Introduction

Natural and anthropogenic radionuclides in soils have been reported to move to other environmental compartments through the food chain, leaching, soil resuspension and erosion processes [1,2,3,4,5,6,7]. The determination of radionuclides in the soil is required to evaluate the negative impacts of radioactive contamination on the environment and human health. Therefore, environmental monitoring of facilities responsible for the use of radioactive materials and management of radioactive wastes is relevant to ensure safe operations and evaluate radiological impacts on the environment and human health.

The Waste Isolation Pilot Plant (WIPP site) is located in southeastern New Mexico, USA and is an underground geologic repository for transuranic (TRU) wastes generated by the United States Department of Energy sites. In 2014, an accidental underground radiation release was caused by a runaway chemical reaction inside a TRU drum, a small portion of the contaminated air bypassed the ventilation filters and was discharged directly to the environment from an exhaust duct. The radiation release was reported to be below any level of public health or environmental concern [8]. Plutonium-239 + 240 (239+240Pu) and americium-241 (241Am) constitute more than 99% of the total radioactivity for disposal in the repository. Therefore, 239+240Pu and 241Am are of interest for the environmental monitoring program of the WIPP site including uranium (238U), potassium-40 (40K), cesium-137 (137Cs) and cobalt-60 (60Co). Like 238U and 40K, radium-226 (226Ra) and thorium-232 (232Th) are known as naturally occurring radionuclides in the soil that contribute to terrestrial background radiation which is one of the main sources for radiation exposure of the World’s population [9,10,11].

The land around the WIPP site is used for livestock grazing, potash mining, and oil/gas production. Workers conduct routine activities that could lead to exposure to radioactivity of the soil due to direct contact with the soil and suspension of fine soil particles during the operation of heavy equipment and strong wind events. Exposure to natural soil radiation takes place through external exposure to ionizing radiation, inhalation and ingestion of soil particles [12]. Also, workers could be exposed to anthropogenic radionuclides in the soil as result of global fallout caused by nuclear test sites and nuclear power plants accidents. Workers are potential human receptors for external exposure to ionizing radiation since they conduct regular outdoor activities across two soil monitoring areas (Near Field and Cactus Flats) located approximately 1 km northwest (downwind) and 19 km southeast (upwind) of the WIPP site. Previous studies reported external exposure to ionizing radiation as the exposure pathway with the highest cancer risk values [13, 14]. Therefore, the determination of potential damage to human health resulting from external exposure to ionizing radiation is necessary for the protection of workers.

The objectives of the study were to evaluate the relationship between radioactivity concentrations and soil properties, and determine the total annual effective dose (AED) for natural and artificial gamma emitting radionuclides detected in two soil monitoring areas located in the vicinity of the WIPP site. Information on the relationship between radionuclide concentrations and soil properties is relevant to understand the behavior of radionuclides in the environment and identify potential exposure pathways. Also the information generated in this study regarding the total AED as result of external exposure to ionizing radiation will be useful to evaluate the potential damage to human health in the vicinity of the WIPP site.

Materials and methods

Sample collection and analysis:

The study site has a non-calcareous, reddish sandy soil originated from eolian sand deposits over the Dewey lake red bed formation that mostly consist of reddish-brown siltstone and fine-grained sandstone The Dewey lake red bed is the uppermost formation in the Ochoan series followed by the Rustler, Salado (location of geologic repository for TRU wastes at approximately 655 m below the ground surface) and Castile formations [15]. A total of 64 (32 samples x 2 years) soil samples were collected from two locations (Site 107-Near Field and Site 108-Cactus Flats) where high-volume air samplers are located around the WIPP site in the Chihuahuan desert of southeastern New Mexico between 2015 and 2018 (Fig. 1). At both locations, a grid was established with sixteen undisturbed soil sampling locations.

Fig. 1
figure 1

Soil and aerosol sampling locations in the Waste Isolation Pilot Plant vicinity

Full size image

Approximately 4 L of soil were collected using a trowel and a 50 × 50 cm frame with a depth of 0–2 cm and stored in plastic bags. All samples were air-dried for 24 h and passed through 2 mm sieve at the Carlsbad Environmental Monitoring and Research Center (CEMRC). A 50 g aliquot was used to determine the soil particle-size distribution using the pipette method described by Gee and Bauder [16]. The rest of the collected soil was oven dried at 105°C for 12 h then aliquots were used for soil organic matter (SOM), gamma and actinide analyses. Aliquots of 50 g of soil for each sample were used to determine the SOM content using the loss on ignition method [17]. Triplicates of 10 g of soil were placed in a muffle furnace at 500°C for 6 h.

