The ABINIT Users Mailing List ( CLOSED )

[Mustafa Uludogan <uludoganmustafa@yahoo.com>]

Dear Dr. Gonze and Dr. Ghosez,
I would like to ask a question about how you obtained the displacement eigenvectors for BaTiO3 in your paper of
Ferroelectrics V206-207, 205 (1998)
as delta(Ba)=-0.002, delta(Ti)=-0.096, delta(O_1)=+0.158 and delta(O_1=O_2)=+0.071.
Here is my result for the first eigenvector displacement at gamma point. 

Eigendisplacements
 (will be given, for each mode : in cartesian coordinates
   for each atom the real part of the displacement vector,
   then the imaginary part of the displacement vector)
  Mode number    1   Energy   -5.341274E-04
 Attention : low frequency mode.
   (Could be unstable or acoustic mode)
;  1  0.00000000E+00  3.38387803E-05  0.00000000E+00
;     0.00000000E+00  0.00000000E+00  0.00000000E+00
;  2  0.00000000E+00  2.30468183E-03 
 0.00000000E+00
;     0.00000000E+00  0.00000000E+00  0.00000000E+00
;  3  0.00000000E+00 -1.95491772E-03  0.00000000E+00
;     0.00000000E+00  0.00000000E+00  0.00000000E+00
;  4  0.00000000E+00 -3.27586692E-03  0.00000000E+00
;     0.00000000E+00  0.00000000E+00  0.00000000E+00
;  5  0.00000000E+00 -1.95684952E-03  0.00000000E+00
;     0.00000000E+00  0.00000000E+00  0.00000000E+00


I also checked my result with test21.out  of Test_v2, the order of eigendisplacements seems to be similar with my results except I have used ecut=50 Ha and ngkpt 4 4 4.
From   phfr3.F90 file, I see that these are normalized eigendisplacements.
I calculated eigenvectors from   sqrt(atomicmass*e_mass)*eigendisplacement 
as defined at the end of phfr3.F90.
I found eige!
 nvectors
 in y axis as
for Ba 0.0169
     Ti 0.6807
      O_1  -0.55944
      O_2=O_3   -0.3338
As far as I know these results have cartesian coordinate dimensions(Bohr?)
These are all normalized results <eta|M|eta>=1
eta=gamma/sqrt(M)

To double check, I eliminated eigendisplacement in phfr3.F90 and printed out eigenvectors, I obtained the above eigenvector results. 
So I will kindly ask how you processed these  eigenvectors  in anaddb (or postanaddb)and reported the values in this paper? I think I might be missing something.
Yours Sincerely

 


Mustafa Uludogan
Chemical Engineering Department
Texas A&M University
College Station, TX 77843

---

Yahoo! Mail
Bring photos to life! New PhotoMail  makes sharing a breeze.