High-throughput design of low-activation, high-strength creep-resistant steels for nuclear-reactor applications

Reduced-activation ferritic/martensitic steels are prime candidate materials for structural applications in nuclear power reactors. However, their creep strength is much lower than that of creep-resistant steel developed for conventional fossil-fired power plants as alloying elements with a high neutron activation cannot be used. To improve the creep strength and to maintain a low activation, a high-throughput computational alloy design model coupling thermodynamics, precipitate-coarsening kinetics and an optimization genetic algorithm, is developed. Twelve relevant alloying elements with either low or high activation are considered simultaneously. The activity levels at 0–10 year after the end of irradiation are taken as optimization parameter. The creep-strength values (after exposure for 10 years at 650 °C) are estimated on the basis of the solid-solution strengthening and the precipitation hardening (taking into account precipitate coarsening). Potential alloy compositions leading to a high austenite fraction or a high percentage of undesirable second phase particles are rejected automatically in the optimization cycle. The newly identified alloys have a much higher precipitation hardening and solid-solution strengthening at the same activity level as existing reduced-activation ferritic/martensitic steels.