Actinides AMS at CIRCE and 236U and Pu measurements of structural and environmental samples from in and around a mothballed nuclear power plant

Nuclear Instruments and Methods in Physics Research B 294 (2013) 152–159

Contents lists available at SciVerse ScienceDirect

Nuclear Instruments and Methods in Physics Research B journal homepage: www.elsevier.com/locate/nimb

Actinides AMS at CIRCE and 236U and Pu measurements of structural and environmental samples from in and around a mothballed nuclear power plant M. De Cesare a,b,⇑, L.K. Fifield c, C. Sabbarese a,b, S.G. Tims c, N. De Cesare d,b, A. D’Onofrio a,b, A. D’Arco a, A.M. Esposito e, A. Petraglia a, V. Roca f,b, F. Terrasi a,b a

CIRCE, INNOVA, and Dipartimento di Scienze Ambientali, Seconda Università di Napoli, via Vivaldi 43, 81100 Caserta, Italy INFN Sezione di Napoli, via Cintia, Edificio G, 80126 Napoli, Italy Department of Nuclear Physics, Research School of Physics and Engineering, Australian National University, ACT 0200, Canberra, Australia d CIRCE, INNOVA, and Dipartimento di Scienze della Vita, Seconda Università di Napoli , via Vivaldi 43, 81100 Caserta, Italy e Società Gestione Impianti Nucleari–SoGIN, via Torino 6, 00184 Roma, Italy f Dipartimento di Scienze Fisiche, Università Federico II, via Cintia, Edificio G, 80126 Napoli, Italy b c

a r t i c l e

i n f o

Article history: Received 1 July 2011 Received in revised form 1 March 2012 Available online 16 May 2012 Keywords: 236 U Pu isotopes Mass and isotopic ratio sensitivities Environmental and structural samples

a b s t r a c t Accelerator mass spectrometry (AMS) is presently the most sensitive technique for the measurement of long-lived actinides, e.g. 236U and 239Pu. A new actinide line is in operation at the Center for Isotopic Research on Cultural and Environmental heritage (CIRCE) in Caserta, Italy. Using the actinide line a uranium mass sensitivity of around 4 lg has been reached measuring with a 16-strip silicon detector, and a 239 Pu background level of below 0.1 fg has been obtained. In this work we also discuss preliminary results for environmental and structural samples from in and around the Garigliano nuclear power plant (GNPP), presently in the decommissioning phase. Measurements on environmental samples from the vicinity of the plant allow the assessment of contamination, if any, over the years. Measurements of structural samples from the plant are relevant to the optimization of the decommissioning program for the GNPP. Ó 2012 Elsevier B.V. All rights reserved.

1. Introduction Long-lived anthropogenic radionuclides have been released into the environment by nuclear weapons testing, nuclear accidents, fuel reprocessing and decommissioning of NPPs. In Italy no NPPs are in operation, but the four shutdown NPPs are now being decommissioned. An overriding concern in the dismantling process is to avoid possible contamination of the site by radionuclide release. At the GNPP several radiological assessment campaigns have been carried out using conventional radioactive decay counting techniques to determine the extent, if any, of environmental contamination by b- and c-emitters. Long-lived a-emitters, specifically uranium-236 and isotopes of plutonium, are however difficult to measure with adequate sensitivity using a-particle counting techniques, and the ultra-sensitivity of AMS is crucial for the evaluation of any contamination from these important nuclides. In particular, the measurement of the activities in structural materials of the plant is required to inform the appropriate procedures for dismantling. ⇑ Corresponding author at: CIRCE, INNOVA, and Dipartimento di Scienze Ambientali, Seconda Università di Napoli, via Vivaldi 43, 81100, Caserta, Italy. E-mail address: [email protected] (M. De Cesare). 0168-583X/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.nimb.2012.05.020

Pu isotopes and 236U are currently used in a broad domain ranging from tracing nuclear releases, through nuclear safeguards, to tracing soil loss and sediment transport [1–3]. The CIRCE [3,4], located in Caserta (Italy), in collaboration with SoGIN (Nuclear Plant Management Company) started a research program to establish a highly sensitive system for measuring the concentration and isotopic ratios of U and Pu isotopes based on AMS, applicable to the analysis of both environmental and structural samples (i.e. from the reactor building and related infrastructures) to quantify and determine the origin of any U or Pu. For 236U, the measured quantity is 236U/238U, and the sensitivity limit for this ratio depends on the mass of uranium in the sample. To measure the isotopic ratio in environmental samples it is therefore desirable to push the sensitivity down to natural abundance levels (236U/238U  10 13) in samples with sizeable amounts of U (1 mg). On the other hand, for anthropogenically influenced samples, higher ratios are expected so the required sensitivity may be relaxed, and significantly smaller amounts of U may then be used [3]. In this paper we present the performance of the CIRCE system, the sensitivity reached and the measurement results on samples from the GNPP site, obtained both at the Australian National University (ANU) [3] and at CIRCE.

