Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Zhigang Mei Project Name: DFT-defects-UO2 Division: NE Project title: Nanoscale Modeling of Defects in Materials for Energy Applications Associated funding: ANL LDRD Other Systems: Carbon: 300000 core-hours Eddy：No time limit Science: The technological importance of actinide dioxides, AO2 (A=U, Pu or Np), largely originates from their application in nuclear reactor fuels. Today many reactor use UO2 as the primary fuel component. High-energy particle (self)irradiation in materials causes damage by creating point defects that can form defect structures such as small clusters, voids, and stacking faults, ultimately leading to material failure. The damage tolerance of a material depends not only on the number of defects created during irradiation but also on their long-time kinetic evolution. Therefore, understanding the evolution of highly interacting defects is very important. Experimentally is difficult to capture the irradiation-induced short time-scale atomistic process; computer simulations are particularly adept at capturing such dynamical process. Evaluating the concentration of point defects in actinide oxides requires a complex evaluation of the energy of formation of each defect type. The traditional mass action law is widely used to calculate the concentration of lattice defects. This formalism assumes that the defects do not interact, and that each defect occupies only one lattice site. However, complex defect structures can occupy multiple sites and the strong Coulomb and elastic repulsions may exclude other complexes from nearby sites. This is a source of great uncertainty in any statistical description, especially when the defect complexes are anisotropic. Roughly, the concentration of point defects increases with the temperature and the departure from stoichiometry, and decreases with pressure. Uranium oxide, UO2+x, which is the basis for most nuclear fuels used in Light Water Reactors, has a fluorite structure and hosts a large set of point defects to include oxygen vacancies, uranium vacancies, oxygen interstitials, as well as more complex defects such as small polarons or Willis defects. In UO2+x, uranium can exhibit valence changes from U4+ to U5+ leading to a wide range of departure from stoichiometry x, up to UO2.3. As first shown by neutron diffraction studies, this is mainly due to excess oxygen ions that are incorporated at interstitial sites and to a lesser degree to uranium vacancies. However, most current experimental and theoretical treatments of uranium oxide focus on isolated, non-interacting defects, which is consistent with small departures form stoichiometry, usually x < 0.1, and provides good descriptions of the thermochemistry of the oxide at low temperatures. The purpose of this project is to develop an understanding of defect formation in UO2+x at high defect concentrations and departures from stoichiometry, 0.1 < x < 0.3. Such an understanding will allow us to develop models of the free energy of the oxide that are valid at high temperatures, consistent with potential accident scenarios in the nuclear reactor. This study involves a comprehensive analysis of the oxide electronic and defects structure at nano-scale and using Density Functional Theory (DFT) and Ab-Initio Molecular Dynamics (AIMD). Since UO2+x is thermodynamically stable over a wide range of oxygen content in which the nature of the majority of defects may change greatly, we will examine the free energy of formation of a variety of defects types using DFT and AIMD. As opposite to previous studies, we will examine the particular cases of closely interacting defects and determine the energetics of these structures. For example, at large oxygen contents, where oxygen interstitial migration is preponderant, we will calculate the formation and activation energy for oxygen atoms diffusion as function of temperature and defect concentration. We will validate the advanced QM methods by predicting lattice constants, thermodynamics and other properties of compounds in the U-O phase diagram. Project description: To achieve the objectives of the study we will run standard density-functional theory (DFT)-based packages such as VASP and CP2K. We propose to apply these tools to model defects in non-stoichiometric uranium oxide, UO2+x (0.1<x<0.2), for the case of high defect concentrations. In addition to defect formation and migration energies, we will calculate entropic contributions to the free energy of various interacting defects to evaluate properties that are suitable for experimental validation, such as oxygen self and chemical (tracer) diffusivity in UO2+x. The LCRC’s high-performance computing cluster Fusion provides an outstanding combination of cutting-edge computational power and the efficient DFT software needed for this project. Due to the strongly correlation of U-5f electrons, more accurate methods, such as the LDA+U method and hybrid functionals, are required to overcome the limitations of conventional LDA/GGA method. These new methods require an order magnitude more of computational time to study the defects formation in UO2+x by DFT calculations. To go beyond point defects and study defect clustering and voids, we will use the AIMD capability of VASP and CP2K to study the formation and evolution of these defects in realistic time scale. With its well-developed pseudopotentials for actinide elements, VASP is extremely adept to study the static behaviors of a system with only a few hundred atoms. In the case of UO2, most of the published DFT works were limited to systems with less than two hundred atoms due to strongly correlated f-electrons. Additional complexity is the multiple metastable states accompanying DFT+U calculations for f-electrons. Approaches, such as monitoring f-electron occupation matrices and U ramping, are required to achieve the correct electronic ground state. These approaches usually require ten times of computational time than standard DFT+U calculations. This may not pose a significant problem for studying point defect formation or migration energies under the assumption of dilute defect. In additional to the formation energies of defects, we will also calculate the vibrational entropy of defect formations and effective vibrational frequencies of diffusing atoms from phonon calculations . These provide all factors entering the vacancy/interstitial-mediate self-diffusion coefficient by first-principles methods. To study the dynamics of defects in highly defected UO2, large cell with more than a thousand atoms is indispensable, which is far beyond the capability of VASP code. By exploiting linear scale method, CP2k was demonstrated to model a system with one million water molecules. Our benchmark calculations show that UO2 with a few thousand atoms can be well handled by CP2K. We will calculate the structure and thermodynamics of UO2 to establish the reliability of U GTH pseudopotential and the hybrid functional. Based on these results, we will model oxygen Frenkel pairs and Schottyky defects with the pseudopotential and functional as an additional check. Once the reference calculations are complete, we will calculate the free energy surface for the formation of the Oxygen Frenkel pair in UO2+x with metadynamics. Three person will be involved in this projected. Dr. Marius Stan is a Senior Scientist in the Nuclear Engineering Division at ANL and a Senior Fellow of the Computational Institute at University of Chciago. Dr. John Low is a Computational Scientist in Mathematics and Computer Science division at ANL. Dr. Zhi-Gang Mei is a postdoc in the Nuclear Engineering Division at ANL. Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 400000 Q1: 100000 Q2: 100000 Q3: 100000 Q4: 100000 Justification: Thank You, The LCRC Accounts System