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


Exposure Assessment of Airborne Nickel and other Metal Particulates from Historic Smelter Operations at the Paducah Gaseous Diffusion Plant



This health consultation uses air dispersion modeling and measured emissions to evaluate the off-siteconcentrations and potential health effects of airborne metal particulates emitted from secondarymetal smelting operations at the Paducah Gaseous Diffusion Plant (PGDP). Between 1952 and1986, the PGDP operated several smelters to recycle scrap metals. These metals included steel,nickel, aluminum, copper, monel (a copper-nickel alloy), cobalt, gold, and silver (DOE, 2000a). Ofthese metals, only nickel and aluminum were smelted in large quantities (more than 11 millionpounds of aluminum and about 37 million pounds of nickel; DOE, 2000a/b). Because nickel ismore toxic than aluminum, this evaluation will initially focus on nickel emissions and potentialexposures. Given that the dispersion and transport of the other recycled metals is basically the sameas for nickel, the concentrations of the recycled metals will be evaluated relative to the toxicity andestimated concentrations of nickel (airborne uranium emissions have been previously evaluated inthe PGDP Public Health Assessment). Estimated maximum hourly and annual nickelconcentrations in areas of potential off-site exposure do not exceed health-based cancer and non-cancer comparison values for either acute or chronic exposures (WKWMA workers and residents;respectively). As the toxicity of other smelted metals is lower than nickel (comparison values willbe higher) and/or the quantity of metals smelted is much lower, estimated airborne concentrations of the particulates of the other metals are also below levels of public health concern.


This health consultation addresses the concentrations and potential health effects of airborne metalparticulates emitted from secondary metal smelting operations at the Paducah Gaseous DiffusionPlant (PGDP) C-746 Facility. A concern about potential community exposures to airborne metalsfrom PGDP metal recycling was transmitted to the Agency for Toxic Substances and DiseaseRegistry (ATSDR) via comments on the public release version of the PGDP Public HealthAssessment (PHA; ATSDR, 2001). Although the results and conclusions of this health consultationwill be included in the pending final release version of the PGDP PHA, the issue of potentialexposures to airborne metals is addressed in this health consultation in order to allow additionalcommunity review and comment. The purpose of this health consultation is to determine whetherairborne metals released from PGDP smelter operations were present in areas of potential off-siteexposure at concentrations of public health concern. This consultation does not address airborneuranium releases from uranium metal operations from uranium smelting and recycling processeswhich were evaluated in the PHA (ATSDR 2001).

PGDP, which is about 10 miles (16 km) west of Paducah, Kentucky, began operation in 1952 toproduce enriched uranium. Between 1952 and 1986, the PGDP operated several smelters to recyclescrap metals. These metals included steel, nickel, aluminum, copper, monel (a copper-nickel alloy),cobalt, gold, and silver (DOE, 2000a). Of these metals, only nickel and aluminum were smelted inlarge quantities (more than 11 million pounds of aluminum and about 37 million pounds of nickel;DOE, 2000a/b). Because nickel is more toxic than aluminum, this evaluation will initially focus onnickel emissions and potential exposures. Given that the dispersion and transport of the otherrecycled metals is basically the same as for nickel, the concentrations of all recycled metals will beevaluated relative to the toxicity and estimated concentrations of nickel.

Nickel has a relatively high melting temperature (1455oC) such that it is very unlikely to occur ingaseous (vapor) form at normal atmospheric temperatures (ATSDR, 1997). Nickel (and othersmelted metals) is most likely to be released from the smelter stack in a solid or particulate form. The PGDP permit application states that nickel particulate emissions occur as nickel oxides (KDEP,1979).

Approximately 17 million pounds of non-radiologically contaminated nickel were smelted from1977 to 1982. From late 1983 to 1986, an additional 20 million pounds of radiologicallycontaminated nickel were smelted. No off-site nickel air concentration data were collected byPGDP or by the Kentucky Department of Environmental Protection (Garrison, pers. comm).Potential off-site exposures to nickel particulates released from the smelting operation are estimatedusing the Breeze Version of the US Environmental Protection Agency (EPA) Industrial SourceComplex (ISC) air dispersion model (Trinity Consultants Version 1.07, 1996; ISC3 Short Term). This model uses meteorological data from the Paducah-Barkley Airport (1989-91), and particulateemission and source data from the PGDP air discharge permit application (KDEP, 1979).

The smelter emission rate (nickel particulates) used in the air dispersion model is based on measuredemission rates described in the air discharge permit application (KDEP, 1979; APC 110B,Attachment V). Three trial runs provide an average emission rate of 160 g/hr (0.044 g/sec) basedon an average smelter charging load of 1520 kg/hr (3350 lbs/hr). Modeled emission rates wererounded up to a value of 0.05 g/sec. The 20 million pounds of nickel smelted from late 1983 to1986 is equal to an average of ~6.7 million pounds/year. This 3 year time frame represents theperiod of highest annual nickel output and will be used to evaluate potential air concentrations.

Although the specific operating schedule of the smelter is not known, an average annual productionof 6.7 million lbs/yr divided by the loading rate of 3350 lbs/hr equates to an operating schedule of 8hrs/day for ~250 days/yr. Variation in the operating schedule, such as 16 hrs/day for 125 days/yr,will not change the instantaneous emission rate nor significantly change the annual dispersedconcentrations. Some variation in short term concentrations may occur as a result of specificweather conditions during smelter operation. However, multiple model runs were conducted usingmeteorological data from three years (1989, 1990, and 1991) to ensure that the range of weatherconditions were evaluated for both short term (one and eight hour maximum averages) and longterm (annual average) concentrations. Meteorological data from the 1984 to 1986 years are notavailable on the EPA website (EPA, 2001b).

In addition to emissions from the PGDP smelters, nickel and other metal particulates were (and arestill) released from combustion of fossil fuels at the PGDP steam plant and the Tennessee ValleyAuthority (TVA) Shawnee facility. Instantaneous emission rates from the TVA facility areestimated from annual releases submitted to the Environmental Protection Agency Toxic ReleaseInventory (TRI; TVA, 2002). The 1999 TRI indicates that 159 pounds of nickel compounds werereleased to air, the 1998 release amount was 137 pounds. In order to account for yearly variations,this evaluation estimates an annual release of 170 pounds which, distributed over the course of anentire year, equals an emission rate of 0.0025 g/sec (released from a 244 m stack).

Airborne emissions released from the PGDP steam plant occur from two 34 m stacks that are inclose proximity to the each other and are modeled as a single source. Nickel emissions from thePGDP steam plant are estimated from maximum monthly coal and fuel oil usage (PGDP, 1986;3300 tons/month coal and 400,000 gallons/year for fuel oil) and standard emission factors (EPA,2001a). The estimated instantaneous maximum nickel emission rate from the steam plant (0.00009g/sec) is more than 500 times lower than smelter emissions. However, the steam plant was assumedto operate 24 hours/day, 7 days/week for an annual emission rate about 100 times lower than thesmelter emission rate.

The nickel releases from all sources are predominantly in a nickel oxide form (ATSDR 1997; EPA2001a). Particulate releases from the smelter are predominately about 30 microns in diameter withsmall fractions in the 10 and 20 micron diameter size ranges (KDEP, 1979). Nickel particulatesfrom the PGDP steam plant and TVA facilities are predominately less than 5 microns in diameter(ATSDR 1997; EPA 2001a).