Determination of gamma radionuclides in soil:

Aluminum paint cans of 300 g capacity were filled with soil, sealed and allowed radium nuclides (226Ra and 228Ra) to reach equilibrium with daughter radionuclides for 25 days before conducting the gamma ray spectrometry analysis. Samples were counted for 48 h using high purity germanium detectors and spectral data was analyzed using Genie 2000 software (Mirion Technologies, Inc.). 232Th activity was determined by using the energies of the progenies 228Ac (338.3, 911.6 and 969.1 keV) and 212Pb (238.6 keV). 226Ra activity was measured by using the energies 295.2 and 351.9 keV of 214Pb and 609.3, 1120.3, 1764.5 keV of 214Bi. 137Cs and 40K activities were measured using the gamma energy lines 661.6 and 1460.8 keV, respectively.

Preparation of soil samples for determination of actinides:

Approximately 4-5 g of grounded soil was placed in the muffle furnace at 500°C for 6 h to eliminate the organic matter before conducting the radiochemical analysis for actinides [18]. Then, each sample was spiked with radioactive tracers (243Am, 242Pu and 232U), sea sand was spiked with 241Am, 239+240Pu and 238U and used as laboratory control. Samples were digested in teflon beakers with 10 mL hydrochloric (HCL), 10 mL nitric (HNO3), and 20 mL hydrofluoric (HF) acids on a hotplate at 200°C. The sample residues were heated with perchloric acid and boric acids to remove HF. Finally, the residues were dissolved in 8 M HNO3 for radiochemical separation.

Separation of plutonium from americium and uranium:

The oxidation state of plutonium was adjusted to plutonium (IV) by adding 1 mL of 1 M ammonium iodide (NH4I) with a 10 min wait step, followed by 1 mL of 2 M sodium nitrite (NaNO2). Plutonium was separated from americium and uranium using an anion exchange resin (AG1-X8, 100–200 mesh, Cl− form; Eichrom Technologies, Inc.) conditioned with 25 mL of 8 M HNO3 [19]. Americium and uranium passed through the anion exchange resin column and the effluent was used for americium and uranium analysis. Thorium was removed from the column with 40 mL of 10 M HCL. Finally, plutonium was eluted with 30 mL of 0.1 M NH4I + 10 M HCL solution that reduced the plutonium (IV) previously retained by the anion exchange resin column to plutonium (III).

The eluted plutonium fraction was evaporated with 3 mL of HNO3 and 0.5 mL of 5 mg mL− 1 iron carrier to complete dryness. The plutonium residue in the samples was dissolved in 4 mL of 2 M HCL, then 0.5 mL of 2 M NaNO2 and 10 mL of HCL were added to complete the oxidation state adjustment of plutonium. The samples were passed through anion exchange resin columns previously conditioned with 25 mL of 8 M HCL to purify the plutonium fraction. The columns were washed with 30 mL of 8 M HCL before eluting plutonium with 30 mL of 0.1 M NH4I + 8 M HCL. The collected plutonium samples were evaporated to complete dryness after adding 3 mL of HNO3 and 0.5 mL of 5 mg mL− 1 iron carrier.

Separation of americium and uranium:

