The ABINIT Users Mailing List ( CLOSED )

["Anglade Pierre-Matthieu" <anglade@gmail.com>]

PS: What is happenning if you set brvltt to 0 ?

On 8/6/07, Anglade Pierre-Matthieu <anglade@gmail.com> wrote:

Hi,

Would you mind sending the input file for reproduction of your problem ?

Regards

PMA

On 8/6/07, Jess Kondor < kondor.jess@gmail.com> wrote:

 Dear all,

  I tried to use abinit geometry optimizer to build whole structure using crystallographic data. I have:
   BiMnO3,  C2 group (#5) , volume = 491.7 Ang^3
a = 9.854 , b =   5.606 , c =  9.532 (Ang) , alpha = gamma= 90.0 ; beta=110.60
11
    Mn1 Mn 4c  0.254 0.062 0.757
    Bi2   Bi  4c  0.359 0.1 0.115
    Mn3 Mn 2a  0.0 0.0 0.0
    O4   O  4c   0.144 0.257 0.106
    O5   O  4c   0.346 0.363 0.415
    O6   O  4c  0.365 0.276 0.915
    O7   O  4c  0.1 0.015 0.838
    O8   O  4c  0.4 0.169 0.672
   Mn9 Mn 2b  0.5 0.096 0.5
   Bi10 Bi  4c  0.135 0.037 0.372
   O11 O  4c  0.149 0.405 0.63

the corresponding part of the input file is :

spgroup 5
brvltt -1
acell  9.854   5.606   9.532  Angstr
angdeg 90.0 110.60 90.0

ntypat 3
natom  20
natrd 11
typat  2*1 3*2 6*3
xred
      0.359 0.1 0.115
      0.135 0.037 0.372
      0.0 0.0 0.0
      0.254 0.062 0.757
      0.5 0.096 0.5
      0.144 0.257 0.106
      0.346 0.363 0.415
      0.365 0.276 0.915
      0.1 0.015 0.838
      0.4 0.169 0.672
      0.149 0.405 0.63


I see in the output file:
 Symmetries : space group C2 (#  5); Bravais mC (1-face-center monocl.)

but the unit cell volume is almost twice smaller and the angles are strange:

 Real(R)+Recip(G) space primitive vectors, cartesian coordinates (Bohr,Bohr^-1):
 R(1)=  9.0064347  5.2959575  0.0000000  G(1)=  0.0555159  0.0944116  0.0208670
 R(2)= -9.0064347  5.2959575  0.0000000  G(2)= -0.0555159  0.0944116 -0.0208670
 R(3)= -6.5484460  0.0000000 17.4218583  G(3)=  0.0000000   0.0000000  0.0573992
 Unit cell volume ucvol=  1.6619650E+03 bohr^3
 Angles (23,13,12)=  7.23445051E+01  1.07655495E+02  1.19087300E+02 degrees

V = ~1661 bohr^3  = ~ 246 Ang^3

Did I miss something?

  Thanks,
     jess


 

--
Pierre-Matthieu Anglade

-- 
Pierre-Matthieu Anglade