- Email: [email protected]
Fate of 60Co at a sludge land application site
Journal of Environmental Radioactivity 99 (2008) 1611–1616
Contents lists available at ScienceDirect
Journal of Environmental Radioactivity journal homepage: www.elsevier.com/locate/jenvrad
Fate of
60
Co at a sludge land application site
Michael A. Smith a, *,1, Ingvar L. Larsen b, 2, Audeen W. Fentiman a, 3 a b
Environmental Science Graduate Program, Ohio State University, Columbus, OH 43210, USA Environmental Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 29 November 2007 Received in revised form 4 June 2008 Accepted 9 June 2008 Available online 22 July 2008
Vertical distributions of 60Co are determined in soil cores obtained from a 10-ha grassland, where anaerobically digested sludge was applied by surface spraying from 1986 to 1995 on the U.S. Department of Energy’s Oak Ridge Reservation. These results, along with historical application records, are used to estimate vertical-migration rates and perform a mass balance. The presence of 60Co results solely from the sludge-application process. Soil, vegetation, and surface-water samples were collected. Eleven soil cores were sectioned into 3-cm increments and analyzed by gamma-ray spectrometry. No 60Co was detected in the vegetation or water samples. The downward migration rate of 60Co in the upper 15 cm of soil ranged from 0.50 to 0.73 cm/yr. About 98%, 0.020 0.011 Bq/cm2, of 60Co remained in the upper 15 cm of soil, which compared favorably with the expected 60Co activity based on historical records of 0.019 0.010 Bq/cm2. Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Biosolid Heavy metal Vertical migration Mass balance Radionuclide 60 Co
1. Introduction The fate of land-applied radionuclides contained in wastewatertreatment sludge has received considerable recent attention (e.g., Bastian et al., 2005; O’Connor et al., 2005; Wolbarst et al., 2006), and the practice of land application of these biosolids is expected to continue to grow because of the perceived economic and societal benefits (Basta et al., 2005). Radionuclides are either present naturally or enter the wastewater stream from numerous anthropogenic sources, such as combustion, fertilizer runoff, mining and smelting operations, leachate from disposal facilities, industrial byproducts, research institutions, and hospitals, as well as from atmospheric fallout. Historically, many investigators have studied the fate of radionuclides in the environment or laboratory, considering wind, soil, surface water, and groundwater, with numerous studies resulting from the 1986 Chernobyl release (e.g., Anspaugh et al., 2002; Arapis et al., 1997; Baes and Sharp, 1983; Boone et al., 1985; Bunzl et al., 1994; Tanaka and Ohnuki, 1996; Whicker and Pinder, 2002). Many of these studies involved acute application rates with high concentrations of radionuclides. In contrast, the study conducted at the Department of Energy’s (DOE) Oak Ridge Reservation focused on * Corresponding author. Tel.: þ1 509 372 4788; fax: þ1 509 375 2019. E-mail address: [email protected] (M.A. Smith). 1 Present address: Pacific Northwest National Laboratory, P.O. Box 999, Battelle for the U.S. DOE, MS K3-54, Richland, WA 99354, USA. 2 Retired. 3 Present address: School of Nuclear Engineering, Purdue University, West Lafayette, IN 47907, USA. 0265-931X/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.jenvrad.2008.06.006
the fate of 60Co resulting from a chronic application rate at a low concentration. Results from previous studies of 60Co are compared to the results from this investigation. A study was undertaken during the summer of 1996 to characterize the movement of sludge-applied 60Co on a land-application site near Oak Ridge, Tennessee, USA. Because of the types of industries in the Oak Ridge area, this sludge contained a variety of radionuclides; however, 60Co was the radionuclide selected for investigation because its presence is unique to the sludge-application site. Anaerobically digested sludge from the Oak Ridge wastewatertreatment plant was applied to 11 grassland or forested sites on the Department of Energy’s Oak Ridge Reservation beginning in 1983. This sludge typically contained 2–4% solids and was applied by either surface spraying or subsurface injection of sludge. One of these application sites, a 10.1-ha pasture designated as the Upper Hayfield site, was selected for study to determine whether all of the applied radionuclides could be accounted for through soil sampling. Sludge application by surface spraying at the Upper Hayfield site occurred intermittently from 1986 to 1995. With state and U.S. Department of Energy (DOE) oversight, the City of Oak Ridge and Oak Ridge National Laboratory worked cooperatively to minimize any adverse effects of potential accumulation of radionuclides in sludge-application areas (Gunderson et al., 1995). 2. Material and methods The presence of 60Co at this site was solely the result of surface spraying, and its occurrence at depth was investigated to determine vertical-migration rates and a mass balance between applied and measured soil activities and vegetative uptake.
1612
M.A. Smith et al. / Journal of Environmental Radioactivity 99 (2008) 1611–1616
In addition, the percentage of applied 60Co remaining in the soil, both in the top 15 cm and at depths below 15 cm, was estimated for the next few decades. Soil pH at the site ranges from about 5.0 to 6.0, and cation-exchange capacity is about 6 cmol/kg (P.R. Jardine, Ph.D., personal communication, 22 August 2007). Three soil samples from the Upper Hayfield site showed an average weight-percent distribution of 19% gravel (>2 mm), 25% sand (2 mm–63 mm), 47% silt (63–2 mm), and 9% clay (<2 mm). The soils are associated with the Knox Group and are generally formed from residuum that usually is uniform cherty silty clay (Hatcher et al., 1992). 2.1. Radionuclides applied to the site Data on sludge application at the Upper Hayfield site between 1986 and 1995 were obtained from the City of Oak Ridge (City of Oak Ridge, 1991–1996c) and the Tennessee Department of Environment and Conservation (TDEC, 1996). A ‘‘sludge-application model’’ was developed to estimate present radionuclide activity resulting from past sludge application at the site. The model used reported application dates, weekly application rates, and weekly specific activity levels in applied sludge. The reported activity of sludge applied to the site during each week of application was decay corrected to 1 June 1996, using the following formula: Asite ¼ VAsludge elW
(1)
where Asite is the 1 June 1996 activity resulting from a prior week’s sludge application (Bq); V is the volume of sludge applied during that week (L); Asludge is the reported specific activity of sludge during that week (Bq/kg); l is the decay constant for the radionuclide of interest (1/week); and W is the number of weeks between the application period and the week of 31 May–6 June 1996. In using Eq. (1), it is assumed that 1 kg of sludge is equivalent to 1 L of sludge, which is a reasonable assumption given the low solids content of the sludge (i.e., 2–4%). The 1 June 1996 activity for 60Co, as predicted by the model, was 19.3 10.3 MBq, or 0.0191 0.0102 Bq/cm2. Variability in the model’s results (i.e., sigma) may be attributed to several sources, including the radionuclide analyses, the sampling process, and actual variability in the source term. Prior to June 1988, a single sample was analyzed monthly. Subsequently, a sample was collected from the first sludge-application shipment made on a given day, from which a single 1-L composite sample was created from equal parts of that week’s daily samples. The variability in sample activity from week-to-week was minimal, and only over multi-month periods is change noted. Results from curve fitting were used to provide weekly input to the model for time periods prior to June 1988. The variability associated with the available data used in the model, no matter the cause, was estimated by calculating the standard deviation of the data about a best-fit line using the following equation:
se ¼
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi P ðy yÞ2 n2
mixed-gamma standard traceable to the National Institute of Standards and Technology. The most recent efficiency calibration used a mixed-gamma standard, QCY48, from Amersham Radiochemical Centre. A known quantity of this material was diluted in a 4-M HCl solution and placed in the 0.5-L and 1-L Marinelli beakers used for calibrating (Larsen and Cutshall, 1981). A calibration check and contamination check were performed daily for each detector. To verify results, a quality-control blank-soil sample, a laboratory-control sample, and quality-assessment samples also were counted during the investigation. As an additional verification measure, Oak Ridge National Laboratory participates routinely in the Quality Assessment Program administered by the DOE Environmental Measurements Laboratory. The program is designed to test the quality of the environmental–radiological measurements being reported by DOE contractors and subcontractors. All samples contained low concentrations of 60Co. Count times of at least 1000 min were used to reduce statistical-counting uncertainties. To determine sample activities, the detector systems first calculated the photopeak count rate, which is a combined count rate from the sample itself and any ambient background. The ambient-background count rates were subtracted to determine the net-count rate introduced by the sample. The background-count rate was determined on a monthly basis by performing at least a 50-h count with no sample on the detector.