153

M. De Cesare et al. / Nuclear Instruments and Methods in Physics Research B 294 (2013) 152–159

2. Results of AMS measurements In a first phase of the project [4–6] aiming to build a beam line dedicated to actinides AMS at CIRCE based on a 3-MV AMS Pelletron tandem system, we have performed preliminary measurements to characterize the system, shown in Fig. 1. In particular it has been shown that the addition of a switching magnet just after the ESA of the CIRCE AMS system was able to reduce the background in the 236U detection from 236U/238U  3.0  10 9 to < 5  10 11, even without the use of a TOF-E detection system, that is in any case planned for the near future. The determination of the background was achieved using a silicon strip detector that provides the spatial distribution of the incident ions to assist in the identification and assessment of isotopic interferences. 2.1. CIRCE accelerator:

236

U and Pu measurement procedures

UO or PuO molecular negative ions were extracted from an NEC 40-sample MC–SNICS cesium sputter ion source and pre-accelerated to a total injection energy of 50 keV. Typical beam currents are 50–300 nA for 238U16O ions. The ions are energy selected by a spherical electrostatic analyzer with a bending angle of 45°, operated up to ±15 kV. The 90° double focusing low energy (LE) injector magnet allows high resolution mass analysis for all stable isotopes in the periodic table; mass resolution is M/DM  500 for a slits aperture of ±1 mm [4]. The insulated stainless steel chamber can be biased up to 15 kV for beam sequencing (e.g. between 238 16 U O , 236U16O or between 239Pu16O , 240Pu16O , 242Pu16O ). An argon gas stripper in the high-voltage terminal of the accelerator is used to dissociate the molecules and strip the resulting atomic ions to high positive charge states. After acceleration, the double focusing 90° high energy (HE) analyzing magnet (M/DM = 725 with

slit opening of ±1 mm both at object and image points), efficiently removes molecular break-up products [4,5]. Subsequently, the two 45° electrostatic spherical analyzers select ions of the correct energy with an energy resolution E/DE = 700 for a typical beam size of 3–4 mm. A switching magnet is positioned after the ESA. Finally the selected ions at 20° are counted in an appropriate detector. The control of the acquisition system is handled by the Fast Intercrate Readout (FaIR) system [7] via Ethernet or AccelNet interfaces. Measurement of the 236U/238U ratio is effected by counting 236U ions in the final detector and periodically measuring the 238U current in the FC04 Faraday cup after the analyzing magnet. The transmission efficiency between the FC04 and FC5 at 20° is 80%, with a 4 mm collimator in place. A tuning of the transport elements up to the Faraday cup in front of the final detectors (LFC) is performed by setting the parameters of the beam line to the detection of 238U. For 238 5+ U an energy of E = 17.3 MeV with a terminal voltage of V = 2.900 MV is reached. The working pressure of the Ar in the stripper is about 1.3 mTorr for 238U5+ at 2.875 MV [5] and the stripping yield achieved for 238U5+ is around 3.1%. Once the setup for the pilot 238U5+ beam is determined, the voltage at the chamber of the injection magnet, the terminal voltage and the voltage of the ESA are scaled to transmit 236U5+. For Pu isotopes, the system is also set up with a 238U5+ pilot beam, and then the same parameters as above are scaled to transmit sequentially 239,240,242Pu. 2.2. Internal calibration and

236

U and

239

Pu mass sensitivity

2.2.1. Internal calibration In order to test the linearity of the response of the system, a series of samples (Ratio Series) with nominal isotopic ratios from  5  10 8 down to  1  10 10 were prepared by mixing different

Fig. 1. Schematic layout of the CIRCE accelerator with the actinides line layout, including the switching magnet, start and stop TOF-E detector and the Ionization Chamber. FC denotes Faraday Cup; arrows indicate system slits and collimators are identified by a C. ERNA is the acronym of European Recoil separator for Nuclear Astrophysics.

154

M. De Cesare et al. / Nuclear Instruments and Methods in Physics Research B 294 (2013) 152–159

Table 1 236 U/238U isotopic ratio series, Vienna-KkU and IRMM-075/5,6: Nominal and measured values. The nominal isotopic ratios were obtained by mixing the ViennaKkU material with various IRMM certified materials with higher isotopic ratio, to produce an internal CIRCE isotopic ratio standard series. The RS nominal isotopic ratio uncertainty arises from the uncertainty of the Vienna-KkU mass and isotopic ratio, and from the dilution of the IRMM-075/1, 2, 3, 4 standards. The RS measured isotopic ratio uncertainty (1 r) is dominated by the low FC5 currents. 236

U/238U nominal values

Normalization samples IRMM-075/5 (1.06519 ± 0.00075)  10 IRMM-075/6 (1.0885 ± 0.0063)  10 9 Average normalization Ratio series samples RS0 (4.56 ± 0.28)  10 RS1 (9.61 ± 0.58)  10 RS2 (4.69 ± 0.28)  10 RS3 (1.12 ± 0.06)  10 RS4 (5.46 ± 0.29)  10 RS5 (9.56 ± 0.36)  10 Vienna-KkU (6.98 ± 0.32)  10

8 9 9 9 10 11 11

236

U/238U measured valuesa

8

(9.38 ± 0.47)  10 (9.40 ± 0.45)  10 1.15 ± 0.04 (5.71 ± 0.45)  10 (8.50 ± 0.67)  10 (4.29 ± 0.34)  10 (9.73 ± 0.80)  10 (4.42 ± 0.36)  10 (9.31 ± 0.93)  10 (8.04 ± 0.82)  10

9 10

8 9 9 10 10 11 11

a

The measured values for the ratio series have been normalized to the average of the two IRMM standards.

