disposal activities have resulted in airborne releases of radioactive andchemical compounds. ATSDR scientists reviewed PGDP processes that produce air contaminantreleases, selected contaminants of concern, identified potentially exposed populations, and evaluatedexposure to those populations. Determination of contaminants of concern is based largely oninformation about process operations and accidents that produce air contamination, releaseinformation, data reported from the perimeter fence and off-site air monitors, and air dispersionmodeling. Descriptions of air dispersion models used to evaluate air releases from the PGDP facilityare included as Appendices E through H.

Airbourne contaminants at PGDP occur or have occurred as a result of primary operations and accidental releases. Determination of contaminants of concern is based largely on information about process operations and accidents, release information, perimeter and off-site monitoring data, and air dispersion modeling.Contaminant releases to the atmosphere at PGDP occur or have occurred as a result of primary operations/processes and accidental releases. These processes include (1) uranium hexafluoride (UF6) enrichment and handling, (2) uranium fluoride manufacturing, (3) uranium metal production, (4) metal finishing and grease removal, (5) recycling and smelting other metals, (4) cooling process water through the water cooling towers, and (5) generation of electricity by the coal-fired steam plant. Other process releases are caused by secondary operations, such as groundwater treatment and cylinder maintenance (cleaning and painting cylinders). Each operation is a source of specific air contaminants; one can determine which contaminants an operation produces by estimating material usage or loss. Table 7 lists contaminants released to the air from processes at the PGDP site, maximum annual releases with the year of release, and major operations or release points. Contaminant sources or release points that no longer exist are noted in the table.

During the uranium enrichment process, UF6 is released to the atmosphere mainly through vents inthe process buildings and through a 61-meter stack at Building C-310 [62]. These releases occurprimarily during the transfer of the gaseous UF6 from the cascade equipment into the storagecylinders. UF6 is the primary contaminant released in this process. Once released into the air, UF6reacts rapidly with atmospheric water to form hydrogen fluoride (HF) gas, uranyl fluorides, anduranium oxides [63]. The year in which the most uranium was released was 1956 [12]. Betterfiltration and operational procedures reduced the total air release of uranium in 1977 to less thanone-tenth the release in 1956. (Refer to Table 8A.) Also, major modifications were made to the C-310 stack in 1983 to reduce uranium emissions further [64].

Large quantities of uranium and fluoride were released from 1952 until 1977, when UF6 wasproduced from uranium oxides at the feed plant (Building C-410) and when UF6 was converted touranium tetrafluoride (UF4) at the uranium metal plant (Building C-340). Most airborne uraniumduring these years originated at the Feed Plant, where UF6 was produced from uranium oxide [65].

Table 7.

Airborne releases (from PGDP processes) and major release sources [41,7]
Air Contaminants Maximum Reported Annual Release (Year) Major Release Sources
Hexavalentchromium1,700 kg (1990)Cooling towers
Hydrogen fluoride
(See Appendix F)Plant operations (includes building vents, seal exhaust/wetexhaust, product/tail withdrawals, and laboratory hoods); C-310stack emissions; U metals plant (past); C-400 UF4 pulverizer(past); C-410 HF storage and production areas (past)
Nickel, aluminum,and other metalparticulates(See Appendix I)Secondary metal smelting operations in C-746A facility (past)
Sulfur dioxide
Nitrogen oxides
399,579 kg (1993)
314,400 kg (1985)
PGDP's coal-burning steam plant
Trichloroethylene62,826 kg (1986)C-400 and C-720 degreasing operations, and spills (past);northeast and northwest water treatment facilities
Technetium 99 6.3 Ci (233.1 GBq)
Plant operations (includes building vents, seal exhaust/wetexhaust, product/tail withdrawals, and laboratory hoods); C-310stack emissions; U metals plant (past); C-400 spray booths, rotaryvacuum dissolver, laundry, cylinder drying station
Uranium 234
Uranium 235
Uranium 238
1.62 Ci (59.9 GBq)
0.08 Ci (2.96 GBq)
3.50 Ci (130 GBq)
Plant Operations (includes building vents, seal exhaust/wetexhaust, product/tail withdrawals, and laboratory hoods); C-310stack emissions; U metals plant (past); C-400 spray booths, rotaryvacuum dissolver, laundry, cylinder drying station; C-400 UF4 pulverizer (past)
UraniumHexafluorideUnknown1"Midnight negatives"2 before mid-1980s
1 Timeframe and frequency of releases unknown; quantities released per event unknown [66].
2 Releases through jets at night to accelerate attainment of adequate UF6 negative (<10 ppm) in system in order to maintain or inspect isolated process gas equipment [66].
Key: Ci = curies; GBq = gigabecquerels; kg = kilograms; HF = hydrogen fluoride; U = uranium; UF4 = uranium tetrafluoride

