Introduction

Radon (Rn) is an element having an atomic number of 86. It is a colorless, odorless, tasteless radioactive noble gas. It occurs, naturally, as an intermediate step in the normal radioactive decay chains of uranium and thorium. Radon is the immediate decay product of radium elements and is considered an important constituent of the natural radiation environment. The amounts of its isotopes in the environment depend primarily upon the concentrations of 238U, 235U, and 232Th in the soil and rocks. These isotopes are radioactive; their half-lives are so long that the time since the formation of universe (since the time these radionuclides were formed) is not sufficient for them to decay to stable elements.

Therefore, the half-life of 238U is estimated to be as the age of the earth. The abundance of 232Th in the earth crust is higher than that of 238U, but because of the longer half-life of 232Th. There are three natural isotopes of radon (Table 1) 219Rn, 220Rn, and 222Rn. All three radon isotopes are alpha-decay radionuclides. 222Rn is the daughter of 226Ra and the parent of 218Po. The most stable isotope is 222Rn. Under standard conditions, it is gaseous and easily inhaled, therefore causing a health hazard. The exposure level to radon gas differs from place to place (El-Gamal et al. 2019; ICRP 2014; Lecomte et al. 2014; Salaheldin et al. 2019, 2021, 2020a; Günay et al. 2019). It is a dangerous contaminant that affects indoor air quality worldwide. Its daughters were associated with lung cancer induction in humans exposed to high concentrations for prolonged periods. Radon causes between 6 and 15% of lung cancers worldwide (Wakelee et al. 2007). The dangers of indoor radon exposure are the subject of a large research work (NRC 1999; Darby et al. 2005; Krewski et al. 2005). The highest non-occupational exposure risk to increased radon gas levels between population includes home dwellers, as gas are trapped indoors (i.e. released into the air from water, soil, building materials, and natural gas use). The public and medical community often underestimates the potential risk of cancer due to radon exposure.

Table 1 Physics of the natural radioactive elements

Building materials made of soil (e.g. clay bricks) or stone always contain uranium and radium. Radon can be released from these materials into the surrounding air. The amount produced depends on radon production speed and the material porosity. The content is usually low, but some materials may contain high levels of radium-226. Radon deposits can reach hundreds of thousands of Becquerel per liters (Bq l−1) in water from wells drilled in regions with granitic rocks (Finland and Sweden). This gives in to the internal radon and exposure via ingestion (El-Taher et al. 2021; Salonen 1994). For ordinary building materials, the volumetric rate of admission is between 0.05 and 50 Bq (m3 h)−1 and the corresponding concentration is between 0.03 and 30 Bq m−3 (for an average rate of air renewal of 0.7 h−1). There are examples of concrete with high concentration of radium, such as that produced from natural “alum shale,” where the binding of radon can reach 1000 Bq m−3 (EC 1999).

Phosphate rocks contain very high levels of radioactive material occurring naturally from the decay of uranium 238U and thorium 232Th (UNSCEAR 1993). With the advancement of the phosphate industry in Egypt, the production of phosphate fertilizers represents ways in which workers and the public are exposed to increased doses of radiation during handling, packing, and fertilizing (UNSCEAR 1993; IAEA 2004; Othman and Al-Masri 2007). The phosphate industry produces dust of high intensity that can affect the respiratory system between industrial workers. Some workers may experience an external gamma increase displayed at doses of up to 0.8 µGy h−1 (Steinhäusler 1984). Radon is more commonly found where uranium is relatively abundant in bedrock at the surface, often in granite, shale, and limestone.

Another source of radon pollution comes from using granite, gypsum, marble, and ceramic in modern houses. Buildings are very important in people’s lives because they spend most of their time (about 80%) in them.

So far, there are no data on radon levels in the granitic rocks of the Central Eastern Desert of Egypt. Therefore, the main goal of this investigation is to determine the natural levels of Radon 222Rn radioactivity in these rocks. The evaluation of the collective impact of radiological hazard indices in terms of the annual effective dose equivalent, surface area, and mass radon exhalation rate in the rock samples should give important issue for the building material used from these five locations.

Material and methods

Study area

The studied granitic samples were collected from the areas of Gebel (mountain) El-Missikat MS (sixteen samples), El-Gidamy GD (seven samples), RieEl-Garra RG (six samples), El-Aradiya AR (five samples), and Kab-Amira KA (ten samples) (Fig. 1).

