1 Introduction

Humans are continuously exposed to natural radiation which comes from cosmogenic radionuclides and primordial radionuclides [1]. The cosmogenic radionuclides are continuously produced in the upper part of the atmosphere by the interaction of the cosmic radiation with atoms or molecules. The primordial radionuclides are the uranium series with as parent the 238U, 235U series or actinium series, thorium (232Th) series, and 40K which is a non-series of disintegration. When dealing with naturally occurring radioactive material (NORM), only the primordial radionuclides are of radiological interest. Depending on the geological origin of the raw materials (rocks, soil, and industrial products), the building materials may contain different amounts of natural radionuclides [2, 3]. As cement is most used in building material, it can become a health and environmental problem for the population. According to Mansoor et al., individuals spend 80% of their time at home or office indoor [4]. It is then important to estimate the natural radioactivity in cement.

There are different exposition ways of humans by radionuclides content in cement: internal and external exposure. Internal exposure is due to the inhalation of radon (222Rn) which emanates from the building material [5]. During the inhalation, radon may decay in the track respiratory conducting to the deposition of its progenies and becomes a permanent source of internal exposure [6]. External exposure is due to the emission of γ-rays by primordial radionuclides (226Ra, 232Th, and their daughters and 40K).

In Senegal, there are three factories of cement, but so far, no data concerning natural radioactivity of cement used are available. To fill this gap, the main objectives of this study are to assess the 226Ra, 232Th, and 40K activity concentrations and the radiological hazards in the types of cement used as a building material.

2 Materials and methods

2.1 Sample collection and preparation

Twenty cement samples were collected from building material suppliers and labeled properly. The net weight of the collected samples was 1 kg. The gray samples were manufactured by three different domestic cement factories (CEM I, CEM II, and CEM III). To remove moisture, the collected samples were dried at 105 °C for 24 h and then transferred to 60 mL of cylindrical containers with an internal diameter of 72 mm.

2.2 Gamma-ray spectrometry

The 226Ra, 232Th, and 40K activity concentrations were measured using a calibrated broad energy germanium (BEGe) detectors with a relative efficiency of 50% at 1332 keV and an energy resolution of 0.7 keV for 60 keV and 1.8 keV for 1332 keV gamma-rays for 241Am and 60Co, respectively. The detectors were calibrated to energy and efficiency using a mixed radionuclides standard containing 241Am, 210Pb, 139Ce, 137Cs, 113Sn, 109Cd, 88Y, 85Sr, 60Co, 57Co, and 51Cr. The sample analysis was conducted by Genie 2000 software program. The samples were measured at least 80,000 s. As in a planar BEGe detector, the sample is placed on the top of the detector, the coincidences summing were corrected using correction factors calculated by the Monte Carlo simulation, using the GESPECOR software package. For self-attenuation correction, a transmission bench was used for the low energies (below 100 keV). The principle of transmission bench consists to detect gamma-rays emitted by a collimated source of 133Ba and 109Cd passing through a container, then a sample within its container. A schematic diagram of the transmission bench is shown in Fig. 1.

Fig.1
figure 1

Diagram of transmission bench

Full size image

The attenuation coefficient (μl) was calculated following the Beer–Lambert equation:

$$N = N_{0} e^{ - \mu \rho x}$$
(1)

where N, N0, ρ, and x represent, respectively, the number of photons having passed through the sample within the geometry, the number of photons detected after crossing the empty geometry, the sample density, and the thickness.

The self-attenuation correction factor (Fatt) was calculated using the following equation:

$$F_{{{\text{att}}}} = \frac{{1 - e^{ - \mu \rho x} }}{\mu \rho x}$$
(2)

where x, μ, and ρ represent the thickness of the sample, the coefficient of attenuation, and the density, respectively.