amounts of the VERA in house standard Vienna-KkU [8] material with IRMM-075 (Institute for Reference Materials and Measurements) certified series material (IRMM-075/1,2,3,4) with higher isotopic ratios. The data, including the Vienna-KkU material itself, are shown in Table 1 and were normalized to the weighted average of the IRMM-075/5 and IRMM-075/6 reference material. A good linear behavior is observed (Fig. 2) and the lowest point (ViennaKkU) is in agreement with the VERA result. We could, however, take the difference between the measured and actual ratios of the Vienna-KkU material as an indication of the background at the 236U settings due to 235U ions. This then corresponds to a background of 1  10 11 in the 236U/238U ratio, or equivalently 1  10 9 (1 ppb) in the 236U/235U ratio (see Section 3.2.3). 2.2.2. 236U mass sensitivity In order to study the 236U concentration sensitivity of the system, we measured a series of samples from the RS1 material containing progressively smaller amounts of uranium. Four samples, DS1, DS2, DS3 and DS4, respectively containing 430, 43, 4.3 and

0.4 lg of U, were measured and compared to sample DS0 which contained 4.3 mg of U. The results are shown in Fig. 3. For samples DS3 and DS4 where no 238U5+ current was measurable we counted 234 5+ U in the final detector as a proxy for 238U, since the isotopic ratio between 234U and 238U is well known. Both methods could be employed for sample DS2, and Fig. 3 shows that good agreement was achieved. For uranium masses down to about 4.3 lg the measured 236U/238U ratio agrees well with the nominal value. The 0.4 lg sample, however, gives a ratio that is almost a factor of two higher than expected (RS1). We conclude that at this level, uranium from the sample matrix, sample holder, and/or ion source becomes significant. Furthermore, this extraneous uranium must contain 236U and have a 236U/238U ratio above 1  10 8. It is perhaps not surprising that the laboratory background is relatively high in 236U, since ‘‘normal uranium’’ as found in present-day soil, surface water, ocean, reagents, and labware typically has 236U/238U ratios in the range 10 9–10 6 as a consequence of the 900 kg of 236 U that was distributed by global fallout [9,18]. For the moment then, reliable measurements of 236U/238U ratios of  1  10 8 can be performed on samples containing as little as 4 lg of uranium, with a corresponding sensitivity of 40 fg of 236U, i.e. 1  108 atoms. 2.2.3. 239Pu mass sensitivity Using the same system and tuning procedure as for U we performed Pu isotope measurements, employing an ionization chamber in place of the Si strip detector. Such detectors are unable to discriminate between 239Pu and 238U ions with the same mE/q2, due to their limited energy resolution. Although it is possible to discriminate between the two with a Time-of-Flight system, this comes at the expense of reduced efficiency and increased complexity. The question then arises, is uranium background likely to be a serious concern in AMS measurements of plutonium with an ionization chamber detector? In order to address this question, we prepared a series of plutonium-free samples with decreasing uranium concentration, and measured the counting rate at the 239Pu setting from each. This can be converted to an ‘‘apparent abundance’’ of 239Pu due to uranium background. A dilution series containing a decreasing amount of natural uranium, from 5 mg to 5 pg, was prepared from Vienna-KkU uranium which had been processed to remove any plutonium that might have been present (Pu occurs naturally in uranium ores [10]). The results are shown in Fig. 4 as the apparent 239 Pu mass vs. the mass of uranium in the sample. For the samples

Fig. 2. Internal calibration series: 236U/238U Ratio Series. Measured vs. nominal values. The nominal isotopic ratios were obtained by mixing Vienna-KkU material IRMM-075 certified material with a higher isotopic ratio, to produce an internal CIRCE isotopic ratio standard series (see text). The 1:1 nominal ratio line along which the samples should fall is also shown.

M. De Cesare et al. / Nuclear Instruments and Methods in Physics Research B 294 (2013) 152–159

155

Fig. 3. Results of 236U/238U measurements for a series of samples of decreasing uranium content. Samples DS0, DS1, DS2, DS3, DS4 contained respectively 4300, 430, 43, 4.3 and 0.4 lg of U with a nominal isotopic ratio of 9.61  10 9. The 238U5+ currents measured in the FC5 and 234U counts per second (cps) are also shown.

containing 0.5 lg–5 mg of uranium, it can be seen that the uranium background at the 239Pu settings is at the level of 1 ppb. Since 238 U background arises from injection into the accelerator of both 238 17 U O and 238U16O1H negative molecular ions, this apparent background will partly depend on the amount of hydrides produced by the ion source. Note also that the abundance sensitivity for mass 239 in the presence of 238 is essentially equivalent to 236 in the presence of 235. The value of 1 ppb given here for 239 Pu/238U is in essential agreement with that deduced in Section 3.2.1 above for 236U/235U. A similar study has been reported previously [11] using the AMS system at ANSTO (Australian Nuclear Science and Technology Organization) in Australia, where the uranium background was found to be at the level of 10 ppm, i.e. a factor of 104 higher than found in the present work. CIRCE, however, has two major advantages over the ANSTO system. First, it has an electrostatic analyzer (ESA) between ion source and low energy magnet which essentially eliminates the high-energy tail of 238U16O ions which would otherwise be injected with 239Pu16O . Secondly, it has substantially better vacuum, 9  10 9 Torr [5], in the high-energy acceleration

