Original Paper Fission Gas Bubbles Characterisation ... - Springer Link

2 downloads 0 Views 197KB Size Report
SIMS and EPMA have been used to measure the radial distribution of xenon and SEM gives information on bubble formation across the fuel pellet radius. Using.
Microchim Acta 155, 183–187 (2006) DOI 10.1007/s00604-006-0540-y

Original Paper Fission Gas Bubbles Characterisation in Irradiated UO2 Fuel by SEM, EPMA and SIMS Je´roˆme Lamontagne , Lionel Desgranges, Christophe Valot, Jean Noirot, Thierry Blay, Ingrid Roure, and Bertrand Pasquet Commissariat a l’Energie Atomique, Centre de Cadarache, DEN=DEC=SA3C=L2EC, B^at 316, F-13108 St Paul Lez Durance, Cedex, France Received May 26, 2005; accepted October 31, 2005; published online April 18, 2006 # Springer-Verlag 2006

Abstract. The behaviour of gases produced by fission is of great importance for nuclear fuel operation. Within this context, an experimental method for the characterisation of the fission gas including gas bubbles in an irradiated UO2 nuclear fuel was developed in our laboratory using SIMS, EPMA and SEM results. SIMS and EPMA have been used to measure the radial distribution of xenon and SEM gives information on bubble formation across the fuel pellet radius. Using SIMS, xenon concentration can be determined in the matrix and in the bubbles. A quantification method, allowing the determination of the total inventory of xenon, is proposed and qualified with EPMA results. It is concluded that the complementary micro-analytical techniques SIMS, EPMA and SEM are very powerful tools for the characterisation of the fission gas bubbles in irradiated nuclear fuel. Key words: SEM; EPMA; SIMS; UO2; xenon.

In the strategy for extending the burn-up of Pressurized Water Reactor (PWR) fuels, a matter of concern remains the behaviour of fission gases. They are released in the free volume and they impact the internal pressure of the fuel rod, the fuel swelling and the  Author for correspondence. E-mail: jerome.lamontagne@ cea.fr

fuel behaviour during incidental or accidental events. For two decades, Electron Probe Micro Analysis (EPMA) and Scanning Electron Microscope (SEM) methods have been applied to study this fission gas behaviour on polished cross-sections of irradiated fuel pellets. SEM allows to study UO2 microstructure changes across the fuel pellet radius and from EPMA measurement the radial distribution of the fission products in the UO2 fuel matrix is determined [1, 2]. This last method is limited by the fact that the EPMA analysis, only involves a layer at the sample surface thinner than 1 mm. The amount of fission gas in bubbles opened by the sample preparation is not negligible as soon as the bubbles are bigger than about 0.1 mm [3]. So there is a lack of detection of the gas from these bubbles [4, 5]. In order to improve the detection of xenon in irradiated fuels some experiments have been conducted with the CEA newly installed shielded Secondary Ions Mass Spectroscopy (SIMS) [6–8]. In a previous paper, it was showed that xenon can be detected both in bubbles and in solid solution in the UO2 matrix of an irradiated nuclear fuel [9]. In this paper, a methodology to characterize the total inventory of xenon along the radius of an irradiated nuclear fuel pellet is proposed using three complementary micro-analysis techniques.

184

Experimental Sample The UO2 sample, with an initial 235U enrichment of 4.5%, is taken from a French PWR fuel rod with Zircaloy cladding, irradiated for five cycles. Volumetric analyses gave a fraction of gas released out of the fuel of 2.5%. The local burn-up at the sampling position was calculated to be 61 GWd=tM. The sample was embedded in a metallic alloy with low melting point to ensure good electrical conductivity and polished. The same sample was used for SEM, EPMA and SIMS examination. The three apparatus are indeed in the same room with shielded tunnels allowing an easy transfer of radioactive material from one device to the other.

J. Lamontagne et al. teristic X-ray with a PET crystal. Xenon radial distribution and mapping were measured with an electron acceleration potential of 25 kV and incident beam current of 250 nA. The acquisition time was 20 s. Quantitative analysis of xenon was carried out [10] using an UO2 (irradiated at low temperature, 25 GWd=t(U)) standard, multi-characterised.