In addition to ongoing operational contaminant releases, accidental and/or inappropriate releases ofUF6 and HF have occurred throughout the operating history of the plant. The largest reportedaccidental release occurred in 1960, when a cylinder ruptured inside Building C-333 as a result ofoverfilling and released 17,800 pounds (8,074 kilograms) of UF6. Another major accident occurredin December 1962 during a fire in Building C-337; in that accident, 5,062 pounds (2,296kilograms) of UF6 were released. There have been several other accidents, but these have involvedmuch smaller quantities [67]. Refer to Appendices E and F for further details. Also, before the mid-1980s, it has been reported that UF6 was released at night through jets on top of the processbuildings to accelerate the reduction of UF6 concentration in the process gas system in order toperform maintenance and inspection on process gas equipment. These releases, called "midnightnegatives", potentially contained significant quantities of uranium and hydrogen fluoride; however,the quantities released and the frequency of releases are unknown [66].

Also, four deliberate releases of UF6 were made as part of experiments to model the behavior ofatmospheric releases of UF6 [61]. There were two releases in 1955 and two releases in 1974; the1955 releases involved 4.4 kilograms and 0.68 kilograms (10 pounds and 1.5 pounds) of UF6, andboth 1974 releases involved about 0.16 kilograms (0.4 pounds).

Fluoride releases were not reported for most of the years of operation. From 1955 until 1993, theAtomic Energy Commission, DOE, and their contractors used off-site air monitors to detect airbornefluorides, uranium, and beta-emitters in the surrounding environment [68]. A continuous stacksampler for HF was installed in the C-310 stack; however, the reviewed reports do not indicate whenit was installed or how much fluoride was released before 1985. Starting in 1985, total annual HFreleases were reported in the annual reports. Fluoride released from the C-310 stack and sulfurdioxide released from the C-600 steam plant were sampled continuously until 1993. Other emissionsfrom plant operations were intermittently sampled. The majority of the releases were determined bymaterial balancing or engineering calculations using emission design factors from EPA'sCompilation of Air Pollutant Emission Factors (cited in 40 CFR 61, Subpart H, Appendix D) andcoal content information provided by the coal supplier [47].

Most of the uranium used at PGDP was extracted from uranium ore and shipped to PGDP asuranium oxide. However, from 1953 to 1975, some reprocessed uranium, containing traces of otherradioactive materials, was fed into the cascade system [4]. The other materials included technetium99 (Tc-99), thorium 230 (Th-230), neptunium 237 (Np-237), and plutonium 239 (Pu-239).Significant quantities of Tc-99 were released to the air as early as 1953, with the largest annualrelease in 1958 [12]. Several reports and studies, from as early as 1957, describe operationalproblems caused by these other radioactive materials [69,70,71,72,73]. Airborne releases of Th-230, Np-237, and Pu-239 were not reported in the annual environmental reports until the 1990s butwere released in much less quantity than Tc-99. (Refer to Table 8B.) These constituents ofreprocessed uranium were in low concentrations when it was received and mostly concentrated inthe flame tower ash during the manufacturing of UF6.

Between 1952 and 1986, PGDP operated several secondary smelters to recycle scrap metals (otherthan uranium) in the C-746A facility. These metals included steel, nickel, aluminum, copper, monel(a copper-nickel alloy), cobalt, gold, and silver [74]. Only nickel and aluminum were smelted inlarge quantities: more than 11 million pounds of aluminum and about 37 million pounds of nickel[74, 75].

PGDP also has operated two vapor degreasers in Building C-400 and one in Building C-720 formetal cleaning and degreasing. Two systems used trichloroethylene (TCE) and one used 1,1,1-trichloroethane (1,1,1-TCA) as organic solvents and degreasers. Both chemicals vaporize orevaporate readily, so PGDP assumed that 90% of the TCE and 1,1,1-TCA was released to theatmosphere [48]. Use of TCE and 1,1,1-TCA was discontinued in 1993; therefore, ATSDRscientists evaluated only past airborne exposures from these sources. Operation of the Northwest andNortheast Groundwater Treatment Systems (beginning in 1995 and 1997, respectively) has alsoresulted in small releases of TCE to the atmosphere.