tube as a consequence of accelerator tube design (metal-bonded ceramic at CIRCE versus epoxy-bonded glass at ANSTO), which reduces the probability of charge-changing collisions with residual gas molecules that can lead to the production of 238U5+ ions of the correct energy to pass through the analyzing magnet [4,5]. Hence, we would expect the background to be significantly lower at CIRCE than at ANSTO. Similar unpublished work at the ANU, which does not have an ESA at the low-energy end either, but does have the advantage of higher vacuum, 2  10 8 Torr, in the accelerator tubes, indicates a background level of 100 ppb, i.e. still 2 orders of magnitude higher than at CIRCE. A comparison between the three systems is shown in Fig. 4. Because uranium concentrations in the environmental sample can readily be reduced to very low levels (<1 ppm, equivalent to <2 ng in a typical 2 mg sample) with appropriate chemistry, we can conclude from these results that we are already sensitive to 239Pu at the level of <0.1 fg, since 500 ng of uranium is required to produce an apparent 239Pu concentration of 0.1 fg. If uranium background is negligible, then the final detection system need only be able to discriminate between Pu ions and lower-energy ions in lower charge states.

Fig. 4. Background from 238U ions at the 239Pu settings as a function of the amount of uranium in the sample. Samples containing a decreasing amount of uranium, from 5 mg to 5 pg, were prepared from Vienna-KkU uranium which had been processed to remove any plutonium that may have been present. The circled CIRCE point at 0.1 fg indicates the limits of the system (see text).

156

2.3.

M. De Cesare et al. / Nuclear Instruments and Methods in Physics Research B 294 (2013) 152–159 236

U and Pu GNPP measurements

2.3.1. Environmental samples The area surrounding the GNPP (41°15’ N, 13°49’ E) has been divided into four circular crowns, each of 1 km radius, from a internal radius of 500 m from the NPP out to 4.5 km, Fig. 5. The two internal crowns were divided into 8 sectors and the outer two into 16 sectors, for a total of 48 sampling points. An additional remote sampling point (BSC2) was selected in the plain of Sele, Salerno province, geologically similar to the Garigliano NPP, but more than 100 km away, and hence should not show any significant influence from the NPP, and indeed the ratios observed from BSC2 are typical of global fallout in southern Europe. At the GNPP various radiological campaigns were carried out which allowed the determination of the environmental contamination attributable to c-ray emitters. Gamma spectrometry of natural (7Be, 40K) and artificial (60Co, 137Cs) radionuclides was performed with a high energy resolution, low background germanium detector [12,13]. The results reported here were obtained from soil samples collected in May–July, 2008 from about 20 cm deep soil pits in areas where there was little evidence of disturbance. Samples were dried at 60 °C for 36 h and at 105 °C for 12 h. They were then pulverized to obtain an homogeneous matrix, sieved at 2 mm and placed in 1.45 L Marinelli containers for c-ray measurements. For those samples found to have the highest 137Cs activities, 20 g sub-samples were taken for measurement of 236U and 239Pu concentrations at the ANU AMS system [3,14]. Details of the actinide chemical procedure are given in [10,12]. Details of the measurements are and will be given elsewhere [12,15]. The results are given in Table 2, Table 3 and Fig. 6. The 240Pu/239Pu ratio for the samples from around the GNPP, as well as from the remote BCS2 site, are all consistent within errors

with the ratio expected from global fallout in southern Europe; the average composition of the 240Pu/239Pu fallout for the latitude zone 30–71° N is 0.180 with a standard deviation of 0.014 (2r) [16]. This indicates that the observed plutonium likely originates from global fallout, and not from the GNPP. On the other hand, 239Pu concentrations vary by a factor of 10, from 32.1 to 396.4  106 atoms g 1. Assuming a soil density of 1.5 g cm 3, these can be converted to inventories of 1.0– 11.9  109 atoms cm 2 in the top 20 cm of soil where most of the Pu is expected to reside. These are of the same order as the value of 5.8  109 atoms cm 2 measured in samples collected by Hardy et al. [17] at Ispra, Italy (the nearest site to the present study site) and reported by Kelley et al. [16]. The range of concentrations observed here may be due to different degrees of disturbance in what is an intensively agricultural area, despite our efforts to collect from what appeared relatively undisturbed locations. The 236U concentrations and 236U/238U isotopic ratios also vary by a factor of 10, ranging respectively from 7.7 to 54.9  106 atoms g 1 and from 0.94 to 11.70  10 9. These results, being of the some order or lower than reported by [2,9,18,19] are all consistent with the values expected from global fallout. Significant variability is also observed in the 137Cs/(239+240)Pu activity ratio and 236U/239Pu atom ratio. The 137Cs/(239+240)Pu activity ratios, for example, vary between 36.7 and 158.1, which are to be compared with the expected global fallout ratio of 26.1 (137Cs decay corrected to October–November, 2008 when the c-ray measurements were performed) [20]. All of the samples measured have higher ratios. It is likely that this variability reflects the influence of Chernobyl which deposited significant amounts of 137Cs, but not Pu or 236U, in this part of Italy [2]. Since the Chernobyl fallout is more recent than the global fallout and hence would have been retained nearer the surface compared to 236U and Pu isotopes [9], its

Fig. 5. Grid around the Garigliano NPP that is in the center of the crown. The crowns were identified by a letter of the alphabet starting with the innermost ring, and the sectors are numbered sequentially going counterclockwise; for the inner crowns, A and B, the sectors cover twice the angular range, and the numbers were paired two by two. For example, the area comprising Maiano is shown as C4, while that with San Venditto is shown as B1516.

