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

[LCRC Accounts] Yearly Allocation Request from Cosmology
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: Joe Bernstein Project Name: Cosmology Division: HEP Project title: Harnessing Computational Cosmology Simulations to Investigate Cosmic Evolution Associated funding: None external to ANL sources Other Systems: ANL Blue Gene/P (renewal recently granted for 800K CPU-hrs over 3 months) Science: The evidence for dark energy represents one of the greatest mysteries of modern science. The proposed research will support probing of the implications of dark energy via analysis of computational cosmology simulations. Several independent observational methods provide evidence pointing to the conclusion that 95% of the current energy density of the Universe exists as part of the "dark sector" that is different from normal, i.e.,"baryonic," matter. In particular, the dark sector is thought to be comprised of 22% "dark matter," which interacts with baryonic matter ostensibly only through gravity, and 73% "dark energy," which experiences gravity as a repulsive force. The nature of both dark sector components are essentially complete unknowns. It is presently an exciting time to be involved in cosmology because planned astronomical surveys will effectively result in dark sector probes becoming systematics-limited, making numerical simulations crucial to placing precisi on constraints on the dark sector. In addition, particle physics theorists predict that the Large Hadron Collider (LHC) and direct dark matter experiments may soon discover a dark matter candidate particle. ANL HEP staff and collaborating U. Illinois faculty are members of the Dark Energy Survey, which will provide the next generation of dark energy and (indirect) dark matter data. Local scientific simulation expertise and high-performance computing facilities further make ANL ideally-suited to pursue answers to questions about the dark sector. The focus of the proposed research is the production and interpretation of computational cosmology simulations in order to elucidate the nature of the dark sector. Project description: The "Cosmology" project on fusion.lcrc.anl.gov (hereafter, "fusion") is in support of the main research effort of the PI as an ANL Computational Postdoctoral Fellow. The number of project members is currently 5. The PI will make use of the requested number of CPU-hrs (10K) in efforts to evaluate code performance, generate initial conditions, and verify scaling results for cosmology simulations run on the ANL Blue Gene/P ("BG/P"). Toward this end, the PI will harness piecewise-parabolic method (PPM) hydrodynamics and N-body gravity solvers, MPI and OpenMP for inter-processor communication, and HDF5 to implement parallel I/O. The project utilizes both the C and FORTRAN90 programming languages. Collaborators on the BG/P cosmology studies include Paul Ricker (U. Illinois), a computational cosmology expert, and Boyana Norris (MCS), a computer scientist. The goals of the project include submitting a DOE INCITE proposal. The ultimate target of the BG/P simulation effort is runs requiring PPM+N-body simulations with a 1 Gpc (1 pc = 3.09x10^16 m) box size, 2048^3 grid cells, and 10^11 dark matter particles. Fusion will be employed to generate initial conditions for a suite of a few dozen BG/P simulations. Generating such initial conditions is memory intensive and cannot currently be done on the BG/P. These requirements also push the limits of existing codes. Verifying scaling results, on a traditional Linux cluster such as fusion, will constitute an important part of the BG/P studies. We are highly confident of fusion's suitability for the proposed simulations based on our FY2010-11 experience. In FY2011, we uncovered scaling problems with the default adaptive mesh refinement (AMR) element of the cosmology simulations that was seen on both BG/P and fusion. In addition, we proved that crashes in the context of non-adaptive, i.e., unigrid, runs were due to a particle breaching it's computational block on both machine. Comparison of results on fusion proved to be crucial to those efforts. Thus, we anticipate continued use of fusion in such a role. Based on the increased understanding of efficient use of fusion gained in the past 6 months, the PI proposes to complete additional development work requiring a reduced amount of resources as compared to the FY2011 proposal. Upon completion of the development work, the PI anticipates requesting additional time for scaling studies as needed. The primary development work proposed includes testing of an updated version of the default AMR module and the pre-production-level parallel implementation of the computational cosmology initial conditions generator provided by collaborator Ricker. The latter is necessary for the generation of initial conditions of sufficient size for the simulations studies discussed above. In support of these efforts, the PI requests 10K fusion CPU-hours for FY2012. Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 10000 Q1: 2500 Q2: 2500 Q3: 2500 Q4: 2500 Justification: Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Yearly Allocation Request from Cat_Biomass
