Natural Radioactivity in Soil and Radiological Risk Assessment in Lișava Uranium Mining Sector, Banat Mountains, Romania

Abstract

The specific activity and spatial distribution of ²³⁸U, ²³²Th and ⁴⁰K were determined in the surface soil from the Lișava uranium mining sector. This sector belongs to the Banat district, an historically important uranium mining area in Romania (an area with closed uranium mines and a radioactive waste dump). Gamma-ray spectrometry using a high-purity germanium (HPGe) detector was used to measure the activity of naturally occurring radionuclides in the soil. The average specific activities of ²³⁸U, ²³²Th and ⁴⁰K in the soil were 197.21 Bq/kg for ²³⁸U, 16.21 Bq/kg for ²³²Th and 543.21 Bq/kg for ⁴⁰K. The mineral contents of selected waste rock samples (sandstones) were examined using a scanning electron microscope (SEM), which revealed that brannerite, pitchblende and coffinite were the most important uranium-bearing minerals. The means of the radiological hazard parameters were calculated to be 262.22 Bq/kg radium equivalent activity (Raeq), 123.72 nGy/h absorbed gamma dose rates (DR), 0.7 external hazard index (Hex) and 1.8 representative level index (RLI). The spatial distribution of the risk assessment indices associated with the investigated soils exceeded the median values provided by UNSCEAR and reflected the geological settings and influences of anthropic activities such as uranium mining practices and the tipping of radioactive mining waste.

1. Introduction

Uranium mining practices usually result in an increase in radionuclide concentrations in the environment, which can lead to increased radiation exposure for the population [1]. Compared to radioactive waste from nuclear power plants, waste from uranium mining activities is characterized by a significantly lower level of radioactivity, but in the case of large volumes of naturally radioactive waste, an increase in the risk of radiation exposure for the local population can occur [2]. This type of waste is generally described as a mixture of host rocks and mineralized rocks that are uneconomic [2,3].

The occurring radionuclides present in all types of rocks, soils, waters and air in uranium mining perimeters consist mainly of the natural radioactive decay series of ²³⁸U, ²³²Th and ⁴⁰K [4,5,6]. The distribution of and variation in radionuclides in surface soil depend on the composition and distribution of radioelements in the bedrock and their physical and mechanical properties (porosity, permeability) [7]. Likewise, the concentrations of natural radioactive elements in soils depend on additional factors such as the stability of the primary radioactive-material-bearing minerals to weathering, the presence of a clay phase, the organic matter content and the physico-chemical characteristics of the soils, as well as the specific geochemical characteristics of uranium, thorium and potassium [4,5,8].

Uranium is a lithophilic metallic element with an average abundance of about 2.7 ppm in the upper continental crust [9]. Like uranium, thorium is a rare element and is the most abundant heavy element, with a continental crustal abundance of 10.5 ppm [9]. In igneous rocks, uranium appears as U⁴⁺ with crystallochemical properties similar to Th⁴⁺ and the light rare earth elements. Under endogenic conditions, both parental radioactive elements (U⁴⁺ and Th⁴⁺) crystallize together as late products of magmatic differentiation and are always concentrated in acidic rocks as accessory minerals. Under exogenic conditions, the chemical properties of uranium and thorium are markedly different. Uranium has a high geochemical mobility; it is oxidized easily from its tetravalent state to its hexavalent state [9,10,11]. Under these circumstances, uranium can be oxidized to uranyl, which is very soluble and easily transportable in groundwater and surface waters. Compared to uranium, thorium cannot be oxidized up to the hexavalent state and therefore has no analogue of the uranyl ion; any amount of thorium dissolved is adsorbed and precipitated as a hydrolyzed ion due to the potential of the tetravalent ions of thorium [10,11].

