Characteristics of Natural Background Radiation in the Lubin Mine, Poland

Abstract

There has been growing interest in using underground locations for applications in various fields, including research. In Poland, for several years, attempts have been made to build an underground laboratory. For this purpose, selecting an appropriate location requires a detailed analysis of the level of natural radioactivity. The present study presents detailed characteristics of the natural background radiation in close vicinity to shaft L-VI of the Lubin mine, at the depth of 910 m (2275 m w.e.). The in situ measurement of the photon flux in the 7–3150 keV energy range was equal to 8.08 ± 0.90 cm⁻²s⁻¹, and the gamma-ray dose rate of 0.070 ± 0.010 µSv/h with the highest contribution from ⁴⁰K and ²¹⁴B isotopes. The thermal neutron flux measured using helium counters was equal to 4.2 ± 0.9 × 10⁻⁶ cm⁻²s⁻¹. The radon concentration in the air measured with the RAD7 monitor showed low values ranging from 0 to 15.3 Bq/m³. Laboratory measurements of rocks using alpha and gamma spectrometry techniques showed a significant variation in the concentration of ²²⁶Ra and ^234,238U isotopes, and the highest concentration values were recorded for shales. The ⁴⁰K, ^234,238U and ²²⁶Ra isotopes make the greatest contribution to the natural radioactivity of analyzed rocks.

1. Introduction

There are several dozen operating underground laboratories (ULs) worldwide, with various applications. They conduct basic research in particle physics, astrophysics, and nuclear physics but also perform technical and biomedical measurements that require a low background from cosmic rays and natural radioactivity. Such conditions are impossible to obtain on the surface of the Earth. Therefore, deep underground environmental conditions create new, unique opportunities for various types of research and measurements, interdisciplinary. There are also underground laboratories that deal with the final disposal of spent nuclear fuel. One of them is Äspö Hard Rock Laboratory, located in Sweden. They investigate, inter alia, the interaction of bentonite and copper canisters with rock under realistic conditions, as well as determine the role of the rock as a barrier [1]. In Europe, the leading laboratories are Gran Sasso, located in Italy, Modane in France, Canfranc in Spain and Boulby in Great Britain.

There is currently no operating underground laboratory in Poland, although efforts to create it began in 2010. At that time, the Polkowice-Sieroszowice mine, owned by KGHM Polska Miedź S.A., was one of the candidates for an underground laboratory to participate in the European project LAGUNA FP7 with a liquid argon detector for physics research [2,3,4]. The natural radioactivity measurements carried out at that time in the salt layer showed an extremely low level of natural radioactivity (lower than in the leading European underground laboratories). However, the idea was rejected at the time due to the strictly mining policy of KGHM’s mines. Currently, intensive work is being carried out in Poland to find a suitable location to create an underground research laboratory characterized by a low level of natural radioactivity. Several scientific and state institutions are involved in this work. One of the considered locations is the Lubin mine, the oldest KGHM Mining Division in the Lower Silesian Copper Basin (LSCB). A large part of the mine workings is out of use, but chosen workings are still maintained for ventilation and transport reasons. The construction of an underground laboratory in these locations could give these parts of the mine a second ‘life’ and better utilization. For this purpose, preliminary measurements of natural background radiation were required to determine if the underground workings of the Lubin mine are suitable for performing chosen non-mining activities. Therefore, a detailed analysis of natural radioactivity was carried out in close vicinity to shaft L-VI of the Lubin mine, at a depth of 910 m (about 2275 m w.e.).

Radionuclides are widespread in the environment in various concentrations. Natural radioactivity can be divided into three groups: long-lived radionuclides, radionuclides present in decay chains, and cosmogenic radionuclides. Most naturally occurring radioactive elements are grouped into three decay chains, and ²³²Th, ²³⁸U, and ²³⁵U isotopes are the protoplasts of the natural radioactive series. However, due to the time of half-life, the ²³⁸U and ²³²Th decay series are of major importance. Cosmic radiation is not only a source of ionizing radiation but also produces many radioactive elements in the atmosphere and on the Earth. Cosmogenic isotopes are formed as a result of the reaction of high-energy particles with the components of the atmosphere (nitrogen, oxygen, and argon) and the dust particles present in it and with the surface layers of the lithosphere [5]. From the point of view of irradiation of living organisms, the most important are: ³H, ⁷Be, ¹⁴C and ²²Na. Elements belonging to the radioactive series of ²³⁸U uranium, thorium and ⁴⁰K isotope are the primary sources of natural ionizing radiation

The aim of the present study was to provide detailed characteristics of the natural radioactivity of the studied location. The research includes in situ measurements with the use of gamma spectrometry, the radon concentration in the air, and thermal neutron flux measurements. The laboratory alpha spectrometry measurements of ^234,238U concentrations in rock and water samples collected in the shaft were performed. Moreover, radium ^226,228Ra and potassium ⁴⁰K concentrations in rock samples were determined using the gamma spectrometry technique. Based on estimated concentrations of ^226,228Ra and potassium ⁴⁰K in analyzed rock samples, potential radiological risk indices have been calculated.