The distance of the C746-A smelter building to the closest residence is more than 1000 meters (m; 0.6 miles). Estimated air concentrations 1000 m from the source will be used to evaluate potential residential exposures. Results of the air dispersion modeling are compared with the ATSDR health-based, non-cancer, comparison value (0.2µg/m3 chronic inhalation EMEG/MRL; ATSDR, 1997; 2002a) to determine if estimated long term air concentrations are at levels of health concern for residents. Potential cancer effects are evaluated by multiplication of the EPA unit risk for nickel subsulfide(1) of 4.8E-4 per µg/m3 (EPA, 2002) and the maximum off-site nickel concentration (and an exposure factor based on duration of exposure of nine years). Health-based comparison values are chemical concentrations that health scientists have determined will not cause adverse health effects, even when assuming very conservative exposure scenarios. The chronic nickel health comparison value used in this consultation is protective for people most sensitive to chemical-induced effects (ATSDR, 1997). This necessarily includes the exposure factors for children, such as increased breathing rates and lower body weights (relative to adults), that may exacerbate the potential health effects of exposure.

WKWMA workers may have been present in buffer areas at distances from 500 to 1000 m from building C746A. Concentrations at points 500 m from C746A will be used to evaluate potential worker exposures. WKWMA worker exposures are evaluated using the National Institute of Occupational Safety and Health (NIOSH) Recommended Exposure Limit of 15µg/m3 for soluble or insoluble nickel (8 hour time weighted average(2) (TWA); ATSDR, 1997). Nickel soil concentrations are compared with estimated air concentrations to determine if estimated nickel dispersion and deposition are reflected in nearby soils.

Activities of radiological contaminants from nickel samples ((Th 230, Th 232, U 235 and Tc 99;DOE, 2000b) and stack emission monitoring (U234, U235, and U238; DOE 1986) are used todetermine the quantity of radionuclides released during nickel smelting. Calculated quantities andrates of radionuclide emissions from nickel smelting are about 10,000 times lower than radionuclideemissions from other PGDP process operations that were used to calculate radiological exposures(ATSDR, 2001). Consequently, radiological emissions from the smelting operations did notconstitute a significant radionuclide source and will not change previously estimated releases or exposures.


The estimated average annual concentrations of nickel particulates (based on 1984-1986 emission data and hourly weather data for 1990) are shown on a site map in Figure 1. The highest estimated annual average nickel concentration of 0.18µg/m3 occurred approximately 200 m north of the smelter building which reflects prevailing southerly winds. The maximum one hour average concentration is estimated to be about 15µg/m3 and the maximum eight hour average is estimated to be about 3.0µg/m3 with both maxima about 200 m north of the stack. Distribution of short term nickel concentrations, although higher to the north, are more uniformly distributed around the C-746-A building and reflect short term weather conditions which may have winds from any direction. Model runs using hourly weather data(3) for the years 1989, 1990 and 1991 produced essentially identical results for both short term (hourly and daily) and long term (annual) nickel dispersion. All of the above maximum air concentrations occurred at areas within the DOE property.

The estimated nickel air concentrations at areas of potential off-site exposure (listed in Table 1)includes several notable trends. First, the short term concentrations are much higher than the longterm concentrations. This trend reflects how short term meteorological conditions control transportand how averaging time affects concentration. One hour maxima are significantly greater than theeight hour maxima and much greater than the annual average concentrations. A second trend is howoverall concentrations drop very rapidly with distance away from the stack. At 1000 m, which is thedistance to the closest residence, all concentrations are about an order of magnitude lower thanconcentrations at 500 m. The third significant trend is the similarity between concentrations anddistributions for different weather years which indicates these data adequately capture the range ofpotential meteorological conditions.

The dispersion modeling indicates that most nickel deposition will occur within 500 m of the smelterbuilding. Measured soil nickel concentrations support the model results. Average soilconcentrations at locations within the security fence average 25 ppm. Areas within the buffer zone(on DOE property, but outside the fence) and off-site areas have statistically similar averages of ~12ppm. These results indicate that onsite soils have elevated nickel concentrations relative to off-siteareas but that there is no significant difference between buffer area and off-site concentrations. Local nickel air concentrations and deposition to soil from the TVA plant may have been higherprior to the increase in stack height in the early 1970's (Garrison, 2001). Potential exposures to allmetals in soil are evaluated in the PGDP public health assessment (ATSDR, 2001).

The highest average annual nickel concentration at 1000 m from the smelter building is 0.004 µg/m3 which is about 50 times lower than the ATSDR health-based comparison value of 0.2 µg/m3. Estimated long term exposures at 1000 m also represent no potential cancer risk(4). Similarly, the maximum short term nickel concentration of 1.1 µg/m3 (at 500 m from source; maximum one hour average) is more than 10 times lower than the worker exposure limit of 15 µg/m3 (which includes consideration of potential occupational cancer risk. The estimated maximum hourly, daily, and annual nickel air concentrations from PGDP smelter operations, with potential contributions from the PGDP steam plant and TVA facility are below levels of health concern.

These estimated off-site nickel concentrations in air include potential contributions from the PGDPnickel smelter emissions, the PGDP steam plant nickel emissions, and the TVA Shawnee coal plantnickel emissions. Modeled emission rates from the PGDP steam plant were based on healthprotective assumptions concerning fuel usage, operating schedules, and emission factors. Actualemission rates are probably lower than modeled rates. All estimated nickel air concentrations,calculated using health protective model assumptions, in areas of potential off-site exposure aremore than 10 times lower than relevant health comparison values.

Table 1.

Maximum estimated nickel air concentrations for different weather years and distances. Concentrations are in µg/m3, distances are meters (m) due north of smelter stack.
  500 m 1000 m 1500 m
1989 Annual Average 0.018 0.003 0.001
1990 Annual Average 0.022 0.004 0.002
1991 Annual Average 0.018 0.003 0.001
1989 1 hour maximum 1.0 0.12 0.05
1990 1 hour maximum 1.0 0.15 0.07
1991 1 hour maximum 1.1 0.17 0.06
1989 8 hour maximum 0.25 0.05 0.02
1990 8 hour maximum 0.23 0.05 0.02
1991 8 hour maximum 0.25 0.05 0.02

Table 2 lists the metals smelted at PGDP along with the estimated total amount of each metal andthe relevant inhalation health comparison values for each metal. The health comparison values forresidential and occupational exposure are a measure of the toxicity of each metal. The most toxicmetals have the lowest health comparison values because it takes a lower amount of the substance tocause sickness or adverse health effects. This evaluation focuses on nickel because it is relativelytoxic (a low health comparison value) and very large quantities of the metal were smelted resultingin the highest potential off-site air concentrations.

As the smelting of other metals will release particulates in proportion to the total amount of materialsmelted and the dispersion of those metals will follow the same dispersion patterns as nickel, itfollows that the off-site concentrations of other metal particulates will be proportional to the amountof metal smelted. Based on the relative toxicities of the various metals and the amounts of eachmetal smelted, off-site concentrations of all metals smelted at PGDP are below levels of healthconcern.

ATSDR considered interactive effects (cumulative, additive, synergistic, and antagonistic) ofchemicals following exposure to multiple chemicals to the extent of the scientific knowledge in thisarea.

