Assessment of Natural Radioactivity in Rivers Sediment and Soil from the Copper Belt Artisanal Mining Region, Democratic Republic of the Congo

The concentrations and distribution of naturally occurring radionuclides (U, U, U, Th, Ra, Pb, Th, Ra, and K) were determined by alpha and gamma spectrometry in soil and sediments collected from Luilu and Dilala rivers located in the mining district of Kolwezi (Lualaba Province) of the Democratic Republic of the Congo. The average concentrations of U and Ra in the analyzed samples were 5 10 times higher than the world average values for soil provided by the United Nations Scientific Committee on the Effects of Atomic Radiation. However, the average concentrations of Th were found similar to the world average. In both river basins, artisanal mining activities and mineral washing sites displayed the higher concentration values of radionuclides. The mean values of health risk indices calculated for those sites were found significantly higher compared to world average levels. Radiation protection measures seem needed to ensure the radiation safety of local populations. Graphical Abstract How to cite this paper: Atibu, E. K., Oliveira, J. M., Malta, M., Santos, M., Mulaji, C. K., Mpiana, P. T., & Carvalho, F. P. (2021). Assessment of Natural Radioactivity in Rivers Sediment and Soil from the Copper Belt Artisanal Mining Region, Democratic Republic of the Congo. Journal of Geoscience and Environment Protection, 9, 1-20. https://doi.org/10.4236/gep.2021.97001 Received: April 26, 2021 Accepted: July 12, 2021 Published: July 15, 2021 Copyright © 2021 by author(s) and Scientific Research Publishing Inc. This work is licensed under the Creative Commons Attribution International License (CC BY 4.0). http://creativecommons.org/licenses/by/4.0/ Open Access


Introduction
Rocks and soil (the lithosphere) generally are the main reservoirs of primordial radioactive elements and sources to their transfer to other environmental compartments, such as the hydrosphere and biosphere (Eisenbud & Gesell, 1997;Rahman & Faheem, 2008). Human activities, such as industrial and artisanal mining, can enhance the concentration of naturally occurring radioelements in soil, particularly through the release of radionuclides with mining effluents, and thus potentially increasing the radiological hazards to the population. It is well known that the enhancement of environmental radioactivity levels may occur in mining and milling of radioactive ores for uranium production  but may occur also in the exploitation of non-radioactive metals like iron, gold, copper, niobium, cobalt, and in the operation of quarries and sand pits (Michalik, 2017).
The former province of Katanga in the Democratic Republic of the Congo (DRC) is well known for its immense copper and cobalt deposits. In the East of the DRC, especially in the copper belt region (ex-Katanga province), mining activities were started during the colonial period by the "Union Minière du Katanga" owned by Belgium (Malaisse et al., 1994;Atibu et al., 2018;Atibu et al., 2016;Atibu et al., 2013). After the independence of the DRC in 1960, the national company "Gécamines" concentrated its mining activities in this province.
Nowadays, many national and international companies implement mining projects in this region, alongside with artisanal mining activities (Atibu et al., 2018;Atibu et al., 2016;Atibu et al., 2013). The ex-Katanga region is known also for

Description of the Study Area
The district of Kolwezi, where the current study was performed, is located in the province of Lualaba (ex-Katanga region), at the southeastern side of the DRC, in Central Africa (Figure 1). The province of Lualaba is known as the richest region of DRC because of the large number of industrial and artisanal mining activities that significantly increased the national income over the last decades. The Kolwezi district, with an area of 213 km 2 and about 453,000 inhabitants is located on the Manika plateau at the average altitude of 1500 m above the sea level. The stratigraphy and geology of this region, known as "Série des Mines" was investigated and described in detail by geologists (Grujensch, 1978). In the copper belt, there are sedimentary layers of clay-talc, and terrigenous layers with occasional elements of magmatic origin, and shales in which metal deposits occur,. Kolwezi holds today the largest and most important cobalt and copper mining centers of the DRC. Uranium deposits do exist in the region and uranium mining was also carried out in the ex-Katanga region up to 1960. The history of mining activities, together with the geological aspects of the district of Kolwezi, were previously described in several reports (Atibu et al., 2018;Atibu et al., 2016;Mees et al., 2013;Cheyns et al., 2014).
The co-occurrence of uranium with non-radioactive metals has been noticed

