Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: John J. Low Project Name: Linr_sclng_catalysis Division: MCS Project title: Linear Scaling Electronic Structure for Catalysis in nanobowls. Associated funding: of Energy, Office of Science, and Office of Basic Energy Sciences Other Systems: None Science: Metal clusters supported in metal oxide nanobowls are being synthesized by the Institute for Atom-efficient Chemical Transformations (IACT). These nanobowls are synthesized by depositing several layers of an oxide (i.e. Al2O3) with atomic layer deposition [i.e. Al(CH3)3] on another a substrate oxide (i.e. SrTiO3). Complexes (i.e. Fe-calixarene) on the substrate oxide form nanobowls in the deposited oxide. These nanobowls have diameters of about 1-2 nm. Modeling the chemistry in 1 nm nanobowls will require cells with a volume of 10 cubic nanometers containing thousands of atoms. The computational effort for VASP, GPAW and other plane wave code scales like the number of electrons to the third power. Although simulations of this magnitude are possible with VASP or GPAW on high performance supercomputers, programs like CP2K implement linear scaling methods of electronic structure require an order of magnitude less computer time than conventional electronic stru cture codes for models containing a thousand atoms. CP2K will allow the efficient modeling of the catalytic chemistry in nanobowls at the atomic level. Project description: Cp2k has been built, tested and put into production on fusion. Currently a metadynamics study of surface segregation in bimetallic clusters is underway. This study will show why the surface of PtMo clusters stay enriched in Mo after oxidation and reduction, in spite of the fact that energy of a reduced cluster with a surface enriched in Pt is lower energy. CP2K is very efficient at modelling metal clusters. For metal cluster containing 50 atoms it is ~100 times faster than NWCHEM for DFT calculations while generating simuliar results. This software could be used to accelerate the simulations of metal clusters in other fusion projects. A seminar is planned to encourage the use of CP2k by more users at Argonne. We are beginning the planned simulations of metal clusters supported in nanobowls to demonstrate the linear scaling of CP2K. The linear scaling program CP2K can handle models containing 1000s of atoms efficiently. CP2K has been shown to scale well to 256 processors for a model containing 1028 water molecules. The elapsed time required for a calculation containing a thousand atoms is approximately 28 hours on 256 processors (7,000 hours) to optimize the geometry for a metal cluster in a nanobowl. We plan to run geometry optimizations for models of nanobowls, metal clusters in nanobowls and chemisorption of H, O and CHx species on metal clusters in nanobowls. Performing these calculations on three different combinations of metal oxides will demonstrate whether metal-support interactions alters the chemistry of a metal cluster in a nanobowl and the utility of this approach to modeling chemistry in nanobowls. The twenty-four calculations described here will require 200,000 hours of CPU. Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 200000 Q1: 50000 Q2: 50000 Q3: 50000 Q4: 50000 Justification: Thank You, The LCRC Accounts System