- Email: [email protected]
Elemental phosphorus slag exposure study in Southeastern Idaho, USA
International Congress Series 1225 (2002) 131 – 138
Elemental phosphorus slag exposure study in Southeastern Idaho, USA J.L. Alvarez a, R. Geddes b, J.E. Rice c, T.F. Gesell d, D. Wells d,* a
Auxier and Associates, 9821 Cogdill Road, Knoxville, TN, 37932, USA b P4 Production, L.L.C., P.O. Box 816, Soda Springs, ID, 83276, USA c FMC Corporation, P.O. Box 4111, Pocatello, ID, 83205, USA d Health Physics Program, Department of Physics, Idaho State University, Campus Box 8106, Pocatello, ID, 83209, USA
Abstract Phosphorus slag containing elevated concentrations of radionuclides in the uranium series with radium-226 ranging from 750 to 1100 Bq/kg has been used in Southeastern Idaho communities as a construction material. An exposure study was conducted in the region to identify residences that incorporated phosphorus slag in the construction material, to determine the dose from the slag to residents, and to find slag in public structures and roads. Measurements were made with calibrated, commercial ‘‘Micro Rem’’ instruments and with commercial Al2O3 TL dosimeters. A total of 1472 residences were surveyed, and we estimate that less than 12% of the residences in Soda Springs contain slag; while no houses in Pocatello or Fort Hall were found to have slag in the construction. Individual dose estimates for nine individuals in this study exceeded 1 mSv/year above background, while the highest estimated individual dose was 1.32 mSv/year. Few of the public structures surveyed were found to contain slag. A significant fraction of the public roads, however, contained slag: 27% in Pocatello, 23% in Soda Springs, and 20% in Fort Hall. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Environmental radiation; Technologically enhanced natural radiation; Background radiation; Mining slag; Building materials
1. Introduction Phosphorus for agricultural purposes is produced in Southeastern Idaho near Soda Springs by Monsanto (now P4 Production, a subsidiary of Monsanto and Solutia), and near Pocatello by FMC. Regional phosphorus ores contain naturally elevated concen*
Corresponding author. Fax: +1-208-282-4649. E-mail address: [email protected] (D. Wells).
0531-5131/02 D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 5 3 1 - 5 1 3 1 ( 0 1 ) 0 0 5 5 0 - 7
132
J.L. Alvarez et al. / International Congress Series 1225 (2002) 131–138
trations of uranium, thorium and their progeny. A byproduct of the phosphorus production process is a vitreous slag. The elevated uranium, thorium and their progeny content of the slag leads to elevated gamma radiation levels from the slag. Radon levels, in contrast, are not enhanced because the vitreous nature of the slag prevents radon emanation [1– 3]. Phosphorus slag has been used in residential construction, streets, sidewalks, and construction fill in Pocatello, Soda Springs, Fort Hall, and nearby areas of Southeastern Idaho. The average natural external dose rate in Southeastern Idaho is approximately 0.10 mSv/h. The average external dose rate above a large volume of slag, in contrast, is approximately 0.60 mSv/h. Dose rates in residences that use construction material containing slag may range from 0.10 to 0.60 mSv/h depending upon the amount of slag in the construction. This potential level of external radiation from construction material led the US Environmental Protection Agency (EPA) to investigate regional slag-based radiation doses [4]. They concluded that some residents could be at significantly increased risk of cancer due to long-term exposure to slag. FMC and P4 Production companies (the Companies) agreed in 1992 to develop protocols for measuring slag-related radiation exposures and to carry out these measurements. The Companies developed methods for measuring individual exposures to radiation from the slag, contributed to the development of graded decision guidelines (GDGs) that establish exposure levels at which corrective action was recommended [3,4] and, for exposure levels above the GDGs recommendations, to implement corrective actions if requested by the property owner. The regional state health department served as the primary interface and coordination between the public and ourselves. Thus, our objectives were to assess the individual doses of volunteer member of the public, identify sources of exposure, locate and inventory slag in the communities of Pocatello, Soda Springs, and Fort Hall and, when requested, implement GDGs. We assessed gamma doses to individuals from phosphorus slag (Residential Surveys) and located slag within public structures and roads (Community Surveys). Aggregate results of measurements of public and private properties and their consequences are discussed below.
