Quantitative analysis of the fission product distribution in a damaged fuel assembly using gamma-spectrometry and computed tomography for the Phebus FPT3 test

摘要

The international Phebus FP programme, initiated in 1988 by the French "Institut de Radioprotection et de Surete Nucleaire" (IRSN), in cooperation with the European Commission (EC) and with financial support from USNRC, Canada, PSI/HSK (Switzerland), Japan and Korea, was aimed at studying severe accident phenomena: the fuel degradation, the release of fission products (FPs) and their transport through the reactor coolant system to the containment building. The FPT3 test, conducted in 2004, was the last of the five light water reactor core meltdown accident tests performed on irradiated fuel rods. After the experiment, the test device was recovered and analysed through a full set of non-destructive examinations performed over the fuel bundle zone, including gamma-scanning, gamma emission tomography, X-ray radiography and X-ray transmission tomography. The gamma-scanning was specifically devoted to the location, identification and amount quantification of the FPs remaining in the bundle. Since the fuel bundle became highly degraded during the experiment, the geometry was different at each level examined, and did not correspond to the well-known initial state. The self-attenuation of the test device and consequently the efficiency correction could then not be estimated by classical means that need to know the geometry of the object. Using the results of the other non-destructive examinations, specific computational tools and methods have therefore been developed to compute the self-attenuation of the bundle (locate and identify the materials and estimate their density with the X-ray tomograms, locate the FP distribution inside the bundle with the gamma emission tomograms) and to automate the processing of the gamma spectra acquired. The specificities of these gamma spectra (high count rate, number of gamma rays, number of measurements, etc.) required in particular to analyse key lines only and needed an original counting loss correction. The method was validated over the pre-test examination of the fuel bundle, through a comparison with the classical gamma analysis method used at the laboratory for objects of known geometry. The final results, given with acceptable uncertainties, gave for all FPs identified (mainly ~(137)Cs, ~(131)I, ~(132)Te, ~(140)Ba, ~(95)Zr, ~(103)Ru, etc.) their quantitative activity profile along the bundle, their retained and released fractions in the bundle, and also some information about their relocation inside the bundle. The results are in very good agreement with other Phebus FPT3 measurements and inventory calculations.