Introduction

Uranium and thorium are radioactive substances that can be found in nature, like soil [1,2,3], sediment [4, 5], water [6,7,8], and building materials [9,10,11]. The U and Th isotopes typically are employed in various tracer investigations to assess their influence on the environment and geology. The major three isotopes of natural uranium are 238U (T1/2 = 4.468 × 109 a), 235U (T1/2 = 704 × 106 a), and 234U (T1/2 = 2.455 × 105 a), all of which are alpha (α) emitters [12]. Thorium has three primary isotopes, all of which are alpha emitters: 232Th (T1/2 = 1.402 × 1010 a), 230Th (T1/2 = 7.54 × 104 a) [13], and 228Th (T1/2 = 1.9126 a) [14].

Higher levels of radioactivity in the environment are connected with increased radiation damage and risk to humans, as evidenced by kidney damage, mutagenicity, bladder and kidney cancer, leukaemia, testis cancer, and lung cancer [15,16,17]. There are several residential locations on the earth with high natural background radiation (HNBR), such as Guarapari in Brazil, Yangjiang in China, Kerala in India, and Ramsar in Iran. In HNBRs regions like Ramsar, the yearly effective dosage rate can approach 260 mSv a−1[2].

Uranium and thorium transfer from one environmental compartment to another in biological systems such as vegetables are dependent on the availability of nuclides in the soil as well as the rate of loss from the interior structure of leaves following translocation. This translocation can add the U and Th to the human food chain. Studies on the HNBR area are necessary because clearer findings about the true risks of radiation should obtained and reasonable levels of radiation protection promoted. Although radiobiological and epidemiological studies in HNBR locations have yet to demonstrate any indication of a substantial increase in health damage compared to that in normal background areas, it is necessary to quantify the U and Th concentrations in biological samples [18], particularly if the linear threshold model is used for dose assessments.

Alpha spectrometry could be used to determine the radioactivity of alpha emitter nuclide in environmental samples. Examples include actinide determination on soil samples and analysis of water samples downstream of uranium mining operations for uranium series radionuclides. The technique is also used for internal dosimetry for workers using the analysis of biological samples such as urine [19].

In recent years, natural radionuclides (U and Th) series determination methods in biological samples have been successfully used in various studies around the world [20,21,22,23,24,25,26,27]. There are different methods for the determination of Th series radionuclides in biological samples, but alpha spectrometry is the most accurate and low minimum detection level (MDL) in comparison with other methods like gamma spectrometry and ICP-mass for short-lived radionuclides. However, in the determination of long-lived 234U, 235U, and 238U, the minimum detection level (MDL) of ICP-MS is better than alpha spectrometry.

This study aimed to determine uranium, thorium, radium, and polonium isotopes in biological samples to provide some basic data on residents living in the HNBR area. For these purposes, the activity concentrations of the most important α -emitters, 238U series (234U, 230Th, 226Ra, and 218Po) and 232Th series (232Th, 228Th, 228Ra, and 224Ra) were measured in biological samples collected in Ramsar, Iran.

Materials and methods

Study area

Ramsar is located at 36° 53′ N, 50° 41′ E, and has an average elevation of 20 m a.s.l. This city is situated in the Caspian Sea’s southern region. Ramesar area has a high natural background radiation (HNBR) area in the Talesh Mahaleh, Ramak, Sadat Mahale, and Chaparsar districts (Fig. 1) [4]. In 2016, the Ramsar area was home to a population of 80,000 residents. However, only approximately 20,000 of them chose to reside in the HNBR zones within the area.

