Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: John J. Low Project Name: polarons Division: CLS Project title: Polarons in Urania Associated funding: DOE Other Systems: Science: Safe nuclear power is important for the country’s energy security. The Fukushima Daiichi nuclear disaster in Japan demonstrated that little is known about how nuclear fuel materials behave during a meltdown of the core in a nuclear reactor. A recent review stated that “because of measurement difficulties and lack of calibration standards, properties of liquid UO2 that are important for reactor-safety assessments still have large uncertainties.” [Fink, J. Nucl. Materials, 279(2000)1]. A study of severe nuclear accidents by the French Atomic Energy Commission show that a better knowledge of “corium”, a lava-like mixture of primarily of molten fuel and its zirconium cladding, is fundamental to understanding and modeling this kind of accident. The objective of this research is to develop a predictive atomistic model of hot solid and molten nuclear systems. The lack of detailed experimental results obtained from costly and challenging experiments at extreme physical conditions makes advanced modelling techniques essential tools for gaining further insight into the material behavior. A recent study of the structure of molten uranium dioxide with X-Ray diffraction and molecular dynamics (MD) by Skinner, et al [Science, 346(2014)984] demonstrated that after fitting the parameters of a MD simulation to the X-Ray structure factors, excellent agreement was obtained between theory and experiment. While it is encouraging that a MD simulation can be fit to the long order of molten Urania, variable stoichiometry and polarons play an important role in the thermodynamics of urania near 3138K, the melting point of urania [Winter and MacInnes, J. Nucl. Mat. 137(1986)161]. These terms were not included in the MD simulation of Skinner, et al. We propose to develop a reactive force-field, to model variable stoichiometry and polarons in high temperature urania. This model will include the physics for an accurate description of urania, which can predict the structure, thermodynamics and properties of urania. Project description: The charge optimized many body (COMB) force-field of Li, et al. [J. Phys.: Condens. Matter 25(2013)50401 provides a fair description of the low temperature thermodynamics, polymorphs and defects of urania. Since the fitting database did not explicitly include the effect of excited states, the high temperature density, lambda transition and melting temperature were not correctly predicted. Polarons in urania involve the disproportionation of U(IV) to U(III) and U(V) and other oxidation states. We proposed to add these excited state atoms as species in the COMB force-field of and use the “Free Hopping Approximation” [Yakub, et al. Journal of Nuclear Materials 389 (2009) 119] to include the effect of polarons in molecular dynamics calculations in LAMMPS. We will need to run 500 1 nansecond trajectories of unit cells containing 10,000 atoms to span the relevant temperatures and compositions. We estimate that this will require 100,000 core hours. We propose to use hybrid density functional to predict the energy of polarons. These energies will be needed to parameterize the force-field for excited states. Hybrid functional density functional theory can be used to make accurate predictions of excited states of materials without an adjustable parameter for every element [Erhart, et al., Phys.Rev.B 90(2014) 035204]. Comparison of results from CP2K and ABINIT (and/or Dorado’s version of VASP) will be made. Gaussian basis sets, efficient implementation of hybrid functionals and excellent parallel scaling make CP2K an efficient choice for hybrid functionals. However, it is difficult to determine how basis set size and unit cell size in CP2K will influence the accuracy of predicted excitation energies. ABINIT (and Dorado’s version of VASP) include occupation matrix control which will be needed to converge to the desired electronic state [Dorado, et al. J. Phys.: Condens. Matter 25(2013)333201]. The results from the plane-wave code will serve as benchmarks for CP2K to determine what level of basis set is needed for accurate polaron energies. CP2K will be used to generate the data for the fitting database for the COMB potentials. The program GARField will be used to fit the COMB force-field to a database of calculated energies for urania polymorphs, defects and polarons [Jaramillo-Botero, et al., J. Chem. Theory Comp. 10(2014)1426]. The CPU time require to generate energies for polaron formation and fit the force-field will require 200,000 hours. There will be four project members in this project Marius Stan(NE), Alejandro Lopez-Bezanilla (MSD) and Laura Ratcliff (ALCF). Industry partnership: Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 300000 Q1: 75000 Q2: 75000 Q3: 75000 Q4: 75000 Justification: Storage requirements: Thank You, The LCRC Accounts System