  • Cumulative effects (the effects associated with concurrent exposure by all relevant pathways and routes of exposure to a group of chemicals that share a common mechanism of toxicity) were addressed on pages 93 and 94 and in Table 22 of the PHA (ATSDR, 2001).

  • Additive effects (the situation in which the combined effect of two chemicals is equal to the sum of the effect of each agent given alone) were considered for radioactive materials in the PHA (ATSDR, 2001; pages 119-120). Of the metals evaluated in this consultation only nickel and monel are considered to have additive effects (because monel is a nickel alloy). However, the amount of monel smelted and emitted is insignificant with regard to the amount of nickel smelted.

  • Existing information is inconclusive with regard to potential Synergistic effects (the situation in which the combined effect of two chemicals is much greater than the sum of the effect of each agent given alone) for the metals evaluated in this consultation.

  • Antagonistic effects (when a chemical reduces the toxicity or uptake of another chemical) were not considered in order to maintain a health-protective screening scenario even though concurrent exposures to magnesium or zinc reduces the toxicity of nickel.

ATSDR has reviewed the scientific literature surrounding chemical interactions and noted that if theestimated exposure doses for individual contaminants detected at the site are below doses shown tocause adverse effects (No Observed Adverse Effect Level; NOAEL), then ATSDR considers that thecombined effect of multiple chemicals is not expected to result in adverse health effects. It should benoted that typical environmental exposure doses to both carcinogenic and non-carcinogenicchemicals are more than 1000 times lower than laboratory-induced effect threshold doses. Thisapproach to chemical interactions is based on the results of numerous studies:

  • Several animal and human studies (Berman et al. 1992; Caprino et al. 1983; Drott et al. 1993; Harris et al. 1984) have reported thresholds for interactions. Studies have shown that exposure to a mixture of chemicals is unlikely to produce adverse health effects as long as components of that mixture are detected at levels below the NOAEL for individual compounds (Seed et al. 1995; Feron et al. 1995).

  • The absence of interactions at doses 10-fold or more below effect thresholds have beendemonstrated by Jonker et al. (1990) and Groten et al. (1991). Specifically, in two separate subacute toxicity studies in rats (Groten et al. 1997; Jonker et al. 1993), adverse effects disappeared altogether as the dose was decreased to below the threshold level.

  • For carcinogens, the interactions are more difficult to quantify due to the large study size (humans or animals) needed for statistical significance at the low doses observed in environmental exposures . In an animal study, Takayama et al. (1989) reported that 40 substances tested in combination at 1/50 of their cancer effect level (CEL) resulted in an increase in cancer. However, Hasegawa et al. (1994) reported no increase in cancer when dosing animals at 1/100 of the CEL for 10 compounds.


Estimated maximum hourly and annual nickel concentrations in areas of potential off-site exposuredo not exceed health-based cancer and non-cancer comparison values for either acute or chronicexposures (WKWMA workers and residents; respectively). As the toxicity of other smelted metalsis lower than nickel (comparison values will be higher) and/or the quantity of metals smelted ismuch lower, estimated airborne concentrations of the particulates of the other metals are also belowlevels of public health concern. Past emissions of metal particulates from PGDP smelter operationsin building C-746-A posed no apparent health hazard. "No apparent health hazard" means thatpeople may have been exposed to metal particulates but that such exposure is not expected to causeany adverse health effects. As metal particulates from smelter operations did not occur atconcentrations of public health concern, recommendations are not warranted.

Table 2.

Comparison of amounts of various metals smelted at PGDP with Inhalation Health Comparison Values (in micrograms per cubic meter; µg/m3).
Metal Estimated Amount of Metal Smelted at PGDP (pounds) 1 Health Comparison Value (Inhalation)
Aluminum > 11,500,000 pounds over more than 21 years 2 8 hr TWA3- 5000 µg/m3
Cobalt 71,500 pounds over 4 years 14-365 day MRL4- 0.03 µg/m3
8 hr TWA- 50 µg/m3
Gold ~5284 pounds over 21 years Not Available
Lead >1,258,990 pounds (no dates) 8 hr TWA- 50 µg/m3
Monel (copper-nickel alloy) 882,440 pounds over 4 years see Nickel
Nickel ~37,000,000 pounds over
6 ˝ years
>365 day MRL- 0.2 µg/m3
8 hr TWA- 15 µg/m3
Silver ~9,630 pounds over 8 years 8 hr TWA- 10 µg/m3
Steel 22,000 pounds (process terminated after test runs) 8 hr TWA- 5000 µg/m3 (as iron oxide dust/fumes)
1 Estimates of total amount of metals smelted are derived from DOE (2000a; 2000b).
2 Internal DOE (or AEC) memos and letters indicate, but do not prove, that aluminum smelting at PGDP began in the 1950's, but that in early years was limited to laboratory and bench testing of methods and processes that were instituted in production scale operations in the 1960's.
3 TWA- is the 8 hour time weighted average occupational exposure limit established by the National Institute of Occupational Safety and Health (NIOSH, 2002).
4 MRL- is the minimal risk level for residential exposures established by ATSDR.

Average annual airborne nickel concentration estimated using EPA ISC Short Term air dispersion model
Figure 1. Average annual airborne nickel concentration estimated using EPA ISC Short Term air dispersion model (1984-1986 emission data; 1990 weather data). Site fence is the PGDP security fence (blue area). Annual nickel concentrations include emissions from PGDP smelter and steam plant and TVA plant. Grid origin (0,0 location) is C-746-A smelter stack.

(NOVEMBER 6, 2001)

The following comments have been received by ATSDR in response to the public release version ofthis health consultation (dated November 6, 2001). These comments have elicited severalmodifications of the original evaluation. Potential source contributions from the PGDP steam plantand TVA Shawnee Facility have been added to the dispersion model. Clarifying information andreferences have been added to the consultation as indicated in each response. The comments areincluded verbatim, although names and personal identifiers have been redacted. ATSDR responsesare in bold type to distinguish community comment from ATSDR responses.

Commentor 1:

I read the recently released Health Consultation titled Exposure Assessment of Airborne Nickel and other Metal Particulates from Historic Smelter Operations at the Paducah Gaseous Diffusion Plant and have the following observations and recommendations for your review.

  1. Item one: Page one paragraph 2--"Between 1952 and 1986 the PGDP operated several smelters to recycle scrap materials"
  2. Comment: During the referenced time frame the PGDP did operate smelters in the C-746-A building for recycling materials but most of the scrap was processed to destroy or disguise the classified nature of the scrap. Another smelter operated in the C-340 complex to produce uranium alloys for use by the DOD and AEC. Though it is presented as a total review of smelting operations at the PGDP I found no mention of the uranium smelter in the document. The processing of other materials is very important since some materials handled in these units was far more dangerous (e.g. organics, beryllium components and uranium) and the cumulative problem presented by all operations must be studied and quantified. Were conditions "beyond the fence" terribly bad? I don't know but I had hoped that those with the expertise to evaluate the situation would be given all available information by the DOE.

    ATSDR Response:
    This consultation specifically addresses releases from the C-746-A smelters. Appropriate revisions have been made to clarify the purpose of the health consultation. Releases from uranium metal operations at the C-340 complex were evaluated as part of total uranium releases from the entire PGDP facility and are included in the PGDP Public Health Assessment (ATSDR, 2001).

