[DFTB-Plus-User] Problem of ZnO bulk structure optimization
yangming
ming_young at msn.com
Fri Jul 24 13:28:54 CEST 2009
Dear all,
I have two problems of DFTB+ calculations as follwing:
(1) I did a geometry optimization of a hydrogen atom on graphene, and found that it takes me about 80 hours already but does not converge yet. However, I did geometry optimization of a similar system, oxygen atom on graphene, and it took me only 1 hour to finish the job. I am really confused, because these two systems are almost the same.
(2) I did geometry optimization of ZnO bulk. After relaxation, I found that the symmetry of relaxed ZnO lowered from originally P63MC(186) to CMC21 (36). This is not consistent with DFT calculation results (VASP). And, I believe that this symmetry difference would lead to later unreliable electronic results such as band structures or DOS.
The input parameters are attached for the calculations. Please take a look, and give me some comments.
Thank you very much!
Kevin
__________________
1. Input parameters for Hydrogen on graphene. (the SKF parameters are based on pbc-0-1)
Geometry = GenFormat {
61 S
C H
1 1 2.822666667 -0.000000000 11.700000000
2 1 0.705666667 1.222500000 11.700000000
3 1 0.000000000 -0.000000000 11.700000000
4 1 2.117000000 1.222500000 11.700000000
5 1 7.056666667 -0.000000000 11.700000000
6 1 4.939666667 1.222500000 11.700000000
7 1 4.234000000 -0.000000000 11.700000000
8 1 6.351000000 1.222500000 11.700000000
9 1 11.290666667 -0.000000000 11.700000000
10 1 9.173666667 1.222500000 11.700000000
11 1 8.468000000 -0.000000000 11.700000000
12 1 10.585000000 1.222500000 11.700000000
13 1 2.822666667 2.445000000 11.700000000
14 1 0.705666667 3.667500000 11.700000000
15 1 0.000000000 2.445000000 11.700000000
16 1 2.117000000 3.667500000 11.700000000
17 1 7.056666667 2.445000000 11.700000000
18 1 4.939666667 3.667500000 11.700000000
19 1 4.234000000 2.445000000 11.700000000
20 1 6.351000000 3.667500000 11.700000000
21 1 11.290666667 2.445000000 11.700000000
22 1 9.173666667 3.667500000 11.700000000
23 1 8.468000000 2.445000000 11.700000000
24 1 10.585000000 3.667500000 11.700000000
25 1 2.822666667 4.890000000 11.700000000
26 1 0.705666667 6.112500000 11.700000000
27 1 0.000000000 4.890000000 11.700000000
28 1 2.117000000 6.112500000 11.700000000
29 1 7.056666667 4.890000000 11.700000000
30 1 4.939666667 6.112500000 11.700000000
31 1 4.234000000 4.890000000 11.700000000
32 1 6.351000000 6.112500000 11.700000000
33 1 11.290666667 4.890000000 11.700000000
34 1 9.173666667 6.112500000 11.700000000
35 1 8.468000000 4.890000000 11.700000000
36 1 10.585000000 6.112500000 11.700000000
37 1 2.822666667 7.335000000 11.700000000
38 1 0.705666667 8.557500000 11.700000000
39 1 0.000000000 7.335000000 11.700000000
40 1 2.117000000 8.557500000 11.700000000
41 1 7.056666667 7.335000000 11.700000000
42 1 4.939666667 8.557500000 11.700000000
43 1 4.234000000 7.335000000 11.700000000
44 1 6.351000000 8.557500000 11.700000000
45 1 11.290666667 7.335000000 11.700000000
46 1 9.173666667 8.557500000 11.700000000
47 1 8.468000000 7.335000000 11.700000000
48 1 10.585000000 8.557500000 11.700000000
49 1 2.822666667 9.780000000 11.700000000
50 1 0.705666667 11.002500000 11.700000000
51 1 0.000000000 9.780000000 11.700000000
52 1 2.117000000 11.002500000 11.700000000
53 1 7.056666667 9.780000000 11.700000000
54 1 4.939666667 11.002500000 11.700000000
55 1 4.234000000 9.780000000 11.700000000
56 1 6.351000000 11.002500000 11.700000000
57 1 11.290666667 9.780000000 11.700000000