Collection of Environmental Samples
Soil and sediment sampling was done in August 2017 and the analytical work in the laboratory was carried out in 2019. Sample identification, GPS coordinates of sampling sites, and human activities carried out at each area are reported in Table 1. Samples were collected from two river basins: 1) the Luilu river that flows through the mining and milling city of "Gécamines-Kapata" receiving effluents from such activities before merging into the Mpingiri river; 2) the Dilala River that separates the Kasulo city (on the left river bank) and the industrial zone, with a brewery, offices and truck garages and workshops for several mining companies located in the town of Manika (on the right river bank) (Figure 1). were collected with a shovel on riverbanks. The area represents agricultural land irrigated with river water and used by population for agriculture and as a ground for human settlements. The river sediments surface layer of (0 -5 cm depth) were manually collected from the riverbed at a distance of about 1 -2 m from the shore and at 10 -30 cm water depth. The population is in daily close contact with sediments and soil because of their mining activities and the use of river sediments (clay) in house construction. Therefore, for the purpose of radiological hazard assessment the soil and sediments were considered together as they are main sources of ambient radiation ( Figure 2).

Sample Pretreatment
Soil and sediment samples were oven dried at 80˚C -100˚C during 24 h, disaggregated and sieved through a 63 μm stainless steel sieve to remove coarse sand, stones, pebbles and other macro-impurities that with their large volume and mass and low radionuclide content would modify radionuclide activity concentrations (Bq·kg −1 ) if not eliminated.
Less than 63 μm grain-size fraction, the largest fraction of the sample (>4/5) and containing most radioactivity, was used for radionuclide analysis in order to allow the comparison between samples Carvalho, 1995).

Sample Analysis by Gamma-Ray Spectrometry
After sample homogenization, aliquots of the less than 63 μm grain-size fraction from soil and sediment samples were placed in Petri dishes and compacted to completely fill the container (no head space was left), and hermetically sealed using cellophane tape. Once in the containers, the samples were kept aside for about one month in order to ensure the formation of secular radioactive equilibrium between 226 Ra and its radioactive decay products 214 Pb and 214 Bi. Afterwards, they were analyzed by gamma-ray spectrometry for the determination of

Sample Analysis by Alpha Spectrometry
Aliquots of the homogenized < 63 μm fraction of samples were used for the determination of 238 U, 235 U, 234 U, 232 Th, 230 Th, and 210 Pb. Among these radionuclides, 210 Pb was calculated through the determination of the alpha emitter 210 Po in secular radioactive equilibrium with 210 Pb, ensured with the sample storage for about 2 years. These radionuclides were determined in all samples applying a sequential radiochemical separation of radioelements, followed by their electrodeposition on metal planchets. Thereafter, radiation measurement was carried out by alpha spectrometry using an OCTETEplus spectrometer from ORTEC, with 450 mm 2 active surface ULTRA AS Mark detectors. The Maestro software from ORTEC was used in spectrum analysis. The procedures followed, including analytical quality control, are described in detail and validated in published reports Oliveira & Carvalho, 2006;Carvalho & Oliveira, 2009). Quality control was ensured through periodic participation in IAEA intercomparison exercises with consistent good results over the years, as referred previously (Povinec et al., 2007;Pham et al., 2014;Pham et al., 2016). Results for radionuclide activity concentration are expressed in Bq·kg −1 (dry weight), with the expanded uncertainty (k = 2).

Evaluation of Radiological Hazard and Health Effects
To estimate the radiation risks for human population from exposure to radionuclides in soil and sediments, several radiological indices were calculated such as, the Radium Equivalent Activity Index (Ra Eq ), the Outdoor Gamma Absorbed Dose Rate (ODRA), the Annual Effective Dose Equivalent (AEDE outdoor ), the Excess Lifetime Cancer Risk (ELCR) and Hazard indices (H ex and H in ). These indices are defined as follows.

