Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: John Russell Project Name: uo2_md Division: MSD Project title: Thermal conductivity of uranium oxide with embedded carbon nanotubes Associated funding: Other Systems: Science: Heat dissipation is an important factor in safely using and storing nuclear fuels. Uranium oxide has a low thermal conductivity which can allow temperatures to rise quickly. One strategy to improve heat flow in uranium oxide is to embed carbon nanomaterials in the oxide. We investigate by molecular dynamics the phonon thermal conductivity of uranium dioxide with embedded carbon nanotubes. In initial calculations, we find a three fold increase in thermal conductivity of a 50 nm uranium oxide model system with an embedded (10,0) CNT. Real values will likely be higher since phonon confinement effects in the relatively small model system result in systematically lower values. Project description: Uranium and uranium oxide spent fuels are stored in water to safely maintain temperatures well below the melting point. Power disruptions from a tsunami interupted cooling of fuel rods at the Fukushima Daichi power plant resulting in a meltdown [1, 2]. Increasing thermal conductivity of fuels will result in a faster cooling rate, which could improve safety. Embedded carbon nanotubes (CNTs) or nanofibers might be a strategy to improve thermal conductivity in bulk UO2. CNTs have large surface areas, high mechanical strength, chemical stability, and thermal conductivity which makes them an intriguing candidate for design of a hybrid material. The goal of this project is calculate structural and thermodynamic properties of UO2 and hybrid materials where experimental values may be unavailable. Our initial approach will be to use a partially ionic model (PIM) of the Born-Mayer-Huggins (BMH) potential to model the UO2 material [3, 4]. The PIM features a reduced ionicity with less than full ionic charge on U and O atoms, and it matches the UO2 lattice constant at temperatures over 1000K, which a full ionic model gets wrong. Likewise, electrostatics will be calculated with a linear-scaling Wolf summation. For coupling terms, C-C interactions will be handled by means of a Tersoff potential [5], and the remaining U-O and U-C interactions will be handled with Morse potentials fit to density functional theory (DFT) reference calculations. With respect to thermodynamic proprerties, we will calculate the thermal conductivity by means of the Muller-Plathe [6] method and the Green-Kubo method. In this way, our approach will capture the salient features of the problem, including ionicity of the UO2 and covalency of the CNTs. Our model systems will include free particles, surfaces and nanocrystalline materials. To calculate properties of our model system, we will conduct molecular dynamics simulations using LAMMPS and DFT calculations will be using CP2K and/or VASP. This modeling approach is well suited for large systems due to the favorable scaling of the potentials and software. LAMMPS is massively parallel with nearly linear scaling with system size which can allow for 100k+ atom models for tens of nanoseconds [7]. CP2K is an electronic structure package with good efficiency for serial and parallel systems and excellent accuracy. [8,9] Depending on the size of the model system, it may make more sense to use VASP with a larger number of K-Points. For this reason, we also include VASP, which is a standard electronic structure package with reasonable parallel scaling up to hundreds of cores. Our team has 5 members: John Russell (MD calculations), John Low (DFT calculations), Peter Zapol, Petr Kral, and Larry Curtiss providing scientific leadership and oversight. John Russell is also collaborating with John Low, Marius Stan, Jianwei Sun, and James Furness on a project related to testing the SCAN functional to see if it can avoid problems with metastable states in uranium oxides. So far, extensive DFT calculations have been conducted with the Abinit and VASP software packages. Conventional DFT calculations with uranium oxides are very challenging and calculations are often stuck in metastable low lying states. There may be some fundamental reasons why certain DFT functionals perform better than others. [1] D. Cyranoski, G. Brumfiel Nature 477, 139-140 (2011). [2] J. Medalia Congressional Research Service, R41728 (2011). [3] T. Arima, S. Yamasaki, Y. Inagaki, K. Idemitsu J. Alloys Compd. 400, 43-50 (2005). [4] T. Arima, S. Yamasaki, K. Idemitsu, Y. Inagaki J. Nucl. Mater. 376, 139-145 (2008). [5] P. Erhart and K. Albe Phys. Rev. B 71 035211 (2005). [6] F. Muller-Plathe J. Chem. Phys. 106 6082-6085 (1997). [7] S. Plimpton J. Comp. Phys. 117, 1-19 (1995). [8] J. VandeVondele, M. Krack, F. Mohamed, M. Parrinello, T. Chassaing and J. Hutter, Comp. Phys. Comm. 167, 103 (2005). Industry partnership: Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 160000 Q1: 40000 Q2: 40000 Q3: 40000 Q4: 40000 Justification: Storage requirements: Thank You, The LCRC Accounts System