(2)
2.4. Vertical migration model To characterize the downward movement of 60Co at the Upper Hayfield site, residence half-times of 60Co were estimated. The residence half-time is the time necessary for the activity present in a soil layer to be reduced by 50%. Ideally, the residence halftimes would be estimated by measuring radionuclide soil-profile changes over time. However, time-sequenced data were not available, so a six-compartment model was used to estimate the 60Co residence half-times based on the historical radionuclideapplication rates and the present radionuclide-depth profile. Five of the ‘‘compartments’’ represented 3-cm segments of the soil cores at depths of 0–3 , 3–6 , 6–9 , 9–12 , and 12–15 cm. The sixth ‘‘compartment’’ represented soil at a depth greater than 15 cm. Many transport processes can contribute to migration of 60Co through soil, but they can be generalized by considering diffusion and mechanical dispersion, advection, radioactive decay, retardation, and a source/sink term (Domenico and Schwartz, 1990). In the approach used for this study, radionuclide concentrations in each soil layer were assumed to be reduced only through radioactive decay and movement into deeper soil. The model estimates migration rates, but does not characterize the transport processes involved (e.g., diffusion and mechanical dispersion, advection, retardation). A visual representation of this simplified model is shown in Fig.1. The 60Co activity in the upper 0–3 cm soil layer is described by the following differential equation: dA1 ¼ D K1 A1 lA1 dt
(3)
P where se is the standard deviation; ðy yÞ2 is the sum of squared deviations from the best fit line y; and n is the number of data points. 2.2. Sample collection Soil samples were collected from the Upper Hayfield (11 cores, comprising 60 samples) and adjacent reference areas where no sludge had been applied (three cores, comprising 16 samples). Flora samples collected from the Upper Hayfield site were primarily above-ground grasses (11 samples), but included a variety of other leaves, stems, roots, and fruits (seven samples). The grass samples consisted primarily of Muhlenbergia schreberi, Setaria faberi, and Panicum dichotomiflorum. Other samples collected included Asclepias syriaca, Allium tricoccum, and various unidentified terrestrial macrophytes. Because there was no surface water on the Upper Hayfield, surface water (three samples) was collected from a pond bounded by another sludge-application site, from a depression area adjacent to the Upper Hayfield site, and from a pond near the Upper Hayfield site. Soil cores were randomly collected to a depth of at least 15 cm when possible. Eleven soil cores were collected from the Upper Hayfield site, and one soil core was collected from a depression area northeast of the Upper Hayfield site. To help characterize radionuclide movement, the soil cores were sectioned into 3-cm increments and collected in 0.5-L Marinelli beakers. The soil-coring device was a stainless-steel pipe with a cross-sectional area of 168 cm2. This coring device was used to collect the 11 soil cores from the Upper Hayfield site. Soil compaction was considered negligible because of the relatively large diameter of the pipe used (i.e., 15 cm), and this conclusion was confirmed by comparing the extracted core lengths with the borehole depths. A soil-coring device with a cross-sectional area of 49 cm2 was used to collect the soil sample from the depression area. 2.3. Sample analysis Samples were analyzed using high-purity germanium and lithium-drifted germanium coaxial-photon detector systems that were calibrated using a certified
Fig. 1. Visual representation of the six-compartment model used to estimate residence half-times at the Upper Hayfield site.
60
Co
M.A. Smith et al. / Journal of Environmental Radioactivity 99 (2008) 1611–1616
Fig. 2.
60
Co application rates used for the STELLAÒ II model of migration rates at the Upper Hayfield site.
where D is the yearly 60Co surface-application rate (Bq/cm2/yr); A1 is the 60Co activity present in the 0–3 cm layer (Bq/cm2); K1 is the fractional rate of downward migration from 0 to 3 cm layer (yr1); and l is the decay constant for 60Co (yr1). The 60 Co activity in each 3-cm layer is the mean activity found in that layer for the 11 core samples collected. The 60Co activity for the 3–6 , 6–9 , 9–12 , and 12–15 cm layers is described by the following differential equation: dAi ¼ Ki1 Ai1 Ki Ai lAi dt
(4)
where i and i 1 represent the soil layers 3–6 cm (i ¼ 2), 6–9 cm (i ¼ 3), 9–12 cm (i ¼ 4), and 12–15 cm (i ¼ 5); Ki is the fractional rate of downward migration from compartment i to i þ 1 (yr1); and Ai is the 60Co activity present in the ith layer (Bq/cm2). The 60Co activity present in the lowest layer, termed >15 cm, is described by the following differential equation: dA6 ¼ K5 A5 lA6 dt
(5)
where K5 is the fractional rate of downward migration from the 12–15 cm layer (yr1), and A5 and A6 are the 60Co activities present in the 12–15 and >15 cm layers (Bq/cm2). Radionuclide decay is the only removal process from the >15 cm layer. The differential equation system was modeled using STELLAÒ II (version 3.06), an object-oriented modeling language designed for use with the Microsoft WindowsÒ operating system. Simulations were conducted through the year 2020 with a time-step of 0.0625 yr. The time-step was sufficiently small to give accurate results and also met the software requirement of time steps of 1/n2, where n is any integer. The fourth order Runge-Kutta method was selected to approximate solutions of the ordinary differential equations. The only unknowns in the system of equations are K1–K5, the compartmental transfer rates. Yearly, decay-corrected 60Co surfaceapplication rates were estimated from weekly historical sludge-application records.
Table 1 Gamma analysis results for Core no.
Units
1 2 3 4 5 6 7 8 9 10 11 Ref. 1 Ref. 2 Ref. 3
mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2
60
1613
The 60Co-application rate input used for the model is shown in Fig. 2. The decay constant used for 60Co in the model was ln 2/5.2714 yr–1. The value of K1 was varied until the compartmental activity A1 matched the average 60Co activity in the 1–3 cm segments from the Upper Hayfield site soil samples. Then, K1 was held constant while K2 was varied until the value of A2 matched the average 60Co activity in the 3–6 cm segments of the soil samples. The process continued, with the K values being varied one at a time, in sequence, until all of the K values were determined. An attempt was made to analyze the variance of the fractional rates of downward migration (K) by running the model for each of the individual soil cores; however, solutions did not exist or were nonsensical for most of the soil cores (i.e., the solutions, when obtainable, required K values that approached zero or were greater than 1).