Fig. 1
figure 1

Sample location of the studied granitic rocks

Before sampling, the rocks were measured using the RS 230, to choose the samples having high radiation emission from uranium, thorium, and potassium.

The granites collected were medium to coarse-grained rocks composed of quartz, feldspar, and micas or hornblende. Microscopically, they were identified as granite. A brief account about the other rock types available in the study area, is given below:

El-Missikat MS area is covered by various types of igneous rocks. The main rocks were metavolcanics, synorogenic plutonic rocks, late orogenic granites, besides pegmatites, aplites, felsites, and quartz veins. The samples selected were the younger pink to red medium-grained granite intruding the older crystalline rocks.

El-Aradiya AR area: the oldest rock units are ophiolitic rocks, followed by metavolcanic rocks, then an early magmatic phase of diorites and granodiorites, followed by felsites, undeformed granites, and finally swarms of mineralized veins and post-granitic dykes. The samples collected were from the undeformed younger granites.

Gebel El-Gidamy GD: the rocks of this area are consisted of amphibolites, younger granite, and older granite rocks. The younger granite is the most dominant rock in Gebel El-Gidamy and the samples were collected from this younger granite.

Gebel RieEl-Garra RG: the rocks of this area are younger granites surrounded by older granitoids, the samples collected only from the younger granites.

Kab-Amira KA area: two concentric parts make up this pluton, consisting of an inner biotite granite and an outer garnetiferous two-mica granite, the samples used in this study were collected from the inner part only i.e. biotite granite.

Sample preparation

The 222Rn activity concentration was measured using a radon monitoring system (Alpha GUARD PQ 2000 PRO). The AlphaGUARD is a measuring system for the continuous determination of radon- and radon progeny concentration in air as well as selected climatic parameters. During operation, this detector measures air radiation by dispersing the gas through a large glass filter area installed in the ionization chamber. It is compatible with continuous radon steps and has a measurement speed of 2.0–2.0 × 106 Bq m−3 (AlphaGUARD 2012).

A three hundred gram of the sample was placed in a container. The instrument is connected to a modeling device via an alpha pump; placed in a continuous flow of 0.1 L min−1. The Alpha GUARD monitor was in “flow” mode, radon concentrations were recorded every 10 min for 1 h. Before measurement, a radon (222Rn) measuring device was placed in this vessel, measured, and repeated several times to get the exact order. The background noise is subtracted from the concentration of 222Rn in each sample (Salaheldin et al. 2020a, b). Radon concentrations using the AlphaGUARD detector are between 2 and 2 000 000 Bq m−3.

Activity concentration

The radon activity concentration CRn(t) at time t during the radon growth inside the chamber. The equilibrium radon concentration Ceq for each sample was calculated from the following Eq. (1),

$${\mathrm C}_{\mathrm{Rn}}={\mathrm C}_{\mathrm{eq}}(1-\exp\left(-\mathrm\lambda\times\mathrm t\right))$$
(1)

where CRn, radon concentration measured at time t (Bq m−3); t, the accumulation time of the released radon from a sample; and Ceq, radon concentration equilibrium (Bq m−3) (Hassan 2014; Salaheldin 2021).

Radiological parameter estimation

The radon surface exhalation rate and mass exhalation rate for any sample, Ea and Em were calculated using the following formula, Eq. (2, 3) (Harb et al. 2016, Salaheldin et al. 2020a, b, ICRP 1993).

$${\mathrm E}_{\mathrm a}=\frac{{\mathrm C}_{\mathrm{eq}}\times\mathrm V\times\mathrm\lambda}{\mathrm S}$$
(2)
$${\mathrm E}_{\mathrm m}=\frac{{\mathrm C}_{\mathrm{eq}}\times\mathrm V\times\mathrm\lambda}{\mathrm M}$$
(3)

where V is the volume of the radon chamber, λ is the decay constant of radon (2.1 × 10–6 s−1), S is the total surface area of the sample surface, and M is the sample weight (kg).