Uranium is a substantial component of several minerals, the most notable of which are uraninite, brannerite and carnotite. Thorium may be found in a variety of minerals, including thorite and thorianite [12]. Both elements (thorium and uranium) coexist in accessory minerals such as zircon, monazite, apatite, allanite and titanite, which are found in igneous and metamorphic rocks, with zircon and monazite being the most resistant to weathering. Uranium may also combine in complex ways with organic materials and phosphate materials [13].

Potassium-40 is a naturally occurring radioactive potassium isotope that is found in trace amounts in all potassium-bearing minerals. Under normal conditions, potassium is a volatile lithophilic metal that is monovalent and has an average concentration of 2.8% in the upper continental crust [9]. Potassium is found in metamorphic and magmatic rocks as a major element of the feldspar mineral series, feldspathoids and micas. In sedimentary rocks, potassium can usually be found in sedimentary rocks due to their content in alkali and plagioclase feldspars [10].

Several natural occurrences and deposits of radioactive mineral ores with a high uranium content are known in Romania. Uranium deposits have been reported in: (i) the Eastern Carpathians as vein-type U deposits hosted in metamorphic rocks [14]; (ii) the Apuseni Mountains (western Romania) as sandstone deposits (such as the Băița Bihor deposit) hosted in Permian sandstone and as a vein-type deposit (such as the Avram Iancu deposit) found in the carbonatic horizon of metamorphic sequences (crystalline schists) belonging to the Biharia unit [14]; and (iii) the Banat mountains with, predominantly, sandstone U-type deposits (Ciudanovița, Natra and Dobrei South) hosted in Permian sedimentary rocks (sandstones, conglomerates and shales) [14].

In Romania, uranium ore has been exploited since the 1950s. In the Banat Mountains, the mining activity started in the Ciudanovița sector (Ciudanovița and Natra deposits) in 1952–1953. The Ciudanovița uranium ore was mined by Sovrom and exported to the USSR for ore processing until 1964. Mining activity at Natra started in 1962 and was suspended in 1988 due to the depletion of the ore deposit. In the Lișava sector (Dobrei North, Dobrei East and Dobrei South ore deposits), mining opened in 1962 at Dobrei South and Natra and in 1978 at the Dobrei North deposit. Mining activity ceased in 1999, followed by a partial process of reconstruction and rehabilitation [15,16].

More than 40 years of prospecting and mining have generated heavily contaminated soil surfaces and numerous waste dumps that include sterile rocks and low-grade ore dumps (economically unprofitable) resulting from radiometric separation [16,17]. However, even though these represent an environmental problem, there is a lack of data regarding the potential impact of radioactivity and the host of radioactive elements. This study was the first to determine the specific activity of ²³⁸U, ²³²Th and ⁴⁰K in the surface soil sampled from the Lișava uranium mining sector and was the first to identify the source of these radionuclides in areas of increased radioactivity. Furthermore, radiological risk parameters such as radium equivalent activity (Raeq), external hazard index (Hex), gamma absorbed dose rates (DR) and gamma representative level index (RLI) were calculated to assess the external exposure. Thus, this study will be a reference point for future environmental studies dealing with radiologic risk assessments and background radiation in the region.

2. Materials and Methods

2.1. Study Area and Sampling

The Lișava uranium mining sector is located about 5 km south Ciudanovița, in the western margin of the Anina Mountain part of the Banat Mountains (southwestern Romania). In the area of interest (45°06′20″ N; 21°03′33″ E), the exploration and mining of uranium ore was undertaken using galleries from the Dobrei North, Dobrei South and Dobrei East deposits. Mining activities were stopped in 1999, and the area has been partly rehabilitated [17]. The perimeter included the Lișava tailing dump, formed from the tailings deposited from the mines mentioned above, and the zone where ore was picked up and transported to the processing plant (Figure 1).

