¹³⁷Cs AND ^239,240Pu ACTIVITY CONCENTRATIONS DISTRIBUTION IN WATERLOGGED AND NON-BOGGY SOILS OF LITHUANIA

2.1. Soil sampling and sampling area

The waterlogged and non-boggy soil cores were taken in different locations of Lithuania from the shores of three lakes (Lake Juodis, Lake Paršelis and Lake Bedugnis). All sampling points were selected on the shores of lakes with undisturbed soils, which have not been processed for many years. In these lakes, the increased water level due to rain and melting snow repeats every year, thereby flooding the nearby coast. This flooded soil is considered to be waterlogged soil. The sampling sites of this type were located at a distance of 1–1.5 m from the water edge of the lake. The collection sites for samples of non-boggy soils were located at a distance of ~10 m from the water edge of a lake and were never flooded. Lake Juodis (N 54°46'54", E 25°26'54") is a small lake in the north eastern part of Vilnius, located about 0.5 km south of Lake Balžis in the Tapeliai Landscape Reserve. Lake Juodis belongs to the Neris River basin. The area of the lake is 9.1 ha. The lake is elongated in shape in the northeastern direction by 0.73 km. The maximum width of the lake is approximately 0.17 km. The shores are high, only the northern coast is low and swampy. The soil sample was taken in 2012. The non-boggy soil core was taken at the western shore of the lake. The waterlogged soil core was taken in the vicinity of the lake in an old canal, which was periodically flooded in spring.

Lake Bedugnis (N 54°11'12", E 24°09'14") is located in the Varėna District (the southern part of Lithuania). The lake has an oval shape, the length from the southwest to the northeast is 0.18 km and the width is up to 0.12 km. The coasts are high, completely overgrown with trees, only the northern shore is low and swampy. The lake is not running and it belongs to the Kempė stream basin. The soil sample was taken in 2018. Impurities of clay substances are characteristic of the region soils. According to the New Classification of Soils of Lithuania [40] soil in the Varėna District is described as arenosol. It consists essentially of more than 70% of sand (0.05–2.0 mm) and less than 15% of clay (<0.002 mm).

Lake Paršelis (N 54°05'0", E 24°41'45") is located in a hollow in the Varėna District. The lake has an oval shape: the length from north to south is 0.5 km and the width is up to 0.35 km. Like Lake Bedugnis, Lake Paršelis has also no bottom sources and is fed exclusively by atmospheric precipitations. Lake shores are with steep slopes and the coastal part is low and swampy. The lake belongs to the Ūla River basin. The soil sample was taken in 2015.

Samples were taken by stabbing a steel tube of 30 cm height (diameter 12.5 cm). In the laboratory, all soil cores were sliced into 1 cm-thick layers. The uppermost layers of the soil samples taken for the study consisted of live moss and needles.

2.2. Radiochemical analysis and measurements

After slicing, the samples were dried for 24 h at 105°C and then homogenized. For the Pu radiochemical analysis, the samples were burned in a muffle furnace at 500°C overnight and then at 700°C for 2 h to decompose the organic matter. The organic matter content was determined on loss of ignition. The samples were spiked with a 0.0125 Bq ²⁴²Pu tracer as a yield monitor. Acid leaching with aqua regia and concentrated HNO₃ was employed. The leachate was filtered (Whatman 114, pore size 20–30 μm) and evaporated until wet residue.

The evaluation of Pu isotope activity concentration requires special procedures of radiochemical separation. Plutonium was pre-concentrated and purified by using ion exchange resin DOWEX 1×8 (NO₃^–), then the samples were processed and electrochemically deposited on stainless steel disks and measured by alpha spectrometry [41–42]. The plutonium activity concentration was measured with a semiconductor alpha spectrometer Octete Plus with the ORTEC large area (450 mm²) Si detectors (BU-020-450-AS) with the resolution capability of 20 keV. The spectra were analyzed applying Alpha Vision and Maestro programs. The detection limit for the counting time of 86400 s was about 10^–3 Bq for plutonium isotopes.

