HI, Marcos,
Thank you for your suggestion. I tried:
%block DM.InitSpin
7 + 30. 30.
%endblock DM.InitSpin
During optimization the spin is always around 0.64 0.37 1.28,
the direction seems to be fixed at 30,30. In the end the energy diffrence
between former 90,0 is 0.0001eV higher. So seems it will not find
a minimum itself, and we have to try different initial spin direction
and compare the energy?
Menghao Wu
# Output options
WriteCoorStep
WriteMullikenPop 1
# Species and atoms
NumberOfSpecies 3
NumberOfAtoms 13
%block ChemicalSpeciesLabel
1 6 C
2 25 Mn
3 1 H
%endblock ChemicalSpeciesLabel
# Basis
PAO.EnergyShift 50 meV
PAO.BasisSize DZP
LatticeConstant 1.00 Ang
%block LatticeVectors
16.0000000000000000 .0000000000000000 .0000000000000000
.0000000000000000 16.0000000000000 .0000000000000000
.0000000000000000 .0000000000000000 3.3700000000000000
%endblock LatticeVectors
KgridCutoff 15. Ang
%block BandLines
1 0.00000 0.000000 0.000000 \Gamma
30 0.00000 0.000000 0.500000 X
%endblock BandLines
%block DM.InitSpin
7 + 30. 30.
%endblock DM.InitSpin
xc.functional LDA # Exchange-correlation functional
xc.authors CA # Exchange-correlation version
SpinPolarized true # Logical parameters are: yes or no
NonCollinearSpin true
MeshCutoff 150. Ry # Mesh cutoff. real space mesh
# SCF options
MaxSCFIterations 500 # Maximum number of SCF iter
DM.MixingWeight 0.1 # New DM amount for next SCF cycle
DM.Tolerance 1.d-3 # Tolerance in maximum difference
# between input and output DM
DM.UseSaveDM true # to use continuation files
DM.NumberPulay 3
SolutionMethod Diagon # OrderN or Diagon
ElectronicTemperature 25 K # Temp. for Fermi smearing
Diag.ParallelOverK no
# MD options
MD.TypeOfRun cg # Type of dynamics:
MD.NumCGsteps 200 # Number of CG steps for
# coordinate optimization
MD.MaxCGDispl 0.2 Ang # Maximum atomic displacement
# in one CG step (Bohr)
MD.iMaxForceTol 0.05 eV/Ang # Tolerance in the maximum
# atomic force (Ry/Bohr)
# Atomic coordinates
AtomicCoordinatesFormat Ang
%block AtomicCoordinatesAndAtomicSpecies
5.984342432 3.596261950 1.908996404 1
4.723873840 4.285144890 1.902371284 1
4.684148487 5.724854135 1.895339145 1
7.203280667 4.354484130 1.901971703 1
7.163821550 5.794305040 1.894546280 1
5.903464827 6.482998694 1.943705236 1
5.944012009 5.038898651 0.232220326 2
6.014619027 2.505118673 1.903154742 3
8.166882755 3.841390919 1.895707998 3
3.791368128 3.717187433 1.896437188 3
5.873242219 7.573943296 1.950529321 3
8.095885216 6.362797445 1.893455905 3
3.720974178 6.238472443 1.894629706 3
%endblock AtomicCoordinatesAndAtomicSpecies
2011/1/26 Marcos Veríssimo Alves <[email protected]>:
> Menghao,
> Whereas I am not sure what exactly is the setup for the angles theta and phi
> (that is, if theta is measured from the z axis or from the x-y plane), it
> could be that theta=90 degrees is already a possible energy minimum. In
> particular, if theta is measured from the x-y plane, then you have your
> previous starting point, which is the spin aligned along the z axis.
> However, in spin-polarized calculations many different minima can exist, and
> it's up to you to try different configurations to find the global minimum.
> If theta=90 deg is already an energy minimum for your system, then nothing
> will change. What happens if you set a different value for theta, say, a
> weird one like 26.73 degrees?
> Marcos
>
> 2011/1/26 吴梦昊 <[email protected]>
>>
>> Hi,
>> Thank you for your advice. When I use %block DM.InitSpin to set
>> a initial spin,
>> I add:
>> %block DM.InitSpin
>> 7 + 90. 0.
>> %endblock DM.InitSpin
>>
>>
>>
>> it seems that the direction will be fixed at angle 90 during
>> optimization---
>> the direction will not optimize itself. So in noncollinear calculation, we
>> have to compare energy of all sort of spin angle, and we cannot let
>> the spin direction optimized itself? Thank you and look forward to
>> advice
>> Sincerely
>> Menghao Wu
>>
>>
>>
>> 2011/1/23 Lucas Fernandez Seivane <[email protected]>:
>> > Hi
>> >
>> > Look for %block DM.InitSpin in the manual. You have to set magnetic
>> > seeds with that options.
>> >
>> > regards
>> > Lucas
>> > 2011/1/24 吴梦昊 <[email protected]>:
>> >> Dear siesta users and developers：
>> >> I found siesta results of noncollinear calculation always leads to
>> >> magnetization
>> >> only in z-direction-----even I switched the directin of magnet! Has
>> >> anyone else found
>> >> something wrong with noncollinear calculation? Or something
>> >> wrong with my operation?
>> >> I put fdf file below.
>> >> I appreciate your suggestions about this
>> >> problem.
