Re: [Wien] 'LAPW2: semicore band-ranges too large error' for spin-orbit calculation

2018-04-23 Thread Peter Blaha 

Remove the RLOs from As. There are no semicore As-p states.

--

  P.Blaha
--
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300 FAX: +43-1-58801-165982
Email: bl...@theochem.tuwien.ac.atWIEN2k: http://www.wien2k.at
WWW:   http://www.imc.tuwien.ac.at/TC_Blaha
--
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] O2 in triplet state?

2018-04-23 Thread chin Sabsu 
 Dear Sir,
I am thankful for the confirmation of the state of O2 molecule.

I am tried to reproduce some results for oxygen deficient system but I see from 
my data that my system is not stable.

I started from the given lattice parameters, exact functionals,(GGA, as 
suggested in the paper) rmt k-mesh etc.

The authors did not mention anything about how they have calculated the 
formation energy and atomization energy. In my previous post your reply with 
some notes, I followed the same data to simulate ground state energy of O2 and 
O but still am not getting the reasonable results.

As I used data for O2 from FAQ and also tried according to the information 
given in the literature paper.

I see there should not be any issue in calculating the O2 energy.


My doubt is somewhere in the calculation of O (-sp) with below data:



Title
F   LATTICE,NONEQUIV.ATOMS:  1
MODE OF CALC=RELA unit=bohr
 28.345900 28.345900 28.345900 90.00 90.00 90.00
ATOM   1: X=0. Y=0. Z=0.
  MULT= 1  ISPLIT= 2
O  NPT=  781  R0=0.0001 RMT= 1.65000 Z:  8.000
LOCAL ROT MATRIX:    1.000 0.000 0.000
 0.000 1.000 0.000
 0.000 0.000 1.000
  48  NUMBER OF SYMMETRY OPERATIONS



and
O
He 3  
2,-1,1.0  N
2,-1,1.0  N
2, 1,1.0  N
2, 1,1.0  N
2,-2,2.0  N
2,-2,0.0  N

 END of input (instgen_lapw)


below are data from O2 and O-atom with GGA

 
| O_atom_rmt_1.75_rkmax_7 | -149.86322972 |
| O_atom_rmt_1.1_rmkax_5.5 | -150.0869798 |
| O2_mol_bondlength_1.21_rkmax_5.5 | -300.1077091 |
| [O2_mol_1.21]\2 | -150.05385455 |
| O3_mol_1.21 | -450.16156365 |
| O2_mol_bondlength_1.219_rkmax_4.6 | -299.95534741 |
| [O2_mol_1.219]\2 | -149.977673705 |
| O3_mol_1.219 | -449.933021115 |

 

In his previous post in response of my query, Prof. Alay advice about 
calculating the ground state energy of O-atom by considering O atom cell as 
orthorhombic to avoid any issue occurring from the occupancy of P-states of 
O-atom. His statement is quoted below:


"Computing the atomic energies of atoms like N and P in an FCC cell is ok, 
however for O atom the high symmetry of the FCC cell results in 1/3  
occupancies (for the 4th p electron of O) in the spin down case. Only using a 
lower symmetry cell (orthorhombic) for O atom eliminates this issue." 
Could you please advise me whether my above data looks good or not.

If I have to follow the suggestion advanced by Prof. Alay, then how to make an 
Orthorhombic cell for O-atom?

I have done three calculations for three materials but I am not getting the 
atomization and formation energy of O2 while the author reported similar 
statements in his papers.


Please help me to simulate the ground state energy of O2 and O taking care of 
occupancy of P orbitals.

Please let me know what additional information I can provide.


thank you very much for a big help.

 Chin S.



 
On Monday, 23 April, 2018, 10:32:22 AM IST, Peter Blaha 
 wrote:  
 
 This is the configuration for a spin-polarized O atom.

And yes, this starting configuration will lead to the triplet state of 
O2 (when you perform spin-polarized calculations.)