Fig. 1
figure 1

The location of biological sampling in high natural background radiation (HNBR) area

Full size image

Samples preparation

The sample preparation includes sampling, dissolution, radiochemical separation, and sample counting [2]. Four urine samples from residency families with ages more than 50 years, each with a volume of 500 ml, were taken from the residence of the HNBR area. After that, 100 ml urine samples were selected to separate possess of each element. 232U (T1/2 = 68.9 a), 229Th (T1/2 = 7880 a), 242Pu (T1/2 = 3.73 × 105 a) and 225Ra (T1/2 = 14.9 d) [13] tracer were added to three blank samples and urine samples as a tracer to measure the separation efficiency. The blank samples were analyzed parallel with urine samples to calculate the difference recovery percentage between blank samples and urine samples. So, there was no significant separation among the alpha-emitting isotopes from the different elements. Also, the concentration of all tracers was corrected for the date of application. In this process, ammonium hydrogen phosphate solution (3 ml 1.3 mol l−1) and 1 ml 1.25 mol l−1 calcium nitrate (50 mg Ca) were added to the solution. In the next step, by adding concentrated ammonium hydroxide and creating an alkaline environment (pH∼9), the actinides in the urine were precipitated as calcium phosphate. The samples were centrifuged at 3000 rpm for ∼5 min. The precipitate was centrifuged again at 3000 rpm for about 5 min after being washed once with approximately 20–30 ml of distilled water, and the supernatant was discarded. The obtained sediment, which contained iron phosphate and actinides, was dissolved in concentrated nitric acid (HNO3 10 ml 6 mol l−1) and 10 ml aluminum nitrate (Al(NO3)3 2 mol l−1) solution. The obtained solution was evaporated. Actinide elements were oxidised by adding 0.5 ml 1.5 mol l−1 sodium nitrite (NaNO2) in an acidic environment. In the oxidation process, the actinides of Pu(III) and U(IV) increased to Pu (IV) and U(VI), respectively.

In the next step, the separation of actinides using a chromatography column containing exchange anionic resin (Dowex 8 × 1) was carried out. In this process, Th, Pu, and U actinides were separated individually from anionic resin and collected in a separate container. After pre-conditioning the resins with 10 ml of 8 mol l−1 HNO3, the sample is loaded onto the columns at a flow rate of ∼1 ml min−1 to extract and separate Th, Pu, U and Ra. The columns are rinsed with ∼10 ml of 8 mol l−1 HNO3 until the effluent turns colorless. The resin in the column was removed, divided into portions, and reloaded into new columns. The resin was blended and then divided equally for individual U, Th Pu and Ra determinations. In addition, the resin fraction was equivalent to about 30 ml of urine. The resin column is rinsed with 10 ml of 12 mol l−1 HCl to remove Th, and then the Pu isotopes are eluted into a 50 ml centrifuge tube using 15 ml of 2 mol l−1 HNO3. The U isotopes are stripped from the resin column with 1.0 ml of 4 mol l−1 HCl followed by 15 ml of 0.01 mol l−1 HCl. Consequently, the Ra was eluted in 5 ml of 3 mol l−1 HNO3.

However, the preparation of a good quality α-source which should be thin, weightless, uniform, and homogenous on a suitable flat and smooth substrate, is one of the most important prerequisites to obtain a good quality α-spectrum with high resolution, and a small low energy tail contribution [28]. The method used in this research was to create sediment (5 μm) through deposition of sediment by gravity sedimentation. Th and U as Phosphate and Pu as Fluoride sediment were co-precipitated [29, 30]. It was obtained uniformly on the membrane filters. Actinide filters were counted by alpha spectrometry for around 16 h; however, depending on the required detection limit, shorter count durations of 1 h can also be achieved for emergency response samples utilizing higher-level tracers [31]. Furthermore, the measurement of 228Ra concentration was conducted to assess the secular equilibrium between 228Ra and 228Ac.