>> >>
>> >>
>> >> With best regards
>> >> Menghao Wu
>> >>
>> >>
>> >> # Output options
>> >>
>> >> WriteCoorStep
>> >> WriteMullikenPop 1
>> >>
>> >> # Species and atoms
>> >>
>> >> NumberOfSpecies 3
>> >> NumberOfAtoms 13
>> >> %block ChemicalSpeciesLabel
>> >> 1 6 C
>> >> 2 25 Mn
>> >> 3 1 H
>> >> %endblock ChemicalSpeciesLabel
>> >>
>> >> # Basis
>> >>
>> >> PAO.EnergyShift 50 meV
>> >> PAO.BasisSize DZP
>> >>
>> >>
>> >> LatticeConstant 1.00 Ang
>> >>
>> >> %block LatticeVectors
>> >> 16.0000000000000000 .0000000000000000 .0000000000000000
>> >> .0000000000000000 16.0000000000000 .0000000000000000
>> >> .0000000000000000 .0000000000000000 3.3700000000000000
>> >> %endblock LatticeVectors
>> >>
>> >> KgridCutoff 15. Ang
>> >>
>> >> %block BandLines
>> >> 1 0.00000 0.000000 0.000000 \Gamma
>> >> 30 0.00000 0.000000 0.500000 X
>> >> %endblock BandLines
>> >>
>> >> xc.functional LDA # Exchange-correlation functional
>> >> xc.authors CA # Exchange-correlation version
>> >>
>> >> SpinPolarized true # Logical parameters are: yes or
>> >> no
>> >> NonCollinearSpin true
>> >> MeshCutoff 150. Ry # Mesh cutoff. real space mesh
>> >>
>> >> # SCF options
>> >> MaxSCFIterations 500 # Maximum number of SCF iter
>> >> DM.MixingWeight 0.1 # New DM amount for next SCF cycle
>> >> DM.Tolerance 1.d-3 # Tolerance in maximum difference
>> >> # between input and output DM
>> >> DM.UseSaveDM true # to use continuation files
>> >> DM.NumberPulay 3
>> >>
>> >> SolutionMethod Diagon # OrderN or Diagon
>> >> ElectronicTemperature 25 K # Temp. for Fermi smearing
>> >> Diag.ParallelOverK no
>> >> # MD options
>> >>
>> >> MD.TypeOfRun cg # Type of dynamics:
>> >> MD.NumCGsteps 200 # Number of CG steps for
>> >> # coordinate optimization
>> >> MD.MaxCGDispl 0.2 Ang # Maximum atomic displacement
>> >> # in one CG step (Bohr)
>> >> MD.iMaxForceTol 0.05 eV/Ang # Tolerance in the maximum
>> >> # atomic force (Ry/Bohr)
>> >>
>> >> # Atomic coordinates
>> >> AtomicCoordinatesFormat Ang
>> >> %block AtomicCoordinatesAndAtomicSpecies
>> >> 5.984342432 3.596261950 1.908996404 1
>> >> 4.723873840 4.285144890 1.902371284 1
>> >> 4.684148487 5.724854135 1.895339145 1
>> >> 7.203280667 4.354484130 1.901971703 1
>> >> 7.163821550 5.794305040 1.894546280 1
>> >> 5.903464827 6.482998694 1.943705236 1
>> >> 5.944012009 5.038898651 0.232220326 2
>> >> 6.014619027 2.505118673 1.903154742 3
>> >> 8.166882755 3.841390919 1.895707998 3
>> >> 3.791368128 3.717187433 1.896437188 3
>> >> 5.873242219 7.573943296 1.950529321 3
>> >> 8.095885216 6.362797445 1.893455905 3
>> >> 3.720974178 6.238472443 1.894629706 3
>> >> %endblock AtomicCoordinatesAndAtomicSpecies
>> >>
>> >>
>> >> The result always has net M in z-direction:
>> >> Total Total 21.13603 2.63240 0.000 0.000 1.491
>> >>
>> >>
>> >> even after I switch the direction of lattice to y-axis
>> >> %block LatticeVectors
>> >> 16.0000000000000000 .0000000000000000 .0000000000000000
>> >> .0000000000000000 3.3700000000000 .0000000000000000
>> >> .0000000000000000 .0000000000000000 16.00000000000000
>> >> %endblock LatticeVectors
>> >>
>> >> # Atomic coordinates
>> >> AtomicCoordinatesFormat Ang
>> >> %block AtomicCoordinatesAndAtomicSpecies
>> >> 5.984342432 1.908996404 3.596261950 1
>> >> 4.723873840 1.902371284 4.285144890 1
>> >> 4.684148487 1.895339145 5.724854135 1
>> >> 7.203280667 1.901971703 4.354484130 1
>> >> 7.163821550 1.894546280 5.794305040 1
>> >> 5.903464827 1.943705236 6.482998694 1
>> >> 5.944012009 0.232220326 5.038898651 2
>> >> 6.014619027 1.903154742 2.505118673 3
>> >> 8.166882755 1.895707998 3.841390919 3
>> >> 3.791368128 1.896437188 3.717187433 3
>> >> 5.873242219 1.950529321 7.573943296 3
>> >> 8.095885216 1.893455905 6.362797445 3
>> >> 3.720974178 1.894629706 6.238472443 3
>> >> %endblock AtomicCoordinatesAndAtomicSpecies
>> >>
>> >
>> >
>> >
>> > --
>> > Lucas Fernández Seivane
>> > Ph. D. Candidate
>> > Universidad de Oviedo - CINN
>> > http://condmat.uniovi.es/~quevedin
>> > http://www.researcherid.com/rid/A-8610-2008
>> >
>
>
