This thread is locked.Only browsing is available.
Top Page > Browsing
Different total energy: X and Z periodicity tested with 1D carbon-chain
Date: 2018/07/19 21:48
Name: Dongzhe Li   <dongzhe.li@uni-konstanz.de>

Dear OpenMX developers and users,

As a new user OpenMX, I encountered a strange problem, I got different total energy for the ame system with different choice of the periodicity. I tested the example in the folder of "openmx3.8/work/negf_example" as follow:


----- Supercell with X periodicity ------------

Atoms.Number 3
Atoms.SpeciesAndCoordinates.Unit Ang # Ang|AU
<Atoms.SpeciesAndCoordinates
1 C -1.5 0.000 0.000 2.0 2.0 ## valance e: up and down
2 C 0.000 0.000 0.000 2.0 2.0
3 C 1.50 0.000 0.000 2.0 2.0
Atoms.SpeciesAndCoordinates>
Atoms.UnitVectors.Unit Ang # Ang|AU
<Atoms.UnitVectors
4.5 0.0 0.0
0.0 10.0 0.0
0.0 0.0 10.0
Atoms.UnitVectors>

scf.Kgrid 200 1 1 # means n1 x n2 x n3

*******************************************************
Total Energy (Hartree) at MD = 1
*******************************************************

Uele = -5.314335569994

Ukin = 11.054942068711
UH0 = -45.715234222346
UH1 = 0.127735338871
Una = -7.134993033403
Unl = -1.320621112207
Uxc0 = -2.622881498184
Uxc1 = -2.622881498184
Ucore = 31.045065274667
Uhub = 0.000000000000
Ucs = 0.000000000000
Uzs = 0.000000000000
Uzo = 0.000000000000
Uef = 0.000000000000
UvdW = 0.000000000000
Utot = -17.188868682078

----------------------------------------------------


----- Supercell with Z periodicity ------------

Atoms.Number 3
Atoms.SpeciesAndCoordinates.Unit Ang # Ang|AU
<Atoms.SpeciesAndCoordinates
1 C 0.000 0.000 -1.500 2.0 2.0 ## valance e: up and down
2 C 0.000 0.000 0.000 2.0 2.0
3 C 0.000 0.000 1.500 2.0 2.0
Atoms.SpeciesAndCoordinates>
Atoms.UnitVectors.Unit Ang # Ang|AU
<Atoms.UnitVectors
0.0 0.0 4.5
10.0 0.0 0.0
0.0 10.0 0.0
Atoms.UnitVectors>

scf.Kgrid 1 1 200 # means n1 x n2 x n3

*******************************************************
Total Energy (Hartree) at MD = 1
*******************************************************

Uele = -5.123669654675

Ukin = 11.260423137399
UH0 = -45.715234222346
UH1 = 0.113523538357
Una = -7.205113183330
Unl = -1.260421485055
Uxc0 = -2.628503177046
Uxc1 = -2.628503177046
Ucore = 31.045065274667
Uhub = 0.000000000000
Ucs = 0.000000000000
Uzs = 0.000000000000
Uzo = 0.000000000000
Uef = 0.000000000000
UvdW = 0.000000000000
Utot = -17.018763294402

----------------------------------------------------

Since in the OpenMX, the a-axis is the transport direction, this problem is kind of important, otherwise I have to rotate my structure...

I just created in a two different ways for the same system, and I also changed properly the k-mesh, why the total energy is still very different?

Any comments will be appreciated.
Best regards,
Dongzhe Li


メンテ
Page: [1]

Re: Different total energy: X and Z periodicity tested with 1D carbon-chain ( No.1 )
Date: 2018/07/20 08:52
Name: Mitsuaki Kawamura  <mkawamura@issp.u-tokyo.ac.jp>

Dear Dongzhe

Hello,

Please use k-mesh (200, 1, 1) for both cases.

Best regards,
Mitsuaki Kawamura
メンテ
Re: Different total energy: X and Z periodicity tested with 1D carbon-chain ( No.2 )
Date: 2018/07/20 14:20
Name: Dongzhe Li  <dongzhe.li@uni-konstanz.de>

Dear Mitsuaki,

Thanks for your answer.

I misunderstood the "OpenMX 3.8" at page 139.

"Then, the direction of (0.0, 0.0, 10.0) becomes the direction of electronic transport. As shown in the
above example, when you change the order of the lattice vectors, please make sure that the keyword
’scf.Kgrid’ has to be changed as well."

Best regards,
Dongzhe
メンテ

Page: [1]