The ABINIT Users Mailing List ( CLOSED )

["LEE Chin Chai" <cc-lee@imre.a-star.edu.sg>]

Hi Matthieu,

Thank for the correction for acell. It should be in angstrom,indeed.

Yes, a and b for rprim are interchanged.

Could you give some examples to demonstrate the different between
optcell=1 and optcell=3?

ChinChai

-----Original Message-----
From: Matthieu Verstraete [mailto:mjv500@york.ac.uk] 
Sent: Monday, October 09, 2006 7:11 PM
To: forum@abinit.org
Subject: Re: [abinit-forum] FW:Problem in using flag optcell


Hi,

It looks like your structure is too contorted. For one, your acell are 
probably in angstroem, which you must specify explicitly (add Angstr)

The Si atoms also look compacted into planes in the CoSi part. Probably 
wrong.

Further, I'm not sure (although maybe) that your rprim is correct: it 
doesn't look like the atoms are in nice hexagonal packing. If your a and
b 
are interchanges it may explain the last 2 points.

In short, abinit is fine, but your atomic positions are wrong somehow,
so 
the relaxation algorithm explodes the structure, and the next Broyden
step 
is all wrong.

Matthieu

-- 
================================================================
Dr. Matthieu Verstraete                 mailto:mjv500@york.ac.uk
Dept. of Physics, University of York,     tel: +44 1904 43 22 08
Heslington, YO10 5DD York, United Kingdom fax: +44 1904 43 22 14