3. Results and discussion Results from soil, flora, and surface-water sampling conducted at or near the Upper Hayfield site were used to characterize the fate of 60Co in applied sludge. Activity data for individual soil cores, by 3-cm segment, are provided in Table 1. The discussion of the fate of 60Co will focus on its downward migration in the soil, as 60 Co was not detected in flora or surface-water samples. The minimum detectable activity for 60Co ranged from approximately 0.35 to 2.2 mBq/g-dry for flora samples and approximately 0.075 to 0.13 Bq/L for water samples. Subsurface horizontal migration was not considered because 60Co migrating away from the sample core area would be replaced by 60Co migrating into the sample core. For the mass balance, sampling results were compared to
Co of 11 soil cores collected at the Upper Hayfield site and three reference soil cores Soil core interval (cm) (0–3)
(3–6)
(6–9)
(9–12)
(12–15)
(15–18)
(18–21)
5.18 0.455 3.67 0.141 8.47 0.540 4.00 0.247 6.77 0.474 8.25 0.940 5.99 0.562 7.29 0.707 4.55 0.274 3.89 0.200 4.70 0.562 <1.38 <1.01 <1.01
6.18 0.574 10.0 0.232 4.48 0.466 10.8 0.343 10.8 0.588 3.77 0.585 5.07 0.588 5.40 0.644 6.77 0.396 12.8 0.603 1.87 0.514 <1.41 <1.05 <1.01
10.3 0.677 2.35 0.159 4.29 0.496 8.10 0.364 4.74 0.814 1.66 0.426 1.19 0.433 2.48 0.625 5.07 0.335 5.85 0.581 1.67 0.492 <1.32 <1.11 <0.958
6.11 0.607 0.500 0.158 2.47 0.570 2.35 0.302 2.28 0.488 1.28 0.477 0.825 0.377 2.11 0.559 2.53 0.307 1.52 0.500 <1.12 <1.25 * <1.04
3.40 0.474 0.403 0.135 2.12 0.592 0.821 0.267 * <1.14 <1.04 <1.05 0.992 0.370 <1.04 <1.07 <1.28 * <1.25
* <1.52 <1.12 <0.316 * * * * 0.818 0.455 0.966 0.451 * * * <1.27
* * * <1.49 * * * * * * * * * <1.11
Results are given for each 3-cm increment in units of mBq/cm2 one-sigma uncertainty. Note: * indicates sample not collected at this depth interval. Data below minimum detectable activity reported as less-than value (<).
1614
M.A. Smith et al. / Journal of Environmental Radioactivity 99 (2008) 1611–1616
Table 2 Results from the STELLAÒ II model for fractional rates of downward migration (K), residence half-times (s), and mean migration rates (v), reported with one standard deviation Depth (cm)
K (1/yr)
s (yr)
v (cm/yr)
0–3 3–6 6–9 9–12 12–15
0.160 0.049 0.166 0.077 0.165 0.113 0.169 0.147 0.115 0.171
4.33 1.34 4.18 1.94 4.20 2.88 4.10 3.57 6.03 8.98
0.692 0.214 0.718 0.332 0.714 0.490 0.731 0.636 0.498 0.742
expected values determined from the reconstructed loading history. 3.1. Vertical migration Model results for fractional rates of downward migration (K), residence half-times (s), and mean migration rates (v) are shown in Table 2, along with uncertainty estimates based on the standard deviation of the activities of the 3-cm soil layers among the 11 soil cores. As described in Section 2.4, an analysis of variance of the fractional rates of downward migration (K) was not performed because K could not be approximated for many of the individual soil cores. Note that values of K are likely variable and dependent on soil type and environmental conditions. Of additional consideration, 60Co can serve as a reasonable surrogate for a number of other radionuclides and toxic metals, including Cd, Ni, and Zn. Not only do these four cation species (i.e., Cd, Co, Ni, and Zn) share similar Kd values over a broad range of soil types, Buchter et al. (1989) also reports, in an investigation that included 13 elements and 11 soil properties, that they share similar correlations for soil pH and cation-exchange capacity. A number of studies also conclude that the mobility of these cation species is highly dependent on soil pH and cation-exchange capacity (e.g., Basta et al., 2005; Janssen et al., 1997; Yasuda et al., 1995), and King (1988) attributes 84–95% of the variability in sorbed Co to soil pH. Other less-controlling parameters for 60Co mobility include sand and clay content, dithionite-extractable Fe, exchangeable Mg, exchangeable Ca, and water content (King, 1988; Yasuda et al., 1995). Sludge-application limits for the City of Oak Ridge were based on annual and cumulative loading-rate limits for organics, inorganics, heavy metals, pathogens, and radionuclides. Historically, sludge-application rates for the City of Oak Ridge were limited by the annual and cumulative nitrogen-loading limits of 9.86 Mg/ ha/yr and 49.3 Mg/ha.
Fig. 4. Percentages of the total
60
Fig. 3. Estimate of 60Co depth profiles through the year 2020 (approximated using Eqs. (3)–(5)).
Alternatively, the 60Co-migration rate can be estimated using the elapsed time (9 yr) between the mean surface-applied activity pulses recorded in 1987 (as seen in Fig. 2) and the maximum mean activity observed in the soil cores near 4.5 cm in 1996 (as seen in Fig. 3). This approach yields an estimated 60Co-migration rate of about 0.5 cm/yr. This result compares favorably with the model results of 0.50–0.73 cm/yr, with a mean of 0.67 cm/yr. At these migration rates, 60Co passes through approximately four half-lives (half life 5.27 yr) while traveling 15 cm, leaving about 6% of the original activity in the soil by the time the 60Co reaches 15 cm. The migration rate (i.e., 0.50–0.73 cm/yr) is comparable to, but slightly higher than those reported in the literature. One study for an undisturbed grassland site reported 60Co-migration rates for three cores ranging from 0.141 to 0.166 cm/yr for very sandy soil with low clay and humus content (Bossew et al., 2004). Another study using a disturbed lysimeter to simulate crop rotation reported 60Comigration rates for four cores ranging from 0.0 to 0.3 cm/yr for soils with pH ranging from 4.6 to 7.5, sand content ranging from 17% to
Co present in 1996 that are expected to remain in the upper 15 cm of soil and to move below the upper 15 cm at the Upper Hayfield site.
M.A. Smith et al. / Journal of Environmental Radioactivity 99 (2008) 1611–1616
61%, and clay content ranging from 9% to 18% (Shinonaga et al., 2005). During the summer of 1996, about 98% of decay-corrected, surface-applied 60Co activity was detected in the upper 15 cm of soil. The STELLAÒ II model was used with Eqs. (3)–(5) to estimate how the activity profile would change over time if no further sludge was applied at the Upper Hayfield site. Plots of the estimated 60Codepth profiles through the year 2020, along with the 1996 60Codepth profile, are shown in Fig. 3. In the year 2020, the 60Co activity is estimated to be only 4% of the 1996 value, and approximately one-half of that percentage is expected to be in the upper 15 cm. Fig. 4 shows the percentages of the total 60Co measured in 1996 that are expected to remain in the upper 15 cm of soil and to move below the upper 15 cm of soil through the year 2036, assuming no further sludge application. Activity changes in the upper 15 cm of soil are the result of radioactive decay and migration loss to soil below the upper 15 cm. Changes below the upper 15 cm of soil are the result of radioactive decay and migration gain from the upper 15 cm of soil.
3.2. Mass balance 60 Co was detected at the Upper Hayfield site but not at the reference sites. This finding simplified development of the mass balance for 60Co, as all the 60Co detected at the Upper Hayfield site could be assumed to result from sludge-application activities. Results from soil sampling at the Upper Hayfield site indicate that about 98%, 0.020 0.011 Bq/cm2, of the 60Co resides in the upper 15 cm. No 60Co was detected in flora or water samples collected at or near the Upper Hayfield site, giving a site activity for 60Co of 0.020 0.011 Bq/cm2. This result compares favorably with the activity of 0.019 0.010 Bq/cm2 predicted by the sludge-application model described in Section 2.1. This approach does not consider the airborne-dust pathway as a removal process, which may be an important consideration in arid or sparsely vegetated regions. However, this location was highly vegetated, receives an annual average precipitation of 137 cm, and has an average annual transpiration of 75 cm (Hatcher, 1989). In addition, 60Co is considered to be quickly sorbed to soil. As such, only sludge that lands on and dries on the flora would be susceptible to airborne loss in this environment.
