Introduction

The Semipalatinsk Test Site (STS) in North-East Kazakhstan, was the former Soviet Union’s first and the main testing grounds for nuclear weapons. At present, the radioactive contaminations at the epicenter areas are represented by fission products such as 137Cs, 90Sr, activation products such as 236U, 60Co, Eu isotopes, 3H, and radionuclides associated with fissile material such as 239+240Pu, 233U, 235U, 241Am [1, 2]. The area is expected to have natural uranium, however some 238U may have originated from the depleted or natural uranium tamper in some bomb designs [3].

The Degelen mountain range site was used for medium- and low-energy nuclear tests. Given that a total of 209 nuclear tests were carried out in 181 tunnels between 1961 and 1989, this site is considered to be a radiation hazard. The Degelen mountain range is part of the regional hydrogeological system on the western bank of the Irtysh River and occupies an area of about 220 km2, where the runoff from the surface water systems mainly flows in northerly and easterly directions. The main creeks situated here are Uzynbulak, Karabulak, Baytles and Toktakushik. The Uzynbulak creek valley is about 20 km in length and divides the Degelen mountain range from south to northwest. The creek itself—one of the longest at STS—and its tributaries are adjoined by approximately 50 adits. Over the course of many years of research, abnormally high uranium concentrations were recorded in the surface waters (up to 62 µg/l) and bottom sediments (up to 540 mg/kg) in Uzynbulak creek [4,5,6,7]. Based on previously conducted studies to assess the level of technogenic uranium content in the natural components at STS, the study of the uranium isotopic ratio here is particularly relevant. According to results from previous studies at STS, areas at this site were identified with altered uranium isotopic compositions, indicative of its technogenic origin. Technogenic uranium in contaminated soils was characterized by a high degree of potential mobility in the area [8]. The Uzynbulak creek basin and its tributaries cover a vast territory, much of which is located at the Degelen mountain range site, where earlier studies have recorded values exceeding the tentatively permissible level of uranium in water on multiple occasions [9]. The uranium isotopic ratio observed in water from almost every well and river is slightly higher than the natural indicators worldwide, indicating its influence by local nuclear testing at least in some watercourses [5]. The transport and redistribution of artificial radionuclides in groundwater from creeks at the Degelen mountain range test site have also previously been studied. The transfer of radionuclides in the water of creeks outside the Degelen mountain range is the main factor behind the formation of radiological conditions in adjacent territories. The radionuclide concentration sharply decreases as the distance from the entrances to the adits increases. According to a study by Yamamoto [11], the uranium content in the water from wells at STS varies significantly from 3.6 to 356 mBq/l (0.3–28.7 μg/l), the higher end of which is close to the maximum recommended uranium concentration of 30 µg/l in drinking water [10].

Currently, many sequential extraction schemes are used to determine the uranium speciation in bottom sediments, which is crucial to determine the mobility of uranium in the environment and the potential for groundwater contamination under changing environmental conditions. Sequential extraction involves the sequential fractionation of chemical elements from a single weighed sample, starting with the "weakest" and ending with the “strongest” extractant. On the basis of various methods, the following extracted fractions were used: water-soluble, exchangeable, weakly specifically sorbed, connected to organic matter, connected to iron and manganese oxides and hydroxides, and the residue, which is normally not available under environmental conditions [12,13,14,15,16]. However, it should be noted that, in general, uranium mobility depends on many factors, e.g. the nature of the element itself, physical and chemical environmental conditions, soil composition, etc. In this regard, changes in the uranium concentration distribution in the forms of various compounds can yield a wide range of data [17,18,19,20,21,22,23].

The purpose of this work is to assess the variation in spatial uranium in the studied forms of the bottom sediments along the Uzynbulak creek in order to determine the extent of uranium contamination as well as the isotopic ratio of 235U and 238U, thereby resolving if anthropogenic uranium from nuclear testing is a significant contributor.

