Isotopic composition and source of plutonium in the Qinghai-Tibet Plateau frozen soils

The 239+240Pu activities and 240Pu/239Pu atom ratios in the frozen soils of the Yellow River Source Area (YRSA) were determined to examine the Pu source and evaluate its environmental risk. The 239+240Pu activities of surface frozen soils in the YRSA, ranging from 0.053 to 0.836 mBq g−1, are comparable to those observed in China elsewhere (0.005–1.990 mBq g−1). The 240Pu/239Pu atom ratios of surface soils in the YRSA are in the range of 0.168–0.201 (average = 0.187 ± 0.012, n = 6), comparable to the global fallout of 0.180 ± 0.014. Based on the latitudinal and spatial distribution of Pu isotopic composition, I clarified that the Pu source is mainly from global fallout at present. The activity levels of Pu in the YRSA do far not cause a Pu toxicity to the downstream drinking water even the frozen soil begins to melt and release Pu to the Yellow River. However, since close-in fallout from Lop Nor where the Chinese nuclear tests were carried out during 1964–1980, high deposition and accumulation of Pu was observed in the Chinese soil cores through synthesizing an expanded Pu dataset, which alerts us it is necessary to further monitor the Pu activity levels in the YRSA soil cores to ensure the safety of downstream drinking water. Finally, I point out that information on Pu isotopes would help in establishing a baseline for future environmental risk assessment.

www.nature.com/scientificreports www.nature.com/scientificreports/ the YRSA are Ngoring Lake (34.767°~35.083°N, 97.533°~98.900°E) with a surface area of ~610 km 2 and a total water volume of 10.8 × 10 9 m 3 , and Gyaring Lake (34.817°~35.001°N, 97.050°~98.450°E) with a surface area of ~525 km 2 and a total water volume of 4.8 × 10 9 m 3 . The YRSA is the river's major source area of streamflow and is also called "water tower" of the Yellow River. The Yellow River, the second largest river of China, has a catchment area of >750 000 km 2 and 5464 km length 22 , which flows from north-west to south-east with an average discharge of 22.6 m 3 s −1 recorded by the Huanghe (Yellow River) hydrological station (34.885°N, 98.172°E) during 1955-2005. The Yellow River, called the "mother river" of China, is critical to water resource and security of China. Meanwhile, the YRSA is located the downwind area from the Chinese nuclear tests in Lop Nor where 22 atmospheric nuclear tests were carried out during the period of 1964-1980 8 (Fig. S1 in the supplementary information). Thereinto, the six large nuclear tests of Lop Nor released a great number of radioactive debris (~18.5 Mt). The man-made radionuclides rather than just global fallout occurred in this area as local fallout. Therefore, the potential impact of artificial radionuclides and radioactive source identification in the YRSA is an issue of great concern to the public. To date, however, there has no Pu data reported for the YRSA frozen soil. This study, for the first time, reported the Pu isotopic ratio and activity level of frozen soil in the YRSA and sought to examine Pu source. Such source identification is further aided by the measurements of the relative abundance of Pu isotopes. For this study, I also aimed at investigating spatial distribution of Pu in the soil of China by synthesizing an expanded Pu dataset. Finally, I contend that information on Pu activities and isotopic ratios would greatly help in establishing a baseline for future environmental radiological risk assessment.

Results and Discussion
Isotopic composition and source of pu in the YRsA. 240 Pu/ 239 Pu atom ratio. Pu in soil tends to partition to special particles rather than to sorb homogeneously to all particles 23 . In order to avoid the effect of particle size, the soil samples (<63 µm) have been homogenized and are primarily composed of clay.