As the matrix and the density of the measured samples were different from that of the calibration source, the self-attenuation of the measured samples and the calibration source is therefore different. The relative self-attenuation correction factor FCself (E) is determined by the ratio of the self-attenuation correction factor of the sample to that of the calibration source.

$${\text{FC}}_{{{\text{self}}}} \left( E \right) = \frac{{F_{{{\text{att}}.{\text{sam}}}} }}{{F_{{{\text{att}}.{\text{ref}}}} }}$$
(3)

where Fatt.sam and Fatt.ref are the self-attenuation correction factors of the sample the calibration source, respectively.

The self-attenuation correction factors calculated for ten energies from the transmission source according to their energies were plotted. The interpolated curve obtained was used to calculate the self-attenuation correction factor at other energies.

The 226Ra activity concentration has been estimated using the 186.2 keV gamma-ray. Since the 235U energy peak at 163 keV has not been identified, the gamma-ray interference between 226Ra at 186.2 keV with a peak of 235U at 185.7 keV was resolved taking into account the ratio of 238U/235U = 21.7 and the full energy peak of 234Th at 63.3 keV [7]. The activity concentration of 232Th was estimated using the full energy peaks of 228Ac at 911.2 keV and 208Tl at 583.2 keV taking into account the braching ratio. For 40 K activity concentration calculation, its peak at 1460.8 keV was used. The following equation was used to calculate the activity concentration of each radionuclide:

$$A\left( {{\text{Bqkg}}^{ - 1} } \right) = \frac{{N_{{{\text{net}}}} }}{{P\left( {E_{{\text{i}}} } \right) \times \varepsilon \left( {E_{{\text{i}}} } \right) \times t \times m \times F{}_{{\text{c}}}}}$$
(4)

where Nnet represents the net counts under the full absorption peak, P(Ei) the emission intensity, ε(Ei) the detector efficiency at energy Ei, t the counting time, m the sample mass, and Fc the corrective factors taking into account the radioactive decay, the self-attenuation, and the coincidence summing.

2.3 Dose parameters and radiological hazards indices

2.3.1 Absorbed gamma dose rate

At a height of 1 m above the ground, the indoor absorbed gamma dose rate (D) due to γ-ray emitted by 226Ra, 232Th, and 40 K is evaluated by the following equation [8, 9]:

$$D\left( {{\text{nGyh}}^{ - 1} } \right) = 0.92A_{{{\text{Ra}} - 226}} + 1.1A_{{{\text{Th}} - 232}} + 0.08A_{{{\text{K}} - 40}}$$
(5)

The conversion coefficients of 226Ra, 232Th, and 40K activity concentrations into dose for materials used as building materials are 0.92, 1.1, and 0.08 in nGy h−1/ Bq kg−1, respectively. These conversion coefficients were calculated by the Monte Carlo method using a standard room model of 2.8 m × 4 m × 5 m in which the wall width is 20 cm and the density 2.35 g.cm−3 [10].

2.3.2 Annual effective dose equivalent

The annual effective dose equivalent (AEDE) was calculated by the following equation [10]:

$${\text{AEDE}}\left( {\mu {\text{Svy}}^{ - 1} } \right) = D\left( {{\text{nGyh}}^{ - 1} } \right) \times 0.7\left( {{\text{SvGy}}^{ - 1} } \right) \times 0.8 \times 8760{\text{hy}}^{ - 1} \times 10^{ - 3}$$
(6)

The conversion coefficient of absorbed dose to an effective dose and the indoor occupancy factor used for the calculation of AEDE were 0.7 Sv Gy−1 and 0.8, respectively [10].