Samples counting

The Alpha spectrometry system (ALPHAQUATTRO spectroscopy, SILENA, Italy) was applied with a multi-channel analyzer alpha spectroscopy workstation. It incorporates four independent high-quality modular vacuum chambers equipped with high-performance silicon ion implanted charged particle detectors, their associated electronics, a multichannel pulse-height analyzer, and a test pulser [32]. The effective cross-section of the detector was 450 mm2 with an ineffective thickness of 100 μm. The detector has a resolution of approximately 30 keV (FWHM) and a counting efficiency of 20% for a 5 mm source-detector distance calibrated for 3 to 8 MeV energy. Genie 2000 software was applied for data acquisition and count processing. The counting time was 16 h for each sample, and the working voltage of the detector was + 20 V with 1024 channels. In addition, the uncertainty components to calculate the activity concentration combined uncertainty are presented in Table 1. The system was calibrated for energy, resolution, and counting effectiveness using a standard electrodeposited mixed source of 241Am, 239Pu, and 244Cm to ensure accurate identification and quantification of alpha-emitting radionuclides. In addition, the sample counting sources and the standard calibration source were physically and chemically identical to ensure the accuracy of the results. The uncertainty of the source activities was 4% at the 95% confidence level.

Table 1 Combined standard uncertainties for activity measurements
Full size table

The activity concentration was calculated by applying the following equation:

$${A}_{238}=\frac{\frac{{C}_{238}{ V}_{232}{ A}_{232}{ I}_{232}}{{I}_{238({C}_{232}-{C}_{228})}}-\frac{{C}_{238B}{ V}_{232B }{A}_{232}{ I}_{232}}{{I}_{238({C}_{232B}-{C}_{228B})}}}{V}$$
(1)

where C is counts, V is the tracer volume (l), A is the tracer special activity (Bq/l), I is the decay probability, and V is the sample volume (l). The B index was related to the blank sample.

The minimum detection limit (MDL) was determined using the following equation:

$$MDL\left[ {\frac{{Bq}}{g}} \right]_{{238}} = \frac{{2.71 + 4.66 \times \sigma _{B} }}{{T_{B} \times \varepsilon \times m_{B} \times 2.22}}$$
(2)

where TB is blank counting time (s), \(\varepsilon\) is measurement system efficiency, mB is mass of blank sample (g), and σB is:

$${\sigma }_{BLK}=\frac{\sqrt{{C}_{238 B}} \times {m}_{238 B} \times {A}_{232} \times {I}_{232}}{{I}_{238} \times ({C}_{232 B }- {C}_{228B})}$$
(3)

where σB estimated from the standard deviation of multiple blank measurements, and other parameters were explained in the above section.

The 228Ra radionuclide was measured by HPGe gamma spectrometry system according to the provirus published paper explained method. The gamma lines of 911.2 keV and 969.0 keV emitted from 228Ac radionuclide were applied to activity calculation (See the method [10]).

Interferences in the spectrum

The major spectrum interferences for measuring 224Ra and 226Ra are predicted to be naturally existing radionuclides in biological samples, such as 228Th, 230Th, and 234U, as well as radium isotope progeny nuclides (226Ra, and 224Ra). Dai et al. [33] report revealed that Th and U distinguish in the urine samples process for 226Ra, employing chromatographic separation stages using coupled cation/anion exchange columns to ensure efficient Th and U interferences eradication. Because the current process used the same cation/anion exchange separation strategy, the spectrum interferences of 230Th/234U on 226Ra and 228Th on 224Ra were negligible.

Because radium isotopes (226Ra and 224Ra) cannot be chemically separated from each other, possible spectrum interferences between these isotopes and their offspring nuclides must be carefully examined to choose adequate regions of interest (ROI) for calculating radium activity. For the 226Ra peak at 4.78 MeV and the 224Ra peak at 5.68 MeV in biological samples, high-energy resolution of the FWHM (full-width at half maximum) ranging from 10 to 30 keV was indicated.