    As you have indicated there was a uranium metal production operation in building C-340. 1985 annual uranium emissions from the C-340 operations are provided in Table 1 of the 1985 Environmental Monitoring Report (DOE, 1986). Annual emission values for U234, U235, and U238 are 7.1E-6, 3.3E-7, 2.6E-5 Ci/yr, respectively. As these values are all at least 10,000 times lower than the emissions from other previously evaluated sources (C-310 stack, C-400 group, seal/wet air exhaust, C-710 laboratory) for earlier years (1956-59), they do not comprise a significant additional source of radionuclide emissions (see Table E-1 in the PGDP Public Health Assessment for uranium values evaluated for specific release years). For example, the previously evaluated 1956 annual releases included 1.6 Ci/yr for U234, 0.08 Ci/yr for U235, and 3.5 Ci/yr for U238.

    As the much higher 1956 releases did not result in off-site exposures of public health concern, the measured 1985 emissions from uranium metal production in C-340 are below levels of public health concern. We will make appropriate revisions in the Public Health Assessment document to ensure that potential releases from C-340 uranium operations are explicitly referenced even though such emissions do not constitute a significant source.

  3. Item two: Page one paragraph 3--"The PGDP permit application…emissions occur as nickel oxides."
  4. Comment: As this entire review appears to be based on the PGDP permit application and ingot samples (addressed under item 5) and since this report is apparently going to be used by the DOE as the "there are no problems from smelting" banner, we must make sure that each data source is as accurate as possible. While it is probably true that nickel particulates reaching the final stack of the C-746-A induction furnace would have been dominantly nickel oxide, releases from other portions of the smelter were composed of other nickel compounds. For example all nickel was run through a large calciner to convert the chemical composition of the scrap before it was fed to the smelter. Particulates and gas from this process were released to a chemical trap system that exhibited a less than stellar ability to capture materials.

    ATSDR Response: The Commentor is correct in indicating that the entire smelting process included several operations in addition to the smelting furnace. Scrap nickel was subjected to physical grinding in order to produce a small, uniform feed material. According to air permit documents and personal communications (Steve Shell, 2002) this process occurred in enclosed containers and emissions were subject to filtration before release to the atmosphere. After grinding, the nickel chips were processed through the calciner to remove hydrogen fluoride (HF). This process was designed to remove gaseous HF and did not release particulate materials. Although both of these processes may contribute to contamination of the on-site processing facilities, only the smelting furnace emissions served as a significant source of nickel particulates for off-site dispersion.

  5. Item three: Page one paragraph 4---"No specific nickel air concentration data were collected.."
  6. Comment: This statement is very important for two reasons. First, it should made clearer through the rest of the document (other instances cited below) that all conclusions are based on modeling which may or may not have been based on valid assumptions. Second, it is widely understood (perhaps as a point of urban legend) at the PGDP that orders for shutting down activities at the C-746-A smelter come down from high ranking DOE folks in DC that were concerned with air emissions…. simply put, it is unlikely that this facility would operated at the PGDP without air emission data being internally collected,….. period. Although records in the building had been largely and obviously purged by the time I conducted research in the building in 1994(there were file cabinets filled with hanging folders with no contents), I did manage to fill two boxes with information which might contain references to air sampling. These two boxes were placed in the document vault in Kevil Kentucky (now under the control of Bechtel Jacobs) and should contain information that would at least point a researcher in the right direction. In addition, several of the folks that managed the operation are still available to discuss if and what data was collected.

    ATSDR Response: We will review the document to ensure that all conclusion statements explicitly indicate that the results are referenced as modeled (or estimated) air concentrations. In the absence of directly measured air concentrations, modeled data represent the best available information. It is also important to point out that this consultation does use measured stack concentrations as the basis for determining the source or emission concentrations which is the source term for the dispersion modeling. We have evaluated extensive amounts of air monitoring data. However, the available data do not include ambient nickel analyses. Even if such data did exist, it is very unlikely that the monitor locations would have been appropriately located so as to assess the spatially restricted areas affected by emissions from the nickel smelter. As our modeled ambient air concentrations are based on measured stack concentrations and are consistent with measured soil concentrations, we are confident that the results of the evaluation are protective of public health.

  7. Item four: Page one paragraph 4--- " This model uses...and source data from the PGDP air discharge permit application (KDEP, 1979)."
  8. Comment: It is unlikely that the permit application was totally forthcoming, though I will try to get a copy and check it out.

    ATSDR Response: According to the information available (KDEP, 1979), stack concentrations were measured for three separate trial runs of the batch smelter using an average of 96% of the maximum charging capacity. The average annual smelter output was ~72% of the annual maximum charging capacity. Therefore, the trial runs were conducted using conservative estimates of operating capacity. ATSDR used an estimated instantaneous nickel emission value that was 13% greater than the measured average trial run value. Overall, these emission and production values indicate that ATSDR employed health protective estimates of annual production and instantaneous emission rates in estimating airborne nickel concentrations to off-site areas.

  9. Item five: Page three paragraph 3---"Activities of radiological contaminants. …from nickel samples (DOE, 2000b) are used to determine the quantity of radionuclides released during nickel smelting".
  10. Comment: Using the level of radioactive materials in the nickel ingots is completely invalid. If this was information provided by DOE then I am deeply saddened and insulted as member of the public. It was demonstrated in lab tests conducted before the major melting campaigns began that radioactive materials were highly concentrated at the top of the melt (subjecting it to greater release than from the body of the melt) and subsequently removed as slag. The highly radioactive nature of the slag was the subject of internal documents and safety guidance. At a minimum the worst-case concentration should be used to see if there might have been a problem (perhaps modeling the elevated concentrations in the slag might show there were no problems at all).

    The odd thing about some of the slag was it's tendency to roll onto the surface of the melt in the form of golf ball sized spheres which were sometimes kept as souvenirs on worker's desk. A more accurate estimation of releases could be made by having these spheres tested (they are still around, no pun intended) or better still performing a mass balance (drums of flakes were weighed and sampled before and ingots were tested after they were cast so total material lost could be estimated). [redacted name] in the C-710 building closely studied distribution of materials within the melting process and his documents detail his findings (a search of the database using his name as the author should kick out some good information, e.g. K/TL 338-339-443 and K/PS230). If nothing is easily available I could provide some of the analytical result from the testing done.

    ATSDR Response: The Commentor is correct that radionuclide concentrations in the nickel buttons underestimate the total radionuclide concentration of the feed material due to segregation in the slag and nickel ingots. However, as stated in the above response to item 1, radionuclide stack emissions were also directly measured. Annual emissions from the C-746-A stack are reported for the calender year 1985 (DOE, 1986). As these values are all at least five orders of magnitude lower than the emissions from other previously evaluated sources for earlier years (1956-59), they do not comprise a significant additional source of radionuclide emissions (see Table E-1 in the PGDP Public Health Assessment for uranium values evaluated for specific release years).

  11. Item six: Page three paragraph 4---"The average annual concentrations of nickel particulates (based on 1984-1986 emission data…."
  12. Comment: According to Page one, no air concentration data were collected??? If quoted statement is based on data please cite the source(s) and add it to the reference list at the end. If this is based on their permit application and production rates for the time frame please clarify. The referenced map also lists the same information sources. I certainly hope that the most significant output from this effort (the map and conclusion that there is not any problem present) is not simply based on a model constructed from a KDEP application when better data is available. The better data should be demanded from DOE before the ATSDR puts their final stamp on this thing.