2.3.3 Annual gonadal dose equivalent

Since the gonads are organs of interest, the annual gonadal dose equivalent (AGDE) was calculated using the following equation [11,12,13,14]:

$${\text{AGDE}}\left( {\mu {\text{Sv}}.{\text{y}}^{ - 1} } \right) = 3.09A_{{{\text{Ra}} - 226}} + 4.18A_{{{\text{Th}} - 232}} + 0.314A_{{{\text{K}} - 40}}$$
(7)

2.3.4 Gamma index and alpha index

The gamma index (Iγ) was calculated to find out if the cements had met the safety requirements for building materials by the following equation [2, 8]:

$$I_{\gamma } = \frac{{A_{{{\text{Ra}} - 226}} }}{300} + \frac{{A_{{{\text{Th}} - 232}} }}{200} + \frac{{A_{{{\text{K}} - 40}} }}{3000}$$
(8)

For material used in a bulky amount, the exemption criterion and the upper dose limit are defined. The exemption criterion (0.3 mSv y−1) corresponds to Iγ ≤ 0.5 and the upper dose limit to Iγ ≤ 1, respectively [8].

The alpha index (Iα) corresponding to the excess alpha radiation due to the 222Rn inhalation from cement was calculated using the following equation [15]:

$$I_{\alpha } = \frac{{A_{{{\text{Ra}} - 226}} }}{200}$$
(9)

The Iα should not exceed the unity since the radon exhalation from cement could cause an indoor radon concentration greater than 200 Bq m−3 [8, 16].

2.3.5 Excess lifetime of cancer risk

The excess lifetime cancer risk (ELCR) was evaluated by the following formula [17, 18]:

$${\text{ELCR}} = {\text{AEDE}} \times {\text{RF}} \times {\text{DL}}$$
(10)

where RF is the fatal cancer risk per Sievert (0.05 Sv−1) and DL the life duration (70 y).

2.3.6 Effective dose rate to different organs and tissues

The effective dose rate to different organs and tissues (Dorg) was calculated using the following equation [19, 20]:

$$D_{{{\text{org}}}} \left( {\mu {\text{Svy}}^{ - 1} } \right) = {\text{AEDE}} \times {\text{CF}}$$
(11)

where CF is the conversion coefficient for the organ dose from air dose (Table 1).

Table 1 Conversion coefficient CF for different organs or tissues [19]
Full size table

3 Results and discussions

3.1 226Ra, 232Th, and 40K activity concentrations

Table 2 reports the range and average of 226Ra, 232Th, and 40K activity concentrations in the types of cement. The 226Ra activity concentration in cement samples varied from 7.09 Bq kg−1 to 150.25 Bq kg−1. The 232Th and 40K activity concentration in cement samples varied from 3.72 Bq kg−1 to 16.09 Bq kg−1 and from 48.67 Bq kg−1 to 133.89 Bq kg−1, respectively.

Table 2 Range and average with their standard deviation (SD) of 226Ra, 232Th, and 40K activity concentrations in the types of cement
Full size table

As shown in Table 2, the highest average value of the 226Ra and 232Th activity concentrations was found in CEM I. The CEM IV (white cement) presented the lowest average value of activity concentration of these two radionuclides. Unlike the 226Ra and 232Th, the highest average activity concentration of 40K was found in CEM IV and the lowest average value in CEM I. To make a comparison between the 226Ra, 232Th, and 40K average activity concentrations and the worldwide average values in building materials, their respective ratios were calculated. The worldwide average values used for comparison were 50 Bq kg−1, 50 Bq kg−1, and 500 Bq kg−1, respectively for 226Ra, 232Th, and 40K [21, 22]. In gray cements, the ratio ranges from 1.84 to 2.72 for 226Ra. Therefore, the 226Ra activity concentration in gray cements was then found two times higher than the worldwide average value. In white cement, the ratio was 0.16. Then, the 226Ra average activity concentration in white cement was lower than the worldwide average value cited above. The ratio ranges from 0.24 to 0.30 for 232Th and from 0.12 to 0.16 for 40K in the gray cement. In white cement, the ratios were 0.10 for 232Th and 0.24 for 40K. The 232Th and 40K average activity concentrations were then below their worldwide average values in all types of cement.