4. Conclusions To characterize the fate of radionuclides in soils at the Upper Hayfield site at Oak Ridge, Tennessee, soil cores were analyzed for radionuclides in separate 3-cm increments. Efforts focused on 60Co, because 60Co was the only radionuclide for which all activity at the site could be attributed exclusively to sludge-application activities. Results from soil sampling at the Upper Hayfield site showed that about 98%, 0.020 0.011 Bq/cm2, of 60Co from the sludge remained in the upper 15 cm of soil. This compared favorably with the 60Co activity calculated by the sludge-application model of 0.019 0.010 Bq/cm2. No 60Co was detected in flora or surfacewater samples from the sludge-application site. Modeling showed that 60Co moves downward at a rate of about 0.7 cm/yr in the upper 15 cm. At this migration rate, 60Co passes through approximately four half-lives while traveling 15 cm, leaving only 1/16, or about 6%, of the original activity in the soil by the time 60Co reaches 15 cm. Results of this study indicate that for loading rates and conditions similar to those at the Upper Hayfield site, 60Co from land-applied sludge remains in the soil, and about 94% of the original 60Co decays to 60Ni before reaching a depth of 15 cm. The loss of 60Co below 15 cm results solely from radioactive decay.
1615
Acknowledgements This research was performed under subcontract ERD-88-793 with the City of Oak Ridge, Tennessee, under Lockheed Martin Energy Research, Inc. contract DE-AC05-96OR22464 with the U.S. Department of Energy, and under appointment to the Applied Health Physics Fellowship Program administered by the Oak Ridge Institute for Science and Education for the U.S. Department of Energy.
References Anspaugh, L.R., Simon, S.L., Gordeev, K.I., Likhtarev, I.A., Maxwell, R.M., Shinkarev, S.M., 2002. Movement of radionuclides in terrestrial ecosystem by physical processes. Health Physics 82, 669–679. Arapis, G., Petrayev, E., Shagalova, E., Zhukova, O., Sokolik, G., Ivanova, T., 1997. Effective migration velocity of 137Cs and 90Sr as a function of the type of soils in Belarus. Journal of Environmental Radioactivity 34, 171–185. Baes III, C.F., Sharp, R.D., 1983. A proposal for estimation of soil leaching and leaching constants for use in assessment models. Journal of Environmental Quality 12, 17–28. Basta, N.T., Ryan, J.A., Chaney, R.L., 2005. Trace element chemistry in residualtreated soil: key concepts and metal bioavailability. Journal of Environmental Quality 34, 49–63. Bastian, R.K., Bachmaier, J.T., Schmidt, D.W., Salomon, S.N., Jones, A., Chiu, W.A., Setlow, L.W., Wolbarst, A.B., Yu, C., Goodman, J., Lenhart, T., 2005. Radioactive materials in biosolids: National survey, dose modeling, and publicly owned treatment works (POTW) guidance. Journal of Environmental Quality 34, 64–74. Boone, F.W., Kantelo, M.V., Mayer, P.G., Palms, J.M., 1985. Residence half-times of 129I in undisturbed surface soils based on measured soil concentration profiles. Health Physics 48, 401–413. Bossew, P., Gastberger, M., Gohla, H., Hofer, P., Hubmer, A., 2004. Vertical distribution of radionuclides in soil of a grassland site in Chernobyl exclusion zone. Journal of Environmental Radioactivity 73, 87–99. Buchter, B., Davidoff, B., Amacher, M.C., Hinz, C., Iskandar, I.K., Selim, H.M., 1989. Correlation of Freundlich Kd and n retention parameters with soils and elements. Soil Science 148, 370–379. Bunzl, K., Fo¨rster, H., Kracke, W., Schimmack, W., 1994. Residence times of fallout 239þ240 Pu, 238Pu, 241Am and 137Cs in the upper horizons of an undisturbed grassland soil. Journal of Environmental Radioactivity 22, 11–27. City of Oak Ridge, 28 January 1991. City of Oak Ridge 1990 Land Application Report: Application of Sludge to Department of Energy Reservation. City of Oak Ridge, 6 April 1992. City of Oak Ridge 1991 Land Application Report: Application of Sludge to Department of Energy Reservation. City of Oak Ridge, March 1993. City of Oak Ridge Waste Water Treatment Plant 1992 Biosolids Management Report. City of Oak Ridge, 15 April 1994. City of Oak Ridge 1993 State of Tennessee Biosolids Management Report. City of Oak Ridge, 17 March 1995. City of Oak Ridge 1994 State Biosolids Management Report. City of Oak Ridge, 15 March 1996a. City of Oak Ridge 1995 State Biosolids Management Report. City of Oak Ridge, 1996b. Digested Sludge Removed Storage Digester, 1986–1989. Oak Ridge Wastewater Treatment Plant Records. Photocopied. City of Oak Ridge, 1996c. Annual Sludge Data Summary 1988 thru 1993, Radionuclide Data for Oak Ridge Digested Sludge. Oak Ridge Wastewater Treatment Plant Records. Photocopied. Domenico, P.A., Schwartz, F.W., 1990. Physical and Chemical Hydrogeology. John Wiley & Sons, New York. Gunderson, C.A., Boston, H.L., Van Meigroet, H., Morris, J.L., Larsen, I.L., Walzer, A.E., Adler, T.C., Bradburn, D.M., Huq, M., 1995. Sludge Application and Monitoring Program on the Oak Ridge Reservation, 1986–1993. U.S. Department of Energy. ORNL/TM-11601. Hatcher, R.D., 1989. Field Guide and Perspective on the Geology and Hydrology of the Oak Ridge Reservation. University of Tennessee, Oak Ridge, Tennessee. Hatcher Jr., R.D., Lemiszki, P.J., Dreier, R.B., Ketelle, R.H., Lee, R.R., Leitzke, D.A., McMaster, W.M., Foreman, J.L., Lee, S.Y., 1992. Status Report on the Geology of the Oak Ridge Reservation. U.S. Department of Energy. ORNL/TM-12074. Janssen, R., Peijnenburg, W., Posthuma, L., Van den Hoop, M., 1997. Equilibrium partitioning of heavy metals in Dutch field soils. I. Relationship between metal partition coefficients and soil characteristics. Environmental Toxicology and Chemistry 16, 2470–2478. King, L.D., 1988. Retention of metals by several soils of the southeastern United States. Journal of Environmental Quality 17, 239–246. Larsen, I.L., Cutshall, N.H., 1981. Direct determination of 7Be in sediments. Earth and Planetary Science Letters 54, 379–384. O’Connor, G.A., Elliot, H.A., Basta, N.T., Bastian, R.K., Pierzynski, G.M., Sims, R.C., Smith Jr., J.E., 2005. Sustainable land application: an overview. Journal of Environmental Quality 34, 7–17. Shinonaga, T., Schimmack, W., Gerzabek, M.H., 2005. Vertical migration of 60Co, 137 Cs and 226Ra in agricultural soils as observed in lysimeters under crop rotation. Journal of Environmental Radioactivity 79, 93–106.
1616
M.A. Smith et al. / Journal of Environmental Radioactivity 99 (2008) 1611–1616
Tanaka, T., Ohnuki, T., 1996. Colloidal migration behavior of radionuclides sorbed on mobile fine soil particles through a sand layer. Journal of Nuclear Science & Technology 33, 62–68. TDEC (Tennessee Department of Environment and Conservation), 1996. TDEC Routine Analysis of ORWWTP Sludge 1986–1988. Division of Radiological Health. Photocopied. Whicker, F.W., Pinder, J.E., 2002. Food chains and biogeochemical pathways: contributions of fallout and other radiotracers. Health Physics 82, 680–689.
Wolbarst, A.B., Chiu, W.A., Yu, C., Aiello, K., Bachmaier, J.T., Bastian, R.K., Cheng, J.-J., Goodman, J., Hogan, R., Jones, A.R., Kamboj, S., Lenhartt, T., Ott, W.R., Rubin, A., Salomon, S.N., Schmidt, D.W., Setlow, L.W., 2006. Radioactive materials in biosolids: dose modeling. Health Physics 90, 16–30. Yasuda, H., Uchida, S., Muramatsu, Y., Yoshida, S., 1995. Sorption of manganese, cobalt, zinc, strontium, and cesium onto agricultural soils: statistical analysis on effects of soil properties. Water, Air, and Soil Pollution 83, 85–96.
