[Pw_forum] vc-relax fixing volume in Base-centered orthorhombic lattice varying A, B, C

[Andrew Supka]

So I am thinking I want to do vc-relax allowing A,B,C to vary in
Base-centered orthorhombic while keeping the volume fixed.

I know we can do cell_dofree in the pw.x input file and can
modify cell_base.f90 if an option that we want isn't available to us.

The issue I'm having is that we can do

in cell_base.f90 I'll keep the cell volume fixed and vary the vectors of
the primitive ORCC cell as so..

1 1 0
1 1 0
0 0 1

with fix_volume=.true.

but if I use that iforceh with fix_volume then I will get cell_parameters
that will vary like this:

(CP11 , CP12 , 0,       ),
(CP11 , CP12 , 0,       ),
(0       , 0        , CP33,),


CP11 != CP12 and CP21 != CP22

so I am thinking I am losing the ORCC primitive cell shape. Ideally I'd
like to make it so

CP11 == CP12 and CP21 == CP22

while varying CP11,CP22, and CP33 and keeping a constant volume.

One option would be to transform the primitive ORCC cell and
atomic_positions to it's conventional counterpart and do cell_dofree='xyz'
with adding fix_volume=.true. to cell_base.f90.

I'd rather not do that just because it would take a while to do a vc-relax
with 90 atoms (and I'm thinking I want to do this for a few dozen chemical
variations of the same system)

Any suggestions?

Thanks,
Andrew
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20160125/802f022a/attachment.html>