150k granted and note sent to run on Fusion - JB On Tue, Jul 28, 2015 at 4:50 PM, [email protected] <[email protected]> wrote:
Hello,
A change in allocation has been requested:
Requester: kclau (Kah Chun Lau) Project: Li_air_battery Title: Computational studies of Li-Air battery components Description: In the coming fiscal year (FY2015), we need 1200 000 core-hour to complete our project. We will complete our current on-going work and continue to explore different problems in Li-air batteries based on experimental input from our colleagues at ANL. To complete the current ongoing study (i.e. the effects of K-doping in carbon electrode to facilitate ORR in Li-O2 cell, solvation and solubility of Li2O2 and LiO2 molecule in DME electrolytes), our experience indicates that 100, 000 hours is required.
For our new study at FY2015, the computational methods will involve largely AIMD (CPMD, VASP, CP2K) and DFT method (Gaussian09, VASP, Quantum Espresso). For Gaussian09, the DFT calculation scales well up to 8-16 processors and for 4-6 nodes is the optimal number of nodes for geometry optimization for system size below 100 atoms. We will require about 200,000 core hours to study the Nuclear Magnetic Resonances (NMR) properties of Li2O2 and LiO2 stoichiometric clusters at different electrolytes in solution phases.
To extend our previous solvation study on LiO2 in DME electrolyte, we will study extend our study on the stability and solvation of LiO2 in different dielectric media with the representation of various electrolytes-based solution (e.g. DMSO, DMA, ionic liquids) based on both combined AIMD simulation and quantum thermochemistry calculations using VASP and Gaussian09 code. For a single Γ-point calculation with a total atom ~ 300-400 atoms AIMD simulation, a parallel efficiency of ~ 60% -75% can be achieved using 256 cores. Thus to explore a moderately large systems (~ 300 atoms) with simulation time up to ~ 5ps in 1 fs time step, 50 000-100 000 core-hour is needed to perform the tasks for each electrolyte. To study a few chosen bulk electrolytes AIMD simulation, this will need 500 000 core-hour to complete these computational intensive simulations.
In addition to the studies on electrolytes, we will spend our significant efforts on LiO2 studies that directly related to our ANL colleagues’ future plan on the experimental studies in LiO2. Based on our understanding of LiO2 crystalline solid, we will carry out a systematic study on various low indexes stoichiometric and non-stoichiometric surfaces of LiO2 (in vacuum and with implicit solvent) to predict its thermodynamic equilibrium Wulff crystalline shapes based on DFT calculations which will take 150 000 core-hour. To extend our basic understanding of the crystal growth of LiO2, we will carry out a systematic study of a promising Li-metal alloy as a template growth for LiO2 crystal. Thus to construct a reasonable DFT model, a simulation cell of about 200-300 atom surface calculations within an optimal DFT accuracy is necessary, and it needs an extra 250 000 core-hour in this problem.
In conclusion, this will need 1200 000 core-hour (1.2 Million core-hour) to complete all these problems in FY2015.
Current: undetermined amount Justification: For the details and evidence how to use the hours effectively, please see the project description and project report (FY2014). Further details of the scaling and the performance of the codes can be found as follows: (1) VASP: http://cms.mpi.univie.ac.at/vasp/vasp/Performance_parallel_code_on_various_m... (2) CPMD: http://cpmd.org/documentation (3) Gaussian09: http://www.gaussian.com/g_prod/g09_glance.htm
Requested: 200000
A specific reason has been given: To complete an on-going five LiO2 VASP surface simulation (i.e. with system size of ~ 100 atoms for geometry optimization and two unfinished NEB calculations) with different stoichiometries that are needed for manuscript write-up.
This needs to be approved and the final allocation amount decided upon.
Thank You, The LCRC Accounts System