Adjusting aluminum elastic scattering data from the TALYS nuclear reaction code based upon fast criticalities

•Testing aluminum data from TALYS.•Selecting multiplication factors of fast-spectrum benchmarks sensitive to aluminum.•Prior results unsatisfactory for these systems having aluminum-based reflectors.•Adjusting aluminum elastic scattering data using consistent covariance data.•Improvement achieved. The suitability of aluminum data produced by means of the TALYS nuclear reaction code, which differently from other libraries is largely based upon the use of theoretical models in the fast energy region, is analyzed by using acceptable benchmark data, indicating severe shortcomings. More precisely, this data is tested based upon a complete sequence of thirteen effective multiplication factors for well documented benchmark cores of fast-spectrum systems reflected by aluminum taken from the International Handbook of Evaluated Criticality Safety Benchmark Experiments (ICSBEP) database which are firstly computed by means of a stochastic technique based upon the Serpent code. The same stochastic approach is then used to perform adjustments in conjunction with the Asymptotic Generalized Linear Least-Squares (AGLLS) assimilation methodology by including the angular distribution of the scattered neutron in the adjustment and considering two suitable assimilation databases which consist of three of these effective multiplication factors, with the remaining ten cores being treated as target setups, in each case. These adjustments of aluminum data only, provide improvements of the calculation results showing decreased mean χ2 and variance of the thirteen individual χ2 s, with a maximum simultaneous reduction factor in the order of 3 to 4 for both parameters. It is questionable whether the strong adjustments coming out as an important result of the analysis may have physical meaning; this issue cannot be answered by this study alone.