Determination of radionuclide concentration and radiological hazard in soil and water near the uranium tailings reservoir in China

ABSTRACT Radioactive levels and radiological hazard assessment associated with exposure to radionuclides from uranium tailings were investigated in this study. Herein, the concentrations of 238U, 226Ra, 232Th, and 40K in 36 samples including 25 soil and 11 water samples were measured using a low background HPGe semiconductor detector. The radionuclide concentrations were much lower outside the tailings reservoir than inside the reservoir; however, both exceeded the national average and world average values. The mean concentrations of 238U, 226Ra, 232Th, and 40K in water samples were higher than the WHO guideline levels. The radionuclide transfer factors showed that the radionuclides in soil of the tailings enter the surrounding water. Additionally, the radiological hazard indices around the reservoir were much higher than the worldwide average. Results suggest that the high level of radiation in the tailings reservoir posed a potential threat to the health of the local residents.


INTRODUCTION
Natural uranium occurs in uranium deposits, which are mined and processed into nuclear materials for military and civilian uses. The modern uranium mining industry began in the 1940s. As the demand for nuclear fuel has increased, uranium mining and milling activities have been extensively conducted in the past decade [1,2]. It is well known that mining of uranium ore produces large amounts of tailings that contain high concentrations of uranium-series radionuclides. The effective dose rate produced by radionuclides that were released from worldwide mining, hydrometallurgy, and tailings from 1998 to 2002 was approximately 0.238 Gy y −1 (United Nations Scientific Committee on the Effects of Atomic Radiation [3]. The radioactivity levels and environmental problems surrounding uranium mines have attracted worldwide attention. Extensive research has been conducted on the characteristics and mobilization of radionuclides in soil, plants, and water, as well as on the transfers between these components at different uranium mining sites worldwide [4][5][6]. For example, the highest uranium concentration in water occurred at the beginning of spring in the former Mondego Sul uranium mine (Portugal) because the highest water flow caused leaching of uranium from secondary minerals to be released into the water [7]. The median uranium concentrations in drinking water sources showed a decreasing trend: streams (135.30 µg L −1 ) > dams (115.62 µg L −1 ) > boreholes (111.31 µg L −1 ) > shallow wells (110.03 µg L −1 ) in the vicinity of a uranium mine in the Siavonga district [8]. In northwestern Greece, minor pollution of the lakes in a lignite mining area was slightly more prevalent during warm periods; this was because of leaching effects following the area's snow melt and rainfall, as well as increased evaporation and industrialization activity, especially during the summer [9]. The accumulation of radioactivity was limited from the irrigation in the paddy-soil around a decommissioned uranium mine in eastern China [10]. There were also weak positive correlations between organic matter and natural radionuclides in soil from Istanbul in the northwest part of Turkey [11].
According to the report of the United Nations Scientific Committee on the Effects of Atomic Radiation, exposures to natural radiation sources account for more than 98% of all radiation doses to humans (excluding medical exposures) [12]. The average annual exposure to natural radiation sources in the world is estimated to have ranged from 1-10 mSv (United Nation Scientific Committee on the Effects of Atomic Radiation [12]. Because of the high levels of background radiation in the vicinity of uranium mining areas, local residents may be affected by this radiation [13,14]. Ionizing radiation produced by decaying nuclides causes biological damage to human organs (United States Food and Drug Administration [15]. Empirical observations and epidemiological research have consistently suggested that ionizing radiation has carcinogenic properties(United States Environmental Protection Agency [13,15] after certain level. Radiological hazard parameters can assist in calculating the possible risk to local populations. Statistical analysis in crude oil and petroleum products of Yemen revealed that accumulation of radioactive materials may lead to future hazards [16]. 40 K in diets from the study in a major mining region of Ghana was the highest contributor (48%) to committed effective dose, followed by 228 Ra (35%), 226 Ra (16%) and 228 Th (1%) respectively [17]. Phosphate rock and phosphogypsum declare higher radiological parameters than the world limit while for phosphoric acid are within the world permissible limits in Phosphate Fertilizers and Chemicals Company, located in Egypt [18]. Therefore, it is essential to assess radioactivity hazards to the population and understand the environmental behavior of natural radionuclides in the different media that surround a mining area. The data can be used as a starting point for research on radionuclides pollution and mobility around the tailings reservoir.
Uranium mining in China began in the 1960s. Tailings reservoirs containing waste rocks and slag are potential sources of long-term radioactive contamination. Owing to physical and chemical processes such as rainfall eluviation and dissolution, the nuclides in the ore sand can be transported to the soil and water in the surrounding environment. Few studies have investigated the radiation levels and their associated doses in the soil and water for a tailings reservoir in China. Therefore, this study investigated the radiation levels in soil and water near a tailings reservoir by measuring the concentrations of the long-lived radionuclides 238 U, 226 Ra, 232 Th, and 40 K in soil and water, assessing the radiological hazard indices, and determining the effective dose to the public from the soil samples.

