2.1 Sites and soil sampling

The Qinshan site, located in Haiyan County, Zhejiang Province, southeast China, consists of three nuclear power plants (NPPs) which are referred to as Plant I, Plant II and Plant III. As the first pressurized water reactor (PWR) (300 MWe) in China, Plant I began its operations in April 1994. Plant II comprises three PWR reactors (600 MWe), and their operations began in April 2002, May 2004, and August 2010, respectively. The two heavy-water reactor (HWRs, CANDU type, 700 MWe) in Plant III, were put into operation in December 2002 and July 2003. The Qinshan area is covered by a yellow soil.

As shown in Fig. 1, three sampling locations (A, B, C) were chosen considering the previling wind and the distance from the stacks of the HWRs in Plant III (P III), which released much more tritium than that of the PWRs. Site A, B and C, were located approximately 1300, 1800 and 2100 m away from the HWRs in the prevailing wind direction, respectively. 11 soil core samples were sampled using soil probes and shovels on 14^th August 2013. There had been no rain and there had been prevailing wind (Grade 3–5) contitions from the HWRs to the sampling sites for at least 5 days before sampling. The soil cores collected from Sites A, B and C were section-cut into serial column samples according to the following depths: 0–5, 5–10, 10–15, 15–20 and 20–25 cm). All the column soil samples were then immediately sealed in aluminum bags to prevent the loss of moisture. The samples were kept frozen at −18°C until required for analysis.

Fig. 1Full-screen View

Sites of collection of soil samples (A, B and C) from the Qinshan NPP (P III).

The physical characteristics of the soils including water content and organic content were determined ^[13,17] and are presented in Table 1.Water content (%) in soil was measured by dividing the difference in weight between a soil sample and the sample after oven-drying at 80°C by the original soil sample weight. Organic content (%) was determined by dividing the soil weight by the sample weight difference before and after combustion at 850°C.

2.2 Sample preparation

The soil samples from the depths of 5–10, 10–15 and 15–20 cm collected form Site A were combined into one sample because of the small amount of sample taken from these depth intervals. Similarly, samples collected from 15–20 and 20–25 cm were mixed for Sites B and C, respectively. The HTO was extracted from each soil sample by freeze-drying for approximately 24 h to determine its activity concentration, then the dried samples were placed into a drying oven at 80°C to ensure that the free water was completely removed. Next, the dried samples were immersed in tritium free rinse water to eliminate the exchangeable OBT and then dried again, following the methods introduced by other reseachers ^[7,13].

The OBT activity was measured in the dry matter remaining after HTO and the exchangeable tritium had been extracted from the sample. The measured OBT activity, therefore, represents the NE-OBT. For the NE-OBT measurement, after the pretreatment described above, each soil sample was combusted using the combustion apparatus illustrated in Fig. 2 ^[18]. Prior to soil combustion, the samples were ground into fine particles and small stones and root debris were removed with tweezers. Approximately 100–150 g of the ground soil was then added into a sample boat to combust for about 8 h at gradually increasing temperature from room temperature to 850°C. Throughout, the working temperature of catalytic zone, which was filled with noble metal catalyst, was fixed to 400°C. Combustion water was collected by a crycooler and approximately 5–8 g of water was generally collected in each combustion experiment.

Fig. 2Full-screen View

Schematic of the combustion apparatus

2.3 Liquid scintillation counting

We used 8 ml extracted water and 5 ml combustion water to determine the activities of HTO and NE-OBT, respectively. The extracted water was mixed with 12 ml of scintillation cocktail “ULTIMA GOLD μLLT” (PerkinElmer Ltd. America) for tritium measurement by using liquid scintillation counting (LSC, Aloka LB7). For the NE-OBT measurement, 15 ml of scintillation cocktail was mixed with 5 ml of combustion water. Low background and low diffusion polyethylene LSC vials (Packard) were used. Distilled water (Nestle Ltd.) was used to determine the background count rate. The counting time was 180 min for each sample and the measurement efficiency (ε) was determined using the inner standard source method ^[19]. The activity concentrations of HTO and NE-OBT (reported for combustion water) were calculated by formula:

where A is the activity concentration of either HTO or NE-OBT, Bq/L; n and n₀ are the gross count rate and background count rate, respectively, cpm; V is the volume of either extracted water or combusted water, ml; and ε is the counting efficiency, %. The limits of detection under the measurement conditions in our laboratory for HTO and NE-OBT are 3.52 Bq/L and 5.63 Bq/L, respectively.

3.1 Vertical profile of HTO and NE-OBT in soil

Both HTO and NE-OBT concentrations were determined using Equation (1), expressed as Bq/L. Figures 3a and b show the HTO and NE-OBT variability with soil layers, respectively. The uncertainty expressed in the figure mainly originates from the the volume of water for LSC and the LSC count rate.

Fig. 3Full-screen View

Vertical profile of (a) HTO and (b) NE-OBT concentrations (Bq/L) in soil. The vertical and horizontal bars represent the depth range of collected samples and the uncertainty (1σ), respectively.

The HTO concentrations at Sites B and C generally initially decreased and then increased with the increasing depths (except at 5–10 cm at Site B). However, the trend was not significant and the differences among the HTO concentrations in different soil layers were small. For Site A, the HTO concentration in the 20–25 cm soil layer was a little higher than those in 0-5 and 5-20cm soil layers in which the HTO concentrations were similar. Kim et al. ^[13] indicated that soil HTO in the surface layers is highly dependent on recent air and precipitation tritium concentrations. Thus, the HTO activities can infer the HTO concentrations in the ambient moisture at each site during the sampling period. The HTO concentrations in air moisture at Site A in the period from 2^nd to 12^th August and 16^th to 26^th August 2013, were routinely determined by the China National Nuclear Corporation (CNNC). The HTO concentrations in these two periods were 13.44 ± 3.84 Bq/L and 30.60 ± 6.70 Bq/L, respectively, which is close to the HTO concentration in the surface soil at Site A. This stuggests that the present results are credible. Among the vertical profiles of HTO, only the HTO concentrations at 5–10 cm soil depth at Site B were abnormal, which should be further studied.