1.1. Description of Deposit

Permian deposits from the described area were deposited in the SE part of the South Permian Basin, the so-called Polish Permian Basin. The basin, developed due to rifting between Late Carboniferous and Early Permian, extends from Great Britain to Poland. It was filled with Lower and Upper Rotliegend strata deposited mostly as clastic and volcanic (in the lower part) rocks [6]. In the top part of the strata, the presence of paleo-dunes has been confirmed [7,8,9]. Those paleomorphological undulations were an important factor influencing the lateral and vertical distribution of younger sediments [7,10]. In the Zechstein times (Late Permian), transgression of the waters from the Boreal Ocean flooded the area of intracontinental depression [11,12]. That resulted in the remobilization of the aeolian sand. The oversaturated sand displacement, from tops to depressions between dunes, reduced the elevations inherited from the previous landscape. In the topmost parts of the white sandstones, sedimentological structures and fauna clearly indicate underwater deposition [13]. The detrital materials of the white sandstones (Weissliegend) are mainly grains of quartz and, less often, feldspars, micas, and igneous and metamorphic rocks fragments [14]. Cements are mainly composed of illite, carbonates, sulfides, and quartz (less often), close to the top of sandstone elevations sulfates [14]. During the next phase, nearshore, calcareous deposits were settled (now several dozen centimeters-thick so-called “basal limestone”). On the regional scale, in the depressions between paleodunes, fine grained muds were deposited. At the same time, on the topmost parts of paleoelevations, small or no deposition of mud took place. That resulted in the elongated (NNW-SSE) areas without Coppershale (Figure 1). Coppershale thickness varies from 0 to 1.7 m [15]. Shales are composed of illite, glauconite, carbonates, quartz, and phengite. Phosphates and feldspars were also identified [14]. During the deposition of the mud, below the wave base, the euxinia existed, which made possible the preservation of up to 40 wt.% of organic matter [16]. Subsequently, muds were gradually replaced by carbonate deposition of Zechstein limestones. The Zechstein Limestones (Ca1) are composed of carbonates with varying dolomite–calcite content. With an upward increase in the volume of carbonates, a decrease in the content of organic matter is observed. In the area of copper mines, three characteristic carbonate facies can be distinguished regionally: carbonate platform, slope and basin (Figure 1). On the platform, more than 100 m of carbonates thickness was revealed; then, on the slope, thickness gradually dropped and reached a few meters in the basinal part.

The main mineralized, and thus mined, lithologies in the KGHM Polska Miedź S.A. mines are described above the lowest part of the Zechstein sediments: white sandstones, shales, and dolomites. On the carbonates, anhydrite layers interbedded with salt were deposited. The Lower Anhydrite (A1a) thickness changes from a few meters to more than 100 m and can vary significantly over short distances. A good negative correlation between the Lower Anhydrite thickness and the Oldest Halite can be observed. The thickness of salts varies from 0 to more than 180 m. Salt layers do not completely cover the mining areas. The present extent of salt layers crosscut the mining area approx. from E to W, and north to the Zechstein Limestone platform facies (Figure 1). Upper anhydrite (A1b), when the salt layer is absent, can be distinguished from Lower Anhydrite thanks to the anhydrite breccia layer (Br1). Upper anhydrite has an average thickness of around 50 m and closes the first evaporite cyclothem–Werra (PZ1).

The periodically isolated Zechstein Sea and dry climate conditions resulted, in the described area, in the development of three classical evaporitic deposition cycles: PZ1–Werra, PZ2–Stassfurt, and PZ3–Leine. The youngest period with more continental influence was deposited under more humid conditions (PZ4–Aller) [18]. Zechstein sediments were covered with thick Mesozoic and Cenozoic deposits.

In general, Permian and Mesozoic sediments are slightly tilted to the NE and NNE. The homocline was formed at the turn of the Mesozoic and Cenozoic. Therefore, especially southern parts of the copper mines, close to the Fore-Sudetic Block, are characterized by dense faulting. Structures created by older tectonic events were also recognized [19,20].

1.2. Site Geology

The Lubin mine is located in the Lower Silesia voivodeship north of the borders of Lubin city. The deposit’s industrial resources amount to over 328 million tonnes of copper ore. The copper content in the ore is 1.28% on average, and the silver content is 54.5 g/Mg. The dominant lithological type of copper ore is sandstone, which accounts for nearly 70% of the resources. The depth of copper ore in the Lubin deposit ranges from 368 to 1006 m. The deposit is relatively shallow, just below loose Cenozoic sediments. In regions where discontinuities such as faults occur, exploitation is very difficult and sometimes simply impossible. The height of workings in the Lubin mine is characterized by variability in the range from 1.5 m up to 5.5 m. The average height of the workings is about 2.8 m. Within the mining level, three types of rocks occur. In the central part of the excavation level, Coppershales are present. These shales are bitumen-carbonate shales containing minerals of copper and other metals, e.g., silver, lead, nickel, cobalt and others. Above the Coppershales, there is a thick layer of carbonate rocks, mainly in the form of dolomites. These rocks are characterized by high strength. In turn, below the Coppershales strata, a thick layer of sandstone is located. The stratigraphic profile within the analyzed area is presented in Figure 2.

2. Materials and Methods

Measurements of the natural radioactivity were carried out at a distance of about 400 m from the L-VI ventilation shaft, at a depth of 910 m (2275 m w.e.) at the level of copper ore deposit excavation. The location of the measuring site, dimensions of the underground excavation and parameters of air in the area of measurements are presented in Figure 3. All measurements were performed with the same equipment and similarly as was done at the Polkowice-Sieroszowice mine in the anhydrite layer and at the GIG Experimental Mine ‘Barbara’ in sandstone rocks and described in detail in previous works [21,22]. The detectors were installed side by side (Figure 4) in the excavation near the Rdg-3/Rd-17.2 switching station chamber.

The height of the chamber was 2.92 m, while the width was 5.80 m, and at the bottom it was 4.77 m. In situ gamma radiation measurements were carried out using a semiconductor portable gamma spectrometer with the high purity germanium detector (HPGe); radon concentration measurements in the air using the RAD7 radon monitor; and the neutron flux was measured with helium counters. To carry out laboratory analysis (at the A. Chełkowski Institute of Physics, the University of Silesia in Katowice, Poland), six rock samples and one water sample were collected from the site where the in situ measurements were taken. The rock samples for laboratory analysis consisted of two samples of sandstone, dolomite, and shale. The concentration of uranium ^234,238U radioisotopes in rock and water samples was measured by alpha spectrometry, while the content of radium ^226,228Ra and potassium ⁴⁰K in rock samples by gamma spectrometry technique.