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: Rajeev Surendran Assary Project Name: Cat_Biomass Division: MSD Project title: Computational Studies of Biomass Catalysis Associated funding: BES Other Systems: CNM Argonne, Blue Gene/P Argonne Science: This work will be done as part of the new Energy Frontier Research center at Argonne and will be done in close collaboration with the experimental part of the Center involving researchers from Argonne, Northwestern University, Purdue University, and the University of Wisconsin. Efficient chemical transformation of biomass is essential to produce sustainable energy and industrial chemicals. Industrial level chemical conversion of naturally abundant sugar molecules to alternative fuel precursor chemicals (hydroxyl-methyl-furfural, γ-valero-lactone, levulinic acid) require accurate molecular level understanding of reaction energetics and kinetics. High-level quantum chemical methods can provide accurate energetics for important chemical transformations of sugar molecules and assist experimental efforts for efficient production of alternative fuels and fuel precursors. Kinetics and thermodynamics of complex sugar decomposition pattern is another area of huge importance that requires fundamental understanding to limit undesired reactions. Due to development of fast computers and reliable, accurate computational methods now allows us to compute the energetics of many important reactions in the Biomass conversion. Micro-kinetic modeling of various reactio ns based from accurate ab-initio calculations is another useful area and need a great deal of computational studies. Industrial level chemical transformation of biomass to useful chemicals requires sustainable, efficient and economical catalysts and this require design of multifunctional catalysts. Mineral or Lewis acid catalyzed reaction is envisioned for the conversion of glucose to hydroxy-methyl-furfural, levulinic acid, γ-valero-lactone and this require detailed understanding of possible reactions of sugar molecules with a variety of acid sites and elucidation of reaction mechanisms Computational modeling of catalyst design, binding modes of sugar molecules and reaction sequences are essential to compliment experimental efforts towards an industrial level biomass conversion Project description: Biomass has the potential to serve as sustainable source of energy and organic carbon for our industrialized society. The key reactions involve the process of the converting the biomass to useful chemicals are hydrolysis, dehydration, isomerization, aldol condensation, reforming, hydrogenation, and oxidation. Investigating these important reactions and modeling efficient catalysts for industrial use, by using computational chemistry methods is the aim of this project. Acid-Base hydrolysis and pyrolysis techniques are currently practiced as the ways to convert biomass to useful platform chemicals. These decomposition reactions are typically complex; include structural rearrangements, dehydration, oxidation and reduction. In order to assess the accurate kinetic and thermodynamic feasibilities of possible reactions, computational modeling is alternative approach to experiments. Application of methods such as G4/G4MP2, a CCSD(T) based energy evaluation method has been employed in the recent past to gain accurate understanding of energetics along with the help of density functional methods to investigate structures. Due to the complex chemical pathways of reactions in the biomass conversion, micro-kinetic models would be more appropriate and needed to assist the experimental interpretations or predict the outcome of such chemical transformations. This would require a significant amount of computational resources. Development of possible micro-k inetic models for following reactions would enhance the fundamental understanding of biomass catalysis: • Acid hydrolysis and Pyrolysis of Glucose • Acid hydrolysis and Pyrolysis of Sucrose • Acid hydrolysis and Pyrolysis of Cellobiose/triose • Understanding the Oligomerization processes during hydrolysis and pyrolysis These studies would lead to building of a thermochemical database and can be used as future reference tool for the thermodynamics for biomass transformations. And computational modeling of these reactions would also enable us to understand the mechanistic details at the atomistic level. Project URL: http://www.anl.gov/catalysis-science/ Current FY Hours Used: undetermined amount New FY Requested allocation: 600000 Q1: 150000 Q2: 150000 Q3: 150000 Q4: 150000 Justification: We gratefully acknowledge the grants for the computer time during the past two years. We have acknowledged LCRC for the computational resources. The following are the list of publications. 1. R.S. Assary, P. Redfern, J. Greeley, L. A Curtiss (2011). Mechanistic insights into the decomposition of fructose to hydroxy-methyl-furfural in neutral and acid environments using high-level quantum chemical methods, 4341-4349. In Journal of Physical Chemistry B. 2. R.S Assary, L.A Curtiss (2011). Theoretical Study of 1,2-Hydride Shift Associated with the Isomerization of Glyceraldehyde to Dihydroxy Acetone by Lewis Acid Active Site Models. In The journal of Physical Chemistry A. 8754-61 3. R.S, Assary, L. A Curtiss (2011). Thermochemistry and reaction barriers for the formation of Levoglucosenone from Cellobiose. Accepted, ChemCatChem (cctc.201100280). 4. T Kim, R. S. Assary, C. Marshall, L.A. Curtiss, P. Stair (2011). Acid Catalyzed Polymerization of Furfuryl alcohol : Combined experimental and Theoretical Study. In ChemCatChem (Cover story), Volume 3, 1369 5. T Kim, R.S. Assary, L.A. Curtiss, C. Marshal, P. Stair (2011). Vibrational properties of levulinic acid and furan derivatives: Raman spectroscopy and theoretical calculations. In Journal of Raman Spectroscopy. (Published Feb 2011,DOI 10.1002/jrs.2951) 6. R.S. Assary, P. Redfern, J. Hammond, J. Greeley, L.A. Curtiss (2010) Computational Studies of the Thermochemistry for Conversion of Glucose to Levulinic Acid. In Journal of Physical Chemistry B. 9002-9009 7. R.S. Assary, P. Redfern, J. Hammond, J. Greeley, L.A. Curtiss (2010) Predicted Thermochemistry for Chemical Conversion of 5-hydroxymethylfurfural. In Chemical Physics Letters. 123-128. Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Yearly Allocation Request from LiBattModeling