From a geological perspective, the Lișava sector belongs to the west Banat region in the southern Carpathians and is part of the Alpine orogenic belt. The regional lithostratigraphic formations are, from the oldest to the newest: a Proterozoic layer, represented by crystalline schists, followed by Carboniferous formations consisting of grey conglomerate layers, sandy layers and organic-rich black shales, sandstone and siltstone [19]. Overlying the Carboniferous layer are Permian formations made up of the Lower Permian (sandstone, siltstone and shale) and Upper Permian (Ciudanovița Series) that hosts uranium deposits, comprising conglomerate, bituminous sandstone, sandstone, siltstone and carbonaceous shale deposited in five cycles. The uppermost formations are Jurassic layers composed of grey limestones and marls lying over conglomerates, as shown in Figure 1A.

The Ciudanovița sedimentation cycle has 30–40 m thick hosted uranium deposits in the western and eastern flank of the Natra-Gîrliște Anticline (Figure 2). The Ciudanovița Series have five sequences in the eastern part of the Natra-Gîrliște Anticline. Each rhythm is between 20 and 200 m thick and begins from the bottom to the top with conglomerate, ending with fine-grained facies, which are predominantly siltstone. Three out of the five cycles contain uranium, and every cycle comprises cross-bedded microconglomerates, gritstone, sandstone and bituminous siltstone of an alluvial and lacustrine–alluvial origin. Intercalated with uraniferous-bearing cycles, barren rocks represented by well-sorted conglomerate and sandstone derived from extrusive igneous rock occur [19,20,21].

The dump material is a mixture of sterile rock and fragments of mineralized rock represented by red sandstones and reddish or grey microconglomerates that incorporate uneconomic small amounts of uranium. The waste deposit is compacted and covered by a layer of surface soil overgrown with vegetation that has grown spontaneously on the entire surface of the waste dump, which is more developed in the western part of the dump.

The soil samples were collected from 47 sampling points that were uniformly distributed over the whole surface of the Lișava uranium mining sector. The sampling point locations are presented in Figure 1B. Soil sampling was conducted by manually boring holes with a diameter of 55 mm drilling at a depth range of 0–15 cm. At each point, five samples were collected from the corners and center of a square with ten-meter sides. The permanent grass cover was excluded. After collection, the samples were mixed, resulting in one composite sample weighing about 1.5 kg, and placed into a plastic bag.

Together with the soil samples, three mineralized rock samples from the dump material were collected. The rock samples were selected based on dose rate measurements in air using a radiation survey meter (Berthold 137 UMO 137). The samples collected were red sandstones (Figure 3).

2.2. Sample Preparation

Samples were dried at 105 °C and sieved using a Teflon sieve with a 0.63 mm mesh [22]. The 0.63 mm fraction was quartered and homogenized in a mill to a grain size below 0.15 mm. Each dry soil sample weighing 300 g was stored in a plastic container with a threaded cap fitting (Marinelli type) for 30 days, i.e., the period required to reach a perfect equilibrium between parent and daughter in the ²³⁸U and ²³²Th series. [23].

In order to evaluate the morphology and structure and to identify the uranium minerals, eight polished sections were prepared from the collected waste rocks.

2.3. Analytical Techniques

The soil samples were analyzed non-destructively using a gamma-ray spectrometry methodology with a high-purity germanium (HPGe) detector connected to digital multichannel analyzer (MCA) with 16,684 channels (ORTEC) [24,25]. The relative efficiency of the HPGe detector was 26%, and the energy resolution for ⁶⁰Co was 1.80 keV and was 0.800 keV for ⁵⁷Co. A cylindrical lead shield with a wall thickness of 10 cm protected the detector from background radiation. Each soil sample was counted for 70,000 s. The detection energy of the HPGe detector was calibrated using certified sources of ¹⁵²Eu, ¹³⁷Cs and ⁶⁰Co. The IAEA reference materials RGU_1, RGTh_1 and RGK_1 were used to calibrate the efficiency of the HPGe detector. The size and geometry of the reference samples were identical to those of the analyzed samples. Gamma energy lines used in the activity calculations in the 238U series were 352 and 295 keV for ²¹⁴Pb, 609 and 1120 keV for ²¹⁴Bi and 186 keV for ²²⁶Ra. The γ-lines of 338.4 and 911.2 keV for ²²⁸Ac, 727 keV for ²¹²Bi and 860 keV for ²⁰⁸Tl were used to calculate the activity for ²³²Th. ⁴⁰K activity was calculated based on its emission at 1460 keV [24,25].

