Introduction

In 2006, the activities of naturally occurring radionuclides in the characteristic rocks of the Modane-Aussois region were measured to compare them with gamma-ray background measurements from the Laboratoire Souterrain de Modane (LSM) within the ILIAS (Integrated Large Infrastructures for Astroparticle Physics) project. The LSM is located in the Frejus highway tunnel connecting Italy and France 1200 m a.s.l. During that time, LSM’s main scientific program included dark matter detection and the neutrinoless double beta decay experiment (NEMO).

Geological setting

The investigated area belongs to the Aussois-Modane region in the French portion of the Western Alps, which is close to the “de Frejus” tunnel entrance connecting the French Modane with the Italian Bardonecchia. Its terrain is situated in the internal arc of the Western Alps formed by an internal zone of the Penninic nappes. The following tectonic units are distinguishable (from inner to outer): the Briançonnais, Subbriançonnais, Valaisan, Cheval Noir flysch (Penninc units), and Dauphinois domains (Fig. 1). A major tectonic boundary separates the Penninic units from the Dauphinois domain. This boundary is called the Pennine/Penninic Front [1–4], and is also referred to as the Pennine/Penninic Frontal Thrust [5–9], the Penninic Basal Contact [10, 11], and the Basal Penninic Fault [12]. In the past this thrust represented a suture zone between the Dauphinois and Briançonnais/Subbriançonnais units that was active during the Eocene period [13]. During the second deformation stage (Oligo-Miocene), the Roselend Thrust (Fig. 1) completely reworked the Penninic Front.

Fig. 1
figure 1

Simplified geological overview of the Western Alps after [10, 13, 15, 26] showing the location of the study area. 1 Molase basin (undifferentiated); 2 Jura belt (undifferentiated), Dauphinois Domain; 3 Triassic/Jurassic; 4 Basement (external massifs), Penninic Domain; 5 Cheval Noir unit; 6 North Penninic (Valaisan); 7 Middle Penninic (Briançonnais/Subbriançonnais); 8 South Penninic (Piemont-Liguria); 9 Austro-/Southalpine Domain (undifferentiated); 10 Penninic Basal Contact; 11 Roselend Thrust; 12 location of the study area

Full size image

In the French-Italian Western Alps, the internal arc originated from an oceanic accretionary wedge, which was incorporated in a wider orogenic wedge during the Paleogene collision [9, 14]. During the Neogene, this arc was subjected to widespread ductile to brittle extension.

The Aussois-Modane region is located in the Briançonnais unit (Middle Penninic [10, 15]. This area is characterized by a very complicated tectonic structure. Basically, the main units are the stratigraphic series from the Lower Triassic to the Paleocene (having very thick Triassic layers, compared to the remaining quite thin ones), and a thick overlaying series of calcschists (schistes lustre) with intercalations of the basic rocks (greenschists). Above Aussois and the “Monolithe” outcrop, the Dent Parah formed from dark Liassic shales and phyllites from other exotic series, blocks of which may have fallen into the slope.

Location of sampling points

Samples of calcschist (no. 1, Table 1), carbonaceous breccia (no. 2), limestone dolomite (no. 3), and dolomite (no. 4) were taken from the large rock exposure locally named the “Monolithe”, located at the holiday resort Aussois (Fig. 2a). Samples of marble (no. 5, Table 1) and quartzite (no. 6) were collected along the road in Modane-Aussois (Fig. 2b). All of the samples were carefully separated from their rock material, and had high purity without any superficial covers.

Table 1 Locations of the samples
Full size table
Fig. 2
figure 2

Location of samples: a outcrop “Monolithe”; b road Modane–Aussois

Full size image

Sample 1 is represented by grey calcschists that are very similar to calcite phyllite. The sample contains firmly tectonized thin layers of calcite and clay minerals. Due to tectonic movements, their surfaces can have a characteristic glittering, which might cause these rocks to be classified as schistes lustres [16].

Sample 2 shows the Upper Triassic carbonaceous breccia, which contains free variation forms of calcites, dolomites, quartz veins, and mainly gypsum evaporates. These rocks appear in many tectonic contacts.