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: Roy Benedek Project Name: LiBattModeling Division: CSE Project title: Lithium-Ion Battery Electrode Modeling Associated funding: DOE Office of Vehicle Technologies, EERE (Batteries for Advanced Transportation Technologies [BATT] Program) Other Systems: NERSC Science: First principles local-density-functional-theory methods are applied to calculate physical and electrochemical properties of lithium-ion-battery-electrode materials. This work complements an experimental program in the Battery Department of CSE led by Michael Thackeray. We intend to address the following systems and properties in FY12: 1. Properties of dopants near lithium manganate spinel (LiMn2O4) surfaces 2. Li-ion transport properties through protective coatings (such as alumina and zirconia) of LiMn2O4 surfaces 3. Phase stability and structure of over-lithiated layered compounds Li1+xMO2 Project description: 1. In previous work, we have found that the interaction of divalent dopants with LiMn2O4 surfaces is repulsive, whereas trivalent and tetravalent dopants are less repulsive, or slightly attractive. The previous calculations were limited to stoichiometric LiM2O4 slabs. We intend to investigate the effect of Li-depletion (which occurs during battery charging) on the stability of dopants and on Mn oxidation states near Li(1-x)M2O4 surfaces. 2. In previous work, the energy of a migrating lithium ion through an amorphous AlF3 coating on LiM2O4 has been mapped. We intend to perform similar calculations for other candidate protective coatings on lithium manganate spinel, such as Al2O3 and ZrO2. 3. Overlithiated layered compounds Li(1+x)M(1-x)O2, attractive for their application as high capacity lithium ion battery cathodes, have complex structures that are poorly understood theoretically. Calculations will be performed to elucidate the tendency of the materials to show either solid-solution or composite structures. Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 120000 Q1: 30000 Q2: 30000 Q3: 30000 Q4: 30000 Justification: Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Yearly Allocation Request from Dissolution
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: Roy Benedek Project Name: Dissolution Division: CSE Project title: Dynamical simulation of lithium manganate spinel dissolution in acid Associated funding: Office of FreedomCar and Vehicle Technologies (Batteries for Advanced Transportation Technologies (BATT) Program), U. S. Dept. of Energy Other Systems: NERSC Science: The acid-promoted dissolution of lithium manganate, LiMn2O4, is an obstacle to its application as a cathode material in lithium-ion batteries. Although dissolution-inhibition strategies are available (such as coating the surface with a thin protective layer), a better fundamental understanding of the dissolution process would help focus battery-design efforts. In work completed in FY2011 (submitted to J. Phys. Chem. C), Blue Moon ensemble simulations were performed of the dissolution of Li, Mn and O ions in neutral water. Benchmark values were obtained of the energy required to break the chemical bonds of the dissolving ions with the substrate. In the next stage, the effect of acid (HF) will be incorporated into the simulations, to determine the extent to which chemical bonding energies are lowered by adsorbed F on cations and H on oxygen. Project description: Experiments Planned Blue Moon ensemble simulations, using the VASP code, will be performed to investigate the effect of acid (HF) on the dissolution of LiMn2O4 spinel. Two steps are planned: 1. Determine adsorption energies for acidic ions to LiMnO2 surfaces (of F to cations Mn and Li, and H to oxygen), and 2. Determine the (activation) energy required to break the chemical bonds of surface cations or anions to the LiMn2O4 substrate, in the presence of adsorbed F (cations) or H (oxygen). In the case of Mn, sites with three-fold and four-fold coordinate will be considered. Once these steps are completed, the resultant energies will be incorporated in phenomenological rate theory models, to predict the rate of dissolution of lithium manganate spinel in acid. Estimates will also be made of the effect of temperature on dissolution rates. Computational Methods and Software Calculations will be performed with VASP, version 5.2.12. Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 160000 Q1: 40000 Q2: 40000 Q3: 40000 Q4: 40000 Justification: Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Yearly Allocation Request from Metastable_Materials