Mineralogical analyses of the polished sections were performed using a tabletop Hitachi TM 3030SEM scanning electron microscope with energy dispersive spectroscopy (SEM-EDS) operating at an acceleration voltage of 15 kV. Elemental analysis (identification and quantification of elemental composition) was conducted using a QUANTAX 70 EDS system from Bruker [26].

2.4. Radiogical Parameters

In order to assess the radiological risk in the study area, the following radiological parameters were calculated: radium equivalent activity (Raeq), absorbed gamma dose rate (DR), external hazard index (Hex) and representative level index (RLI). The calculation of hazard indexes was based on the specific activities of uranium, thorium and potassium measured in the soil samples, and the equations used are presented in

Table 1. The calculation equations for radiological parameters.

Radiological Parameter	Equation	References
RRadium equivalent activity (Raeq)	Raeq = AU + 1.43ATh + 0.077AK (Bq/kg)	[27,28]
Absorbed gamma dose rate (DR)	DR = 0.462AU + 0.604ATh + 0.042AK (nGy/h)	[28,29]
External hazard index (Hex)	Hex = (AU/370) + (ATh/259) + (AK/4810)	[27,30]
Representative level index (RLI)	RLI = (1/150AU) + (1/100ATh) + (1/1500AK)	[30,31]

A_U, A_Th and A_K: specific activity of ²³⁸U, ²³²Th and ⁴⁰K (Bq/kg); recommended values: Ra_{eq =} 370 Bq/kg; D_R = 84 nGy/h; Hex > 1; and RLI > 1 [27].

.

2.5. Statistical Analysis and Mapping

Statistical analyses were performed using the SPSS software, version 17. All individual variables were estimated in terms of minimum, maximum, standard deviation (SD), skewness and kurtosis. The correlation analysis was based on Pearson’s coefficients and the spatial distribution maps for the natural radionuclides and radiological hazard indexes were built using the inverse-distance weighting (IDW) interpolation method, generated in ArcGIS software, version 10.2.

3. Results and Discussion

3.1. The Activity Concentrations of Naturally Occurring Radionuclides (²³⁸U, ²³²Th and ⁴⁰K) in Soil

The statistical parameters for the specific activity of ²³⁸U, ²³²Th and ⁴⁰K in the collected samples from the Lișava uranium mining sector are presented in

Table 2. Descriptive statistics for ²³⁸U, ²³²Th, ⁴⁰K, Ra_eq, D_R, H_ex and RLI in soil samples; Ra_eq stands for radium equivalent activity, D_R for absorbed gamma dose rate, H_ex for external hazard index and RLI for representative level index.

	N	238U (Bq/kg)	232Th (Bq/kg)	40K (Bq/kg)	Raeq (Bq/kg)	DR (nGy/h)	Hex	RLI
Minimum	47	32.120	7.230	497.320	118.331	57.558	0.320	0.877
Maximum	47	396.230	48.160	615.320	458.314	214.552	1.239	3.148
Mean	47	197.213	16.215	543.213	262.228	123.721	0.708	1.839
Std. Deviation	47	92.491	10.785	32.622	88.915	41.254	0.240	0.586
Skewness	47	0.119	1.913	0.569	0.329	0.318	0.329	0.343
Kurtosis	47	−0.639	2.595	−0.665	−0.700	−0.700	−0.700	−0.641

N—no. of samples.