58 1 9.173666667 11.002500000 11.700000000
59 1 8.468000000 9.780000000 11.700000000
60 1 10.585000000 11.002500000 11.700000000
61 2 5.645333277 6.112500000 13.000002000
0.000000000 0.000000000 0.000000000
12.70200000 0.000000000 0.000000000
0.000000000 12.22500000 0.000000000
0.000000000 0.000000000 20.00000000
}
Driver = ConjugateGradient {
MovedAtoms = Range { 1 -1 }
MaxForceComponent = 1.0e-4
MaxSteps = 1000
OutputPrefix = "geom.out"
}
Hamiltonian = DFTB {
SCC = Yes
SCCTolerance = 1.0e-5
MaxSCCIterations = 1000
Mixer = Broyden {
MixingParameter = 0.2
CachedIterations = -1
InverseJacobiWeight = 1.000000000000000E-002
MinimalWeight = 1.00000000000000
MaximalWeight = 100000.000000000
WeightFactor = 1.000000000000000E-002
}
KPointsAndWeights = SupercellFolding {
3 0 0
0 3 0
0 0 1
0 0 0
}
SlaterKosterFiles = {
C-C = "C-C.skf"
C-H = "C-H.skf"
H-C = "H-C.skf"
H-H = "H-H.skf"
}
MaxAngularMomentum = {
C = "p"
H = "s"
}
Charge = 0.0
SpinPolarisation = {}
Filling = Fermi {
Temperature = 1.000000000000000E-008
IndependentKFilling = No
}
ElectricField = {}
OrbitalResolvedSCC = No
ReadInitialCharges = No
Eigensolver = DivideAndConquer {}
OldSKInterpolation = No
OrbitalPotential = {}
Dispersion = {}
}
ParserOptions = {
ParserVersion = 3
WriteHSDInput = Yes
WriteXMLInput = No
StopAfterParsing = No
}
Options = {
WriteEigenvectors = No
WriteAutotestTag = No
WriteDetailedXML = No
WriteResultsTag = Yes
WriteDetailedOut = Yes
WriteBandOut = Yes
AtomResolvedEnergies = Yes
RestartFrequency = 20
RandomSeed = 0
}
______________________
2. Input parameters for ZnO bulk structure optimization:
#skf file is based znorg-0-1.
_______________________________________
Geometry = GenFormat {
4 S
Zn O
1 1 -0.000000000 1.842293516 0.010849207
2 2 -0.000000000 1.842293516 1.963057252
3 1 1.595472986 0.921146758 2.590444683
4 2 1.595472986 0.921146758 4.542652729
0.00000000 0.00000000 0.00000000
2.763440274226130 -1.595472986213910 0.000000000000000
0.000000000000000 3.190945972427830 0.000000000000000
0.000000000000000 0.000000000000000 5.159190953394310
}
Driver = ConjugateGradient {
MovedAtoms = Range {
1 -1
}
MaxForceComponent = 1.0e-4
MaxSteps = 1000
OutputPrefix = "geom.out"
AppendGeometries = No
Constraints = {}
}
Hamiltonian = DFTB {
SCC = Yes
SCCTolerance = 1.0e-5
Mixer = Broyden {
MixingParameter = 0.2
CachedIterations = -1
InverseJacobiWeight = 1.000000000000000E-002
MinimalWeight = 1.00000000000000
MaximalWeight = 100000.000000000
WeightFactor = 1.000000000000000E-002
}
KPointsAndWeights = SupercellFolding {
9 0 0
0 9 0
0 0 6
0.0 0.0 0.5
}
SlaterKosterFiles = {
Zn-Zn = "Zn-Zn.skf"
Zn-O = "Zn-O.skf"
O-Zn = "O-Zn.skf"
O-O = "O-O.skf"
}
MaxAngularMomentum = {
Zn = "d"
O = "p"
}
Charge = 0.0
SpinPolarisation = {}
Filling = Fermi {
Temperature = 1.000000000000000E-008
IndependentKFilling = No
}
MaxSCCIterations = 1000
ElectricField = {}
OrbitalResolvedSCC = No
ReadInitialCharges = No
Eigensolver = DivideAndConquer {}
OldSKInterpolation = No
OrbitalPotential = {}
Dispersion = {}
}
ParserOptions = {
ParserVersion = 3
WriteHSDInput = Yes
WriteXMLInput = No
StopAfterParsing = No
}
Options = {
WriteEigenvectors = No
WriteAutotestTag = No
WriteDetailedXML = No
WriteResultsTag = Yes
WriteDetailedOut = Yes
WriteBandOut = Yes
AtomResolvedEnergies = Yes
RestartFrequency = 20
RandomSeed = 0
}
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.zfn.uni-bremen.de/pipermail/dftb-plus-user/attachments/20090724/9a4ffb8e/attachment.htm>
More information about the DFTB-Plus-User
mailing list