Scanning Electron Microscope SEM was performed with a shielded XL 30 model (PHILIPS) with a Centaurus KE developments BSE detector and a SIS ADDA image acquisition device to obtain 4096 4096 pixels images covering large areas with a high definition.

Electron Probe Micro Analysis

Secondary Ions Mass Spectroscopy

EPMA was performed with a shielded CAMEBAX model (CAMECA). Xenon was analysed using the L line of the charac-

SIMS was performed with a shielded IMS 6f (CAMECA). 132Xe depth profile was measured with a 20 nA oxygen primary beam

Fig. 1. (a) Photograph of the sample. The location where the measurements were performed by SEM, EPMA and SIMS are located. (b) EPMA radial measurement of Xe compared to creation. (c) SIMS radial measurement of 132Xe

Fission Gas Bubbles Characterisation in Irradiated UO2 Fuel by SEM, EPMA and SIMS defocused on a 30 mm diameter area. The ion sputtering lasted for 1000 s and the 132Xe signal was measured every 1 s.

Results and Discussion A global view of the sample is shown in Fig. 1a. The location where the SEM, EPMA and SIMS measurements were performed are indicated.

EPMA Results Figure 1b shows the EPMA xenon measurement profile from the periphery to the centre of the fuel pellet. The xenon creation profile in this figure is an estimation where shape is given by the Nd (which is generally considered as directly proportional to the local burn-up) relative profile and where intensity is fitted to the overall xenon calculation at this level of the rod. Three mains areas appear in this profile. The first one (0–100 mm) is located at the periphery of the pellet

185

where the rim effect occurred (i.e. changes of the UO2 structure into High Burn-up Structure (HBS) [2]). The second one is located between 100 mm to 3000 mm. In this mid-radius area, because of the low thermal level, no bubbles are formed during irradiation. As a consequence the xenon measured with EPMA is equal to the quantity created during irradiation (about 0.8% in mass). The third area is located at the centre of the pellet. In this area, there is a decrease in xenon concentration due to the fission gas precipitation in the form of bubbles opened by the sample preparation. In both central part and rim area, the level of the measured xenon is lower than the created one. SEM Results Together with that observation, BSE images from SEM (Fig. 2a) show that there is a major gas precipitation, leading to the formation of almost micrometric bubbles. In the centre of the pellet, these bubbles are

Fig. 2. (a) SEM BSE images (181  181 mm) with a detail area of each image. (b)

132

Xe depth profiles using SIMS

186

a consequence of the local high temperature. At the periphery, they are a consequence of the HBS formation with the very high local burn-up. Most of the missing gas (difference between measured and calculated xenon) was actually in the bubbles before sample preparation, as it is expected from the gas release fraction 2.5% and as it will be demonstrated with SIMS. SIMS Results On the same areas as the ones observed by SEM, some xenon depth profiles are obtained by SIMS. They are presented in Fig. 2b. The measurements presented here have been performed with 132Xe isotope and it was shown that it was impossible to obtain accurate measurements with other isotopes of xenon because of the interferences. A calculation demonstrates that the ratio 132Xe=(total Xe) varies between 0.204 and 0.209 from the pellet centre to its periphery [9]. We will consider that this ratio is approximately equal to 0.204, because the value of 0.209 corresponds to the extreme periphery (the RIM area) and represents a small surface. Depending on the radial position, the depth profiles are very different. At the periphery and in the central part, the depth profiles exhibit many peaks. In the mid-radius area, no peaks are detected. The results obtained with SIMS are consistent with the SEM observations. The number of peaks is in relation with the number of bubbles observed by SEM. This relation was already evidenced in [11]. Elsewhere, two components are seen: – a baseline which is associated to 132Xe solid solution, – several peaks. It has been shown [11] that the peaks correspond to the 132Xe coming from bubbles (and in solution xenon from the baseline). Using the depth profile data, the two components, baseline and peaks, can be estimated separately. On the one hand, the total inventory is defined as the integration over the depth profile of all points, on the other hand the baseline intensity is obtained after computation of the peaks size of the distribution diagram [12]. The total and baseline intensities are correlated to the 132Xe inventory and solid solution in the matrix respectively. In order to characterise this two quantities on an actual irradiated fuel, several depth profiles were performed along a radius of our sample.

