Analysis of zircaloy rewetting experiments and their implication for reactor fuel

This report describes the analysis of single-tube rewetting experiments in the Reflex rig. The primary aim was to apply the Grenoble k2-rewetting model to study rewetting of zircaloy tubes with controlled oxide-layers up to 75 μm thick and develop a correlation of the parameter k2 and assess the implication of these experiments to reactor fuel. In the absence of a two-dimensional conduction code at Winfrith containing the k2-heat flux boundary condition, it was decided to modify three existing codes TRAC-PD2, QUENCH and REWET to incorporate such a boundary condition, the first being a transient code and the others quasi-steady-state. All three codes are now satisfactorily validated and have contributed to the analysis of the REFLEX data. It was found that the Grenoble correlation for k2 was unable to predict the data particularly at the lower powers and sub-coolings. A new correlation has therefore been developed which involves the rewetting temperature as well as the system variable pressure, mass flow rate and sub-cooling at the quench front. In the higher power ERSEC experiments at Grenoble the new correlation should reduce to the original k2-correlation as the rewetting temperatures are thought to be nearly constant. The code(s) have also been used to study the effect of gas-gap and oxide layers on PWR fuel-pin rewetting