[Wien] Mismatch in the Fermi level between Wien2K open core calculations and VASP

[Anup Shakya]

Dear All,

 I have been trying to perform calculations for PrBi by keeping the
Pr 4f electrons in core. For this, I have modified the case.inc by
increasing the number of orbitals from 14 to 15 i.e, adding one line 4, 3,
3 to keep the three Pr 4f electrons in the core. I also added a shift of
0.7 in the case.inc file. After that in case.in1 file I kept the Pr 4f
electrons at a low energy of -2.00 with no search and then decreased the
number of electron from 28 to 25 in case.in2 file. I have then performed
GGA as well as GGA+SOC (spin orbit coupling) calculations. There are
already some report for DFT calculations of PrBi using VASP where they have
also performed the calculations by keeping the Pr 4f electrons in the core.
The calculations performed by me are in very good agreement with the paper [PRB
99, 245131, 2019] and also *Commun Phys* 1, 71 (2018) except that there is
a shift of the Fermi level in my calculations by around ~ 1eV. (I could not
attach the figures in this email but if any more information is needed
please let me know).  Could anyone please explain the reason behind the
mismatch in the Fermi level in the calculations performed by me and the
reported results.  Is there any way to overcome this problem? I do not want
to perform LDA+U or GGA+U calculations as I have already done it and it
does not match with the ARPES data.  I would be grateful if anyone could
provide some suggestions.
Sincerely,
Anup Pradhan Sakhya (Ph.D.)
Visiting Post-Doctoral Fellow
DCMP, TIFR, Mumbai
___
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] "LAPW2: semicore band-ranges too large" error in LSDA+SO

[Dibyendu DEY]

Dear Prof. Blaha,

Thanks for your response.
I have tried the procedure you mentioned in your earlier mail, but I am
getting the same error if I choose the magnetization axis 1 0 0.

With regards,
Dibyendu

On Tue, Jan 28, 2020 at 1:55 AM Peter Blaha 
wrote:

> I suggest you move the saved lsda calculation (no so) to a new
> directory, restore it and do the init_so with the correct magnetization
> direction again.
>
>
> On 1/27/20 9:37 PM, Dibyendu DEY wrote:
> > Dear Wien2k users,
> >
> > Recently, I performed DFT calculations on VI3 monolayer in Wien2k-18.2.
> >
> > With spin-orbit coupling, if I choose the magnetization axis along 0 0 1
> > (easy axis), LSDA+SO, and LSDA+SO+U calculations run perfectly without
> > any error, and I get desirable values of the magnetic (1.87 \mu_B) and
> > orbital moments (-1.08 \mu_B) at the V site. However, if I choose the
> > magnetization axis to be 1 0 0, LSDA+SO calculations stop with the
> > following error at the beginning.
> >
> > LAPW2: semicore band-ranges too large
> > cp: cannot stat ‘.in.tmp’: No such file or directory
> > set: No match.
> >
> > We found large QTL-B values cause this problem. I changed the energy
> > parameters for the respective atom and L value in case.in1 file, but I
> > could not able to resolve the issue.
> >
> > Original case.struct and case.in1 files have been attached.
> > In calculations, RKmax value was set to 7.0.
> >
> > Any suggestions in this regard would be appreciated.
> >
> > With best regards,
> > Dibyendu
> > --
> > *Dibyendu Dey
> > Postdoctoral Research Scholar
> > **Department of Physics, Arizona State University *
> > *Tempe, AZ 85287, United States*
> > *Ph: +1-480-427-9970**| Personal Email: dibyendu@gmail.com
> > *
> >
> > ___
> > 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
> >
>
> --
>
>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
>


-- 


*​Dibyendu DeyPostdoctoral Research Scholar**Department of Physics, Arizona
State University *
*Tempe, AZ 85287, United States*
*Ph: +1-480-427-9970** | Personal Email: dibyendu@gmail.com
*
___
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] Ce does not converge

[Tran, Fabien]

Maybe you already know our paper (see appendix):
https://journals.aps.org/prb/abstract/10.1103/PhysRevB.89.155106



From: Wien  on behalf of Peter Blaha 

Sent: Wednesday, January 29, 2020 6:13 PM
To: wien@zeus.theochem.tuwien.ac.at
Subject: Re: [Wien] Ce does not converge

A "lanthanide" is NEVER simple.

