September 2011 - Allocations-admins - mailman3.cels.anl.gov

[LCRC Accounts] Project Allocation Request
by accounts＠lcrc.anl.gov 15 Sep '11

15 Sep '11

Hello, A change in allocation has been requested: Requester: ssom (Sibendu Som) Project: cfd_enginemodeling Title: Computational Fluid Dynamics Modeling of Petrodiesel and Biodiesel for Compression Ignition engine applications Description: 1) Finite volume schemes 2) LES and RANS turbulence models used 3) Detailed chemical kinetic models for diesel and biodiesel fuels 4) CONVERGE software is being used - 48 licenses available 5) Scales well up-to 48 processors 6) Detailed spray models used for different spray processes 7) Full-cycle engine simulations Current: undetermined amount Justification: Requested: 495000 A specific reason has been given: Last year we used up 387,500 hours and there is still negative balance on the account. The usage for this year is expected to be even higher. This needs to be approved and the final allocation amount decided upon. Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Project Request: Comp_Soft_Catalysts
by accounts＠lcrc.anl.gov 15 Sep '11

15 Sep '11

Hello, A new project on the LCRC cluster has been requested. Please forward the information on to the LCRC Allocation sub-committee. Applicant's name: Juan Lopez-Encarnacion Applicant's institution: ANL Applicant's division: CSE Project Name: Comp_Soft_Catalysts Project title: Soft catalysts for green chemistry and energy applications Associated funding: Strategic Initiative on Materials for Energy (LDRD-2012-060-R1) Other Systems: None Science: The objective of this project is to perform extensive modeling and computational studies that will aid in design and synthesis of new catalytic materials for efficient transformation of abundant C1 sources (carbon dioxide (CO2) and methane (CH4), in particular) into chemical feedstocks and fuels. The planned theoretical/computational work will be performed in close contact with the ongoing experimental effort on the subject at CSE. The goal is to design and synthesize a new type of catalytic materials that will combine the best characteristics of both homogeneous catalysts (high selectivity at low temperature, tunability) and heterogeneous catalysts (stability, ease of separation). Such catalytic systems are not available at present. Project description: The targeted class of new catalytic materials is envisioned as one that integrates highly active and selective catalytic centers in the form of organometallic active sites with porous organic polymer (POP) frameworks as their supports. The computational studies will focus on atomic (compositional, structural) and electronic features of these systems, their geometrical characteristics (surface area, pore sizes/volume), mechanical aspects (density, bulk modulus, stress, etc.), thermal stability, and dynamics/kinetics of the catalytic reactions that will be explored experimentally. First, we will study the properties and characteristics of the building blocks - organometallic reaction centers, polymer struts, POP networks, elementary reaction steps, and then will address the full complexity of the problem at hand, i.e. reaction centers incorporated into POP networks and their catalytic functionality for reactants and products of interest. Two types o f POP networks will be considered – crystalline and amorphous. The structural and electronic aspects will be explored and characterized using state-of-the-art density functional theory. The number of atoms treated at a time will vary from a few to a few tens (e.g., in organometallic reaction centers), from tens to about a hundred (e.g., in struts), and to a few hundreds (e.g., in POP networks). The bulk of the planned computations will be performed with the SIESTA code, which is an efficient order-N algorithm (i.e., the computational time scales linearly with the size of the system); with the help of John Low, we recently installed on Fusion the latest version of SIESTA (version 3.1) and it is currently being tested. In the case of amorphous POP networks, the studies will require inclusion of thousands of atoms, and they will be performed using efficient and at the same time adequate semiempirical potentials and/or force fields to characterize their low-energy structure s. Then single-point DFT treatments will be performed on these structures to characterize their electronic features. The dynamical and thermal behavior will be explored using both first principles based and semiempirical potentials/force fields. Development and implementation of new such potentials/force fields will be part of the task. The kinetic aspects will be evaluated and characterized using transition state theory. In addition to the SIESTA code we will also use other packages (e.g., CP2K and NWChem) as appropriate. The requested time is based on estimates performed on Hopper at NERSC (see below). The systems we will study are considerably larger and more complex (contain heavier atoms such as N, O, Si, and metals such as Pd, Ta, and Pt) than those used in the test. The expected number of project members is three (3) or more. Project URL: http://www.cse.anl.gov/catalysis_energy_conversion/homogeneous/index.html Requested allocation: 1000000 Q1: 250000 Q2: 250000 Q3: 250000 Q4: 250000 Justification: A test on a fixed geometry so-called PAF-1 structure, a building block of a porous organic polymer, which contains 128 atoms of H (5 numerical basis function per each H) and 200 atoms of C (13 numerical basis functions per each C) was performed with the SIESTA code using 60, 120, 192, and 240 cores. The performance results are summarized in the table below. Substantial speed-up is achieved with the increase in the number of cores. Further improvements in the parallelization efficiency can and will be achieved through a variety of techniques, for example, by making an optimal distribution of fixed-grid operations between the cores. Cores Wall Time (s) 60 628.893 120 484.702 192 301.916 240 213.681 The requester has used undetermined amount hours of their initial startup project. In addition to approving an initial amount, please specify a Category and Subcategory for this project. For a list of the current selection of approved categories, please see: https://wiki.lcrc.anl.gov/wiki/Processes/Categories Once the Allocation committee has approved the project, please go to the Project Management page to create it: https://accounts.lcrc.anl.gov/projects.php Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Yearly Allocation Request from genomic_distribution
by accounts＠lcrc.anl.gov 15 Sep '11

