The 4^th ASTEC international Users’ Club was organized by IRSN and GRS and hosted by GRS in Köln on October 11-14. During this 4-day workshop, discussion took place on the progress of the technical work done in the WP4 “ASTEC” between IRSN-GRS and 45 users from 24 organizations (some of them being non–SARNET partners, like Kurchatov Institute in Russia). Twenty-seven presentations were performed, along with an on-line demonstration on the ATLAS graphical post-processing tool.

Discussions focused mainly on the assessment of the ASTEC V2.0 version versus experimental data (the Deliverable on the synthesis of this work has been released by the end of 2010). It covered the experiments shown in the following table, plus the 4 phases of the TMI2 accident.

The main conclusions are summarised hereafter:

• For primary and secondary circuit thermal-hydraulics: good results but modelling shortcomings on a PACTEL-ISP33 integral test (VVER design).
• For core degradation: good results for early-phase models (core heat-up, oxidation and hydrogen cumulated production) at least up to the final quenching, if any. Conversely, the large hydrogen peaks observed during the quenching late phase of CORA-13 and LOFT-LP-FP2 tests and TMI2 are not reproduced. For late phase models, good results have been obtained on Phébus FPT4 (using the magma-debris models in the core region) and LIVE-L1.
• For FP release: very good results that underline a significant improvement with respect to previous ASTEC V1 versions, in particular on release of semi-volatile FPs and SIC (Silver-Indium-Cadmium) materials from control rods.
• For FP/aerosol transport in the primary circuit: reasonable results. Anyway, the crucial importance of gas phase chemistry has been underlined again in the Phébus FPT1 applications, in particular for iodine behaviour. Indeed, though the development of a detailed gas chemistry modelling made the assessment much more complex, SOPHAEROS has proven to be a very useful tool for such analyses of aerosols and vapour behaviour simultaneously with speciation. Work will continue in that direction, in particular with the validation of the new ASTEC V2 chemistry kinetics modelling on the CHIP IRSN experiments (in ISTP frame).
• For the containment: good results on thermal-hydraulics, behaviour of dry aerosols, and gas combustion in lean atmosphere. To confirm such very promising results, further work has to be done by validating this model against experiments with different conditions like combustion in steam enriched atmosphere and with downward directed flame propagation. Conversely, much less good results about the retention of aerosols in pools have undermined the need of model improvements.
• For DCH: reasonable results but the models are too parametric and too geometry-dependent.
• For iodine in containment: good reproduction of global trends like the effect of pH or silver on iodine volatility but nevertheless in terms of quantity, the RI concentration in gas phase is often underestimated. In spite of that, the RI production model in liquid phase seems to be adequate to reproduce experimental data.
• For MCCI: models at the State of the Art but they showed (like all other codes) a lack of predictability. Anyway, the good agreement with experimental data shows basically the relevance of the set of assumptions and models used; it means that a comprehensive approach using the ASTEC V2 models seems to be able to reproduce the MCCI phenomena on diverse concrete types.

In summary, the main modelling efforts must focus in priority on the reflooding of degraded cores (in particular the corresponding hydrogen production), on MCCI (in particular on the coolability aspects), and in lower priority on DCH and on pool-scrubbing phenomena in the containment. The ASTEC V2 assessment work is continuing during the SARNET2 2nd current period, using the successive code revisions.

Contact: Patrick Chatelard
patrick.chatelard@irsn.fr