Study area
The uranium ore mine sampled in this study is located in the southern part of China, which contains the largest volcanogenic orebodies in the country. A valley-dammed uranium tailings reservoir that stores a large amount of tailing sand is located in the eastern part of the uranium mining area. Waste rocks and effluent generated from the uranium mill tailings have led to a variety of radioactive pollution problems in the area. The uranium tailings reservoir is surrounded by mountains and is located in a subtropical humid and rainy climate. The weather in the region is continental, with temperatures varying between −6°C (winter) and 39°C (summer).

Sample collection and processing
A total of 25 soil samples and 11 water samples were collected around the uranium tailings reservoir. The 25 soil samples included 16 samples from the uranium tailings reservoir (S-1 to S-16) and 9 samples from outside the tailings reservoir (S-17 to S-25). The 11 water samples included 3 samples of water discharging from the bottom of the tailings dam (W-4, W-5, and W-6) and 8 samples from nearby surface water bodies (W-1, W-2, W-3 and W-7, W-8, W-9, W-10, W-11). The study area and sampling locations are depicted in Figure 1.
Each individual ~1 kg soil sample was carefully excavated at a depth of 5 cm and mixed homogeneously, then was immediately placed into labeled plastic bags. The samples were taken to the laboratory for subsequent treatment and analysis. Radionuclides in the soil samples were determined via gamma spectrometry, following which the samples were processed [19]. After removing grass, stones, and organic matter, the samples were ground to less than 0.15 mm. They were then dried in an oven at 100°C for 24 h to remove moisture content. Then, the samples were packed and sealed in cylindrical polyethylene beakers for at least 4 weeks to establish secular equilibrium between 226 Ra, and their decay products.
Radionuclides were also determined in the water samples via gamma spectrometry, following which the samples were preprocessed [20]. The water samples were collected at the surface directly into polyethylene after being filtered by 0.45 μm-pore-sized filters on-site. The filtered samples were then acidified with HNO 3 to a pH value below 2 for laboratory analyses. Parameters including Temperature, pH, and Eh of the water samples were measured in situ using a portable multipleparameter meter (HACHHQ40d, USA).