15 Sep '11

Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Will Trimble Project Name: genomic_distribution Division: IGSB Project title: Distribution of kmers in metagenomic data Associated funding: Other Systems: Science: Next-generation sequencing technologies are currently producing data from nucleotide sequences extracted from the environment at an unprecedented rate. It is possible that probabilistic models of the character of DNA sequences can be used to substantially reduce the computational burden of sequence similarity searching and de novo assembly. This project aims to do exploratory work in characterizing the probabilistic properties of real and artificial metagenomic data and their usefulness in handling the same. Project description: Initial exploration of sequence-redundancy has shown promise for rapidly predicting the genome size in high-throughput genome sequencing projects and for determining the likely assembleability of metagenomic datasets. Error-correction in shotgun data sequencing data, a step that could improve the quality of assemblies, requires a memory- and compute- intensive traversal of the data. The desired all-against-all comparisons for error correction, assembly, and probing for redundancy at present require large memory (32Gb+) machines for input datasets in excess of 1 Gbase. Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 32000 Q1: 8000 Q2: 8000 Q3: 8000 Q4: 8000 Justification: Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Project Allocation Request
by accounts＠lcrc.anl.gov 15 Sep '11

15 Sep '11

Hello, A change in allocation has been requested: Requester: hereld (Mark Hereld) Project: Neocortex_sim Title: Neocortical Seizure Simulation Description: Development of new scalable methods. Current: undetermined amount Justification: Requested: 5000 A specific reason has been given: Improved scaling that has enabled 1M+ neuron simulations on another large platform. New scale makes simulating realistic tissue slices feasible. Need to develop methods to compare simulation to optical areal measurements of samples in lab. This needs to be approved and the final allocation amount decided upon. Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Yearly Allocation Request from md_pls_dft_for_TiO2
by accounts＠lcrc.anl.gov 15 Sep '11

