The ABINIT Users Mailing List ( CLOSED )

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

Dear Chun,

The short answer is that the piezoelectric tensor in the RF .out file
is in atomic units, bohr^-2.

The long answer is that for any non-trivial structure (wurtzite is
non-trivial), the physical result is the relaxed-ion piezoelectic
tensor.  This is calculated by doing an RF abinis run, followed
by an anaddb run.  This both incorporates the ionic relaxation effects,
which are usually large, and converts to C/m^2 (so labeled in the
.out file).  The anaddb .out file also gives the rigid-ion tensor
converted to C/m^2.

To understand the theory behind this, read Infos/Theory/vanderbilt-
anaddb-notes.pdf in the latest release (4.4.2).  To see how to set up
your calculations, look at Test_v4/README for the discussion of
t65-70, the t65-70.in files, and the Test_v4/Refs/t65-70.out files.
These tests are also present in 4.3.3.

To complete the story, the entries in the .DDB file corresponding
to the piezoelectric tensor are the mixed 2nd derivatives of the
energy per unit cell with respect to reduced electric field and
strain, and are dimensionless.  I believe the relevant reduced
units in Abinit are also discussed in the vanderbilt notes.

Best regards,
Don Hamann

Chun Li wrote:

Dear Hamann,

Please apologize my disturbing you with this trivial question.
Since I got some incorrect results (the difference is about 2 orders of
magnitude) of piezoelectric tensor for the wurtzite structure, I wonder
if the unit of the components of piezoelectric tensors obtained by RF
calculations with ABINIT is always C/m^2. I cannot find definite answer
in the help file. Could you give me authoritative answer? Thank you very
much.

Best regards.

Chun Li


  

-- 
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