The 4f states in GGA are all at EF and form a very narrow band. Thus
tiny changes in the potential move the 4f states relative to the broad
s,d states and huge internal charge transfer can occur.

RKmax=7 for a 4f element is MUCH too small. Use at least  8, better 9-10

Use TEMP(S) with 6 mRy (at the beginning). It damps oszillations.

Use a good k-mesh. 1000 k-points are by a factor of 10 too small.

I'm not sure if GGA+U is very good for Ce metal, at least not for the
low-volume fcc phase.

Am 29.01.2020 um 17:26 schrieb delamora:
> Dear WIEN2k community;
> I tried to make a spin polarized calculation the Ce metal;
> FCC a=b=c=5.16A with a U=5eV
> and also an antiferromagnetic calculation
> a=b=c/sqrt(2), c=5.16A
> Ce up; 0,0,0
> Ce dn; 1/2, 1/2, 1/2
> and they did not converge with cc=0.001
>
> for the ferromagnetic calculation I did 140 iterations with RxK=7 and
> #K=10,000
> :ENERGY convergence:  0 0 .69395000
> :ENERGY convergence:  0 0 .00010918
> :ENERGY convergence:  0 0 .00020010
> :DIS  :  CHARGE DISTANCE   ( 0.0377909 for atom1 spin 1)
>   0.0820411
> :DIS  :  CHARGE DISTANCE   ( 0.0301614 for atom1 spin 1)
>   0.0685493
> :DIS  :  CHARGE DISTANCE   ( 0.0278653 for atom1 spin 1)
>   0.0641289
>
> for the antiferromagnetic calculation I had RxK=7 and #K=1,000 and I did
> 63 iterations
>
> I find confusing that a simple calculation for a lanthanide element does
> not converge.
>
> Pablo
>
> ___
> 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
___
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] Ce does not converge

[Laurence Marks]

Additional point to what Peter said: if you are not optimizing positions,
sometimes MSEC3 is more stable than MSR1.

On Wed, Jan 29, 2020 at 11:13 AM Peter Blaha 
wrote:

> A "lanthanide" is NEVER simple.
>
> The 4f states in GGA are all at EF and form a very narrow band. Thus
> tiny changes in the potential move the 4f states relative to the broad
> s,d states and huge internal charge transfer can occur.
>
> RKmax=7 for a 4f element is MUCH too small. Use at least  8, better 9-10
>
> Use TEMP(S) with 6 mRy (at the beginning). It damps oszillations.
>
> Use a good k-mesh. 1000 k-points are by a factor of 10 too small.
>
> I'm not sure if GGA+U is very good for Ce metal, at least not for the
> low-volume fcc phase.
>
> Am 29.01.2020 um 17:26 schrieb delamora:
> > Dear WIEN2k community;
> > I tried to make a spin polarized calculation the Ce metal;
> > FCC a=b=c=5.16A with a U=5eV
> > and also an antiferromagnetic calculation
> > a=b=c/sqrt(2), c=5.16A
> > Ce up; 0,0,0
> > Ce dn; 1/2, 1/2, 1/2
> > and they did not converge with cc=0.001
> >
> > for the ferromagnetic calculation I did 140 iterations with RxK=7 and
> > #K=10,000
> > :ENERGY convergence:  0 0 .69395000
> > :ENERGY convergence:  0 0 .00010918
> > :ENERGY convergence:  0 0 .00020010
> > :DIS  :  CHARGE DISTANCE   ( 0.0377909 for atom1 spin 1)
> >   0.0820411
> > :DIS  :  CHARGE DISTANCE   ( 0.0301614 for atom1 spin 1)
> >   0.0685493
> > :DIS  :  CHARGE DISTANCE   ( 0.0278653 for atom1 spin 1)
> >   0.0641289
> >
> > for the antiferromagnetic calculation I had RxK=7 and #K=1,000 and I did
> > 63 iterations
> >
> > I find confusing that a simple calculation for a lanthanide element does
> > not converge.
> >
> > Pablo
> >
> > ___
> > Wien mailing list
> > Wien@zeus.theochem.tuwien.ac.at
> >
> https://urldefense.proofpoint.com/v2/url?u=http-3A__zeus.theochem.tuwien.ac.at_mailman_listinfo_wien=DwIF-g=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws=U_T4PL6jwANfAy4rnxTj8IUxm818jnvqKFdqWLwmqg0=XnmJR_h3fBx3KIFwHrCwC7ne-XAPmvCPBUDahPzXwWk=h6dKg9SNfK3M_EQM5F2a2lZjppYJFKT9SPk92h1-AWU=
> > SEARCH the MAILING-LIST at:
> https://urldefense.proofpoint.com/v2/url?u=http-3A__www.mail-2Darchive.com_wien-40zeus.theochem.tuwien.ac.at_index.html=DwIF-g=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws=U_T4PL6jwANfAy4rnxTj8IUxm818jnvqKFdqWLwmqg0=XnmJR_h3fBx3KIFwHrCwC7ne-XAPmvCPBUDahPzXwWk=2hOwqJa-CpkPJPePm-bq1ynIOKfNkSI6k4X8RPLOdeA=
> >
>
> --
> --
> 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:
> https://urldefense.proofpoint.com/v2/url?u=http-3A__www.wien2k.at=DwIF-g=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws=U_T4PL6jwANfAy4rnxTj8IUxm818jnvqKFdqWLwmqg0=XnmJR_h3fBx3KIFwHrCwC7ne-XAPmvCPBUDahPzXwWk=U_fnl36pmcPoG7H7-PyNwfyn5XhEXkOrz00EP_drQ7E=
> WWW:
>
> https://urldefense.proofpoint.com/v2/url?u=http-3A__www.imc.tuwien.ac.at_tc-5Fblaha-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D-2D=DwIF-g=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws=U_T4PL6jwANfAy4rnxTj8IUxm818jnvqKFdqWLwmqg0=XnmJR_h3fBx3KIFwHrCwC7ne-XAPmvCPBUDahPzXwWk=obvbuxHeMS2k70__sA5uFA-ozShH3Yh9AGaVrVjkS74=
>
>
> ___
> Wien mailing list
> Wien@zeus.theochem.tuwien.ac.at
>
> https://urldefense.proofpoint.com/v2/url?u=http-3A__zeus.theochem.tuwien.ac.at_mailman_listinfo_wien=DwIF-g=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws=U_T4PL6jwANfAy4rnxTj8IUxm818jnvqKFdqWLwmqg0=XnmJR_h3fBx3KIFwHrCwC7ne-XAPmvCPBUDahPzXwWk=h6dKg9SNfK3M_EQM5F2a2lZjppYJFKT9SPk92h1-AWU=
> SEARCH the MAILING-LIST at:
> https://urldefense.proofpoint.com/v2/url?u=http-3A__www.mail-2Darchive.com_wien-40zeus.theochem.tuwien.ac.at_index.html=DwIF-g=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws=U_T4PL6jwANfAy4rnxTj8IUxm818jnvqKFdqWLwmqg0=XnmJR_h3fBx3KIFwHrCwC7ne-XAPmvCPBUDahPzXwWk=2hOwqJa-CpkPJPePm-bq1ynIOKfNkSI6k4X8RPLOdeA=
>


-- 
Professor Laurence Marks
Department of Materials Science and Engineering
Northwestern University
www.numis.northwestern.edu
Corrosion in 4D: www.numis.northwestern.edu/MURI
Co-Editor, Acta Cryst A
"Research is to see what everybody else has seen, and to think what nobody
else has thought"
Albert Szent-Gyorgi
___
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] Ce does not converge

[Peter Blaha]

A "lanthanide" is NEVER simple.

The 4f states in GGA are all at EF and form a very narrow band. Thus 
tiny changes in the potential move the 4f states relative to the broad 
s,d states and huge internal charge transfer can occur.


RKmax=7 for a 4f element is MUCH too small. Use at least  8, better 9-10

Use TEMP(S) with 6 mRy (at the beginning). It damps oszillations.

Use a good k-mesh. 1000 k-points are by a factor of 10 too small.