15 Sep '11

Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Handan Yildirim Project Name: md_pls_dft_for_TiO2 Division: CNM Project title: First Principles and Classical Molecular Dynamics Calculations of Lithium Intercalation into TiO2 Surfaces and Nanostructures Associated funding: BES Other Systems: Carbon 150 000 Science: First principles calculations based on Density Functional Theory (DFT) and classical Molecular Dynamic (MD) simulations (using polarizable potentials) will be performed to study diffusion and intercalation of Li in TiO2 crystalline and amorphous nanoparticles and nanotubes of several sizes. These simulations are directly relevant to the exploration of the properties of titania as an alternative electrode for energy storage for Li ion batteries. Our calculations will be the extension of our earlier study at which Li diffusion and intercalation characteristics are derived for bulk TiO2 polymorphs. The effect of Li concentration on the diffusion energetics and dynamics in the nanostructures will be studied for better understanding of their electrochemical performance. We will compare our results obtained those for the bulk anatase, rutile and amorphous TiO2 to extract the role of nanostructuring and morphology on the diffusivity of Li ions. Additionally, the structural transitions upon certain concentration of Li intercalation into these structures will be evaluated, and compared to that observed in the amorphous TiO2. We will also extend our study to explore pressure-induced structural transitions and role of nanostructuring on the transition pressure. The present study requires detailed analysis of the energetics, dynamics of Li in TiO2 under various concentration and temperature conditions. For these calculations, we will vastly benefit from the ability of classical MD simulations that allow simulation of several thousands of atoms. We will explore the size effects on Li intercalation properties. We will use MD simulations (for temperatures ranging form 300K to 900K) for a long enough time (~ 5ns at the current studies) to explore the details of the Li diffusion mechanisms in both crystalline and amorphous TiO2 nanoparticles and nanotubes for concentrations ranging from 25% to 100%. The observed mechanisms of diffusion will be supplie d as an input to calculate the corresponding energetics and dynamics using periodic Density Functional Theory calculations. Project description: In our preliminary simulations, we have obtained insights regarding the diffusivity of Li in the TiO2 polymorphs including amorphous. The effects of concentration, mechanism for diffusion, and host morphology on the diffusivity of Li ions are explored by comparing the results obtained for anatase, rutile and amorphous TiO2. The structural changes upon certain concentration of Li intercalation into these systems are studied extensively. The structural properties and the diffusivity of Li ions in the new crystal phase are explored in details. The results of MD simulations at relatively low temperatures are analyzed to identify Li diffusion mechanisms in these systems, and the obtained mechanism is used to calculate the corresponding isolated Li diffusion energetics using DFT. Our MD results, in summary, showed that Li ions diffuse with different mechanism at each system that is mostly determined by the host lattice structure. For instance, in anatase, the diffusion is 3D in nature, while for rutile, the diffusion proceeds in 1D through the open channels provided by the intrinsic characteristic of the host lattice. Our DFT calculations also confirmed this observation. For the systems under study, for sufficient convergence of the total energy of about 100-150 atoms system with the chosen parameters required 100 plus ionic iterations. For our simulations, we have mostly used 32 CPU’s per job, and at least 5 jobs, on average, run simultaneously; we expect bit more higher distribution of jobs for the extension of our study, as we need to explore several sizes for tubes and nanoparticles. Note that we will also explore Li adsorption and sub-surface diffusion on these nanoparticles and nanotubes thus it is inevitable that we require significant computational resources to complete this part of the proposed study. We performed classical MD simulations to study Li diffusion energetics and dynamics under several conditions using the DL-POLY code. We have gained significant insights into the role of concentration and morphology on the diffusivity of Li ions in the bulk TiO2 polymorphs. Additionally, phase transition with high Li load and under certain pressure range is observed with these studied for nanosecond time scales. These simulations are indeed computationally intensive as the systems have large number of atoms. Our calculations for this part of the study are performed with a very fine time step (~0.2 fs in general), requiring long simulation hours even for 5ns (due to the type of the potential used). For the calculations of nearly 5000 to 10000 atoms for simulating 5ns, 25 to 50 millions simulation steps are needed. As our simulations extend to several temperature and concentration ranges, the need for intensive computational time can be understood. Our proposed simulations will require more computational resources as we plan to study size effects, and the effect of thickness, outer radius and length of the nanotubes on the diffusivity of Li ions and the structural changes upon lithiation. We will explore phase transitions for both nanoparticles and nanotubes for wide ranges of pressure values. The effect of the size of the nanoparticles will be evaluated to determine the role of nanostructuring on the transition pressure. The sizes will range between 2nm to 20nm and the simulations will be performed at least 1ns timescale for each size. These simulations will clearly require significant computational resources. We anticipate that we will be running of about 10 to 20 jobs per day to complete this part of the proposed research. Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 500000 Q1: 175000 Q2: 125000 Q3: 100000 Q4: 100000 Justification: In roughly three months, we required over 250,000 CPU hours for the combination of MD and DFT calculations including testing of classical potential used, and the convergence tests for DFT studies. These calculations have been performed on the CNM’s carbon cluster, but this resource is fully subscribed and is being increasingly used for external user projects. Thus, we require computational resources on Fusion to complete our proposed project. For a full year, we estimate that approximately 500,000 CPU hours will be sufficient. Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Yearly Allocation Request from Linr_sclng_catalysis
by accounts＠lcrc.anl.gov 14 Sep '11