157

M. De Cesare et al. / Nuclear Instruments and Methods in Physics Research B 294 (2013) 152–159 Table 2 Concentrations of 137Cs, 239Pu, 240Pu and 236U, inventories of 239Pu, 236U and isotopic atom ratios of 240Pu/239Pu, from around the GNPP. Errors correspond to one standard deviation, and reflect statistical uncertainties alone. 137 Cs 106 atoms g

A78 B56 B34 C7 C13 D12 BSC2

239

Pu 106 atoms g

1

15.9 ± 0.3 23.3 ± 0.4 7.9 ± 0.3 4.2 ± 0.2 32.0 ± 0.9 32.6 ± 0.5 7.4 ± 0.2

240

Pu 106 atoms g

1

106.8 ± 3.1 342.5 ± 8.6 85.5 ± 5.4 32.1 ± 2.6 92.3 ± 5.1 396.4 ± 7.2 95.6 ± 6.2

1

22.2 ± 1.2 63.7 ± 2.5 18.8 ± 3.9 5.6 ± 1.5 19.2 ± 1.8 67.0 ± 1.8 18.0 ± 2.0

239 Pu 109 atoms cm

240

Pu/239Pu

2

3.2 ± 0.4 10.3 ± 1.3 2.6 ± 0.3 1.0 ± 0.1 2.8 ± 0.4 11.9 ± 1.4 2.9 ± 0.4

0.21 ± 0.01 0.19 ± 0.01 0.22 ± 0.05 0.17 ± 0.05 0.21 ± 0.02 0.17 ± 0.01 0.19 ± 0.02

236

U/238U as well as

236 U 106 atoms g

1

15.4 ± 0.9 37.7 ± 1.9 27.2 ± 1.7 10.0 ± 0.8 7.7 ± 1.1 54.9 ± 2.0 10.1 ± 0.7

236

U/239Pu for the environmental samples

236 U 109 atoms cm

236 2

0.46 ± 0.06 1.13 ± 0.15 0.82 ± 0.11 0.30 ± 0.04 0.23 ± 0.04 1.65 ± 0.21 0.30 ± 0.04

10

236

U/238U

U/239Pu

9

4.61 ± 0.20 6.03 ± 0.24 2.18 ± 0.19 0.94 ± 0.07 1.30 ± 0.09 11.70 ± 0.27 4.67 ± 0.27

0.14 ± 0.01 0.11 ± 0.01 0.32 ± 0.03 0.31 ± 0.04 0.08 ± 0.01 0.14 ± 0.01 0.11 ± 0.01

Table 3 Activity concentrations of 137Cs, 239Pu, 236U, 239+240Pu, and activity ratio of 137Cs /(239+240)Pu as well as 239Pu, 239+240Pu and 236U inventories for the environmental samples from around the GNPP. Errors correspond to one standard deviation, and reflect statistical uncertainties alone. 137

A78 B56 B34 C7 C13 D12 BSC2

Cs (mBq g

11.6 ± 0.2 17.0 ± 0.3 5.8 ± 0.2 3.1 ± 0.1 23.4 ± 0.7 23.8 ± 0.3 5.4 ± 0.2

1

)

239

Pu (mBq g

0.097 ± 0.003 0.312 ± 0.008 0.078 ± 0.006 0.027 ± 0.004 0.084 ± 0.005 0.361 ± 0.007 0.087 ± 0.006

1

)

236

U (lBq g

1

)

0.0144 ± 0.0008 0.0355 ± 0.0018 0.0255 ± 0.0016 0.0093 ± 0.0008 0.0072 ± 0.0010 0.0515 ± 0.0018 0.0095 ± 0.0006

239+240

Pu (mBq g

0.172 ± 0.005 0.525 ± 0.011 0.150 ± 0.072 0.045 ± 0.006 0.148 ± 0.008 0.585 ± 0.009 0.147 ± 0.009

1

)

137

Cs/(239+240)Pu

67.4 ± 3.0 32.4 ± 1.2 38.7 ± 18.8 68.9 ± 11.1 158.1 ± 13.1 40.7 ± 1.1 36.7 ± 3.5

239

Pu (mBq cm

2.92 ± 0.09 9.36 ± 0.23 2.30 ± 0.29 0.80 ± 0.12 2.52 ± 0.14 10.83 ± 0.20 2.61 ± 0.17

2

)

236

U (lBq cm

0.43 ± 0.06 1.07 ± 0.14 0.77 ± 0.10 0.28 ± 0.04 0.22 ± 0.04 1.55 ± 0.19 0.29 ± 0.04

2

)

239+240

Pu (mBq cm

2

)

5.1 ± 0.1 15.8 ± 0.3 4.2 ± 0.5 1.4 ± 0.2 4.5 ± 0.2 17.6 ± 0.3 4.4 ± 0.3

Fig. 6. The 137Cs, 239Pu and 236U concentrations and 240Pu/239Pu ratios are respectively shown for the environmental samples. The solid line corresponds to the global fall out value 0.18.