I'm not sure if GGA+U is very good for Ce metal, at least not for the 
low-volume fcc phase.


Am 29.01.2020 um 17:26 schrieb delamora:

Dear WIEN2k community;
I tried to make a spin polarized calculation the Ce metal;
FCC a=b=c=5.16A with a U=5eV
and also an antiferromagnetic calculation
a=b=c/sqrt(2), c=5.16A
Ce up; 0,0,0
Ce dn; 1/2, 1/2, 1/2
and they did not converge with cc=0.001

for the ferromagnetic calculation I did 140 iterations with RxK=7 and 
#K=10,000

:ENERGY convergence:  0 0 .69395000
:ENERGY convergence:  0 0 .00010918
:ENERGY convergence:  0 0 .00020010
:DIS  :  CHARGE DISTANCE       ( 0.0377909 for atom    1 spin 1) 
  0.0820411
:DIS  :  CHARGE DISTANCE       ( 0.0301614 for atom    1 spin 1) 
  0.0685493
:DIS  :  CHARGE DISTANCE       ( 0.0278653 for atom    1 spin 1) 
  0.0641289


for the antiferromagnetic calculation I had RxK=7 and #K=1,000 and I did 
63 iterations


I find confusing that a simple calculation for a lanthanide element does 
not converge.


Pablo

___
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] Ce does not converge

[Tran, Fabien]

The problem may also be due that the calculation tries to escape from a local 
minimum (in the electronic configurations space) but cannot, or switches 
between two electronic configurations.

Maybe it can help to do a calculation with -orbc (the case.vorbup/dn files are 
kept fixed):

1) copy the attached dmat files in your directory (they were obtained from a 
converged NM calculation with RMT=2.2 bohr)

2) execute "x orb -up" and "x orb -dn" to generate case.vorbup/dn

3) runsp_lapw -orbc ...

4) save_lapw ...

5) runsp_lapw -orb



From: Wien  on behalf of delamora 

Sent: Wednesday, January 29, 2020 5:26 PM
To: A Mailing list for WIEN2k users
Subject: [Wien] Ce does not converge

Dear WIEN2k community;
I tried to make a spin polarized calculation the Ce metal;
FCC a=b=c=5.16A with a U=5eV
and also an antiferromagnetic calculation
a=b=c/sqrt(2), c=5.16A
Ce up; 0,0,0
Ce dn; 1/2, 1/2, 1/2
and they did not converge with cc=0.001

for the ferromagnetic calculation I did 140 iterations with RxK=7 and #K=10,000
:ENERGY convergence:  0 0 .69395000
:ENERGY convergence:  0 0 .00010918
:ENERGY convergence:  0 0 .00020010
:DIS  :  CHARGE DISTANCE   ( 0.0377909 for atom1 spin 1)  0.0820411
:DIS  :  CHARGE DISTANCE   ( 0.0301614 for atom1 spin 1)  0.0685493
:DIS  :  CHARGE DISTANCE   ( 0.0278653 for atom1 spin 1)  0.0641289

for the antiferromagnetic calculation I had RxK=7 and #K=1,000 and I did 63 
iterations

I find confusing that a simple calculation for a lanthanide element does not 
converge.

Pablo


Ce.dmatup
Description: Ce.dmatup


Ce.dmatdn
Description: Ce.dmatdn
___
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] Ce does not converge

[Laurence Marks]

RMTs ?

These are probably large, HDLO may be needed., larger RKMax and VNS.

On Wed, Jan 29, 2020 at 10:26 AM delamora  wrote:

> Dear WIEN2k community;
> I tried to make a spin polarized calculation the Ce metal;
> FCC a=b=c=5.16A with a U=5eV
> and also an antiferromagnetic calculation
> a=b=c/sqrt(2), c=5.16A
> Ce up; 0,0,0
> Ce dn; 1/2, 1/2, 1/2
> and they did not converge with cc=0.001
>
> for the ferromagnetic calculation I did 140 iterations with RxK=7 and
> #K=10,000
> :ENERGY convergence:  0 0 .69395000
> :ENERGY convergence:  0 0 .00010918
> :ENERGY convergence:  0 0 .00020010
> :DIS  :  CHARGE DISTANCE   ( 0.0377909 for atom1 spin 1)
>  0.0820411
> :DIS  :  CHARGE DISTANCE   ( 0.0301614 for atom1 spin 1)
>  0.0685493
> :DIS  :  CHARGE DISTANCE   ( 0.0278653 for atom1 spin 1)
>  0.0641289
>
> for the antiferromagnetic calculation I had RxK=7 and #K=1,000 and I did
> 63 iterations
>
> I find confusing that a simple calculation for a lanthanide element does
> not converge.
>
> Pablo
> ___
> Wien mailing list
> Wien@zeus.theochem.tuwien.ac.at
>
> https://urldefense.proofpoint.com/v2/url?u=http-3A__zeus.theochem.tuwien.ac.at_mailman_listinfo_wien=DwICAg=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws=U_T4PL6jwANfAy4rnxTj8IUxm818jnvqKFdqWLwmqg0=DN4M3D3RgnV9E27Pf6cF9lWeHIP3n4_jDt89MuqTWOo=qCXbf5if0YN1NxO2pk-UvxStRIZeLq_0YrJAzK8NBpg=
> SEARCH the MAILING-LIST at:
> https://urldefense.proofpoint.com/v2/url?u=http-3A__www.mail-2Darchive.com_wien-40zeus.theochem.tuwien.ac.at_index.html=DwICAg=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws=U_T4PL6jwANfAy4rnxTj8IUxm818jnvqKFdqWLwmqg0=DN4M3D3RgnV9E27Pf6cF9lWeHIP3n4_jDt89MuqTWOo=Bns4fXybWZ-T5HZkY_Ut2I74kuIt3FmlBvyxGfe4cY4=
>


-- 
Professor Laurence Marks
Department of Materials Science and Engineering
Northwestern University
www.numis.northwestern.edu
Corrosion in 4D: www.numis.northwestern.edu/MURI
Co-Editor, Acta Cryst A
"Research is to see what everybody else has seen, and to think what nobody
else has thought"
Albert Szent-Gyorgi
___
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] Ce does not converge

[delamora]

Dear WIEN2k community;
I tried to make a spin polarized calculation the Ce metal;
FCC a=b=c=5.16A with a U=5eV
and also an antiferromagnetic calculation
a=b=c/sqrt(2), c=5.16A
Ce up; 0,0,0
Ce dn; 1/2, 1/2, 1/2
and they did not converge with cc=0.001

for the ferromagnetic calculation I did 140 iterations with RxK=7 and #K=10,000
:ENERGY convergence:  0 0 .69395000
:ENERGY convergence:  0 0 .00010918
:ENERGY convergence:  0 0 .00020010
:DIS  :  CHARGE DISTANCE   ( 0.0377909 for atom1 spin 1)  0.0820411
:DIS  :  CHARGE DISTANCE   ( 0.0301614 for atom1 spin 1)  0.0685493
:DIS  :  CHARGE DISTANCE   ( 0.0278653 for atom1 spin 1)  0.0641289

for the antiferromagnetic calculation I had RxK=7 and #K=1,000 and I did 63 
iterations

I find confusing that a simple calculation for a lanthanide element does not 
converge.

Pablo
___
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 type II

[pieper]

Dear Gerhard,

nice ... thanks a lot for the references!

Best regards,

Martin


---
Dr. Martin Pieper
Karl-Franzens University
Institute of Physics
Universitätsplatz 5
A-8010 Graz
Austria
Tel.: +43-(0)316-380-8564


Am 2020-01-29 13:16, schrieb Fecher, Gerhard:

Dear Martin,
this concerns your remark:
"With two magnetic species, say, Mn and Cu, you would wind up with
different size of the moment on Mn and Cu. I know of no case where 
exact

compensation into an AFM structure occures by accident in such a
situation."

You may have the situation of a completely compensated ferrimagnet 
exampels are:

CrMnSb (or VFeSb) in the cubic C1b structure
H. van Leuken and R. A. de Groot, Phys. Rev. Lett. 74, 1171 (1995)
or more complicated
Mn1.5FeV0.5Al
Rolf Stinshoff et al; Phys. Rev. B 95, 060410(R) (2017)

However, it is by purpose rather than by accident.

