This thread is locked.Only browsing is available.
Top Page > Browsing
Fe/MgO/Fe calculation
Date: 2012/12/01 00:55
Name: Kamaram Munira   <kmunira@mint.ua.edu>

Dear Prof. Ozaki,

I am trying to reproduce the results of the Fe/MgO/Fe system that you had in Phys. Rev. B 81, 035116 (2010).

I am having trouble converging the channel system. Here is my input file:

#
# Atoms
#

Atoms.Number 8
Atoms.SpeciesAndCoordinates.Unit Ang # Ang|AU
<Atoms.SpeciesAndCoordinates
1 Mg 0.0000 0.0000 6.459 4.00 4.00
2 O 1.4330 1.4330 6.459 3.00 3.00
3 O 0.0000 0.0000 8.486 3.00 3.00
4 Mg 1.4330 1.4330 8.486 4.00 4.00
5 Mg 0.0000 0.0000 10.513 4.00 4.00
6 O 1.4330 1.4330 10.513 3.00 3.00
7 O 0.0000 0.0000 12.54 3.00 3.00
8 Mg 1.4330 1.4330 12.54 4.00 4.00
Atoms.SpeciesAndCoordinates>

#
# Lead-Left
#

LeftLeadAtoms.Number 4
<LeftLeadAtoms.SpeciesAndCoordinates # Unit=Ang.
1 Fe 0.0000 0.0000 0.0000 8.00 6.00
2 Fe 1.4330 1.4330 1.4330 8.00 6.00
3 Fe 0.0000 0.0000 2.8660 8.00 6.00
4 Fe 1.4330 1.4330 4.2990 8.00 6.00
LeftLeadAtoms.SpeciesAndCoordinates>


#
# Lead-Right
#

RightLeadAtoms.Number 4
<RightLeadAtoms.SpeciesAndCoordinates # Unit=Ang.
1 Fe 0.0000 0.0000 14.7 8.00 6.00
2 Fe 1.4330 1.4330 16.133 8.00 6.00
3 Fe 0.0000 0.0000 17.566 8.00 6.00
4 Fe 1.4330 1.4330 18.999 8.00 6.00
RightLeadAtoms.SpeciesAndCoordinates>



#
# SCF or Electronic System
#

scf.XcType GGA-PBE # LDA|LSDA-CA|LSDA-PW|GGA-PBE
scf.SpinPolarization on # On|Off|NC
scf.ElectronicTemperature 1000.0 # default=300 (K)
scf.energycutoff 350.0 # default=150 (Ry)
scf.maxIter 200 # default=40
scf.EigenvalueSolver NEGF # DC|GDC|Cluster|Band
scf.lapack.dste dstevx # dstegr|dstedc|dstevx, default=dstegr
scf.Kgrid 1 2 2 # means n1 x n2 x n3
scf.Mixing.Type Rmm-Diisk # Simple|Rmm-Diis|Gr-Pulay|Kerker|Rmm-Diisk
scf.Init.Mixing.Weight 0.03 # default=0.30
scf.Min.Mixing.Weight 0.001 # default=0.001
scf.Max.Mixing.Weight 0.005 # default=0.40
scf.Mixing.History 30 # default=5
scf.Mixing.StartPulay 3001 # default
scf.Kerker.factor 0.05 # default=1.0
scf.criterion 1.0e-4 # default=1.0e-6 (Hartre
scf.restart on


The NormRD is hovering around 0.147:
<DFT> NormRD = 0.147153674869 Criterion = 0.000100000000
<DFT> NormRD = 0.147160639848 Criterion = 0.000100000000
<DFT> NormRD = 0.147167640155 Criterion = 0.000100000000
<DFT> NormRD = 0.147174675029 Criterion = 0.000100000000
<DFT> NormRD = 0.147181743735 Criterion = 0.000100000000
<DFT> NormRD = 0.147188845559 Criterion = 0.000100000000
<DFT> NormRD = 0.147195979809 Criterion = 0.000100000000
<DFT> NormRD = 0.147203145813 Criterion = 0.000100000000
<DFT> NormRD = 0.147210342919 Criterion = 0.000100000000
<DFT> NormRD = 0.147217570494 Criterion = 0.000100000000
<DFT> NormRD = 0.147224827922 Criterion = 0.000100000000
<DFT> NormRD = 0.147232114608 Criterion = 0.000100000000
<DFT> NormRD = 0.147239429970 Criterion = 0.000100000000
<DFT> NormRD = 0.147246773445 Criterion = 0.000100000000
<DFT> NormRD = 0.147254144485 Criterion = 0.000100000000
<DFT> NormRD = 0.147261542556 Criterion = 0.000100000000
<DFT> NormRD = 0.147268967141 Criterion = 0.000100000000
<DFT> NormRD = 0.147276417735 Criterion = 0.000100000000
<DFT> NormRD = 0.147283893848 Criterion = 0.000100000000
<DFT> NormRD = 0.147291395003 Criterion = 0.000100000000
<DFT> NormRD = 0.147298920736 Criterion = 0.000100000000
<DFT> NormRD = 0.147306470595 Criterion = 0.000100000000
<DFT> NormRD = 0.147314044138 Criterion = 0.000100000000
<DFT> NormRD = 0.147321640939 Criterion = 0.000100000000
<DFT> NormRD = 0.147329260580 Criterion = 0.000100000000
<DFT> NormRD = 0.147336902654 Criterion = 0.000100000000
<DFT> NormRD = 0.147344566765 Criterion = 0.000100000000
<DFT> NormRD = 0.147352252527 Criterion = 0.000100000000

My questions are:
1) We can't do for NormRD thats in the order of 1e-6. What range should we try to converge to?

2) And how?

Thanks in advance for your help.
-Kamaram

e
Page: [1]