[Pw_forum] vc-relax question

[Samin, Adib J.]

Dear Quantum Espresso,


I am trying to run a test simulation on the silicon unit cell using vc-relax. When I run the simulation using "ibrav=2" the simulation converges in 4 scf cycles. However, when I attempt to run the simulation using "ibrav = 0" it doesn't converge in 18 scf cycles. Moreover, I always receive the error that I need to use a larger cell factor. Finally, and when using a larger cell factor, the simulation runs longer and doesn't converge and also gives me the same error again.

I am using this test because I will need to run vc-relax in a situation where I will need to use "ibrav=0" so it is important to me that I get it to work properly.

I have attached my input file below. Any clarifications on this issue would be greatly appreciated.


 &control
    calculation     =  'vc-relax'
    restart_mode    =  'from_scratch'
    prefix          =  'si'
    tprnfor         =  .true.
    wf_collect      =  .true.
    pseudo_dir      =  '/usr/local/espresso/espresso-5.0.3/pseudo/'
    outdir          =  '/fs/lustre/osu7834/si/'
/
 &system
    ibrav           =  0
    nat             =   2
    ntyp            =   1
    ecutwfc         =  25.0
/
 &electrons
    conv_thr        =   1.0d-7
    diagonalization =  'cg'
/
&IONS
/
&CELL
cell_factor = 1.6
/
CELL_PARAMETERS {bohr}
10.40 0.00 0.00
0.00 10.40 0.00
0.00 0.00 10.40
ATOMIC_SPECIES
 Si  28  Si.pbe-n-van.UPF
ATOMIC_POSITIONS {crystal}
Si  -0.25   0.75  -0.25
Si   0.00   0.00    0.00
K_POINTS {automatic}
4 4 4 0 0 0?



+++++++++++++++++++++++++++++++++

Adib Samin, PhD

Postdoctoral Researcher

The Department of Aerospace and Mechanical Engineering

The Ohio State University


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20141126/e608608d/attachment.html>