Am 22.04.2018 um 08:16 schrieb chin Sabsu:
> Dear Users,
> 
> 
> Could you please advice me whether below *.inst form O2 in triplet 
> state? three e- in dn and one e- in up state?
> 
> 
> O
> He 3
> 2,-1,1.0  N
> 2,-1,1.0  N
> 2, 1,1.0  N
> 2, 1,1.0  N
> 2,-2,2.0  N
> 2,-2,0.0  N
> 
>  END of input (instgen_lapw)
> 
> 
> Thanks and best regards,
> 
> Chin S.
> 
> 
> ___
> Wien mailing list
> Wien@zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:  
> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
> 

-- 
--
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300            FAX: +43-1-58801-165982
Email: bl...@theochem.tuwien.ac.at    WIEN2k: http://www.wien2k.at
WWW: 
http://www.imc.tuwien.ac.at/tc_blaha-
 

___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
  ___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] AFM calculations

2018-04-23 Thread Lawal Mohammed 
Dear Prof. Peter,

Thanks a lot for the explanation. I have another question in this regard.
Please why do we have to do spin-polarized calculation for O2 (or for 
non-closed shell elements) as mentioned in the FAQ page under  Calculations of 
cohesive or formation energies?

Thanks very much for your time.

Kind regards.

Lawal 




 

On Monday, April 23, 2018, 1:10:32 PM GMT+8, Peter Blaha 
 wrote:  
 
 Without SO: You can either use runafm (if you can figure out the correct 
symmetry operation which transforms spin-up into spin-dn atoms) OR 
runsp_lapw (takes twice as much cpu time, but is "simpler").

With SO you must use runsp. runafm does not support spin-orbit.


Am 20.04.2018 um 13:24 schrieb Lawal Mohammed:
> Dear respected Developers and Users,
> 
> I am trying to understand how to do AFM calculations with SO. I read 
> section 4.5.4 of the UG and check some threads in the wien list.
> 
> The way I understand it, one can choose either of the two options.
> 
> 1-run runsp_lapw and then do scf with SO
> 
> OR
> 
> 2-runafm_lapw and then do SO
> 
> I may probably be wrong. I want to test run with Fe2O3.
> 
> Any advice is highly appreciated.
> 
> Regards
> 
> */Lawal
> /*
> 
> 
> 
> 
> 
> ___
> Wien mailing list
> Wien@zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:  
> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
> 

-- 
--
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300            FAX: +43-1-58801-165982
Email: bl...@theochem.tuwien.ac.at    WIEN2k: http://www.wien2k.at
WWW: 
http://www.imc.tuwien.ac.at/tc_blaha-
 

___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
  ___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] AFM calculations

2018-04-23 Thread Peter Blaha 
Because for cohesive energies you need the ground state energies of the 
atoms/molecules involved.


And the ground state of O2 has 2 unpaired electrons, i.e. it requeres a 
spin polarized calculation.


Test it yourself: Do a spin and non-spinpolarized calculationfor O2 (in 
2 directories) with otherwise identical parameters. Which energy is lower ?


Am 24.04.2018 um 05:19 schrieb Lawal Mohammed:

Dear Prof. Peter,

Thanks a lot for the explanation. I have another question in this regard.
Please why do we have to do spin-polarized calculation for O2 *(or for 
non-closed shell elements)* as mentioned in the FAQ page under 
*Calculations of cohesive or formation energies 
*?


Thanks very much for your time.

Kind regards.

*/Lawal
/*





On Monday, April 23, 2018, 1:10:32 PM GMT+8, Peter Blaha 
 wrote:



Without SO: You can either use runafm (if you can figure out the correct
symmetry operation which transforms spin-up into spin-dn atoms) OR
runsp_lapw (takes twice as much cpu time, but is "simpler").

With SO you must use runsp. runafm does not support spin-orbit.