14 Sep '11

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

1 0

[LCRC Accounts] Project Request: TMix
by accounts＠lcrc.anl.gov 14 Sep '11

14 Sep '11

Hello, A new project on the LCRC cluster has been requested. Please forward the information on to the LCRC Allocation sub-committee. Applicant's name: Rui Hu Applicant's institution: ANL Applicant's division: NE Project Name: TMix Project title: Translating Petascale Simulation into Engineering Design Tools: Application to Mixing Associated funding: Argonne-LDRD Other Systems: Science: Mixing is a phenomenon that is central to a broad range of DOE science and engineering problems, such as convective heat transfer in energy systems, chemical deposition and materials processing, dispersion of contaminants in the ocean and in urban canyons. The extreme nonlinearity of the systems coupled with minimum dissipation leads to a broad range of scales that make solution of the governing equations of flow mixing intractable by analytic means and extremely demanding from a computational standpoint. Although the Department of Energy's leadership computing facilities provide simulation capabilities that allow scientists to explore a significant range of scales with unprecedented details, it is too expensive to be used for parameter-space exploration or design analysis. This project will develop a hierarchical methodology to model turbulent mixing in any given wall-bounded flow system. Using DOE’s leadership computing facilities, high fidelity simulations based on direct numerical simulation or large eddy simulation will be performed to probe the micro-details of the turbulent mixing phenomenon. The fine-scale simulation results will be used to develop a model to accurately estimate mixing through lower fidelity Reynolds-Averaged Navier-Stokes calculations on a small cluster such as Fusion. Such an approach would allow exploring a wider range of parametric design options for a given flow system. Through the wide exploration of the parameter space it will be possible to construct a general model for mixing to be used in reduced order methods for engineering design analysis. This approach provides a critical bridge among the micro-, meso-, and macro- scale modeling and simulation methods. Thus, the high-fidelity simulations on DOE exsa- or p eta- scale computing facilities can be integrated into quotidian design process. Project description: High fidelity calculations using direct numerical and large eddy simulations (DNS/LES) based on the ANL code Nek5000 will be performed in prototype geometries and a full rod-bundle in a reactor core and other cases. The probability density functions for each velocity component will be computed locally. From this data, one can evaluate the mixing parameter through Reynolds stresses in RANS based simulations with the commercial CFD code, STAR-CCM+. While the high fidelity calculation of the full rod-bundle simulation has to be executed in Blue Gene/P, the calculation of prototypical geometry can be run in Fusion. Once the pipeline from DNS/LES to RANS is established and validated, RANS will be used to examine more complex problems in their relevant parameter space: the geometric parameters, the Reynolds number, the Peclet Number, etc., which require intense use of computation time in a small clusters. Access to the Fusion cluster will allow us to explore a wide range of parameter space and to develop models/correlations to be used in engineering design tools. Nek5000 is a mature research DNS/LES computational fluid dynamics code using spectral element method. STAR-CCM+ is a commercial CFD tool using the finite volume formulation of the RANS method for thermal fluid dynamics simulation. Both Nek5000 and STAR-CCM+ has already been installed and used in Fusion. Nek5000 employs the MPI standard for parallelism. In 1999 the code was recognized with the Gordon Bell prize for algorithmic quality and sustained parallel performance. It has been scaled on Blue Gene/P to over 100,000 cores. The CFD codes STAR-CCM+ are routinely applied at Argonne to simulations that use 8-200 cores, and good parallel performance has been observed. As problem size increases, i.e. for finer meshes and for large pin bundles, larger numbers of cores would be utilized and based on data from the developers good parallel performance will continue to be expected. All jobs run on Fusion will be run in parallel. A typical case in parameter exploration may cost 8 cores/node * 25 nodes * 20 hours of run time = 4000 core-hours. Project URL: Requested allocation: 200000 Q1: 50000 Q2: 50000 Q3: 50000 Q4: 50000 Justification: The requester has used undetermined amount hours of their initial startup project. In addition to approving an initial amount, please specify a Category and Subcategory for this project. For a list of the current selection of approved categories, please see: https://wiki.lcrc.anl.gov/wiki/Processes/Categories Once the Allocation committee has approved the project, please go to the Project Management page to create it: https://accounts.lcrc.anl.gov/projects.php Thank You, The LCRC Accounts System