The 240 Pu/ 239 Pu atom ratios of surface soils in the YRSA ranged from 0.168 to 0.201 (average = 0.187 ± 0.012, n = 6), which were comparable to those observed in the Qinghai Lake (0.172-0.221, average = 0.189 ± 0.028, n = 3) 24 , at the distance of about 300 km from our studied area. Within the distance of 400 km from our investigated area, the 240 Pu/ 239 Pu atom ratios in the Sugan and Shuangta Lakes located in the downwind of Lop Nor were reported to be 0.166-0.188, with an average of 0.177 ± 0.011 (n = 3) 25 . At the northwest (upwind) 250 km of Lop Nor, the 240 Pu/ 239 Pu atom ratios in the Bosten Lake were observed to be 0.168-0.184, with a mean of 0.178 ± 0.012 (n = 3) 26 . Therefore, the surface 240 Pu/ 239 Pu atom ratios of soils/sediments around the YRSA are comparable to that of global fallout. This implies that: (1) the Pu in this area is mainly sourced from the global fallout; (2) the 240 Pu/ 239 Pu atom ratio, if there has any additional Pu source, is similar as that of global fallout. The possible additional sources in this area are consist of the nuclear accidents (e.g., Chernobyl and Fukushima), Semipalatinsk and Lop Nor nuclear tests. The 240 Pu/ 239 Pu atom ratio of Chernobyl nuclear accident was characterized by 0.408 14 and Kim et al. 27 confirmed that the Pu contribution derived from the Chernobyl nuclear accident was negligible in the marginal seas of the North Pacific Ocean. The 240 Pu/ 239 Pu atom ratio was also featured by 0.30-0.38 for the Fukushima nuclear accident 7,16 and our previous work in the China Sea suggested that there had no Pu signature from the Fukushima nuclear accident 9 . Meanwhile, the YRSA is far away from Chernobyl and Fukushima nuclear accidents than the China Sea. The 240 Pu/ 239 Pu atom ratios (~0.18) in the Chinese profile soils reveal that the direct input of close-in fallout from the Semipalatinsk nuclear tests did not lead to any significant Pu contribution in these areas 28,29 , in view of its distinctive 240 Pu/ 239 Pu atom ratios of 0.03-0.05 30 . Therefore, we can expel the Pu signatures from the Chernobyl and Fukushima nuclear accidents, and Semipalatinsk nuclear tests. Nevertheless, the 240 Pu/ 239 Pu characteristic atom ratio of Lop Nor has not been well defined up to date due to be lack of systematic studies. Leifer and Toonkel 31 measured high 240 Pu/ 239 Pu atom ratio (~0.224) in the atmospheric debris after the largest testing of Lop Nor (~4 Mt) on November 17, 1976. However, low 240 Pu/ 239 Pu atom ratios (0.080-0.103) in deep sediments were also observed in the downwind of Lop Nor, indicating the Pu atom ratio of Lop Nor could be lower than that of global fallout at this period 25,26 . Accordingly, Bu et al. 32 inferred that the 240 Pu/ 239 Pu atom ratios of Lop Nor were possibly in the range of 0.059-0.224 based on their observation in the downwind from the Lop Nor site in the Jiuquan area and the above published Pu data 25,26,31 . Therefore, they cannot well define the 240 Pu/ 239 Pu characteristic atom ratio of Lop Nor because of the sporadic reports and/or Pu dataset. Here, I combined the most complete Pu dataset in the Chinese surface soils up to date and did a statistical analysis in order to well define the 240 Pu/ 239 Pu characteristic atom ratio of Lop Nor.
Indeed, our 240 Pu/ 239 Pu atom ratios (0.187 ± 0.012, n = 6) were also comparable to the previous observations made elsewhere in the Chinese surface soils (0.183 ± 0.018, n = 71; Fig. 1), namely, Northwest China (0.178 ± 0.013, n = 21) [24][25][26]32 , Southwest China (0.187 ± 0.005, n = 6) 29,33-35 , South China (0.181 ± 0.008, n = 10) 9,36 , Central China (0.190 ± 0.025, n = 2) 37 , East China (0.177 ± 0.012, n = 5) 35,38 and Northeast China (0.187 ± 0.023, n = 27) 39,40 . As shown in Fig. 1, the relationship between 240 Pu/ 239 Pu atom ratio and the reciprocal of 239+240 Pu activity in the Chinese surface soils suggested the global fallout was a dominated Pu source. It is noted that the 240 Pu/ 239 Pu atom ratios have the different uncertainties in the different laboratories, which were potentially caused by the measuring instruments (e.g., MC-ICP-MS, SF-ICP-MS, AMS, TIMS). Whilst, our Pu data with the low uncertainties are mainly attributed to the MC-ICP-MS with multiple ion counting detectors (1-4) using synchronous measurements. In addition, I further analyzed the big dataset of Pu in the Chinese surface soils and plotted the frequency distribution of 240 Pu/ 239 Pu atom ratios in Fig. 2a, showing a typical Gaussian/ normal distribution. Among the 77 surface soil samples, the ranges of 0.17-0.18 and 0.18-0.19 account for 22% and 29%, respectively. Approximate 86% of 240 Pu/ 239 Pu atom ratios fall within the range of global fallout, which further confirmed the major Pu source in Chinese surface soils was from the Pu deposition of global fallout. The 240 Pu/ 239 Pu atom ratios in the Chinese surface soils are in good agreement with those observed in Japan surface soils collected around 1970s (0.183 ± 0.018) where the global fallout is also the dominated Pu source as suggested by Yang et al. 41 . In summary, I suggest, if the Pu additional source in the surface soil from Lop Nor, that www.nature.com/scientificreports www.nature.com/scientificreports/ the Pu atom ratio of Lop Nor should be identical to that of global fallout (0.180 ± 0.014). Indeed, a similar case occurred in the Enewetak atoll, the low 240 Pu/ 239 Pu atom ratios in the Lujor island (0.111-0.117) and Runit island (0.062-0.070) were also observed, but the 240 Pu/ 239 Pu atom ratios in the Marshall Islands were characterized by 0.300-0.360 because the high 240 Pu/ 239 Pu atom ratios were produced from the mainly large tests 42 .