2.1. In Situ Gamma Spectrometry

In situ gamma radiation measurements were made using a portable gamma spectrometer with the HPGe detector placed directly on the “floor”, with the window facing the closest wall about 80 cm away (Figure 4). The entire measuring equipment consisted of the portable spectrometer equipped with an HPGe coaxial detector (model: GR4020), In Spector™ 2000 multichannel analyzer (for data collection) and Genie™ 2000 v.3.2.1 software package (for spectra analysis). The Ge crystal has a mass of about 0.98 kg and a diameter equal to 61 mm. The HPGe detector has a relative efficiency of 40% and is equipped with a 0.6 mm-thick carbon window; the FWHM is 2.1 keV for the 1.33 MeV ⁶⁰Co line, and the peak-to-Compton ratio (P-to-C) is 57/1. The spectrometer is cooled with liquid nitrogen. The measuring range of the gamma radiation spectra was 7–3150 keV with a gain of 5. Before in situ measurements, energy and efficiency calibration of the detector was performed. Energy calibration was performed using the following sources: ¹³³Ba, ¹³⁷Cs, ⁵⁴Mn, ⁵⁷Co, ¹⁰⁹Cd, ²²Na, and ⁶⁰Co, having a diameter of 1 cm and about 10 kBq activity. The efficiency calibration was performed using ISOCS™ software (Canberra Industries, Inc., Meriden, CT, USA). A room/box geometry with internal surface contamination was used. It was modeled in Geometry Composer software (Canberra Industries, Inc.) in the energy range and 10–7000 keV for mono-energy photons. The uncertainty of the photon detection efficiency ranged from 4 to 15%. The software modeled that the radioactivity was distributed evenly between the four walls, while the radioactivity from the two far ends was assumed not to contribute to the measurement. The equation of detection efficiency based on photopeak net counts as a function of gamma radiation energy [keV] is described using a fifth-order polynomial as follows:

ln(ε) = −10.88 − 8.884ln(E) + 5.996ln(E)² − 1.428ln(E)³ + 0.1452ln(E)⁴ − 0.005426ln(E)⁵.

(1)

In the recorded gamma-ray spectrum, radioisotopes were identified based on the energy of the photopeaks. In Genie^TM 2000 v.3.2.1 software, the areas under consecutive peaks were determined automatically by an analysis sequence, including an unidentified second difference fit and a nonlinear least-squares fit (LSQ).

Based on the gamma-ray spectra, the gamma-ray flux density, the effective dose rate and the total number of counts per second (cps) were calculated. The effective dose rate at the site of in situ measurements was determined using the photon flux to dose conversion factors taken from the ICRP report [23]. The photon flux determination method, effective dose rate and uncertainty estimation were described in our previous work [21].

2.2. In Situ Radon Measurement in Air

The concentration of ²²²Rn in the air was measured using a RAD7 portable detector (Durridge Company, Inc., Billerica, MA, USA). The detector was placed next to a gamma spectrometer and helium counters (Figure 4). This detector has several preprogrammed setups which can be applied to obtain ²²²Rn concentration in the air. The monitor uses a semiconductor detector (Planar Silicon detector with a passivated ion) to generate an electrical signal from the alpha radiation and, through spectral analysis, differentiate alpha particles based on their energy. Moreover, the RAD7 monitor allows for radon concentration measurement in the air below 4 Bq/m³ due to the low intrinsic background of the device lower than 0.2 Bq/m³. The measurements were performed using a 2-day protocol with a 1 h measurement cycle. First, in Sniff mode, radon concentration is calculated based on the daughter isotope of ²¹⁸Po. After three hours, for higher precision, the Sniff mode changes automatically to the Normal mode. In this mode, radon concentration is measured based on the signals originating from ²¹⁴Po and ²¹⁸Po isotopes.

2.3. α and 𝛾 Laboratory Spectrometry Techniques

One water sample was collected in a 1.5 L polyethene bottle and transferred to the laboratory for ^234,238U analysis using the alpha spectrometry technique. The sample was acidified with nitric acid to avoid the precipitation of iron on the wall of the bottle. Then, the known amount of the ²³²U tracer was added to the sample. The chemical analysis was based on the procedure worked by Suomela [24]. Uranium was separated from the other alpha radionuclides on the anion exchange resin Dowex^® 1 × 8 ((Cl) type, 200–400 mesh). Uranium fraction was coprecipitated with NdF₃ [25] and then filtered using a polypropylene filter (Pall Corporation, New York, NY, USA). A blank sample was prepared using 0.5 L of distilled water. Minimum Detectable Activity (MDA) equaled 0.5 mBq/L for both ^234,238U isotopes and two days of blank sample measurement time. The measurements were performed with the use of alpha spectrometers 7401VR and Alpha Analyst™ (Mirion Technologies (Canberra), Inc., Meriden, CT, USA) with semiconductor detectors (passivated implanted planar silicon (PIPS) detectors).