2. Methods We conducted exposure surveys and assessed doses to individuals from gamma radiation due to elemental phosphorus slag (Residential Surveys) and located slag within the communities (Community Surveys). Public participation of volunteers was solicited through public meetings, radio announcements, television advertisements, newspaper ads, and mass mailings. Residential surveys were generally conducted in two phases. Phase I initial screening survey was performed to determine the possible presence of slag. Volunteer participants were offered a Phase II detailed dose assessment, if the Phase I screening survey identified dose rates greater than or equal to 0.20 mSv/h. Volunteers also had the option to proceed to a Phase II dose assessment without having a screening survey. Doses greater than or equal to 1.0 mSv/year above the local background were eligible for dose reduction evaluations under the GDGs. Locations for community surveys of roads, parking lots, public buildings, etc., were selected based upon the prior aerial gammasurveys and recommendations from local governmental authorities.
J.L. Alvarez et al. / International Congress Series 1225 (2002) 131–138
133
The initial Phase I residential screening surveys were performed to determine the possible presence of slag. Screening measurements were made with a Bicron Micro Rem meter. Ten-second measurements were taken at approximately 1 m (3 ft) above the floor at the center of high-occupancy rooms within the home. This survey included two measurements on the ground floor of the residence and two in the basement (if present). Measurements were also taken in transit between rooms in order to observe any anomalous dose rates that could be associated with concentrations of slag. Any measurement greater than or equal to 0.20 mSv/h was presumed to be due to the presence of slag. All Phase I results were recorded and submitted to the resident. A recommendation for Phase II dose assessment was provided to the resident whenever the dose rates were greater than or equal to 0.20 mSv/h, while a recommendation of no further action was provided whenever all readings were below 0.20 mSv/h. When volunteer residents opted to proceed to a Phase II dose assessment, one or both of two distinct methods were employed. One method employed thermoluminescent dosimeters (TLD) worn by residents to screen for individual dose. For this method, each volunteer wore a TLD (Al2O3, Landauer Model X-9 Environmental/Low Level dosimeters) for approximately 90 days. Because our protocol for TLD-based dose assessments included doses outside the home, such as the workplace, these measurements overestimated the slag contribution to dose. Thus, eligibility for consideration under the GDGs required a dose determination by the second dose assessment method, which we call the dose-rate/time-log method. The Phase II TLD measurements included 1402 TLDs that were issued to volunteer participants, either as an alternative to actual surveys of residences or to supplement such surveys. More than half of the participants requested both methods. The second Phase II method consisted of dose-rate/time-logging (DRTL)-used portable Bicron Micro Rem meters. In this case, a systematic survey of dose rate was performed throughout the residence, and time logs of occupancy were recorded. Individual doses were then assessed by summing the products of occupancy times dose rate. At each survey location with elevated direct radiation levels, the probable presence of slag was confirmed by gamma spectroscopy. The major disadvantages to DRTL surveys are that many locations must be measured for total dose, and residents must supply a corresponding time estimate (of unknown uncertainty) for each location. The DRTL surveys began with an inspection for slag. Measurements with a Bicron Micro Rem meter was taken during the inspection for slag. Slag was sought at high doserate locations, which generally occurred in the basements. A geologist trained all the personnel performing the measurements for this study in the identification of a slag in concrete by physical appearance. Gamma spectra were taken with an Aptec Odyssey 5 multichannel analyzer and a 7.5 7.5 cm NaI(Tl) detector at locations of highest dose rate. If exposed concrete could not be found in the dwelling, gamma spectroscopy was the only slag identification measurement performed. The presence of 200 or more excess counts in the 1764 keV photo-peak over a count time of 600 s indicated the presence of abovebackground concentrations of Ra-226 and probably the entire uranium decay series. The DRTL surveys for individual dose rates were also performed with the Bicron Micro Rem meter. Measurements were taken at approximately 1 m above the floor and while walking about each room of the residence. The minimum coverage of the room
134
J.L. Alvarez et al. / International Congress Series 1225 (2002) 131–138