The effluent solution containing the americium and uranium fractions was evaporated to dryness and dissolved in 10 mL of 2 M HNO3. Uranium was separated from americium on TRU chromatography resin (tri n-butyl phosphate, N-diisobutyl carbamoxyl methyl phosphine oxide; Eichrom Technologies, Inc.) previously conditioned with 10 mL of 2 M HNO3[19]. The presence of iron (III) in the samples interferes in the separation of americium (III) on the TRU column, therefore its presence was tested by adding 1 drop of ammonium thiocyanate (NH4SCN) solution. Samples turned color red if the test was positive and approximately 10 drops of 1 M ascorbic acid solution were added until the red color disappeared indicating that iron (III) was reduced to iron (II). The samples were loaded into the TRU columns, washed with 10 mL of 2 M HNO3 and washed a second time with 10 mL of 2 M HNO3 + 0.05 M NaNO2 to oxidize any plutonium (III) to plutonium (IV). Then americium (III) was eluted with 20 mL of 4 M HCL and uranium was eluted with 20 mL of 0.1 M ammonium bioxalate. After separation, the uranium fraction was evaporated to dryness with 5 mL of HNO3 and 1 mL of 10% sodium sulfate (Na2SO4). Then the uranium samples were dissolved in 10 mL of 8 M HCL and purified by passing them through anion exchange resin columns previously conditioned with 25 mL of 8 M HCL. The columns were washed with 30 mL of 8 M HCL before eluting uranium with 30 mL of 1 M HCL. The uranium fraction was evaporated to dryness after adding 1 mL of 10% Na2SO4 and 5 mL of HNO3.

The americium fraction collected from the TRU column was evaporated to dryness with 1 mL of 50% sulfuric acid and 3 mL of perchloric acid to destroy any TRU resin material present in the solution. The dried americium fraction was dissolved in 10 mL of 3 M NH4SCN + 0.1 M formic acid (HCOOH). The solution was loaded onto a TEVA resin (Aliquot 336; Eichrom Technologies, Inc.) previously conditioned with 10 mL of 3 M NH4SCN. The TEVA resin column was washed with 12 mL of 1.5 M NH4SCN + 0.1 M HCOOH to remove lanthanides that could be present. Americium was eluted with 15 mL of 2 M HCL and the remaining NH4SCN was destroyed by heating with 8–10 mL of HNO3: HCL (1:3). Finally, the americium samples were evaporated to dryness. The dried americium, plutonium and uranium samples from the purification process were dissolved with 4 mL of 2 M HCL and transferred to 50 mL polycarbonate centrifuge tubes with DI water. The samples were then micro-coprecipitated using 0.1 mL of 0.5 mg mL− 1 neodymium-carrier and 1 mL of HF on stainless steel planchettes that were counted for five days using an APEX-Alpha spectrometer (Mirion Technologies, Inc.) [20].

Statistical analysis

Statistical significant difference of radionuclide concentrations, soil texture, pH and SOM in samples collected from both sites was determined using the t-test (SPSS software, IBM Corp, Armonk, NY, USA). Correlation analysis was used to evaluate the relationship between radionuclide concentrations and soil properties. Data collected in 2015 and 2018 were combined to conduct the correlation analysis.

Radiological characterization of 226Ra, 232Th40K and 137Cs:

The radiological characterization of 226Ra, 232Th 40K and 137Cs was conducted using the soil activity concentrations detected in 2018. Soil activity concentrations of 226Ra, 232Th, 40K and 137Cs were used to calculate the total absorbed gamma dose rate (D; nGy h− 1) in air at 1 m above the ground level using the conversion factors in Eq. (1):

$$\text{D}=0.462{\text{A}}_{\text{R}\text{a}}+0.604{\text{A}}_{\text{T}\text{h}}+0.0417{\text{A}}_{\text{K}}+0.1125{\text{A}}_{\text{C}\text{s}}$$
(1)

where ARa, ATh, AK and ACs are the activity concentrations (Bq kg− 1) of 226Ra, 232Th, 40K and 137Cs, respectively [9, 21,22,23].

The D values were converted to total annual effective doses using the method presented by UNSCEAR [9]. Total annual effective dose (AED; µSv y− 1) was calculated using Eq. (2) as:

$$\text{A}\text{E}\text{D}=\text{D} \text{x} 8760 \text{x} 0.2 \text{x} 0.7 \text{x} {10}^{-3}$$
(2)

where D (nGy h− 1) is total absorbed gamma dose rate in air at 1 m above the ground level, 8760 (h yr− 1) is the time conversion factor, 0.2 is the outdoor occupancy factor for a person spending 20% of time outdoors, 0.7 (Sv Gy− 1) is the conversion coefficient factor to convert outdoor gamma absorbed dose rate in air to effective dose received by adults and 10− 3 is a conversion factor [9, 24].

