Modeling of Rayleigh-Taylor instability for steam direct contact condensation

Passive systems are commonly employed in nuclear reactors to enhance safety during postulated accidents. However these systems, such as direct contact condensation, might introduce complex and unexpected phenomena, possibly reducing the efficiency in case of long transients. A similar event is expected during the Fukushima Daiichi accident in the suppression pool, where the reduction in steam condensation might have resulted in local steam by-pass increasing the containment pressure. Detailed analyses of such complex phenomena are therefore necessary and various attempts have been performed in the past to develop related two-phase two-fluid Computational Fluid Dynamics (CFD) models. The most successful CFD investigations of the related phenomena focused on the definition of a heat transfer coefficient from the water side, based on the surface renewal theory and the turbulent scales. Little effort was however attempted to add a general model of the interfacial area based on theoretical and experimental evidences. In the present work it is proposed a method to treat the surface with growing instabilities based on the Rayleigh-Taylor theory, leading to the break-up of the surface and triggering steam depressurization typical of chugging phenomena. The preliminary results of the implemented technique have demonstrated the capability to predict the characteristic phases of chugging (bubble collapse, depressurization and steam flow upward in the pipe) with a considerable improvement in the prediction compared to previous investigations. The result assumes remarkable interest in the perspective of the employment of multiphase CFD for the prediction of the water temperature distribution within the suppression pool since the presence of chugging introduces large water mixing at the onset of every water discharge cycle.