Am 20.04.2018 um 13:24 schrieb Lawal Mohammed:
 > Dear respected Developers and Users,
 >
 > I am trying to understand how to do AFM calculations with SO. I read
 > section 4.5.4 of the UG and check some threads in the wien list.
 >
 > The way I understand it, one can choose either of the two options.
 >
 > 1-run runsp_lapw and then do scf with SO
 >
 > OR
 >
 > 2-runafm_lapw and then do SO
 >
 > I may probably be wrong. I want to test run with Fe2O3.
 >
 > Any advice is highly appreciated.
 >
 > Regards
 >
 > */Lawal

 > /*
 >
 >
 >
 >
 >
 > ___
 > Wien mailing list
 > Wien@zeus.theochem.tuwien.ac.at 
 > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
 > SEARCH the MAILING-LIST at: 
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

 >

--
--
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300            FAX: +43-1-58801-165982
Email: bl...@theochem.tuwien.ac.at   
   WIEN2k: http://www.wien2k.at

WWW:
http://www.imc.tuwien.ac.at/tc_blaha- 



___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at 
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at: 
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html




___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html



--
--
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300 FAX: +43-1-58801-165982
Email: bl...@theochem.tuwien.ac.atWIEN2k: http://www.wien2k.at
WWW: 
http://www.imc.tuwien.ac.at/tc_blaha- 


___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] O2 in triplet state?

2018-04-23 Thread Peter Blaha 

I do not find much sense in the data you sent ???

When you want to calculate the O2 binding energies, the  RMTs and RKmax 
values of O and O2 need to be IDENTICAL. And yes, break a bit the 
symmetry (use slightly different a,b,c) to get the lowest energy for 
atoms according to Hunds rule.

-

If you want to calculate cohesive/formation energies, you must use 
identical RMTs and "equivalent RKmax".


So for a Me_xO_y compound, you would first optimize the structure. 
Optimization in genergal involves lattice parameter (at least voluem) 
AND internal positions (forces).
Then you need to calculate the O2 energy and the Me (typical this is the 
metallic phase of this element, like bcc Fe or fcc Al,...).


To do so, you need to use a small RMT for O2 because of the samll bond 
length. Optimize the O2-distance eg. with 1 k-point and RMT 1.2 and 
RKMAX 5.5.


Then repeat your Me-O compound (in the relaxed minimum structure) with 
this same O-RMT (1.2) and also RKmax=5.5 (but a good k-mesh).


Finally do the Me phase (with the same Me-RMT as in your compound) and a 
RKmax= 5.5 / 1.2 * RMT(Me) !


Form the energy difference to find the cohesive energy.

Now repeat the O2, MeO and Me calculations, where you increase RKmax 
from 5.5 to eg. 6.0. Form again the difference. If it is stable, you are 
done, if not, increase RKMax again (6.6 or 7, depends also on your Me 
and compound) until you get a stable cohesive energy.


PS: When increasing RKmax, check the forces on the atoms if they remain 
small (usually they do unless you started much too small).





Am 23.04.2018 um 12:14 schrieb chin Sabsu:

Dear Sir,
I am thankful for the confirmation of the state of O2 molecule.

I am tried to reproduce some results for oxygen deficient system but I 
see from my data that my system is not stable.


I started from the given lattice parameters, exact functionals,(GGA, as 
suggested in the paper) rmt k-mesh etc.


The authors did not mention anything about how they have calculated the 
formation energy and atomization energy. In my previous post your reply 
with some notes, I followed the same data to simulate ground state 
energy of O2 and O but still am not getting the reasonable results.


As I used data for O2 from FAQ and also tried according to the 
information given in the literature paper.


I see there should not be any issue in calculating the O2 energy.


