Introduction

The measurement of naturally occurring radioactive materials (NORMs) in environmental samples has received increasing attention recently. This is primarily because of the concern about human exposure resulting from the radioactive contamination of the environment. The environmental radiation background is a representation of the natural radioisotopes of 238U, 232Th, primordial 40K, and other artificial radionuclides such as 137Cs. Natural radiation accounts for approximately 85% of the human effective dose (UNSCEAR 2000; IAEA 1996). Its presence at certain levels can affect the human body, depending on location and geological characteristics. The activity levels of these radionuclides can be determined by a gamma radiation hazards assessment. The natural radiation to which the human body is exposed mainly originates from the decay of radium isotope (226Ra), which produces radon gas (222Rn) and its decay progeny (Yuvi 1988; Abdullahi et al. 2019). Radon is a noble gas that appears in different geological formations such as soil, rocks, the earth’s crust, and building materials. The local environmental radon level is mainly determined by the radon exhalation rate from the soil, and the content of 226Ra in soil plays an important role in the accumulation of 222Rn gas in the outdoor and indoor environments. However, 226Ra and 222Rn are the decay products of 238U with an estimated average of 2.7 mg kg−1 (Lide 1994). Therefore, environmental radiation background surveys are important for understanding radiological and geological characteristics, and also for identifying the biochemical and geochemical traces in the environment. Gross α and gross β activities in environmental samples (e.g., rocks, sand, soil, and water) have attracted considerable attention in recent years, because our living environment has been contaminated by natural radionuclides. Furthermore, elevated levels of β radiation may be a sign of the accidental atmospheric release of radioactivity from the Chernobyl nuclear reactor or nuclear test explosions. The total radioactivity of α emitters in the soil of a particular area determines the area’s gross α activity, which depends on geological formations, mineral distribution, and the type of activities in the selected area.

An artificial radioactive element such as cesium is produced in nuclear power plants and disposed in wastewater. The Fukushima Daiichi nuclear power plant accident in Japan increased the activity concentrations of 134Cs and 137Cs, with the respective half-lives of 2.06 and 30.4 years, in the soil, air, and water to dangerous levels (Shizuma et al. 2018). Several studies have described successful methods for removing radioactive elements such as cesium (134Cs and 137Cs) and cobalt (Co(II)) from nuclear wastewater. These studies involved the development of ligand-based nanomaterials such as mesoporous silica and conjugate adsorbent (Awual at al. 2014a), inorganic and conjugate adsorbent (Awual et al. 2014b), DSDH ligand (Shahat et al. 2015), nano-conjugate adsorbent (Awual et al. 2015), organic–inorganic mesoporous hybrid (Awual et al. 2016a), crown ether-based mesoporous adsorbent (Awual 2016), Schiff base ligand (Awual et al. 2017), and macrocyclic ligand (Awual et al. 2016b).

No study has yet reported the levels of natural radioactivity in the soils of the Al-Qassim, Al-Ghat, Al-Majmaah, and Al-Zulfi regions of Saudi Arabia. The present study aimed to estimate the natural radioactivity concentrations, radon concentrations, and gross α and β activities in these regions and to provide a radiation map as a reference for future radiological evaluations. The activity concentrations of 238U, 226Ra, 232Th, and 40K in 138 soil samples across the four regions were determined and then used to estimate the associated radiological hazards. The measurements were performed using a high-purity germanium (HPGe) gamma-ray detector. The gross α and β activities in the soil samples were estimated using low-background proportional detectors. The radiological hazards and risk assessment due to naturally occurring radionuclides (232Th, 226Ra, and 40K) in the samples were estimated and compared with available results from other countries.

Materials and methods

Geology of the study area

This study was conducted in four different regions (Al-Qassim, Al-Ghat, Al-Majmaah, and Al-Zulfi) of Saudi Arabia. The study regions (latitude 24–27° N and longitude 41–44° E) are located in the center of the Arabian Peninsula, north-west of Riyadh, and have a population of 1.9 million (Fig. 1). The geology of the study area is not uniform; it consists of sand, clay, and limestone. The Rummah Valley or Wadi Rummah, which has various dry riverbeds, crosses the entire Al-Qassim region and is one of the longest valleys in the Arabian Peninsula, covering a distance of over 600 km starting from the west near Medina to the border of Kuwait in the north-east of the Arabian Peninsula. In the Al-Qassim province, the soil consists of loose deposits in its natural state, but it can be compacted to attain relatively high, dry densities (Al-Refeai and Al-Ghamdy 1994).