included the perimeters and diagonals of the room. If there were no obvious gradients in dose rate, then a single 10-s reading in the center of the room was used for the room dose rate. When the room exhibited an obvious gradient, 10-s measurements were made every meter along the gradient and the average of the measurements was used for the room dose rate. Doses to residents were estimated by subtracting the background dose rate from the measured room dose rates. A resident-supplied estimate of average time spent in each room was obtained for each individual living in the residence. The product of the abovebackground dose rate, average hours per day, and days per year yielded the annual dose per room. The sum of doses for all rooms was the total annual above-background dose for the individual resulting from slag in the residence. In addition to these gamma-based dose measurements, 265 residences requested radon measurements. A total of 518 separate radon measurements were made with charcoal radon test kits and following vendor instructions. Community Survey locations were chosen from results of previous investigations [5] and areas suggested by local governmental officials. Locations in the Pocatello and Soda Springs communities exceeding 0.17 mSv/h were identified from the 1986 aerial surveys of Berry et al. At the request of community officials, two municipal facilities were surveyed in Pocatello and Chubbuck, 18 public buildings and other facilities were surveyed in Soda Springs, and 17 public buildings were surveyed in Fort Hall. Road surveys were initially done from a moving vehicle; the vehicle speed was maintained at approximately 10 m/s or less, with the survey instrument on the floorboard inside the vehicle (approximately 0.5 m above the road surface). Radiation measurements were recorded directly onto Community maps. Visible changes in road surface and changes in dose rate (>50%) led to reduced vehicle speeds and/or walking surveys to better delineate boundaries of areas of elevated direct radiation. At locations where vehicle monitoring identified levels equal to or greater than 0.20 mSv/h, the vehicle was stopped and radiation levels were measured outside the vehicle at a height of 1 m above the road. In all, approximately 200 km in Fort Hall, 416 km in Pocatello/Chubbuck, and 42 km in Soda Springs were surveyed in this manner. In general, slag-containing street surfaces of more than 20 m in length, which exceeded 0.20 mSv/h, could be accurately identified from a vehicle traveling at 10 m/s. A total of 3550 TLDs were employed in this investigation and stored in the lead shields when they were not in use. Standard quality assurance measures were taken to assess storage, transit and background doses and reproducibility. Background measurements were performed with TLDs in the three study areas. Ten background measurement locations each were selected in Pocatello, Soda Springs, and Fort Hall. These background locations were selected to be away from direct radiation from slag and were surveyed using a Micro Rem meter to ensure that the surroundings were unaffected by slag, other radiation sources and, were further of constant dose rate. Two TLDs (measurement and duplicate) were placed in plastic containers with security seals to identify tampering and deployed at the locations for 3 or more months. The background measurements were performed four times for Pocatello and Fort Hall and three times for Soda Springs at the same locations and at different times of the year to measure seasonal variations. Further details of our methods can be found in Refs. [6,7].
J.L. Alvarez et al. / International Congress Series 1225 (2002) 131–138
135
3. Results Survey data were used to estimate the fraction of residences of each town that have slag in their construction. We estimate this fraction using: X p 1=2N z ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi : pð1 pÞ=N
ð1Þ
In Eq. (1), z is the ± 0.975 normal quantile, X is the number of houses with dose rate of 0.20 mSv/h or greater divided by N, the number of houses surveyed, and p is the probability limit of z. The 1/2N in the equation is + for the upper limit and for the lower limit. This estimate assumes that the sample of houses screened was randomly selected and all houses have an equal chance of containing slag. However, the houses were selected on a voluntary basis and the bias would be expected to be toward a likelihood of slag. Thus, the number of houses that would be expected to have slag as a construction material is probably lower. We estimate that the fraction of homes in Pocatello and Fort Hall that have slag as a construction material is between 0% and 0.5% (95% confidence interval). In contrast, the fraction of residences in Soda Springs that we estimate to have slag as a construction material is between 28% and 43%. It should be noted, however, that slag was only available for construction from 1960 to 1976. Approximately 25% of the houses in Soda Springs were built during this time. If it is assumed that 35% of the houses built between 1960 and 1976 (25% of the total houses) contain slag, then 7 –11% of all houses in Soda Springs are likely to contain slag. TLDs were carried on the key rings of volunteers in order to keep the TLD near their person for the wear period. Thus, TLD measurements included some occupational and medical exposures away from the home. The results of the individual measurements for the three communities compared to the background measurements are shown in Figs. 1 and 2. Individual doses were generally higher than the background measurements and their
Fig. 1. Total dose to individuals in Pocatello as measured by TLD compared to annual background dose.
136
J.L. Alvarez et al. / International Congress Series 1225 (2002) 131–138
Fig. 2. Total dose to individuals in Soda Springs as measured by TLD compared to annual background dose.