Radioactivity measurements
The activity concentrations of 238 U, 226 Ra, 232 Th, and 40 K in the soil samples were measured using a low background high purity germanium (HPGe) semiconductor detector (ORTEC GMX40P, USA) with a resolution of 2.1 keV for a 1332 keV gamma ray emission of 60 Co. Data acquisition and spectral analyses were performed using the MAESTRO-32 software by ORTEC. The detector was calibrated using gamma rays emitted during the decay of 137 Cs and 60 Co at energies of 662 keV, 1173 keV and 1332 keV. The activity concentration of 238 U was determined via its daughter 234 Th (92.6 keV), whereas 226 Ra was obtained from the determinations of 214 Pb (351.9 keV) and 214 Bi (609.3 keV). The activity concentration of 232 Th was determined from 208 Tl (583.4 keV) and 228 Ac (911.2 keV), and the concentration of 40 K was determined from an emission line of 1460.8 keV. Quality assurance was guaranteed by the regular participations in national and IAEA laboratory intercomparison exercises. The expanded uncertainties of 238 U, 226 Ra, 232 Th and 40 K concentrations in soil samples were less than 3.59%, 5.50%, 3.30%, and 3.50%, respectively. They all meet the requirements of GB/T 11,743-2013 [19] ( 238 U less than 20%; 226 Ra, 232 Th, and 40 K less than 10%). The counting uncertainties of 238 U, 226 Ra, 232 Th and 40 K concentrations in water samples were less than 5.0%, 4.9%, 4.5%, and 4.9%, respectively. They all meet the requirements of GB/T 11,713-2015 [21] (less than 10%).
Soil samples and standard sources were placed into cylindrical polyethylene containers (diameter 75 mm, height 70 mm). In addition, water samples and calibration sources were placed into 1 L polyethylene bottles. To establish a secular equilibrium between the parent nuclides of uranium and thorium and their short-lived products, the samples were measured after being sealed for 20 days. The background gamma levels and sample spectra were measured for 24 h.
According to the relative method, the activity concentration of radionuclides in the soil sample can be calculated using Eq. (1) in GB/T 11,743-2013 [19,22]: Where, A s is the activity concentration of the nuclide in the soil sample (Bq g −1 ); A j is the activity of the nuclide in the standard source of soil monitoring efficiency (Bq), derived from the National Institute of Metrology (No. TRH1608023). The radioactive activity of 238 U, 226 Ra, 232 Th, and 40 K is 5.01E+02 Bq, 1.56E+02 Bq, 1.16E+02 Bq and 1.063E+03 Bq, respectively. S j , S 0 and S b are the gross counting rates of the characteristic peaks of the nuclide in the standard source, the sample and the background, respectively (s −1 ), M is the volume of the soil sample (335.0 g). D j is the decay correction factor of the nuclide, and the value of D j is 1. The detection limits of the instrument were 0.0032 Bq g −1 , 0.0018 Bq g −1 , 0.0015 Bq g −1 , and 0.0092 Bq g −1 for 238 U, 226 Ra, 232 Th, and 40 K, respectively.
The activity concentrations of the nuclides in the water samples were calculated using Eq. (2) in GB/T 16,140-2018 [20]: Where, A s is the activity concentration of the nuclide in the sample (Bq L −1 ); A j is the activity of the nuclide in the calibration source (Bq), derived from the National Institute of Metrology (No. YMLH1608022). The radioactive activity of 238 U, 226 Ra, 232 Th, and 40 K are 1.886E+03 Bq, 6.22 E + 02 Bq, 2.74 E + 02 Bq and 6.89 E + 02 Bq, respectively. S j , S 0 and S b are the gross counting rates of the characteristic peaks of the nuclide in the calibration source, the sample and the background, respectively (s −1 ); V is the volume of the water sample (1 L). DF j is the decay correction factor of the nuclide, and the value of DF j is 1. The detection limits were 1.032 Bq L −1 , 0.463 Bq L −1 , 0.312 Bq L −1 , and 2.141 Bq L −1 for 238 U, 226 Ra, 232 Th, and 40 K, respectively.

Calculation of radiological hazard indices
Four radiological indices for the absorbed dose rate (D γ ) , the outdoor annual effective dose equivalent (AEDE out ) , the outdoor excess lifetime cancer risk (ELCR out ), and the annual gonad dose equivalent (AGDE) were calculated to quantify the radiation hazard introduced by radionuclides in the topsoil to the local population.

(1) Absorbed dose rate in air (D γ )
The absorbed dose rate in air caused by terrestrial gamma rays at 1 m above the ground was calculated from the 226 Ra, 232 Th, and 40 K activity concentrations in the soil. The absorbed dose rate is given as follows [23]: where A Ra , A Th , and A K are the activity concentrations of 226 Ra, 232 Th, and 40 K, respectively.

(2) Annual effective dose equivalent for outdoor (AEDE out )
The outdoor terrestrial gamma radiation value is needed to calculate the outdoor annual effective dose equivalent (AEDE out ). D γ can be converted to AEDE out using a conversion factor and an outdoor occupancy factor as follows [23]: where F is the dose conversion factor (0.7 Sv/Gy) [24], and T is the value that converts the time from years to hours (8760 h/y). The fraction of time spent outdoors is 0.2 [23].