Contents lists available at ScienceDirect
Journal of Environmental Radioactivity journal homepage: www.elsevier.com/locate/jenvrad
Fate of
60
Co at a sludge land application site
Michael A. Smith a, *,1, Ingvar L. Larsen b, 2, Audeen W. Fentiman a, 3 a b
Environmental Science Graduate Program, Ohio State University, Columbus, OH 43210, USA Environmental Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 29 November 2007 Received in revised form 4 June 2008 Accepted 9 June 2008 Available online 22 July 2008
Vertical distributions of 60Co are determined in soil cores obtained from a 10-ha grassland, where anaerobically digested sludge was applied by surface spraying from 1986 to 1995 on the U.S. Department of Energy’s Oak Ridge Reservation. These results, along with historical application records, are used to estimate vertical-migration rates and perform a mass balance. The presence of 60Co results solely from the sludge-application process. Soil, vegetation, and surface-water samples were collected. Eleven soil cores were sectioned into 3-cm increments and analyzed by gamma-ray spectrometry. No 60Co was detected in the vegetation or water samples. The downward migration rate of 60Co in the upper 15 cm of soil ranged from 0.50 to 0.73 cm/yr. About 98%, 0.020 0.011 Bq/cm2, of 60Co remained in the upper 15 cm of soil, which compared favorably with the expected 60Co activity based on historical records of 0.019 0.010 Bq/cm2. Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Biosolid Heavy metal Vertical migration Mass balance Radionuclide 60 Co
1. Introduction The fate of land-applied radionuclides contained in wastewatertreatment sludge has received considerable recent attention (e.g., Bastian et al., 2005; O’Connor et al., 2005; Wolbarst et al., 2006), and the practice of land application of these biosolids is expected to continue to grow because of the perceived economic and societal benefits (Basta et al., 2005). Radionuclides are either present naturally or enter the wastewater stream from numerous anthropogenic sources, such as combustion, fertilizer runoff, mining and smelting operations, leachate from disposal facilities, industrial byproducts, research institutions, and hospitals, as well as from atmospheric fallout. Historically, many investigators have studied the fate of radionuclides in the environment or laboratory, considering wind, soil, surface water, and groundwater, with numerous studies resulting from the 1986 Chernobyl release (e.g., Anspaugh et al., 2002; Arapis et al., 1997; Baes and Sharp, 1983; Boone et al., 1985; Bunzl et al., 1994; Tanaka and Ohnuki, 1996; Whicker and Pinder, 2002). Many of these studies involved acute application rates with high concentrations of radionuclides. In contrast, the study conducted at the Department of Energy’s (DOE) Oak Ridge Reservation focused on * Corresponding author. Tel.: þ1 509 372 4788; fax: þ1 509 375 2019. E-mail address: [email protected] (M.A. Smith). 1 Present address: Pacific Northwest National Laboratory, P.O. Box 999, Battelle for the U.S. DOE, MS K3-54, Richland, WA 99354, USA. 2 Retired. 3 Present address: School of Nuclear Engineering, Purdue University, West Lafayette, IN 47907, USA. 0265-931X/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.jenvrad.2008.06.006
the fate of 60Co resulting from a chronic application rate at a low concentration. Results from previous studies of 60Co are compared to the results from this investigation. A study was undertaken during the summer of 1996 to characterize the movement of sludge-applied 60Co on a land-application site near Oak Ridge, Tennessee, USA. Because of the types of industries in the Oak Ridge area, this sludge contained a variety of radionuclides; however, 60Co was the radionuclide selected for investigation because its presence is unique to the sludge-application site. Anaerobically digested sludge from the Oak Ridge wastewatertreatment plant was applied to 11 grassland or forested sites on the Department of Energy’s Oak Ridge Reservation beginning in 1983. This sludge typically contained 2–4% solids and was applied by either surface spraying or subsurface injection of sludge. One of these application sites, a 10.1-ha pasture designated as the Upper Hayfield site, was selected for study to determine whether all of the applied radionuclides could be accounted for through soil sampling. Sludge application by surface spraying at the Upper Hayfield site occurred intermittently from 1986 to 1995. With state and U.S. Department of Energy (DOE) oversight, the City of Oak Ridge and Oak Ridge National Laboratory worked cooperatively to minimize any adverse effects of potential accumulation of radionuclides in sludge-application areas (Gunderson et al., 1995). 2. Material and methods The presence of 60Co at this site was solely the result of surface spraying, and its occurrence at depth was investigated to determine vertical-migration rates and a mass balance between applied and measured soil activities and vegetative uptake.
1612
M.A. Smith et al. / Journal of Environmental Radioactivity 99 (2008) 1611–1616
In addition, the percentage of applied 60Co remaining in the soil, both in the top 15 cm and at depths below 15 cm, was estimated for the next few decades. Soil pH at the site ranges from about 5.0 to 6.0, and cation-exchange capacity is about 6 cmol/kg (P.R. Jardine, Ph.D., personal communication, 22 August 2007). Three soil samples from the Upper Hayfield site showed an average weight-percent distribution of 19% gravel (>2 mm), 25% sand (2 mm–63 mm), 47% silt (63–2 mm), and 9% clay (<2 mm). The soils are associated with the Knox Group and are generally formed from residuum that usually is uniform cherty silty clay (Hatcher et al., 1992). 2.1. Radionuclides applied to the site Data on sludge application at the Upper Hayfield site between 1986 and 1995 were obtained from the City of Oak Ridge (City of Oak Ridge, 1991–1996c) and the Tennessee Department of Environment and Conservation (TDEC, 1996). A ‘‘sludge-application model’’ was developed to estimate present radionuclide activity resulting from past sludge application at the site. The model used reported application dates, weekly application rates, and weekly specific activity levels in applied sludge. The reported activity of sludge applied to the site during each week of application was decay corrected to 1 June 1996, using the following formula: Asite ¼ VAsludge elW
(1)
where Asite is the 1 June 1996 activity resulting from a prior week’s sludge application (Bq); V is the volume of sludge applied during that week (L); Asludge is the reported specific activity of sludge during that week (Bq/kg); l is the decay constant for the radionuclide of interest (1/week); and W is the number of weeks between the application period and the week of 31 May–6 June 1996. In using Eq. (1), it is assumed that 1 kg of sludge is equivalent to 1 L of sludge, which is a reasonable assumption given the low solids content of the sludge (i.e., 2–4%). The 1 June 1996 activity for 60Co, as predicted by the model, was 19.3 10.3 MBq, or 0.0191 0.0102 Bq/cm2. Variability in the model’s results (i.e., sigma) may be attributed to several sources, including the radionuclide analyses, the sampling process, and actual variability in the source term. Prior to June 1988, a single sample was analyzed monthly. Subsequently, a sample was collected from the first sludge-application shipment made on a given day, from which a single 1-L composite sample was created from equal parts of that week’s daily samples. The variability in sample activity from week-to-week was minimal, and only over multi-month periods is change noted. Results from curve fitting were used to provide weekly input to the model for time periods prior to June 1988. The variability associated with the available data used in the model, no matter the cause, was estimated by calculating the standard deviation of the data about a best-fit line using the following equation:
se ¼
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi P ðy yÞ2 n2
mixed-gamma standard traceable to the National Institute of Standards and Technology. The most recent efficiency calibration used a mixed-gamma standard, QCY48, from Amersham Radiochemical Centre. A known quantity of this material was diluted in a 4-M HCl solution and placed in the 0.5-L and 1-L Marinelli beakers used for calibrating (Larsen and Cutshall, 1981). A calibration check and contamination check were performed daily for each detector. To verify results, a quality-control blank-soil sample, a laboratory-control sample, and quality-assessment samples also were counted during the investigation. As an additional verification measure, Oak Ridge National Laboratory participates routinely in the Quality Assessment Program administered by the DOE Environmental Measurements Laboratory. The program is designed to test the quality of the environmental–radiological measurements being reported by DOE contractors and subcontractors. All samples contained low concentrations of 60Co. Count times of at least 1000 min were used to reduce statistical-counting uncertainties. To determine sample activities, the detector systems first calculated the photopeak count rate, which is a combined count rate from the sample itself and any ambient background. The ambient-background count rates were subtracted to determine the net-count rate introduced by the sample. The background-count rate was determined on a monthly basis by performing at least a 50-h count with no sample on the detector.