Fig. 1
figure 1

The map of Saudi Arabia showing the study area

Full size image

Sample collection and preparation

There are several approaches for collection of soil samples include random, judgmental, stratified random, systematic random grid, systematic grid, search, and transect sampling (IAEA 2004). After field visit, the most suitable approach for the current study is systematic random grid due to some restrictions of sample collections in some grid points. The total area of the Al-Qassim province, Al-Ghat, Al-Majmaah, and Al-Zulfi was subdivided using square grids, and the soil samples were collected from within each cell using the random selection procedures. Soil samples were collected from 138 different sites at a depth of 0.2–0.3 m from ground level to obtain undistributed and pure soil samples from the four study regions, as shown in Fig. 2. Approximately 2–3 kg of soil from each location was collected in a polyethylene bag. The samples were sieved using a 2-mm mesh to remove unwanted materials. The samples were then dried in an oven at 90 °C for 24 h to remove any moisture. Additional sieving was performed using a 500-μm sieving machine (Octagon digital) to achieve homogeneity between samples. The dried, sieved samples were weighed and transferred into Marinelli beakers (500 mL) and sealed with adhesive tape. The beakers were then stored for 30 days to achieve secular equilibrium between radium (226Ra) and radon (222Rn) and radon’s most abundant daughter products.

Fig. 2
figure 2

Location of sampling points (blue circles) and study regions: Al-Qassim (red), Al-Zulfi (green), Al-Ghat (yellow), Al-Majmaah (blue), and unmeasured areas (gray).

Full size image

To measure the gross α and β activities, approximately 2 g of each dried sample was placed on a 5.08-cm-diameter stainless-steel planchet, and a small amount of UHU glue diluted with acetone was spread over the sample surface. The planchets were then kept under an infrared lamp for a few hours until the samples were fully glued.

Experimental technique

The activity concentration of radionuclides in the soil samples was determined by using a gamma-ray spectrometry technique. The gamma-ray spectrometry system comprised a high-resolution (n-type) HPGe coaxial detector (ORTEC) with a pre-amplifier, amplifier, high-voltage power supply, and multi-channel analyzer (MCA) coupled to a PC. To reduce the thermally induced leakage current, the HPGe detector was cooled using an X-COOLER system (ORTEC). The HPGe detector was placed in a 10-cm-thick cylindrical Pb shielding that was internally lined with a sheet of Cu (1.6 mm) and Sb (0.5 mm) to reduce the external gamma-ray radiation background in the measured spectrum. The HPGe detector efficiency curve based on the experimental data was calculated from the following equation:

$$ {\varepsilon}_c=\frac{N_c}{t_c\times {A}_c\times {P}_{\gamma }}, $$

where εc is the photopeak efficiency corresponding to a specific radionuclide, Nc is the net count in the photopeak, tc is the measurement time (s) of the calibration source, Ac is the present activity of the source, and Pγ is the probability of γ-emission of the radionuclide. The energy resolution (FWHM) of the spectrometer was 0.8 keV and 2.2 keV for the 122-keV 57Co and 1332-keV 60Co gamma-ray lines, respectively. The energy and efficiency calibrations of the detector were performed using a mixed radionuclide source of 210Pb, 88Y, 241Am, 137Cs, 57Co, 109Cd, 139Ce, 203Hg, 113Sn, 60Co, and 85Sr. The source was kept inside a 500-ml Marinelli beaker with the same geometry as that used for sample measurements. The efficiencies and activities were determined from the counts in the photopeak of the specific nuclides using the gamma vision software. Each soil sample was counted for 60,000 s. The background spectra, which were subtracted from the sample spectra, were collected for 48 h using deionized water in the Marinelli beaker.