PGDP uses four recirculating water cooling systems to dissipate heat generated by the diffusionprocess. Moisture in the air flow (drifts) from PGDP's cooling towers contains elements found inthese recirculating water systems. These elements come from chemicals used as corrosion inhibitors,algicides, etc.; the corrosion inhibitor used at PGDP until 1993 was a chromate-zinc-phosphatecompound. Of the contaminants released from the cooling towers, the two with the greatest potentialenvironmental impact are hexavalent chromium and zinc. These were investigated by Oak RidgeNational Laboratory in 1978 [76]. At that time, the hexavalent chromium was detected onvegetation and in the soil at a distance of about 0.9 miles (1,500 meters) from the cooling towers(extending outside the PGDP boundary). We do not have the release quantities for hexavalentchromium before 1988, so we cannot compare releases in 1978 with quantities seen in the off-siteenvironment in 1978. Available data indicate that the highest annual quantity was released in 1992.In 1993, use of the chromium-zinc-phosphate anti-corrosion compound was discontinued [48].

PGDP operates a coal-burning steam plant to provide steam and for process operations and heating.In 1974 and 1975, two of the three boilers at the steam plant were converted to burn low-sulfur coaland oil instead of natural gas. Electrostatic precipitators with 97% efficiency for the capture ofparticulates were installed [48]. One boiler continues to use natural gas and oil. PGDP reportsreleases of sulfur dioxide, nitrogen oxides, particulates, carbon monoxide, non-methane volatileorganic compounds, and methane. Sulfur dioxide was continuously monitored at the steam plantstacks from 1979 until 1993 [77]. The reported results were based on the quantity released per unitof heat (BTU) produced or total released for the year. Other emissions were calculated from fuelusage and emission factors [47]. Annual releases have been reported in DOE's annual reports for1985 through 1993. No off-site air monitoring for sulfur dioxide and nitrogen oxides has beenperformed near the PGDP site.

Release quantities in Tables 8A and 8B were estimated by DOE, their predecessors, or theircontractors, mainly through material balance records or by engineering calculations. Values listedwith "est." in Table 8A are estimated by ATSDR using available information. Table 8A includesannual estimated release quantities for uranium and Tc-99. Table 8B includes annual estimatedreleases of other radioactive materials and chemicals from 1985 through 1996. The amount ofmaterial smelted in the C-746A facility are not included in these tables, but are listed in Appendix I.

Table 1.

Annual estimated release quantities of uranium and technetium 99 from process operations at PGDP for 1952 through 1993 and 1996 [12,68,4]
Year U (in kg) U (in Ci) U-234 (in Ci) U-235 (in Ci) U-238 (in Ci) Tc-99 (in Ci)
1952 30 0.02
est. 0.0056
est. 0.0003
est. 0.010
1953 500 0.25
est. 0.078
est. 0.0038
est. 0.167
1954 4,800 2.4
est. 0.75
est. 0.04
est. 1.60
1955 8,400 4.2
est. 1.31
est. 0.06
est. 2.81
1956 10,500 5.2
est. 1.62
est. 0.08
est. 3.50
1957 3,900 2.4
est. 1.10
est. 0.05
est. 1.20
1958 3,500 2.2
est. 0.98
est. 0.05
est. 1.17
1959 3,300 2.1
est. 0.93
est. 0.04
est. 1.10
1960 3,000 2.0
est. 0.94
est. 0.05
est. 1.00
1961 3,600 2.4
est. 1.12
est. 0.05
est. 1.20
1962 2,400 1.3
est. 0.45
est. 0.02
est. 0.80
1963 2,400 1.3
est. 0.45
est. 0.02
est. 0.80
1964 900 0.6
est. 0.28
est. 0.01
est. 0.30
1965 20 0.02
est. 0.01
est. 0.00
est. 0.01
1966 30 0.02
est. 0.0056
est. 0.0003
est. 0.01
1967 20 0.02
est. 0.01
est. 0.00
est. 0.01
1968 600 0.3
est. 0.11
est. 0.006
est. 0.20
1969 1,800 1.0
est. 0.34
est. 0.02
est. 0.60
1970 900 0.5
est. 0.17
est. 0.01
est. 0.30
1971 1,200 0.7
est. 0.30
est. 0.01
est. 0.40
1972 1,200 0.7
est. 0.30
est. 0.01
est. 0.40
1973 1,400 0.8
est. 0.35
est. 0.02
est. 0.47
1974 1,100 0.6
est. 0.17
est. 0.01
est. 0.37
1975 1,100 0.7
1976 1,500 1.0
1977 610 0.4
1978 96 0.06
1979 48 0.03
1980 22 0.02
1981 140 0.06
1982 300 0.14
1983 11 0.0045
1984 3.2 0.0019
1985 4.4 3.7E-03
1986 0.79 3.6E-04
1987 <1.0 2.9E-04
1988 0.14 0.6E-04
1989 0.2 2.9E-04
1990 0.03 3.37E-05
1991 0.005 6.63E-06
1992 1.42 2.12E-03
19931 3.06 3.19E-03
1996 ----- 4.37E-03
1 Uranium and uranium isotope values reported for 1993 were not consistent in the annual environmental report; maximum values were used.
Key: Ci = curies; kg = kilograms; est. = estimated; U = uranium; U-234, U-235, and U-238 = uranium 234, uranium 235, and uranium 238; Tc-99 = technetium 99