(2)
2.4. Vertical migration model To characterize the downward movement of 60Co at the Upper Hayfield site, residence half-times of 60Co were estimated. The residence half-time is the time necessary for the activity present in a soil layer to be reduced by 50%. Ideally, the residence halftimes would be estimated by measuring radionuclide soil-profile changes over time. However, time-sequenced data were not available, so a six-compartment model was used to estimate the 60Co residence half-times based on the historical radionuclideapplication rates and the present radionuclide-depth profile. Five of the ‘‘compartments’’ represented 3-cm segments of the soil cores at depths of 0–3 , 3–6 , 6–9 , 9–12 , and 12–15 cm. The sixth ‘‘compartment’’ represented soil at a depth greater than 15 cm. Many transport processes can contribute to migration of 60Co through soil, but they can be generalized by considering diffusion and mechanical dispersion, advection, radioactive decay, retardation, and a source/sink term (Domenico and Schwartz, 1990). In the approach used for this study, radionuclide concentrations in each soil layer were assumed to be reduced only through radioactive decay and movement into deeper soil. The model estimates migration rates, but does not characterize the transport processes involved (e.g., diffusion and mechanical dispersion, advection, retardation). A visual representation of this simplified model is shown in Fig.1. The 60Co activity in the upper 0–3 cm soil layer is described by the following differential equation: dA1 ¼ D K1 A1 lA1 dt
(3)
P where se is the standard deviation; ðy yÞ2 is the sum of squared deviations from the best fit line y; and n is the number of data points. 2.2. Sample collection Soil samples were collected from the Upper Hayfield (11 cores, comprising 60 samples) and adjacent reference areas where no sludge had been applied (three cores, comprising 16 samples). Flora samples collected from the Upper Hayfield site were primarily above-ground grasses (11 samples), but included a variety of other leaves, stems, roots, and fruits (seven samples). The grass samples consisted primarily of Muhlenbergia schreberi, Setaria faberi, and Panicum dichotomiflorum. Other samples collected included Asclepias syriaca, Allium tricoccum, and various unidentified terrestrial macrophytes. Because there was no surface water on the Upper Hayfield, surface water (three samples) was collected from a pond bounded by another sludge-application site, from a depression area adjacent to the Upper Hayfield site, and from a pond near the Upper Hayfield site. Soil cores were randomly collected to a depth of at least 15 cm when possible. Eleven soil cores were collected from the Upper Hayfield site, and one soil core was collected from a depression area northeast of the Upper Hayfield site. To help characterize radionuclide movement, the soil cores were sectioned into 3-cm increments and collected in 0.5-L Marinelli beakers. The soil-coring device was a stainless-steel pipe with a cross-sectional area of 168 cm2. This coring device was used to collect the 11 soil cores from the Upper Hayfield site. Soil compaction was considered negligible because of the relatively large diameter of the pipe used (i.e., 15 cm), and this conclusion was confirmed by comparing the extracted core lengths with the borehole depths. A soil-coring device with a cross-sectional area of 49 cm2 was used to collect the soil sample from the depression area. 2.3. Sample analysis Samples were analyzed using high-purity germanium and lithium-drifted germanium coaxial-photon detector systems that were calibrated using a certified
Fig. 1. Visual representation of the six-compartment model used to estimate residence half-times at the Upper Hayfield site.
60
Co
M.A. Smith et al. / Journal of Environmental Radioactivity 99 (2008) 1611–1616
Fig. 2.
60
Co application rates used for the STELLAÒ II model of migration rates at the Upper Hayfield site.
where D is the yearly 60Co surface-application rate (Bq/cm2/yr); A1 is the 60Co activity present in the 0–3 cm layer (Bq/cm2); K1 is the fractional rate of downward migration from 0 to 3 cm layer (yr1); and l is the decay constant for 60Co (yr1). The 60 Co activity in each 3-cm layer is the mean activity found in that layer for the 11 core samples collected. The 60Co activity for the 3–6 , 6–9 , 9–12 , and 12–15 cm layers is described by the following differential equation: dAi ¼ Ki1 Ai1 Ki Ai lAi dt
(4)
where i and i 1 represent the soil layers 3–6 cm (i ¼ 2), 6–9 cm (i ¼ 3), 9–12 cm (i ¼ 4), and 12–15 cm (i ¼ 5); Ki is the fractional rate of downward migration from compartment i to i þ 1 (yr1); and Ai is the 60Co activity present in the ith layer (Bq/cm2). The 60Co activity present in the lowest layer, termed >15 cm, is described by the following differential equation: dA6 ¼ K5 A5 lA6 dt
(5)
where K5 is the fractional rate of downward migration from the 12–15 cm layer (yr1), and A5 and A6 are the 60Co activities present in the 12–15 and >15 cm layers (Bq/cm2). Radionuclide decay is the only removal process from the >15 cm layer. The differential equation system was modeled using STELLAÒ II (version 3.06), an object-oriented modeling language designed for use with the Microsoft WindowsÒ operating system. Simulations were conducted through the year 2020 with a time-step of 0.0625 yr. The time-step was sufficiently small to give accurate results and also met the software requirement of time steps of 1/n2, where n is any integer. The fourth order Runge-Kutta method was selected to approximate solutions of the ordinary differential equations. The only unknowns in the system of equations are K1–K5, the compartmental transfer rates. Yearly, decay-corrected 60Co surfaceapplication rates were estimated from weekly historical sludge-application records.
Table 1 Gamma analysis results for Core no.
Units
1 2 3 4 5 6 7 8 9 10 11 Ref. 1 Ref. 2 Ref. 3
mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2 mBq/cm2
60
1613
The 60Co-application rate input used for the model is shown in Fig. 2. The decay constant used for 60Co in the model was ln 2/5.2714 yr–1. The value of K1 was varied until the compartmental activity A1 matched the average 60Co activity in the 1–3 cm segments from the Upper Hayfield site soil samples. Then, K1 was held constant while K2 was varied until the value of A2 matched the average 60Co activity in the 3–6 cm segments of the soil samples. The process continued, with the K values being varied one at a time, in sequence, until all of the K values were determined. An attempt was made to analyze the variance of the fractional rates of downward migration (K) by running the model for each of the individual soil cores; however, solutions did not exist or were nonsensical for most of the soil cores (i.e., the solutions, when obtainable, required K values that approached zero or were greater than 1).