Radium Equivalent Activity Index (RaEq)
Defined by Beretka and Mathew (1985), the Radium Equivalent Activity Index (Ra Eq ) allows a single number to describe the gamma output and the radiation hazards associated with different mixtures of 238 U, 232 Th and 40 K in samples. It was estimated using a well-established relation (Taqi et al., 2018; UNSCEAR,

Annual Effective Dose Equivalent (AEDE)
The Annual Effective Dose Equivalent (AEDE) received by a person living in the area, was calculated using the following equation: where DCF is the dose conversion factor (0.7 Sv·Gy −1 ). The DCF is used to convert the absorbed dose rate to the human effective dose equivalent with an outdoor occupancy factor (OF) of 20% (UNSCEAR, 1993) and T is the exposure time (8760 h·y −1 ) (UNSCEAR, 2000). So the AEDE can be calculated using the following equation: The world average value of AEDE is 0.07 mSv·y −1 (UNSCEAR, 2000; UNSCEAR, 2010).

Excess Lifetime Cancer Risk (ELCR)
The Excess Lifetime Cancer Risk (ELCR) estimates the probability of cancer incidence in members of a human population for a specific lifetime, due to expo- where DL is the duration of life (assumed as 70 years), and RF is the risk factor of contracting a fatal cancer per sievert (Sv −1 ) received. For stochastic effects, the ICRP 60 recommends the use of RF = 0.05 for the members of the public (ICRP, 1991;Taskin et al., 2009). The world average value of ELCR is 0.00029 (UNSCEAR, 2000; UNSCEAR, 2010).

Hazard Indices
To make an assessment of the health effects from the internal and external exposure to radioactivity from earth's surface materials containing 238 U, 232 Th and 40 K, Beretka and Mathew (1985) defined two indices which objective is to compare the radiation dose received to a dose equivalent limit of 1 mSv·y −1 recommended for members of the public (ICRP, 1991 Th and 40 K respectively (Taqi et al., 2018;Agbalagba & Onoja, 2011). The H in value must not exceed 1 to be accepted as negligible.

Data Analysis
Statistical treatment of data, using Spearman's rank order correlation, was made using SigmaStat 12.5 (Systat Software, Inc., USA).