Six samples (two samples of dolomite, sandstone and shale) were taken at the site where in situ measurements were carried out. In the laboratory, samples were dried, ground and homogenized. The samples were placed in Marinelli containers, sealed to avoid radon leakage, and left for a month to achieve secular equilibrium between ²²⁶Ra, ²²⁸Ra and their daughters in the uranium and thorium series. The activities of ^226,228Ra and ⁴⁰K concentrations were calculated using a standard prepared from certificated materials obtained from Central Laboratory for Radiological Protection in Poland. The activity concentration of the ⁴⁰K isotope was calculated based on a 1460.8 keV photopeak. The concentrations of ²²⁶Ra and ²²⁸Ra isotopes were calculated as the weighted mean of the activities of daughter isotopes ²¹⁴Pb (295.2, 351.9 keV), ²¹⁴Bi (609.3, 1120.3 keV) and ²²⁸Ac (338.3, 911.1 keV), respectively. The laboratory measurements were performed using a gamma spectrometer from Canberra-Packard. The spectrometer consists of an HPGe coaxial detector (type: GC2018) of 60.7 mm crystal diameter, a mass of about 0.52 kg with a relative efficiency of 20% and Cryo-Pulse 5 Plus—an electrically powered cryostat. The detector was connected to a Digital Signal Analyser (DSA-LX). To reduce the external radiation background, the detector was placed in a low-background lead–copper shield manufactured by Ortec. The data acquisition and analysis was performed using Genie™ 2000 v.3.4.1 software from Canberra-Packard. The detector had the P-to-C equal to 53.8/1 and an energy resolution (FWHM) of 1.7 keV for a 1.33 MeV gamma line from ⁶⁰Co.

The sub-samples of about 1 g were prepared for the alpha-spectrometric measurements of ^234,238U isotopes. Samples were transferred to PTFE pressure decomposition vessels and digested using a microwave unit MAGNUM II (ERTEC-Poland). At the beginning of the chemical preparation known activity of ²³²U was added to each sample. Wet mineralization of analyzed rocks was performed with the use of the following hot acids: HF, HNO₃, HCl with H₃BO₃. Then, uranium was pre-concentrated with iron and co-precipitated using ammonia at pH 9. The precipitate, dissolved in 8 M HCl, was passed through an ion-exchange column. The next steps of the chemical preparation were the same as those applied to the water sample. Details on the chemical analysis of samples and determination of the analyzed isotopes can be found in previous publications [22,26,27].

2.4. Neutron Flux Measurements

To measure the thermal neutron flux, the same apparatus and the same method were used as in our earlier measurements at the Polkowice-Sieroszowice and GIG Experimental ‘Barbara’ mines [22]. It consists of measuring the counting rate using a setup of helium counters (sensitive to thermal neutrons), and then comparing the result with a simulation in which the setup was illuminated with an isotropic flux of thermal neutrons. The comparison of the measurement and simulation results allows for determining the real neutron flux at the measurement site. The assumption in the Monte Carlo simulation of an isotropic flux is a simplification, which, however, seems reasonable when a single measurement is made in the center of the mine tunnel.

The measurement setup consisted of helium counters (Figure 4), i.e., proportional gas counters filled with helium-3. The helium-3 nucleus captures a neutron invisible normally to the counter and then decays into two charged products: proton and tritium, which the counter can register. In this process, 764 keV of energy is released, which means that in the spectrum of amplitudes of the recorded events, neutron events create an easily identifiable peak.

We used two types of counters: Centronic 50He3/190/50MS and ZDAJ NEM425A50. Both types had the classic form of 50 cm steel pipes, but differed in diameter: 5 cm for Centronic and 2.5 cm for ZDAJ. Four counters of each type were used in the measuring setup. All the counters were placed vertically in the form of a flat tray: first, 4 Centronic counters, then 30 cm break, then 4 ZDAJ counters; additionally, there were 5 cm intervals between the counters (Figure 4), which reduces the mutual shading effect.

When there was a trigger (exceeding the comparison threshold), the oscilloscope-like waveform was recorded for each counter, sampled with a frequency of 10 MHz and ADC dynamics of 10 bits. The length of the waveform was dynamically adjusted to the length of the appearing signal, but for correct neutron events, the length was 128 or 192 samples. The oscilloscope-like waveform allowed off-line analysis to distinguish neutron events from various disturbances by means of the pulse shape analysis. In view of the low counting rate (several counts per hour) and difficult mining conditions, shape analysis is critical for the precision of the measurement.

Finally, the recorded counting rate was compared with Monte Carlo simulations carried out using the Geant4 package [28] version 10.07 using the physical package QGSP_BERT_HP and NeutronHPThermalScattering, which is important for low energy neutron interactions with nucleons bound in nuclei. The comparison of the measurement and simulation results made it possible to determine the neutron flux at the measurement site.

3. Results and Discussion

3.1. Analysis of the Gamma-Ray Spectrum

Figure 5 shows a 48 h gamma-ray spectrum recorded in shaft L-VI of the Lubin mine at a depth of 910 m. The total counting rate, similar to our previous works [21,22], was normalized to the day (d), energy (keV), and mass of the germanium crystal (kg). The total count rate per second in the energy range from 7 to 3150 keV is 224.59 ± 0.05. This result obtained in the Lubin mine is about 4 to 8 times higher than in the three main underground European laboratories (Modane, Boulby and Gran Sasso) [29,30,31], and 40 to 100 times higher than in the salt chamber in the Polkowice-Sieroszowice mine [32], but comparable to the anhydrite layer in the same mine [21].

Table 1. Apparent activity in [Bq/cm²], effective dose rate (EDR) in [pSv/s], with the calculated contribution (%) of each isotope to EDR estimated based on registered in situ gamma-ray spectrum. Uncertainties (denoted as Δ) are also presented.