contribution would have been more sensitive to small amounts of soil loss or gain, particularly in the initial period after deposition, which could lead to considerable variability in the ratio. The 236U/239Pu ratios vary between 0.08 and 0.32, which are to be compared to the global fallout ratio results in the range of 0.05– 0.50 [21], 0.212–0.253 [18] and 0.04–0.78 [19]. The wide range obtained by Ketterer et al. [21] and Srncik et al. [19] was attributed to the evidently higher mobility of 236U fallout compared to 239Pu. However, this migration behavior of U and Pu was not observed by Sakaguchi et al. [18]. Our data seems to confirm the higher mobility. The lower ratios may indicate that a significant amount of the 236U has been leached from the soil due to the greater solubility of uranium relative to plutonium. The two samples with the higher ratios, B34 and C7, are the ones with the lowest Pu concentrations; perhaps in this case we are seeing 236U that has moved

down the soil column while the high Pu surface material has been lost. We conclude, therefore, that the data presented here indicate that there is little evidence of contamination by radionuclides from the GNPP in the surrounding environment. The uranium isotopic ratio and concentrations results also confirm this statement. A similar conclusion was drawn from measurements on sediment samples from the GNPP drain channel and Garigliano River [2,8].

2.3.2. Structural samples A first attempt at quantitative measurements of 236U and plutonium isotopes on structural material from the GNNP was performed at CIRCE to provide information valuable for the decommissioning program. The details of these measurements

158

M. De Cesare et al. / Nuclear Instruments and Methods in Physics Research B 294 (2013) 152–159

Table 4 239 Pu concentrations and inventories and Pu and U isotopic ratio results from concrete from the chimney, external wall and pipe cooling circuit (see text). Errors correspond to one standard deviation, and reflect statistical uncertainties alone. The densities to convert concentrations to inventories are 2.1 and 7.8 g cm 3 for the concrete and metal (steel) and the depths are 1 mm for the internal, 5 mm for the external as well as the wall, and 0.6 mm for the metal. 239 Pu 106 ats g

I I8 I12 I16 E1.5 E5_1 2 3 4 5 EW_1 2 3 4 MCC1 2 3

734 ± 82 340 ± 54 559 ± 69 449 ± 70 25 ± 14 45 ± 16 46 ± 8 30 ± 6 30 ± 6 42 ± 9 13 ± 3 3±2 14 ± 4 8±2 270 ± 17 89 ± 7 48 ± 5

239 1

Pu 107 ats cm

15.4 ± 2.4 7.1 ± 1.4 11.7 ± 2.0 9.4 ± 1.8 2.6 ± 1.5 4.8 ± 1.8 4.9 ± 1.0 3.1 ± 0.7 3.1 ± 0.7 4.4 ± 1.0 1.4 ± 0.3 0.4 ± 0.2 1.4 ± 0.4 0.8 ± 0.3 – – –

240

Pu/239Pu

2

236

10 0.32 ± 0.05 0.33 ± 0.08 0.18 ± 0.04 0.19 ± 0.06 0.51 ± 0.42 0.24 ± 0.15 0.70 ± 0.16 0.46 ± 0.16 0.38 ± 0.12 0.31 ± 0.14 0.73 ± 0.27 0.69 ± 0.56 1.31 ± 0.49 0.48 ± 0.27 0.41 ± 0.02 0.41 ± 0.04 0.51 ± 0.07

U/238U

- Metal from the cooling circuit of the reactor (one sample was progressive leached): 1. MCC1: First leach of metal, 1 M HCl at room temperature for 10 h 2. MCC2: Second leach of metal, 5 M HCl at room temperature for 10 h 3. MCC3: Third leach of metal, 5 M HCl on heating plate for 2 h

8

5.7 ± 4.0 5.7 ± 5.7 7.5 ± 3.8 14.3 ± 5.4 5.9 ± 4.2 0.7 ± 0.1 0.3 ± 0.1 0.9 ± 0.7 0.17 ± 0.08 1.4 ± 1.4 0.3 ± 0.1 – – 3.3 ± 3.3 18.1 ± 10.5 50.4 ± 13.1 5.8 ± 5.8

are and will be given elsewhere [12,15]. The specific materials targeted for this proof-of-principle study were: - Concrete from the ventilation chimney of the reactor system: four samples were taken from the inner surface, starting at the chimney foundations, 4 m below ground level, and then at 8, 12 and 16 m above this point (samples I, I8, I12, and I16). An additional sample was collected from the external surface, 1.5 m above the foundations (sample E1.5). A further five external surface samples (E5_1, 2, 3, 4, 5) were taken at 5 m above the foundations, with horizontal spacing of 15 cm and vertical distance of 15 cm. The chimney is considered to have been the most likely route for any contamination of the environment. - Concrete from the external wall of the G-22 building close to the ventilation chimney: four samples (EW_1, 2, 3, 4) were taken at 1.5 m above the ground level, with horizontal spacing of 30 cm and vertical distance of 8 cm.