respective distributions differ, as can be seen in the figures. The origin of these differences presumably stems from at least three sources. First, individual doses were an integration of doses received from all the locations the individual visited, while the background doses were received at a single, fixed location. Second, individual doses included occupational and medical doses, while background doses did not. Third, some individual doses included slag from housing construction. Our physical and gamma-spectroscopic measurements, however, revealed no evidence of slag in Pocatello and Fort Hall housing construction. Thus, we suspect that the additional dose observed in these data stem from medical or occupational sources. Net background-subtracted TLD dose distributions were assessed by subtracting the measured normally distributed backgrounds with Monte-Carlo techniques. The resulting above-background distributions were generally normal or log-normal. Results for Pocatello, Soda Springs and Fort Hall are listed in Table 1. Additional measurements to more accurately estimate the slag-related above-background doses were performed at 28 homes with DRTL method. We found eight homes, all in Soda Springs, that produced at least one individual with an estimated above-background dose greater than 1.00 mSv/year. The highest estimated, annual, above-background dose was 1.32 mSv. These DRTL surveys also yielded above-background dose estimates for 100 individuals in Soda Springs. We estimate that their mean dose from residential slag is
Table 1 Estimated above-background dose estimates from TLD data with 95% confidence limits Location
Mean dose (mSv/year)
Geometric mean dose (mSv/year)
Median dose (mSv/year)
Estimated 95th percentile dose (mSv/year)
Pocatello Soda Springs Fort hall
0.63 ± 0.30 NA 0.18 ± 0.35
0.055 (1.7) 0.57 (2.3) NA
0.029 ( 0.019, 0.074) 0.53 (0.39, 0.66) 0.11 (0.06, 0.17)
1.70 2.75 1.20
These are over-estimates because they include occupational and medical exposures.
J.L. Alvarez et al. / International Congress Series 1225 (2002) 131–138
137
0.42 mSv/year with a 95th-percentile expectation of 1.40 mSv/year. The comparable TLD estimates are a mean of 0.63 mSv/year and 95th-percentile expectation of 2.75 mSv/year. Subsequent radon measurements in these homes found no correlation between radon and dose rates due to slag; thus confirming that slag does not contribute to indoor radon. The maximum levels measured on roads in the three communities were: Pocatello/ Chubbuck, 0.45 mSv/h; Soda Springs, 0.50 mSv/h; and Fort Hall, 0.35 mSv/h. Approximately one-fourth of the paved roads exhibited radiation levels equal to or greater than 0.20 mSv/h. The fraction of slag-infused road construction in each community were 25.5% in Pocatello/Chubbuck, 23.1% in Soda Springs, and 28.5% in Fort Hall. These fractions are somewhat smaller than the 1986 aerial survey. This is likely due to a combination of attrition through routine resurfacing with non-slag material, application of surface sealer coats, and complete replacement of paving and paving base during various construction activities. Of the 37 community buildings and facilities surveyed, three had small areas of elevated direct radiation exceeding 0.20 mSv/h. The highest level measured at these locations was 0.45 mSv/h. More detailed results can be found in Ref. [8].
4. Conclusions We conducted an exposure study of phosphorus slag that was used in Southeastern Idaho communities as a construction material. We found that slag was commonly used in road-bed construction; 27% in Pocatello, 23% in Soda Springs, and 20% in Fort Hall. We found slag in residential construction, however, only in Soda Springs, ID. We estimate that less than 12% of the residences in Soda Springs contain slag. Individual dose estimates for nine individuals in this study exceeded 1 mSv/year above background, while the highest estimated individual dose was 1.32 mSv/year.
Acknowledgements This research was funded by FMC and the Monsanto.
References [1] IT Corporation, Methods development study work plan (draft), Prepared by IT Corporation/Nuclear Sciences, Prepared for FMC/Monsanto, 1993. [2] IT Corporation, Methods development study report, Prepared by IT Corporation/Nuclear Sciences, Prepared for FMC/Monsanto, 1994. [3] Southeast Idaho Technical Work Group on Slag, Graded decision guidelines for individuals exposed to gamma radiation from slag, Prepared by Southeast Idaho Technical Work Group on Slag, Prepared for the US EPA, 1995. [4] US Environmental Protection Agency, Administrative Order on Consent Docket No. 1092-11-04-7003, In the matter of: elemental phosphorus slag, 1992. [5] Berry, H.A., An aerial radiological survey of Pocatello and Soda Springs, ID and surrounding area, EPA8613, prepared for EPA by EG&G Energy Measurements, Las Vegas, NV, 1987.
138
J.L. Alvarez et al. / International Congress Series 1225 (2002) 131–138
[6] Auxier and Associates, 1996, Exposure study work plan, Prepared by Auxier and Associates, Prepared for FMC/Monsanto, Revised September 17, 1996 and April 1, 1997. [7] Auxier and Associates, Exposure study Phase II work plan, Prepared by Auxier and Associates, Prepared for FMC/Monsanto, December 31, 1998. [8] Auxier and Associates, Elemental phosphorus slag exposure study: Phase I final report, Prepared for FMC/ Monsanto, 1999.