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: Weronika Walkosz Project Name: Metastable_Materials Division: MSD Project title: First-principles studies of growth mechanisms for metastable materials Associated funding: ANL Strategic LDRD, DOE Other Systems: NERSC, pnnl-Chinook Science: The main objective of the project is to develop first-principles models describing the growth of Group-III nitrides in order to establish a firm basis for far-from-equilibrium nitride synthesis in a wide range of compositions. In particular, the growth of InxGa1-xN alloys will be extensively studied in this work because of their widespread applications in solid-state lighting, optoelectronics, and photovoltaics. Although metal organic chemical vapor deposition (MOCVD) is the most efficient and clean deposition process for production of nitride solid-state devices, the growth process for InxGa1-xN via this technique has been observed to display a complex behavior characterized by self-sustaining spatiotemporal chemical oscillations between nanocrystalline islands of InN and elemental In droplets. While these results suggest that the nitrogen activity produced by surface-catalyzed ammonia (NH3) decomposition may be responsible for the observed unusual behavior, an at omistic model of the gas phase and surface chemistries, and catalytic reactions that govern such behavior needs to be developed. We propose calculations aimed at development of a kinetic model, which will help optimize synthesis conditions to grow InxGa1-xN materials of much higher quality for solid-state lighting. The parameters of the model describing oscillatory behavior of InN islands/In droplets during the growth will be determined from the results. Identification of rate-limiting reaction steps will lead to optimization of synthesis conditions and precursors based on this model. The results of the modeling will be validated by comparison to the experimental data obtained in ANL. This model is expected to be applicable to describe kinetically controlled processes in synthesis of other materials. Project description: We will continue our systematic study of the non-equilibrium growth of Group-III nitrides with NH3, In(CH3)3, and Ga(CH3)3 as N, In, and Ga precursors, respectively. So far, we have investigated the interaction of NH3 and its moieties with the (0001) surfaces of InN and GaN, focusing on their adsorption energies and decomposition barriers. The next step in our modeling will be the incorporation of In and Ga from In(CH3)3 and Ga(CH3)3 onto the nitride surfaces. First, gas phase decomposition rates will be calculated for these compounds to establish the precursor composition interacting with the surfaces at relevant conditions. The species will be then placed on the surfaces along with appropriate NHx (x=0-3) precursors and the reaction energies will be calculated. Decomposition barriers for NH3 will be also computed on the nitride surfaces in the presence of the In and Ga species and compared to our earlier calculations on bare surfaces. Additional ly, we are planning to calculate kinetic barriers for NH3 decomposition on strained surfaces since the nitrides are currently grown on substrates with a large lattice mismatch by our experimental collaborators. Surface step edges and kinks will be considered in our modeling as well. Lastly, we will investigate atomic and electronic structure of the (0001) surface of ZrB2, which is a promising substrate for the growth of GaN due to a good lattice match and similar thermal expansion coefficients. Binding sites and energies, and diffusion barriers for adsorbed Ga and NHx species on two different terminations of ZrB2 (0001) surface will be calculated. Also, vibrational frequency calculations will be performed for the lowest energy ZrB2 surfaces in the presence of NHx (x=0-3) and will be compared to the ongoing experimental measurements. The predicted precursor decomposition barriers will be used to develop a kinetic model aiming at optimizing synthesis conditions to grow ni tride materials of much higher quality. All calculations will be performed with Vienna Ab-initio Simulation Package (VASP), using plane wave basis sets and PAW potentials. The kinetic barriers for NH3 on the nitride surfaces and ZrB2 will be computed using the Nudged Elastic Band Method. The scale of the proposed calculations exceeds small computer cluster capacity. Therefore, we are requesting the LCRC resources to perform the outlined calculations. The speed and reliability of the Fusion cluster will allow us to accomplish the work in a timely manner. The calculations of the InN, GaN, and ZrB2 surfaces in the presence of N and In (Ga) precursors will take approximately 48 hours each when ran on 48 processors. At least 50 different structures will be calculated. The frequency calculations for the ZrB2 configurations will take approximately 96 hours each when ran on 48 processors. We are planning to run about 25 of them. The calculations of the reaction energy barriers (NEB), on the other hand, will require about 96 hours each when ran on 48 processors based on our previous work. At least 60 reaction barriers for NH3 decomposition (i) in the presence of In (Ga) precursors, (ii) on strained surfaces and (iii) miscut surfaces will be performed for InN and GaN surfaces. Therefore, a total allocation of 700,000 processor hours will be needed to perform the proposed calculations. Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 700000 Q1: 250000 Q2: 150000 Q3: 150000 Q4: 150000 Justification: This work pursues development of models to describe kinetically controlled growth of group-III nitrides based on first-principles approaches. The project is part of Argonne strategic Materials for Energy initiative. Three different surfaces (i.e. the (0001) surfaces of InN, GaN, and ZrB2) will be investigated in detail in the presence of MOCVD precursors. Since VASP does not scale very well beyond 48 processors for our systems, it precludes the use of short-time large-scale capabilities of leadership class computer facilities. In fact, the scaling of the calculations is dramatically reduced with the number of processors beyond 48. The use of Fusion cluster will allow us to overcome the computational challenges posed by the significantly large system sizes necessary for modeling various surfaces as well as decomposition and diffusion processes, which are computationally very demanding. The detailed estimate of the required computer time is given at the end of the project description. Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Project Request: MnO2_interfaces
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: Roy Benedek Applicant's institution: ANL Applicant's division: CSE Project Name: MnO2_interfaces Project title: Dynamical simulation of structure and reactions at alpha-MnO2-electrolyte interfaces Associated funding: Grand Challenge LDRD Other Systems: Science: Lithium-air batteries are presently the subject of a Grand Challenge LDRD project at Argonne. Within this program, alpha-MnO2, a tunneled form of manganese dioxide, is under active investigation as a possible electrocatalyst, which appears to facilitate the charge step (delithiation of the cathode), a critical process for cyclability. In addition to alpha-MnO2, another crucial component of lithium-air batteries is the electrolyte (solvent + salt), with which it must function stably. Among the electrolytes under investigation are propylene carbonate and siloxane-based systems. The electrochemical reactions that the battery is designed to exploit occur at, or in the vicinity of, the interface between electrolyte and electrocatalyst (alpha-MnO2). Characterization of the alpha-MnO2-electrolyte interface, and how it affects the electrochemical reaction, therefore, would be a great benefit for gaining an understanding of battery operation, and for engineering improved performance. Atomic scale simulations can provide valuable assistance in this characterization. First principles simulations have already been performed for alpha-MnO2 (M. K. Chan, CNM) and for electrolyte (K. C. Lau, MSD) separately. Furthermore, simulations of the possible discharge reaction products, lithium peroxide and oxide, have also been performed. Simulations of the alpha-MnO2/electrolyte interface, however, have not yet been attempted, with or without the possible interphase of the discharge reaction product. Calculation of the structure and energies of the alpha-MnO2/electrolyte interface, possibly in the presence of discharge products, is therefore a key part of this project. The simulations of the alpha-MnO2/electrolyte interface are a necessary preliminary to the simulation of the electrocatalytic reactions that may occur in its vicinity. Simulation tools designed to model rare events (which occur infrequently on typical time scale of classical molecular dynamics simulations), such as the Blue-Moon ensemble and metadynamics, are expected to be helpful in the prediction of activation energies, and their dependence on environment (proximity of aMO, possibly activated by defects) for electrochemical reactions at the cathode. Both the interface structure simulations and the rare event simulations will be performed at the first-principles level. In this project we intend to explore charge-step (delithiation) reactions, for example the oxidation of lithium peroxide, to obtain lithium ions and LiO2 (superoxide) anions. These rare-event simulations are intended to reveal atomic scale mechanisms as well as activation energies that determine the kinetics of the process. Project description: Experiments Planned The project can be divided into three phases: 1. Surface structure of alpha-MnO2 Work by M. K. Chan shows that a corrugated (110) surface [which is closely related to the 2x2 tunnel structure of aMO] is a low energy surface. This appears to be a logical surface to start with. Besides the (110) surface, other low-index surfaces, such as (001) [perpendicular to the tunnels] and (100) will be addressed. Dynamical as well as static structure simulations will be performed. 2. Alpha-MnO2/electrolyte interface PC is a candidate electrolyte with relatively small molecular size, and is therefore attractive as a first system to consider. Siloxane-based electrolytes, with somewhat larger molecules will also be considered. 