On using alternative perturbation theory methodologies in the evaluation of reactivity effects in ADS cores

Core-average Doppler and coolant void reactivity coefficients, as well as the kinetic parameters (βeff and Λ), have been determined for sub-critical accelerator-driven systems employing lead–bismuth eutectic (LBE) and helium gas coolants. To determine these parameters use is made of the standard procedure for analyzing critical reactors, which is based on “perturbation-theory” (PT), while in addition two dedicated methodologies for sub-critical systems, i.e. “inhomogeneous perturbation-theory” (IPT) and “heuristically based generalized perturbation-theory” (HGPT), have been employed to compute these parameters in a more rigorous manner. The two methods (PT and IPT/HGPT) are found to give similar results for each application and despite a smaller target keff-value, the sensitivity of the method is small in the case of the gas-cooled system, thus confirming the adequacy of the standard procedure. As regards the coolant void reactivity coefficient in the gas-cooled ADS, this finding can mostly be attributed to the fact that the core is always transparent with respect to the source neutrons, irrespective of the specific helium content. The sensitivity of the Doppler coefficient is also rather low in the case of the LBE cooled system. However, the dedicated methods are needed for the correct prediction of the coolant void reactivity coefficient, especially if minor actinides are introduced into the core. More important, in this case, is the fact that the PT-approach does not produce conservative results. Finally the sensitivity of the reactivity and kinetic parameters to the different methods is of the same order as that due to uncertainties in nuclear data and therefore these will need to be included in any overall evaluation of the impact of uncertainties on steady-state and transient ADS performance.