I also firstly presented the latitudinal distribution of 240 Pu/ 239 Pu atom ratios in the Chinese surface soils in Fig. 3a, showing a slight increase with the latitude. For example, they increased from 0.180 at the latitudinal  www.nature.com/scientificreports www.nature.com/scientificreports/ zone of 20-30° to 0.186 at the latitudinal zone of 30-45°. Such pattern is also very consistent with the latitudinal distribution of global fallout, indicating the Pu in the surface soil of China is mainly sourced from the global fallout. Combined with the most complete Pu dataset up to now, the spatial distribution of 240 Pu/ 239 Pu atom ratio in the Chinese surface soils are also presented in Fig. 4a, showing a positive relationship with the latitude, as similar as the deposition of global fallout. Meanwhile, I did statistical analysis using a simple F-test and T-test in detail described in elsewhere 43 . The result shows the 240 Pu/ 239 Pu atom ratios between the Chinese surface soils and the global fallout had no significant difference (student t-test, p = 0.06 > 0.05). The horizontal distribution of 240 Pu/ 239 Pu atom ratio in the Chinese surface soils overall showed a uniform distribution pattern in the seven zones of China. Therefore, all this implies that the Pu of surface soils in these areas is mainly sourced from the global fallout at present. And then, I further examine the activity levels and inventories of 239+240 Pu in the Chinese soils.
239+240 Pu activity. Frozen soils are usually formed under long-term weathering, abrasion, handling and sedimentation conditions 44 . The mechanical properties of structure in the frozen soils are different from the other soil types 45 . There has no report about Pu isotopic composition in the frozen soil up to date. Here, I present the 239+240 Pu activities of surface frozen soils in the YRSA in Table S1, ranging from 0.053 to 0.836 mBq g −1 , an average of 0.37 ± 0.29 mBq g −1 (n = 6). Such 239+240 Pu activity levels were also comparable to those observed in the vicinity of our investigated area (within 500 km), for example, Qinghai Lake (0.48-0.53 mBq g −1 , n = 3) 24 39,40 . The surface 239+240 Pu activity levels in the frozen soils were comparable to those in other soil types, which is potentially related to their similar organic carbon content and grain size 29,35,39,47,48 . Overall, the 239+240 Pu activities in Chinese surface soils, ranging from 0.005 to 1.990 mBq g −1 , were comparable with the activity levels of Pu in soils in Japan (0.07-4.31 mBq g −1 ) 41 and in Korea (0.24-1.1 mBq g −1 ) 49 before the Fukushima Daiichi nuclear accident. The spatial distribution of 239+240 Pu activities in the Chinese surface soils are also presented in Fig. 4b, showing quite heterogeneous pattern, which was potentially caused by the multifarious factors, such as the physical and chemical properties of soil (e.g., mineral composition and the contents of organic matter), bacteria activity (e.g., anaerobic sulfate-reducing bacteria), precipitation, Eh, pH, the oxidation state and ionic size of Pu isotopes 11,[49][50][51] . In general, the 239+240 Pu activity is positive correlation with the precipitation and the concentration of organic matter 28,29,39,49 . The Pu speciation and its partition with soil particles size were influenced by the pH of soil. It is reported that the K d (distribution coefficient) of Pu in the bentonite was ~40% lower at 3 of pH than that at pH = 7 because the Pu at higher pH would help in being associated with soil particles 52 . In addition, the 239+240 Pu activities in the Chinese surface soils also show a wider variability range at the latitudinal zone of 20-45° and their latitudinal distribution also corresponds The 240 Pu/ 239 Pu atom ratios in those sediment/soil cores ranged from 0.038 to 0.273 (average = 0.178 ± 0.025, n = 343), with a wide range. As shown in Fig. 5a, two notable features have been observed: 1) the 240 Pu/ 239 Pu atom ratios in the entire soil cores correspond well with the range of global fallout, indicating the dominated Pu source was from the global fallout in the Chinese soils at the past decades. Furthermore, the frequency distribution of 240 Pu/ 239 Pu atom ratios in the Chinese profile samples also showed a typical Gaussian distribution. Among the 343 soil samples, the ranges of 0.17-0.18 and 0.18-0.19 account for 23% and 28%, respectively. About 77% of 240 Pu/ 239 Pu atom ratios correspond well to the global fallout, further suggesting the Pu was mainly sourced from the global fallout in the last sixty years. 2) the lower 240 Pu/ 239 Pu atom ratios observed in the subsurface soils indicated the Pu signature from Lop Nor as suggested by Wu et al. 25 , Liao et al. 26 and Bu et al. 32 . They also calculated the Pu contribution of Lop Nor to be 20-70%, showing high uncertainty due to be absent of Pu source term value and the accurately estimated method.
