Appendix G to 10 CFR 50 requires that licensees evaluate all reactor beltline materials to determine whether the Upper Shelf Energy (USE) screening criterion will be exceeded prior to the end-of-license (EOL). Measurement of the unirradiated USE and estimation of the EOL irradiated USE level are key elements of the screening criterion evaluation. The current Regulatory Guide 1.99 (Revision 2) USE drop model does not provide adequate representation of the actual shelf drop behavior for some reactor materials. Therefore, research was conducted to identify the important factors which should be included in the development of a physically-based USE trend curve model and to test the validity of the material-specific modelling approach.
As a result of this investigation, it has been concluded that there are three radiation damage mechanisms which may have significant influence on the ductile fracture process: precipitation and particle coarsening due to radiation enhanced diffusivity; cascade induced matrix damage; and possibly surface active element transport to grain boundaries and particle interfaces. At present, it is believed that the largest micromechanical effect is due to the impact of precipitates and matrix damage on the hardening of the material bridges between voids. However, additional experimental data are needed to confirm the dominant mechanisms which contribute to the USE drop.
In the absence of an experimentally verified microstructural model, work has been performed to test the efficacy of the material-specific approach. The validity of the material-specific approach has been demonstrated by reducing the Light Water Reactor (LWR) database to produce a data set with comparable chemical composition, heat treatment, and neutron flux spectra. Based on the work reported herein, it has been observed that a drop in the unirradiated USE requires an incubation dose. The reduced data set correlates well with the square root of the dose, indicating the importance of the fine scale precipitates and the matrix damage component.