Experimental data

Sampling and preparation of the water and bottom sediments

Sampling of the water and bottom sediments was conducted to assess the spatial distribution of uranium as well as its speciation along the course of the Uzynbulak creek in July 2015. Water was also sampled from the permanent watercourse of adit № 104, as a clearly formed channel was observed at its confluence with the Uzynbulak creek approximately 9 km from the source. In total, 22 samples were taken every 500 m up to 9 km from the source of the creek to the influx from adit № 104 as well as every 250 m from 9 to 13 km away (Fig. 1). It should be noted that samples were not collected along the entire length of the creek due to the channel drying out every summer. Water in the creek only flows continuously during the spring or heavy periods of rain, which renders sampling difficult if not unfeasible.

Fig. 1
figure 1

Sampling scheme for the Uzynbulak creek

Full size image

During water sampling, the following operations were carried out: water filtration to remove mechanical impurities through an ashless "blue tape" quantitative filter paper of grade 589/3; and sample preservation by adding concentrated nitric acid of very high purity at a rate of 3 ml of HNO3 per liter of water. Filtration and preservation were performed at the sampling site.

Bottom-sediment sampling was carried out by point sampling at a depth of 0–5 cm over 100 cm2 at locations where exchange of the pollutant between the water mass and bottom sediments could be characterized by extreme values where abnormally high uranium concentrations in surface waters were measured. Selected samples of bottom sediments weighing at least 1 kg were dried in a drying cabinet at 105 °C for 3–6 h before being passed through a sieve with holes of 1 mm in diameter. A 200 g sub-sample was taken by the quartering method and ground on a “Pulverisette 9” disc mill (FRITSCH, Germany) for 20 min at a rotational speed of 1000 rpm. A 50 g sub-sample was repeatedly extracted from the reground homogeneous sample by quartering and grinded once more for an additional 20 min before a final 5 g sub-sample was taken by quartering from the reground homogeneous sub-sample.

Sequential extraction method

Uranium speciation was determined by sequential extraction from a bottom-sediment sample taken at each of the 22 sampling points. The ratio of bottom sediments to extractant was 1:10 and the exposure time was 1 h over which the sample was shaken on a rotator. Next, the solution was separated from the solid phase by centrifugation using a Digicen 21 centrifuge (“OrtoAlresa”, Spain) for 10 min at a rotational speed of 4000–6000 rpm. After each extractant had been separated, the samples were washed with distilled water. The resulting solution was diluted in a 1:10 ratio. In the process, fractions were isolated [24, 25], as shown in Table 1. Sequential extraction involves sequential fractionation of chemical elements from a sample, starting with the “weakest” extractant and ending with the strongest one.

Table 1 The six environmental compartments of uranium compounds and their extraction conditions from bottom sediments
Full size table

Conducting analytical work

The uranium content of the samples was determined by inductively coupled plasma mass spectrometry (ICP-MS) using an Agilent 7700x-type quadrupole mass spectrometer by Agilent Technologies and an “iCAP 6300 Duo” inductively coupled plasma atomic emission spectrometer (ICP-AES) by Thermo Scientific. Multi-element standard solutions were used to plot calibration curves.

The concentration and ratio of uranium isotopes (235U and 238U) were determined by ICP-MS using a “ELAN 9000” quadrupole mass spectrometer by PerkinElmer, USA. Solutions prepared from a standard solution containing 300 mg/l of uranium ions were used to calibrate the mass spectrometer. Calibration was carried out using three standard solutions of uranium at concentrations of 30, 150 and 300 μg/l. After every ten measurements, control determination of the certified aqueous standard, namely a 150 μg/l uranium solution, was performed. The uranium limit of quantification (10σ, n = 10) was 0.0002 mg/kg.

Quality control of the measurements was carried out by measuring the calibration solution after every 10 samples. In the case of unsatisfactory calibration results (8–10% deviation from the calibration curve), the instrument was recalibrated by taking into account the new background parameters.

Results and discussion

Study of the spatial distribution of uranium in the “water-bottom sediments” system

The formation processes and chemical composition with regard to the geochemical characteristics of water are the same when considering homogeneous hydrogeochemical complexes. The levels of accumulation of chemical elements within these complexes can be estimated [33] at different stages of the interaction of water with rocks, moreover, the water is enriched with chemical elements. Since the degree of transformation is determined by the nature of water exchange, so is the stage of interaction in the “water-rock” system [34,35,36].