1 0

Re: [allocations-admins] [LCRC Accounts] Yearly Allocation Request from aimd_catalysis
by John Valdes 14 Sep '11

14 Sep '11

Hi John, FYI, I corrected the values in the database (*100) for this renewal. John On Wed, Sep 14, 2011 at 03:54:35PM -0500, accounts(a)lcrc.anl.gov wrote: > > A yearly allocation for the LCRC cluster has been requested with the > following updated information: > > Submitter/PI: John J. Low > Project Name: aimd_catalysis > [...] > New FY Requested allocation: 200 > Q1: 50 > Q2: 50 > Q3: 50 > Q4: 50

1 0

[LCRC Accounts] Project Request: MBM
by accounts＠lcrc.anl.gov 14 Sep '11

14 Sep '11

Hello, A new project on the LCRC cluster has been requested. Please forward the information on to the LCRC Allocation sub-committee. Applicant's name: hakim iddir Applicant's institution: ANL Applicant's division: MSD, CNM Project Name: MBM Project title: Modeling Battery Materials (MBM) Associated funding: EFRC, EERE Other Systems: EMSL-PNNL, CNM-ANL Science: Rechargeable Li-ion batteries (LIB) have become one of the most important energy storage devices for portable electronics, electric and plug-in hybrid vehicles, owing to their high energy density and design flexibility 1. Graphitic based materials are usually used for the anode in commercial LIB because of their low cost, relatively high capacity, non-toxicity, and improved thermal stability and safety over lithium metal anodes2-5, as a result of a passivation film (solid-electrolyte interface (SEI)) that forms on graphite6. Previous studies suggested that defects in disordered/amorphous graphene sheets might be at the origin of their greatly enhanced storage capacity5,7,8. A fundamental understanding of both SEI formation, structure and properties, as well as Li and Li+ diffusion across different defects in graphene planes is a prerequisite to better understand the overall capacity and transport behavior of several graphitic anode materials and conf igurations in LIB, such as carbon nanospheres 9, carbon nanotubes (CNT)10, highly oriented pyrolitic graphite and graphene paper11. This work will be closely coupled with extensive experimental efforts conducted by two groups from the Chemical Sciences and Engineering (CSE) division, as well as research groups from Northwestern University (NU) and the University of Illinois at Urbana-Champaign (UIUC). This project is being done as part of the ANL EFRC on energy storage and an EERE funded program. (1) Tarascon, J. M.; Armand, M. Nature 2001, 414, 359. (2) Noel, M.; Santhanam, R. Journal of Power Sources 1998, 72, 53. (3) Fauteux, D.; Koksbang, R. Journal of Applied Electrochemistry 1993, 23, 1. (4) Yazami, R.; Touzain, P. Journal of Power Sources 1983, 9, 365. (5) Yoo, E.; Kim, J.; Hosono, E.; Zhou, H.-s.; Kudo, T.; Honma, I. Nano Letters 2008, 8, 2277. (6) Aurbach, D. Journal of Power Sources 2003, 119-121, 497. (7) Pan, D.; Wang, S.; Zhao, B.; Wu, M.; Zhang, H.; Wang, Y.; Jiao, Z. Chemistry of Materials 2009, 21, 3136. (8) Sato, K.; Noguchi, M.; Demachi, A.; Oki, N.; Endo, M. Science 1994, 264, 556. (9) Pol, S. V.; Pol, V. G.; Sherman, D.; Gedanken, A. Green Chemistry 2009, 11, 448. (10) Frackowiak, E.; Béguin, F. Carbon 2002, 40, 1775. (11) Abouimrane, A.; Compton, O. C.; Amine, K.; Nguyen, S. T. The Journal