Results and discussion

Activity concentration

The activity concentration value of the 234U, 230Th, 226Ra, 228Th, 228Ra, 224Ra and 239Pu measured in biological (Urine) samples and sample codes area is given in Table 2. The region of interest area technique was applied to measure low-activity concentration radionuclides. The minimum and maximum value of 234U, 230Th, 226Ra,232Th,228Th,228Ra, 224Ra and 239Pu were obtained 8011 to 1079 mBq l−1, 379 to 681 mBq l−1, 4815 to 8830 mBq l−1, 4.98 to 7.25 mBq l−1, 6.27 to 13.15 mBq l−1, 7.18 to 12.17 mBq l−1, 4.25 to 8.95 mBq l−1, and < MDL, respectively. The order of average concentrations of radionuclides recorded in these samples was as follows; 234U > 226Ra > 230Th > 228Ra > 228Th > 224Ra > 232Th > 239Pu. 234U and 226Ra were significantly higher than the other measured radionuclides, while in the case of 239Pu, the average concentrations were lower than MDL (= 5 mBq l−1). Table 3 presents the average ratios of 224Ra/226Ra, 228Ra/226Ra, 230Th/232Th, and 228Th/232Th in biological samples were calculated. The higher average value was the 230Th/232Th ratio, which shows the difference concentration value of the 228U series (230Th) and 232Th series. The lower average value was obtained at the 224Ra/226Ra ratio that 224Ra is from the 232Th series with low concentration and 226Ra is from the 228U series with high concentration in study samples. As per some investigation in the study area, 238U daughter (226Ra) concentration was reported to be higher than the worldwide average [34,35,36].

Table 2 The activity concentration of 238U and 232Th series in biological samples of the study area
Full size table
Table 3 The activity concentration ratio of 224Ra/226Ra, 228Ra/226Ra, 230Th/232Th, 228Th/232Th; and 230Th/234U, 226Ra/234U, 228Ra/232Th and 224Ra/232Th in biological samples of study area
Full size table

Table 3 indicated the ratios of 238U daughters/238U and 232Th daughters/232Th to show disequilibrium in the 238U and 232Th decay chains. As demonstrated in Table 3, the average equilibrium ratio in the 232Th series and its progeny was found to be greater than one, whereas the equilibrium ratio in the 238U series and its progeny was less than one. Those results show that the radionuclide retention and excretion biokinetic results from this study are consistent with the ICRP models [37].

The comparison of the measured activity concentration of all radionuclides showed that the activity concentration of 234U was higher than other radionuclides. The activity concentration of 226Ra in spring water was reported to be more than 146,000 mBq L−1 in the study area [34]. Therefore, 226Ra with a high concentration of water can enter the human organs through the food chain.

The activity concentration percentage of 238U series (234U, 228Th, 226Ra, and 224Ra) and 232Th series (232Th, 228Th, 228Ra, and 224Ra) radionuclides in biological samples are presented in Fig. 2-a and Fig. 2b, respectively. As shown in Fig. 2a, more than 95% of activity concentration was related to two elements of 234U and 226Ra. Whiles in the 232Th series, the activity concentration percentage was distributed between 15 to 40% in measured radionuclides (Fig. 2b).

Fig. 2
figure 2

The activity concentration percentage of a 238U series (234U, 230Th, 226Ra and 218Po) and b 232Th series (232Th, 228Th, 228Ra and 224Ra) in biological samples

Full size image

Statistical interpretation

To estimate, the correlations between the activity concentration of 226Ra–234U, 226Ra–228Ra, and 232Th–224Ra, 232Th–228Th, 232Th–230Th in biological samples, Spearman correlation analysis was applied. This kind of investigation is beneficial for determining whether there are any probable links between observed activity concentrations of all primordial radionuclides. According to Spearman’s correlation, a strong positive correlation was observed between 234U and 226Ra (r = 0.977). The correlation coefficient between two sets of variables was determined, and the significance level was decided depending on the sample size. A highly significant correlation was with p < 0.023, whereas, p > 0.05 was not a significant correlation. For the highly significant correlations regression line (with R2 = 95.47%) and the prediction were plotted in Fig. 3a. While there is no positive correlation between them and 228Ra and 226Ra. This correlation is negative (r = -0.501) with a p-value of more than 0.05 (p = 0.488) and an R-square value of R2 = 26.18%. The regression and prediction were plotted in Fig. 3b.