It should also be noted that bottom sediments can be a source of secondary pollution in aquatic ecosystems [37], therefore, studying the migration of heavy metals in the “water-bottom sediment” system is very much relevant.

The spatial distribution of U in the “water-bottom sediment” system during the summer period is shown in Fig. 2 with some data points exceeding the MAC (maximum acceptable concentration).

Fig. 2
figure 2

U content in the “water-bottom sediment” system of the Uzynbulak creek (summer period)

Full size image

The Fig. 2 and Table 2 shows that in areas where the concentrations of uranium in water were high, this element accumulates in the bottom sediments, starting from the source of the creek and throughout the watercourse. In addition, the maximum uranium content in bottom sediments was shifted relative to the peak of the concentration of this element in the water caused by the influence of inflowing water from tunnels [38]. The mechanism determining the behavior of uranium in water shows that this element is not immediately deposited in bottom sediments, rather migrates downstream and is actively sorbed by bottom sediments. The direction in which micro-components migrate depends not only on the chemical properties of the elements but also on the physical and chemical properties of bottom sediments.

Table 2 Uranium contentrations in the "water - bottom sediment" system of the Uzynbulak creek (summer period)
Full size table

A special role in the adsorption of heavy metals is played by humus as well as fine dust and silt fractions. It is known that between 9 and 13 km from the headwaters of the Uzynbulak creek, the content of fine dust and silt fractions increases by several orders of magnitude. The organic matter content over this stretch is due to the abundance of vegetation, which is a source of organic matter deposition [39].

Based on the data obtained, it can be assumed that the main mechanism resulting in the occurrence of a high uranium content in the surface waters of the Uzynbulak creek is the inflow of water of a different chemical composition. Particular attention should be paid to the inflow of groundwater with a different redox potential. The ORP (Oxidation Reduction Potential) of water in Uzynbulak creek fluctuates throughout the channel from − 44 to + 51 mV. The redox potential of water characterizes the ratio of oxidized to reduced forms of all elements with variable valency that it contains. In reductive (oxygen-free) water, iron ions are in the form of Fe2+, determining their ability to migrate in the creek. Conversely, in oxygenated water, iron is in the form of Fe3+, forming minerals that are barely soluble.

Figure 3 shows the behavior of Fe depending on the redox reaction that occurs in the Uzynbulak creek during the summer period.

Fig. 3
figure 3

The behavior of Fe depending on the redox reaction of the Uzynbulak creek in the summer period

Full size image

Figure 3 shows that Fe concentration is strongly dependent on changes in the redox potential. In all likelihood, the increased content of this element in water is connected with it being leached from bottom sediments due to the inflow of groundwater with a different redox potential. As a result, when the redox potential decreases by 55 mV, the Fe concentration increases and becomes several orders of magnitude higher than on average.

Abnormally high concentrations of uranium at the source of the Uzynbulak creek were detected as well as peak concentrations of these elements over the stretch 9–11 km from the source. No outflows of this element were observed beyond the border of the Degelen mountain range site and its content was below the MAC. In general, the increase in the concentration of this element in the surface water of the creek is connected with the influence of groundwater and water from tunnels. Deposition in bottom sediments is associated with the mechanism of U sorption from water in the creek.

Uranium speciation in the bottom sediments of Uzynbulak creek

Since no environmental standards for the uranium content in the bottom sediments have been approved (Table 3), this parameter in the various environmental compartments in the bottom sediments of the creek according to sequential extraction was compared with the Clarke numbers of uranium in soil according to Vinogradov [26].

Table 3 Uranium speciation in bottom sediments of the creek during sequential extraction (n = 22), mg/kg
Full size table

The gross uranium content exceeds its worldwide average in soils by 4.8 times. The greatest content of the studied element was measured in stable, environmentally less available forms bound to iron and manganese hydroxides and oxides or, to a lesser degree, was associated with the residual form that typically cannot be leached. It should also be noted that a significant degree of accumulation in the weakly specifically sorbed form was found.