Table 8B.

nnual estimated release quantities of major airborne contaminants other than uranium and technetium 99 for 1985 through 1993 and 1996 [68,4]
Year Np-237
(in Ci)
(in Ci)
(in Ci)
(in kg)
Hexavalent Chromium
(in kg)
(in kg)
(in kg)
(in kg)
** Quantities were not reported in the documents reviewed.
Key: Np-237 = neptunium 237; Pu-239 = plutonium 239; Th-230 = thorium 230; TCE=trichloroethylene; SO2 = sulfur dioxide; NOx = nitrogen oxides; Ci = curies; kg = kilograms

Since 1993, when USEC became responsible for the process facilities, DOE has not reported processrelease information or off-site air monitoring data. DOE retains responsibility for four air emissionsources, which are the Northwest Groundwater Treatment Facility, the C-337 Cooling Tower (aspart of the Northeast Groundwater Treatment System), the cylinder refurbishment operations, andtwo separate fluorescent lamp crushers [78]. The groundwater treatment systems releasedapproximately 2 tons of TCE in 1997. Only the cylinder refurbishment operations require a permitfrom the Kentucky Division of Air Quality (KDAQ); these operations are the largest current sourceof non-radioactive air emissions [78]. Sandblasting of UF6 cylinders produced an estimated 4.5 tonsof particulates or dust in 1996 [1] and approximately 5 tons in 1997 [78]. Cylinder paintingreleased up to 3.4 tons of volatile organic compounds in 1996 [1] and 3.5 tons in 1997 [78]. Theseoperations are minor sources of hazardous air pollutants under the Clean Air Act, because they havelimited potential for public health effects. DOE is also responsible for four empty TCE tanks but hasno plans to use these tanks at this time [78].

In 1988 and 1989, KDAQ cited PGDP for excessive dust emissions from the C-726 sandblastingfacility [55,56]. The facility was shut down in May 1989. PGDP installed a dust collection and filtersystem, but the facility has been used infrequently since that time.

Contaminants of Concern

Each release of a contaminant to the air represents a potential human exposure. ATSDR scientistsused information about contaminant releases to identify and select possible contaminants of concernfor air exposure pathways. Contaminant concentrations in off-site areas are used to determinecontaminants of concern. Additional criteria used to select contaminants of concern were (1)maximum concentrations exceeding media-specific comparison values, (2) toxicity andradioactivity, and (3) community concerns. Modeling was used to estimate off-site air concentrationsfor those contaminants that did not have adequate off-site air monitoring. We compared modelingresults to ambient air monitoring measurements, when possible, in order to evaluate the accuracy ofmodel predictions.

In estimating the airborne release concentrations and the potential exposure doses for radioactivematerials other than uranium and Tc-99, we assumed that process operations released thesematerials into the air in the same proportion to uranium as in the materials shipped to Paducah fromother DOE facilities. Actually, most of these radioactive materials are removed in the ash residuewhen UF4 is converted to UF6. About 25% of the neptunium and trace amounts of the plutonium areconverted to hexafluoride compounds and processed with the UF6. Under the original, conservativeassumption, the contribution to the dose estimates from radioactive materials other than uraniumand Tc-99 would be at least an order of magnitude smaller than the contribution from uraniumisotopes, and would not add significantly to the dose estimates [71].