Activity concentration determination by gamma-ray spectroscopy

The activity concentration of 226Ra was calculated from the gamma-ray lines of energies 338.3, 911.1, and 969.11 keV (228Ac); 609.3 keV (214Bi); and 351.9 keV (214Pb). The gamma-ray lines of 238.63 keV (212Pb), 727.3 keV (212Bi), and 583.19 keV (208Tl) were used to determine the activity concentration of 232Th. The gamma-ray photopeaks at 661.6 keV and 1460.8 keV were used to determine the activity concentrations of 137Cs and 40K, respectively. The minimum detectable activities (MDAs) for the soil samples were 0.40 ± 0.12 Bg kg−1 for 226Ra, 0.32 ± 0.08 Bg kg−1 for 232Th, 1.68 ± 0.52 Bg kg−1 for 40K, and 0.03 ± 0.02 Bg kg−1 for 137Cs. The activity concentrations of specific radioisotopes were determined using the following expression (Dovlete and Povinec 2004):

$$ A=\frac{C_n}{\varepsilon_f{P}_{\gamma }{t}_sm}, $$

where A is the activity concentration of a specific radionuclide in Bq kg−1, Cn is the net count of the corresponding peak, Pγ is the absolute emission probability for a specific gamma-ray energy, ɛn is the absolute energy efficiency, m is the mass of the sample in kg, and ts is the counting time in s.

Determination of radon concentration

The radon (222Rn) concentration of the sampling locations was measured using a RAD7 semiconductor-based detector connected to a soil gas probe. At each sampling location, the stainless-steel probe, with holes near the tip, was immersed in the soil to a depth of 1 m. The probe was connected to the RAD7 detector with a desiccant tube, followed by inert filters for absorbing the soil gas. The counting time was 30 min for each location (Durridge Co 2017; Thu et al. 2019).

Gross α and β activity measurements by proportional counting system

An advanced low-background α/β proportional system (LB4200, CANBERRA) was used to determine the gross α and β activities. The multi-detector counting system was equipped with 16 gas flow proportional detectors using a mixed P10 gas (90% Ar and 10% CH4). The system was calibrated simultaneously using disk-shaped standard sources (90Sr for β and 241Am for α particles). The counting efficiency of the low-background proportional counting system was found to be 2.94% for α particles (based on the guaranteed reagent U3O8 standard) and 28.92% for β particles (based on the high-purity reagent KCl standard) using 2 g of the standard reference sample. The background and samples were counted for 24 h. For the soil samples, the MDAs were 0.11 Bg kg−1 for gross α and 0.16 Bg kg−1 for gross β.

Radiological risk assessment

The specific activity concentrations of the 226Ra, 232Th, and 40K nuclides were used to estimate the outdoor external dose rates for the measured samples. The internal hazard index (Hint), external hazard index (Hext), radium equivalent activity (Raeq), and annual effective dose rate equivalent (AEDE) were calculated using the following equations (Berekta and Mathew 1985; UNSCEAR 2008):

$$ {Ra}_{eq}={A}_U+\left({A}_{Th}\times 1.43\right)+\left({A}_K\times 0.077\right) $$
$$ DR=\left(0.462\times {A}_U\right)+\left(0.604\times {A}_{Th}\right)+\left(0.0417\times {A}_K\right) $$
$$ \mathrm{AEDE}=\mathrm{D}\times 1.23\times \frac{10^{-3} mSv}{y} $$
$$ {\mathrm{H}}_{\mathrm{int}}=\left(\frac{A_U}{185}+\frac{A_{Th}}{259}+\frac{A_K}{4810}\right), $$

where DR is the absorbed dose rate in nGy h−1 in the air 1 m above the ground surface; Raeq is in Bq kg−1; AEDE is in mSv year−1; and AU, ATh, and AK are the activity concentrations of 238U, 232Th, and 40K in Bq kg−1 in the soil samples, respectively.

Results and discussion

Radionuclides concentration

Table 1 lists the range of activity concentrations and average activity concentrations of the 238U, 232Th, 226Ra, and 40K nuclides for all soil samples. The overall average activity concentrations of 238U, 232Th, and 226Ra in the soil samples were 23.80 ± 11.61, 24.33 ± 17.63, and 23.83 ± 11.70 Bq kg−1, respectively. These average activity concentrations are lower than the worldwide average values of 35, 30, and 35 Bq kg−1, respectively, whereas for 40K, the mean activity concentration was found to be 790 ± 398 Bq kg−1, which exceeds the worldwide average of 400 Bq kg−1 (UNSCEAR 2000). The activity concentrations of 238U, 232Th, 226Ra, and 40K at the sampling points in the Al-Qassim region were higher compared with those in the other three regions (Al-Ghat, Al-Zulfi, and Al-Majmaah). This may be caused by the heavy mineral groups from the Paleozoic Era that are dominated by the ultra-stable minerals zircon, tourmaline, and rutile in the Al-Qassim region. Other heavy minerals, including garnet, monazite, staurolite, and kyanite, were also found in small quantities. Among light minerals, alkali feldspars were more abundant than plagioclase, owing to the significantly higher K2O concentrations, controlled by the presence of K-feldspar, K-mica, and glauconite (Bassis et al. 2016). The activity levels of 238U, 232Th, and 40K in the soil samples are compared with those from other countries in Table 2. The variations in the activity levels can be explained by the differences in geological and geographical conditions, as well as by the geochemical characteristics. Figure 3 shows the activity concentration distributions of 238U, 232Th, 226Ra, and 40K in each soil sample compared with their worldwide average values, indicated by red lines.

