[Wien] Different "total force" with or without Forceconvergence and Supercell of hexagonal structure.

Donghui Guo dhguo at cat.hokudai.ac.jp
Thu Jan 8 07:27:55 CET 2009


Dear Prof. Blaha,

Thank you for your reply.
I did as you mentioned and I can initialize the structure successfully.
Unfortunately, I got a "lapw2.error" in the frist iteration. It shows:

   'FERMI'
   'FERMI'
   'FERMI'
   'FERMI'
   'FERMI'
   'FERMI'
   'FERMI'
   'FERMI'
   'FERMI'
   'FERMI'
   'FERMI'
   'FERMI'

Best regards, 

Yours	
Donghui Guo
2009-01-08


  
======= At 2009-01-07, 17:04:30 you wrote: =======

>ad 1) Only with force convergence, the scripts will change "TOT" to "FOR" in lapw2 and
>calculate the Pulay-forces. Thus, only the "total force" is ok, while when it says
>"partial forces" in the scf file, these forces are NOT complete.
>
>ad 2) Just accept the struct-file from nn and continue. The initialization analyses the symmetry
>and modifies the struct file.
>
>> 1) I have a Ni2P structure from ICSD database. After I ran SCF with the energy convergence parameter (0.0001), I got a total force about 30 mRy/a.u.for Ni atom. If I use both energy convergence parameter and force convergence parameter (1 mRy/a.u.), the total force of Ni decreased to about 3 mRy/a.u.. Meanwhile, the total energy in both calculation is nearly same. It is confused me since the structures are same in these two calculations.
>> 
>> 2) I want to calculate the bulk Ni2P with some P site vacancy defects. First I build a Ni2P unit cell with symmetry. The space group of Ni2P is P-62m and there are 4 inequivalent atoms in one unit cell. Then I build the supercell by "x supercell" with 2x2x2. The "x supercell" program showed:
>>      " Current sutructure has lattice type H
>>        Target lattice type will by H
>>        Add vacuum in z-direction for surface slab [bohr]:"
>> I chose "0" since I am interested in bulk. Then I got the new structure with 32 atoms.
>> I remove one P atom randomly in W2web interface and save this 31-atoms structure with lattice type H.
>> The problem is when I run NN program, it showed there are too much atoms are equivalent. I cannot find out what's wrong in the supercell-making process. 
>> 
>> Thank you in advance and Happy New Year.  
>> 
>> Best regards, 
>> Yours, 
>> Donghui Guo
>> 2009-01-07
>> 
>> 
>> ************************************************
>> Donghui Guo 
>> Catalysis Research Center (CRC)
>> Hokkaido University
>> 21-10 Kita, Sapporo 001-0021
>> Japan
>> dhguo at cat.hokudai.ac.jp
>> TEL: +81-11-706 9114 (office)
>> ******************************************************
>> 
>> _______________________________________________
>> Wien mailing list
>> Wien at zeus.theochem.tuwien.ac.at
>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>
>-- 
>
>                                       P.Blaha
>--------------------------------------------------------------------------
>Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
>Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
>Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
>--------------------------------------------------------------------------
>_______________________________________________
>Wien mailing list
>Wien at zeus.theochem.tuwien.ac.at
>http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien

= = = = = = = = = = = = = = = = = = = =

************************************************
Donghui Guo 
Catalysis Research Center (CRC)
Hokkaido University
21-10 Kita, Sapporo 001-0021
Japan
dhguo at cat.hokudai.ac.jp
TEL: +81-11-706 9114 (office)
******************************************************



More information about the Wien mailing list