The ABINIT Users Mailing List ( CLOSED )

["Chun Li " <chun.li@mf.mpg.de>]

Dear Hamann,

You are right. The piezoelectric constants of the supercell are
different for different width of the vacuum layer. Is there any material
that introduces the specific arithmetic of this kind of calculations in
abinit? 
One more question, I wonder if the polarization tensor can be obtained
by abinit rf calculation followed with anaddb.

Thank you and best regards.

Chun Li

-----Original Message-----
From: D. R. Hamann [mailto:drh@bell-labs.com] 
Sent: Friday, October 29, 2004 4:55 PM
To: forum@abinit.org
Subject: Re: [abinit-forum] RE: about DDB file

Dear Chun,

Abinit can do the calculation.  The results, since they are
non-physical,
will depend on the width of the vacuum region.  No experiment can
measure 
the effect of a surface on the elastic tensor of a material.  While
there
was debate in the literature about the role of surface charge in
piezoelectricity, it was firmly established to be a bulk property by 
Richard Martin about 20 years ago.

D. R. Hamann


Chun Li wrote:

Dear Hamann,

Thank you for your reply.

The problem has been resolved by using mrgddb to merge the DDB files
into a complete one.

About the serious issue you mentioned, I wonder if it is only for bulk
model that the elastic tensor and piezoelectric tensor can be calculated
meaningfully in ABINIT. If it is true, how to calculate the
piezoelectricity of film or surface structure models using abinit? My
intent is just to perform such simulations. Could you give me some
ideas? Thanks a lot.

Best regards.

Chun Li





-----Original Message-----
From: D. R. Hamann [mailto:drh@bell-labs.com] 
Sent: Thursday, October 28, 2004 6:32 PM
To: Chun Li; abinit_forum
Subject: Re: about DDB file

Dear Chun,

Please send your Abinit questions to the forum rather than my personal
email.

I think your anaddb issue relates to your first run being cut off. 
Abinit
only writes the DDB file after in completes all the perturbations in
a given dataset, so I believe that is why part was missing.  You can
divide perturbations among different datasets, or even among different
runs and use the mrgddb utility.

There is a more serious issue.  If I am interpreting your data
correctly,
it looks as if you have set up a slab surface model of ZnO with a vacuum
region.  The elastic tensor and piezoelectric tensor are bulk
properties,
and are meaningless for such a model.

D. R. Hamann


  


-- 
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