The soil samples were analysed for ¹³⁷Cs using a Canberra gamma-spectrometric system with a HPGe detector (model GC2520, 26.2% relative efficiency, resolution 1.76 keV/1.33 MeV) according to the gamma line at 661.62 keV of ^137mBa (a daughter product of ¹³⁷Cs). Measurements of the soil samples were carried out in standard geometry (‘dental’ plastic containers) and at known efficiencies (~1.6 – ~1.8%), which depended on the packing parameters of the sample as well as on the sample weight according to the density of samples. The radionuclide mixtures (¹⁵²Eu + ¹³⁷Cs) with two different densities (1.0 and 1.45 kg/L) prepared by the Russian Scientific Research Institute of Physical-Technical and Radiometric Measurements (Moscow, Russia) were used for efficiency calibrations. The calibration quality was certified using IAEA standards. Measurement errors of the ¹³⁷Cs activity concentration in the samples (standard deviation) were evaluated using the GENIE software program. They were less than 5% (standard deviation) for the measuring time of about 86400 s for the active samples corresponding to the soil layers with the maximal radiocesium activity concentrations in the respective vertical profiles and were not higher than 20% for less active samples (usually, the samples of low activity, corresponding to the tails of the respective vertical profiles) when the above errors were achieved during long-term measurements up to 500000 s.

Clay minerals were separated using the sieves with pore sizes of 1, 0.25, 0.1 and 0.063 mm instead of the usual sedimentation method due to a too small quantity of the soil. The last (smallest) fraction was used for X-ray diffraction (XRD) measurements. The XRD measurements were performed using a X-ray diffractometer Smart Lab (Rigaku) with a 9 kW X-ray source with a rotating Cu anode. The results of the XRD measurements allowed us to identify clay impurity, however, the composition of soil mineral content could not be determined due to the deficiency of the samples.

The results of the XRD measurements allowed us to identify the presence of clay impurities (muscovite KAl₃Si₃O₁₀(OH)₂) in the studied samples. The studied samples also contained a large amount of quartz and other minerals such as albite, anorthite and orthoclase.

3.1. Vertical profiles of ¹³⁷Cs and ^239,240Pu activity concentration in non-boggy and waterlogged soil cores

The vertical profiles of ¹³⁷Cs activity concentration as well as the organic matter content in the waterlogged and non-boggy soil cores were measured down to the 30 cm depth. The respective plutonium vertical profiles were measured down to the 30 cm depth in the soil cores near Lake Bedugnis and in the waterlogged soil core taken at the shore of Lake Juodis. In the non-boggy soil core taken near Lake Juodis, these measurements were carried out down to the 22 cm depth and in the both cores taken near Lake Paršelis down to the 15 cm depth. The activity concentrations of ^239,240Pu and ¹³⁷Cs in the waterlogged and non-boggy soil samples next to Lake Juodis, Lake Bedugnis and Lake Paršelis are presented in Figs. 1–3. According to the presented data (Figs. 1–3), the highest activity concentrations of ¹³⁷Cs and ^239,240Pu were in the surface layers down to the 5–8 cm depths.

The analysis of the vertical profiles of the radionuclide activity concentration in the non-boggy and waterlogged soil cores showed two more or less pronounced peaks of activity concentration. Two peaks of ^239,240Pu activity concentrations were determined in all profiles of the studied waterlogged and non-boggy soil profiles and three in the profile of the waterlogged soil of Lake Bedugnis (Figs. 1(a), 2(a) and 3(a)). All peaks of ^239,240Pu activity concentration are clearly visible, because these layers contain a relatively large amount of organic matter (see Fig. 4). Two peaks of ¹³⁷Cs activity concentrations were determined in the soil profiles of the waterlogged and non-boggy soil of Lake Juodis and Lake Bedugnis and one peak in the profiles of the waterlogged and non-boggy soil of Lake Paršelis (Figs. 1(b), 2(b) and 3(b)). The secondary peaks of ¹³⁷Cs activity concentration are not clearly visible on the non-boggy soil profiles of Lake Juodis and Lake Bedugnis, because there is a very small amount of organic matter in these layers (see Fig. 4).

Different depths of ^239,240Pu and ¹³⁷Cs activity concentrations peaks in the respective vertical profiles of waterlogged and non-boggy soil cores may indicate different migration abilities of the radionuclides in those media. Thus, more enhanced migration of both radionuclides is evident in the waterlogged soil of an old canal in the vicinity of Lake Juodis (Fig. 1(a, b)). In the absence of adsorbing substances in the soil, radionuclides can migrate deep into the soil, reaching the groundwater level. The same feature can be seen for radiocesium and plutonium in the waterlogged soil at the shore of Lake Paršelis (Fig. 2(a, b)). In the waterlogged soil core taken near Lake Bedugnis (Fig. 3(b)), the radiocesium peak is located deeper than that in the non-boggy soil. This can also be considered as the evidence of the higher mobility of radiocesium in the waterlogged soil. However, this result is only apparent. The fact is that in waterlogged soils during spring floods an additional layer of sediments may appear, contributing to the deepening of the peak of the radionuclide activity concentration. Also, it seems that in the waterlogged soil, radiocesium may be more mobile than plutonium, since its peak is located deeper. However, this may also be due to the radiocesium dominant uptake in these layers (6–10 cm) by impurities of clay materials, which are characteristic of these soils. Also, as can easily be seen from Figs. 1–3, the secondary peaks are most pronounced in the distributions of plutonium in the waterlogged soil. The appearance of the secondary peaks on the deep slope of the distribution also indicates the migration ability of radionuclides in the soil. This possibility arises during the decomposition of the organic substances and organic forms of the radionuclides due to the microbiological activity in soils when the soil was flooded with lake water. These processes are followed by the decomposition of other potentially unstable carbonate and oxide physicochemical forms of radionuclides and the release of radionuclides to the pore water. The black colour of the waterlogged soil samples indicates the possible presence of FeS compounds – typical compounds in the anaerobic soil regime. Comparing the mobility of radiocesium and plutonium in terms of possible sizes of the secondary peaks in Fig. 3 (soil core near Lake Bedugnis), we can conclude that radiocesium is more mobile in the nonboggy soil than plutonium.