Sample 3 is a Middle Triassic light grey limestone dolomite, mainly composed of dolomite with minor amounts of calcite and other clay minerals. The thin calcite veins cross dolomite in some sections. Although barely visible, the rock shows a coarse schist structure.

Sample 4 represents light grey (Middle Triassic) massive dolomites. The finely grained structure and lack of schist formation are characteristics of this rock.

Sample 5 is a Middle Triassic light grey massive marble, mainly calcite. The rock is strongly tectonized with a finely grained structure. It contains numerous examples of mutual differently-orientated mesofault surfaces, with distinctly visible tectonic striate and steps (associated with slickensides).

Sample 6 is a typical, strongly tectonized Lower Triassic thick-bedded massive quartz, and includes a slight occurrence of muscovite and sericite schists, some quartzite schists, and numerous tectonic slickensides with tectoglyphs.

Materials and methods

The rock samples, both after crushing and 6 months after collection, were measured in a Marinelli 450 using a gamma-ray GX3020 (Canberra Industries) spectrometry system. The system is based on a coaxial HPGe Extended Range detector, with 32% relative efficiency, energy resolutions of 0.86 keV at 122 keV and 1.76 keV at 1332 keV, and a detector bias voltage of 4000 V. The LabSOCS (Laboratory Sourceless Calibration Software) and Genie 2000 v.3 software packages were used for the efficiency calibration and the determination of radionuclides and their activities. The energy calibration of the spectrometer was performed using homogeneously dispersed 241Am, 109Cd, 139Ce, 57Co, 60Co, 137Cs, 113Sn, 85Sr, 88Y, and 203Hg radioisotopes in silicone resin (certificate source type MBSS2). The activities of the examined radionuclides were calculated from the following gamma transitions (energy in keV): 40K (1460.8), 208Tl (277.4, 510.8, 583.1, 860.6, and 2614.5), 212Pb (238.6 and 300.0), 214Pb (241.9, 295.2, and 351.9), 214Bi (609.3, 768.3, 1120.3, 1764.5, and 2204.2), 226Ra (186.2), and 228Ac (338.3, 911.6, 964.6, and 969.1). The total duration of a single measurement was 72 h.

Results and discussion

The results of the gamma-ray activities of the 40K, 208Tl, 212Pb, 214Pb, 214Bi, 226Ra, and 228Ac samples are provided in Table 2. The gamma-ray spectra are shown in Fig. 3.

Table 2 Activity concentrations of primordial radionuclides in the examined samples
Full size table
Fig. 3
figure 3

Gamma-ray spectra for all the samples. The characteristic gamma-ray emitters are marked above the corresponding peaks

Full size image

40K

The highest 40K activity concentrations occurred in sample 6 (quartzite) at 572 Bqkg−1, and sample 1 (calcschist) at 392 Bqkg−1 (Table 2, Fig. 4). The lowest activities occurred in samples 3 (limestone dolomite) and 5 (marble), at 18 and 78 Bqkg−1, respectively. Intermediate values were observed in sample 2 (carbonaceous breccia) at 111 Bqkg−1, and in sample 5 (dolomite) at 129 Bqkg−1. As shown in Fig. 4, an average value of 40K activity over all samples is 217 Bqkg−1. This value is relatively high compared to the average in typical carbonate rocks, 70 Bqkg−1. The higher value results from increased activity concentrations of potassium-40 in calcschist and quartzite. However, the average activity calculated only from samples 2, 3, 4, and 5 (“pure carbonates”), 84 Bqkg−1, is close to that observed in typical carbonate rocks (Fig. 4). It should be emphasized that the 40K activity concentrations in the samples, with the exception of quartzite, are below the average activity of 40K observed in typical soils, ca 400 Bqkg−1 [17, 18].