(3) Outdoor excess lifetime cancer risk (ELCR out )
The excess lifetime cancer risk is the chances that an individual would develop cancer in their lifetime when considering a specified exposure level. This risk can be estimated by the following [25]: where AEDE out is the outdoor annual effective dose equivalent, DL is the average life duration (74.33 years for the residents of the study area) (National Health and Family Planning Commission [26], and RF is the risk factor (0.05 Sv −1 ) [27].

(4) Annual gonad dose equivalent (AGDE)
The gonads, active bone marrow, and bone surface cells are considered the organs of interest [27]. Therefore, the annual gonad dose equivalent (AGDE) of local people can be calculated as follows [28]: where A Ra , A Th and A K are the activity concentrations of 226 Ra, 232 Th, and 40 K, respectively. Th, and 2.89-8.60 for 40 K. There are two reasons for the high levels of 226 Ra observed in the study area. One is that most of the ore remains in the tailings, with only a small amount entering the solution during the metallurgical process [29]. The other reason is that 226 Ra is a decay product of 238 U. High levels of 226 Ra were produced by 238 U in the soil, as reported by E. Charro [30]. The average concentrations of 238 U, 226 Ra, 232 Th, and 40 K were 20.69 Bq g −1 , 12.85 Bq g −1 , 1.21 Bq g −1 , and 5.14 Bq•g −1 , respectively, which are 545, 347, 22, and 9 times the national average value [31]. In addition, radionuclides and their daughters emit α and β rays during the decay process, which can lead to high levels of radioactivity in the reservoir area. The ranges in the coefficients of variation (CV) of the nuclides in the tailings reservoir were approximately 10-25%, with small differences, thereby demonstrating that variations in the activity concentrations of the radionuclides were weak in the reservoir.

Radionuclide activity concentrations in soil
The average activity concentrations of 238 U, 226 Ra, 232 Th, and 40 K in the soil samples outside the tailings reservoir were 7.97 Bq g −1 , 7.00 Bq g −1 , 0.67 Bq g −1 , and 2.65 Bq g −1 , respectively, which are 210, 189, 12, and 5 times more than the national average values, respectively. The radionuclide contents were obviously much higher than those of the national average value, as well as the world average value obtained by UNSCEAR for reference soils [23]; however, they were much lower than the concentrations inside the tailings reservoir. The average activity concentrations of the four radionuclides differed significantly, whereas the coefficients of variation (CV) were moderate. In addition, the CVs of the nuclides were higher than those inside the reservoir, which indicated that the nuclides outside the reservoir were unevenly distributed. This could be caused by soil properties such as soil size and organic matter type, as well as human activities. The mean concentrations of the four radionuclides inside and outside the tailings reservoir are shown in Figure 2. The radionuclide concentrations in the soil decreased from inside the reservoir to the outside of the reservoir area, as has also been reported in other studies [32]. The 238 U concentrations in the tailings reservoir exceeded the 226 Ra concentrations inside the reservoir, but the 238 U concentrations were similar to the 226 Ra concentrations outside the reservoir. The concentrations of 232 Th and 40 K were reduced by approximately half from inside the reservoir to its outside. Most of the 238 U, 226 Ra, 232 Th, and 40 K concentrations obtained in this study were several times higher than those obtained from uranium sites in other countries, as listed in Table 2. Specifically, concentrations of 238 U in the study area were much higher compared to other countries. This can be attributed to the large volume of uranium tailings contained within the study area. This indicated that the tailings reservoir was a potential radioactive source, which caused the accumulation of nuclides in the soil.

