Nuclear Technology / Volume 163 / Number 2 / August 2008 / Pages 209-227
Technical Paper / Reactor Safety / dx.doi.org/10.13182/NT08-A3982
Analyses of severe accidents in nuclear power plants by using integral codes are necessary in order to develop accident management strategies that prevent such accidents or mitigate their consequences for the environment. The most important requirement for the development of integral codes is to achieve good predictability of a given accident scenario through the understanding and quantification of severe accident phenomena and their underlying physical and chemical processes. In this paper, the progress in modeling the processes related to the radioactive source term, and in particular progress related to the release and transport of fission products in the circuit and containment, is demonstrated by the assessment of integral and detailed codes using the experimental results of the in-pile Phebus fission product tests (FPTs). It is shown that the integral codes are good in predicting both the hydrogen release and the total release of volatile fission products from the bundle.
It is also shown that the commonly used fission product transport codes overestimate the deposited aerosol mass in the Phebus steam generator. However, by using an improved model for the thermophoretic aerosol particle deposition, it has been possible to reproduce the aerosol mass deposited in the steam generator more accurately. The containment analyses carried out with both lumped-parameter and multidimensional computational fluid dynamics codes showed that the measured thermal-hydraulic data are accurately reproduced. The aerosol behavior in the containment estimated from the lumped-parameter codes corresponded satisfactorily to the experimental data. The iodine chemistry codes highlighted the substantial role of silver released from the degraded absorber rod (Ag-In-Cd), as it was observed experimentally; however, the temporal dependence of the gaseous iodine concentration in the containment atmosphere was poorly calculated. There are plans to improve the modeling in order to reproduce better the fission product release from the bundle, the fission product transport in the primary circuit duct, and the gas phase chemistry in the containment, with particular emphasis on gaseous iodine species. Further plans include the analysis of Phebus FPT3, which was the last in the series of Phebus tests, with its boron-carbide control rod.