The 239+240 Pu activities in the sediment/soil cores ranged from 0.005 to 6.993 mBq g −1 , with an average of 0.659 ± 0.889 mBq g −1 (n = 343). As shown in Fig. 5b, three features should be noted: (1) the 239+240 Pu activities showed an exponential decline with an increase of soil depth, indicating there was dramatic influence by the radioactive fallout from the atmospheric nuclear tests since 1940s. (2) the maximum value of 239+240 Pu activities appeared at the subsurface was indicative of the maximum Pu deposition of global fallout in 1963 because the large-scale nuclear tests were conducted during the period 1961-1962 53 . (3) in the same latitudinal zone, the 239+240 Pu activities in sediments of Lake were significantly higher than those obtained in the soils, indicating the different migration behaviors in between. The mobility of Pu in the lake sediments was controlled by the changes of pH and oxidation state [54][55][56] . In contrast, the remigration of Pu in soils was mainly influenced by the Pu partitioning to colloidal and particulate matter 57,58 .
239+240 Pu inventory in Chinese soils. The 239+240 Pu inventory in the core is a useful indicator for evaluating the sedimentation and accumulation processes and it aids in deciphering the source function, deposition, soil erosion www.nature.com/scientificreports www.nature.com/scientificreports/ and lateral dispersion of Pu 9 . In this study, 239+240 Pu inventories in dry soils were calculated by summing their respective activities at each layer, according to the following equation 59 : where I represents the inventories of 239+240 Pu in the dry soils (mBq cm −2 ), N is the number of sampling layers, ρ s is the solid phase dry density, X is the thickness of the sampling interval i (cm), and A is the activity of the sampled interval (Bq kg −1 ). Uncertainties on inventories are the sum of the propagated error determined for each of the sampling intervals. Combined the 239+240 Pu inventories in Chinese sediments/soils, the latitudinal distribution of 239+240 Pu inventory in Chinese sediment/soil cores is presented in Fig. 6. In the latitudinal zone of 20°-30°N, the 239+240 Pu inventories in soils were slightly higher than the integrated global fallout of 36 Bq m −2 published by UNSCEAR 8 . In the latitudinal zone of 30°-45°N, the 239+240 Pu inventories in sediments/soils were higher than that of global fallout. The high accumulation of Pu in Chinese sediments/soils was potentially caused by the input of Pu signature from Lop Nor in 1964-1980, particularly in the high latitude close to the site of Lop Nor. However, the 239+240 Pu inventories in Chinese sediment/soil cores were lower than those observed in China Sea (South China Sea 9 : 365.6 ± 3.0 Bq m −2 ; East China Sea 38,60 : 333-407 Bq m −2 ). The latter source received Pu from Pacific Proving Grounds continuously transported into the China Sea via North Equatorial Current and Kuroshio current 9,10 . It is noted that 239+240 Pu inventories in lake sediments were significantly higher than those obtained in soils in the same latitudinal zone. In general, Pu in anaerobic sediments increases with increasing depth where plenty of organic matter has continuously been supplied, and Pu is likely to disperse and migrate over the whole core because of low K d values in the strongly anaerobic environment 51 . Implication of pu in the YRsA. The frozen soil, closely related to the terrestrial ecosystem, accumulates the man-made radionuclides originating from the atmospheric nuclear tests and/or the nuclear facilities. Among these radionuclides, Pu is considered as a very radiotoxic nuclide 61 . Pu behavior in soil is a matter of concern in the view of environmental radioactivity because of their high toxicity, long half-lives and a large risk for internal radiation exposure. It is thus important to understand the Pu isotopic composition and source in the YRSA in order to further evaluate their environmental risk. In this study, both 240 Pu/ 239 Pu atom ratio and 239+240 Pu activity have all pointed towards the Pu source was mainly from the global fallout at present. The activity level of 239+240 Pu in the YRSA was now between in low values or close to background of global fallout, which does far not cause a Pu toxicity to the downstream drinking water even the frozen soil begins to melt and release Pu to the Yellow River. However, the high deposition and accumulation of Pu was observed in the Chinese sediment/soil cores since the Chinese nuclear tests conducted in the period of 1964-1980, particularly in the downwind of Lop Nor. Note that, the deteriorated soil erosion and associated sedimentation arising from human activities have become a prominent problem, which would potentially result in downstream sedimentation in fields, land degradation, floodplains and water bodies, consequently affect the safety of water quality 39 . Meanwhile, Pu in the environment is serious in two exposure pathways, namely, the Pu in the soils released from the resuspension for inhalation and plant uptake for ingestion should be considered. It is therefore necessary to further monitor the Pu activity levels in the YRSA soil column to ensure the safety of downstream drinking water.

