Sensitivity and uncertainty in the effective delayed neutron fraction ()

Precise knowledge of the effective delayed neutron fraction ( β eff ) and the corresponding uncertainty is important for nuclear reactor safety analysis. The interest in developing the methodology for estimating the uncertainty in β eff was expressed in the scope of the UAM project of the OECD/NEA. The sensitivity and uncertainty analysis of β eff performed using the standard first-order perturbation code SUSD3D is presented. The sensitivity coefficients of β eff with respect to the basic nuclear data were calculated by deriving Bretscherʹs k-ratio formula. The procedure was applied to several fast neutron benchmark experiments selected from the ICSBEP and IRPhE databases. According to the JENDL-4.0m covariance matrices and taking into account the uncertainties in the cross-sections and in the prompt and delayed fission spectra the total uncertainty in β eff was found to be in general around 3%, and up to ∼ 7 % for the 233U benchmarks. An approximation was applied to investigate the uncertainty due to the delayed fission neutron spectra. The β eff sensitivity and uncertainty analyses are furthermore demonstrated to be useful for the better understanding and interpretation of the physical phenomena involved. Due to their specific sensitivity profiles the β eff measurements are shown to provide valuable complementary information which could be used in combination with the criticality ( k eff ) measurements for the evaluation and validation of certain nuclear reaction data, such as for example the delayed (and prompt) fission neutron yields and interestingly also the 238U inelastic and elastic scattering cross-sections.