Skip to Content.
Sympa Menu

forum - Re: [abinit-forum] Question on crystal building

forum@abinit.org

Subject: The ABINIT Users Mailing List ( CLOSED )

List archive

Re: [abinit-forum] Question on crystal building


Chronological Thread 
  • From: mmikami@rc.m-kagaku.co.jp
  • To: forum@abinit.org
  • Subject: Re: [abinit-forum] Question on crystal building
  • Date: Fri, 29 Aug 2003 16:51:21 +0900

Hello,

From the Internet, I picked up a structure of Al2O3.
(Your interest might be a different structure,
so please check your system for sure ...)
Looking into the page "Crystal Lattice*Structures"
(http://cst-www.nrl.navy.mil/lattice/index.html
this URL is found in ~ABINIT/Infos/Miscellaneous/Crystals),
then following the hyperlinks (Prototype-> Al2O3(alpha)),
we will reach the page "Corundum (D51)" :
http://cst-www.nrl.navy.mil/lattice/struk/d5_1.html

The primitive cell information is given in the next URL:
http://cst-www.nrl.navy.mil/lattice/struk.xmol/d5_1.pos
The conventional cell contains 40 atoms, while the primitive
cell contains only 10 atoms.
Thus the input parameters regarding the geometry would be
as follows for the primitive cell:
acell
5.1278736 5.1278736 5.1278736 anstrom
angdeg 55.30000 55.30000 55.30000
natom 10
typat1 4*1 6*2
xred
.35228000 .35228000 .35228000
-.35228000 -.35228000 -.35228000
.14772000 .14772000 .14772000
-.14772000 -.14772000 -.14772000
-.05640000 .55640000 .25000000
.05640000 -.55640000 -.25000000
.55640000 .25000000 -.05640000
-.55640000 -.25000000 .05640000
.25000000 -.05640000 .55640000
-.25000000 .05640000 -.55640000
znucl 13.0 8.0

In this case, "spgroup" cannot be used, as mentioned in
http://www.abinit.org/wws/arc/forum/2003-07/msg00052.html
Instead of assigning "angdeg", we can assign "rpim" based on the
information of "Primitive vectors" :
a(1) = .25698400 3.62139800 3.62139800
a(2) = 3.62139800 .25698400 3.62139800
a(3) = 3.62139800 3.62139800 .25698400
"rprim" can be obtained by dividing the above values
with acell "5.1278736".

If you still prefer the conventional cell ... (it would be much
heavier calculation ...). you might want to read the following e-mail
that appeared on the Forum ML for your reference :
http://www.abinit.org/wws/arc/forum/2003-08/msg00000.html (*)
In this case, the lattice constant, alpha, beta, gamma (i.e. the angles
for the lattice vectors) and the positons of the inequivalent atoms are
necessary. Then, by assigning "spgroup", all the atom position will be
produced. (Kindly see the example file "MgZn2.in" attached in the
above Mail (*).) ... the rest is left for the questioner's task ...

Regards,
Masayoshi

On 2003.8.29, at 14:30 Asia/Tokyo, sundar.amancherla@geind.ge.com wrote:

Hello

I am a new user of ABINIT and also electronic structure calculations. I need a little help in constructing a crystal of alpha-alumina. If I do not have all the atomic positions (10 atoms) considering a primitive rhombohedral cell, would it be difficult to create the crystal.

I have gone through the tutorial and examples' input files, and could not get how to use the spgroup,brvlatt,natom combination.

I have these info -

Spacegroup 167 (R-3c)
a = 3.621A; b = 0.257A

Do I need any other info? Could you point me towards an example. Thanks very much in advance for the help...

--
Aman





Archive powered by MHonArc 2.6.16.

Top of Page