[Pw_forum] [EXTERNAL] Biaxial Strian Silicon

[Zimmerman, Jonathan A]

Hi Cameron,

I'm a relatively new user of PWscf, but I'll take a try to help you out. I see two potential problems with your setup:

  1.  I'm not sure that you have the correct ATOMIC_POSITIONS and CELL_PARAMETERS, especially for the ibrav=2 structure type. If you examine the online documentation (http://www.quantum-espresso.org/wp-content/uploads/Doc/INPUT_PW.html), you'll notice that the lattice directions are not the <100>/cubic directions (ibrav=1), but rather <110>-type directions. Thus, I'm not sure that the values you've listed are correct. In particular, it's odd that a few of your CELL_PARAMETERS values are negative. I'm not sure what that's supposed to mean.
  2.  Secondly, I believe the '2Dxy' specification for cell_dofree is opposite to your intent. I believe you want to specify, and then fix during relaxation, the x-y dimensions of your system and only allow z to change. If that's the case, then the correct option would be cell_dofree=z.

Good luck,

Jon Zimmerman

From: Cameron Foss <cjfoss at umass.edu<mailto:cjfoss at umass.edu>>
Reply-To: PWSCF Forum <pw_forum at pwscf.org<mailto:pw_forum at pwscf.org>>
Date: Thursday, August 14, 2014 at 2:28 PM
To: "pw_forum at pwscf.org<mailto:pw_forum at pwscf.org>" <pw_forum at pwscf.org<mailto:pw_forum at pwscf.org>>
Subject: [EXTERNAL] [Pw_forum] Biaxial Strian Silicon

Hello all,

I am trying to study phonons in silicon under biaxial strain, which occurs when you grow thin films of silicon over a germanium(or SiGe) bulk, which has a slightly larger lattice constant than silicon, thus stretching the silicon out. This type of procedure stretches silicon in the xy directions and slightly compresses it in the z direction. This consequently distorts the symmetry of the crystal.

I have ran some strained cases; in which I "strain" the crystal by changing the lattice vectors in the CELL_PARAMETERS CARD and the atomic positions like so:
ATOMIC_POSITIONS {alat}
Si      0       0       0
Si      0.255   0.255   0.25
CELL_PARAMETERS {alat}
  -0.51     0.00    0.5
   0.00     0.51    0.5
  -0.51     0.51    0.00

Whilst the lattice constant remained the equilibrium lattice constant 5.431Angs.

However, I would like to use the vc-relax calculation to relax the structure to the strained amount...so I know that the structure is optimized. Although I keep getting an orthogonal operation error.

Here is the vc-relax input code and the error:
&control
    calculation='vc-relax'
    restart_mode='from_scratch',
    outdir='/home/cameron/QE/espresso-5.1/wktmp',
    pseudo_dir='/home/cameron/QE/espresso-5.1/pseudo',
    prefix='relax'
/
&system
    ibrav=2, celldm(1)=10.263, nat=2, ntyp=1,
    ecutwfc =24.0
/
&electrons
    conv_thr =  1.0d-10
    mixing_beta = 0.4
/
&ions
    ion_dynamics='damp'
/
&cell
    cell_dynamics='damp-w',
    press=55.0,
    cell_dofree='2Dxy'
/
ATOMIC_SPECIES
 Si  28.0855 Si.pz-vbc.UPF
ATOMIC_POSITIONS
 Si 0.00 0.00 0.00
 Si 0.25 0.25 0.25
K_POINTS AUTOMATIC
4 4 4 1 1 1

 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     Error in routine checkallsym (2):
     not orthogonal operation
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

I have two main questions: Did I incorrectly manually "strain" the lattice structure in the first case?
And how can I use/fix the vc-relax code to simulate the strained structure?

Best,
Cameron
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20140814/71a038e0/attachment.html>