of Physical Chemistry C, 114, 12800. Project description: We propose to investigate the above mentioned properties (SEI formation, structure, Li transport, and reactions at the interface) on model systems using density functional theory (DFT) as implemented in the Vienna Ab Initio Simulation Package (VASP) currently available on FUSION. We have successfully investigated the growth of Li2CO3 films on graphite surfaces, as well as the diffusion of Li in Li2CO3 bulk material (TIES project). We have started to investigate the interaction of solvent molecules (EC) with both edge and basal planes of graphite, as well as Cu and Au model surfaces. During this next period we would like to continue the study of the interaction of solvent molecules with different surfaces, a work that will contribute to better understand the SEI formation and properties. This specific part of the project will be conducted in close collaboration with the group from UIUC using the Sum Generation Frequency (SFG) technique. We will partic ularly investigate the difference in interaction and diffusion properties between Li/Li+ and different graphite and graphene surfaces with several defect configurations (single, double and triple-vacancies and Stone-Wales type defect) and different types of CNT’s (semiconductor and metallic). Similar studies will be performed on Al2O3/graphite interfaces. In this case, separate MD studies will provide us with the starting structures of the interfaces that will be further optimized using VASP. This part of the project, originally proposed in TIES (2010) was not accomplished and will be proposed within this program. A total amount of 800000 hours would be necessary to complete the proposed work. Project URL: Requested allocation: 800000 Q1: 200000 Q2: 200000 Q3: 200000 Q4: 200000 Justification: The study of solvent molecules interaction with surfaces with frequency calculations will require the investigation of about 20 different configurations per solvent molecule and surface models. For each couple solvent/surface system about 50000 hours would be needed, we anticipate the investigation of at least four types of solvent molecules (EC, VC, VEC, and PC) and four different surfaces (Cu(111), Au(111) and two graphite surfaces) hence about 200000 core hours would be necessary to complete this part of the project. Li and Li+ diffusion calculations through one defect type of graphene will require about 80000 core hours. The diffusion calculations for different defects and surfaces (graphene, graphite (top and inner layers, CNT’s and Al2O3) would require about 600000 hours. A total of 800000 core hours would be needed to perform these calculations. The requester has used undetermined amount hours of their initial startup project. In addition to approving an initial amount, please specify a Category and Subcategory for this project. For a list of the current selection of approved categories, please see: https://wiki.lcrc.anl.gov/wiki/Processes/Categories Once the Allocation committee has approved the project, please go to the Project Management page to create it: https://accounts.lcrc.anl.gov/projects.php Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Yearly Allocation Request from xenon
by accounts＠lcrc.anl.gov 14 Sep '11