The average percentage distribution of U between the various environmental compartments is compared to the total content in Fig. 4.

Fig. 4
figure 4

Forms of U compounds as a gross content percentage. Note 1—water-soluble, 2—exchangeable, 3—weakly specifically sorbed, 4—connected with organic matter, 5—connected with oxides/hydroxides of iron and manganese, 6—residual

Full size image

It was found that uranium in the Uzynbulak creek bed is characterized by a low migration ability of no more than 7% of the gross content in the easily accessible forms. For this element, the maximum degree of accumulation of 73% was in the sedimentary form and assumed to be adsorbed on the surface of Fe and Mn oxides and hydroxides [27,28,29,30], which combined with the 18% in the residual form suggests that 91% of the uranium is stable. Fe and Mn oxides and hydroxides that are thermodynamically unstable under anoxic conditions [40], which are atypical of running creeks, can occur under hypereutrophic conditions, where decaying plant matter uses up all the available oxygen on the bed, or from intrusions of anoxic groundwater. On the other hand, even if this situation were to arise, uranium might still be retained on organic matter and adsorbed to particulate organic carbon due to complexation under anoxic conditions [41].

Spatial distribution of uranium speciation in the bottom sediments along Uzynbulak creek

The spatial distribution of uranium speciation in the bottom sediments will now be considered in detail. Changes in the spatial distribution of uranium speciation in the bottom sediments along Uzynbulak creek are presented in Fig. 5 below.

Fig. 5
figure 5

Spatial changes in uranium speciation compounds (a, b, c) and gross U content (d) in bottom sediments of the Uzynbulak creek. Note H2O—water-soluble, Ca(NO3)2—exchangeable, CH3COOH—weakly specifically sorbed, NaOH—connected with organic matter, Tamma (Oxalic acid/ammonium oxalate)—connected with oxides/hydroxides of iron and manganese, HF + HNO3—residual

Full size image

Figure 5 shows that uranium accumulates to a lesser degree in mobile forms since the water-soluble and exchangeable fractions hold less than 2% of its total concentration along the observed stretch. Therefore, U in the bottom sediments studied is present in a firmly bound form and has a low migration ability. The weakly sorbed fraction fluctuates somewhat between 1 and 10%, while the organically bound one starts to increase before the 10 km mark. The highest total U concentrations were observed at the source of the creek, probably due to the dissolution of minerals underground between 9 and 11 km as a result of an influx from adit № 104.

Uranium isotope ratio in water flowing through Uzynbulak creek and from adit № 104

After identifying high U concentrations in the bottom sediments of Uzynbulak creek, it was important to determine whether its origins were natural or man-made. Table 4 presents data on the uranium isotopic composition of water in the creek as well as in adit № 104, which is located within the boundaries surrounding the Degelen mountain range test site and has the same watercourse as Uzynbulak creek, where water from the creek dilutes the incoming high uranium concentrations.

Table 4 Isotopic composition of uranium (235U and 238U) in the Uzynbulak creek water and the watercourse of adit № 104
Full size table

The U isotopic composition indicates that the observed samples are of natural origin which, according to previous studies, corresponds to the geochemical background of the Degelen mountain range. Subsequently, no signs of the presence or migration of man-made uranium in the studied water bodies were found.

In the analyzed bottom sediments, the average U content is several orders of magnitude higher than its average Clarke value in soils worldwide. This element mainly accumulated in the most stable forms of compounds, namely bound to Fe and Mn oxides and hydroxides, and in the residual form. The transfer of it from bottom sediments to surface water depends on the physical and chemical conditions at the interface between the solid and liquid phases. For example, when the equilibrium conditions—first the pH, Eh and other general physicochemical characteristics—are disturbed, the migration and mobility of heavy elements, including uranium, can increase, transferring them into the aquatic environment. For instance, Fe and Mn oxides and hydroxides are thermodynamically unstable under anoxic conditions as mentioned in Section “Sequential extraction method”.