This was already found by Neel in his work on antiferromagnets
(probably it is mentioned in the Nobel lecture)



Ciao
Gerhard

DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy:
"I think the problem, to be quite honest with you,
is that you have never actually known what the question is."


Dr. Gerhard H. Fecher
Institut of Inorganic and Analytical Chemistry
Johannes Gutenberg - University
55099 Mainz
and
Max Planck Institute for Chemical Physics of Solids
01187 Dresden

Von: Wien [wien-boun...@zeus.theochem.tuwien.ac.at] im Auftrag von
pieper [pie...@ifp.tuwien.ac.at]
Gesendet: Mittwoch, 29. Januar 2020 12:49
An: A Mailing list for WIEN2k users
Betreff: Re: [Wien] AFM type II

No one can give you an honest answer without knowing the structure you
put these elements in. Zr, S, Se are almost certainely non-magnetic, 
but
there are quite a few structures with magnetic moments on Cr, Cu, and 
of

course on Mn.

To make terminology more complicated, remember that AFM means fully
compensated magnetic moments. The net magnetization of a unit cell of
some AFM structure is zero. This does not happen by coincidence, it is
because by symmetry all moments have the same size, and there are as
many of them pointing in one direction as there are pointing in exactly
the opposite direction.

So, IF your compound REALLY is AFM by experiment (NO net 
magnetization),

you almost certainely have only one magnetic species in there (probably
Mn). With two magnetic species, say, Mn and Cu, you would wind up with
different size of the moment on Mn and Cu. I know of no case where 
exact

compensation into an AFM structure occures by accident in such a
situation. You always get something with net moment - and these are
called ferrimagnetic structures. And since we are at it: there are
canted and helical strucutures where the moments are not collinear (not
within the scope of Wien2k), there are spin density waves, ...

Scanning this thread my advice would be to study a (good) book on solid
state physics, with special attention payed to its chapter discussing
magnetic order. If it doesn't have such a chapter its not a good book -
at least not for you. Do NOT use wikipedia or this mailing list and its
archive as a substitute for such a reading. It will not work.


---
Dr. Martin Pieper
Karl-Franzens University
Institute of Physics
Universitätsplatz 5
A-8010 Graz
Austria
Tel.: +43-(0)316-380-8564


Am 2020-01-27 17:45, schrieb djamel slamnia:

WHEN I STUDY A COMPOUND CONTAINS THIS ELEMENETS CU  MN   CR  ZR S SE

BETWEEN THEM WITCH ONE TO PUT IT SPIN UP OR DOWN AND NON-MAGNETIC ???

 Le lundi 27 janvier 2020 à 14:22:39 UTC+1, Gavin Abo
 a écrit :

 As previously mentioned [1], a short literature survey showed that
AFM type II and III are terms used for _fcc_ and _bcc_ lattices.
Since spacegroup 156 is not one of those, it might be inappropriate to
use those terms for spacegroup 156 having a _primitive_ lattice [2] of
the hexagonal crystal family.  If you do a more extensive literature
survey yourself and find a paper (article, book, etc.) that defines
the AFM magnetic orders for spacegroup 156, then reference and use it
for what the AFM order is.  If there is not any notations and terms
for AFM magnetic orders for spacegroup 156, you might have to make
your own figure or write in your own words what the definition is
should any AFM magnetic orders exist for it.
Keep in mind that as mentioned before in the mailing list archive, the
initial configuration can be set in case.inst with "instgen_lapw -ask"
[3].

Though, you need to check the final magnetic order that comes out of
the scf [4], because the configuration set in case.inst with
instgen_lapw is just the initial one that could change [5,6].

It is also possible to try to force a magnetic order using dmatup/dn
matrices but the final magnetic order is still what comes out of the
scf and could be different [6-11].

Therefore, it likely not beneficial to name the AFM order before
starting a calculation such that you would likely want to identify the
name 

Re: [Wien] AFM type II

[Fecher, Gerhard]

Dear Martin,
this concerns your remark:
"With two magnetic species, say, Mn and Cu, you would wind up with
different size of the moment on Mn and Cu. I know of no case where exact
compensation into an AFM structure occures by accident in such a
situation."