Fig. 4
figure 4

Activity concentrations of 40K. Thick solid line average 40K activity in typical carbonates; thin solid line average 40K activity using all the samples; thin dotted line = average activity of 40K using only samples 2, 3, 4, and 5

Full size image

232Th series (228Ac, 212Pb, 208Tl)

As seen in Table 2, radioactive equilibrium is achieved for each sample between progenies in the 232Th series. Since 228Ac is the second radionuclide in the thorium series, the activity concentration of 232Th can be assumed to be equal to the 228Ac activity. As seen in Fig. 5, the measured activities associated with the 232Th series were the highest in calcschist (18 Bqkg−1) and quartzite (10.4 Bqkg−1). By far, the lowest 232Th activity below 1 Bqkg−1 was observed in sample 3 (limestone dolomite). Intermediate values of 2.4, 3.8, and 5.5 Bqkg−1 were found in samples 5, 4, and 2, respectively (Table 2). The 232Th activity averaged over all the samples is 7 Bqkg−1 (Fig. 5). Considering activities from the “pure carbonates” (samples 2, 3, 4, and 5), the resulting average value of 3 Bqkg−1 is nearly three times lower than the average reported for carbonates, 8 Bqkg−1 [17]. Similar to the results for 40K, the average activity of 232Th in the investigated rocks is clearly below the averages of 232Th observed in a typical soil (37 Bqkg−1) and in the continental crust (44 Bqkg−1).

Fig. 5
figure 5

Activity concentrations of 232Th. Thick solid line average 232Th activity in typical carbonates; thin solid line average 232Th activity using all the samples; thin dotted line average activity of 232Th using only samples 2, 3, 4, and 5

Full size image

238U series (214Pb, 214Bi, 226Ra)

Activity concentrations of 238U were estimated assuming radioactive equilibrium in the 238U → 226Ra → 222Rn → 214Pb → 214Bi decay chain (occurring in the vast majority of minerals and rocks) [19]. As shown in Table 2, the mean values of the 226Ra activity calculated from 214Pb and 214Bi correspond to the values obtained directly from line 186.2 keV (226Ra). Due to a strong overlap between the line 186.2 keV and the line 185.7 keV (235U), and the relatively high background in this energy range, the corresponding cumulative uncertainties are naturally higher than those calculated from the averaged values for 214Pb and 214Bi. Therefore, in our work, the 226Ra (238U) estimate was determined from 214Pb and 214Bi activities.

Table 2 shows the highest activity concentrations of 238U in dolomite (29 Bqkg−1), with the next highest in limestone dolomite and marble (ca 26 Bqkg−1 in both). Unlike with 40K and 232Th, the lowest activity concentrations of 238U were observed in quartzite (9.5 Bqkg−1) and calcschist (14.4 Bqkg−1; Table 2 and Fig. 6). An intermediate value was observed in carbonaceous breccia (22.5 Bqkg−1). As seen in Fig. 5, an average value of 238U for all investigated rocks is 21 Bqkg−1. An average using just samples 2, 3, 4, and 5 is 26 Bqkg−1; this value is the same as the average reported for typical carbonate rocks (25 Bqkg−1), within measurement uncertainty.

Fig. 6
figure 6

Activity concentrations of 238U. Thick solid line average 238U activity in typical carbonates; thin solid line average 238U activity using all the samples; thin dotted line average activity of 238U using only samples 2, 3, 4, and 5

Full size image

Concentrations of K (%), 232Th (ppm), and 238U (ppm)

The concentrations of K (%), 232Th (ppm), and 238U (ppm) were calculated from the activity concentrations of 40K, 228Ac, and 226Ra (Table 3). Also seen in Table 3, the highest concentrations of K were found in quartzite (1.89%) and calcschist (1.29%). An exceptionally low concentration, 0.06%, was calculated for limestone dolomite. The average concentration of K in the investigated rocks was 0.71%.

Table 3 Concentrations of K (%), 232Th (ppm), and 238U (ppm)
Full size table

Concentrations of 232Th varied from 0.16 ppm in limestone dolomite to 4.38 ppm in calcschist. The concentration of 232Th averaged over all the samples is 1.65 ppm. Concentrations of 238U for samples 2, 3, 4, and 5 vary within a narrow range from 1.8 ppm (carbonaceous breccia) to 2.3 ppm (dolomite). Similar to the activity levels, the lowest concentrations of 238U were obtained in quartzite (0.73 ppm) and calcschist (1.2 ppm; Table 3). The average concentration of 238U is 1.7 ppm.