Results and discussion

Relationship between radioactivity concentrations and soil properties

241Am, 239+240Pu, 238U, 137Cs, 40K, 232Th and 226Ra were detected in soil samples collected from both locations, except for 60Co. The radioactivity concentrations of 241Am, 239+240Pu, and 137Cs were representative of background levels due to global and regional (Nevada Test Site) fallout, and possibly soil resuspension from the Gnome site (underground nuclear test) located 8.8 Km southwest of the WIPP site [25, 26]. Radiological analysis of air filter samples collected from 1998 to 2010 showed that the exhaust air from the WIPP underground repository has not been a source of radiological contamination in the study site [27]. After the 2014 radiation release event at the WIPP site, 241Am and 239+240Pu were reported to be the dominant radionuclides detected in air monitoring stations located within and outside the WIPP [8]. Detected 241Am and 239+240Pu concentrations in air filters were low and no adverse health effects on WIPP workers were observed after the 2014 radiation release event [28]. Thakur [29] reported that 239+240Pu and 241Am concentrations in air filter samples reached pre-release levels two weeks later after the 2014 radiation release event. This confirms that 239+240Pu and 241Am concentrations detected in ambient air filter samples were originated due to the resuspension of soil particles contaminated by nuclear weapon fallout.

Most of the soil radionuclide concentrations were significantly higher in Cactus Flats than in Near Field (p < 0.05), except for 40K and 226Ra concentrations that were similar at both locations while 241Am concentrations were not significantly different in 2018 (Figs. 2 and 3). Differences of concentrations in both locations could be explained by differences in soil properties. The soils in Near Filed and Cactus Flats were well-drained, non-calcareous with a carbonate content < 1% and developed from eolian sand parent materials [30, 31]. Both locations had a sand texture classification, however silt, clay and SOM contents were significantly greater in Cactus Flats than in Near Field, while the sand content was greater in Near Field (p < 0.05) (Table 1).

Fig. 2
figure 2

Average plutonium-239 + 240 (a), americium-241 (b) and uranium-238 (c) activity concentrations in soil samples collected from Near Field (white columns) and Cactus Flats (grey columns) in 2015 and 2018. Error bars represent standard error

Full size image
Fig. 3
figure 3

Average potassium-40 (a), radium-226 (b), throrium-232 (c) and cesium-137 (d) concentrations in soil samples collected from Near Field (white columns) and Cactus Flats (grey columns) in 2015 and 2018. Error bars represent standard error

Full size image
Table 1 Average soil properties of samples collected from Near Field and Cactus Flats
Full size table

Sand showed a negative correlation with silt (r = -0.80, p < 0.01), clay (r = -0.88, p < 0.01), SOM (r = -0.25, p = 0.17) and pH (r = -0.40, p < 0.05) in Near Field. While similar relationships were obtained for sand with silt (r = -0.75, p < 0.01), clay (r = -0.71, p < 0.01), SOM (r = -0.58, < 0.01) and pH (r = -0.14, p = 0.47) in Cactus Flats. Clay content was positively correlated with SOM in Cactus Flats (r = 0.46, p < 0.05), however the correlation was not significant in Near Field (r = 0.24, p = 0.19). Silt and clay contents were positively correlated in Near Field (0.67, p < 0.01) and Cactus Flats (0.46, p < 0.05). Fine soil particles (clay and silt) are able to interact with radionuclides through the adsorption process for having greater surface area compared to sand particles. Clay is the finest of the soil particles, therefore it has more surface area available for sorption sites for solutes [32, 33]. Studies have reported that soils with high clay content have better capacity to preserve organic matter than those having low clay content [34, 35].

Kirchner [25] reported strong correlations between radionuclide concentrations and soil texture, but the relationship between radionuclide concentrations and SOM was not examined for soil samples collected from Near Field and Cactus Flats in 1998. Similarly, Thakur [26] reported that 239+240Pu concentration was positively correlated with the aerosol mass retained in ambient air filters from air monitoring stations located at the WIPP site, Near Field and Cactus Flats during 1998–2010. Also 239+240Pu and 137Cs concentrations showed a decreasing trend with increasing soil depth in the WIPP vicinity. 239+240Pu, 241Am and 137Cs concentrations in soil samples collected from Near Field after the 2014 radiation release event were not adverse to human health, however, radionuclide concentrations in soil samples from the Cactus Flats location and the influence of soil properties on radionuclide concentrations were not evaluated [28].

