100k granted and note sent to use the blues biggpu and or Fusion. - JB On Tue, Jul 28, 2015 at 10:29 AM, [email protected] <[email protected]> wrote:
Hello,
A change in allocation has been requested:
Requester: rbenedek (Roy Benedek) Project: LiBattModeling Title: Lithium-Ion Battery Electrode Modeling Description: We will perform simulations to evaluate substitutions such as Mo and Ru to partially replace Mn in the Li2MnO3 component of xLi2MnO3•(1-x)LiMO2 composites. The simplest choices of M are M=Co and M=Mn0.5Ni0.5. Relevant replacement fractions may lie in the range y=0.25 to 0.50; higher values may result in a structure significantly different from that of xLi2MnO3•(1-x)LiMO2.
A. Stability. Simulations will be performed to elucidate the structure and stability of Li2(M’yMn1-y)O3, both separately and in xLi2(M’yMn1-y)O3•(1-x)LiMO2 composites. Of interest, for example, would be determination of the pseudo-binary phase diagram of Li2(M’yMn1-y)O3 in the monoclinic structure. Although a full phase diagram calculation (e.g., by the cluster expansion method) would require a huge computational effort, perhaps some features of the phase diagram can be established with only moderate effort.
B. Migration of M’. Since the migration from the metal layer to the lithium layer in the activated standard material xLi2MnO3•(1-x)LiMO2 leads to degradation, stability of M’ against such degradation is desired. Simulation of the energy along a migration path between octahedral and tetrahedral sites will provide an indication of the stability of M’ candidates in tetrahedral sites.
C. The performance of M’ in xLi2(M’yMn1-y)O3•(1-x)LiMO2 cathodes depends on the redox energies associated with M’. Simulations of the voltage as a function of delithiation for Li2(M’yMn1-y)O3, both as an isolated system and as part of a composite, will provide an estimate of the capacity and energy density as a function of upper charge voltage.
Current: undetermined amount Justification: The VASP software employed in this work is known to provide relatively good scaling, for the types of simulations proposed, up to a number of cores roughly equal to the number of atoms in the periodic cell. Typically, our cells include up to about 400 atoms.
Requested: 100000
A specific reason has been given: John Low has recently compiled the beta test form of the GPU version of VASP (which we have access to through my VASP license), and tests are desired. My allocation (and that of my collaborator, Hakim Iddir) for the current allocation year, however, are exhausted. I am therefore requesting a supplemental allocation to enable me to test the GPU form of VASP.
This needs to be approved and the final allocation amount decided upon.
Thank You, The LCRC Accounts System