J. Lamontagne et al.

For each depth profile, the total and baseline intensities were calculated. The results are presented in Fig. 1c as a function of the radial position on the pellet. The top line is the total inventory (i.e. the total intensity of xenon), the bottom line is the measurement apart from the peaks signal (i.e. the solid solution xenon in the matrix). The difference between the total inventory and the baseline is the xenon in the bubbles, i.e. approximately the xenon from the bubbles not analysed by EPMA due to the sample preparation. The EPMA xenon profile and SIMS baseline xenon profile can be compared as they can be both interpreted as an analysis of solid solution xenon with the very small bubbles. Moreover, it is shown that the HBS formation in the rim area did not induce any major gas release. Quantification of the Xenon Total Inventory The experimental results obtained with SIMS can lead to a quantitative evaluation of the xenon local concentration. The method for xenon quantification uses the mid-radius position. In this position the xenon is totally considered in solid solution because no bubbles was detected by SIMS or observed by SEM (though part of it may be in nanobubbles). In this position both EPMA and SIMS measurement are equal to the quantity created and can be directly compared and correlated. Thus, this correlation can be applied in the central part where a part of the xenon is precipitated in bubbles. In this area, the SIMS measurements show that approximately 30% of the total gas is in the bubbles. Conclusion A new method for the determination of the total inventory of xenon is proposed using SIMS. It is indeed shown that the xenon coming from the matrix and from bubbles can be detected with SIMS in irradiated nuclear fuel. A quantification method is proposed using EPMA results. This very promising method provides quantitative radial profile on irradiated nuclear fuel with an accuracy of 10%, depending on the counting statistics, the standard characterisation and the analysed surface. Experiments are in progress. The quantification of xenon in the bubbles by SIMS coupled to a study of the size of the bubbles by SEM will make it possible to estimate the pressure in these bubbles.

Fission Gas Bubbles Characterisation in Irradiated UO2 Fuel by SEM, EPMA and SIMS

References [1] [2] [3] [4] [5] [6] [7] [8]

Walker C T (1999) J Anal At Spectrom 14: 447 Ronchi C, Walker C T (1980) J Phys D 13: 2175 Verwerft M (2000) J Nucl Mat 282: 97 Lozano N, Desgranges L, Aymes D, Niepce J C (1998) J Nucl Mat 257: 78 Manzel R, Walker C T (2002) J Nucl Mat 301: 170 Rasser B, Desgranges L, Pasquet B (2003) Appl Surf Sci 203–204: 673 Desgranges L, Pasquet B (2004) Nucl Instr Meth 215: 545 Noirot J, Noirot L, Desgranges L, Lamontagne J, Blay Th, Pasquet B (2004) Fission gas inventory in PWR High burnup fuel: experimental characterisation and modelling. Proceedings of the 2004 International Meeting on LWR Fuel Performance. Orlando, Florida, September 19–22, 2004

187

[9] Noirot J, Desgranges L, Marimbeau P (2000) Contribution of the rim to the overall fission gas release: what do isotopic analyses reveal. Fission gas behaviour in water reactor fuels. NEA Seminar Proceedings. Cadarache, France, September 26–29, 2000 [10] Boidron M, Tourasse M, Boussard F, Piron J-P, Pasquet B (1991) Microprobe measurement calibration of absorbed xenon in a PWR fuel, ANS, International Topical Meeting on LWR Fuel Performance. Avignon, France, p 21 [11] Lamontagne J, Noirot J, Desgranges L, Blay Th, Pasquet B, Roure I (2004) Microchim Acta 145: 91 [12] Valot Ch, Desgranges L, Pasquet B (2005) Characterisation of irradiated nuclear fuel with SIMS (submitted) Proceedings of the 2005 Secondary Ion Mass Spectrometry XV. Manchester, UK, September 12–16, 2005