The results are shown in Table 4 and Fig. 7. Clearly, there is plutonium in the concrete from the inner surface of the chimney, although the inventories are lower than in the soils surrounding the plant. The concentration appears to be a maximum at the base, but otherwise does not vary substantially with height. Although the 240Pu/239Pu ratios appear to show differences between the two lower and the two upper samples, the statistical precisions is very poor due to a low chemical yield of the extraction process, and the differences are not statistically significant. The exterior of the chimney as well as the external wall of the adjacent building have very little plutonium and it was not possible to determine a reliable 240Pu/239Pu ratio. A clear decreasing trend between the inner surface of the chimney, the exterior surface of the chimney and the external wall is visible. The weighted means of the 239Pu concentrations are respectively (480 ± 33)  106atoms g 1, (35 ± 3)  106atoms g 1 and (7 ± 1)  106atoms g 1. For the metal sample from the cooling circuit, a higher chemical yield was achieved for extraction of the plutonium, and hence the statistical precisions of both the concentration and especially the 240 Pu/239Pu ratio are substantially better. The metal leaching is clearly removing more Pu at each step, with significant Pu still being removed at the third step. The summed 239Pu concentrations (239Pu = (407 ± 19)  106 atoms g 1) from the three leaches are comparable with the concentrations on the interior surface of the chimney, while the 240Pu/239Pu weighted means of 0.41 ± 0.02 are characteristic of high burn-up fuel, and in marked contrast to the global fallout ratio observed in the soil samples from around the plant. Since, for the metal leaching, significant Pu is still removed at the third step (Fig. 7) and since the contamination probably resides mainly on the surface of the materials, we report our results also as number of atoms per cm2, which appears to be a more appropriate unit in order to compare the activities of different surfaces exposed to contamination. The inventory of the summed 239Pu concentrations is slightly higher

Fig. 7. The 239Pu concentrations and 240Pu/239Pu ratios are respectively shown for the structural samples. The solid lines correspond respectively to the mean weight of the 239 Pu concentrations of the inner surface of the chimney, exterior surface of the chimney and external wall, as well as a line at the global fall out value 0.180.

M. De Cesare et al. / Nuclear Instruments and Methods in Physics Research B 294 (2013) 152–159

(239Pu = (19.0 ± 2.3)  107 atoms cm 2) respect to the interior surface of the chimney. The 236U/238U ratios range from about 2  10 9 to 5  10 7. There was, however, very little uranium beam from these samples, suggesting that the uranium extraction efficiency was very low. One consequence of the small uranium beams is the possibility that a significant fraction of the observed 238U beam may have been coming from natural uranium intrinsic to the sample holders or to the iron oxide and silver of the sample matrix, and accordingly the present ratios represent lower limits. Even so, the observed ratios are higher than in the environmental soil samples, again supporting the conclusion that any releases from the plant have been minimal. During sample preparation, all solutions were retained, so it should be possible to find where the uranium resides and to recover it for future measurements. In conclusion, this very preliminary study shows a clear excess of plutonium and 236U in the GNPP structural samples, relative to environmental samples from the surrounding area.

3. Summary and conclusions An AMS system for measurement of U and Pu isotopes, based on the CIRCE pelletron accelerator, has been established. The present set up, using a Si strip and/or ionization chamber as the final detector for U ions, and an ionization chamber for Pu measurements, has been characterized both for concentration and isotopic ratio sensitivity. A TOF-E detector, presently under installation, will push the 236 U detection limits towards those needed to measure natural abundances and isotopic ratios [12,22]. It has been shown 236U/238U ratios of 1  10 8 can be determined on samples containing as little as 4 lg of uranium, with a corresponding sensitivity of 40 fg of 236U (i.e. 1  108 atoms). For Pu, an overall detection rate of 5  10 3 cps per fg was achieved. Background from uranium is <0.1 fg of 239Pu equivalent, provided that the sample contains less than 500 ng of uranium. With the present configuration, preliminary results on environmental and structural samples from the decommissioned GNPP have been shown. The data presented here for environmental sample indicate values consistent with global fallout and an additional 137Cs contribution from Chernobyl. There is little evidence of contamination by radionuclides from the GNPP in the surrounding environment. A similar conclusion was drawn from measurements on sediment samples from the GNPP drain channel and Garigliano River [2,8]. The structural (concrete and metal samples) results show the characteristics of high burn-up fuel (0.32) in marked contrast to the global fallout (0.18). This further reinforces the conclusion that the environmental signal is due to global fallout, and not to releases from the plant, although the inventories are lower than in the soils surrounding the plant. For the metal sample from the cooling circuit the sum of the 239Pu concentrations (400  106 atoms g 1) is comparable with the concentrations on the interior surface of the chimney, while the weighted means of 240 Pu/239Pu ratio (0.41) is also characteristic of high burn-up fuel, and in marked contrast to the global fallout ratio observed in the soil samples. The metal leaching is clearly removing more Pu at each step, with significant Pu still being removed at the third step and the inventory of the summed 239Pu concentrations is slightly higher respect to the interior surface of the chimney.

