The ABINIT Users Mailing List ( CLOSED )

["Josef W. Zwanziger" <jzwanzig@jzwanzig.org>]

In the "3 atom case" as you call it (it's more officially called "the 
asymmetric unit"), note that the input is

spgroup 194
angdeg 90.0 90.0 120.0
natom 12
natrd 3
xred (3 lines, because natrd = 3)

You have to tell it which space group to use, that there will be 12 atoms in 
the primitive unit cell, and that it should read only 3 atom positions 
because there are only 3 in the asymmetric unit. Abinit correctly determined 
how to fill out the cell based on the space group, and also computes rprim 
correctly.


Then, later, you can either keep using this input, or, what I usually do, cut 
and paste into a new input file the OUTPUT of the above calculation. In the 
new file (what you call the "12 atom case"), you will have 

rprim (copied from above output)
acell (copied from above output)
natom 12
xred (now 12 lines, copied from above output)

Note in this case that spgroup, angdeg, and natrd DO NOT appear, instead you 
have rprim and the full set of xred. Abinit
will now figure out the space group and apply all the symmetries correctly.

The above procedure is safest because there are multiple ways to order the 
cell axes and so forth, and if you let abinit do the work for you as much as 
possible, it minimizes problems and also the structure obeys by definition 
the conventions that abinit wants to follow.

 Josef W. Zwanziger
Professor of Chemistry
Canada Research Chair in NMR Studies of Materials
Director, Atlantic Region Magnetic Resonance Centre
Department of Chemistry
Dalhousie University
6274 Coburg Road
Halifax, Nova Scotia B3H 4J3 Canada
tel: +1.902.494.1960
fax: +1.902.494.1310
web: http://jwz.chem.dal.ca
jzwanzig@jzwanzig.org,jzwanzig@dal.ca



----- Original Message ----
> From: 程迎春 <watermall@126.com>
> To: forum <forum@abinit.org>
> Sent: Wednesday, December 17, 2008 11:12:29 AM
> Subject: {Spam?} Re: Re: {Spam?} [abinit-forum] {Spam?} optimize structure 
> under pressure
> 
> Dear Professor Zwanziger，
>        Thank you for your help.
>        This structure is also CaLi2 and I use the *.FHI LDA psp. Actually, 
> I 
> have use Materials Studio to build CaLi2, and I get the entries in the 
> *.cell 
> file.  There are 12 atoms in the primitive cell, so I convert the *.cell 
> file to 
> meet Abinit's cystal building needs.  In your input, there are only three 
> atoms. 
> I tried to run scf caulculation from your input file, and I find that 
> abinit 
> convert the cystal system to :
>      acell    1.1844803351E+01  1.1844803351E+01  1.9311111271E+01 Bohr
>      rprim    1.0000000000E+00  0.0000000000E+00  0.0000000000E+00
>              -5.0000000000E-01  8.6602540378E-01  0.0000000000E+00
>               0.0000000000E+00  0.0000000000E+00  1.0000000000E+00
>       xred    3.3333333333E-01  6.6666666667E-01  6.2700000000E-02
>               0.0000000000E+00  0.0000000000E+00  0.0000000000E+00
>               1.6700000000E-01  3.3400000000E-01  2.5000000000E-01
>               6.6666666667E-01  3.3333333333E-01  5.6270000000E-01
>               0.0000000000E+00  0.0000000000E+00  5.0000000000E-01
>               8.3300000000E-01  1.6700000000E-01  7.5000000000E-01
>               3.3400000000E-01  1.6700000000E-01  7.5000000000E-01
>               6.6600000000E-01  8.3300000000E-01  2.5000000000E-01
>               1.6700000000E-01  8.3300000000E-01  2.5000000000E-01
>               8.3300000000E-01  6.6600000000E-01  7.5000000000E-01
>               6.6666666667E-01  3.3333333333E-01  9.3730000000E-01
>               3.3333333333E-01  6.6666666667E-01  4.3730000000E-01
> I am a little confused about this result.  Indeed, abinit convert your 
> input 
> correctly for it can actually find 12 atoms in the primitive cell. But I 
> have 
> some questions about ABINT cystal build system:
> 1.  If I optimize the cystal structure under pressure, and     finally I 
> will 
> got the "12 atom" entries, HOW to convert "12 atom" input to "3 atom" input?
> 2.  If I want to calculate Phonon band structure,  can I use "3 atom" 
> entries in 
> the input file? 
> 
> Thank you in advance.
> 
> 
> 　　　　　　　　
>                 Sincerely Yours，     
> 
> 　　　　　　　　  CHENG Yingchun
>