The 226Ra, 232Th, and 40K average activity concentrations in the gray and white cements were also compared with results from other countries (Table 3). The 226Ra average activity concentration in gray cement was lower than that of Albania [23] and China [24]. It was higher compared to other countries. For 232Th, its average activity concentration was below the cited results in other countries. Regarding the average activity concentration of 40K, this study was only lower than that of Albania [23]. The 226Ra and 232Th average activity concentrations found in white cement in this study had the lowest values. The result of the 40K activity concentration in Malaysia was only greater than this study [25].

Table 3 Comparison of 226Ra, 232Th, and 40K average activity concentrations of gray and white cements with those obtained in other countries
Full size table

3.2 Dose parameters and radiological hazard indices

The dose parameters and radiological hazards indices were evaluated and presented in Table 4.

Table 4 Average with their standard deviation of dose parameters and radiological hazards indices according to the types of cement
Full size table

The average values of the indoor absorbed gamma dose ranged from 22.09 ± 2.10 nGy h−1 (CEM IV) to 146.41 ± 6.98 nGy h−1 (CEM I). The indoor absorbed gamma dose rate in air of the types of cement exceeds the population-weighted average of 84 nGy h−1, except for CEM IV [10]. From Table 5, it can be seen that the average value of AEDE varied from 108.42 ± 10.32 µSv y−1 (CEM IV) to 718.72 ± 34.26 µSv y−1 (CEM I). The average value of AEDE of all types of cement is lower than the permissible limit which is 1000 µSv y−1. Among the natural radionuclides, the principal contributor to the indoor AEDE was the 226Ra, with a contribution of 84% in gray cement. The 226Ra was followed by 232Th (11%) and 40K (5%). In white cement, the principal contributor was 40K (43%), followed by 226Ra (34%) and 232Th (23%).

Table 5 Effective dose rate to different organs and tissues according to types of cement
Full size table

The average value of AGDE of the types of cement ranged from 81.84 ± 7.88 µSv y−1 (CEM IV) to 501.75 ± 23.15 µSv y−1 (CEM I). The average value of AGDE of CEM I, CEM II, and CEM III was higher than the worldwide average value of 300 µSv y−1 calculated by considering a house containing worldwide average activity concentrations of 226Ra, 232Th, and 40K in soil [10, 36].

Table 4 shows that the average value of the gamma Index (Iγ) and the alpha index (Iα) varied from 0.09 ± 0.01 (CEM IV) to 0.55 ± 0.03 (CEM I) and from 0.040 ± 0.004 (CEM IV) to 0.68 ± 0.04 (CEM I), respectively. The Iγ in all types of cement was in the range of the exemption criterion (Iγ < 0.5) except the CEM I which was slightly greater but below the recommended limit (Iγ = 1) [8]. The Iα of the types of cement was lower than the recommended limit value of 1 [8].

The ELCR for each type of cement is presented in Table 4 with an average value ranging from (0.38 ± 0.04) 10–3 (CEM IV) to (2.52 ± 0.12) 10–3 (CEM I). The excess lifetime cancer risk in all types of cement was higher than the worldwide average value which is 0.29 10–3 [10]. The values of ELCR equivalent to 1000, 100, 10, and 1 µSv y−1 will increase the risk of developing mortal cancer by 4%, 0.4%, 0.04%, and 0.004%, respectively [37, 38]. Even all ELCR calculated are higher than the worldwide value, the chances to increase the risks of cancer in life duration remains negligible.

The values of Dorg evaluated in different types of organs and tissues according to the types of cement shown in Table 5 were less than the set limit. The calculated values of Dorg showed that the testes were more sensitive to the radiations, and the ovaries were less sensitive.

3.3 Statistical analysis

3.3.1 Descriptive statistics

A descriptive statistics was performed to describe and also to have a better understanding of the statistical characteristic of the activity concentrations of the natural radionuclides (226Ra, 232Th, and 40K). As the 226Ra and 232Th activity concentrations in white cement samples are found very low compared to gray cements, then they cause a large deviation of the distribution of the activity concentrations. Therefore, they are not used for the statistical analysis of the data. The results of the statisitical anlysis are reported in Table 6.