For ^239,240Pu, taking into account additional deposits in the surface layer of waterlogged soil due to repeated floods, both peaks in the non-boggy and waterlogged soils near Lake Bedugnis coincide well (Fig. 3(a)). The vertical distribution of radionuclides showed an exponential decrease in the ^239,240Pu and ¹³⁷Cs activity concentration with depth (Figs. 1–3). These results agree with those of Koarashi et al. 2017 [43].

It is clearly visible (Figs. 1–3) that the ^239,240Pu and ¹³⁷Cs activity concentration in the waterlogged and non-boggy soil cores taken at the shores of Lake Juodis, Lake Paršelis and Lake Bedugnis noticeably differ. Thus, ¹³⁷Cs activity concentrations in the waterlogged soil core taken in the vicinity of Lake Juodis are much larger than those in the non-boggy soil. For Lake Paršelis, they are slightly larger than those in the non-boggy soil ones. And vice versa, in the non-boggy soil core taken near Lake Bedugnis, the activity concentrations of both radionuclides were noticeably larger than those in the waterlogged one. There is nothing unexpected in these results. These differences may be mainly due to different densities of the soil layer content. Also, the radioactive load of waterlogged soil depends on the conditions of deposition of the radioactive fallout on its surface. If the surface of such soil is covered with a layer of water for most time of the year, or, as it was in the case of the Chernobyl NPP accident – during the spring flood, then radionuclides fall accordingly to the water surface, which is usually covered by a film of surfactants. In this case, part of the radionuclides is removed from this surface due to the mechanism of surface self-cleaning [44]. Radionuclides that penetrate directly into the water are distributed depending on the stratification of the aquatic environment in significant water volumes. At the same time, their concentration in the water layer above the soil surface decreases, which leads to a decrease in the soil radioactive load. However, due to the water surface self-cleaning mechanism, in some sites of the lake shore, the so-called ‘hot spots’ may be created [44]. These are areas with a high concentration of radionuclides. They are usually located in the areas of waterlogged soil near the water edge. Their location on the coast is determined by the direction of dominant winds at the time of the arrival of radioactive air masses, since ‘hot spots’ are placed where radioactive foam was thrown onto the shore of the water body under the influence of the wind. Thus, on the coast of the same lake, in the areas of waterlogged soil, there may be zones with increased loads of radionuclides.

To compare the ¹³⁷Cs loads in soil cores from different lakes, the data for Lake Juodis and Lake Paršelis were corrected for the date when the loads of the soil cores from Lake Bedugnis were estimated (by the beginning of 2019). The maximum ¹³⁷Cs load in the core was estimated in the waterlogged soil sample taken in an old canal in the vicinity of Lake Juodis (2697 Bq/m²). In the respective non-boggy soil core, the load was lower and amounted to 1078 Bq/m². In other cases, the radioactive loads were smallest for Lake Paršelis and for the cores of the waterlogged soil were somewhat lower than those for the non-boggy ones. Thus, the loads in the waterlogged and non-boggy soil cores taken near Lake Paršelis amounted to 1306 and 1585 Bq/m², respectively. For the soil cores taken near Lake Bedugnis, they amounted to 1638 and 1811 Bq/m², respectively. Plutonium (^239,240Pu) loads in the waterlogged soil cores were also less than those in the non-boggy ones. Thus, the loads in the waterlogged soil cores taken near Lakes Juodis, Lake Bedugnis and Lake Paršelis amounted to 89, 57 and 18 Bq/m², respectively. The load in the non-boggy soil core taken near Lake Bedugnis amounted to 72 Bq/m². The load in the non-boggy core taken near Lake Juodis was measured down to the 22-cm depth and amounted to 47 Bq/m². Also, in the respective core taken near Lake Paršelis, the load was measured down to the 15-cm depth and was almost 8 Bq/m².