Decay Chain	Nuclide	Apparent [Bq/cm2]	Δ [Bq/cm2]	EDR [pSv/s]	Δ [pSv/s]	% to EDR
Uranium	234Th	0.24	0.05	0.004	0.001	0.02
	234mPa	1.62	0.36	0.042	0.007	0.22
	226Ra	1.04	0.28	0.020	0.003	0.10
	214Bi	1.60	0.29	6.678	0.978	34.23
	214Pb	0.75	0.08	0.494	0.072	2.53
Thorium	228Ac	0.49	0.18	1.134	0.166	5.81
	208Tl	0.23	0.07	2.847	0.404	14.59
	212Pb	0.30	0.04	0.094	0.013	0.48
	212Bi	0.66	0.19	0.184	0.032	0.94
None	40K	15.36	1.68	7.733	1.095	39.64
	231Th	0.51	0.07	0.008	0.001	0.04
	X-rays			0.188	0.027	0.96
	Annihilation			0.083	0.012	0.42

presents the apparent activity and effective dose rate estimated based on the registered in situ gamma-ray spectrum. The calculated gamma-ray flux and the effective dose rate based on the 48 h measurement of gamma-ray is 8.08 ± 0.90 cm⁻²s⁻¹ and 0.070 ± 0.010 µSv/h (19.5 ± 2.8 pSv/s), respectively. The largest contribution to the effective dose rate came from ⁴⁰K (39.64%), ²¹⁴Bi (34.23%) from the uranium series, and ²⁰⁸Tl (14.59%) from the thorium series. The gamma-ray flux is approximately 13 times greater than in the Polkowice-Sieroszowice mine in the anhydrite layer [21] at a depth of 1014.4 m and approximately 65 times higher than in the salt chamber at a similar depth (930 m) [33]. On the other hand, the effective dose rate is about 9 times higher than in the anhydrite layer and 39 times higher than in the salt chamber.

In addition, the effective dose rate is about 3.5 times higher than in Modane (one of the main European ULs) [34]. These results show that a salt chamber in the Polkowice-Sieroszowice mine is a better place to establish a low-level underground laboratory. On the other hand, Lubin excavations can be used for other purposes where such low background radiation is not needed.

3.2. The Ratio of K-40/Bi-214

Similarly, as in our previous works [21,22,35,36], the ratio of ⁴⁰K/²¹⁴Bi was determined. The ratio of the number of counts under the 1460.8 keV (⁴⁰K) and 1764.5 keV (²¹⁴Bi) lines helps to distinguish between shallow and deep underground locations. ²¹⁴Bi isotope comes from the uranium series in the surrounding rock and also is derived from the radon, while ⁴⁰K is incorporated in rocks only. In the investigated location, the ⁴⁰K/²¹⁴Bi ratio is equal to 7.1, which points out that we are dealing with a deep location.

3.3. Radon Concentration in Air

The measurements were performed from 21 March to 23 March using a 2-day protocol. The concentration was in the range of 0 to 15.3 Bq/m³, with the arithmetic mean equal to 7.1 ± 0.5 Bq/m³ and the median equal to 7.1 Bq/m³. Moreover, no significant changes in radon concentration in the air over measurement time were observed. The previous measurement of radon concentration in the air in the Polkowice-Sieroszowice mine, within the anhydrite layer, also showed very low values [21], with the median equal to 4.8 Bq/m³. Exposure to high radon concentrations in the air can lead to lung cancer. Therefore, a limit on radon level in buildings intended for people and in workplaces has been introduced to be equal to 300 Bq/m³ [37,38,39]. The measured radon concentration in the air is significantly lower than the established reference level. The comparison of underground laboratories in terms of radon concentration previously showed [40] that the main factors determining its concentration are: radionuclide concentration in the rock, ventilation, temperature, and humidity. Low concentrations of radionuclides, including radon, are crucial in research conducted by underground laboratories [41,42], and therefore special constructions are being developed dedicated to radon reduction [43,44].

3.4. Radioactivity Content in Water and Rocks Samples

The ²²⁶Ra activity concentrations in investigated rocks samples ranged from 37.3 ± 1.4 Bq/kg to 247 ± 9 Bq/kg (

Table 2. Activity concentration of ²²⁶Ra, ²²⁸Ra, ⁴⁰K, ^238,234U, uranium content, ²³⁴U/²³⁸U, and ²²⁶Ra/²³⁸U activity ratios in investigated rock samples.

Sample	226Ra [Bq/kg]	228Ra(228Ac) [Bq/kg]	40K [Bq/kg]	238U [Bq/kg]	234U [Bq/kg]	U [mg/kg]	234U/238U	226Ra/238U
sandstone	40.1 ± 1.5	16.5 ± 0.5	586 ± 10	58.1 ± 3.2	58.2 ± 3.2	4.7 ± 0.3	1.0 ± 0.1	0.7 ± 0.1
sandstone	37.3 ± 1.4	16.1 ± 0.4	531 ± 9	45.5 ± 4.7	43.4 ± 4.6	3.7 ± 0.4	1.0 ± 0.1	0.8 ± 0.1
dolomite	65.9 ± 2.4	17.2 ± 0.4	641 ± 10	61.6 ± 4.5	55.5 ± 4.1	5.0 ± 0.4	0.9 ± 0.1	1.1 ± 0.1
dolomite	79.0 ± 2.9	16.1 ± 0.4	580 ± 10	87.3 ± 5.7	86.1 ± 5.6	7.1 ± 0.5	1.0 ± 0.1	0.9 ± 0.1
shale	186 ± 7	21.5 ± 0.5	747 ± 12	129 ± 8	124 ± 7	10.5 ± 0.6	1.0 ± 0.1	1.4 ± 0.1
shale	247 ± 9	30.0 ± 0.8	927 ± 15	249 ± 19	247 ± 18	20.2 ± 1.5	1.0 ± 0.1	1.0 ± 0.1

