Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Julius Jellinek Project Name: Comp_Nanocatalysis Division: CSE Project title: Theoretical/Computational Studies of the Fundamentals of One-Component and Alloy Metal Based Nanocatalysis Relevant to Alternative Sources of Energy Associated funding: EFRC/IACT funded by BES/DOE Other Systems: NERSC, ~600000 core-hours Science: The central objective of this project is to perform modeling and computational studies that aim at understanding and characterization of the chemical reactivity and catalytic functionality of pure and alloy metal nanoparticles at the microscopic (i.e., atomic and electronic) level. The systems and processes of interest are those defined in the project “Institute for Atom-Efficient Chemical Transformation”, a DOE Energy Frontier Research Center. The overall emphasis is on finding new ways of generating fuels from biomass. Specific examples of transformations of interest include reforming of glycerol and formic acid into H2. The goal of the planned computational studies is to aid the ongoing experimental investigations on synthesis of novel, primarily alloy, nanocatalysts and characterization of their various physico-chemical properties and catalytic characteristics. The aim is not only to explain the experimental findings but also to provide guidance to future ex periments. Project description: The planned work will continue and expand our studies performed last year, which addressed and answered a host of issues related to structural, electronic and chemical reactivity properties of pure Pt and Mo and mixed Pt/Mo clusters in the size range of n=2-7, 13, and 55 atoms. In the case of mixed clusters, these were obtained and characterized for different Pt/Mo composition for each cluster size. Among the major achievements is the clarification of the reasons for and mechanisms of the experimental observation that admixing Mo to Pt substantially improves the catalytic performance of the latter. Another important outcome of our computational studies is the explanation why the synthesis process used in the experimental part of the project leads to mixed Pt/Mo nanoparticles with Mo in the surface even though energetically it would be preferable for Mo to be in the interior of the particles. Our planned computations will expand the current studies in three aspects. First, we will include nanocatalysts of larger sizes (on the order of hundred atoms and beyond). As a consequence, the space of different structural forms, both geometric and homotopic (different placement of atoms of different elements between the sites of a given geometric conformation) will increase immensely. The same is true for the scope of the electronic structure problem at hand since the number of electrons that will have to be treated explicitly will also increase substantially. Second, we will extend the studies to nanocatalysts of other elements. The candidates include Pd/Pt, Pd/Cu, Pd/Zr, and others. Third, we will incorporate into the studies the supports of the nanocatalysts. The supports of relevance are oxides such as SiO2, TiO2, Al2O3, etc. All three of these aspects substantially increase the scope of the computational task at hand and the needed core-time allocation. The expected n umber of project members is 4. We will continue to use the parallelized version of the NWChem code, which proved to be very efficient in our studies performed last year. In addition we will also use the Crystal package (we have purchased the license for its latest version and recently with the help of John Low installed it on Fusion). We have also purchased the license for the VASP package, which is available on Fusion, and we intend to use it as well. Project URL: http://www.anl.gov/catalysis-science/index.html Current FY Hours Used: undetermined amount New FY Requested allocation: 1200000 Q1: 300000 Q2: 300000 Q3: 300000 Q4: 300000 Justification: The scaling efficiency of the NWChem, Crystal, and VASP codes with an increase in the number of cores is well documented for the types of computations that are being and will be performed within the scope of this project. Examples include: 1) Scalability of NWChem on Fusion as evaluated by us for the Pt12Mo-CO system (total of 554 basis functions): Cores Average SCF cycle (s) 64 9 128 6 192 5 2) Scalability of MPP_Crystal on Hector (UK) as evaluated for TiO2(3x3x3) supercell (648 atoms, 13608 basis functions) [1]: Cores Average SCF cycle (s) 256 739 512 398 1024 213 2048 135 4096 85 [1]I. J. Bush et al., Proc. Royal. Soc. 467, 2112 (2011). Thank You, The LCRC Accounts System