3.2. Distribution of organic matter in waterlogged and non-boggy soil cores

The uppermost layers (1–5 cm) of the waterlogged soil and non-boggy soil cores were rich in organic matter. The soil surface layers in the waterlogged shores of lakes Juodis and Paršelis contained leaves, needles and some moss and their organic matter content was highest (94.2 and 93.6%, respectively) (Fig. 4). The uppermost layer of the waterlogged soil core near Lake Bedugnis contained only needles and some moss and its organic matter content was ~66.0%. The deeper layers of the non-boggy soil samples taken near lakes Juodis, Paršelis and Bedugnis were composed mainly of sand and their organic content was rather small.

It is clearly visible from Fig. 4 that the difference in organic matter content between the waterlogged and non-boggy soil cores is significant. The waterlogged soil core taken near Lake Juodis differed from the other samples having an elevated amount of organics in soil surface layers. The soil from Lake Juodis had an approximately uniform distribution of organic matter in depth where it varied in a quite narrow range of 85–76%. This core differed from others because it was sampled in a dry swamp of the old canal.

Suggesting that the studied radionuclides (¹³⁷Cs and ^239,240Pu) have an affinity to organic matter, the correlation relationship between these radionuclides and organic matter was assessed. However, no consistent pattern of correlations could be established in our case. A statistically significant (p-value <0.05) linear correlation for ^239,240Pu activity concentrations with organic matter was obtained (Table 1) for two waterlogged soil profiles of the cores taken near Lake Paršelis (r = 0.89) and Lake Bedugnis (r = 0.6) (Table 1). Small correlation coefficients (r = 0.02 for ¹³⁷Cs) were obtained for the waterlogged soil profile near Lake Juodis and for two non-boggy soil profiles of the cores taken near Lake Paršelis (r = –0.19 for ^239,240Pu) and Lake Bedugnis (r = 0.48 for ^239,240Pu). For ¹³⁷Cs, a significant correlation was obtained for the non-boggy soil core taken near Lake Juodis (r = 0.95), Lake Paršelis (r = 0.77) and Lake Bedugnis (r = 0.83) and two waterlogged ones taken at the shores of Lakes Paršelis (r = 0.88) and Lake Bedugnis (r = 0.7). The corresponding correlation for ¹³⁷Cs and ^239.240Pu in the waterlogged soil core taken near Lake Juodis was insignificant (Table 1). The studies conducted in contaminated soils in Fukushima showed that ¹³⁷Cs was retained in the top 5 cm of surface soils, with considerably reduced amounts below the 5 cm depth [43]. Soil organic matter also has the ability to retain ¹³⁷Cs [45]. Several studies have shown that ¹³⁷Cs is more easily available to biological systems in organic matter-rich soils [46].

In a vertical profile in the forest soil from southern Germany, more than 72% of the ^239,240Pu, originating from nuclear weapon tests, was concentrated in the litter/humus (decomposed organic matter in soils) layer and the uppermost soil layer. The sequential extraction revealed that 81% of that Pu was in an organically bound form [47]. Thus, the vertical distribution of Pu can be attributed to the affinity of plutonium for natural organic matter. In the Chernobyl forest, a large portion of plutonium was also found in an organic matrix [35]. It means that the greatest influence on Pu sorption is due to organic matter. The fallout of Pu isotopes investigated in the surface soil samples taken from north western Italy [48–50] showed that the content of organic matter is the main factor influencing the Pu activity concentration, while other soil parameters are of lesser significance. Americium and Pu have higher affinity for organic matter than Cs, which enhances their retention in organic-rich soil. Pu is also known to bind strongly to organic matter in soils [34]. Humic substances and their components (fulvic and humic acids, and some others) which cover various individual compounds, have a large number of functional groups that might bind some kind of trace elements and radionuclides. Therefore, humic substances have been recognized as decisive materials in affecting plutonium behaviour in the environment [35]. The complexation may influence the mobility of fallout radionuclides. Hence, investigating the association of fallout radionuclides with organic acids, it may be possible to interpret the mobility of fallout radionuclides in soils [37].