Elemental phosphorus slag exposure study in Southeastern Idaho, USA J.L. Alvarez a, R. Geddes b, J.E. Rice c, T.F. Gesell d, D. Wells d,* a
Auxier and Associates, 9821 Cogdill Road, Knoxville, TN, 37932, USA b P4 Production, L.L.C., P.O. Box 816, Soda Springs, ID, 83276, USA c FMC Corporation, P.O. Box 4111, Pocatello, ID, 83205, USA d Health Physics Program, Department of Physics, Idaho State University, Campus Box 8106, Pocatello, ID, 83209, USA
Abstract Phosphorus slag containing elevated concentrations of radionuclides in the uranium series with radium-226 ranging from 750 to 1100 Bq/kg has been used in Southeastern Idaho communities as a construction material. An exposure study was conducted in the region to identify residences that incorporated phosphorus slag in the construction material, to determine the dose from the slag to residents, and to find slag in public structures and roads. Measurements were made with calibrated, commercial ‘‘Micro Rem’’ instruments and with commercial Al2O3 TL dosimeters. A total of 1472 residences were surveyed, and we estimate that less than 12% of the residences in Soda Springs contain slag; while no houses in Pocatello or Fort Hall were found to have slag in the construction. Individual dose estimates for nine individuals in this study exceeded 1 mSv/year above background, while the highest estimated individual dose was 1.32 mSv/year. Few of the public structures surveyed were found to contain slag. A significant fraction of the public roads, however, contained slag: 27% in Pocatello, 23% in Soda Springs, and 20% in Fort Hall. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Environmental radiation; Technologically enhanced natural radiation; Background radiation; Mining slag; Building materials
1. Introduction Phosphorus for agricultural purposes is produced in Southeastern Idaho near Soda Springs by Monsanto (now P4 Production, a subsidiary of Monsanto and Solutia), and near Pocatello by FMC. Regional phosphorus ores contain naturally elevated concen*
Corresponding author. Fax: +1-208-282-4649. E-mail address: [email protected] (D. Wells).
0531-5131/02 D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 5 3 1 - 5 1 3 1 ( 0 1 ) 0 0 5 5 0 - 7
132
J.L. Alvarez et al. / International Congress Series 1225 (2002) 131–138
trations of uranium, thorium and their progeny. A byproduct of the phosphorus production process is a vitreous slag. The elevated uranium, thorium and their progeny content of the slag leads to elevated gamma radiation levels from the slag. Radon levels, in contrast, are not enhanced because the vitreous nature of the slag prevents radon emanation [1– 3]. Phosphorus slag has been used in residential construction, streets, sidewalks, and construction fill in Pocatello, Soda Springs, Fort Hall, and nearby areas of Southeastern Idaho. The average natural external dose rate in Southeastern Idaho is approximately 0.10 mSv/h. The average external dose rate above a large volume of slag, in contrast, is approximately 0.60 mSv/h. Dose rates in residences that use construction material containing slag may range from 0.10 to 0.60 mSv/h depending upon the amount of slag in the construction. This potential level of external radiation from construction material led the US Environmental Protection Agency (EPA) to investigate regional slag-based radiation doses [4]. They concluded that some residents could be at significantly increased risk of cancer due to long-term exposure to slag. FMC and P4 Production companies (the Companies) agreed in 1992 to develop protocols for measuring slag-related radiation exposures and to carry out these measurements. The Companies developed methods for measuring individual exposures to radiation from the slag, contributed to the development of graded decision guidelines (GDGs) that establish exposure levels at which corrective action was recommended [3,4] and, for exposure levels above the GDGs recommendations, to implement corrective actions if requested by the property owner. The regional state health department served as the primary interface and coordination between the public and ourselves. Thus, our objectives were to assess the individual doses of volunteer member of the public, identify sources of exposure, locate and inventory slag in the communities of Pocatello, Soda Springs, and Fort Hall and, when requested, implement GDGs. We assessed gamma doses to individuals from phosphorus slag (Residential Surveys) and located slag within public structures and roads (Community Surveys). Aggregate results of measurements of public and private properties and their consequences are discussed below.