Table 6 Descriptive statistics of 226Ra, 232Th, and 40K activity concentrations in cement samples
Full size table

The skewness, kurtosis, and the p value using the Shapiro–Wilks test were calculated to have a comprehensive understanding of the distribution of the data. The skewness characterizes the degree of asymmetry of the data [39]. In the theory of probability for a normal distribution, the skewness is equal to zero [40]. However, the data points are not always perfectly symmetric. The absolute magnitude of the ratio between the skewness and its standard error was calculated. If the ratio is less than two, the probability distribution can be assumed to be normally distributed. The distributions of 232Th and 40K activity concentrations in this study had a weak positive skewness, whereas a weak negative skewness was observed for the distribution of 226Ra activity concentration. The ratios between the skewness and its standard error of the distributions of 226Ra 232Th and 40K activity concentrations were found lower than two.

The kurtosis measures the extent of which data points cluster around the center of the distribution. For a normal distribution, the kurtosis is equal to zero. Negative kurtosis indicates that the data points are less clustered around the center, and the distribution has a thicker tail [41]. Unlike a negative kurtosis, the data points are clustered around the center of the distribution for a positive kurtosis and a thinner tail can be observed [41]. The ratio between the absolute magnitude and the standard error of the kurtosis must be less than two for a normal distribution. These ratios were found less than two for the distributions of 226Ra, 232Th, and 40K activity concentrations.

The Shapiro–Wilks was also used to test the normality of the data. In the Shapiro–Wilks test, if the p value is less than or equal to 0.05, the distribution will not be normal. The found results in Table 3 show that the p values were greater than 0.05 for the distributions of 226Ra, 232Th, and 40K activity concentrations.

The calculated values of the skewness, the kurtosis, and the p value found by the Shapiro–Wilks test showed that the 226Ra, 232Th, and 40K activity concentrations can be assumed normally distributed.

3.3.2 Pearson correlation

Pearson correlation analysis was performed to determine the interdependency and the strength of the relation between the natural radionuclides (226Ra, 232Th, and 40K) in cement samples. The results of the Pearson correlation analysis are shown in Table 7. Between 226Ra and 40K, a weak negative correlation was observed, whereas a high positive correlation was observed between 226Ra and 232Th. It indicates that 226Ra and 232Th have a common source which in general due to the mineralogical components [42]. A high negative correlation was found between 232Th and 40K which can be due to the mineral composition in cement that can affect the mobility of radionuclides [42].

Table 7 Pearson correlation matrix for variables
Full size table

4 Conclusion

The 226Ra, 232Th, and 40K activity concentrations were assessed in different types of cement available in Senegal by gamma-ray spectrometry. All activity concentrations were less than the worldwide values except the 226Ra in the gray cement. The Raeq was found less than the recommended value. The absorbed dose rate was greater than the worldwide average value of 84 nGy h−1 only in the gray cement. The AEDE was less than the recommended limit of 1 mSv y−1. In the CEM I, CEM II, and CEM III, the AGDE was higher compared to the worldwide average of 300 µSv y−1. The Iγ of CEM I was only greater than the exemption criterion, but it was found below the recommended limit. The Iα was slightly greater than the recommended exemption level value in the CEM I and CEM II. However, the alpha index of these two types of cement was below the recommended limit which indicates that the concentration of radon will be less than 200 Bq m−3. The calculated ELCR was greater than the worldwide value; however, the chances to increase the risks of cancer in lifetime remain negligible. The absorbed dose rate according to different types of organs or tissues showed that the testes are more radiosensitive and the higher values of different organs or tissues are found in the CEM I. The contribution of the radiological hazard from the cement under this study is not significant. However, the activity concentration of 226Ra was found greater compared to its worldwide average value, which could serve as an alert to the radiation protection authority.