Zinc was released from the cooling tower with chromium. Zinc concentrations in foliage werestudied at Oak Ridge Gaseous Diffusion Plant (ORGDP), where zinc and chromium concentrationswere a little higher than at PGDP [79]. Beyond 660 feet (200 meters) from the base of the ORGDPcooling towers, zinc could not be differentiated from background levels. At PGDP, 660 feet from thebase of the cooling towers would still be on site. Also, a study of mature tree cross sections showedthat the zinc emissions were uniform over the past 20 years of cooling tower operations. Thisindicates that off-site air concentrations have not changed significantly. Based on the estimatedquantities of zinc emissions and the probability that zinc would not be seen off site at PGDP, zincwas not selected as a contaminant of concern.

Lastly, 1,1,1-TCA was not selected as a contaminant of concern, because the small quantityreleased to the atmosphere would not produce adverse health effects off site.

ATSDR scientists used computer modeling to predict chronic and acute off-site concentrations ofseveral contaminants known to be released from the PGDP site. These include uranium isotopes,Tc-99, uranium (as a chemical), HF, TCE, sulfur dioxide, nitrogen oxides, hexavalent chromium,and nickel and other metal particulates. For chemical contaminants, maximum off-siteconcentrations were compared to media-specific comparison values to determine whether thecontaminants should be selected as contaminants of concern for the air exposure pathways. For radioactive contaminants, ATSDR scientists estimated total committed effective doses.

Uranium Isotopes and Technetium 99

Historically, the PGDP site has released uranium isotopes (primarily U-234, U-235, and U-238) andTc-99 into the air. ATSDR scientists evaluated off-site exposures (committed effective doses) toairborne radioactive materials using the Clean Air Act Assessment Package-1988 (CAP88-PC)[80,81]. This computer model was developed by EPA for assessing regulatory compliance withEPA's National Emission Standards for Hazardous Air Pollutants (NESHAP). The model has anumber of adjustable parameters, which are discussed in Appendix E. Annual committed effectivedoses were based on air emissions of U-234, U-235, U-238, and Tc-99, and were calculated forchronic exposures using data from the highest-release years (1956, 1957, 1958, and 1959) andfrom 1996 (a recent year for which we have complete information). These estimated chronic doses(shown in Table 9) were calculated for the maximally exposed individual (the closest residentdownwind, about 0.9 miles--1,500 meters--north of the source; CAP88 assumes the source is inthe center of the site [30]). The total estimated annual committed effective doses for 1960 through1963 were also calculated; these were between 100 and 150 millirems per year (1.0 to 1.5millisieverts per year). The estimated committed effective dose from other constituents inreprocessed uranium could add up to 10% to these total doses.

The total estimated annual committed effective doses, shown in Table 9, were not compared to dosescalculated from ambient air monitoring data: the air monitoring stations were located at the securityfence perimeter, whereas the doses in Table 9 are estimated for the closest downwindresidence--again, approximately 1,500 meters north of the fence. Both calculations result inpotential exposures greater than 10 millirems per year, the current NESHAP emission standard forradionuclides (40 CFR 61). Therefore, these radionuclides were selected, as a group, ascontaminants of concern for the air exposure pathway.

For acute (short-term) exposure, the highest estimated off-site exposure occurred when a cylinderruptured in Building C-333 at 4:00 a.m. on November 17, 1960. According to our modeling of thisaccident (refer to Appendix E), an estimated uranium inhaled dose of 1.5 rems (0.015 sieverts)could have been received by the maximally exposed resident southeast of the site. Therefore, theuranium isotopes were selected as contaminants of concern for the air exposure pathway.

Table 9.

Off-site estimated annual committed effective doses from uranium 234, uranium 235, uranium 238, and technetium 99 air releases from PGDP to the maximally exposed individual (approximately 1,500 meters north of the source)1
Year Releases in curies (gigabecquerels) Total estimated annual committed effective dose2 from releases in millirems (millisieverts) [60]
U-234 U-235 U-238 Tc-99
1956 1.62 (59.9) 0.08 (2.96) 3.50 (129.5) 2.6 (96.2) 340 (3.40)
1957 1.10 (40.7) 0.05 (1.85) 1.20 (44.4) 4.8 (177.6) 156 (1.56)
1958 1.09 (40.3) 0.05 (1.85) 1.16 (42.9) 6.3 (233.1) 147 (1.47)
1959 0.93 (34.4) 0.04 (1.48) 1.10 (40.7) 5.1 (188.7) 132 (1.32)
1996 2.9E-03 (0.107) 1.2E-04 (0.044) 1.4E-03 (0.052) 3.6E-02 (1.332) 0.43 (0.0043)
1 Predominant wind direction from south-southwest.
2 Estimated using the Clean Air Act Assessment Package-1988 (CAP-88-PC) [81].
Key: U-234 = uranium 234; U-235 = uranium 235; U-238 = uranium 238; Tc-99 = technetium 99