The emanation coefficient is calculated by Eq. (4) (Salaheldin 2021; Raslan et al. 2010).

$$\mathrm f=\frac{{\mathrm C}_{\mathrm{eq}}\times\mathrm V}{{\mathrm A}_{\mathrm{Ra}}\times\mathrm M}$$
(4)

where f is the emanation coefficient of radon, ARa is the concentration of radium measured using HPGe detector (Bq kg−1), V is gas empty volume (volume of chamber minus the sample volume, m3).

The annual effective dose equivalent calculated from the concentration of radon using UNSCEAR recommended conversion factor of 9 nSv (Bq h m−3)−1 (UNSCEAR 2000). Assuming 7000 h year−1 indoors and an equilibrium coefficient of 0.4 indoors (Salaheldin 2021), the effective dose for radon exposure during 1 year is calculated from Eq. (5) (Hassan et al. 2011):

$$\mathrm{The}\;\mathrm{annual}\;\mathrm{effective}\;\mathrm{dose}=\mathrm\epsilon\times{\mathrm f}_{\mathrm{Rn}}\times\mathrm T{\times\mathrm C}_{\mathrm{eq}}$$
(5)

Results and discussion

The measurements of surface radon concentration, emanation coefficient, mass radon exhalation rate, and annual effective dose are listed in Table 2. The concentration of 222Rn in El Missikat MS samples ranges from 49,597 ± 8773 to 236,029 ± 22,397 Bq m−3 with an average value 107,935 ± 14,359 Bq m−3. Surface radon exhalation rate (Ea) varied from (12.8 ± 2.1) × 10–3 to (61.0 ± 5.4) × 10–3 Bq m−2 s−1 with an average value of (27.9 ± 3.4) × 10–3 Bq m−2 s−1. The mass radon exhalation rate (Em) fluctuated between (20.9 ± 3.5) × 10–5 to (99.7 ± 8.8) × 10–5 Bq kg−1 s−1 with an average value of (45.6 ± 5.7) × 10–5 Bq kg−1 s−1. The emanation coefficient (f) changed from (69.2 ± 0.01) × 10–3 to (1802.4 ± 0.38) × 10–3 with a mean value of (703.9 ± 0.17) × 10–3. The annual effective dose indoors oscillated between 1249.86 and 5947.93 mSv year−1 with an average value of 2719.96 mSv year−1. The values of radon concentration for El-Gidamy GD and RieEl-Garra RG samples varied between 39,658 ± 8341 and 188,666 ± 20,545 Bq m−3 (mean value 92,280 ± 13,102 Bq m−3) and from 39,854 ± 7645 to 91,393 ± 13,429 Bq m−3 (mean value 56,142 ± 9702 Bq m−3) respectively. The surface and mass radon exhalation rate changed from (10.3 ± 2.0) × 10–3 to (48.8 ± 4.9) × 10–3 Bq m−2 s−1 (mean value (23.9 ± 3.2) × 10–3 Bq m−2 s−1) and from (16.7 ± 3.3) × 10–5 to (79.7 ± 8.1) × 10–5 Bq kg−1 s−1 (mean value (38.9 ± 5.2) × 10–5 Bq kg−1 s−1) for El-Gidamy GD samples, respectively. RieEl-Garra RG samples varied from (10.3 ± 1.8) × 10–3 to (23.6 ± 3.2) × 10–3 Bq m−2 s−1 (mean value (14.5 ± 2.3) × 10–3 Bq m−2 s−1) and from (16.8 ± 3.0) × 10–5 to (38.6 ± 5.3) × 10−5 Bq kg−1 s−1 (mean value (23.7 ± 3.8) × 10–5 Bq kg−1 s−1) respectively. The emanation coefficient (f) changed from (362.0 ± 0.10) × 10–3 to (1161.1 ± 0.24) × 10–3 and from (673.5 ± 0.20) × 10–3 to (1495.9 ± 0.37) × 10–3 with an average (735.9 ± 0.19) × 10–3 and (974.3 ± 0.27) × 10–3 for El-Gidamy and RieEl-Garra samples, respectively. The annual effective dose indoors varied from 999.39 to 4754.39 mSv year−1 and from 1004.33 to 2303.09 mSv year−1 (average 2325.45 mSv year−1 and 1414.77 mSv year−1) for El-Gidamy and RieEl-Garra granites, respectively. The radon concentration for El-Aradiya AR and Kab-Amira KA samples span the range from 45,951 ± 9488 Bq m−3 to 121,935 ± 14,869 Bq m−3 (mean value 83,143 ± 12,262 Bq m−3) and 33,209 ± 6989 Bq m−3 to 251,398 ± 25,759 Bq m−3(mean value 69,141 ± 10,786 Bq m−3), respectively. The surface and mass radon exhalation rate for El-Aradiya granites changed from (11.9 ± 2.3) × 10–3 to (31.5 ± 3.6) × 10–3 Bq m−2 s−1 (mean value (21.5 ± 2.9) × 10–3 Bq m−2 s−1) and from (19.4 ± 3.7) × 10–5 to (51.5 ± 5.9) × 10–5 Bq kg−1 s−1 (mean value (35.1 ± 4.8) × 10–5 Bq kg−1 s−1) respectively, while the variation in Kab-Amira samples was from (8.6 ± 1.7) × 10–3 to (65.0 ± 6.2) × 10–3 Bq m−2 s−1 (mean value (17.9 ± 2.6) × 10–3 Bq m−2 s−1) and from (14.0 ± 2.8) × 10–5 to (106.2 ± 10.2) × 10–5 Bq kg−1 s−1 (mean value (29.2 ± 4.3) × 10–5 Bq kg−1 s−1), respectively.