3. Results and discussion Results from soil, flora, and surface-water sampling conducted at or near the Upper Hayfield site were used to characterize the fate of 60Co in applied sludge. Activity data for individual soil cores, by 3-cm segment, are provided in Table 1. The discussion of the fate of 60Co will focus on its downward migration in the soil, as 60 Co was not detected in flora or surface-water samples. The minimum detectable activity for 60Co ranged from approximately 0.35 to 2.2 mBq/g-dry for flora samples and approximately 0.075 to 0.13 Bq/L for water samples. Subsurface horizontal migration was not considered because 60Co migrating away from the sample core area would be replaced by 60Co migrating into the sample core. For the mass balance, sampling results were compared to
Co of 11 soil cores collected at the Upper Hayfield site and three reference soil cores Soil core interval (cm) (0–3)
(3–6)
(6–9)
(9–12)
(12–15)
(15–18)
(18–21)
5.18 0.455 3.67 0.141 8.47 0.540 4.00 0.247 6.77 0.474 8.25 0.940 5.99 0.562 7.29 0.707 4.55 0.274 3.89 0.200 4.70 0.562 <1.38 <1.01 <1.01
6.18 0.574 10.0 0.232 4.48 0.466 10.8 0.343 10.8 0.588 3.77 0.585 5.07 0.588 5.40 0.644 6.77 0.396 12.8 0.603 1.87 0.514 <1.41 <1.05 <1.01
10.3 0.677 2.35 0.159 4.29 0.496 8.10 0.364 4.74 0.814 1.66 0.426 1.19 0.433 2.48 0.625 5.07 0.335 5.85 0.581 1.67 0.492 <1.32 <1.11 <0.958
6.11 0.607 0.500 0.158 2.47 0.570 2.35 0.302 2.28 0.488 1.28 0.477 0.825 0.377 2.11 0.559 2.53 0.307 1.52 0.500 <1.12 <1.25 * <1.04
3.40 0.474 0.403 0.135 2.12 0.592 0.821 0.267 * <1.14 <1.04 <1.05 0.992 0.370 <1.04 <1.07 <1.28 * <1.25
* <1.52 <1.12 <0.316 * * * * 0.818 0.455 0.966 0.451 * * * <1.27
* * * <1.49 * * * * * * * * * <1.11
Results are given for each 3-cm increment in units of mBq/cm2 one-sigma uncertainty. Note: * indicates sample not collected at this depth interval. Data below minimum detectable activity reported as less-than value (<).
1614
M.A. Smith et al. / Journal of Environmental Radioactivity 99 (2008) 1611–1616
Table 2 Results from the STELLAÒ II model for fractional rates of downward migration (K), residence half-times (s), and mean migration rates (v), reported with one standard deviation Depth (cm)
K (1/yr)
s (yr)
v (cm/yr)
0–3 3–6 6–9 9–12 12–15
0.160 0.049 0.166 0.077 0.165 0.113 0.169 0.147 0.115 0.171
4.33 1.34 4.18 1.94 4.20 2.88 4.10 3.57 6.03 8.98
0.692 0.214 0.718 0.332 0.714 0.490 0.731 0.636 0.498 0.742
expected values determined from the reconstructed loading history. 3.1. Vertical migration Model results for fractional rates of downward migration (K), residence half-times (s), and mean migration rates (v) are shown in Table 2, along with uncertainty estimates based on the standard deviation of the activities of the 3-cm soil layers among the 11 soil cores. As described in Section 2.4, an analysis of variance of the fractional rates of downward migration (K) was not performed because K could not be approximated for many of the individual soil cores. Note that values of K are likely variable and dependent on soil type and environmental conditions. Of additional consideration, 60Co can serve as a reasonable surrogate for a number of other radionuclides and toxic metals, including Cd, Ni, and Zn. Not only do these four cation species (i.e., Cd, Co, Ni, and Zn) share similar Kd values over a broad range of soil types, Buchter et al. (1989) also reports, in an investigation that included 13 elements and 11 soil properties, that they share similar correlations for soil pH and cation-exchange capacity. A number of studies also conclude that the mobility of these cation species is highly dependent on soil pH and cation-exchange capacity (e.g., Basta et al., 2005; Janssen et al., 1997; Yasuda et al., 1995), and King (1988) attributes 84–95% of the variability in sorbed Co to soil pH. Other less-controlling parameters for 60Co mobility include sand and clay content, dithionite-extractable Fe, exchangeable Mg, exchangeable Ca, and water content (King, 1988; Yasuda et al., 1995). Sludge-application limits for the City of Oak Ridge were based on annual and cumulative loading-rate limits for organics, inorganics, heavy metals, pathogens, and radionuclides. Historically, sludge-application rates for the City of Oak Ridge were limited by the annual and cumulative nitrogen-loading limits of 9.86 Mg/ ha/yr and 49.3 Mg/ha.
Fig. 4. Percentages of the total
60
Fig. 3. Estimate of 60Co depth profiles through the year 2020 (approximated using Eqs. (3)–(5)).
Alternatively, the 60Co-migration rate can be estimated using the elapsed time (9 yr) between the mean surface-applied activity pulses recorded in 1987 (as seen in Fig. 2) and the maximum mean activity observed in the soil cores near 4.5 cm in 1996 (as seen in Fig. 3). This approach yields an estimated 60Co-migration rate of about 0.5 cm/yr. This result compares favorably with the model results of 0.50–0.73 cm/yr, with a mean of 0.67 cm/yr. At these migration rates, 60Co passes through approximately four half-lives (half life 5.27 yr) while traveling 15 cm, leaving about 6% of the original activity in the soil by the time the 60Co reaches 15 cm. The migration rate (i.e., 0.50–0.73 cm/yr) is comparable to, but slightly higher than those reported in the literature. One study for an undisturbed grassland site reported 60Co-migration rates for three cores ranging from 0.141 to 0.166 cm/yr for very sandy soil with low clay and humus content (Bossew et al., 2004). Another study using a disturbed lysimeter to simulate crop rotation reported 60Comigration rates for four cores ranging from 0.0 to 0.3 cm/yr for soils with pH ranging from 4.6 to 7.5, sand content ranging from 17% to
Co present in 1996 that are expected to remain in the upper 15 cm of soil and to move below the upper 15 cm at the Upper Hayfield site.
M.A. Smith et al. / Journal of Environmental Radioactivity 99 (2008) 1611–1616
61%, and clay content ranging from 9% to 18% (Shinonaga et al., 2005). During the summer of 1996, about 98% of decay-corrected, surface-applied 60Co activity was detected in the upper 15 cm of soil. The STELLAÒ II model was used with Eqs. (3)–(5) to estimate how the activity profile would change over time if no further sludge was applied at the Upper Hayfield site. Plots of the estimated 60Codepth profiles through the year 2020, along with the 1996 60Codepth profile, are shown in Fig. 3. In the year 2020, the 60Co activity is estimated to be only 4% of the 1996 value, and approximately one-half of that percentage is expected to be in the upper 15 cm. Fig. 4 shows the percentages of the total 60Co measured in 1996 that are expected to remain in the upper 15 cm of soil and to move below the upper 15 cm of soil through the year 2036, assuming no further sludge application. Activity changes in the upper 15 cm of soil are the result of radioactive decay and migration loss to soil below the upper 15 cm. Changes below the upper 15 cm of soil are the result of radioactive decay and migration gain from the upper 15 cm of soil.
3.2. Mass balance 60 Co was detected at the Upper Hayfield site but not at the reference sites. This finding simplified development of the mass balance for 60Co, as all the 60Co detected at the Upper Hayfield site could be assumed to result from sludge-application activities. Results from soil sampling at the Upper Hayfield site indicate that about 98%, 0.020 0.011 Bq/cm2, of the 60Co resides in the upper 15 cm. No 60Co was detected in flora or water samples collected at or near the Upper Hayfield site, giving a site activity for 60Co of 0.020 0.011 Bq/cm2. This result compares favorably with the activity of 0.019 0.010 Bq/cm2 predicted by the sludge-application model described in Section 2.1. This approach does not consider the airborne-dust pathway as a removal process, which may be an important consideration in arid or sparsely vegetated regions. However, this location was highly vegetated, receives an annual average precipitation of 137 cm, and has an average annual transpiration of 75 cm (Hatcher, 1989). In addition, 60Co is considered to be quickly sorbed to soil. As such, only sludge that lands on and dries on the flora would be susceptible to airborne loss in this environment.