. Appendix A shows the individual results of the gamma spectroscopy of the analyzed soil samples from the study area. The specific activity or activity concentration (activity per unit mass) is expressed in Bq/kg of radionuclides in soil, and the conversions recommended between Bq/kg and ppm are: 1 ppm U = 12.35 Bq/kg, 1 ppm Th = 4.06 Bq/kg and 1% K = 313 Bq/kg [10]. The histograms, Q–Q plots and box plots for these elements are presented in Figure 4. The frequency distribution charts were used to assess the normality distribution of the ²³⁸U, ²³²Th and ⁴⁰K data sets, and the patterns displayed were multimodal for ²³⁴U and ⁴⁰K and unimodal for ²³²Th. The fitting curves in the histograms show that the distribution was asymmetrical; they were skewed towards the right for all three elements, but the skewness was more significant for thorium compared to uranium and potassium (the distribution’s peak is off center toward the limit, and a tail stretches away from it to the right). The Q–Q plots confirmed the asymmetric distribution for ²³⁸U, ²³²Th and ⁴⁰K, with the actual data shown on the x-axis and the predicted values for a normal distribution shown on the y-axis. The concavity in the plots denote the lognormal distribution, with low and high values placed below the regression line. The box plots depict the average, minimum and maximum values of the data sets for ²³⁸U, ²³²Th and ⁴⁰K.

The specific activity of naturally occurring radionuclides varied significantly in the studied area due to the geological features and human activity (uranium mining practices). In all the soil samples, the specific activities were in the following order: ⁴⁰K > ²³⁸U > ²³²Th.

The activity concentration of ²³⁸U in all the soil samples showed the biggest variation, ranging from 32.12 to 396.23 Bq/kg with a mean concentration of 197.21 Bq/kg. The calculated statistical parameters (Table 2) and graphic representations for ²³⁸U (Figure 4) indicated a large variation in uranium in the soil and suggested a spatially inhomogeneous distribution. The standard deviation (SD) value for ²³⁸U was relatively high (92.49 Bq/kg) and indicated an asymmetric distribution of the ²³⁸U in the investigated surface soils.

The skewness had a positive value, which revealed a positive asymmetry to the right for uranium, whereas the kurtosis parameter had a negative value. The ²³⁸U histogram pointed to a frequency of the activity concentration class between 170 and 200 Bq/kg followed by those with concentrations slightly above 300 Bq/kg. The elevated ²³⁸U concentration and high variation in the soil reflected the long history of uranium mining practices in the area, such as the exploitation and exploration of uranium ore in the galleries, the dumping of mine tailings and the temporary storage of uranium ore before transport to a processing plant.

The activity concentration of ²³²Th in all the soil samples from the Lișava area ranged within relatively narrow limits from 7.23 to 48.16 Bq/kg with a mean concentration of 16.21 Bq/kg. The ²³²Th analyses displayed an asymmetric distribution, although the variations in the specific activities of ²³²Th were relatively low. The statistical parameters calculated for ²³²Th showed moderate variation. The box plot and normal Q–Q plot diagrams highlighted outlier values in the soil samples with 2, 5, 6 and 13 IDs. These sampling points were located at the extremities of the study area, where the anthropic influence of uranium mining was decreased and reflected the natural background of thorium content in the soil. The samples with higher ²³⁸U specific activities exhibited low ²³²Th specific activities, sometimes being below the average reported for the crustal abundance of 10.5 ppm [9].

The potassium specific activities varied from 497.32 to 615.32 Bq/kg as a result of the lithological features in the investigated perimeter with a mean specific activity of 543.21 Bq/kg. Like ²³⁸U, the descriptive statistics for ⁴⁰K (Table 2 and Figure 4) showed wide variations in terms of specific activity and an asymmetric distribution of content values.

The average specific activities of ²³⁸U, ²³²Th and ⁴⁰K found in the analyzed soils are presented in

Table 3. Average values of ²³⁸U, ²³²Th and ⁴⁰K in investigated area compared with data reported by similar studies.