239+240Pu, 238U, 40K, showed a positive correlation with SOM at both locations (Table 2). 241Am concentration was positively correlated with SOM, while 40K concentration also showed a positive correlation with silt content, however there was no significant correlation between soil properties and the rest of radionuclides (137Cs, 232Th and 226Ra) in Near Field. In contrast, 238U, 40K and 137Cs, 232Th and 226Ra were negatively correlated with sand content and positively correlated with SOM in Cactus Flats. Also most of the radionuclides were significantly correlated with silt and clay contents in Cactus Flats. Previous studies reported that concentrations of natural gamma-emitting radionuclide (232Th, 40K, and 226Ra) and 137Cs were positively correlated with clay and SOM [33, 36, 37]. Also 241Am, 239+240Pu, 238U have shown greater affinity for finer soil particles and SOM during laboratory and field studies that used soils from different climate regions [25, 38,39,40,41,42,43,44,45].

Table 2 Correlation between radioactivity concentrations and soil properties
Full size table

Generally, there was no significant relationship between pH and radionuclide concentrations. The lack of correlation between pH and radionuclide concentrations could be due to the narrow range of soil pH values [46]. Previous studies have reported weak or no correlations between pH and radionuclide concentrations in different soil types [26, 47,48,49]. In contrast, other studies reported a negative correlation between pH and radionuclide concentrations [37, 50, 51].

Radiological characterization of detected gamma emitting radionuclides

Average 226Ra, 232Th, 40K and 137Cs concentrations in Near Field were 16.3 ± 2.7, 8.6 ± 1.2, 208 ± 29 and 1.8 ± 0.9, respectively, while 226Ra, 232Th, 40K and 137Cs concentrations in Cactus Flats were 19.2 ± 4, 11.3 ± 2, 216 ± 35 and 2.5 ± 0.9, respectively (Table 3). Concentrations at both sites were lower than the worldwide average values for 226Ra (35 Bq kg− 1), 232Th (30 Bq kg− 1), 40K (400 Bq kg− 1) and 137Cs (51 Bq kg− 1) [9, 11, 23, 52]. 40K was the major contributor to the calculated D and AED followed by 226Ra > 232Th > 137Cs. Similarly, Hannan [22] reported that even with the additional contribution of 137Cs the D and AED values were lower than the global average D (59 nGy h− 1) and AED (70 µSv y− 1) values calculated based on the contribution from natural gamma-emitting radionuclides. The average D values of 21.6 ± 3.1 nGy h− 1 for Near Field and 24.9 ± 4.2 nGy h− 1 for Cactus Flats were significantly different (p < 0.05). All of the AED values estimated for each of the grid nodes were less than 1mSv y− 1, with average AED values of 26.5 ± 3.8 and 30.6 ± 5.2 µSv y− 1 for Near Field and Cactus Flats, respectively (Table 4). The estimated AED values were relatively similar compared to a study conducted in Qatar (27 µSv y− 1), but lower than AED values from Texas-USA (59 µSv y− 1), Taiwan (60.5 µSv y− 1), Nigeria (70 µSv y− 1); and Malaysia (169 µSv y− 1) [10, 11, 22, 33, 50].

Table 3 Radioactivity concentrations of gamma-emitting radionuclides and their radiological risk characterization parameters for soil samples from Near Field (Grid A-B) and Cactus Flats (Grid C-D)
Full size table
Table 4 Comparison of radioactivity concentrations of gamma-emitting radionuclides and their radiological risk characterization parameters for soil samples collected from different countries of the world
Full size table

Conclusions

The Cactus Flats location had higher soil radioactivity concentrations, SOM and percentage of fine soil particles compared to the Near Field location. Therefore, stronger relationships between soil radioactivity concentrations and soil properties were obtained in Cactus Flats. Soil radioactivity concentrations showed a positive correlation with silt, clay, and SOM, however they were negatively correlated with sand content and either not or weakly correlated with soil pH. The AED values were lower than the recommended limit, meaning that soil radioactivity concentrations from detected natural and artificial gamma-emitting radionuclides in the soil around the WIPP site are safe to human health.