2. Methods We conducted exposure surveys and assessed doses to individuals from gamma radiation due to elemental phosphorus slag (Residential Surveys) and located slag within the communities (Community Surveys). Public participation of volunteers was solicited through public meetings, radio announcements, television advertisements, newspaper ads, and mass mailings. Residential surveys were generally conducted in two phases. Phase I initial screening survey was performed to determine the possible presence of slag. Volunteer participants were offered a Phase II detailed dose assessment, if the Phase I screening survey identified dose rates greater than or equal to 0.20 mSv/h. Volunteers also had the option to proceed to a Phase II dose assessment without having a screening survey. Doses greater than or equal to 1.0 mSv/year above the local background were eligible for dose reduction evaluations under the GDGs. Locations for community surveys of roads, parking lots, public buildings, etc., were selected based upon the prior aerial gammasurveys and recommendations from local governmental authorities.
J.L. Alvarez et al. / International Congress Series 1225 (2002) 131–138
133
The initial Phase I residential screening surveys were performed to determine the possible presence of slag. Screening measurements were made with a Bicron Micro Rem meter. Ten-second measurements were taken at approximately 1 m (3 ft) above the floor at the center of high-occupancy rooms within the home. This survey included two measurements on the ground floor of the residence and two in the basement (if present). Measurements were also taken in transit between rooms in order to observe any anomalous dose rates that could be associated with concentrations of slag. Any measurement greater than or equal to 0.20 mSv/h was presumed to be due to the presence of slag. All Phase I results were recorded and submitted to the resident. A recommendation for Phase II dose assessment was provided to the resident whenever the dose rates were greater than or equal to 0.20 mSv/h, while a recommendation of no further action was provided whenever all readings were below 0.20 mSv/h. When volunteer residents opted to proceed to a Phase II dose assessment, one or both of two distinct methods were employed. One method employed thermoluminescent dosimeters (TLD) worn by residents to screen for individual dose. For this method, each volunteer wore a TLD (Al2O3, Landauer Model X-9 Environmental/Low Level dosimeters) for approximately 90 days. Because our protocol for TLD-based dose assessments included doses outside the home, such as the workplace, these measurements overestimated the slag contribution to dose. Thus, eligibility for consideration under the GDGs required a dose determination by the second dose assessment method, which we call the dose-rate/time-log method. The Phase II TLD measurements included 1402 TLDs that were issued to volunteer participants, either as an alternative to actual surveys of residences or to supplement such surveys. More than half of the participants requested both methods. The second Phase II method consisted of dose-rate/time-logging (DRTL)-used portable Bicron Micro Rem meters. In this case, a systematic survey of dose rate was performed throughout the residence, and time logs of occupancy were recorded. Individual doses were then assessed by summing the products of occupancy times dose rate. At each survey location with elevated direct radiation levels, the probable presence of slag was confirmed by gamma spectroscopy. The major disadvantages to DRTL surveys are that many locations must be measured for total dose, and residents must supply a corresponding time estimate (of unknown uncertainty) for each location. The DRTL surveys began with an inspection for slag. Measurements with a Bicron Micro Rem meter was taken during the inspection for slag. Slag was sought at high doserate locations, which generally occurred in the basements. A geologist trained all the personnel performing the measurements for this study in the identification of a slag in concrete by physical appearance. Gamma spectra were taken with an Aptec Odyssey 5 multichannel analyzer and a 7.5 7.5 cm NaI(Tl) detector at locations of highest dose rate. If exposed concrete could not be found in the dwelling, gamma spectroscopy was the only slag identification measurement performed. The presence of 200 or more excess counts in the 1764 keV photo-peak over a count time of 600 s indicated the presence of abovebackground concentrations of Ra-226 and probably the entire uranium decay series. The DRTL surveys for individual dose rates were also performed with the Bicron Micro Rem meter. Measurements were taken at approximately 1 m above the floor and while walking about each room of the residence. The minimum coverage of the room
134
J.L. Alvarez et al. / International Congress Series 1225 (2002) 131–138
