[Pw_forum] ESM and spin polarized ionic relaxations

Alejandro Suarez ams751 at psu.edu
Fri Sep 30 20:42:38 CEST 2011


Hello,

I have recently tried running an ionic relaxation with a cell using the
effective screening medium (ESM) method, but have run into a hanging issue.

Whenever I try running the relaxation, the program hangs after the first scf
minimization converges and before the forces on the atoms are calculated.
The output stops immediately after the ESM charge and potential are printed.
What's interesting is this error also happens when I set the esm boundary
conditions to "pbc" which should remove any implementation of ESM should it
not? The relaxation runs without problems when all esm related variables are
removed.

This issue is reproducible on 4.3.1 and 4.3.2. Is ESM just not yet able to
do 'relax' runs?

Thank you for any help you may provide.

My input file is attached below:

--------------------------------------------
&control
 calculation='relax',
 restart_mode='from_scratch',
 prefix='qe-test',
 pseudo_dir='/usr/global/qe/4.3.2/pseudo',
 outdir='/gpfs/home/ams751/scratch/test',
 tprnfor = .TRUE.
/
&system
    ibrav = 0, celldm(1) = 1.889725989,
    tot_charge=0,
    nat= 2, ntyp= 2,
    ecutwfc = 22.05,
    starting_magnetization(2) = 1.0
    nspin = 2,
    nosym=.TRUE.
    occupations='smearing', smearing='gauss', degauss=0.004
    assume_isolated='esm', esm_bc='pbc'
/
&electrons
    mixing_beta = 0.7
/
&ions
/
ATOMIC_SPECIES
 C   12.000000   C.pbe-rrkjus.UPF
 H    1.000000   H.pbe-rrkjus.UPF
CELL_PARAMETERS
 12.3000000000000000    0.0000000000000000    0.0000000000000000
 -6.1500000000000000   10.6521120000000000    0.0000000000000000
  0.0000000000000000    0.0000000000000000   20.0000000000000000
ATOMIC_POSITIONS (crystal)
C 0.46667 0.53333 0.00000
H 0.46667 0.53333 0.05000

K_POINTS automatic
 1 1 1 0 0 0



Alejandro Suarez
Graduate Student
Department of Physics
Penn State University
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20110930/ecd25a70/attachment.html>


More information about the users mailing list