Country	238U (Bq/kg)	232Th (Bq/kg)	40K (Bq/kg)	References
This study	197.21	16.21	543.21	-
Romania (Crucea)	559.7	91.52	-	[32]
Serbia (Gabrovnica)	249	69	960	[33]
China	79.3	101.0	535.8	[23]
Armenia	45.69	37.25	294.35	[1]
Nigeria	63.29	226.67	832.59	[34]
Egypt	137	82	1082	[31]
Median concentration values	35	30	400	[28]

alongside the data reported in similar studies. This data showed that thorium and potassium presented lower or comparable averages of specific activity compared with the reference data, while uranium showed a specific activity average that was much higher (197.21 Bq/kg). All the average values of ²³⁸U calculated in the present study were comparable to the values reported for soils around uranium waste deposits from the Crucea ore deposit in Romania (559.7 Bq/kg) and the averages found in the soils around a former uranium mine in Gabrovnica, Serbia (249 Bq/kg) [32,33]. This comparison showed that the natural radioactivity in the surface soil from the Lisava uranium mining sector was higher as a result of uranium mining.

3.2. Spatial Distribution of ²³⁸U, ²³²Th and ⁴⁰K in Soil from Lișava Uranium Mining Sector

A spatial distribution map of natural radionuclides is useful for identifying areas with high radionuclide concentrations and the possible sources of these elements. The spatial distribution maps for ²³⁸U, ²³²Th and ⁴⁰K are presented in Figure 5, Figure 6 and Figure 7.

Uranium was the radioactive element with the widest spatial variation in terms of concentration in the study area (Figure 5). The ²³⁸U map shows an apex of uranium specific activity (red areas on the map) in the northeastern part of the perimeter, in the soils developed on waste rocks and at the bottom of the waste dump slope. Furthermore, it shows that the area where uranium enrichment at the soil surface was higher than recommended values [28] was large. The spatial distribution analysis of ²³⁸U activity concentration noted an association between the high values of uranium content and the site of mining activities of uranium deposits (exploitation, transportation and temporary deposition). The increases in uranium concentration resulted from the leaching pf metal from the waste material. The observed lower values in the western part of the study area were spatially correlated with regions outside the uranium mining perimeter, where anthropic influences were decreased.

The spatial distribution of ²³²Th (Figure 6) in the soils from the Lișava mining sector indicated very low values (green zone on the map) in most of the study area, even/especially where many soils exhibited high uranium concentrations. Here, the specific activities of ²³²Th were under the recommended values [28]. Two zones with moderate activity concentrations (up to approximately 48 Bq/kg) were observed in the northeastern and northwestern parts of the perimeter (Figure 6), where the mining footprint was decreased and the relative thorium enrichment was correlated with uranium depletion. This correlation was closely related to the geological features of the study area.

Regarding the distribution map of ⁴⁰K (Figure 7), the highest values were reported in the northern part of the study area, with these displaying specific activity values up to 650 Bq/kg. The area with a high level of ⁴⁰K had approximately the same shape as the waste dump. The lower concentration zone (center and south of the perimeter) overlapped the lower content zone for thorium and was slightly higher in uranium content. Although all the values of the specific activity of ⁴⁰K reported exceeded the median concentration value (400 Bq/kg) provided by UNESCEAR [28], they reflected the local lithological settings.

3.3. Environmental Radiological Risk

The Lișava uranium mining sector is not an inhabited area; the closest village is about 5 km north. This area is included in a larger region where uranium mining has been the most prevalent activity for more than 40 years. The present study revealed that the soils from the Lișava sector had a relatively high concentration of ²³⁸U. Therefore, the radiological risk was assessed by calculating the radium equivalent activity (Ra_eq), absorbed gamma dose rate (DR), external hazard index (Hex) and representative level index (RLI) using the specific activities of ²³⁸U, ²³²Th and ⁴⁰K measured in the soil and comparing the calculated values to the median values provided by UNESCEAR. A descriptive statistic of the radiological parameters is presented in Table 1, and the individual values for each parameter are accessible in Appendix A.