Uranium as a Chemical

Accidental uranium releases were also evaluated for uranium as a chemical--that is, based on itschemical toxicity as a heavy metal, not its radioactivity. Only one reported accident (same as above)could have caused significant off-site uranium exposures: an accident at Building C-333 at 4:00a.m. on November 17, 1960. According to estimated exposures from modeling this accident (seeAppendix E), the highest potential dose of inhaled uranium was 5 to 20 milligrams (mg) forresidents living southeast of the site, 0.9 to 2.5 miles (1,500 to 4,000 meters) from Building C-333.The U.S. Nuclear Regulatory Commission's action level for intake of soluble uranium is 10 mg. (Atthis action level, residents may be instructed to evacuate or to stay indoors with windows closed.)Uranium as a chemical contaminant was selected as a contaminant of concern for the airexposure pathway.

Hydrogen Fluoride

HF (including fluoride and fluorine) was released at the PGDP site. This contaminant is also a notedcommunity concern. Although fluoride emission data from the plant are limited, there is a strongcorrelation between uranium and fluoride releases, and there is historical information on uraniumreleases. Current fluoride releases are much smaller than past releases, because process and filtrationequipment have changed and chemical manufacturing at the PGDP facilities has been discontinued.

Fluoride releases from the plant were not reported in annual reports until 1986. Results from off-siteair monitoring were reported in the annual environmental reports from 1958 until 1993. However,results for each monitoring location and sampling period were not reported. For the first 3 years, themedian value for all off-site air concentrations was reported. From 1961 until 1993, the mean valuesfor each year were reported [68].

Because there is a strong correlation between uranium releases and ambient air concentrations ofHF, ATSDR scientists assumed that the largest chronic HF release coincided with the highest annualuranium release in 1956. To evaluate off-site HF exposures for 1956, one must estimate or modelHF emissions from periods of consistent data reporting when the processes on site were similar (e.g.,1962 through 1970). The method we used is discussed in Appendix F.

For chronic HF exposures, the maximally exposed individual is assumed to be at the perimeter northair monitoring station. This station is closer to the fluoride processing facility than others [68], and isdownwind of the processing facility with respect to prevailing south-southwest winds [1].Kentucky's ambient air standard for average annual exposure is 500 parts per billion (ppb) [82],and ATSDR's provisional guidance for long-term exposure (365 days or more) is 10 microgramsper cubic meter (g/m3), or 12 ppb [83].

Modeling results indicate that long-term estimated HF concentrations at the north perimeter stationdid not exceed the Kentucky ambient air standard but exceeded ATSDR's provisional guidance in1955, 1956, and 1961. However, at 1 mile north of the perimeter, the concentrations did not exceedATSDR's provisional guidance. The estimated annual average airborne HF concentration for 1956at the nearest houses to the site was 22 ppb, which was above ATSDR's provisional guidance butapproximately 25 times lower than Kentucky's standard. 1956 was the only year that the estimatedconcentration exceeded the provisional guidance at this location.

Of the documented accidental HF releases, the largest release occurred on November 17, 1960,when a cylinder ruptured in Building C-333. During this accident, 8,074 kilograms (17,800pounds) of UF6 were released at approximately 4:00 a.m. As discussed in Appendix F, modeling ofthis accident estimates short-term hazardous HF concentrations more than a kilometer to thesoutheast of Building C-333, which would include property immediately off site. Therefore,hydrogen fluoride was selected as a contaminant of concern for the air exposure pathway.


Past operations at PGDP involved large quantities of TCE as an organic solvent and degreaser. Historical air releases of TCE were several orders of magnitude larger than current releases from the groundwater treatment facilities. Although significant amounts of TCE were released to the groundwater in the past, most operational releases of TCE volatilized into the atmosphere [48]. To determine if airborne releases presented a potential inhalation exposure to nearby residents, ATSDR scientists estimated the air dispersion of TCE using the Industrial Source Complex (ISC3) model [84], the maximum annual quantities of TCE released from the site, and very conservative assumptions about dispersion, plume rise, and atmospheric degradation. The model and assumptions are discussed in Appendix G.

