Can OpenMX do geometry and lattice optimization simultaneously? |
- Date: 2015/12/09 21:41
- Name: Yang Jin-Hoon
- Dear OpenMX users and Admin.
I'm new user of OpenMX and i successfully set up OpenMX on my cluster just a week ago.
And now, I'm trying to do some test whether it is correctly installed or not.
As I think, it properly working. The most of examples in work directory are going well.
However, I couldn't find the way how I can optimize the lattice and geometry of bulk material simultaneously.
I tested distorted Si unit cell as follows
Species.Number 1
<Definition.of.Atomic.Species
Si Si7.0-s2p2d1 Si_PBE13
Definition.of.Atomic.Species>
Atoms.Number 8 # Total number of atoms
Atoms.SpeciesAndCoordinates.Unit FRAC # Ang | AU | FRAC
<Atoms.SpeciesAndCoordinates
1 Si 0.10 0.00 0.00 2 2 # distorted along x axis
2 Si 0.50 0.50 0.00 2 2
3 Si 0.50 0.00 0.50 2 2
4 Si 0.00 0.50 0.50 2 2
5 Si 0.25 0.25 0.25 2 2
6 Si 0.75 0.75 0.25 2 2
7 Si 0.25 0.75 0.75 2 2
8 Si 0.75 0.25 0.75 2 2
Atoms.SpeciesAndCoordinates>
Atoms.UnitVectors.Unit Ang # Ang|AU
<Atoms.UnitVectors
5.41 0.00 0.00 # distorted manually
0.00 5.31 0.00
0.00 0.00 5.43
Atoms.UnitVectors>
However, lattice constant dose not changes with 'MD.Type' of Opt, DIIS, EF, RF, whatever.
I want to know there is a way to optimize geometry and lattice constant simultaneously.
If not, should I calculate Energy vs. Lattice constant using MD.Type of EvsLC ? and then perform geometry optimization?
Please know me how can I get optimized geometry and lattice constant
Best regards
 | |
This thread is locked.Only browsing is available.
Re: Can OpenMX do geometry and lattice optimization simultaneously?