The radium equivalent (Ra_eq) is a commonly used risk index and is calculated using the specific activities of ²³U, ²³²Th and ⁴⁰K. The radium equivalent (Ra_eq) ranged between 118.331 (Bq/kg) and 450.314 (Bq/kg) with a mean value of 262.228 (Bq/kg). Although the mean value was lower than the median value, in about 50% of the sampling points the calculated values exceeded 370 Bq/kg (Figure 8). The standard deviation (SD) of 88.91 Bq/kg indicated wide variations, suggesting a log distribution for Ra_eq.

The absorbed gamma dose rate (DR) had a minimum value of 57.55 (nGy/h) and a maximum value of 214. 552 (nGy/h). The mean of the absorbed gamma dose rate (DR) was 123.721 (nGy/h). Furthermore, at almost all the sampling points, the DR was assessed to be above the median value of 84 nGy/h provided by UNESCEAR

In Figure 9, the external hazard index (Hex) values calculated for all the sampling locations are presented. The values of Hex varied from 0.32 to 1.39 and had a mean value of 0.708. Except for the values calculated for sampling points 11 and 23, in all the other sampled points, the values were below 1.

The representative level index (RLI) of gamma radiation hazard associated with ²³⁸U, ²³²Th and ⁴⁰K from the soil varied from 0.877 to 3.148 with a mean value of 1.839. Almost all the calculated values were higher than the median value of one (Figure 9).

3.4. Pearson’s Correlation Analysis

We used Pearson’s correlation analysis to assess the similarity of the ²³⁸U, ²³²Th and ⁴⁰K specific activity concentrations and their related radiological parameters (Raeq, DR, Hex and RLI) in the soils from the Lișava uranium mining area. Pearson’s correlation coefficients classify the linear relationship between all variables by the confidence intervals of the coefficients as follows: weak (0.00–0.19), moderate (0.2–0.39), strong (0.4–0.79) and very strong (0.8–1.00) [35,36]. The Pearson’s correlation coefficients from this study are displayed in

Table 4. Pearson’s correlation coefficients matrix between ²³⁸U, ²³²Th and ⁴⁰K and Raeq, DR, Hex and RLI.

	238U	232Th	40K	Raeq	DR	Hex	RLI
238U	1
232Th	−0.350	1
40K	0.215	0.091	1
Raeq	0.986	−0.189	0.268	1
DR	0.119	−0.202	0.270	1.000	1
Hex	0.986	−0.189	0.267	1.000	1.000	1
RLI	0.567	−0.157	0.193	0.563	0.564	0.563	1

.

²³⁸U showed a very strong correlation with Raeq and Hex, a strong correlation with RLI and a moderate correlation with ⁴⁰K and DR. ²³²Th exhibited negative correlations with almost all the variables with the exception of ⁴⁰K. ⁴⁰K presented a weak to moderate correlation with all the detected variables. The very strong correlation between ²³⁸U and the radiological hazard indices suggested that uranium was the dominant contributor to the radiological hazard and risks assigned to gamma rays emitted from the ²³⁸U decay series in the soil. In addition, the negative correlation between ²³⁸U and ²³²Th and the weak correlation with ⁴⁰K reflected the impact of uranium mining on the increased radiation levels and radiological risk in the study area.

The negative correlation between uranium and thorium implied a different origin of the two elements. Thus, the main sources of uranium were accessory minerals enriched in U but not Th. Part of the observed U values were due to the leaching of U-bearing phases that incorporated large amount of uranium, whereas the rest was attributed to contamination resulting from uranium mining activities [37]. The positive correlation between potassium, uranium and thorium, although moderate and weak, revealed that both the radionuclides were derived from natural sources. Potassium is a common element in sedimentary rocks with an increased content in K-felspar, mica and clay minerals.

