Evaluation of high temperature material models for high temperature advanced reactor component analysis

The high temperature environments in proposed advanced reactor designs pose unique challenges for the integrity of structural components. Capabilities to model metals in this environment are being developed in Grizzly, a finite element-based structural simulation code developed at Idaho National Laboratory. To represent the constitutive behavior of metals at high temperatures, Grizzly offers two distinctly different approaches. The first is to use phenomenological engineering models with a unified representation of creep and plasticity. These models are provided through a library developed at Argonne National Laboratory. These models represent the effects of multiple important phenomena on the aspects of the material response important to engineering analysis and are calibrated to engineering-scale material data. The second modeling approach is being developed at Los Alamos National Laboratory and is based on mesoscale models that directly account for the effects of material texture. These models directly represent the underlying physical phenomena that affect the engineering-scale material response. Through the use of data analytics, reduced order models based on the mesoscale models have been developed. These models can be efficiently used in engineering calculations, while faithfully representing the behavior of the high fidelity mesocale models. This paper provides brief summaries of both modeling approaches, and demonstrates their application on a common set of benchmark problems to assess their applicability to component-level simulations.