[Wien] Problems in Mixer

Tue Aug 5 09:56:27 CEST 2008

Dear Prof. Stefaan

Again I am sorry for late. I was just waiting for my calc to come out. 

I followed the procedure you mentioned in your earlier mail. But again I am getting the same case.dmatup and case.dmatdn files. Then I  apply U
only to the atom nr-3 and ask to calculate density matrix for atom
nr.-3 only. Now the prograam is not coming out after first iteration.
So it's mean there must be some problem with atom nr.-2. For your
information, in Ferromagnetic LSDA calc, Magnetic moment in atom
nr-2(Co1-0.35)  is much lesser than the mag. moment in atom
nr-3(Co2-2.68). What might be the reason, for this problem? 

In your earlier mail, you have suggested to do the LSDA+U calc,
starting from the converged  LSDA calc. If somebody use -orb in the
first iteration of a freshly initialized case, what is wrong there?
Kindly explain this point.

With thanks

> Here I have attached case.dmatup and case.dmatdn files. But there is 
> nothing in case.dmatud files. I suspect,that this error might be due 
> to this case.dmatud file. 
No, case.dmatud should be empty unless you use the -ud switch (don't do 
that).

The problem is in the dmatup as well as in the dmatdn file for atom nr. 
2, which shows things like:

    2********************  0.000000 L, Lx,Ly,Lz in global orthogonal system
 -0.26809832+205  0.54867493+205    0.42184995+221 -0.11013654+222
 -0.36300519+237  0.71222078+237    0.17418686+253 -0.45123460+237
  0.21331738+237 -0.50097105+221

There are '***' instead of numbers, and very unphysical orbital 
occupations (the first number tells you you have -0.268 E+205 electrons 
in the m=2 d-orbital of Co....). Compare this to your atom nur. 3, where 
everything is OK:

    2  0.000000  0.000000  0.000000 L,

 Lx,Ly,Lz in global orthogonal system
  0.94814295E+00  0.25077213E-36    0.55565361E-18  0.00000000E+00
  0.30069670E-18  0.00000000E+00   -0.51617112E-02  0.44469230E-20
 -0.11790699E-17  0.50821977E-20

Here you have 0.948 (=almost 1.000) electrons in that orbital.

You said you successfully did an LDA calculation for this material. 
After that successful calculation, run 'x lapwdm -up' and 'x lapwdm 
-dn', and see which dmat files you get. Are they OK? If so, then 
continue with -orb in the same directory. Perhaps you tried to use -orb 
in the first iteration of a freshly initialized case? Don't do that, 
first converge with LDA.

Stefaan

--- On Wed, 30/7/08, rup_swarup at yahoo.com <rup_swarup at yahoo.com> wrote:

From: rup_swarup at yahoo.com <rup_swarup at yahoo.com>
Subject: Re: Problems in Mixer
To: "Wien" <wien at zeus.theochem.tuwien.ac.at>
Cc: stefaan.cottenier at fys.kuleuven.be
Date: Wednesday, 30 July, 2008, 4:17 PM

Dear Steffan

Thanks for your reply and I am very sorry for so late to respond.

Here I have attached case.dmatup and case.dmatdn files. But there is
nothing in case.dmatud files. I suspect,that this error might be due to this
case.dmatud file. 
What do you suggest now? I will be waiting for your suggestion.
With best regards

--- On Tue, 29/7/08, rup_swarup at yahoo.com <rup_swarup at yahoo.com> wrote:
From: rup_swarup at yahoo.com <rup_swarup at yahoo.com>
Subject: Problems in Mixer
To: "Wien" <wien at zeus.theochem.tuwien.ac.at>
Date: Tuesday, 29 July, 2008, 7:38 PM

Dear Wien2k Users

I
am a very new user of Wien2k. I was trying to do a LSDA+U band
structure calculation of Ca3Co2O6. I did a spin polarized calculation
in FPAPW+lo method. I took U=5 eV and J=0.9 eV for two inequivalent 'Co'. When I run the program, in the first iteration it stop, showing the following error.........

LAPW0 END
 LAPW1 END
 LAPW1 END
 LAPW2 END
 LAPW2 END
LAPWDM END
LAPWDM END
 CORE  END
 CORE  END
forrtl: severe (59): list-directed I/O syntax error, unit 71, file /home/swarup/Calc_Wien2k/Ca3Co2O6_test/LSDAU/U5J.9/U5J.9.dmatup
Image              PC                Routine            Line        Source             
mixer              0000000000898246  Unknown               Unknown  Unknown
mixer              0000000000897446  Unknown              
 Unknown  Unknown
mixer              0000000000854446  Unknown               Unknown  Unknown
mixer              00000000008187EA  Unknown               Unknown  Unknown
mixer              0000000000817DF6  Unknown               Unknown 

 Unknown
mixer              000000000083E1D3  Unknown               Unknown  Unknown
mixer              000000000083CEB9  Unknown               Unknown  Unknown
mixer              0000000000423AB6  read_denmat_               37  read_denmat.f
mixer              0000000000412D37  MAIN__                    364  mixer.F
mixer              000000000040A7E2  Unknown               Unknown  Unknown
libc.so.6          00002B4821210B54  Unknown               Unknown  Unknown
mixer              000000000040A729  Unknown               Unknown  Unknown

>   stop error

My Case.inorb file is the following

  1  2  0                     nmod, natorb, ipr
PRATT  1.0                    BROYD/PRATT, mixing
  2 1 2                          iatom nlorb, lorb
  3 1 2                          iatom nlorb,

 lorb
  1                              nsic 0..AFM, 1..SIC, 2..HFM
   0.367 0.066        U J (Ry)   Note: we recommend to use U_eff = U-J and J=0
   0.367 0.066        U J
~ 
case.indm  

-9.                      Emin cutoff energy
 2                       number of atoms for which density matrix is calculated
 2  1  2      index of 1st atom, number of L's, L1
 3  1  2      dtto for 2nd atom, repeat NATOM times
 0 0           r-index, (l,s)index
~                                   
I was able to converge the LSDA calculation. That's why I feel 
this problem might be related to case.indm and case.inorb files.
Can you please help me to solve this.

With regards
 Swarup Kr. Panda 

      Did you know? You can CHAT without downloading messenger. Go to http://in.webmessenger.yahoo.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20080805/c78fef0c/attachment.html