    ATSDR Response: The section containing the cited paragraph is entitled "Evaluation of Modeled Concentrations of Airborne Metals Particulates." While this clearly indicates that the data in question are model results, we will add the term "modeled" or "estimated" to the cited sentences. The comment is not correct in stating that "better data" are available which should be demanded from DOE. As historic off-site airborne nickel concentrations were not collected for the express purpose of quantifying potential off-site exposures, any additional data will be subject to inherent uncertainty which limits its utility for exposure assessment. The best response to such uncertainty is to seek consistency from different types of data. Because our modeled off-site concentrations are based on measured stack concentrations and are consistent with measured soil concentrations, we are confident that the results of the evaluation are protective of public health.

  13. Item seven: Page 3, footnote one---"Some nickel compounds, such as refinery dust …are known or probable human carcinogens (ATSDR, 1997). Nickel oxides, which is the chemical form released….".\
  14. Comment: I suggest changing the wording to "Nickel oxides, which is the dominant chemical form reported to have been released (citation)". What it appears by the statement in the document is that it is a fact that only nickel oxides were released. It is quite likely that material more similar to refinery dust was released in the processing (shredding and calcining) stages and I wouldn't be amazed if the smelting process itself produced some interesting nickel compounds given the composition of the metal and observations made by former operators.

    ATSDR Response: The suggested revision has been incorporated into the health consultation. It is important to note the distinction between emissions from primary and secondary (such as PGDP) nickel smelting operations. More than 90% of emissions from secondary smelters are in an oxide form (ATSDR, 1997). Refinery dust (primarily nickel subsulfide) is more typically produced during primary nickel smelting operations (i.e., during the production of nickel metal from nickel ore). Nickel operations at PGDP may have produced limited quantities of the non-oxide forms. However, from a quantitative perspective, most PGDP emissions were in an oxide form.

  15. Item eight: Page 5, paragraph one---"As the toxicity of other smelted metals is lower than nickel…"
  16. Comment: Since the emissions of uranium from the uranium furnace were apparently not reviewed in this project and there were other nastier materials (metals and organics which would flash off before complete combustion) fed to the C-746-A furnace than is mentioned in the text, this is a rather bold statement to be made, even though it is fairly well documented that nickel was the largest in volume by far the absolute nature of this statement should be reduced.

    ATSDR Response: Information supporting and clarifying this statement has been added to the consultation. The toxicity of all of the metals smelted was assessed using health comparison values and quantities of smelted metals. These comparison values are stated in terms of a dose or contaminant mass per unit body weight per unit time (mg/kg/day). The more toxic metals have lower comparison values, i.e., it takes less of the metal to cause illness. As nickel was by far the most toxic metal that was smelted in quantity, it was used as a benchmark for assessing levels of significant exposure. Cobalt was the only smelted metal with a higher toxicity than nickel. Based on ATSDR and EPA health comparison values, cobalt is approximately 10 times more toxic than nickel. However, only about 38,000 pounds of cobalt were smelted which is about 1700 times lower than the 6.7 million pounds of annual nickel production. Because smelting operations of relatively toxic nickel did not produce off-site concentrations at levels of health concern, smelting of much smaller quantities of other toxic metals will also be below levels of health concern. As indicated in the response to item one, uranium releases and potential health effects have been evaluated in the PHA (ATSDR, 2001).

  17. Item nine: Page 5, paragraph one---" Because the estimated concentration of all metal particulates from PGDP are below health concern, no adverse health effects are likely from potential residential or WKWMA worker exposures. As metal particulates from smelter operations are not identified as contaminants of concern and did not occur at concentrations of public concern…."
  18. Comment: I did not see a section in the text wherein the concentration of "all metal particulates from PGDP" was reviewed. It is the cumulative affect of PGDP sources on the surrounding environment that would have affected it negatively. I am very much interested in the review of the cumulative affect of materials released from the PGDP. Back reference again to such broad references being made based on a permit application and samples taken from a final "cleaner" product. It should be noted that it has not been determined definitively that metal particulates "did not occur at concentrations of public health concern" since the conclusion is based on review of a model and that model is based on data that is not entirely correct.

    ATSDR Response: ATSDR considered cumulative, additive, synergistic, and antagonistic effects of exposure to the extent of scientific knowledge in this area. Information and references related to ATSDR's process of evaluation have been added to the consultation.


A valiant effort was made given the amount of information available/provided to theresearchers preparing the report. Additional effort needs to be made to add the samplinginformation available at DOE to the review and models. A huge amount of process informationwas removed from C-746-A at some time between 1987 and 1994. Managers of the smeltercould indicate what types of records were kept and document archivists for the site couldindicate where this information was taken. Some information was removed from the buildingin 1995 and uniquely "stored" in drums destined for disposal.

Studies such as these need to be put together to ascertain the cumulative affect of all of thesources of contamination and see if off-site residents would have been within the radius ofinfluence. Doing this in a location-by-location manner is akin to informing the family of atorture victim that everything should be fine because each of the thousand cuts was studied and found to be insignificant.

ATSDR Response: See above for response to assessing interactive effects to differentcontaminants. The PGDP Public Health Assessment does assess potential cumulative dosesand health effects to community members that may be exposed to contaminants throughmultiple pathways. Potential exposure doses to nickel (and other metals) were estimatedfrom ground water, surface water, soil, and food items. None of these other pathwaysresulted in nickel exposure doses of public health concern. It is also important to note thathow the nickel is taken into your body affects the toxicity. Inhaled nickel, via airborneparticulates, is much more toxic than the same amount of nickel ingested with food orwater. Also, nickel particles larger than 5 microns are not considered respirable (ATSDR,1997) and are routed through the gastro-intestinal system rather than the lungs. As theestimated inhaled nickel dose is below a level of public health concern, adding a minoramount of less toxic ingested nickel will not significantly increase the total estimated nickeldose.

Commentor 2:

I thank you for the opportunity to comment on the Health Consultation, "Exposure Assessment ofAirborne Nickel and other Metal Particulates from Historic Smelter Operations at the PaducahGaseous Diffusion Plant." (Public Comment Release- November 6, 2001) I am pleased that theinvestigation found no cause for public concern. I have the following comments on the report.

It would be helpful to have a statement of no public concern near the beginning of the report. Isuggest adding to the first paragraph- "Concentrations of concern were not found and norecommendations are warranted."

The rationale for not including metal particulates from the TVA Shawnee Steam Plant is not clear. While the emissions of the steam plant and smelter stacks will not be additive during any shortperiod, the annual average concentrations would be additive. The lack of any significantcontribution from the steam plant may be deduced from the results of the soil sampling. The 12ppm nickel found in off-site soils probably represents deposition from steam plant effluents duringits nearly 50 years of operation, during many of which the stack heights were only about 60 meters. The significantly higher soil concentrations on the PGDP site is evidence that the contribution,however small, of the nickel smelter to airborne concentrations was significantly higher than thatfrom the steam plant and that from the steam plant may be ignored.

