Monte Carlo N-Particle Transport Code System (MCNP) criticality calculations were performed on a library of critical benchmark experiments to obtain preliminary bias values and subcritical margins to be utilized in licensing calculations for high-level radioactive waste disposal.

The critical experiments library includes a broad range of system physical and neutronic characteristics that are representative of a range of potential criticality configurations relevant to long-term deep geological disposal. Two hundred and eighty-nine critical benchmark experiments were selected and grouped into 20 critical experiment classifications.

From the results of this study, an applicable subcritical margin or maximum allowable keff can be selected for preliminary repository criticality analysis based on the similarity between the physical and neutronic characteristics of the system being analyzed and the relevant library classification. The results of this study provide quantification of both the confidence associated with the MCNP code and the presented conservative method for performing criticality evaluations relevant to repository emplacement of high-level radioactive waste.