), with the mean equal to 109 Bq/kg and the median of 72 Bq/kg. ²²⁸Ra concentrations are in the range of 16.1 ± 0.4–30.0 ± 0.8 Bq/kg, with the mean and the median equal to 17 Bq/kg and 20 Bq/kg, respectively. For the ⁴⁰K isotope, the concentrations reached the value of 927 ± 15 Bq/kg (mean: 669 Bq/kg, median: 613 Bq/kg). According to the literature data of the European Commission [45], the average worldwide ²²⁶Ra, ²³²Th, and ⁴⁰K concentrations in the earth’s crust are equal to 40 Bq/kg, 40 Bq/kg and 400 Bq/kg, respectively. Analyzing the data provided in Table 2, the highest concentrations of the investigated isotopes are observed for shales. The lowest concentrations of ^226,228Ra and ⁴⁰K isotopes were observed in the sandstone samples. Moreover, sandstone and dolomite samples are characterized by similar concentrations of the ²²⁸Ra and ⁴⁰K isotopes and different values of the ²²⁶Ra isotope concentrations (higher in the case of dolomite than in sandstone). The ^234,238U concentrations ranged from 43.4 ± 4.6 Bq/kg to 247 ± 18 Bq/kg and from 45.5 ± 4.7 Bq/kg to 249 ± 19 Bq/kg, with the median equal to 72 Bq/kg and 74 Bq/kg, respectively. A previous study [21] of radionuclide content in anhydrite rocks in the Polkowice-Sieroszowice mine, at a depth of 1014.4 m, showed lower concentrations with the median equal to 1.1 Bq/kg, 0.2 Bq/kg and 8.9 Bq/kg for ²²⁶Ra, ²²⁸Ra and ⁴⁰K, respectively. The uranium content, calculated based on ²³⁸U concentration, is in the range of 3.7 ± 0.4 mg/kg to 20.2 ± 1.5 mg/kg. Additionally, a similar trend in the variability of the ^234,238U concentrations in the analyzed samples was observed, as in the case of the ²²⁶Ra isotope. The previous studies of uranium content in copper ore deposits in the Fore-Sudetic Monocline in Poland showed values up to 163 ppm [46]. Sun Y-Z et al. [47] indicated that uranium accumulates in the Coppershale due to the redox boundary (due to the presence of organic matter) between the Coppershale and the Rotliegende. Uranium concentrations in the analyzed samples (Table 2) are 3 to 8 times higher than ²²⁸Ra and confirm previous studies [16,46]. Moreover, Piestrzyński [16] noted that thorium is present in the Coppershale in small amounts up to 5 ppm and observed a weak negative correlation (r = −0.25) of thorium with uranium.

The analysis of the ²³⁴U/²³⁸U activity ratio for analyzed rock samples indicates a radioactivity equilibrium between these two isotopes. On the other hand, the ²²⁶Ra/²³⁸U activity ratio varied from 0.7 ± 0.1 to 1.4 ± 0.1, with the mean and median equal to 1.0 and 0.9, respectively. Slightly lower values of the ²²⁶Ra/²³⁸U activity ratios were observed in the case of sandstones.

The concentration of ^234,238U isotopes in the analyzed water sample was equal to 220 ± 15 mBq/L and 111 ± 8 mBq/L, respectively. The uranium content, calculated based on ²³⁸U concentration, was equal to 9.0 ± 0.7 µg/L. The occurrence of uranium in water results from leaching this element from the reservoir rocks, and the main factors determining the presence of uranium in waters are the content of radionuclides in rocks, reservoir rocks composition, pH, oxidation potential, mineral dissolution, partial pressure of CO₂, reservoir rocks composition [48]. The uranium ²³⁴U/²³⁸U activity ratio was equal to 2.0 ± 0.2. This is due to the selective leaching of the ²³⁴U isotope from the reservoir rocks or the α-recoil transfer of the ²³⁴Th isotope from the rock to water. The processes responsible for observed disequilibrium between ^234,238U isotopes in waters are extensively discussed in the literature [49,50].

3.5. Qualitative Neutron Activation Analysis

Additionally, the neutron activation of three samples of the collected rocks (anhydrite, shale, and sandstone) from the investigated location was performed. These were the same samples that were tested for the concentrations of uranium, potassium, and radium radioisotopes described in Section 3.4.

Each sample was activated with a neutron flux from a source of californium (²⁵²Cf). The activation time was similar for all samples and was about 21 days. Immediately after the completion of neutron activation, each sample was placed on the HPGe detector in a lead–copper shield (described in Section 2.3). Then, the spectrum of gamma radiation was measured. First, short measurement time was applied to identify short-lived radioisotopes, and subsequently, long measurements were taken to identify long-lived radioisotopes formed in the (n,γ) reaction.

The following isotopes were identified for each analyzed rock sample: ⁵⁶Mn, ²⁴Na, and ⁴²K. The isotope of ⁶⁵Zn was observed in the case of dolomite and shale. The presence of ⁶⁰Co was confirmed in the analysis of shale and sandstone samples. In the case of sandstone, the following isotopes were additionally identified: ⁸²Br and ⁷⁶As. Meanwhile, in the shale sample, the ⁷⁶As, ⁶⁴Cu, ^110mAg and ⁴⁶Sc isotopes were additionally observed.

Based on the analysis of neutron activation, it can be concluded that the radioisotopes were produced by the reaction (n,γ) with minerals of copper and other metals present in the tested samples. In this way, the resulting radioisotopes can increase radioactivity in the investigated location.

3.6. Neutron Flux Results

Data were collected over two days, in two series of 20 h. The average rate of neutron registration by a single counter was 7.1 ± 2.2 h⁻¹ for Centronic counters and 3.4 ± 1.2 h⁻¹ for ZDAJ counters, which corresponds well with the fact that the Centronic counter has twice the surface area than ZDAJ counter. The results for the individual counters are given in

Table 3. The neutron counting rate per hour, for particular Centronic and ZDAJ counters.

Centronic	ZDAJ
8.1 ± 1.8	3.9 ± 1.1
7.6 ± 2.1	3.5 ± 1.1
7.3 ± 2.7	3.3 ± 1.0
7.8 ± 2.3	3.5 ± 1.7

.

