August 2016 - Allocations-admins - mailman3.cels.anl.gov

New LCRC report submitted for StructJuMP
by support＠lcrc.anl.gov 31 Aug '16

31 Aug '16

Full project report at https://accounts.lcrc.anl.gov/reporting/project/2016/?project=StructJuMP

1 0

New LCRC report submitted for StructJuMP
by support＠lcrc.anl.gov 31 Aug '16

31 Aug '16

Full project report at https://accounts.lcrc.anl.gov/reporting/project/2016/?project=StructJuMP

1 0

New LCRC report submitted for HEIGHTS-3D
by support＠lcrc.anl.gov 31 Aug '16

31 Aug '16

Full project report at https://accounts.lcrc.anl.gov/reporting/project/2016/?project=HEIGHTS-3D

1 0

New LCRC report submitted for UNEDF_Optimization
by support＠lcrc.anl.gov 31 Aug '16

31 Aug '16

Full project report at https://accounts.lcrc.anl.gov/reporting/project/2016/?project=UNEDF_Optimiz…

1 0

New LCRC report submitted for PROTEUS_2015
by support＠lcrc.anl.gov 31 Aug '16

31 Aug '16

Full project report at https://accounts.lcrc.anl.gov/reporting/project/2016/?project=PROTEUS_2015

1 0

New LCRC report submitted for ESMF
by support＠lcrc.anl.gov 31 Aug '16

31 Aug '16

Full project report at https://accounts.lcrc.anl.gov/reporting/project/2016/?project=ESMF

1 0

[LCRC Accounts] Yearly Allocation Request for MBM_CEES2
by accounts＠lcrc.anl.gov 31 Aug '16

31 Aug '16

Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: hakim iddir Project Name: MBM_CEES2 Division: MSD Project title: Modeling Battery Materials CEES2 Associated funding: DOE, EFRC Other Systems: NERSC Science: As a part of continuing effort to contribute to the battery research here at Argonne, and to the EFRC (CEES II) project. We will perform theoretical work based on density functional theory using “VASP”, to simulate the equilibrium particle shapes of non-stoichiometric and doped LiMn2O4 particles. As stated in the CEES II proposal, it is central to this task to understand the electrode structure instabilities induced electrochemically. LiMn2O4 is ideal as a high-capacity Li-ion battery cathode material by virtue of its low toxicity, low cost, and the high natural abundance of Mn. However, this material suffers from Mn dissolution problem that not only induces a capacity fade [1], but can also poison the anode. The dissolution is associated with the disproportionation of Mn3+, and is dependent on the surface structure, hence the crystal orientation as well [2, 3]. For example, (111) is expected to be more stable as the reconstruction moves Mn to the surface sub-la yers[4]. Previous works show that (111) surface is more stable than (100) and (110) surfaces with respect to Mn dissolution [3, 5]. To improve its stability with respect to Mn dissolution, one would like to predict the synthesis conditions that would preferentially enhance the (111) surface ratio over that of the other facets, not only in pristine stoichiometric state, but also non-stoichiometric and doped surfaces. However, it is also known that (111) is not favorable for Li diffusion, hence reducing the rate capability [3]. A truncated cubo-octahedron was proposed as a good compromise between stability and rate capability[3]. Karim et al. determined the equilibrium particle shape of pure stoichiometric LiMn2O4 particles, in which they predicted a stable reconstructed (111) facet [4], in agreement with the observed shapes at high temperatures for Li1+xMn2-xO4 (x=0.125)[6]. However, several experimental results show that the particle shapes obtained do not all exhibit the predicted shape by the model. This discrepancy could be due to the nature and stoichiometry of the different surfaces, which depend on the experimental conditions the particles were synthesized in. In this work we will investigate the particles shape for non-stoichiometric as well as with dopants, with the goal being an optimized particle shape that will provide both stability and good rate capability. This work is in line with the two of the DOE Basic Research Needs Report on Electrical Energy Storage: • Rational design of interfaces and interphases in chemical energy storage. • Novel designs and strategies for electrochemical systems. References: 1. Tarascon, J.M. and D. Guyomard, Li Metal‐Free Rechargeable Batteries Based on Li1 + x Mn2 O 4 Cathodes ( 0 ≤ x ≤ 1 ) and Carbon Anodes. Journal of The Electrochemical Society, 1991. 138(10): p. 2864-2868. 2. Hirayama, M., et al., Dynamic Structural Changes at LiMn2O4/Electrolyte Interface during Lithium Battery Reaction. Journal of the American Chemical Society, 2010. 132(43): p. 15268-15276. 3. Kim, J.-S., et al., A Truncated Manganese Spinel Cathode for Excellent Power and Lifetime in Lithium-Ion Batteries. Nano Letters, 2012. 12(12): p. 6358-6365. 4. Karim, A., S. Fosse, and K.A. Persson, Surface structure and equilibrium particle shape of the LiMn2O4 spinel from first-principles calculations. Physical Review B, 2013. 87(7): p. 075322. 5. Thackeray, M.M., Manganese oxides for lithium batteries. Progress in Solid State Chemistry, 1997. 25(1–2): p. 1-71. 6. Takada, T., et al., Structure and electrochemical characterization of Li1+xMn2−xO4 spinels for rechargeable lithium batteries. Journal of Power Sources, 1999. 81–82(0): p. 505-509. As mentioned in our 2016 LCRC report for this project, we have completed the surface calculations of LiMn2O4 (LMO) and determined the particle morphology as a function of Li and O stoichiometry. We have also started investigating the dopants effects on Mn oxidation state in LMO, in order to mitigate Mn disproportionation that is believed to be the root cause of dissolution. The system sizes used so far are quite limited, and do not allow us to get a better picture of the actual dopant profile (surface affinity) with respect to LMO(100) surface. In this project, we propose to continue investigating in details the transition metal doping of LMO, with particular emphasis on the surface affinity calculations, using larger system sizes and different computational techniques (Cluster Expansion CE) coupled with Mote Carlo (MC) simulations to determine the dopants concentration profiles at different dopants concentration levels and temperatures. Project description: Our goal is to improve the methods for determining low energy structures of Li-ion battery materials by using a combination of ab initio calculations with the cluster expansion and Monte Carlo simulations (MC). Several dopant candidates are currently being investigated in our MBM project (for EFRC CEES2 DOE project), using density functional theory with rather limited system sizes. In this project we intend to increase the system size and use less expansive techniques to investigate the dopants surface segregation profiles at different concentrations and temperatures. For a given LiMn2O4 surface (100, 110), using cluster expansion (CE) method [1], we generate a large number of configurations where Mn will be substituted with other candidate transition metals such as Ni, Co, Al, and Ti. The validated set of the CE coefficients is then used to perform MC simulations for large supercells and to address the relative energies of configurations and the effect s of ordering at different concentrations and temperatures. The programming model that will be used in this work is based on a Fortran and Monte-Carlo code, already written by Anh Ngo. Our study will provide fundamental information on electronic properties of this important class of complex oxides and guide the ongoing experimental efforts within the CEES2 EFRC projects at Argonne. A total of four members will be involved in this project. Industry partnership: Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 921600 Q1: 250000 Q2: 250000 Q3: 250000 Q4: 171600 Justification: To generate the interaction coefficients needed to build the CE, density functional theory (DFT) calculations using the VASP package, with periodic boundary conditions, gradient corrected PBE functional will be used. The codes and programs needed to carry out this work are currently available on NERSC computers and scale reasonably well on parallel processors. Energy optimization of a doped LiMn2O4 (~120 atoms) using GGA+U takes approximately 3072 CPU hours (16*4*48) and we plan roughly 300 of such calculations for ab initio calculations with the cluster expansion . For MC calculations, we will perform 5 different concentrations and 10 temperatures calculations. This takes roughly 768 CPU hours (16*1*48) The total amount of computer allocation that will be needed to accomplish this work is: 300*3072 + (16*2*24)*15 = 921600 core hours. Storage requirements: Thank You, The LCRC Accounts System

