Numerical Monte Carlo analysis of the void coefficient in Pavia TRIGA Mark II reactor.

In nuclear reactor safety assessment studies, void formation in the coolant, both in nominal operation and in accidental scenarios, must be considered since bubble nucleation can lead to fast reactivity changes, thus putting a strain on the reactor control. The evaluation of such a phenomenon is not trivial due to the non-linearities related to the competing phenomena (e.g. neutron absorption and scattering) and the spatial effects. This work aims to analyse the impact of void formation on the multiplication factor in the Pavia (Italy) TRIGA Mark II reactor, focusing on spatial effects. A model of the TRIGA has been developed using Serpent Monte Carlo code and validated against experimental measurements conducted at the Pavia reactor. Two approaches have been adopted. The first consists of directly evaluating the multiplication factor of several configurations, each featured by a different water density to mimic a homogeneous void formation. The second approach consists of a perturbative procedure, i.e., a first-order sensitivity analysis, which allows the gathering of more information on the aforementioned competing phenomena. Both cases subdivide the core into several radial and axial regions to recover the spatial effect. The two approaches appear to be complementary in the information they provide. The results show that the void coefficient is negative in the core and strongly dependent on both position and void fraction. Moreover, the results show the strong influence of the fuel elements type (101 and 103), their location, and the experimental methods adopted on the void coefficient. • Monte Carlo analysis of void feedback coefficient is performed. • The sensitivity tool has been exploited. • Experimental data have been used to validate the numerical model. • Experimental procedures for assessing void coefficient have been analysed. [ABSTRACT FROM AUTHOR]