Table 1 Range and average values of radionuclide (238U, 226Ra, 232Th, and 40K) concentrations, radon concentration, and gross α and gross β activities in the soil samples of the selected regions
Full size table
Table 2 Comparison between the levels of 238U, 232Th, and 40K in the soil samples and available data for other countries
Full size table
Fig. 3
figure 3

Activity concentration distributions of 238U, 232Th, 226Ra, and 40K in the soil samples; red lines represent worldwide average values

Full size image

Figure 4 shows the distribution of the measured activity levels of 40K from the present work. The highest activities were found in the southwestern Al-Qassim region, and it is clear that its activity decreases from west to east. The maximum concentration of 40K found in this work was below the highest worldwide concentration. The high levels of 40K concentrations in the south-west region may be attributable to the high content of heavy and lighter minerals in red to pink sedimentary rocks and partially micaceous silty sandstones. Moreover, there are various dry riverbeds, many crossing the entire region from the west to the north-east (Al-Refeai and Al-Ghamdy 1994; Abu Shayeb et al. 2017).

Fig. 4
figure 4

Radiation map of activity concentration distributions of 40K for selected regions (Al-Qassim province, Al-Ghat, Al-Zulfi, and Al-Majmaah)

Full size image

The results from the activity concentration measurements were used to generate a radiological map of the selected regions (138 sampling points), as depicted in Figs. 4, 5, 6, 7 for 40K, 238U, 226Ra, and 232Th, respectively. The activity distribution of 40K, 238U, 226Ra, and 232Th of the selected regions was decreased in a way west-south > north-west > north-east > east-south. However, radionuclide concentrations in the south-west region of Al-Qassim were higher than those in the other regions due to being highly enriched in zircon (Bassis et al. 2016). Benaafi and Abdullatif (2015) reported that the mineralogical composition of the sand dunes in the Al-Qassim region has a high percentage of SiO2 (quartz) and a very low percentage of feldspars. However, in some areas, feldspars and calcite were found significant minerals. A reasonable hypothesis would be that uranium, thorium, and potassium-enriched crystallinity is the main source of heavy and lighter minerals in the sandstones or sedimentary rocks of the Al-Qassim region during the Ordovician age (Pitkin and Huffman 1986).

Fig. 5
figure 5

Radiation map of activity concentration distributions of 238U for selected regions (Al-Qassim province, Al-Ghat, Al-Zulfi, and Al-Majmaah)

Full size image
Fig. 6
figure 6

Radiation map of activity concentration distributions of 226Ra for selected regions (Al-Qassim province, Al-Ghat, Al-Zulfi, and Al-Majmaah)

Full size image
Fig. 7
figure 7

Radiation map of activity concentration distributions of 232Th for selected regions (Al-Qassim province, Al-Ghat, Al-Zulfi, and Al-Majmaah)

Full size image

The measured activity concentrations were correlated for all samples for 238U/226Ra and 232Th/226Ra. The ratio of 238U/226Ra and 232Th/226Ra varied from 0.6 to 1.5 and 0.4 to 2.2, respectively. The average values of 238U/226Ra and 232Th/226Ra were 1.0 ± 0.2 and 1.1 ± 0.3 respectively. There is a slight variation in 232Th/226Ra activity ratios across the soil samples, which may be due to the fact that uranium or thorium has been absorbed by clay minerals or organic matters in soil.