included the perimeters and diagonals of the room. If there were no obvious gradients in dose rate, then a single 10-s reading in the center of the room was used for the room dose rate. When the room exhibited an obvious gradient, 10-s measurements were made every meter along the gradient and the average of the measurements was used for the room dose rate. Doses to residents were estimated by subtracting the background dose rate from the measured room dose rates. A resident-supplied estimate of average time spent in each room was obtained for each individual living in the residence. The product of the abovebackground dose rate, average hours per day, and days per year yielded the annual dose per room. The sum of doses for all rooms was the total annual above-background dose for the individual resulting from slag in the residence. In addition to these gamma-based dose measurements, 265 residences requested radon measurements. A total of 518 separate radon measurements were made with charcoal radon test kits and following vendor instructions. Community Survey locations were chosen from results of previous investigations [5] and areas suggested by local governmental officials. Locations in the Pocatello and Soda Springs communities exceeding 0.17 mSv/h were identified from the 1986 aerial surveys of Berry et al. At the request of community officials, two municipal facilities were surveyed in Pocatello and Chubbuck, 18 public buildings and other facilities were surveyed in Soda Springs, and 17 public buildings were surveyed in Fort Hall. Road surveys were initially done from a moving vehicle; the vehicle speed was maintained at approximately 10 m/s or less, with the survey instrument on the floorboard inside the vehicle (approximately 0.5 m above the road surface). Radiation measurements were recorded directly onto Community maps. Visible changes in road surface and changes in dose rate (>50%) led to reduced vehicle speeds and/or walking surveys to better delineate boundaries of areas of elevated direct radiation. At locations where vehicle monitoring identified levels equal to or greater than 0.20 mSv/h, the vehicle was stopped and radiation levels were measured outside the vehicle at a height of 1 m above the road. In all, approximately 200 km in Fort Hall, 416 km in Pocatello/Chubbuck, and 42 km in Soda Springs were surveyed in this manner. In general, slag-containing street surfaces of more than 20 m in length, which exceeded 0.20 mSv/h, could be accurately identified from a vehicle traveling at 10 m/s. A total of 3550 TLDs were employed in this investigation and stored in the lead shields when they were not in use. Standard quality assurance measures were taken to assess storage, transit and background doses and reproducibility. Background measurements were performed with TLDs in the three study areas. Ten background measurement locations each were selected in Pocatello, Soda Springs, and Fort Hall. These background locations were selected to be away from direct radiation from slag and were surveyed using a Micro Rem meter to ensure that the surroundings were unaffected by slag, other radiation sources and, were further of constant dose rate. Two TLDs (measurement and duplicate) were placed in plastic containers with security seals to identify tampering and deployed at the locations for 3 or more months. The background measurements were performed four times for Pocatello and Fort Hall and three times for Soda Springs at the same locations and at different times of the year to measure seasonal variations. Further details of our methods can be found in Refs. [6,7].
J.L. Alvarez et al. / International Congress Series 1225 (2002) 131–138
135
3. Results Survey data were used to estimate the fraction of residences of each town that have slag in their construction. We estimate this fraction using: X p 1=2N z ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi : pð1 pÞ=N
ð1Þ
In Eq. (1), z is the ± 0.975 normal quantile, X is the number of houses with dose rate of 0.20 mSv/h or greater divided by N, the number of houses surveyed, and p is the probability limit of z. The 1/2N in the equation is + for the upper limit and for the lower limit. This estimate assumes that the sample of houses screened was randomly selected and all houses have an equal chance of containing slag. However, the houses were selected on a voluntary basis and the bias would be expected to be toward a likelihood of slag. Thus, the number of houses that would be expected to have slag as a construction material is probably lower. We estimate that the fraction of homes in Pocatello and Fort Hall that have slag as a construction material is between 0% and 0.5% (95% confidence interval). In contrast, the fraction of residences in Soda Springs that we estimate to have slag as a construction material is between 28% and 43%. It should be noted, however, that slag was only available for construction from 1960 to 1976. Approximately 25% of the houses in Soda Springs were built during this time. If it is assumed that 35% of the houses built between 1960 and 1976 (25% of the total houses) contain slag, then 7 –11% of all houses in Soda Springs are likely to contain slag. TLDs were carried on the key rings of volunteers in order to keep the TLD near their person for the wear period. Thus, TLD measurements included some occupational and medical exposures away from the home. The results of the individual measurements for the three communities compared to the background measurements are shown in Figs. 1 and 2. Individual doses were generally higher than the background measurements and their
Fig. 1. Total dose to individuals in Pocatello as measured by TLD compared to annual background dose.
136
J.L. Alvarez et al. / International Congress Series 1225 (2002) 131–138
Fig. 2. Total dose to individuals in Soda Springs as measured by TLD compared to annual background dose.