The concentration ratios of 238U/232Th in the investigated rocks are shown in Fig. 7. These ratios ranged from 0.26 in calcschist to 12.9 in limestone dolomite. The ratio averaged over all the samples is 3.5, while the average just from samples 2, 3, 4, and 5 is 5.1. Both of these values are noticeable higher than the average ratio in typical carbonates, i.e., 1.3 [20].

Fig. 7
figure 7

Concentration ratios of 238U/232Th. Thick solid line average reported for the 238U/232Th ratio in typical carbonates; thin solid line average 238U/232Th ratio using all the samples; thin dotted line average 238U/232Th ratio using only samples 2, 3, 4, and 5

Full size image

Values higher than these presented here for “pure limestone”, samples 2, 3, 4 and 5, were measured in limestone rocks in northern Iraq [21]. The activity concentrations of 40K and 238U in these rocks changed from 28 to 64 and 290 to 637 Bqkg−1, respectively.

Comparable values to marble from Aussois, sample 5, have been reported for marbles available in the Rawalpindi/Islamabad area [22]. The measured activities of 40K, 232Th and 238U varied in the range of 6–159, 1.2–6.3 and 1.4–29 Bqkg−1, respectively. Similar activity of 238U, 33 Bqkg−1, was found in marble from Italy [23], whereas activity concentrations of 40K (< 1 Bqkg−1) and 232Th (0.4 Bqkq−1) in the same rock were noticeably lower.

Activities similar to these observed in sample 6 (quartzite) have been recorded in quartzite presented in paper [24]. Significantly higher activity concentrations of primordial radionuclides have been measured in quartzite from India [25]. The average activity concentrations of 40K, 232Th and 238U in these rocks were 741, 48 and 112 Bqkg−1, respectively.

Intercolerrations of K (%), 232Th (ppm), and 238U (ppm)

Figure 8 shows distinct intercolerrations between the K, 232Th, and 238U concentrations in the investigated rocks. A positive correlation is seen between 232Th and K (Fig. 8a). Even though points 1 and 6 (representing calcschist and quartzite) show noticeable deviations from the fitted line, the correlation coefficient is relatively high, R = 0.78. A strong negative correlation is clearly visible between the 238U and K concentrations (Fig. 8b), with a correlation coefficient R = −0.94. Only point 4, representing dolomite, shows a slight deviation from a linear fit. As may be expected, a negative correlation is also observed between 232Th and 238U, with R = −0.78 (Fig. 8c). Similar to the 232Th–K dependence, there are noticeable deviations from the fitted line for calcschist (point 1) and quartzite (point 6).

Fig. 8
figure 8

Correlations between: a 232Th (ppm) vs. 40K (%); b 238U (ppm) vs. 40K (%); and c 232Th (ppm) vs. 238U (ppm). The solid lines represent linear fits: 232Th (ppm) = 1.72 × 40K (%) + 0.42; 238U (ppm) = −0.83 × 40K (%) + 2.29; and 232Th (ppm) = −1.97 × 238U (ppm) + 4.99. The correlation coefficients are R = 0.78, −0.94, and −0.78, respectively

Full size image

Conclusions

Characteristic rocks of the Modane-Aussois region showed relatively low activity concentrations of 40K, 232Th, and 238U. The highest activities for 40K and 232Th were measured in quartzite and calcschist (572 Bqkg−1 and 18 Bqkg−1, respectively). The highest activity of 238U (29 Bqkg−1) was noted in dolomite. The concentration ratios of 238U/232Th in the investigated rocks varied from 0.26 in calcschist to 12.9 in limestone dolomite, with an average value of 3.5. A positive correlation was noted between the 40K and 232Th concentrations, whereas correlations between 40K and 238U and between 238U and 232Th were noticeably negative.