Table 2 222Rn concentration levels, 222Rn exhalation rate, emanation coefficient, and annual effective dose in all studied samples

The emanation coefficient (f) varied from (206.8 ± 0.06) × 10–3 to (1873.1 ± 0.42) × 10–3 and from (496.3 ± 0.17) × 10–3 to (2415.4 ± 0.75) × 10–3 (average (892.2 ± 0.22) × 10–3 and (1334.1 ± 0.39) × 10–3) for El-Aradiya and Kab-Amira granites, respectively.

The annual effective dose indoors fluctuated between 1157.97 and 3072.77 mSv year−1 and from 836.86 to 6335.24 mSv year−1 (average 2095.21 mSv year−1 and 1742.36 mSv year−1) for the rocks of El-Aradiya and Kab-Amira, respectively. The radon concentration for all studied samples is higher than the national reference level of 300 Bq m−3 recommended by ICRP (ICRP 2014). The International Commission on Radiological Protection recommended the radon concentration range to be from 200 to 600 and 500 to 1500 Bq m−3 for public and occupation, respectively. The effective annual dose is higher than the world average value of 1.1 and 20 mSv year−1, for public and occupation, respectively (Gupta et al. 2011). Therefore, the investigated rocks can cause health risks for workers handling these rocks and when used as building material as well.

The commission recommended upper values for the derived reference levels of 600 Bq m−3 for homes and 1500 Bq m−3 for workplaces. In response to its review of radon epidemiology, and the increase of the nominal risk coefficient by approximately a factor of 2 (Publication 115, Tirmarche et al. 2010), the Commission reduced the upper reference level for homes to 300 Bq m−3 in the associated statement on radon. The radon concentration of 300 Bq m−3 in homes correspond to an annual dose of approximately 10 mSv using the dose conversion convention, based on the revised nominal risk coefficient. The Statement on Radon also referred to a level of 1000 Bq m−3 as an entry point for applying occupational radiological protection requirements, replacing the upper reference level of 1500 Bq m−3 (ICRP 2014).

El-Gamal et al. (2018) reported the mean value of the activity concentration (measured with HPGe detector) of 226Ra to be 572.32 ± 279.43 Bq kg−1, 169.34 ± 24.50 Bq kg−1, 77.45 ± 5.19 Bq kg−1, 178.57 ± 58.86 Bq kg−1, and 82 ± 17.96 Bq kg−1 for the granitic samples of El-Missikat MS, El-Gidamy GD, RieEl-Garra RG, El-Aradiya AR, and Kab-Amira KA, respectively. The correlations between 222Rn versus 226Ra were plotted in Fig. 2.

Fig. 2
figure 2

Correlations between 222Rn with 226Ra concentration of the samples under investigation a El-Gidamy (GD), Ria El-Garra (RG), El-Aradiya (AR), and Kab Amira (KA); b Gebel El-Missikat

The results showed strong relationships between most values of 222Rn versus 226Ra with correlation coefficient values (R2 = 0.8349, 0.6307, and 0.5919) for El-Missikat, Kab-Amira, and El-Gidamy, respectively. The interrelation between 226Ra and 222Rn displayed a strong positive correlation, due to their existence in the same decay series (238U) where 222Rn is a product of the 226Ra. The correlation coefficient values are 0.3624 and 0.1011 for RieEl-Garra RG and El-Aradiya AR, respectively, i.e. a moderate to weak positive correlation, respectively.