respective distributions differ, as can be seen in the figures. The origin of these differences presumably stems from at least three sources. First, individual doses were an integration of doses received from all the locations the individual visited, while the background doses were received at a single, fixed location. Second, individual doses included occupational and medical doses, while background doses did not. Third, some individual doses included slag from housing construction. Our physical and gamma-spectroscopic measurements, however, revealed no evidence of slag in Pocatello and Fort Hall housing construction. Thus, we suspect that the additional dose observed in these data stem from medical or occupational sources. Net background-subtracted TLD dose distributions were assessed by subtracting the measured normally distributed backgrounds with Monte-Carlo techniques. The resulting above-background distributions were generally normal or log-normal. Results for Pocatello, Soda Springs and Fort Hall are listed in Table 1. Additional measurements to more accurately estimate the slag-related above-background doses were performed at 28 homes with DRTL method. We found eight homes, all in Soda Springs, that produced at least one individual with an estimated above-background dose greater than 1.00 mSv/year. The highest estimated, annual, above-background dose was 1.32 mSv. These DRTL surveys also yielded above-background dose estimates for 100 individuals in Soda Springs. We estimate that their mean dose from residential slag is
Table 1 Estimated above-background dose estimates from TLD data with 95% confidence limits Location
Mean dose (mSv/year)
Geometric mean dose (mSv/year)
Median dose (mSv/year)
Estimated 95th percentile dose (mSv/year)
Pocatello Soda Springs Fort hall
0.63 ± 0.30 NA 0.18 ± 0.35
0.055 (1.7) 0.57 (2.3) NA
0.029 ( 0.019, 0.074) 0.53 (0.39, 0.66) 0.11 (0.06, 0.17)
1.70 2.75 1.20
These are over-estimates because they include occupational and medical exposures.
J.L. Alvarez et al. / International Congress Series 1225 (2002) 131–138
137
0.42 mSv/year with a 95th-percentile expectation of 1.40 mSv/year. The comparable TLD estimates are a mean of 0.63 mSv/year and 95th-percentile expectation of 2.75 mSv/year. Subsequent radon measurements in these homes found no correlation between radon and dose rates due to slag; thus confirming that slag does not contribute to indoor radon. The maximum levels measured on roads in the three communities were: Pocatello/ Chubbuck, 0.45 mSv/h; Soda Springs, 0.50 mSv/h; and Fort Hall, 0.35 mSv/h. Approximately one-fourth of the paved roads exhibited radiation levels equal to or greater than 0.20 mSv/h. The fraction of slag-infused road construction in each community were 25.5% in Pocatello/Chubbuck, 23.1% in Soda Springs, and 28.5% in Fort Hall. These fractions are somewhat smaller than the 1986 aerial survey. This is likely due to a combination of attrition through routine resurfacing with non-slag material, application of surface sealer coats, and complete replacement of paving and paving base during various construction activities. Of the 37 community buildings and facilities surveyed, three had small areas of elevated direct radiation exceeding 0.20 mSv/h. The highest level measured at these locations was 0.45 mSv/h. More detailed results can be found in Ref. [8].
4. Conclusions We conducted an exposure study of phosphorus slag that was used in Southeastern Idaho communities as a construction material. We found that slag was commonly used in road-bed construction; 27% in Pocatello, 23% in Soda Springs, and 20% in Fort Hall. We found slag in residential construction, however, only in Soda Springs, ID. We estimate that less than 12% of the residences in Soda Springs contain slag. Individual dose estimates for nine individuals in this study exceeded 1 mSv/year above background, while the highest estimated individual dose was 1.32 mSv/year.
Acknowledgements This research was funded by FMC and the Monsanto.
References [1] IT Corporation, Methods development study work plan (draft), Prepared by IT Corporation/Nuclear Sciences, Prepared for FMC/Monsanto, 1993. [2] IT Corporation, Methods development study report, Prepared by IT Corporation/Nuclear Sciences, Prepared for FMC/Monsanto, 1994. [3] Southeast Idaho Technical Work Group on Slag, Graded decision guidelines for individuals exposed to gamma radiation from slag, Prepared by Southeast Idaho Technical Work Group on Slag, Prepared for the US EPA, 1995. [4] US Environmental Protection Agency, Administrative Order on Consent Docket No. 1092-11-04-7003, In the matter of: elemental phosphorus slag, 1992. [5] Berry, H.A., An aerial radiological survey of Pocatello and Soda Springs, ID and surrounding area, EPA8613, prepared for EPA by EG&G Energy Measurements, Las Vegas, NV, 1987.
138
J.L. Alvarez et al. / International Congress Series 1225 (2002) 131–138
[6] Auxier and Associates, 1996, Exposure study work plan, Prepared by Auxier and Associates, Prepared for FMC/Monsanto, Revised September 17, 1996 and April 1, 1997. [7] Auxier and Associates, Exposure study Phase II work plan, Prepared by Auxier and Associates, Prepared for FMC/Monsanto, December 31, 1998. [8] Auxier and Associates, Elemental phosphorus slag exposure study: Phase I final report, Prepared for FMC/ Monsanto, 1999.