ABINIT Discussion Forums

I'm trying to set the orthorhombic face centered lattice by using the input as follows:
ntypat 1
znucl 1
natom 2
typat 1 1

xcart
0.0 0.0 0.0
a 0.0 0.0

rprim a 0.000 0.0
0.000 b/2 -c
0.000 b/2 c
scalecart 3*1 acell 3*1

here a b and c are lattice parameters. But abinit fails with this input. What I am doing wrong?

Hello,

Could you be more specific about how abinit is failing ? Add the end of the log file, output files, ...
At first sight, it seems that you put the 2 atoms at the same place (0 0 0 and a 0 0 correspond to the same coordinate)

David

Thank you David for kind reply! Actually I am using v_sim to visualize the structure. It calls abinit and plots the structure. The error is as follows:

chkprimit : ERROR -
According to the symmetry finder, the unit cell is
NOT primitive. The multiplicity is 2 .
The use of non-primitive unit cells is allowed
only when the input variable chkprim is 0.
Action : either change your unit cell (rprim or angdeg),
or set chkprim to 0.

I'm trying new input set:

ntypat 1
znucl 1
natom 2
typat 1 1
xcart
.000000 .000000 0.0
0.00000 6.00000 10.0

rprim 8.0 -6.0 0.0
8.0 6.0 0.0
0.000 0.0 20.0
scalecart 3*1 acell 3*1

and it fails again. The position (0.0 6.0 10.0) is not the same place as (0 0 0) as far the c parameter is 20, not 10. Skipping second atom in xcart matrix means to set simple orthorhombic but I need face-centered one, thus I have two molecules(atoms) in primitive cell.

Hi,

You don't need two atoms to define a face-centered orthorhombic (FCO) structure. The variable rprim will define the system as FCO, not the number of atoms. In your case, I think you don't need the second atom. In the last case, it's clear why abinit failed : xcart = a 0 0 is equivalent to 0 0 0 because of the first rprim. I think you should not put a second atom. What does abinit say if you use
rprim a 0.000 0.0
0.000 b/2 -c
0.000 b/2 c
and
xcart 0 0 0
with only 1 atom?

Simon

Dear Simon,
It says:
space group Cm m m (# 65); Bravais oC (1-face-center ortho.)

but I need #64 C m c a face and base centered, but not body centered #63 C m c m.

I guess I'm starting to understand. If the previous mentioned rprim set produces 1-face centered lattice then adding second molecule to primitive cell fills the other face centered positions but also fills the face positions already filled. Thus I should use rprim set which yelds simple ortho cell and only than I can use second molecule in primitive cell. Is this the case?

upd: No I was wrong. the current rprim set gives base centered lattice, adding second molecule to the primitive cell can't crossover with the base centered molecules. I'm stack

Ok I made the following:

ntypat 2
znucl 17 80 # the second atom is to indicate by other color in v_sim there the secondatom placed to.

natom 2
typat 1 2
xcart
.000000 .000000 0.0
0.00000 6.0000 10.0

rprim 8.0 -6.0 0.0
8.0 6.0 0.0
0.0 0.0 20.0

I made a visualisation in V_Sim software of such structure. Hg was marked red and Cl was gray.
I see the Cl atoms fills the main positions in cell plus base centered. But Hg are face centered only. No Hg was placed in base center position. The structure seems just fine. But when I skip Hg to Cl atoms as follows:

ntypat 1
znucl 17
natom 1
typat 1
xcart
.000000 .000000 0.0
0.00000 6.0000 10.0

rprim 8.0 -6.0 0.0
8.0 6.0 0.0
0.0 0.0 20.0
I still getting the above mentioned error.
chkprimit : ERROR -
According to the symmetry finder, the unit cell is
NOT primitive. The multiplicity is 2 .
The use of non-primitive unit cells is allowed
only when the input variable chkprim is 0.
Action : either change your unit cell (rprim or angdeg),
or set chkprim to 0.

later on...
setting xcart
.000000 .000000 0.0
0.00000 6.0000 9.99
i.e. shifting second atom from the symmetric position breaks symmetry but works...
It's so wierd!

I think you have to change your last rprim to 0 6 10. It cuts your unit cell volume by half and should yield the same lattice.

Thank you, I'll try it. I also gonna try this rprim:
a/2 b/2 0
0 b/2 c/2
a/2 0 c/2

Should I?

Indeed, you should use rprim similar to a FCC system:
0 b/2 c/2
a/2 0 c/2
a/2 b/2 0
Keep us posted if it doesn't work.

blackburn, yes it seems to works like a charm. But the space group abinit founds to be 69 (Fmmm) but I was always thinking it is 64 (Cmca)... Hmmm where was I mistaken?
One more question. Did abinit provide the primitive basis set for the reciprocal lattice?
Can I use the set found at bilbao server?
http://www.cryst.ehu.es/cgi-bin/cryst/p ... g=fmmmfagr

p.s. Once again. Many thx for your advices!!!