My doubt is somewhere in the calculation of O (-sp) with below data:



Title
F   LATTICE,NONEQUIV.ATOMS:  1
MODE OF CALC=RELA unit=bohr
  28.345900 28.345900 28.345900 90.00 90.00 90.00
ATOM   1: X=0. Y=0. Z=0.
   MULT= 1  ISPLIT= 2
O  NPT=  781  R0=0.0001 RMT= 1.65000 Z:  8.000
LOCAL ROT MATRIX:    1.000 0.000 0.000
  0.000 1.000 0.000
  0.000 0.000 1.000
   48  NUMBER OF SYMMETRY OPERATIONS



and
O
He 3
2,-1,1.0  N
2,-1,1.0  N
2, 1,1.0  N
2, 1,1.0  N
2,-2,2.0  N
2,-2,0.0  N

 END of input (instgen_lapw)


below are data from O2 and O-atom with GGA

O_atom_rmt_1.75_rkmax_7 -149.86322972
O_atom_rmt_1.1_rmkax_5.5-150.0869798
O2_mol_bondlength_1.21_rkmax_5.5-300.1077091
[O2_mol_1.21]\2 -150.05385455
O3_mol_1.21 -450.16156365
O2_mol_bondlength_1.219_rkmax_4.6   -299.95534741
[O2_mol_1.219]\2-149.977673705
O3_mol_1.219-449.933021115



In his previous post in response of my query, Prof. Alay advice about 
calculating the ground state energy of O-atom by considering O atom cell 
as orthorhombic to avoid any issue occurring from the occupancy of 
P-states of O-atom. His statement is quoted below:



"Computing the atomic energies of atoms like N and P in an FCC cell is 
ok, however for O atom the high symmetry of the FCC cell results in 1/3  
occupancies (for the 4th p electron of O) in the spin down case. Only using


a lower symmetry cell (orthorhombic) for O atom eliminates this issue."


Could you please advise me whether my above data looks good or not.

If I have to follow the suggestion advanced by Prof. Alay, then how to 
make an Orthorhombic cell for O-atom?


I have done three calculations for three materials but I am not getting 
the atomization and formation energy of O2 while the author reported 
similar statements in his papers.



Please help me to simulate the ground state energy of O2 and O taking 
care of occupancy of P orbitals.


Please let me know what additional information I can provide.


thank you very much for a big help.

Chin S.




On Monday, 23 April, 2018, 10:32:22 AM IST, Peter Blaha 
 wrote:



This is the configuration for a spin-polarized O atom.

And yes, this starting configuration will lead to the triplet state of
O2 (when you perform spin-polarized calculations.)

Am 22.04.2018 um 08:16 schrieb chin Sabsu:
 > Dear Users,
 >
 >
 > Could you please advice me whether below

Re: [Wien] O2 in triplet state?

2018-04-23 Thread chin Sabsu 
Thanks Sir for the detailed reply,

Let me do a set of calculation according to your suggestions.
Get back to you soon with results.

Regards
Chin S.
 

On Tuesday, 24 April, 2018, 11:22:55 AM IST, Peter Blaha 
 wrote:  
 
 I do not find much sense in the data you sent ???

When you want to calculate the O2 binding energies, the  RMTs and RKmax 
values of O and O2 need to be IDENTICAL. And yes, break a bit the 
symmetry (use slightly different a,b,c) to get the lowest energy for 
atoms according to Hunds rule.
-

If you want to calculate cohesive/formation energies, you must use 
identical RMTs and "equivalent RKmax".

So for a Me_xO_y compound, you would first optimize the structure. 
Optimization in genergal involves lattice parameter (at least voluem) 
AND internal positions (forces).
Then you need to calculate the O2 energy and the Me (typical this is the 
metallic phase of this element, like bcc Fe or fcc Al,...).

To do so, you need to use a small RMT for O2 because of the samll bond 
length. Optimize the O2-distance eg. with 1 k-point and RMT 1.2 and 
RKMAX 5.5.

Then repeat your Me-O compound (in the relaxed minimum structure) with 
this same O-RMT (1.2) and also RKmax=5.5 (but a good k-mesh).

Finally do the Me phase (with the same Me-RMT as in your compound) and a 
RKmax= 5.5 / 1.2 * RMT(Me) !