Fig. 3
figure 3

The Spearman’s correlation and prediction plot of a 234U and 226Ra, b 228Ra and 226Ra

Full size image

Figure 4a shows the relation between 232Th concentrations and 224Ra as a linear model plot. As can be seen from the Figure, our results show a direct significant relationship between 224Ra levels and 232Th (r = 0.963, p-value < 0.037). The R-square value (92.77%) and predication plot are shown a strong correlation between 232Th and 224Ra concentration. These results are also not surprising since both radionuclides are part of the same decay series.

Fig. 4
figure 4figure 4

The Spearman’s correlation and prediction plot of a 224Ra and 232Th, b 228Th and 232Th, c 230Th and 232Th

Full size image

The narrow width of the prediction plot shows this relation (see Fig. 4b). This correlation was expected due to the sameness of the radionuclides in the series and the absence of gaseous elements between these two radionuclides in the decay chain.

The correlation between 232 and 230Th concentrations fitted in the liner model and presented in Fig. 4-c. As can be seen, the measured concentration indicates that there is no significant relation between 232 and 230Th. A p-value of more than 0.05 (0.395) and r = 65% indicate this weak relation. The correlation between radionuclides is presented in Table 4.

Table 4 The concentration correlation between measured radionuclides
Full size table

Within the 238U and 232Th decay chains, there exist certain parent and daughter relationships that exhibit relatively weak correlations. Furthermore, there are some surprisingly strong and unexpected correlations between isotopes across the 238U and 232Th chains. These discrepant relationships are generally explained by invoking the rationale of chemical effects within the human body and by local geochemical fractionation processes. The alternative rationale for the presence or lack of correlations among the radionuclides within and across the 238U and 232Th decay chains is due to the limited statistical strength of the study with only n = 4 test donors. With these interpretations, we conclude that the importance of the first scenario (biokenetik) was more than that of the second scenario (limited statistical strength). The strong correlations found in Figs. 4 and 5 indicate, that the results in Table 3 indicate a significant correlation. In addition, the cluster similarity diagram of 234U, 228Th, 226Ra, 224Ra, 232Th, 228Th, 228Ra, 224Ra and 239Pu are presented in Fig. 5. The high similarity in concentration trend was observed in two groups of radionuclides. Those radionuclide groups are indicated in Fig. 5.

Fig. 5
figure 5

The cluster similarity diagram of measured radionuclides in the biological sample

Full size image

In addition, Fig. 5 and Table 4 results indicate that 228Ra was more correlated with the 238U decay chain than the 232Th chain. Because of chemical reactions within the human body and local geochemical fractionation, this phenomenon can be explained.

Conclusion

In this study, 238U series (234U, 230Th, and 226Ra), 232Th series (232Th, 228Th, 228Ra, and 224Ra), and 239Pu of HNBR (Ramsar) were analyzed in biological samples (urine) by radiochemical separation followed by alpha-spectrometry spectra. The significant concentration of 234U and 226Ra was measured in biological samples (urine), while 239Pu concentration was not determined and the concentration level was less than MDL. Also, the activity concentration ratio of 224Ra/226Ra, 228Ra/226Ra, 230Th/232Th and 228Th/232Th was calculated. The significant ratio value was obtained for 230Th/232Th radionuclides. Overall, the analysis of activity concentration ratios of radionuclides in decay series provides valuable information about the processes controlling their distribution in the environment, their mobility, and the timescales involved in their radioactive decay. The statistical analysis was carried out to determine the correlation between radioelements. The correlation results indicate that the radionuclide retention and excretion biokinetic ICRP model is the dominant process of being in the same group in a radioactive decay series. According to the findings of this results living in areas with high natural radiation background (HNRB) can pose unique health and safety considerations for residents. Residents who are living in these areas are suggested to undergo periodic health checks.