Abstract
Uranium nitride fuel is considered for fast reactors (GEN-IV generation and space reactors) and for light water reactors as a high-density fuel option. Despite this large interest, there is a lack of information about its behavior for in-pile and out-of-pile conditions. From the present literature, it is known that C and O impurities have significant influence on the fuel performance. Here we perform a systematic study of these impurities in the UN matrix using electronic-structure calculations of solute-defect interactions and microscopic jump frequencies. These quantities were calculated in the DFT+U approximation combined with the occupation matrix control scheme, to avoid convergence to metastable states for the 5f levels. The transport coefficients of the system were evaluated with the self-consistent mean-field theory. It is demonstrated that carbon and oxygen impurities have different diffusion properties in the UN matrix, with O atoms having a higher mobility, and C atoms showing a strong flux coupling anisotropy. The kinetic interplay between solutes and vacancies is expected to be the main cause for surface segregation, as incorporation energies show no strong thermodynamic segregation preference for (001) surfaces compared with the bulk.
| Original language | English |
|---|---|
| Article number | 100505 |
| Journal | Physical Review B |
| Volume | 95 |
| Issue number | 10 |
| DOIs | |
| State | Published - Mar 27 2017 |
| Externally published | Yes |
Funding
The authors acknowledge support from the UIUC-KTH collaboration project INSPIRE. The authors gratefully acknowledge discussions with Maylise Nastar, Luca Messina, Dallas Trinkle, and Pascal Bellon about the SCMF method. Work by T.S. was supported by the U.S. Department of Energy (DOE), Office of Science, Basic Energy Sciences (BES) under Award DE-FG02-05ER46217. Michel Freyss' help on the OMC scheme was very important for the completion of this article, and we are grateful to him, Emerson Vathonne, and Marjorie Bertolus at CEA/DE, and Boris Dorado at CEA/DAM for the implementation of this method in VASP. Funding from Svensk Krnbrnslehantering AB is acknowledged. The computations were performed on resources provided by the Swedish National Infrastructure for Computing (SNIC) at PDC and NSC.