In order to check the validity of the correlation between radionuclides and the content of organic matter in the studied soils, the physicochemical forms of radionuclides were determined in the sixth layer of the waterlogged soil core sampled near Lake Bedugnis. This is a layer with a high organic matter content (86.8%) and sufficiently high activity concentrations of radionuclides (18.6 Bq/kg for ¹³⁷Cs and 0.89 Bq/kg for ^239,240Pu). So, it turned out that in this layer the fraction of plutonium associated with organic matter was 49.5% while 28.5% of the radionuclide was associated with the residual fraction. These results confirm a great potential for plutonium migration in this soil. As for radiocesium, the results of this analysis were unexpected – almost 99.3% of radiocesium was in its residual fraction. Apparently, this was caused by the presence of an admixture of clay materials in the soil sample. At the same time, the organic fraction of radiocesium was practically absent. The waterlogged soil of the sample was acidic (pH 4.28); therefore, the fraction associated with carbonates was also absent. Similar results in a waterlogged soil sample may also indicate the presence of anaerobic conditions in the soil where a stagnant pore water regime is dominant. These data indicate the limited migration possibilities of radiocesium in the presence of clay substances in the anaerobic waterlogged soil. It can also be concluded that the distribution of radiocesium by depth in the waterlogged soil near Lake Bedugnis is stationary. It follows from these data that the above-described correlations of radiocesium with the content of organic matter in the sample, as well as other similar results described in the literature, should be treated with caution. Objectively, data on such correlations can be obtained only when determining the physicochemical forms of radionuclides in the soil, as well as on the basis of knowledge about the composition of the soil and its oxidation regime.

Table 1. Correlation coefficients (r) between ¹³⁷Cs and ^239,240Pu activity concentrations and the organic matter content in the non-boggy and waterlogged soil profiles for different soil core sampling sites.

3.3. Mean ratios (²³⁸Pu/^239,240Pu and ¹³⁷Cs/^239,240Pu) of the radionuclide activity concentrations in the vertical profiles of the studied waterlogged and non-boggy soil cores

The radionuclide activity ratios in soil vary with the contamination sources and can be utilized for their identification. The mean concentration ratios of ²³⁸Pu and ¹³⁷Cs with respect to ^239,240Pu in the waterlogged and non-boggy soils are shown in Table 2 and Fig. 5, respectively. To date, many approaches have been proposed for identifying dispersion sources of plutonium isotopes.

Thus, the results shown in Table 2 indicate that the ²³⁸Pu/^239,240Pu ratios in all analyzed soils are on an average 0.04, which is consistent with the global fallout ratio 0.04 [4, 6]. The highly refractory oxides of plutonium formed at very high temperatures of weapon explosions remained mostly in the environment as solid particles for very long periods of time [12].

Table 2. ²³⁸Pu/^239,240Pu activity concentration in the waterlogged and non-boggy soil cores.

It means that the appearance of radioactive air masses related to the Chernobyl accident over the studied area was not followed by new plutonium deposits. This result was very helpful in identifying the origin of radiocesium in vertical profiles of the waterlogged and non-boggy soils. The vertical profile of the radiocesium to plutonium activity concentration ratio in the non-boggy soil core sampled near Lake Juodis showed that mainly the surface layers down to the 3 cm depth were affected by the Chernobyl NPP deposits. For the waterlogged soil core, the presence of Chernobyl NPP radiocesium is traced throughout the entire depth of the vertical profile (Fig. 5). Since the current value of the ratio of activity concentration of radiocaesium to plutonium is estimated as 28, the ratios exceeding this value indicate the presence of Chernobyl NPP radiocesium. The presence of Chernobyl NPP radiocesium in deep soil layers in a sample taken near Lake Juodis indicates its increased migration. It is easy to see from Fig. 5 that enhanced radiocesium mobility is also characteristic of the non-boggy soils of the shores of Lake Bedugnis and the waterlogged soils of the shores of Lake Paršelis.

Thus, the vertical profile of the radiocesium to plutonium ratio in the waterlogged soil core sampled near Lake Bedugnis showed that the soil layers down to a depth of 15 cm are exposed to Chernobyl NPP radiocesium (Fig. 5). In the non-boggy soil core taken near Lake Paršelis, only the surface layers down to the 5 cm depth were affected by the Chernobyl NPP deposits.

Thus, the present data on plutonium and radiocesium vertical profiles in the soil core also imply that deeper-lying of radionuclide activities are related to the mobile fractions of the radionuclides in soils.

¹³⁷Cs ir ^239,240Pu AKTYVUMO KONCENTRACIJŲ PASISKIRSTYMAS SAUSAME IR UŽLIEJAMAME LIETUVOS DIRVOŽEMYJE

L. Kazakevičiūtė-Jakučiūnienė, N. Tarasiuk, E. Maceika, R. Druteikienė, M. Konstantinova, Z. Žukauskaitė, R. Gvozdaitė, Š. Buivydas

Fizinių ir technologijos mokslų centras, Vilnius, Lietuva