Radon concentration

The range and average value of the radon (222Rn) concentration in all soil sample locations from the present work are listed in Table 1. The measured 222Rn concentration varied from 39 to 508 Bq m−3 with an average of 145.0 ± 52.9 Bq m−3. The measured values are in agreement with the typical values (UNSCEAR 2000; Eisenbud and Gesell 1997). The levels of 222Rn in the samples from the Al-Qassim region were higher than those from the other regions, which can be attributed to the different soil types and compact packing of the sandstones/bedrocks. Based on the Sweden Criteria radon hazard assessment (Lara et al. 2015), a radon activity concentration below 10 kBq m−3 is classified as “low risk,” a concentration between 10 and 50 kBq m−3 is classified as “normal risk,” and a concentration greater than 50 kBq m−3 is classified as “high risk.” Results from the current work show that all values classified as “low risk.”

Gross α and gross β activities

The gross α and β activities of the samples from the selected regions are summarized in Table 1. The gross β activity in groundwater samples from the same regions (Al-Qassim, Al-Zulfi, and Al-Majmaah) was found to be approximately 5–100 times higher than the gross α activity, owing to the high concentrations of 228Ra and 40K, compared with the 226Ra concentration (Alharbi et al. 2018). The gross α activity depends only on the levels of natural α-emitters (238U, 234U, 232Th, 230Th, 228Th, 226Ra, and 210Po), while the gross β activity depends on the levels of natural β-emitters (228Ra, 210Pb, and 40K). It was found that the levels of gross α and gross β activities at the sampling points of the Al-Qassim region were higher than those of the other three regions (Al-Ghat, Al-Zulfi, and Al-Majmaah), which may be due to the high concentrations of 238U, 235U, 232Th, 226Ra, and 40K found in this study. Table 3 compares the levels of gross α and gross β activities in the soil samples with those from other countries. The different ranges of gross α and gross β activities are due to the activity concentrations of uranium (234U and 238U), radium (226Ra and 224Ra), polonium (210Po), lead (210Pb), and potassium (40K), which normally depend on the geological and geochemical characteristics of the soil.

Table 3 Comparison of the range of gross α and gross β activities in the soil samples with data from other countries
Full size table

Radiological hazards and risk assessment

The evaluated radiological risk indices for the soil samples from the present work are summarized in Table 4. The gamma-absorbed dose rate (DR) in the air 1 m above ground level is defined as the radiation dose received by a person from gamma-rays emitted by the radionuclides (232Th, 226Ra, 40K, and 137Cs) present in the soil. The DR values of the samples varied from 11.05 to 202.04 nGy h−1, with an average value of 58.88 ± 29.21 nGy h−1. Radium equivalent activity (Raeq) was used to assess the hazards associated with soil that contains 232Th, 226Ra, and 40K; it is determined by assuming that 370 Bq kg−1 of 226Ra, 260 Bq kg−1 of 232Th, and 4810 Bq kg−1 of 40K produce the same gamma dose rate (UNSCEAR 2000). The range of Raeq for the samples was 23.6–147.8 Bq kg−1, with an average value of 118.48 ± 59.65 Bq kg−1. Hint was used to evaluate the internal radiation exposure to the respiratory system through the inhalation of radioactive gasses (radon and thoron) and their short-lived daughter products, which emit α particles. The Hint values of the samples varied from 0.07 to 1.45 with an estimated mean value of 0.4 ± 0.2. Hext was used to estimate the hazards due to exposure to gamma-ray radiation associated with the 232Th, 226Ra, and 40K in soil. The Hext values of the samples varied from 0.1 to 1.2, with an estimated mean value of 0.3 ± 0.2. The AEDE values of the samples ranged from 0.01 to 0.25 mSv year−1, with an estimated average of 0.07 ± 0.04 mSv year−1, which is below the maximum recommended worldwide level of 1 mSv year−1 (UNSCEAR 2000).

Table 4 Radiological risk factors measured for all samples from the current work
Full size table

Conclusion

The activity concentrations of radionuclides (238U, 232Th, 226Ra, and 40K) and a gamma radiation hazard assessment were obtained for 138 soil samples using an HPGe gamma-ray spectrometer. The average activity concentrations of 238U, 232Th, and 226Ra in the samples were found to be lower than the worldwide average values, whereas the average activity of 40K was much higher than the worldwide average value. The radon concentration from the sample locations was within the typical range and can be considered “low risk.” A multi-detector low-background α/β counting system was used to determine the gross α and β activities in the samples, and the values obtained ranged from 29.72 ± 1.43 to 1546.64 ± 115.06 Bq kg−1 and 146.37 ± 24.95 to 1651.32 ± 44.15 Bq kg−1, respectively. Finally, the radiological hazards associated with the natural radioactivity in the soil samples were estimated.