The NE-OBT concentrations in the surface soils (0–5 cm) were considerably lower than those in the deep soils (5–25 cm) at all three sites. In the deep soil layers, the NE-OBT concentrations changed little at Sites A and C, while a decreasing trend in NE-OBT with increasing soil depth was observed at Site B. Kim et al.^[13] indicated that the highest NE-OBT activities occurred in the 0–5 cm layer. However, as shown in Fig. 5b, our result is contrary to theirs. Several factors may be responsible for the difference: firstly, the sites at which Kim et al. sampled are located near a historical HT release site, while our sampling sites were located at an NPP where HTO and HT are routinely released, which may lead to different fates of tritium in soil. Secondly, the surface soil may have been disturbed by human activities. Finally, the soil type in Qinshan area is different from the soil type that reported by Kim et al, which implies a different distribution of microorganisms between the two different soils. In order to better understand the NE-OBT distribution, further study should be conducted on the various effects of historical and routine release and the relationship between the NE-OBT and the distribution of microorganisms.

Fig. 5Full-screen View

Spatial distribution of HTO, NE-OBT concentrations and NE-OBT/HTO ratios at the three sampling sites (Sites A, B and C).

The NE-OBT/HTO ratios were determined according to the results shown in Fig. 3, and their vertical profiles are shown in Fig. 4. We suggest that the NE-OBT/HTO ratios are able to reflect the fate of tritium during this particular period of time because of the stable meteorological conditions prior to sampling.

Fig. 4Full-screen View

The vertical profile of NE-OBT/HTO ratios in soil. The vertical and horizontal error bars represent the depth range of collected samples and the uncertainty (1σ), respectively.

The NE-OBT/HTO ratios were between 0.42 ± 0.2 and 2.49 ± 0.1, and the arithmetic mean value for all 11 samples was 1.28 ± 0.27. This is higher than 0.7, which is the value that environmental transfer models approved for regulatory assessments of plant OBT from measured plant HTO activity ^[14,20]. The ratio is expected to be less than unity because of isotopic discrimination in chemical and biological processes, such as photosynthesis, since tritium has a greater mass than hydrogen ^[10,14]. However, Thompson et al. ^[14]measured the NE-OBT/HTO ratios for soil and vegetation samples at background locations and around the four nuclear facilities and they found that the values are generally above the expected value of 0.7, which has also been shown in some other studies ^[12,21,22]. According to the work of Thompson et al., the mean NE-OBT/HTO ratios for soil samples collected near other facilities ranged from 0.9 ± 1.7 to 1.1 ± 2.1, which is in agreement with our results.

The NE-OBT/HTO ratio shows an apparent vertical profile, increasing initially and then decreasing with increasing soil depth; however, the variations in the deep soils (5–25 cm) were minimal at Sites B and C. The minimum and maximum values appeared at 0–5 and 10–15 cm soil depth, respectively. The NE-OBT/HTO ratio can indicate the ability of organisms to accumulate tritium into the organic fraction ^[14]. Thus, the vertical profile results may be related to the distribution of the various types of organisms, especially microorganisms, in different soil layers. However, further research should be performed to validate this assumption. The NE-OBT/HTO ratios (Fig. 4), decreasing in the order Site A > Site C > Site B, also show a clear spatial distribution, which is further discussed in the following section.

3.2 Spatial distribution of HTO and NE-OBT in soil

Figure 5 shows a comprison of the HTO and NE-OBT concentrations and the NE-OBT/HTO ratios between Sites A, B and C. Because there were no samples from 5–10, 10–15 and 15–25 cm at Site A, only the samples from Sites B and C could be compared at these three soil depths (Fig. 5b to d).

Research has indicated that HTO concentrations generally decrease with increasing distance to a nuclear facility ^[14,23]. However, in the current study, the HTO in surface soil (0–5 cm) did not follow this trend (Fig. 5a). The HTO concentrations in soil are related to the HTO in surrounding environment at the moment of sampling. Thus, many factors other than distance from the facility, such as topography or evaporation may affect the diffusion, deposition and re-emission of tritium and thus affect the HTO concentrations in soil. However, we found that the HTO concentrations were larger at Site B than at Site C in all layers from 5–25 cm soil depth (Fig. 5b to d). This suggests that the tritium released from the NPP affected the spatial distribution of HTO in soil to some degree.

The spatial distribution results for NE-OBT in surface soil indicate that the soil NE-OBT concentrations at Sites B and C are similar. However, they were considerably lower than those of Site A, indicating that the distance to the HWRs is a major factor contributing to the observed spatial distribution of NE-OBT. The NE-OBT concentrations in deep soil also supported this conclusion, since the NE-OBT at Site B was higher than that at Site C in all of the deeper soil layers (5–25 cm).

As shown in Fig. 5, the NE-OBT/HTO ratios in surface soil (0–5 cm) at Site A were higher than those at Sites B and C. The NE-OBT/HTO ratios at Site B were a little higher than those at Site C in all layers. The NE-OBT/HTO ratio is an indicator of the ability of organisms to accumulate tritium into the organic fraction, so further studies are necessary to investigate whether the spatial profile arises from the distribution of various microorganisms at the different sites.