Radionuclide activity concentrations in water
The statistics of the 238 U, 226 Ra, 232 Th, and 40 K concentrations in the water samples are detailed in Table 3. The radionuclides 238 U and 226 Ra in the uranium decay series were the main radioactive contaminants in the water, which indicated evident contamination from the uranium mine. The activity concentrations of 238 U, 226 Ra, 232 Th, and 40 K near the uranium tailings reservoir ranged from 1.82-16.99, 0.27-13.52, 0.06-1.31, and 0.36-3.48 Bq•L −1 , respectively, with average values of 4.73, 3.92, 0.44, and 2.17 Bq L −1 , respectively. The mean activity concentrations of 226 Ra was higher than the WHO guideline levels (World Health Organization [33]. The highest contamination was observed in the water discharged near the tailings reservoir. All 238 U concentrations in the water samples, except W-4, were below the guidance level of 10 Bq L −1 . The 238 U contents of the water samples in the reservoir were slightly higher than the concentrations of the water samples outside the reservoir with the exception of W-4. The uranium occurred as uranyl ions (UO 2 2+ ) in large amounts in water from the open pit. These uranyl ions tend to occur in oxidized surface waters and form stable and readily soluble complexes [34]. Similar results have been obtained [35,36] in previous studies in similar regions. The 232 Th concentrations at two sampling locations in this study exceeded the guideline levels by approximately 1.3 and 1.2 times. The radionuclide concentrations in the soluble phase varied widely, which may have been dependent on the sampling station and particlewater partitioning of the radionuclides [37]. In addition, high 226 Ra contents were found in samples W-1,  W-2, and W-3 inside the reservoir, which exceeded guidelines level by 12, 12, and 7 times, respectively. This indicates that high levels of radium are still present in the tailings. According to the CV, the activity concentrations of 238 U, 226 Ra, and 232 Th exhibited considerably more inconsistencies, whereas the 40 K activity concentration had a weak CV. This indicates that the 238 U, 226 Ra, and 232 Th originated from the reservoir, whereas 40 K was not heavily affected by the reservoir area. Table 4 presents the main chemical parameters of the water samples. The pH values in the study area range from 4.59-8.74, which suggests that the water is acidic to alkaline. The pH values of the water near the tailings reservoir varied widely. The electrical conductivity (EC) had a minimum value of 31.9 μS/cm and a maximum value of 2642 μS/cm, thereby suggesting the EC values of the water samples were uneven. The samples with the highest pH values were W-1, W-2, and W-3, all of which were collected from the open pond of the tailings reservoir. The pond also had high EC and 226 Ra, 232 Th, and 40 K activity concentrations. The high radionuclide concentrations may be attributed to radionuclide migration from the tailings sand to the water via precipitation leaching. The high EC of the water samples from the tailings reservoir indicates the presence of more soluble ions. The lowest pH value was W-7, which was recorded in the surface water at the upper reaches of the channel. The nuclide content at W-7 were also low, which was likely caused by continuous runoff. Compared with the discharged water samples, the nuclide concentrations and EC values of the surface water near the tailings reservoir were lower. This may be related to the continuous inflow and adsorption of nuclides by sediments at the bottom of the riverbed. In addition, hydraulic properties of the water body could have also lowered the concentrations [38]. Therefore, hydrological processes are responsible for off-site contaminant transport in the study area [7]. To address the relationship between the nuclide concentrations and pH, pH as a function of the nuclide concentrations in both discharged water and surface water samples was determined ( Figure 3). In general, the nuclide contents increased with pH, except for 232 Th, in the water samples. The 238 U contents of the discharged water were higher than those of the surface water. By filtration and dissolution, the radioactive uranium from the tailings entered the surrounding water. The water discharged from the tailings has a high potential to disperse nuclides, especially 238 U and 226 Ra, and consequently is considerably likely to contaminate the surrounding water bodies [7]. The continuous flow of water carries away the components, producing low nuclide contents in the surface water. The concentrations of 226 Ra and 40 K were significantly positively correlated with pH, which indicates that pH is a key factor for nuclide migration. The main mechanism includes ion exchange at low pH value and surface complexation and surface precipitation at high pH value [39]. Therefore, the chemical properties of the water largely influence the nuclide migration process.