ATSDR Response: A summary has been added to the revised document. We have alsoreconfigured the ISC air dispersion model to include potential contributions from bothPGDP steam plant and the TVA facility. Based on 1998 and 1999 Toxic Release Inventorydata for the TVA facility (TVA, 2002, theinstantaneous nickel air emissions average ~0.0025 g/sec. Assuming that those later yearsare similar to the years of PGDP nickel smelter emissions, the instantaneous TVA release isabout 20 times lower than the PGDP smelter release (0.05 g/sec) and is dispersed from an800+ foot stack. Similarly, using standard emission factors for coal and oil combustion andannual fuel usage rates, instantaneous nickel emissions from the PGDP steam plantaveraged 0.00009 g/sec, which is 500 times lower than the smelter emissions (dispersedfrom two adjacent 110 foot stacks). The results of the revised air dispersion modelsindicate no differences in the short term or annual distributions of nickel. All estimatedconcentrations are still well below levels of health concern.

Commentor 3:


  1. Kentucky Environmental Oversight: List of Hazardous Waste and Radionuclide ParametersCurrently Analyzed and Monitored by the Commonwealth of Kentucky at the PaducahGaseous Diffusion Plant (Name Redacted) includes nickel, nickel compounds, nickelcarbonyl, and nickel cyanide. What PGDP operation(s) produces this waste stream? Didthis Health Consultation and/or the Public Health Assessment for PGDP include this sourceof nickel? Is this nickel monitored in air, soil, or water?

ATSDR Response: ATSDR has reviewed all of the data collected by or through theKentucky Federal Facility Oversight Unit that they have provided to ATSDR. Although thedata sets do contain some measurements of nickel (or nickel compounds) in various media,we have not reviewed any specific nickel cyanide or nickel carbonyl measurements in theavailable data. Several different PGDP process operations may result in nickel emissions. As indicated in the Health Consultation, past smelting operations and fossil fuel combustion release(d) nickel and nickel oxides. Limited quantities of nickel carbonyl or nickel cyanideas liquid effluents may be released from metal cleaning/degreasing operations in C-400building. However, these compounds are not stable in the environment (nickel carbonyl is aliquid at ambient temperatures) and will quickly be transformed to more stable compounds( nickel hydrates in water or nickel ferrite in soil; ATSDR, 1997).

In determining the contaminants of concern for the Public Health Assessment, ATSDR didnot attempt to correlate specific sources of nickel for each environmental media. The PHAevaluation was based on the measured concentrations of the contaminants in various media,such as soil, water, and sediment. The rationale for conducting this health consultation isthat we did not include metal particulates released from smelting operations in assessingairborne contaminant distributions. Nickel, as dissolved or total nickel, is monitored insurface water, ground water, and soil. This Health Consultation used modeled airbornenickel concentrations to estimate off-site distributions because nickel was not monitored atoff-site locations.

  1. Phase II Independent Investigation of the Paducah Gaseous Diffusion Plant ....February2000, section 3.2.5 smelting, refers to scrap metal recovery records being "...removed byanother DOE team investigating scrap metal recovery at the plant." Has ATSDR reviewedthose records? Additionally, the same section of Phase II refers to "a 1972 study ofradionuclides in scrap..."; did ATSDR review that study? Further, a chart on page 74 ofPhase II indicates smelting activities at C-746 (A), PGDP, from the 1950's to 1990. ThisHealth Consultation only considers nickel smelting from 1977 to 1986; was nickel smeltedat PGDP at any other time? Thus, was the 37 million pounds smelted during that periodthe total amount of nickel ever smelted at PGDP?

ATSDR Response: ATSDR has reviewed a number of documents relating to various aspectsof the aluminum and nickel smelting processes including the referenced DOE report on the"Metals Recovery Program" (DOE, 2000a). While that document does discuss the types ofdocuments that were reviewed, neither that document or the Phase II Report (DOE, 2000c)contain specific references or bibliographies. Consequently, ATSDR cannot determine ifwe have reviewed those same documents. However, ATSDR did review, and the HealthConsultation does reference, the report produced by that investigative team (as well as thereport of another team investigating related issues; DOE, 2000a; DOE, 2000b). ATSDR hasnot reviewed the "1972 study of radionuclides in scrap." Although, it is mentioned, asstated by the commentor, in the Phase II Report, it is not specifically referenced so ATSDRhas no way of finding or tracking that document.

As stated in the Health Consultation, smelting activities in C-746 did apparently commencein the 1950's. However, the summary statement describing the years of smelting operationare the only reference to smelting in the 1950's (other documents suggest that earliersmelting activities mainly involved aluminum). Based on available reports, the years ofsignificant nickel smelting were restricted to 1977 to 1986 (inclusive). While limitedquantities of nickel may have been smelted during other years, the exposure assessment inthis consultation focuses on those years of maximum nickel smelting (1977-86) and airborneemissions. As the airborne nickel concentrations estimated using conservative (healthprotective) assumptions during years of peak smelter operation are below levels of healthconcern, emissions during any earlier years of lower emission concentrations are also below levels of public health concern.

  1. ATSDR, 2001 Public Health Assessment fo the Paducah Gaseous Diffusion Plant(USDOE), May 14, 2001. (see page 38) When determining the contaminants of concernfor the May 14, 2001 health assessmentdid the ATSDR review the historical releases fromthe nickel smelting and lead "recycling" operations at PGDP? (see page 57-Table 11)What chemical form of nickel is considered here? (see page 68-Table 15A) When theATSDR determined off-site soil concentrations of nickel did not exceed health guidelineswere they able to determine the source of that nickel contamination? What was the source? (see page 17-Table 2) Nickel was detected in 42 of 110 off-site groundwater samples. What was the source of this nickel? (see page 19) "...nickel (was) only detected downwindof the PGDP smelter and mainly up-wind of the TVA plant. What was the source of thatnickel? (see page 65-Table 14A) When the ATSDR estimated nickel posed no healthproblem to workers and visitors of the WKWMA did they consider the concentration ofnickel in ash spread on the roads throughout the refuge and how nickel laden dust risesfrom roads when vehicles travel them? (see page 74-Table 17A) What was the source ofthe nickel contaminating the deer liver and muscle? (see page 146 #3) " can alwaysask for it to be explained in the form that is easiest for you to understand." (Thus) From1977 to 1982 how many pounds of non-radiologically contaminated nickel was releasedfrom the PGDP nickel smelter stack? From 1983 to 1986 how many pounds ofradiologically contaminated nickel was released from the PGDP nickel smelter stack?
  2. At the time ATSDR prepared the May 2001 Health Assessment were they aware nickelsmelting and lead "recycling" operations were conducted at PGDP? If so, why was theHealth Consultation for nickel an addendum to that report?

ATSDR Response: In determining the contaminants of concern for the Public HealthAssessment, ATSDR did not evaluate specific sources of nickel or lead. That evaluation wasbased on the measured concentration of the contaminants in various media, such as soil,water, and sediment. The rationale for conducting this health consultation is that we didnot include metal particulates released from smelting operations in assessing airbornecontaminant distributions. The form of nickel evaluated in the Public Health Assessmentfor the ground water, surface water, soil and sediment, and biota pathways is total nickel. For the air pathway in this consultation, the chemical form is, as referenced, nickel oxide(s),however, the nickel health comparison values used are applicable to elemental or metallicnickel, and both soluble and insoluble forms.