The maximum estimated airborne TCE concentration at 1,000 meters (3,280 feet) north of BuildingC-400 is 112 g/m3 for a 1-hour averaging period (i.e., an acute exposure) and 3 g/m3 for anannual averaging period. Estimated concentrations were several times lower than the media-specificcomparison values for TCE in air (10,920 g/m3 for acute exposure and 546 g/m3 forintermediate-duration exposure) [24]. Therefore, TCE was not selected as a contaminant ofconcern for air exposure pathways.

Sulfur Dioxide and Nitrogen Oxides

Sulfur dioxide and nitrogen oxides are both released to air from the PGDP site. Off-site monitoringfor these contaminants has not been performed near the site; therefore, ATSDR scientists used theISC3 model to estimate off-site concentrations of sulfur dioxide and nitrogen oxides from the on-sitecoal-burning steam plant. Modeling results indicated that off-site estimated concentrations of sulfurdioxide and nitrogen oxides for chronic exposure are not likely to exceed comparison values.Therefore, sulfur dioxide and nitrogen oxides were not selected as contaminants of concern forair exposure pathways.


Although airborne releases of hexavalent chromium from the cooling towers at PGDP were notreported until 1988, the plant used hexavalent chromium since the early days of operations. From1988 until 1993, releases of hexavalent chromium from the cooling towers were estimated fromconcentrations in the cooling water and known annual quantities of chromium compound added tothe water. As early as 1958, hexavalent chromium was considered a potential environmentalcontaminant [86]; however, only surface water samples were analyzed for chromium.

Union Carbide studied chromium contamination in the late 1970s: they evaluated chromiumreleases from the cooling towers to assess the potential for chromium transport and accumulation inthe terrestrial environment [76,79]. Although the Union Carbide studies did not specifically addressairborne hexavalent chromium concentrations, they do provide information about deposition ofhexavalent chromium on vegetation and soil from the airborne releases. The studies also indicatethat chromium deposited from the drift cloud was present as hexavalent chromium; therefore, it isreasonable to assume that airborne chromium was also present in the hexavalent form. Hexavalentchromium was detected on vegetation and in soil at a distance of 1,500 meters (about 0.9 miles)from the towers in 1978. Beyond 1,500 meters, soil and vegetation concentrations could not bedifferentiated from background chromium concentrations [79].

According to release data from 1988 and later, the maximum annual release of airborne hexavalentchromium occurred in 1992. Therefore, ATSDR scientists used 1992 release data and the ISC3 airdispersion model [84] to estimate maximum air concentrations on site, immediately off site, and atthe closest downwind off-site residence for 1-hour, 8-hour, 24-hour, and annual exposures.Appendix H describes the model and the estimated exposure concentrations. Maximum off-site airconcentrations were estimated to be at least 100 times lower than ATSDR's comparison values forhexavalent chromium in air. Therefore, chromium was not selected as a contaminant of concernfor air exposure pathways.

Nickel and Other Metal Particulates

Because nickel is more toxic than the other metals smelted in the C-746A facility, the evaluation ofthese operations focuses initially on nickel emissions and potential off-site exposures. Appendix Idescribes the model and the estimated exposure concentrations. Estimated maximum hourly andannual nickel concentrations in areas of potential off-site exposure do not exceed health-basedcancer and non-cancer comparison values for acute or chronic exposures. As the toxicity of othersmelted metals is lower than nickel and/or the quantity of metals smelted is much lower, estimatedairborne concentrations of the other metal particulates are also below the comparison values for eachof these metals in air. Therefore, nickel and these other metal particulates smelted in the C-746Afacility were not selected as contaminants of concern for the air exposure pathways.

Airborne Exposure Pathways

ATSDR scientists identified completed and potential exposure pathways for past, current, andpotential future exposure to air contaminants. Contaminants of concern in these exposure pathwayswill be evaluated further in the public health implications section of this report.

Current Exposure

Off-site airborne radioactive material concentrations are currently being monitored by the KentuckyDepartment of Health's Radiation Control Program. Since monitoring began in 1996, noconcentrations of radioactive materials have been detected above emission standards [85]. ATSDR'sestimated committed effective dose for radioactive releases in 1996 is 0.43 millirems per year(0.0043 millisieverts per year)--more than 20 times smaller than the NESHAP requirement of 10millirems per year (0.10 millisieverts per year) whole body dose.