Figure 3 shows the average, minimum and maximum values for radon activity concentration for the granites of the five areas (El-Missikat MS, El-Gidamy GD, Kab-Amira KA, RieEl-Garra RG, and Kab-Amira KA).

Fig. 3
figure 3

The activity concentrations of radon for different areas under investigation (Gebel El-Missikat (MS), El-Gidamy (GD), Ria El-Garra (RG), El-Aradiya (AR), and Kab Amira(KA))

Two studies showed definitive evidence for higher levels of radon from other areas in Egypt (Salaheldin 2021, 2020). Salaheldin (2021) reported Radon concentration in Abu El-Hassan area in the northeastern part of Egypt, where the values are higher than the global recommended. The average 222Rn concentration was 10,900 ± 3480 Bq m−3. The mean surface and mass Radon exhalation rate were (2.80 ± 0.89) × 10–3 Bq m−2 s−1 and (4.57 ± 1.46) × 10–5 Bq kg−1 s−1. The mean emanation coefficient was 115 ± 0.058 × 10–3. Salaheldin et al. (2020a, b) studied radon exposure in the Neoproterozoic rocks of Wadi Um Huytat in the Central-Eastern Desert of Egypt, where the average concentration of 222Rn and the radon exhalation rate were higher than the world recommended values. Those studies showed positive evidence for higher levels of radon from other areas in Egypt (Salaheldin 2021, 2020), confirming the radon health risks to miners who breathed radon for a period of years.

Table 3 shows the average, sum, and variance values of 222Rn concentration for the studied granitic samples from the five sites (MS, GD, RG, AR, and KA), where the largest variance values are noticed in the KA area (4.29E + 09) and the lowest in RG area (3.53E + 08). The MS area is represented by 16 samples with an average value of 107,935.1 and variance 4.026E + 09. The GD area (7 samples) has an average value of 92,279.58, and variance 3.38E + 09. The RG area (6 samples) shows average value of 56,141.83 and variance 3.53E + 08. The AR area (5 samples) contains an average value of 83,143.36 with variance 8.87E + 08. The KA area is represented by 10 samples with an average value of 69,141.39 and variance 4.29E + 09.The statistical study of the granitic samples under investigation, using the ANOVA (Table 3), showed that the radioactivity varies according to the site, as indicated from the P value where a high-significant difference exist between the five areas.

Table 3 Mean, sum, and variance values for the studied areas

Conclusion

Radon distribution in the granitic rocks of the Central-Eastern Desert, Egypt, was measured from five areas (El-Missikat MS, El-Gidamy GD, Kab-Amira KA, RieEl-Garra RG, and Kab-Amira KA). In order to assess the radiologic effect of radon concentrations, the activity concentrations have been measured and the radiological hazard parameters were calculated.

When considering the average, minimum and maximum radon activity concentration for the granites of the five areas, El-Missikat MS has the highest average concentration of 222Rn with concentrations 107,935 ± 14,359 Bq m−3, while RieEl-Garra RG has the lowest 222Rn (56,142 ± 9702 Bq m−3). Kab-Amira KA has the highest concentration of 222Rn (251,398 ± 25,759 Bq m−3) while RieEl-Garra RG has the lowest 222Rn (91,393 ± 13,429 Bq m−3). El-Missikat MS has the highest concentration of 222Rn (49,597 ± 8773 Bq m−3), whereas Kab-Amira KA has the lowest 222Rn (33,209 ± 6989 Bq m−3).

Based on these results, we can conclude that the studied granitic rocks from the Central Eastern Desert cause high risk to human, so we recommend not to use these high bearing radon rocks as ornamental, in decoration works, as building material as well as in the manufacture of household appliances. Radon gas release presents a particular hazard as it can escape from the solid rock through pores and cracks to reach the surface. It may then collect in buildings and present risks to people. The results of this study, besides others, can be used to draw Egypt’s radiological map, as reference for radon values within the granitic rocks of the Central Eastern Desert of Egypt.