Radionuclide Concentrations in Samples
The activity concentrations of primordial radionuclides (     Sediments from the Luilu river basin showed an increasing trend in concentrations of uranium ( 238 U) from Lsed7 to Lsed1, i.e., going downstream, indicating uranium rich geologic deposits in the lower catchment, while for thorium ( 232 Th) no concentration trend was observed. With 40 K, when compared with uranium, the opposite trend was observed, i.e., higher concentrations in the upper catchment of the river and decreasing downstream, thus suggesting that rocks in the upper zone of the river basin are a different geologic formation, probably potassium rich magmatic rocks (Grujensch, 1978). In the Dilala river sediments, there were no regular trends in uranium series along the river path, but sediments at station Dsed4 showed a clear occurrence of elevated uranium concentrations, while 232 Th and 40 K concentrations showed no geographic trend. Agricultural soil samples from the Luilu river basin displayed slightly lower uranium and uranium daughter concentrations, but similar concentrations for other radionuclides when compared to river sediments ( Table 2).
The wide difference of average radionuclide concentrations between the two river basins can be explained by the different geology of the basins, and likely also by physical, chemical and geo-chemical parameters acting at specific sampling sites (Issa et al., 2013;Ramasamy et al., 2011). It is noteworthy that sediments from Luilu River basin globally displayed much lower concentrations of 238 U and 234 U isotopes (uranium series) than those from the Dilala River basin, although both have similar concentrations of 232 Th (thorium series) ( Table 2). The Dilala River basin is therefore richer in uranium minerals and, actually, the 238 U concentration at Dsed4, 973 ± 29 Bq·kg −1 , as well as the accompanying 234 U concentration, are much higher than in all other samples. The 238 U activity concentration at Dsed4 corresponds to a mass concentration of uranium of 79 ± 2.4 mg·kg −1 , much higher than the average uranium concentration of 3 mg·kg −1 in the earth crust, and similar to concentrations commonly found in uranium bearing regions only (Choppin et al., 1995). Interestingly, for 40 K the samples from river basins show the opposite ranking when compared to uranium, with the terrains of Luilu River much richer in potassium than those from Dilala River (Table 2), thus reinforcing the idea of river basins with different geological settings. Figure 4 highlights these differences between river basins through the graphic plot of radionuclide ratios using 232 Th for normalization: while in Luilu basin high 40 K concentrations are associated with high 238 U concentrations, in the Dilala basin high 40 K is associated with low 238 U concentrations (Table 3).
Compared with the world average values given by the UNSCEAR (2000), the average concentrations of 238 U and 226 Ra in sediments samples from Luilu and Dilala rivers are 5 -10 times higher than worldwide averages, while the average 232 Th in sediments from both rivers are similar to the world 232 Th average value (Table 2). However, the sampling sites Dsed4 and Lsed1 where artisanal mining and mineral washing is performed, displayed very high concentration values for 238 U, 235 U, 234 U, 230 Th, 210 Pb and 226 Ra respectively. This suggests that there is natural occurrence of hot spots but also artisanal mining activities can enhance the activity concentration of radionuclides in soil and sediments.   Establishing a criterion for deciding what the natural radioactive background is, and what are hot spots (natural high concentration values) and areas with enhanced (modified) radioactive background is not of easy consensus when one deals with naturally occurring radionuclides. However, based on the present data set it is suggested that sites of natural radioactive background in this region are those with 40 K/ 232 Th < 15 and 238 U/ 232 Th < 10 ( Figure 4). The outliers, i.e., sampling points above typical background concentrations are thus one point in Luilu basin, Lsed3, and two points in the Dilala basin, Dsed4 and Dsed5 (Table 3) and these were also the outliers in probability data plots (not shown).
Radionuclide ratios calculated for several radionuclide pairs are shown in Table 3. For the purpose of this discussion, the results for primordial radionuclides in soil were grouped and averaged with sediments as soil samples of the Luilu river basin were similar to radionuclide concentrations in sediments, In Table 3, the activity concentration ratios of uranium daughters with the parent radionuclide, i.e., 234 U/ 238 U, 230 Th/ 238 U, 226 Ra/ 238 U, 210 Pb/ 238 U, in sediment samples give information about radioelement behavior in the aquatic environment. The 234 U/ 238 U ratios in sediments were around 1 which indicate near secular radioactive equilibrium. On the other hand, the values of other radionuclide ratios, although fluctuating, were consistently lower than 1 indicating dissolution and removal by river water. This has been described for other rivers. For example, Vigier et al. (2001) reported that radionuclide ratios in sediments from Canadian rivers were found in the ranges of 0.897 -0.992 for 234 U/ 238 U, and 0.653 -1.002 for 226 Ra/ 238 U, indicating 234 U and 226 Ra removal from sediments. This is indicative to the departure from secular radioactive equilibrium due to dissolution/erosion along the river path. In the same report, concentration ratios that included 230 Th were even further below unity (0.653 -1.002 for 230 Th/ 238 U and 0.596 -1.020 for 226 Ra/ 230 Th), which was found now in Congolese rivers as well. Such deviations of concentration ratio values from unity relate to the different solubility of radionuclides. In this case, 238 U, especially 234 U (parent of 230 Th) and 226 Ra, that were much more soluble when compared with thorium ( 230 Th), as pointed out by those authors.
In the case of the two Congolese rivers, the radionuclide activity concentration ratios in sediments along the river path also showed that uranium from sediments was gradually dissolved while thorium (nearly insoluble) was retained in the solid phase. But this ratio was occasionally reversed probably due to uranium precipitation. It is likely that geochemical conditions change along the path of these two rivers, with episodes of dissolution followed by reprecipitation of radionuclides, probably in relationship with discharges from mining companies and stirring of sediments by artisanal mining. In such conditions, there is no consistent trend in activity ratios along the river path and the exposure to radiation may vary from one location to the next.
The specific activities of 238 U, 232 Th, 40 K and 226 Ra determined in this study are compared in Table 4 with results reported in the literature. This comparison highlights that average concentrations in both river basins in the Congo were   higher than values for other regions. Moreover, maximum values measured in sediments of Dilala river are similar to concentrations typically found only in uranium rich areas and in uranium mining waste.