159

The work on the structural samples has, however, revealed that significant improvements to the chemical extraction procedures for both plutonium and uranium are required, particularly for concrete samples. Acknowledgment We kindly thank VERA (Vienna Environmental Research Accelerator) laboratory for providing the Vienna-KkU samples. References [1] L.K. Fifield, Accelerator mass spectrometry of the actinides, Quatern. Geochronol. 3 (2008) 276. [2] F. Quinto, P. Steier, G. Wallner, A. Wallner, M. Srncik, M. Bichler, W. Kutschera, F. Terrasi, A. Petraglia, C. Sabbarese, The first use of 236U in the general environment and near a shutdown nuclear power plant, Appl. Radiat. Isotopes 67 (2009) 1775. [3] M. De Cesare, Origin and Detection of Actinides: Where Do We Stand with the Accelerator Mass Spectrometry Technique?, Nuclear Power – Control: Reliability and Human Factors, ISBN 978-953-307-599-0, 2011. [4] M. De Cesare, L. Gialanella, D. Rogalla, A. Petraglia, Y. Guan, N. De Cesare, A. D’Onofrio, F. Quinto, V. Roca, C. Sabbarese, et al., Actinides AMS at CIRCE in Caserta (Italy), Nucl. Instrum. Meth. Phys. Res. B 268 (2010) 779. [5] M. De Cesare, Y. Guan, F. Quinto, C. Sabbarese, N. De Cesare, A. D’Onofrio, L. Gialanella, A. Petraglia, V. Roca, F. Terrasi, Optimization of 236U AMS at CIRCE, Radiocarbon 52 (2010) 286. [6] Yong-Jing GUAN, M. De Cesare, F. Terrasi, F. Quinto, C. Sabbarese, N. De Cesare, A. D’Onofrio, Hui-Juan WANG, 236U AMS measurement at CIRCE, Chin. Phys. C 34 (2010) 1729. [7] A. Ordine, A. Boiano, E. Vardaci, A. Zaghi, A. Brondi, FAIR: a new fast trigger and readout bus system, IEEE Trans. Nucl. Sci. 45 (1998) 873. [8] P. Steier, M. Bichler, L.K. Fifield, R. Golser, W. Kutschera, A. Priller, F. Quinto, S. Richter, M. Srncik, F. Terrasi, et al., Natural and anthropogenic 236U in environmental samples, Nucl. Instr. Meth. Phys. Res. B 266 (2008) 2246. [9] A. Sakaguchi, K. Kawai, P. Steier, T. Imanaka, M. Hoshi, S. Endo, K. Zhumadilov, M. Yamamoto, Feasibility of using 236U to reconstruct close-in fallout deposition from the Hiroshima atomic bomb, Sci. Total Environ. 408 (2010) 5392. [10] K.M. Wilcken, T.T. Barrows, L.K. Fifield, S.G. Tims, P. Steier, AMS of natural 236U and 239Pu produced in uranium ores, Nucl. Instr. Meth. Phys. Res. B 259 (2007) 727. [11] D.P. Child, M.A.C. Hotchkis, M.L. Williams, Improvements in actinide and fission product analysis by AMS-raising the bar and lowering detection limits in heavy element AMS, in: Proceedings of the 46th Annual Meeting of the Institute for Nuclear Materials Management, July 10-14, 2005, AZ, USA. [12] M. De Cesare, PhD thesis, Accelerator Mass Spectrometry of actinides at CIRCE, Department of Environmental Science - II University of Naples, 2006–2009. [13] A. Petraglia, C. Sabbarese, M. De Cesare, N. De Cesare, F. Quinto, F. Terrasi, A. D’Onofrio, P. Steier, L. K. Fifield, A. M. Esposito, Assessment of the radiological impact of a decommissioning nuclear power plant in Italy, Accepted by Radioprotection, 2012. [14] L.K. Fifield, R.G. Cresswell, M.L. di Tada, T.R. Ophel, J.P. Day, A.P. Clacher, S.J. King, N. D Priest, Accelerator mass spectrometry of plutonium, Nucl. Instr. Meth. Phys. Res. B 117 (1996) 295. [15] M. De Cesare et al. (to be published). [16] J.M. Kelley, L.A. Bond, T.M. Beasley, Priest, Global distribution of Pu isotopes and 237Np, Sci. Total Environ. 23 (7/238) (1999) 483. [17] E.P. Hardy, P.W. Krey, H.L. Volchok, Global inventory and distribution of fallout plutonium, Nature 241 (1973) 444. [18] A. Sakaguchi, K. Kawai, P. Steier, F. Quinto, K. Mino, J. Tomita, M. Hoshi, N. Whitehead, M. Yamamoto, First results on 236U levels in global fallout, Sci. Total Environ. 407 (2009) 4238. [19] M. Srncik, P. Steier, G. Wallner, Depth profile of 236U/238U in soil samples in La Palma, Canary Islands, J. Environ. Radioactiv. 102 (2011) 614. [20] S.E. Everett, S.G. Tims, G.J. Hancock, R. Bartley, L.K. Fifield, Comparison of Pu and 137Cs as tracers of soil and sediment transport in a terrestrial environment, J. Environ. Radioactiv. 99 (2008) 383. [21] M.E. Ketterer, A.D. Groves, B.J. Strick, 236U inventories, 236U/238U, and 236 U/239Pu: the stratospheric fallout signature, Goldschmidt conference abstracts. Geochim. Cosmochim. Ac. 71 (2007) A480. [22] L.K.Fifield, S.G.Tims, J.O.Stone, D.C.Argento, M. De Cesare, Ultra-sensitive measurements of 36CI and 236U at the Australian National University. Accepted by Nucl. Instr. Meth. Phys. Res. B, 2012.