Form the energy difference to find the cohesive energy.

Now repeat the O2, MeO and Me calculations, where you increase RKmax 
from 5.5 to eg. 6.0. Form again the difference. If it is stable, you are 
done, if not, increase RKMax again (6.6 or 7, depends also on your Me 
and compound) until you get a stable cohesive energy.

PS: When increasing RKmax, check the forces on the atoms if they remain 
small (usually they do unless you started much too small).




Am 23.04.2018 um 12:14 schrieb chin Sabsu:
> Dear Sir,
> I am thankful for the confirmation of the state of O2 molecule.
> 
> I am tried to reproduce some results for oxygen deficient system but I 
> see from my data that my system is not stable.
> 
> I started from the given lattice parameters, exact functionals,(GGA, as 
> suggested in the paper) rmt k-mesh etc.
> 
> The authors did not mention anything about how they have calculated the 
> formation energy and atomization energy. In my previous post your reply 
> with some notes, I followed the same data to simulate ground state 
> energy of O2 and O but still am not getting the reasonable results.
> 
> As I used data for O2 from FAQ and also tried according to the 
> information given in the literature paper.
> 
> I see there should not be any issue in calculating the O2 energy.
> 
> 
> My doubt is somewhere in the calculation of O (-sp) with below data:
> 
> 
> 
> Title
> F   LATTICE,NONEQUIV.ATOMS:  1
> MODE OF CALC=RELA unit=bohr
>   28.345900 28.345900 28.345900 90.00 90.00 90.00
> ATOM   1: X=0. Y=0. Z=0.
>    MULT= 1  ISPLIT= 2
> O  NPT=  781  R0=0.0001 RMT= 1.65000 Z:  8.000
> LOCAL ROT MATRIX:    1.000 0.000 0.000
>   0.000 1.000 0.000
>   0.000 0.000 1.000
>    48  NUMBER OF SYMMETRY OPERATIONS
> 
> 
> 
> and
> O
> He 3
> 2,-1,1.0  N
> 2,-1,1.0  N
> 2, 1,1.0  N
> 2, 1,1.0  N
> 2,-2,2.0  N
> 2,-2,0.0  N
> 
>  END of input (instgen_lapw)
> 
> 
> below are data from O2 and O-atom with GGA
> 
> O_atom_rmt_1.75_rkmax_7     -149.86322972
> O_atom_rmt_1.1_rmkax_5.5     -150.0869798
> O2_mol_bondlength_1.21_rkmax_5.5     -300.1077091
> [O2_mol_1.21]\2     -150.05385455
> O3_mol_1.21     -450.16156365
> O2_mol_bondlength_1.219_rkmax_4.6     -299.95534741
> [O2_mol_1.219]\2     -149.977673705
> O3_mol_1.219     -449.933021115
> 
> 
> 
> In his previous post in response of my query, Prof. Alay advice about 
> calculating the ground state energy of O-atom by considering O atom cell 
> as orthorhombic to avoid any issue occurring from the occupancy of 
> P-states of O-atom. His statement is quoted below:
> 
> 
> "Computing the atomic energies of atoms like N and P in an FCC cell is 
> ok, however for O atom the high symmetry of the FCC cell results in 1/3  
> occupancies (for the 4th p electron of O) in the spin down case. Only using
> 
> a lower symmetry cell (orthorhombic) for O atom eliminates this issue."
> 
> 
> Could you please advise me whether my above data looks good or not.
> 
> If I have to follow the suggestion advanced by Prof. Alay, then how to 
> make an Orthorhombic cell for O-atom?
> 
> I have done three calculations for three materials but I am not getting 
> the atomization and formation energy of O2 while the author reported 
> similar statements in his papers.
> 
> 
> Please help me to simulate the ground state energy of O2 and O taking 
> care of occupancy of P orbitals.
> 
> Please let me know what additional information I can provide.
> 
> 
> thank you very much for