You may have the situation of a completely compensated ferrimagnet exampels are:
CrMnSb (or VFeSb) in the cubic C1b structure
H. van Leuken and R. A. de Groot, Phys. Rev. Lett. 74, 1171 (1995)
or more complicated
Mn1.5FeV0.5Al
Rolf Stinshoff et al; Phys. Rev. B 95, 060410(R) (2017)

However, it is by purpose rather than by accident.

This was already found by Neel in his work on antiferromagnets (probably it is 
mentioned in the Nobel lecture)



Ciao
Gerhard

DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy:
"I think the problem, to be quite honest with you,
is that you have never actually known what the question is."


Dr. Gerhard H. Fecher
Institut of Inorganic and Analytical Chemistry
Johannes Gutenberg - University
55099 Mainz
and
Max Planck Institute for Chemical Physics of Solids
01187 Dresden

Von: Wien [wien-boun...@zeus.theochem.tuwien.ac.at] im Auftrag von pieper 
[pie...@ifp.tuwien.ac.at]
Gesendet: Mittwoch, 29. Januar 2020 12:49
An: A Mailing list for WIEN2k users
Betreff: Re: [Wien] AFM type II

No one can give you an honest answer without knowing the structure you
put these elements in. Zr, S, Se are almost certainely non-magnetic, but
there are quite a few structures with magnetic moments on Cr, Cu, and of
course on Mn.

To make terminology more complicated, remember that AFM means fully
compensated magnetic moments. The net magnetization of a unit cell of
some AFM structure is zero. This does not happen by coincidence, it is
because by symmetry all moments have the same size, and there are as
many of them pointing in one direction as there are pointing in exactly
the opposite direction.

So, IF your compound REALLY is AFM by experiment (NO net magnetization),
you almost certainely have only one magnetic species in there (probably
Mn). With two magnetic species, say, Mn and Cu, you would wind up with
different size of the moment on Mn and Cu. I know of no case where exact
compensation into an AFM structure occures by accident in such a
situation. You always get something with net moment - and these are
called ferrimagnetic structures. And since we are at it: there are
canted and helical strucutures where the moments are not collinear (not
within the scope of Wien2k), there are spin density waves, ...

Scanning this thread my advice would be to study a (good) book on solid
state physics, with special attention payed to its chapter discussing
magnetic order. If it doesn't have such a chapter its not a good book -
at least not for you. Do NOT use wikipedia or this mailing list and its
archive as a substitute for such a reading. It will not work.


---
Dr. Martin Pieper
Karl-Franzens University
Institute of Physics
Universitätsplatz 5
A-8010 Graz
Austria
Tel.: +43-(0)316-380-8564


Am 2020-01-27 17:45, schrieb djamel slamnia:
> WHEN I STUDY A COMPOUND CONTAINS THIS ELEMENETS CU  MN   CR  ZR S SE
>
> BETWEEN THEM WITCH ONE TO PUT IT SPIN UP OR DOWN AND NON-MAGNETIC ???
>
>  Le lundi 27 janvier 2020 à 14:22:39 UTC+1, Gavin Abo
>  a écrit :
>
>  As previously mentioned [1], a short literature survey showed that
> AFM type II and III are terms used for _fcc_ and _bcc_ lattices.
> Since spacegroup 156 is not one of those, it might be inappropriate to
> use those terms for spacegroup 156 having a _primitive_ lattice [2] of
> the hexagonal crystal family.  If you do a more extensive literature
> survey yourself and find a paper (article, book, etc.) that defines
> the AFM magnetic orders for spacegroup 156, then reference and use it
> for what the AFM order is.  If there is not any notations and terms
> for AFM magnetic orders for spacegroup 156, you might have to make
> your own figure or write in your own words what the definition is
> should any AFM magnetic orders exist for it.
> Keep in mind that as mentioned before in the mailing list archive, the
> initial configuration can be set in case.inst with "instgen_lapw -ask"
> [3].
>
> Though, you need to check the final magnetic order that comes out of
> the scf [4], because the configuration set in case.inst with
> instgen_lapw is just the initial one that could change [5,6].
>
> It is also possible to try to force a magnetic order using dmatup/dn
> matrices but the final magnetic order is still what comes out of the
> scf and could be different [6-11].
>
> Therefore, it likely not beneficial to name the AFM order before
> starting a calculation such that you would likely want to identify the
> name of the magnetic order after having finished the converged
> calculation.
>  [1]
> https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg19515.html
> [2] 

Re: [Wien] AFM type II

[pieper]

No one can give you an honest answer without knowing the structure you 
put these elements in. Zr, S, Se are almost certainely non-magnetic, but 
there are quite a few structures with magnetic moments on Cr, Cu, and of 
course on Mn.