Assessment of the radiation hazard of the bottom sediments and Uzynbulak creek

The level of intervention resulting from the activity concentration of uranium in water and the minimum significant specific activity values are set so that under normal conditions the annual effective dose the adult population is exposed to would not exceed 0.1 mSv [31]. For natural uranium isotopes, the levels of intervention are 2.9 and 3.0 Bq/l for 235U and 238U, respectively, according to regulations in Kazakhstan [Order of the Minister of Health of the Republic of Kazakhstan dated August 2, 2022, № ҚP ДCM-71/QR DSM-71] [42]. At these activity concentrations, the chemical toxicity of 238U is a more significant hazard than its radiotoxicity. As can be seen in Table 4, the activity concentrations in the water in adit № 104 exceed regulatory values [32], including the WHO standards for drinking water [31].

The water flow of adit № 104 also affects the uranium concentration in the bottom sediments, which is particularly visible 9 km from the source at the confluence of the water flow from adit № 104 with Uzynbulak creek. As is shown in the pie chart presented in Fig. 6, at several sampling points (1, 6-1, 7, 7-1, 8, 9, 9-1), the minimum significant specific activity (MSA) was exceeded. MSA is critical according to the national regulations [Order of the Minister of Health of the Republic of Kazakhstan dated August 2, 2022, № ҚP ДCM-71/QR DSM-71] [42], which ensures that the individual annual effective dose the population is exposed to does not exceed 1 mSv nor does the collective effective dose exceed 1 manSv under any conditions.

Fig. 6
figure 6

Specific activity of 238U in the bottom sediments at different sampling points

Full size image

In cases where the MSA was exceeded in bottom sediments over these stretches of the creek is associated with the sorption mechanism in the “water-bottom sediment” system. The specific activity values of uranium vary considerably (from 20 to 6000 Bq/kg), the median of which is 156 Bq/kg. The average concentration of uranium in the Earth’s crust is 2.7 mg/kg or 68 Bq/kg and the maximum concentrations observed are similar to those seen in black shale or low-grade uranium ore.

The uranium activity concentration measured of water in Uzynbulak creek did not exceed 0.4 Bq/l, which is less than the level of intervention, moreover, well within the range observed in surface waters and water wells used as sources of drinking water in several countries [43]. Consequently, the influx of uranium from adits along the observed stretch and the migration processes of this element, e.g. the transition from bottom sediments to the aquatic environment, currently do not have a significant economic impact on the level of risk originating from the potential radiation and chemical toxicity of water in the creek despite significant uranium concentrations in the influx from adit № 104. It probably would be worthwhile reevaluating this issue in the future to check if it is stable or the capacity of the creek bed is sufficient to retain the incoming uranium in the long term.

Conclusions

Uranium speciation and its probable migration ability were evaluated in the bottom sediments sampled along the Uzynbulak creek. The highest concentrations of this element in both water and bottom sediments here were observed between 9 and 11 km from the source, starting from the confluence of the watercourse of adit № 104 with Uzynbulak creek.

From analysing the obtained research results, it can be concluded that uranium speciation is heterogeneous in bottom sediments found in Uzynbulak creek. The absolute concentrations and relative weight associated with the environmental compartments of the studied uranium compounds (with insignificant deviations) were in the following order:

$$ {\text{CE}}_{{{\text{exc}}}} < {\text{CE}}_{{{\text{water}}}} = {\text{CE}}_{{{\text{org}}}} < {\text{CE}}_{{{\text{sp}}.{\text{sorb}}}} < {\text{CE}}_{{{\text{res}}}} < {\text{CE}}_{{{\text{ox}}/{\text{hydr Fe }}\& {\text{ Mn}}}} . $$

It should be noted that to all intents and purposes, uranium was not converted into water-soluble and exchangeable forms easily accessible to plants. However, the gross content of this element exceeds the worldwide average concentrations in soil by a factor of 4.8. The greatest accumulation of U (73%) occurred in forms of compounds that were expected to be stable in the area, that is, iron and manganese oxides and hydroxides, so the risk of migration is low.

Based on the results of the gross uranium content and its isotopic ratio, it can be concluded that it probably naturally originated as a consequence of the geochemical features of the Degelen mountain range, namely the natural weathering of rocks. Signs of the presence or migration of man-made uranium in the studied water body were not observed.