Comparison of nuclides in the soil and water
The average radionuclide activity concentrations of the soil and water samples are shown in Figure 2. The average contents of 238 U, 226 Ra, 232 Th, and 40 K of the soil were higher than those of the water. Enrichments of 238 U and 226 Ra were observed in the soil-water samples and were related to the high radioactive background level in the tailings. Additionally, various factors can affect the radionuclide concentrations in water, including the physiochemical parameters of water, the leaching process of rocks, and the alpha recoil effect of the daughter product during the decay process [40].   Table 5 lists the TFs of the radionuclide concentrations in the soil to those in water at the points closest to the dam and furthest from the dam. Consistent with Figure 2, the concentrations of radionuclides in the soil were higher than those in water samples of W-4 and W-9. The TFs of the radionuclides 238 U, 226 Ra, 232 Th, and 40 K in W-4 ranged from 1.06E-04 to 8.21E-04. At W-9, the TFs were 8.78E-05, 3.46E-05, 4.71E-05, and 6.78E-04 for 238 U, 226 Ra, 232 Th, and 40 K, respectively. From these values, we can further conclude that radionuclides were transferred from the soil to the water. Sites located farther away from the tailings contained fewer radionuclides in the water.

Radiological hazard assessment of the soil
The radiological hazard indices (D γ , AEDE out , AGDE, and ELCR out ) were determined for the study area, and the results are summarized in Table 6. The radioactive hazard doses inside the reservoir were higher than those outside the reservoir and were mainly determined by the concentrations of nuclides in the soil in the study area. For D γ in air, the values obtained in the study ranged from 937-10955 nGy h −1 , with an average of 4764 nGy h −1 . The high D γ values near the tailings reservoir were caused by the high U and Ra concentrations. This D γ result was much higher than the average values of 53, 56, 76, 84, 66 and 62 nGy h −1 obtained in Japan, India, Spain, Portugal, Pakistan and China, respectively [12,42,43]. It is also much higher than the worldwide average of 57 nGyh −1 [12]. The mean AEDE out values were 8.4 mSv y −1 inside the reservoir and 4.6 μSv y −1 outside the reservoir, both of which are above the worldwide average of 0.07 mSv y −1 . Compared with other non-mining locations, the value of the study area was higher than that of the average value 0.1 mSv y −1 from four districts of the Punjab Province, Pakistan [44]. AGDE in the study area ranged from 6 to 74 mSv y −1 , with an average of 32 mSv y −1 , which is above the worldwide average of 0.30 nGyh −1    [12] 0.07 [12] 0.30 [45] 0.29 [25] [45] and approximately 13 times larger than 2.39 mSv y −1 reported for the Eastern Desert of Egypt [28]. This has been attributed to the high levels of 226 Ra found in the mine tailings. The mean D γ , AEDE out , AGDE, and ELCR out values in soil outside the reservoir were approximately half of the average values inside the reservoir. Thus, the radiological hazards near the tailings reservoir are higher than the worldwide average, thereby posing a severe radiological threat to the local population.

CONCLUSION
The activity concentrations of 238 U, 226 Ra, 232 Th, and 40 K were measured in soil and water samples in and around a uranium tailings reservoir using a low background HPGe semiconductor detector. Consequently, 238 U and 232 Th showed the highest and lowest activities in the soil samples from inside and outside the reservoir. The activity concentrations of the four radionuclides in the soil were all higher than the national average and the world average value. Because all the samples belonged to the same source and environment, the CVs of the radionuclide concentrations were weak in the reservoir, whereas the CVs outside the reservoir were moderate. This suggests that the nuclides in the soil outside the reservoir are non-uniform because of the influences of soil properties and human activity. The mean activity concentrations of 226 Ra in the water samples was higher than the WHO guidelines. In addition, the nuclide concentrations in the surface water around the tailings reservoir were lower than that in the discharged water. This may be caused by a continuous inflow of water and the adsorption of nuclides by riverbed sediments. The correlation analysis indicated that pH is a crucial factor that affects nuclide transport. The TFs of the radionuclides showed that the radioactive substances in the soil of tailing entered the surrounding water.
All the radiological hazard parameters (D γ , AEDE out , AGDE, and ELCR out ) exceeded the worldwide averages because of the high concentrations of 238 U and 226 Ra. From the inside to the outside of the reservoir, the mean hazard values in the soil were approximately halved. These results can serve as a baseline for radiation levels in a soil-water system and provide basic data for the appropriate assessment of radiation exposure to the local population of the study area.