3. Rare event simulations of interface reactions Both metadynamics and blue-moon ensemble simulations will be utilized. A primary objective is to address the delithiation (charge-step) process (for example, of lithium peroxide) at the interface, which frees oxygen at the cathode (oxygen-evolution reaction). Computational Methods and Software The density functional theory simulation code VASP will be employed in most of the work. The code CP2K, which has favorable scaling, but is less documented, will also be considered. Visualization The public domain graphics packages VMD and VESTA are convenient for depicting three dimensional atomic configurations, and for making movies of dynamical processes. Scalability Issues The smallest cells for which the effects of interest can realistically be simulated will contain about five hundred atoms, with perhaps a 200 atom slab of alpha-MnO2, and 300 atoms of electrolyte. All numerical shortcuts (gamma-point only, soft pseudopotentials) will be exploited. We expect to run on 64-256 processors. Calculation size Of the calculations proposed here, the rare event simulations (item 3) will have the largest CPU-time requirements per unit MD-time step of 1 fs. The constrained MD simulations performed in blue-moon-ensemble simulations may require about 1 processor-hr per time step, for a cell size of about 500 atoms (using perhaps 128 processors). Obtaining a force curve for blue-moon-ensemble activation free energy calculation may require of the order of 100 -200 k processor hrs. Simulations of the surface structure (item 1) are expected to require of order 1 processor-minute per relaxation step. Considering three different surfaces and three temperatures, and of order 2x10**5 steps for each case, the total time for this step would also be of order 100k processor hrs. The interface structure simulations (item 2) will have time step requirements, and total processor time requirements, comparable to those for item 3. Thus, requirements for the entire project are thought to be of order 500,000 processor hours. Project URL: Requested allocation: 475000 Q1: 100000 Q2: 125000 Q3: 125000 Q4: 125000 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

[LCRC Accounts] Yearly Allocation Request from nanocatalysis
by accounts＠lcrc.anl.gov 13 Sep '11

13 Sep '11

Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Peter Zapol Project Name: nanocatalysis Division: MSD Project title: Catalysis in nanostructured materials Associated funding: BES-Chemical Sciences; Other Systems: Science: Our recent theoretical work on Pt and Pd nanoparticles and subnanometer clusters coupled to experiments at Argonne have resulted in several intriguing results on size and shape dependence of rectivity. The indications of modified selective catalytic activity in these systems require computational studies to understand catalytic reaction mechanisms and their dependence on the size and structure. This will involve large-scale calculations using molecular dynamics and electronic structure methods to find out optimized structures , reaction energies and barriers of catalytic reactions, diffusion barriers along the surfaces. Project description: Molecular dynamics studies using many-body potentials as implemented in DL_POLY and LAMMPS computer codes and DFT studies in VASP will be used to simulate support structures. Radial distribution functions will be compared to available experiments. The MD codes will uses 32 to 64 PEs for about 50,000 atom calculations. The geometries from molecular dynamics simulations will be used in studies of catalytic activity. We will study reactions on supported metal subnanometer clusters, extending the composition from Pt and Pd to other transition metals. We will also continue investigations on using thermodynamics to modify shape of catalytic particles, in particular in electrochemical environment. Density functional studies will be performed using VASP, NWChem and GAUSSIAN codes to determine reaction energies and barriers. The VASP calculations typically use 24-64 processors for 24 hours to 48 hours. Expected number of members in the project is three to six. Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 600000 Q1: 150000 Q2: 150000 Q3: 150000 Q4: 150000 Justification: The allocation is used by several postdoctoral and staff scientists to perform calculations on DOE-funded projects. From previous year's experience, we have used efficiently a large allocation using well-scaling codes (VASP, NWChem, DL-POLY). Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Yearly Allocation Request from compnano
by accounts＠lcrc.anl.gov 13 Sep '11

13 Sep '11

Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Peter Zapol Project Name: compnano Division: MSD Project title: Computational Nanoscience Associated funding: BES-DOE Other Systems: Science: Electronic structure calculations of chemical reactivity of nanomaterials is a complex problem involving large scale computations. It is important in catalysis, nanocrystalline materials, bio-inorganic interfaces, energy transfer and conversion, etc. Typically, explicit treatment of hundreds to thousands of atoms is required for accurate results on reaction energetics, structures of transition states and reaction mechanisms. Project description: Computational studies aimed at materials by design will be performed using molecular orbital theory, density functional and molecular dynamics methods as implemented in Gaussian09, VASP, and NWChem computer codes. We have successfully performed NWChem, G09 and VASP calculations on Fusion last year. Good scalability of the codes in 16 to 256 PE range makes Fusion an ideal match to our needs. We will continue our studies of chemical reactivity on the surfaces and interfaces of nanomaterials in FY12. In particular, we will perform DFT calculations coupled to thermodynatic treatments to investigate properties of nanoparticles as a function of their size and shape for metals and metal oxides. We will continue work on computational screening of dopants in oxide-based materials for light to energy