The current maximum estimated or modeled off-site concentrations of HF, TCE, sulfur dioxide,nitrogen oxides, and hexavalent chromium are low and do not exceed their respective health-basedcomparison values.

Therefore, ATSDR scientists did not identify any contaminants of concern for current airexposure pathways.

Past Exposure

For the air exposure pathway, radionuclides (uranium 234, uranium 235, uranium 238, and technetium 99), uranium ( as a chemical), and hydrogen fluoride were identified as contaminants of concern for past exposures. They will be discussed further in the public health implications section of this report.In the past, TCE, sulfur dioxide, nitrogen oxides, and hexavalent chromium were released from PGDP; however, the maximum estimated off-site concentrations did not exceed health-based comparison values. As discussed previously, radionuclides (U-234, U-235, U-238, and Tc-99), uranium (as a chemical), and HF were identified as contaminants of concern for past exposure via completed and potential air exposure pathways. These contaminants will be discussed further in the public health implications section.

Potential Future Exposure

As long as on-site processes remain similar to the current operations and no major accidentoccurs, future off-site releases of airborne contaminants should continue at current levels, whichdo not exceed health-based comparison values. Therefore, we did not identify any airbornecontaminants of concern for potential future chronic exposure. Future releases of airbornecontaminants from the processing facility will be the responsibility of USEC under the currentprivatization plans.

DOE is considering alternatives for the management of the aging depleted uranium cylinders atPGDP. Should any chemical processing or additional handling of the cylinders or storage of wastefrom remedial projects be done at this site, the potential for airborne releases will be part of theenvironmental impact considerations during the planning for the new operations. If on-site activitiesand operations change, then the potential for off-site exposure should be re-evaluated.

Table 10.

Summary of completed and potential exposure pathways for airborne contaminants of concern
Major Sources Radioactive Contaminants Point of Exposure Route of Exposure Exposed Population Period of Time Maximum Estimated Annual Committed Effective Dose
Process operations,
Bldg. C-310 stack,
Bldg. C-410 feed plant,
Bldg. C-340 metals plant
U-234, U-235, U-238

Downwind off-site ambient air (mainly north of the site) Inhalation Residents living within ~500 meters (~0.3 miles) north of fence 1954 to 1963

1955 to 1965; 1970, 1971, 1973 and 1974

Maximum in 1956:
340 mrem
(3.4 mSv)

Major acute releases:

1960 cylinder rupture


U-234, U-235, U-238


Downwind off-site ambient air
(Both accidents toward the southeast)




Residents outdoors within 1.5 to 3 kilometers (~0.9 to 1.8 miles) of accident, which happened at Building C-333


4:00 a.m.


0.5 to 1.5 rem
(5 to 15 mSv)

1962 fire U-234, U-235, U-238 Residents outdoors more than 0.2 kilometers (0.12 miles) from the accident 12/13/62
4:00 p.m.
< 1 mrem
(< 0.01mSv)
Major Sources Chemical Contaminants Point of Exposure Route of Exposure Exposed Population Period of Time Maximum Estimated Exposure Dose (Chemicals)
Process operations Hydrogen fluoride Downwind off-site ambient air (north of site) Inhalation Resident living within ~500 meters (~0.3 miles) north of fence 1956 only 22 ppb average hydrogen fluoride concentration

Major acute releases:

1960 cylinder rupture




Downwind off-site ambient air (toward the southeast)




Residents outdoors within 1.5 to 4 kilometers (~0.9 to ~2.5 miles) of accident, which happened at Building C-333


4:00 a.m.


5 to 20 mg uranium

Hydrogen fluoride 1 to 5 ppm hydrogen fluoride
Key: mg = milligrams; mrem = millirems; mSv = millisieverts; ppm = parts per million; ppb = parts per billion; U-234= uranium 234; U-235 = uranium 235; U-238 = uranium 238; Tc-99 = technetium 99

Next Section     Table of Contents The U.S. Government's Official Web PortalDepartment of Health and Human Services
Agency for Toxic Substances and Disease Registry, 4770 Buford Hwy NE, Atlanta, GA 30341
Contact CDC: 800-232-4636 / TTY: 888-232-6348

A-Z Index

  1. A
  2. B
  3. C
  4. D
  5. E
  6. F
  7. G
  8. H
  9. I
  10. J
  11. K
  12. L
  13. M
  14. N
  15. O
  16. P
  17. Q
  18. R
  19. S
  20. T
  21. U
  22. V
  23. W
  24. X
  25. Y
  26. Z
  27. #