On the Physics of a Core Disruptive Accident in a Heavy Liquid Metal Fast Reactor - Case Study: MYRRHA

A fast-spectrum reactor core is not designed to operate in its most reactive configuration. As a consequence, the reactivity of such system is sensitive to changes in geometry and/or the rearrangement of fuel material. It is therefore possible that a core degradation event results in a super-promptcritical reactivity transient, excessive power buildup and the hydrodynamic disassembly of the reactor core. This sequence, referred to as a Core Disruptive Accident (CDA), has traditionally been analysed for public consequence considerations in Sodium Fast Reactor (SFR) technology. In the framework of this research, a CDA occurring in a Heavy Liquid Metal Fast Reactor (HLMFR) is investigated in detail, with a particular focus on the Multipurpose hYbrid Research Reactor for High-tech Application - MYRRHA.

A core degradation sequence in an HLMFR may involve a change in geometry and/or the rearrangement of fuel material and, as such, holds the potential to lead to a CDA. However, the high boiling point of Heavy Liquid Metal (HLM) coolant implies that no extensive coolant vaporisation is expected to take place during the core degradation sequence and that the coolant remains present in the system. Furthermore, the similar densities of oxide fuel and HLM coolant imply that the gravitational collapse of the fuel (traditionally assumed to lead to fuel compaction and the associated increase in system reactivity in an SFR) can hardly lead to a sustained reactivity insertion at a high rate. These considerations leave important doubts about the possibility and probability of a CDA taking place in an HLMFR.

In the absence of a tool suitable to provide a reliable assessment of core degradation scenarios in an HLMFR, a coherent fuel compaction leading to a CDA is postulated in order to envelop all the sequences that may take place following a core degradation event. This is done to assess the impact and implications of a CDA on the structural integrity of confinement structures and, in doing so, to support the safety demonstration of MYRRHA and the HLMFR technology.

Starting from fundamental laws and principles of physics, a CDA in an HLMFR is investigated in order to obtain an understanding of the governing physical mechanisms, provide an estimate of the associated fission energy release and examine its dependence on different system and accident parameters. To do so, a carefully selected set of mathematical models is established, and a dedicated multiscale, multiphysics simulation tool is developed to obtain its solution.

The assessment of a range of mechanisms postulated to lead to reactivity insertion in a degraded reactor core of an HLMFR has demonstrated that coherent fuel compaction by forced coolant flow holds the potential to insert reactivity at the highest rate, which amounts to ~150 $𝑠⁄. In comparison, the gravitational collapse of the fuel in an SFR leads to a reactivity insertion rate in the range of ~50 $/s to ~60 $/s.

The dynamics of the power buildup that occurs following the achievement of promptcriticality is determined by a variety of system and accident parameters. Owing to the thermophysical properties of the fuel material (i.e., low thermal diffusivity), the majority of deposited fission energy is retained within the fuel, with only a fraction (of the order of a few %’s of the total fission energy input) transferred to the surrounding HLM coolant. The phenomena that occur in the fuel therefore play the most important role in determining the overall dynamics of the transient. A particularly important event during a CDA in an HLMFR is the melting of the fuel, the rapid expansion of which has a significant effect on the course of the accident.

The thermophysical properties of the HLM coolant (i.e., high volumetric expansion coefficient) imply that the limited amount of heat transferred to the coolant nevertheless results in its important thermal expansion. This expansion provides a substantial contribution to the overall expansion of the degraded core and further contributes to the introduction of the associated reactivity effect. Heat transfer phenomena are therefore identified to play an important role in determining the overall dynamics of a CDA sequence in an HLMFR.

The potential presence of non-condensable gas in a degraded reactor core, attributed to the gaseous fission products and filling gas initially contained in fuel pins, is furthermore demonstrated to have an important impact on the dynamics of system expansion and the associated reactivity effect. Before the effective displacement of the fuel material can take place, the material expansion must initially occur internally in order to compensate for the presence of the non-condensable gas. This implies that almost no net expansion of the degraded system takes place before the non-condensable gas is sufficiently compressed, thus introducing a delay in the reactor core expansion.

Following the identification of the most important phenomena that govern the accident sequence, the developed simulation tool was employed to quantify the impact of different system and accident parameters on the fission energy release during a CDA. An extensive application of this model has helped identify the phenomena and parameters to which the energy release is most sensitive. These include the concentration of non-condensable gas in the degraded reactor core, initial system power and the reactivity insertion rate.

In addition to providing a fundamental understanding of the physics of a CDA in an HLMFR, this research demonstrated that the associated fission energy release, normalised to the nominal power level of the reactor core, ranges from ~10 J/W to ~160 J/W. If estimated by models of similar maturity, the normalised fission energy release in an SFR is an order of magnitude higher, thus seemingly proving an advantage of the HLMFR technology in terms of its response to a CDA.

The potential of a CDA to inflict damage on the primary system of MYRRHA is assessed in terms of the mechanical energy conversion factor (i.e., the fraction of fission energy released during a CDA transformed into mechanical energy contained within the primary system) and the increase of static pressure in the primary system. A low value of the mechanical energy conversion factor means that only a limited fraction (i.e., ~0.1 %) of fission energy released during a CDA remains available to mechanically load the primary system. The pressurisation of the primary system following the worst-case CDA, on the other hand, yields pressure levels that may not be covered by the current design requirements but are likely not to impair the structural integrity of the primary system. The outcome of this research therefore hints at the possibility of successful confinement of the nuclear fuel following a core degradation event, thus providing a strong basis for a successful safety demonstration of MYRRHA and the HLMFR technology in general.