With regard to the chemical form of nickel compounds, only the nickel carbonyl form has asignificantly higher toxicity than other forms (and only via inhalation). As nickel carbonylis not thermodynamically stable in soil, air, or water, it will be converted to other morethermodynamically stable forms, such as nickel hydrates in water and nickel ferrite in soil(ATSDR, 1997). Nickel also tends to strongly adsorb onto larger particles in both soilsediment which limits its subsequent mobility (ATSDR, 1997). Although in some cases, itmay be important to assess the potential distribution and concentrations of the most toxicchemical forms, the limited chemical stability of nickel carbonyl precludes significant off-site transport and exposure. For the determination of the off-site contaminants of concernat PGDP, if the total environmental nickel concentrations are below levels of healthconcern, it is not necessary to delineate the specific sources of the nickel.

With regard to the total nickel distribution in the environment surrounding PGDP, thereare a number of potential sources including naturally occurring background levels in allmedia (nickel is the 24th most common element in the earth's crust). In addition to thePGDP smelter and steam plant and the TVA facility, minor sources include forest fires,volcanoes, fuel oil combustion, and other manufacturing processes. Any measurement ofnickel in any environmental media will include contributions from all sources. Neither thepublic health assessment nor the health consultation attempted to discern the specificsource of nickel in any media. The objective in these contaminant evaluations is todetermine if the total exposures result in doses of public health concern. If the doses are atconcentrations of public health concern, it may become necessary to determine the specificsources in order to implement a dose reduction strategy.

The annual smelter nickel emissions for the years 1983 to 1986 is estimated to be about 794pounds per year (based on an instantaneous emission rate of 0.05 g/sec for 8 hours per dayand 250 days per year). This operating schedule is in agreement with estimates of annualsmelter production of 6.7 million pounds per year (for the radiologically contaminatednickel). Using an average smelter loading rate of 3350 pounds per hour for 8 hours day,250 days per year equals a smelter output of 6.7 million pounds per year which is very closeto the average annual nickel production. It is assumed that releases of non-contaminatedand contaminated nickel are similar even though the maximum amount of non-contaminated nickel produced was stated to be just over 5 million pounds per year.

Although ATSDR was aware that metal smelting operations had taken place at PGDP, wehad not included such emissions in the public health assessment. This consultation is aresponse to comments received on the public health assessment document.

  1. Did the ATSDR assess the emissions data from and disposal path of nickel compoundsformed during the nickel stripping evaporation (tank) process in C-400-C at PGDP?

ATSDR Response: Nickel stripping operations in C-400-C consisted primarily in use oforganic solvents (either trichloroethylene or trichloroethane) to remove surfacecontamination from nickel components. This process will not result in the release ofsignificant quantities of nickel. Emissions of and potential exposures to the organic solventswere evaluated in the PHA.

  1. As part of this Health Consultation did the ATSDR review any health data for members of the community surrounding PGDP. Non-smoking individuals in the residential communityhave records showing an elevated body burden for nickel.

ATSDR Response: Any such health data is confidential information. ATSDR has no accessto such data unless provided by the individual in question and cannot review or evaluatewithout such access.


  1. The Toxic Release Inventory (USEPA) data reports for nickel compounds released by theTVA Shawnee Fossil Plant show releases to air, water, and land. For decades nickelcompounds were released via this facilities 244 m stack into the surrounding environment. While the height of the TVA stack facilitates the dispersion of these pollutants away fromPGDP during times of prevailing winds, the opposite occurs during a significant number ofdays each year. Figure 1 of this Health Consultation demonstrates changes in winddirection and resulting dispersion patterns. Further TRI data for TVA shows a significantlevel of nickel compounds in ash generated at the Shawnee plant. Historically, this ash(cinders) was spread on roadways in the WKWMA, dramatically increasing the levels ofnickel in the Health Consultation study area. Consequently, the nickel compounds releasedfrom the TVA facility should be included in this assessment. Additionally, while the PGDPsteam plant emissions of nickel compounds are far less than those from the TVA plant,PGDP releases should not be excluded when calculating TOTAL NICKEL at this locale. To date ash piles from PGDP continue to pollute the environs of the North-South DiversionDitch and the Little Bayou Creek watershed. TOTAL NICKEL from this toxic triad affectsthe health of the residents in the surrounding community.

ATSDR Response: We have reconfigured the ISC air dispersion model to include potentialcontributions from both PGDP steam plant and the TVA facility. Based on 1998 and 1999Toxic Release Inventory data for the TVA facility (TVA, 2002, the instantaneous nickel air emissions average~0.0025 g/sec. Assuming that those later years are similar to the years of PGDP nickelsmelter emissions, the instantaneous TVA release is about 20 times lower than the PGDPsmelter release (0.05 g/sec) and is dispersed from an 800+ foot stack. Similarly, usingstandard emission factors for coal and oil combustion and annual fuel usage rates,instantaneous nickel emissions from the PGDP steam plant averaged 0.00009 g/sec(dispersed from two adjacent 110 foot stacks), which is 500 times lower than the smelteremissions. The results of the revised air dispersion models indicate no differences in theshort term or annual distributions of nickel. All estimated concentrations are still wellbelow levels of health concern.

While the revised air model does not include potential re-suspension from roadways,particulates from road dust are relatively large particles with limited capacity forwidespread dispersion. Additionally, exposure to road dust within the non-residentialWKWMA roadways would be a relatively infrequent type of exposure. As a result of thelimited capacity for significant dispersion and the infrequent exposure, nickel in road dustalong WKWMA roadways is not an important contributor to the total airborne nickel load.

  1. Nickel particulates from TVA, PGDP, and nickel smelting could remain suspended in airfor approximately one week.

ATSDR Response: While this comment is theoretically true, based on availabledocumentation, particulates from the smelter are much larger than those from the PGDPsteam plant and TVA facilities and much more likely to settle out in a short distance fromthe release point. Conversely, the smaller particle sizes from the fossil fuel facilities andmuch higher release points (stack heights), which leads to longer suspension times, meansthat nickel from those sources is dispersed over a much larger area and does notsignificantly contribute to localized air concentrations. The models include appropriateparameters for particle size that largely determine settling rates and travel times. Themodeled dispersion also conservatively neglects wet deposition. Wet deposition occursduring precipitation (rain, snow events) and serves to limit dispersion. Neglecting this term increases the range and concentration of potential dispersion.

  1. Nickel can be transported in air adsorbed to other particles.

  2. Nickel particulates adsorbed to another variety of particulate can be inhaled as one particle.

  3. Nickel oxide is insoluble in water hence, there exists the likely-hood of particles becoming airborne again and again.

  4. When inhaled, nickel oxide is retained longer in the lung than any other form of nickel.

  5. Nickel can cross the placental barrier.

  6. Nickel is a mutagen (human)

ATSDR Response: The above factors have all been considered in the development of thehealth comparison values, the Public Health Assessment, and by reference, in this healthconsultation.

  1. Once again it is apparent that the ATSDR, while recognizing the variety and concentrationof contaminants at PGDP, has failed to address their combined effect when concluding"...estimated concentrations of all metal particulates from PGDP smelter operations arebelow levels of health concern, (and that) no adverse health effects are likely...." In fact, inhaling a mixture of pollutants , individually or in combined form, can magnify the effect several times over the effect of inhaling any one pollutant individually.

ATSDR Response: See above for response to assessing cumulative effects to differentcontaminants. ATSDR considered cumulative, additive, synergistic, and antagonisticeffects of exposure to the extent of scientific knowledge in this area. Information andreferences related to ATSDR's process of evaluation have been added to the consultation.