4. Conclusions To characterize the fate of radionuclides in soils at the Upper Hayfield site at Oak Ridge, Tennessee, soil cores were analyzed for radionuclides in separate 3-cm increments. Efforts focused on 60Co, because 60Co was the only radionuclide for which all activity at the site could be attributed exclusively to sludge-application activities. Results from soil sampling at the Upper Hayfield site showed that about 98%, 0.020 0.011 Bq/cm2, of 60Co from the sludge remained in the upper 15 cm of soil. This compared favorably with the 60Co activity calculated by the sludge-application model of 0.019 0.010 Bq/cm2. No 60Co was detected in flora or surfacewater samples from the sludge-application site. Modeling showed that 60Co moves downward at a rate of about 0.7 cm/yr in the upper 15 cm. At this migration rate, 60Co passes through approximately four half-lives while traveling 15 cm, leaving only 1/16, or about 6%, of the original activity in the soil by the time 60Co reaches 15 cm. Results of this study indicate that for loading rates and conditions similar to those at the Upper Hayfield site, 60Co from land-applied sludge remains in the soil, and about 94% of the original 60Co decays to 60Ni before reaching a depth of 15 cm. The loss of 60Co below 15 cm results solely from radioactive decay.
1615
Acknowledgements This research was performed under subcontract ERD-88-793 with the City of Oak Ridge, Tennessee, under Lockheed Martin Energy Research, Inc. contract DE-AC05-96OR22464 with the U.S. Department of Energy, and under appointment to the Applied Health Physics Fellowship Program administered by the Oak Ridge Institute for Science and Education for the U.S. Department of Energy.
References Anspaugh, L.R., Simon, S.L., Gordeev, K.I., Likhtarev, I.A., Maxwell, R.M., Shinkarev, S.M., 2002. Movement of radionuclides in terrestrial ecosystem by physical processes. Health Physics 82, 669–679. Arapis, G., Petrayev, E., Shagalova, E., Zhukova, O., Sokolik, G., Ivanova, T., 1997. Effective migration velocity of 137Cs and 90Sr as a function of the type of soils in Belarus. Journal of Environmental Radioactivity 34, 171–185. Baes III, C.F., Sharp, R.D., 1983. A proposal for estimation of soil leaching and leaching constants for use in assessment models. Journal of Environmental Quality 12, 17–28. Basta, N.T., Ryan, J.A., Chaney, R.L., 2005. Trace element chemistry in residualtreated soil: key concepts and metal bioavailability. Journal of Environmental Quality 34, 49–63. Bastian, R.K., Bachmaier, J.T., Schmidt, D.W., Salomon, S.N., Jones, A., Chiu, W.A., Setlow, L.W., Wolbarst, A.B., Yu, C., Goodman, J., Lenhart, T., 2005. Radioactive materials in biosolids: National survey, dose modeling, and publicly owned treatment works (POTW) guidance. Journal of Environmental Quality 34, 64–74. Boone, F.W., Kantelo, M.V., Mayer, P.G., Palms, J.M., 1985. Residence half-times of 129I in undisturbed surface soils based on measured soil concentration profiles. Health Physics 48, 401–413. Bossew, P., Gastberger, M., Gohla, H., Hofer, P., Hubmer, A., 2004. Vertical distribution of radionuclides in soil of a grassland site in Chernobyl exclusion zone. Journal of Environmental Radioactivity 73, 87–99. Buchter, B., Davidoff, B., Amacher, M.C., Hinz, C., Iskandar, I.K., Selim, H.M., 1989. Correlation of Freundlich Kd and n retention parameters with soils and elements. Soil Science 148, 370–379. Bunzl, K., Fo¨rster, H., Kracke, W., Schimmack, W., 1994. Residence times of fallout 239þ240 Pu, 238Pu, 241Am and 137Cs in the upper horizons of an undisturbed grassland soil. Journal of Environmental Radioactivity 22, 11–27. City of Oak Ridge, 28 January 1991. City of Oak Ridge 1990 Land Application Report: Application of Sludge to Department of Energy Reservation. City of Oak Ridge, 6 April 1992. City of Oak Ridge 1991 Land Application Report: Application of Sludge to Department of Energy Reservation. City of Oak Ridge, March 1993. City of Oak Ridge Waste Water Treatment Plant 1992 Biosolids Management Report. City of Oak Ridge, 15 April 1994. City of Oak Ridge 1993 State of Tennessee Biosolids Management Report. City of Oak Ridge, 17 March 1995. City of Oak Ridge 1994 State Biosolids Management Report. City of Oak Ridge, 15 March 1996a. City of Oak Ridge 1995 State Biosolids Management Report. City of Oak Ridge, 1996b. Digested Sludge Removed Storage Digester, 1986–1989. Oak Ridge Wastewater Treatment Plant Records. Photocopied. City of Oak Ridge, 1996c. Annual Sludge Data Summary 1988 thru 1993, Radionuclide Data for Oak Ridge Digested Sludge. Oak Ridge Wastewater Treatment Plant Records. Photocopied. Domenico, P.A., Schwartz, F.W., 1990. Physical and Chemical Hydrogeology. John Wiley & Sons, New York. Gunderson, C.A., Boston, H.L., Van Meigroet, H., Morris, J.L., Larsen, I.L., Walzer, A.E., Adler, T.C., Bradburn, D.M., Huq, M., 1995. Sludge Application and Monitoring Program on the Oak Ridge Reservation, 1986–1993. U.S. Department of Energy. ORNL/TM-11601. Hatcher, R.D., 1989. Field Guide and Perspective on the Geology and Hydrology of the Oak Ridge Reservation. University of Tennessee, Oak Ridge, Tennessee. Hatcher Jr., R.D., Lemiszki, P.J., Dreier, R.B., Ketelle, R.H., Lee, R.R., Leitzke, D.A., McMaster, W.M., Foreman, J.L., Lee, S.Y., 1992. Status Report on the Geology of the Oak Ridge Reservation. U.S. Department of Energy. ORNL/TM-12074. Janssen, R., Peijnenburg, W., Posthuma, L., Van den Hoop, M., 1997. Equilibrium partitioning of heavy metals in Dutch field soils. I. Relationship between metal partition coefficients and soil characteristics. Environmental Toxicology and Chemistry 16, 2470–2478. King, L.D., 1988. Retention of metals by several soils of the southeastern United States. Journal of Environmental Quality 17, 239–246. Larsen, I.L., Cutshall, N.H., 1981. Direct determination of 7Be in sediments. Earth and Planetary Science Letters 54, 379–384. O’Connor, G.A., Elliot, H.A., Basta, N.T., Bastian, R.K., Pierzynski, G.M., Sims, R.C., Smith Jr., J.E., 2005. Sustainable land application: an overview. Journal of Environmental Quality 34, 7–17. Shinonaga, T., Schimmack, W., Gerzabek, M.H., 2005. Vertical migration of 60Co, 137 Cs and 226Ra in agricultural soils as observed in lysimeters under crop rotation. Journal of Environmental Radioactivity 79, 93–106.
1616
M.A. Smith et al. / Journal of Environmental Radioactivity 99 (2008) 1611–1616
Tanaka, T., Ohnuki, T., 1996. Colloidal migration behavior of radionuclides sorbed on mobile fine soil particles through a sand layer. Journal of Nuclear Science & Technology 33, 62–68. TDEC (Tennessee Department of Environment and Conservation), 1996. TDEC Routine Analysis of ORWWTP Sludge 1986–1988. Division of Radiological Health. Photocopied. Whicker, F.W., Pinder, J.E., 2002. Food chains and biogeochemical pathways: contributions of fallout and other radiotracers. Health Physics 82, 680–689.
Wolbarst, A.B., Chiu, W.A., Yu, C., Aiello, K., Bachmaier, J.T., Bastian, R.K., Cheng, J.-J., Goodman, J., Hogan, R., Jones, A.R., Kamboj, S., Lenhartt, T., Ott, W.R., Rubin, A., Salomon, S.N., Schmidt, D.W., Setlow, L.W., 2006. Radioactive materials in biosolids: dose modeling. Health Physics 90, 16–30. Yasuda, H., Uchida, S., Muramatsu, Y., Yoshida, S., 1995. Sorption of manganese, cobalt, zinc, strontium, and cesium onto agricultural soils: statistical analysis on effects of soil properties. Water, Air, and Soil Pollution 83, 85–96.