conversion that resulted in useful predictions for surface modification. We plan to extend these studies to reaction pathways from partially reduced products such as formaldehyde to useful hydrocarbons. We will extend these studies to different size and shape of oxide particles. In addition, we will work on carbon nanomaterials using them both as support for heterophase nanostructures with novel chemical properties to investigate their potential for electrochemical applications in energy conversion and storage. Here we will build on our last year work for supported subnanomet er clusters. A typical Gaussian calculation will take for 4 to 16 nodes anywhere from 1-2 hours to several days. VASP and NWChem calculations may involve 16 to 256 nodes and up to several days depending upon the size of the problem. This project will involve 5-7 members of the Molecular Materials Group. Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 640000 Q1: 160000 Q2: 160000 Q3: 160000 Q4: 160000 Justification: The codes that we will use have been benchmarked on fusion and have shown excellent scalability for the number of processors used. From previous year's experience, we have used efficiently a large allocation, about a third of time for Gaussian and NWChem, and the rest for VASP calculations, to perform calculations on DOE-funded projects. Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Project Request: kinetic
by accounts＠lcrc.anl.gov 13 Sep '11

13 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: Zeke Insepov Applicant's institution: ANL Applicant's division: MCS Project Name: kinetic Project title: Kinetic Monte Carlo simulation of radiation defects evolution and thin film deposition Associated funding: RERTR AFCI Other Systems: Science: As a part of the continuing effort to study the microstructure evolution of Molybdenum (Mo) under irradiation of Xenon (Xe) Ions at different temperature regimes and different radiation doses, an on-lattice Kinetic Monte Carlo model is being built for that purpose. The first objective of this model is to investigate the bimodal gas bubble size distribution observed experimentally by simulating defects diffusion through the matrix during the so-called kinetic phase, in contrast to ballistic phase. The “ballistic phase” accounts for the creation of defects through cascade of collisions made by a primary knock-on atom (PKA), while the “kinetic phase”, accounts for the diffusion-controlled dynamical evolution of such defects. The on-lattice model employs space discretization technique and hence allows the defect to exist only on a defined set of points in the space depending on the type of the defect. Normally, this set of points consists of certain interstitial sites, in addition to lattice sites. KMC algorithm is then used to simulate defects hops between these points. Understanding the mechanisms involved in defects diffusion process is so crucial, as different mechanisms,at the same temperature and defect population, could result in completely different behavior. The existence of thermally activated slipping systems in Mo, which can at finite temperature compete with the crowdion and <111> dumbbell diffusion mechanism, adds more complication to the simulation of interstitial diffusion. KMC gives the capability of isolating different mechanisms, in order to explore the contribution of individual mechanisms to a certain phenomenon. The first step focused on studying the embryonic stage (incubation and nucleation stage) of voids and followed by tracking the spatial arrangement and the void size distribution as a function of temperature and displacements per atom (dpa). The objective of this part is to investigate the possibility of void superlattice formation and the role played by hypothesized shadowing effect in this case. Project description: We will use the kinetic Monte Carlo method that was realized as a SPPARKS package at SANDIA Labs. SPPARKS is a parallel Monte Carlo code for on-lattice and off-lattice models that includes algorithms for kinetic Monte Carlo (KMC), rejection kinetic Monte Carlo (rKMC), and Metropolis Monte Carlo (MMC), coupling of this code with MD-Synchronous parallel kMC. The scaling properties of KMC solvers currently available in SPPARKS : linear search, O(N) tree search, O(logN) composition-rejection search, O(1) We plan to run 150-200 jobs of 256 cores x 5-10 hours. For validation and fortification of this KMC model, comparison with experimental results is needed. The in situ TEM study of Xe implantation in pure Mo, took place at ANL facilities, gives valuable data on Xe bubble size distribution in Mo, which can be used as a reference data to be compared with the results of this model. Accelerator experiments are planned to be conducted at Argonne IVEM and Atlas machines, to verify the simulation results. SPPARKS will run on any parallel machine that compiles C++ and supports the MPI message-passing library. This includes distributed- or shared-memory machines. We expect 4-5 users that will be working on this project. Project URL: Requested allocation: 450000 Q1: 50000 Q2: 150000 Q3: 150000 Q4: 100000 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

[LCRC Accounts] Yearly Allocation Request from glassforms
by accounts＠lcrc.anl.gov 13 Sep '11

13 Sep '11

Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Haiying He Project Name: glassforms Division: MSD Project title: First-principles Calculations of Mineral and Glass Dissolution under Various pH Conditions Associated funding: DOE-NE Other Systems: Science: The dissolution of glass forms in nuclear waste disposal determines their long term stability, which is critical for evaluating disposal scenarios. The dissolution behavior of minerals and glasses has been found to strongly depend on the pH value of the aqueous environment. The delicate balance of ions at the solid-water interface in aqueous solutions is still a big challenge both for accurate computational modeling and for unambiguous experimental measurements. This project aims to develop physical models to simulate the water-borosilicate glass forms interface at the atomic level under various pH conditions and to calculate the rates of dissolution for Si, Al and other constituting elements under acidic, neutral and basic conditions. These studies are based on first-principles calculations. The calculated reaction barriers together with the surface model can then be used to estimate the overall dissolution rates and compare with experimental values. These data wil l also serve as a valuable dataset to feed into kinetic models for better understanding of the dissolution behavior and prediction of dissolution rates. Project description: Task I: The task of our research is to include the bulk water effect in the process of glass and mineral dissolution. In the past year, we have calculated hydrolysis reaction barriers of the Si-O-Si bridges and the Si-O-Al bridges at surface sites of orthoclase feldspar (KAlSi3O8) as a model system in different protonation states – neutral (≡SiOH), protonated (≡SiOH2+) and deprotonated (≡SiO-) using the Nudge Elastic Band (NEB) method. In most of the calculations, we have only considered a single H2O attack on the Si-O-Si or Al-O-Si bridge that causes bond rupture with the simultaneous H2O dissociation. Studies of the bulk water effect will determine the role of proton shuttle in aqueous solution and the entropic effect of water. Both effects contribute to the free energy reaction barriers. This task will be accomplished by including explicit H2O molecules in a large supercell and performing constrained molecular dynamics (MD) simulations. Hereby , the free-energy profile is a thermodynamic integration over the restoring forces along a parameterized reaction coordinate using the rare-event ensemble technique, which is newly implemented in VASP. We will also use this technique to determine the chemical potentials of protons in solution for the surface site distribution model which we have developed. Task II: The second task for this year is to extend our study from a model crystalline material to glasses. Using initial structures from classical molecular dynamics, we will continue with first-principles calculations of reaction energies and barriers. Multi-component amorphous structures are some complexities that we need to address in the modeling of glasses. It suggests that we need large supercells and a large number of sampling points. The major components that we will consider in glasses are Si, Al, B (network formers) and Na, Ca (network modifiers). We will carry out similar investigations in terms of the surface protonation and hydrolysis reaction barriers under various pH conditions for bridges involving these elements. The properties that we are going to calculate in this part include structures and energies of reactants, intermediates and products, as well as reaction barriers (based on the NEB method). In some cases, we will validate results using smaller clus ters at high levels of ab initio theory such as CCSD(T) as implemented in NWChem and Gaussian09 packages. The calculated reaction barriers together with the surface site distributions determined from the surface model will then be used in kinetic modeling to calculate the dissolution rates. This project will primarily use the density functional theory (DFT) implemented in electronic structure code VASP. The system sizes of our calculations ranges from 100 to 300 atoms. VASP has been tested to scale very well in parallel processing on Fusion. The optimum number of processors used for these systems is 16-64 processors per calculation. Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 800000 Q1: 150000 Q2: 250000 Q3: 200000 Q4: 200000 Justification: The first-principles calculations will be carried out using the electronic-structure code VASP. In Task I, ab initio MD simulations are conducted, where each trajectory runs for 30-50 ps and it takes 16 processors to run for 96-160 hours (average 2048 core hours per trajectory). We will need to run 100 trajectories for five sets of calculations. That requires 200,048 core hours. In Task II, NEB calculations are conducted to calculate the reaction pathways for a variety of composing elements under acidic, neutral and basic conditions. A single optimization calculation takes 48 hours on 32 processors. We are considering 30 supercell samplings of silicate glasses including multiple components (Si, Al, B, Na, K). The calculation for possible reactants, intermediates, products will take about 138,240 core hours. It takes about 48 hours on 160 processors (5 images) to calculate a reaction barrier along the reaction pathway. The total estimated computer time for bar rier calculation is 460,800 core hours. In sum, the total requested allocation for this project is 800,000 core hours. Thank You, The LCRC Accounts System

1 0