  1. The Health Consultation did not include specific mailing instructions for individualssending comments. Since a telephone inquiry seeking those instructions was not returned Ihave sent our comments to the same address as comments for the PGDP Health Assessment.

ATSDR Response: Instructions for mailing of written comments will be added to futureATSDR consultation cover pages (Health Consultation: A Note of Explanation) for healthconsultations. Thank you for bringing this to our attention.


ATSDR, 2002a. Substance Comparison Value Data Base. Agency for Toxic Substances andDisease Registry, US Department of Health and Human Services; Atlanta, GA, January 2002.

ATSDR, 2002b. Public Health Assessment for the Paducah Gaseous Diffusion Plant (USDOE),Paducah, McCracken County, Kentucky (Final Release). Agency for Toxic Substances andDisease Registry, US Department of Health and Human Services; Atlanta, GA.

ATSDR, 2001. Public Health Assessment for the Paducah Gaseous Diffusion Plant (USDOE),Paducah, McCracken County, Kentucky (Public Comment Release). Agency for Toxic Substancesand Disease Registry, US Department of Health and Human Services; Atlanta, GA, May 14, 2001.

ATSDR, 1997. Toxicological Profile for Nickel. Agency for Toxic Substances and DiseaseRegistry, US Department of Health and Human Services; Atlanta, GA, September 1997.

Berman E, House DE, Allis JW, et al. 1992. Hepatotoxic interactions of ethanol with allyl alcoholor carbon tetrachloride in rats. J Toxicol Environ Health 37(1): 161-176.

Caprino L, Borrelli F, Anonetti, et al. 1983. Sex-related toxicity of somatostatin and its interactionwith pentobarbital and strychnine. Toxicol Let 17:145-149.

DOE, 2000a. Report on the Paducah Gaseous Diffusion Plant Metals Recovery Program, USDepartment of Energy, Oak Ridge Operations, DOE/ORO-2105, Oak Ridge TN, December 2000.

DOE, 2000b. Report on the Paducah Gaseous Diffusion Plant "Work for Others" Programincluding Weapons Support and Disposition, US Department of Energy, Prepared for Oak RidgeOperations by AIMS, Inc., December 2000.

DOE, 2000c. Phase II Independent Investigation of the Paducah Gaseous Diffusion Plant. Preparedby the Office of Oversight, Office of Environment, Safety and Health, U.S. Department of Energy. Downloaded from on April 27, 2001.

DOE, 1986. Environmental Monitoring Report, United States Department of Energy PaducahGaseous Diffusion Plant, Calender Year 1985. Prepared by Martin Marietta Energy Systems, Inc.Paducah KY, KY-755.

Drott P, Meurling S, Gebre-Medhin M. 1993. Interactions of vitamins A and E and retinol-bindingprotein to healthy Swedish children---evidence of thresholds of essentiality and toxicity. Scand JClin Lab Invest 53:275-280.

EPA, 2002. Integrated Risk Information System. Nickel, soluble salts, Nickel Carbonyl, andNickel Subsulfide, Environmental Protection Agency. Downloaded January 31, 2002.

Feron VJ, Groten JP, van Zorge JA, Cassee FR, Jonker D, van Bladeren PJ. (1995). Toxicitystudies in rats of simple mixtures of chemicals with the same or different target organs. Toxicol Let,82-83, 505-512.

EPA, 2001a. Factor Information Retrieval (FIRE) v. 6.2.3; data base and associated files. downloaded 11/01/01.

EPA, 1995; 2001b. PCRAMMET User's Guide (and associated software, 1995). Meteorological data downloaded September 2001.

Garrison, L, personal communication. E-mail from Larry Garrison, Manager, Technical ServicesBranch, Division for Air Quality, KDEP to Mark Evans, ATSDR, Thursday, Sept. 6, 2001.

Groten JP, Schoen ED, van Bladeren PJ, Kuper CF, van Zorge JA, Feron VJ. (1997). Subacutetoxicity of a mixture of nine chemicals in rats: detecting interactive effects with a fractionated two-level factorial design. Fundam Appl Toxicol, 36(1), 15-29.

Groten JP, Sinkeldam EJ, Muys T, Luten JB, van Bladeren PJ. (1991). Interaction of dietary Ca,P, Mg, Mn, Cu, Fe, Zn and Se with the accumulation and oral toxicity of cadmium in rats. FoodChem Toxicol, 29(4), 249-258.

Harris LW, Lennox WJ, Talbot BG, et al. 1984. Toxicity of anticholinesterases: Interactions ofpyridostigmine and physostigmine with soman. Drug Chem Toxicol 7:507-526.

Jonker D, Woutersen RA, van Bladeren PJ, Til HP, Feron VJ. (1993). Subacute (4-wk) oraltoxicity of a combination of four nephrotoxins in rats: comparison with the toxicity of theindividual compounds. Food Chem Toxicol, 31(2), 125-136.

Jonker D, Woutersen RA, van Bladeren PJ, Til HP, Feron VJ. (1990). 4-week oral toxicity studyof a combination of eight chemicals in rats: comparison with the toxicity of the individualcompounds. Food Chem Toxicol, 28(9), 623-631.

KDEP, 1979. Permit Application for Air Contaminant Source (including attachments).Department for Natural Resources and Environmental Protection, Division of Air Pollution,Frankfort KY. Prepared by Department of Energy, Paducah Gaseous Diffusion Plant. November,1979.

NIOSH, 2002. Pocket Guide to Chemical Hazards. U.S. Department of Health and HumanServices, National Institute for Occupational Safety and Health. January 17, 2002.

PGDP, 1986. Environmental surveillance of the U.S. Department of Energy Paducah Reservationand surrounding environs during 1986. Prepared by Martin Marietta Energy Systems, Inc., for U.S.DOE, Paducah KY, ES/ESH-1/V3, April 1987.

Seed J, Brown RP, Olin SS, and Foran JA. (1995). Chemical mixtures: Current risk assessmentmethodologies and future directions. Regul. Toxicol. Pharmacol, 22, 76-94.

TVA, 2002. Toxic Release Inventory, Shawnee Fossil Plant, Tennessee Valley Authority,Downloaded January 3, 2002.

Trinity Consultants, 1996. Breeze Air Suite (ISCST3), Version 1.07. Trinity Consultants, Inc., Dallas, TX.

1. The cancer risks for other forms of nickel, such as nickel carbonyl, nickel oxides, and elemental nickel have not been estimated by the various oversight agencies (EPA, ATSDR, etc.).
2. Because air concentrations at any location vary over time due to changing emission and weather conditions both health comparison values and estimated air concentrations are based on time weighted averages. The average times used in this consultation are annual averages for evaluating long term exposures and 1 or 8 hour averages for short term exposures.
3. Hourly weather data for the Paducah-Barkley Airport were obtained from the US Environmental Protection Agency (EPA) Support Center for Regulatory Air Models (SCRAM) website ( and converted to the ISC3 format using EPA PCRAMMET software (EPA, 1995; 2001b).
4. No potential cancer risk is based on nine years of exposure to a concentration of 0.022 µg/m3 and a unit risk of 4.8E-4 per µg/m3 (which assumes 70 years of exposure; EPA, 2002).

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. #