Dear All,
I write you about an issue I found in the use of the cell_dofree option.
I want to make a variable cell optimization by fixing the lattice angles and
varying only
the lattice axis lengths. It should correspond to cell_dofree = 'xyz' but it
doesn't work
for any cells I tested.
Below you can find the input.
Moreover, a more general question concerns the constrained optimization: is it
possible to
impose a constrain on two or more lattice parameters so that they can vary of
the same
amount? This for example would allow to optimize a cubic cell that for
convenience becomes
orthorombic by applying a supercell: after a constrained optimization where the
lattice
parameters ratio are constrained, you will get still a cubic symmetry even
though you
worked on an orthorombic cell. Is this possible in your opinion?
Thanking you in advance,
Federico Gallino
!test
&CONTROL
title = 'sampleA' ,
calculation = 'vc-relax' ,
pseudo_dir = '/QUANTUM_ESPRESSO/espresso-5.0.2/pseudo/' ,
prefix = 'sampleA' ,
verbosity = 'default' ,
/
&SYSTEM
ibrav = 0,
celldm(1) = 16.11503102 ,
nat = 32,
ntyp = 3,
ecutwfc = 27 ,
ecutrho = 270 ,
occupations = 'smearing' ,
smearing = 'cold',
degauss = 0.03 ,
/
&ELECTRONS
electron_maxstep = 200,
startingpot = 'atomic' ,
startingwfc = 'atomic' ,
mixing_mode = 'local-TF' ,
mixing_beta = 0.2 ,
diagonalization = 'david' ,
/
&IONS
ion_dynamics = 'bfgs' ,
phase_space = 'full' ,
/
&CELL
cell_dynamics = 'bfgs' ,
cell_dofree = 'xyz' ,
/
CELL_PARAMETERS alat
0.000000 0.707107 0.707107
0.707107 0.000000 0.707107
0.707107 0.707107 0.000000
ATOMIC_SPECIES
......
ATOMIC_POSITIONS angstrom
.....
K_POINTS automatic
3 3 3 0 0 0
This message and any attachments are confidential, privileged and/or otherwise
protected from disclosure to anyone other than its intended recipient(s).
Any unauthorized review, use, reproduction, disclosure, modification and/or
dissemination is strictly prohibited. If you are not the intended recipient,
please notify the sender by replying immediately to this message.
Do not disclose the contents to any other person, use it for any purpose, or
store or copy the information in any medium.
_______________________________________________
Pw_forum mailing list
[email protected]
http://pwscf.org/mailman/listinfo/pw_forum