Materials and Methods
sample collection. Six surface soil samples were collected in the YRSA (close to Ngoring Lake and Gyaring Lake), which is about 800-1000 km southeast of Lop Nor, the Chinese nuclear test site during the period of 2014. Detailed sampling locations are marked in Fig. 7. Longitudes, latitudes and sampling date at the stations are presented in Table S1. After collection, they were dried at 110 °C for 24 h and pulverized using agate mortar and pestle sets. Then, they were calcined in a muffle furnace at 500 °C for 6 h to decompose the organic matter in preparation for Pu isotope analysis.
Pu isotope analysis. The separation of Pu has been used the two-stage ion-exchange chromatography.
Briefly, soil samples (~5.0 g) were weighed out and a known amount of 242 Pu (IRMM-085, European Commission Joint Research Centre, Belgium) was added to the soil samples as a yield monitor. The spiked samples were digested by heating on a hot-plate at 180-200 °C for ~4 h using 30 mL concentrated (conc.) HNO 3 in a sealed Teflon digestion tube. Pu in the sample solution was subsequently purified by the two-stage anion-exchange columns using AG 1-X8 and AG MP-1M (Bio-Rad). A small drop of the final sample solution was dissolved in 4% ultrapure HNO 3 (1.0 mL) and filtered for MC-ICP-MS analysis. The determination of Pu isotopes was conducted using MC-ICP-MS (Nu plasma HR, Nu Instruments Ltd., England) in a low resolution mode in order to obtain the maximal sensitivity in Xiamen University. The DSN-100 high efficiency sample introduction system with a membrane desolvation unit and a conical concentric nebulizer was used. The flow chart of analytical procedure for the Pu isotopes in soil/sediment samples are presented in Fig. S2. The chemical yield for Pu resulting from this analytical procedure was 66.4% ± 5.6%. In addition, for Pu measurements with MC-ICP-MS, the most significant interferences are usually caused by the formation of isobaric uranium hydrides ( 238 UH + ) and peak tailing from the 238 U + peak, resulting in overestimation of the 239 Pu signal. Our analytical procedure employed in this www.nature.com/scientificreports www.nature.com/scientificreports/ work was able to effectively eliminate the U interferences by achieving an extremely high U decontamination factor of 6.0 × 10 7 , which was comparable to previously reported values (3.0 × 10 7 -1.0 × 10 8 ) 62 .
The data quality and the mass bias correction were assured by regular analyses of the IAEA-443 (Irish Seawater) certified reference material ( 240 Pu/ 239 Pu = 0.228 ± 0.005, n = 2, verified value: 0.229 ± 0.006). The analytical method was also validated by analyzing with other reference materials: IAEA-384 (Fangataufa Lagoon Sediment) and IAEA-385 (Irish Sea Sediment) (International Atomic Energy Agency). The accuracies of the 239+240 Pu activities and the 240 Pu/ 239 Pu atom ratios were in good agreement with the certified and previously reported values (Table S2). In addition, the operational blank count rates for 239 Pu and 240 Pu were analyzed following the same chemical procedure for Pu determination in soil. The limit of detection (LOD) was calculated based on the International Union of Pure and Applied Chemistry recommendations 63 . The LOD was calculated to be 0.44 fg mL −1 for 239 Pu and 0.36 fg mL −1 for 240 Pu.