3.5. Mineralogical Futures

In the Lișava sector, the uranium deposits are of a tabular sandstone type, and the mineralization is described as consisting of two generations of pitchblende, sooty pitchblende and uraniferous anthraxolite associated with minor, galena, chalcopyrite, pyrite and sphalerite, which are frequently related with bitumen [19,20].

In the oxidized zone, torbernite, autunite and uranium black products appear [19]. The organic carbon content in this zone can be up to seven percent. Pitchblende occurs as finely dispersed inclusions mainly in solid bitumen and carbonaceous matter, in fine- and medium-grained sandstones, siltstones and carbonaceous and bituminous shales. Locally, pitchblende is found in the gritstone and sandstone matrix. Rarely, uranium black products and pitchblende, which are located near bitumen, form bands or small layers and strings that fill cracks along fissures [19,20].

The mineralogical analysis of the rock samples (uraniferous red sandstone) collected from the dump in the vicinity of soil sampling points 11 and 23, where the uranium concentration was highest, reflected some of the mineralogical characteristics of the uranium ore (Figure 10). Uranium-bearing minerals were found to be less associated with coaly matter than expected. Instead, the carbonaceous matter contained frequent sulfide grains, sometimes displaying biogenic structures such as framboidal aggregates or mineralized vegetal cell structures. The most frequent sulfides were sphalerite, galena, pyrite, and subordinate chalcopyrite, which is sometimes associated with barite, which also appears in pore-filling aggregates. Uranium minerals were mostly observed in discontinuous pore-filling aggregates associated with rutile and possibly anatase. The mineral phases identified in the available samples were brannerite, coffinite and pitchblende, usually forming crusts on rutile or cementing it. As a rule, the U-minerals found were fine-grained and involved in intergrowth with other phases, typically with rutile.

4. Conclusions

The specific activities of the naturally radioactive elements ²³⁸U, ²³²Th and ⁴⁰K were measured in 47 soil samples collected from the Lișava uranium mining sector. According to the data obtained, these varied significantly from 32.12 to 396.23 Bq/kg for ²³⁸U and from 497.32 Bq/kg to 615.32 Bq/kg for ⁴⁰K, whereas the specific activity of ²³²Th varied insignificantly from 7.23 to 48.16 Bq/kg. In this study, ²³²Th and ⁴⁰K presented averages lower than or comparable with the worldwide average, while the ²³²U average was higher than the median value (35 Bq/kg). The radiological hazard indexes ranged considerably for Raeq (from 118.33 to 450.31 (Bq/kg)), DR (from 57.55 to 214.55 (nGy/h)), Hex (from 0.32 to 1.239) and RIL (from 0.877 to 3.148). The distribution of the risk assessment indices associated with the investigated soils exceeded the median values given by UNSCEAR. The investigated area had a moderate to elevated level of natural radioactivity. The very strong correlation between ²³⁸U and the radiological hazard indices suggested that uranium was the dominant contributor to the radiological hazards and risks assigned to the gamma rays emitted from the ²³⁸U decay series in the soil. The uranium mineral phases identified in the waste rock samples were brannerite, coffinite and pitchblende, usually forming crusts on rutile or cementing it.

All the data indicated that the investigated area had a moderate to high level of natural radioactivity, and the high concentration of uranium in the soils reflected the richness of uranium in the parent rocks from which the soil was derived as well as uranium enrichment due to previous uranium mining activities.

A possible solution to the high level of radioactivity observed is the rehabilitation of the waste dump and the surrounding land that is exhibiting signs of degradation. Any work that will be undertaken needs to stabilize the uranium waste and reduce the contamination of surrounding areas. Furthermore, further investigations should be carried out to establish the level of contamination in the surface and ground waters in order to clarify any issues related to uranium mobility.