After comparing the measured registration rate with the setup simulation, we obtain that the neutron flux is 4.2 ± 1.2 × 10⁻⁶ cm⁻² s⁻¹ for Centronic counters and 4.1 ± 1.5 × 10⁻⁶ cm⁻² s⁻¹ for ZDAJ counters. So, as the final result of the measurement, we will take their average: 4.2 ± 0.9 × 10⁻⁶ cm⁻² s⁻¹.

This result, compared with the results of measurements carried out in other locations using a similar measurement set, is presented in

Table 4. Comparison of thermal neutron flux measurement in different European underground locations.

Location	Flux of Thermal Neutrons (×10−6 cm −2 s −1)
Lubin (Poland) (this work)	4.2 ± 0.9
Polkowice-Sieroszowice (Poland)	2.0 ± 0.2
Barbara (Poland)	8.6 ± 1.1
Gran Sasso, (Italy)	0.56 ± 0.22
Slanic Prahova, (Romania)	0.12 ± 0.05
Freiberg, (Germany)	3.12 ± 0.10
Pyhäsalmi; Lab2 (Finland)	17.30 ± 0.10

(extended version based on [22]).

The neutron flux in the Lubin mine is approximately twice as high as that obtained in Polkowice-Sieroszowice [21], even though the measurements were made using the same apparatus and in the same deposit. The most likely explanation for this difference is the different rock compositions in both locations. However, the different ventilation efficiency cannot be ignored in explaining the observed difference in the neutron flux.

3.7. Radiological Risk Assessment

Several risk indicators are used in the literature to estimate the potential radiation hazard from exposure to external gamma radiation emitted from building materials, rocks and soils [51,52]. Among them, the following risk indicators are often used: external hazard index (H_ex), radium equivalent activity (Ra_eq); these parameters can be calculated based on the equations [51,53]:

$${\mathrm{Ra}}_{\mathrm{eq}}={}_{}{}^{226}\mathrm{R}\mathrm{a}+1.43{}_{}{}^{232}\mathrm{T}\mathrm{h}+0.077{}_{}{}^{40}\mathrm{K}.$$

(2)

$${\mathrm{H}}_{\mathrm{ex}}=\frac{{}_{}{}^{226}\mathrm{R}\mathrm{a}}{370}+\frac{{}_{}{}^{232}\mathrm{T}\mathrm{h}}{259}+\frac{{}_{}{}^{40}\mathrm{K}}{4810}.$$

(3)

where the activity concentration of ²²⁶Ra, ²³²Th, and ⁴⁰K isotopes are expressed in Bq/kg. In risk estimation, the ²²⁸Ra concentration was used to express ²³²Th activity. The calculated Ra_eq and H_ex values are presented in

Table 5. Radium equivalent activity (Ra_eq), external hazard index (H_ex), absorbed dose rate (D) [nGy/h], and annual effective dose (ED) [mSv/y] estimated based on the ^226,228Ra and ⁴⁰K concentrations in collected rock samples.

Sample	Raeq [Bq/kg]	Hex	D [nGy/h]	ED [mSv/y]
sandstone	109	0.29	373	0.43
sandstone	101	0.27	339	0.39
dolomite	140	0.38	418	0.48
dolomite	147	0.40	387	0.45
shale	274	0.74	538	0.62
shale	362	0.98	675	0.78

.

The maximum allowed risk indicator should not exceed 370 Bq/kg and 1 for Ra_eq and H_ex, respectively, to keep the external gamma radiation dose below 1.5 mSv/y [51,53]. The highest Ra_eq and H_ex values were obtained for shale samples, which are close to established limits. The median values of Ra_eq and H_ex are equal to 143 Bq/kg and 0.39, respectively. It is worth emphasizing that the highest ^226,228Ra and ⁴⁰K concentrations and evaluated risk indicators were obtained for shales, which constitute a negligible fraction (about 1%) of the entire rocks present in the measurement location.

The absorbed dose rate (D) expressed in [nGy/h] can be calculated based on (Equation (4)) [51,54] and characterize the absorbed gamma dose rate in the air at a high of 1 m above the ground level from uniformly distributed ²²⁶Ra, ²³²Th, and ⁴⁰K radionuclides:

$$\mathrm{D}\left[\frac{\mathrm{nGy}}{\mathrm{h}}\right]=0.462{}_{}{}^{226}\mathrm{R}\mathrm{a}+0.604{}_{}{}^{232}\mathrm{T}\mathrm{h}+0.0417{}_{}{}^{40}\mathrm{K}.$$

(4)

The United Nations Scientific Committee on the Effects of Atomic Radiation [55] indicates that population-weighted and the median value of the absorbed dose rate in the air outdoor from terrestrial ⁴⁰K, ²²⁶Ra, and ²³²Th radionuclides in the soil is equal to 60 and 51 [nGy/h], respectively. However, the calculated values of absorbed dose rate (Table 5) are higher than the above-cited values in the range of values presented in the literature [56,57]. The values of the absorbed dose rate calculated based on laboratory measurements of rock samples are higher than the above-reported dose rate estimated based on the photon flux to dose conversion factors taken from the ICRP report [23].