To make terminology more complicated, remember that AFM means fully 
compensated magnetic moments. The net magnetization of a unit cell of 
some AFM structure is zero. This does not happen by coincidence, it is 
because by symmetry all moments have the same size, and there are as 
many of them pointing in one direction as there are pointing in exactly 
the opposite direction.


So, IF your compound REALLY is AFM by experiment (NO net magnetization), 
you almost certainely have only one magnetic species in there (probably 
Mn). With two magnetic species, say, Mn and Cu, you would wind up with 
different size of the moment on Mn and Cu. I know of no case where exact 
compensation into an AFM structure occures by accident in such a 
situation. You always get something with net moment - and these are 
called ferrimagnetic structures. And since we are at it: there are 
canted and helical strucutures where the moments are not collinear (not 
within the scope of Wien2k), there are spin density waves, ...


Scanning this thread my advice would be to study a (good) book on solid 
state physics, with special attention payed to its chapter discussing 
magnetic order. If it doesn't have such a chapter its not a good book - 
at least not for you. Do NOT use wikipedia or this mailing list and its 
archive as a substitute for such a reading. It will not work.



---
Dr. Martin Pieper
Karl-Franzens University
Institute of Physics
Universitätsplatz 5
A-8010 Graz
Austria
Tel.: +43-(0)316-380-8564


Am 2020-01-27 17:45, schrieb djamel slamnia:

WHEN I STUDY A COMPOUND CONTAINS THIS ELEMENETS CU  MN   CR  ZR S SE

BETWEEN THEM WITCH ONE TO PUT IT SPIN UP OR DOWN AND NON-MAGNETIC ???

 Le lundi 27 janvier 2020 à 14:22:39 UTC+1, Gavin Abo
 a écrit :

 As previously mentioned [1], a short literature survey showed that
AFM type II and III are terms used for _fcc_ and _bcc_ lattices.
Since spacegroup 156 is not one of those, it might be inappropriate to
use those terms for spacegroup 156 having a _primitive_ lattice [2] of
the hexagonal crystal family.  If you do a more extensive literature
survey yourself and find a paper (article, book, etc.) that defines
the AFM magnetic orders for spacegroup 156, then reference and use it
for what the AFM order is.  If there is not any notations and terms
for AFM magnetic orders for spacegroup 156, you might have to make
your own figure or write in your own words what the definition is
should any AFM magnetic orders exist for it.
Keep in mind that as mentioned before in the mailing list archive, the
initial configuration can be set in case.inst with "instgen_lapw -ask"
[3].

Though, you need to check the final magnetic order that comes out of
the scf [4], because the configuration set in case.inst with
instgen_lapw is just the initial one that could change [5,6].

It is also possible to try to force a magnetic order using dmatup/dn
matrices but the final magnetic order is still what comes out of the
scf and could be different [6-11].

Therefore, it likely not beneficial to name the AFM order before
starting a calculation such that you would likely want to identify the
name of the magnetic order after having finished the converged
calculation.
 [1]
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg19515.html
[2] https://en.wikipedia.org/wiki/Crystal_structure#Lattice_systems
[3]
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg10044.html

[4]
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg17516.html
[5]
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg03243.html
[6]
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg06739.html
[7]
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg14259.html
[8]
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg05054.html
[9]
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg13124.html
[10]
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg16281.html
[11]
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg16286.html

On 1/26/2020 1:01 PM, djamel slamnia wrote:





THANKS AGAIN SIR

I NEED TO KNOW WHAT IS THE AFM ORDERS FOR P3M1 (156) ??? TYPE II OR
III

THANKS IN ADVANCE





___
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