14 Sep '11

Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Zeke Insepov Project Name: xenon Division: MCS, NE Project title: Multi-scale simulation of radiation damage in nuclear fuels Associated funding: RERTR, AFCI Other Systems: Blue Gene Science: Summary of the project: The formation of defects in bcc Mo lattice as a result of 300 keV Xe ion bombardment will be studied via classical atomistic simulation molecular dynamics and kinetic Monte Carlo simulations, with an interatomic potential developed using the force-matching Ab initio based approach. Radiation defect formation via electronic and nuclear stopping power and evolution of the defects in the cascade will be described. Diffusion and interaction of Frenkel pairs, such as self-interstitials, vacancies, and clusters of defects will be analyzed. Stable forms of clusters of defects will be studied that show one-dimensional diffusion with a very high diffusion coefficient and that escape quickly to an open surface. Single vacancies have lower diffusivities and will not aggregate into large clusters. Large prismatic vacancy loop formation near the impact surface will be studied that may be considered as a result of fast recrystallization. The mobility of the vacancy dislocation loop segments will be found, the motion of the entire loops will be studied that strongly hinders the neighbor point defects. A new multiscale concept for irradiated materials simulation will further be developed that will be based on coupling molecular dynamics simulations (MD) where the potential was obtained from Ab initio data of energies of the basic defect structures, with kinetic mesoscale models. The evolution of a system containing self-interstitial atoms (SIAs) and vacancies in crystalline molybdenum will be investigated by means of MD. The kinetics of formation of a dimer clusters (built of two self-interstitial atoms) (di-SIA) and SIA–vacancy recombination will be analyzed via approaches used in the kinetic theory of radiation ageing. The effects of 1D diffusion of SIAs, temperature, and defect concentrations on the reaction rates will also be studied. This approach can validate both the kinetic mechanisms and the appropriate kinetic coefficients, offering the potential to significantly reduce the uncertainty of the kinetic methodology and providing a powerful predictive tool for simulating irradiation behavior of nuclear materials. Project description: The following calculations will be performed for interatomic potentials Mo-Mo, Mo-Xe, U-U in the form of EAM potential: 1. Choosing a structure fitting the potential, calculations of energies and forces on atoms in this configurations from DFT by VASP. 2. Creating the potential by force-matching software PotFit. 3. Verification of the potential obtained by comparison with the DFT data: formation energies of vacancies, self-interstitial atoms (SIA) and their clusters, energies of the basic structures of SIAs, their hierarchy. 4. Verification of the potential with the experimental data on diffusion of defects (diffusion constants, threshold temperatures to induce long-range migration) and production of defects in radiation cascades (structure of the defects and sizes). 5.Improvement of the reference structures to reduce the discrepancies revealed in 3 and 4. computational methods: Density Functional Molecular Dynamics simulations using the siesta method programming model: parallel experiments planned: Argonne IVEM and Atlas experiments, Electron Microscopy software requirements: C++, fortran compiler (e.g. openmpi) sizes of calculations run on and/or planned for LCRC or other systems: hundreds of nano-seconds time scale for classical MD simulations on 1e7-1e8 atomic systems expected number of project members: 7 We expect a total of 7 project members (two from University of Illinois, and several members from the New University of Astana and Russian Academy of Sciences. Results of assessments are expected to lead to joint proposals to NSF, DOE (Office of Science, and NE), and DARPA. Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 840000 Q1: 210000 Q2: 210000 Q3: 210000 Q4: 210000 Justification: Scalability and Computational Efficiency for the simulations using VASP and LAMMPS codes were investigated. Good scaling was seen up to 8 nodes for the problems studied by VASP and up to 64 nodes for LAMMPS simulations. Thank You, The LCRC Accounts System

1 0