The ABINIT Users Mailing List ( CLOSED )

["D. R. Hamann" <drh@bell-labs.com>]

Dear Rongquin,

The abinis results from t65-66 for the piezoelectric tensor are
incomplete (for speed).  Only one electric field direction is used.
The results are in reduced units I won't try to explain here, and
the components have the basic Abinit indexing: pert = natom+2
for the electric field perturbation, natom+3 for uniaxial strains,
and natom+4 for shear strains.  The "dir=1,2,3" index corresponds to
primitive lattice directions for the electric field and pairs of
cartesian coordinates for the strains, (xx, yy, zz)  or (yz, xz, xy).

If this has you thoroughly confused, download version 4.3.2, look at
tests Test_v4/t67-70, and read Infos/Theory/lr.pdf.  In 4.3.x,
post-processing of the abinis output by anaddb has been added to
give the piezoelectric tensor in straight cartesian coordinates
and real units (C/m^2).  You will see from t68.out, for example,
that there are 3X6 piezoelectric tensor elements as you expect
(this was a full calculation).  The strain indexing (rows) is as
above (standard Voigt).  In addition, the new anaddb capabilities
allow the very important ion-relaxation contributions to the tensor
to be calculated by combining several sets of abinit results
(also described in lr.pdf).  Good luck.

Don Hamann


Wu Rongqin wrote:

Dear all

The piezoelectric tensor, in general has 6 components for each
direction. However, there is little documents about the output format so
as to get e11, e12,.... and so on. Can anybody give some directions on
this matter, using the results of 
Test_v3/t65.in, t66.in ?

Thanks

Rongqin WU
 

  

-- 
D. R. Hamann				Phone: 908-582-4454
Director, Theoretical Materials    	Fax:   908-582-4702
  Physics Research (retired)		email: drh@physics.bell-labs.com
Bell Laboratories 
Lucent Technologies
700 Mountain Ave, Room 1D-371
Murray Hill, NJ 07974-0636 USA