1 0

New LCRC report submitted for LCRC_on_demand
by support＠lcrc.anl.gov 31 Aug '16

31 Aug '16

Full project report at https://accounts.lcrc.anl.gov/reporting/project/2016/?project=LCRC_on_demand

1 0

New LCRC report submitted for Comp_POP_Catalysts
by support＠lcrc.anl.gov 31 Aug '16

31 Aug '16

Full project report at https://accounts.lcrc.anl.gov/reporting/project/2016/?project=Comp_POP_Cata…

1 0

[LCRC Accounts] Yearly Allocation Request for CAT-engine-modeling
by accounts＠lcrc.anl.gov 31 Aug '16

31 Aug '16

Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Sibendu Som Project Name: CAT-engine-modeling Division: ES Project title: Single Cylinder engine simulations to understand fuel property effects Associated funding: DOE Office of vehicle technologies under the co-optima program Other Systems: We also expect to get core hours on the NREL Peregrine system Science: This project will use an Eulerian-Lagrangian Computational Fluid Dynamics (CFD) software called CONVERGE to simulate internal combustion (IC) engines using High Performance Computing (HPC). The effect of fuel physical properties on a single cylinder engine performance will be assessed. Project description: It is well known that fuel physical and chemical properties effect engine performance and emissions. The goal of the co-optima program of DOE is to co-optimize fuel properties together with engine performance to obtain maximum benefits of either. We plan to simulate different fuels under various low-temperature combustion conditions to understand the influence and relative importance of different fuel physical properties on engine performance. For instance, the goal is to understand if fuel density or viscosity or boiling point is more critical in defining its ignition delay or amount of soot emission in the engine. The first step will be to validate the engine simulations against available experimental data under DOE’s co-optima program. Then we plan to run 1000s of engine sector simulations by varying all the fuel properties to a certain extent. These engine simulations will then be post-processed using a global sensitivity analysis tool to further o ur understanding on the influence of fuel properties on engine performance. Industry partnership: None Project URL: http://verifi.anl.gov/ Current FY Hours Used: undetermined amount New FY Requested allocation: 400000 Q1: 100000 Q2: 100000 Q3: 100000 Q4: 100000 Justification: Storage requirements: 5 TB Thank You, The LCRC Accounts System

1 0