Based on the absorbed dose rate (D), the annual effective dose rate (ED) in [mSv/y] can be calculated according to Equation (5):

$$\mathrm{ED}\left[\frac{\mathrm{mSv}}{\mathrm{y}}\right]=\mathrm{D}\times 0.7\times \mathrm{OF}\times {10}^{-6}.$$

(5)

where the occupancy factor was calculated as follows:

$$\mathrm{OF}=\frac{7\mathrm{h}}{\mathrm{day}}\frac{5\mathrm{days}}{\mathrm{week}}\frac{47\mathrm{weeks}}{\mathrm{y}}=\frac{1645\mathrm{h}}{\mathrm{y}}.$$

(6)

According to the UNSCEAR report [55], the conversion coefficient from the absorbed dose in the air to the effective dose received by adults is equal to 0.7 [Sv/Gy]. In dose assessment, the daily seven-hour working time was assumed (staying in the place from which the samples were taken). The calculated effective radiation doses are presented in Table 5, and for the shale sample reaches the value of 0.78 mSv/y. The calculated weighted mean value of the effective radiation dose is equal to 0.44 mSv/y, which is significantly below the limit of 1 mSv/y [58]. It should be emphasized that the highest risk indicators were recorded for shale samples (Table 5), which constitute a negligible fraction (about 1%) of rocks.

3.8. Relation between Local Geology and Radiation Level

There are three main elements responsible for natural radiation in sedimentary basins: potassium, uranium and thorium. Potassium in sedimentary rocks is concentrated in sylvite, carnallite and polyhalite but can be found, in smaller amounts, in more common minerals such as some clays and feldspars. Uranium can be adsorbed onto organic matter or associated with phosphates. Thorium can be adsorbed onto terrestrial clays such as kaolinite [59].

The main uranium carrier in or close to ore-bearing horizons is secondary thucholite. As uranium probably was remobilized from thick red sandstones [60] by ascending brines, the biggest concentrations are recorded on or close to the Weissliegend and Coppershale boundary [16,61]. Gamma radiation anomaly zones were identified in mining areas (Figure 1).

Locally, a “thucholitic shale”, up to 4 cm thick, with average U content of 180 ppm and Th of 30 ppm, was recognized by Kucha [61]. An average uranium content of 61.5 ppm and 5.1 ppm for thorium in Coppershale, decreasing to avg. 10.6 ppm U and 0.9 ppm Th in Zechstein Limestone (Lubin-Sieroszowice district), were noted [16]. In general, fragments abundant with organic matter are enriched in uranium. Additionally, phosphates in shales also show significant amounts of U up to 0.24 wt.% [16].

Sylvite, carnallite, and polyhalite are minerals precipitated in the late stage of the evaporation cycle or diagenetic minerals connected to that deposition. They occur in Zechstein sediments in the described basin but not in the mining district area (at least not in significant volumes).

In good agreement with the described literature data of radiation distribution, in the described lithology profile, are natural gamma-ray logs. They provide information on combined K, U, and Th gamma radiation, or they measure specific wavelengths of gamma-ray output from radioactive elements stored in the rock formation. The measured natural radioactivity of rocks is compared with that of a known standard rock and presented in [API] unit, defined by American Petroleum Institute [59].

Based on wire logs data reports made by Geofizyka Toruń and Geofizyka Kraków as part of the KGHM Polska Miedź S.A. exploration program (2015–2021)—north of the mining areas—radiation in lithological profiles can be described (Figure 6).

In general, according to gamma logs (Figure 6), a decrease in natural radiation values is observed from the upper parts of sandstones through carbonates to anhydrites and salts of the Werra cyclothem. If Coppershale occurs in the profile, the high gamma-ray response is observed within or at the base of the thin shale layer. In a few profiles from Zechstein Limestone platform facies, another intensive, occasionally the strongest, radioactivity is observed—2–8 m below the sandstone-shale boundary. In drillholes, from the basin and less often from slope facies, a much smaller anomaly close to the carbonate–anhydrite boundary can be observed.

If instead of a salt layer, anhydrite breccia was drilled, another peak could be observed, as this rock formation consists of an important admixture of clay minerals. In general, gamma logs provide information about layers with clay minerals or abundant in organic matter. In some cases, it can be related to secondary uranium concentrations.

4. Conclusions

For years, attempts have been made to build an underground laboratory in Poland and to search for a convenient location. Therefore, a detailed characterization in terms of the level of natural radioactivity of the underground location in the Lubin mine has been carried out.

The laboratory analysis showed the variability of ^234,238U, ^226,228Ra and ⁴⁰K concentrations in the analyzed rock samples, and the highest values were observed in the case of shales. Moreover, the concentrations of ^234,238U and ²²⁶Ra isotopes were higher than ²²⁸Ra. The rock samples were characterized by higher concentrations of the analyzed radionuclides in relation to the rocks collected in the anhydrite layer (at a depth of 2941.8 m w.e.) and in the salt cavern (at a depth of 2700 m w.e.) in the Polkowice-Sieroszowice mine.

The concentration of radon in the air in the analyzed location was very low and comparable with the values obtained in the leading European laboratories. Moreover, the radon concentration was lower compared to the level in the salt cavern and comparable to the radon concentration measured in the anhydrite layer of the Polkowice-Sieroszowice mine.

Comparing the results obtained in the present study with the results of measurements in two other potential locations (the Polkowice-Sieroszowice mine anhydrite layer and salt chamber) of the underground laboratory (also owned by KGHM Polska Miedź S.A.), it can be concluded that the Lubin mine is characterized by a higher level of natural radioactivity than in the other locations which were under consideration. On the other hand, the analyzed location in the Lubin mine is characterized by several advantages related mostly to the safety of operation and easy access to the site. The development of an underground laboratory within the hard and stable stratum of dolomite and sandstone, with proper roof support, ensures the stability of the excavated chamber for dozen years. Another advantage of the potential location of UL in the Lubin mine is its close vicinity to one of the main ventilation shafts, significantly reducing the problem of high temperature and lack of efficient airflow. This fact also affects the possibility of transporting large components and parts of any detector. Moreover, a location at a depth of 2275 m w.e. ensures a significant reduction in muon flow. Unfortunately, up till now, the measurements of the muon flux have not been carried out. However, based on the literature data [65], the roughly estimated muon flux is about 0.0014 m⁻²s⁻¹.