Evaluation of Radiological Hazards and Health Effect Indices
The radiological hazard and health effect indices were calculated for all sampling stations and results are shown in  (UNSCEAR, 2000). The ODRA values were ranged between 50 (Lsed5) and 472.6 nGy·h −1 (Dsed4) with an average value of 121.1 nGy·h −1 , which was two times higher than the world average value of 59 nGy·h −1 (UNSCEAR, 2000). The calculated values of AEDE varied from 0.06 (Lsed5) to 0.58 mSv·y −1 (Dsed4). The average level (0.2 mSv·y −1 ) of AEDE values was 2.9 times higher than the world average of 0.07 mSv·y −1 (UNSCEAR, 2000;UNSCEAR, 2010). The cancer risk parameter (ELCR) values ranged from 0.21 (Lsed5) to 2.03 (Dsed4), and the average value of 0.6 for all sampling sites was three orders of magnitude higher than the world mean level of 0.00029 (UNSCEAR, 2000;UNSCEAR, 2010 (Table 1). Other sites, such as Lsed3, were occasionally high for some parameters but not consistently to all or at least the majority of them. Therefore, radiological hazards and the risk of health effects on the riverine population were consistently higher in the Dilala river basin in comparison with Luilu river. Although this is primarily due to the natural occurrence of uranium and uranium daughters, the artisanal mining activities at these sites increased E. K. Atibu et al.
the exposure to radiation and contributed to the dispersion of radionuclides in the river system.

Correlation between Parameters
In order to determine the relationships and strength of association between pairs of variables, the Spearman rank order correlation was calculated. The results are shown in Table 6.  A positive correlation, significant at p < 0.05, was observed between 238 U and 226 Ra, which is not surprising because both radionuclides are members of the same radioactive decay series (uranium series) and thus originate from the same geological sources.
The negative correlation between 238 U and 232 Th, between 238 U and 40 K, and between 226 Ra and 40 K can be explained by the fact that radionuclides of each pair are not from the same radioactive decay series and do not originate from the same geological sources. Furthermore, they are likely to be transferred through different pathways in the environment, i.e., the negative correlation suggests that they are not associated in their environmental cycling.
Positive and statistically significant correlation coefficients (p < 0.05) were observed also between 238 U and 226 Ra concentrations with all radiological indices (Ra Eq , ODRA, AEDE, ELCR, H ex , H in ). These correlations follow the very strong contribution from these two radionuclides to the radiological parameter indices.
Indeed, it is clear from the data that these two radionuclides are the main contributors to the gamma dose at all sampling sites. Despite its contribution to the external radiation dose, 40 K displayed negative correlation with the radiological hazard parameters. The radioactive potassium is not considered a radiological threat through ingestion due to the homeostatic control of potassium concentration in the organism.

Conclusion and Research Outlook
The activity concentrations and environmental distribution of a selection of key natural radionuclides ( 238 U, 235 U, 234 U, 232 Th, 230 Th, 228 Ra, 226 Ra, 210 Pb, and 40 K) were determined using gamma-ray and alpha-particle spectrometry for the analysis of soil and sediment samples from Luilu and Dilala river basins, at the Kolwezi district in the Democratic Republic of the Congo. In general, the concentrations of the four primordial radionuclides 238 U, 232 Th, 226 Ra and 40 K were within the worldwide average concentrations reported by UNSCEAR. Furthermore, the highest concentration values of these radionuclides, well above the average natural background, were determined at locations impacted by artisanal mining activities, confirming the co-occurrence of uranium with non-radioactive metals targeted by the miners. The results from this assessment of radiological hazards and health risks confirmed that the artisanal mining activity in this region, although targeting E. K. Atibu et al. Journal of Geoscience and Environment Protection non-nuclear minerals, enhanced radiation exposure and radiological hazards to riverine populations. As the population engaged in artisanal mining is much wider in the country, a full assessment of radiation exposure should be carried out in other provinces also, and the results should be used to design a suitable radiation protection policy. This study is the first in this field for the DRC and it is a starting point for use in future monitoring and investigations on environmental radioactivity aiming to enhance the radiation protection of population.

Statement on Consent
Photos were taken with permission and their use was consented by people portrayed.