[Pw_forum] vc-relax calculation
Julen Larrucea
julenlist at gmail.com
Mon Nov 11 12:29:16 CET 2013
Dear Ehsan,
As Sanjeev pointed out, it is a good to attach your error message when
asking for help.
After running your example, I get the following error message:
.....................................................................................................
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Error in routine read_namelists (1):
reading namelist cell
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
stopping ...
.....................................................................................................
Which raises due to the fact that you are performing a cell relax
calculation, but you have not included the &CELL section (an empty one is
OK too). You can skip it for a normal "relax" or "scf" calculation, but not
for "vc-relax".
So modify your input as:
.....................................................................................................
...
&IONS
/
&CELL
/
CELL_PARAMETERS {angstrom}
9.83804859 0.00000000 0.00000000
4.91902429 8.52000000 0.00000000
0.00000000 0.00000000 12.00000000
...
.....................................................................................................
Let me add you a couple of suggestions for your input.
- Your conv_thr is way too big. 1.D-6 should be enough for a relaxation,
and if you want higher accuracy for the electronic properties, just run an
scf on the last structure with 1.D-8.
- When you run a cell relaxation, you should use about 3 times the
recommended ecutwfc and ecutrho, so yours is way too small in any case.
- The pseudo potential file in the QE website you are using does not
provide any suggestions for minimum ecutwfc and ecutrho, so I just used a
simple script to check it:
-10.7675 ++-----+------+-------+------+------+------+-------+------+-----++
+ + './energies.ecutwfc.C.pw91-van_ak.UPF' u 1:2
****** +
**** |
-10.768 ++ **
++
|
* |
-10.7685 ++ **
++
|
* |
|
** |
-10.769 ++ *
++
|
* |
|
|
-10.7695 ++
**** ++
|
*** |
|
* |
-10.77 ++
************** ++
|
******* |
-10.7705 ++
******** ++
|
******* |
+ + + + + + + +
********
-10.771 ++-----+------+-------+------+------+------+-------+------+-----++
30 40 50 60 70 80 90 100 110
120
(It does not look very clear, but you can run this script by yourself and
check it in better quality, it only takes a few minutes to run.
http://www.larrucea.eu/checking-optimum-cutoff-qe/)
So, you should use an ecutwfc of around 110 Ry for your normal calculations
and about 330 Ry for the cell relaxations.
Best regards
Julen
On Mon, Nov 11, 2013 at 2:18 AM, Sanjeev Gupta <physics.skgupta at gmail.com>wrote:
> can you post the error?
>
> bests
> sanjeev
>
>
> On Sun, Nov 10, 2013 at 11:43 AM, ehsan targholi <targholi at gmail.com>wrote:
>
>>
>> *hi*
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> *dear all, thanks for reply to my earlier question. i want to optimize
>> graphene , but i have problem in vc-relax calculation. my input is :
>> &control calculation = 'vc-relax' , restart_mode='from_scratch'
>> prefix='gs32' wf_collect =.true pseudo_dir =
>> '/home/ehsan/espresso/upf_files/', outdir
>> ='/home/ehsan/espresso/tmp/'/&system ibrav=0 nat=32 ntyp=1 nbnd=128
>> ecutwfc=45 ecutrho=450/&electrons mixing_beta=0.7 conv_thr = 1.0e-12
>> /&IONS/CELL_PARAMETERS {angstrom} 9.83804859 0.00000000
>> 0.00000000 4.91902429 8.52000000 0.00000000 0.00000000
>> 0.00000000 12.00000000ATOMIC_SPECIESC 12.011 C.pw91-van_ak.UPF
>> ATOMIC_POSITIONS {angstrom} C 3.07439018 1.77499927
>> 0.00160788 C 0.61487804 0.35499985
>> 0.00032158 C 5.53390233 1.77499927
>> 0.00160788 C 7.99341448 1.77499927 0.00160788
>> C 10.45292662 1.77499927
>> 0.00160788 C 3.07439018 0.35499985
>> 0.00032158 C 5.53390233 0.35499985
>> 0.00032158 C 7.99341448 0.35499985 0.00032158
>> C 4.30414626 3.90499840
>> 0.00353734 C 5.53390233 6.03499752
>> 0.00546680 C 6.76365840 8.16499665
>> 0.00739626 C 1.84463411 2.48499898 0.00225103
>> C 3.07439018 4.61499811
>> 0.00418049 C 4.30414626 6.74499723
>> 0.00610995 C 6.76365840 3.90499840
>> 0.00353734 C 7.99341448 6.03499752 0.00546680
>> C 9.22317055 8.16499665
>> 0.00739626 C 9.22317055 3.90499840
>> 0.00353734 C 10.45292662 6.03499752
>> 0.00546680 C 11.68268270 8.16499665 0.00739626
>> C 11.68268270 3.90499840
>> 0.00353734 C 12.91243877 6.03499752
>> 0.00546680 C 14.14219484 8.16499665
>> 0.00739626 C 4.30414626 2.48499898 0.00225103
>> C 5.53390233 4.61499811
>> 0.00418049 C 6.76365840 6.74499723
>> 0.00610995 C 6.76365840 2.48499898
>> 0.00225103 C 7.99341448 4.61499811 0.00418049
>> C 9.22317055 6.74499723
>> 0.00610995 C 9.22317055 2.48499898
>> 0.00225103 C 10.45292662 4.61499811
>> 0.00418049 C 11.68268270 6.74499723 0.00610995
>> K_POINTS {automatic}16 16 1 0 0 0*
>>
>>
>> *but espresso give error. please help me to insert correct input for
>> variable relax calculation. *
>>
>> *ehsan targholi *
>>
>> *graduate student of chemistry department of iust *
>>
>>
>> _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://pwscf.org/mailman/listinfo/pw_forum
>>
>
>
>
> --
> With Best Regards,
>
> ------------------------------------
> Dr. Sanjeev Kumar Gupta
> Fulbright Post-Doctoral Scholar
> Dept. of Physics
> Michigan Technological University
> 1400 Townsend Drive, Houghton
> MI 49931, USA
> ------------------------------------
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum
>
--
--
Dr. Julen Larrucea
Postdoctoral researcher,
BCCMS, HMI Group, University of Bremen
Phone: +49 421 218 64582
Fax: +49 421 218 64599
http://www.larrucea.eu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20131111/493bc9fb/attachment.html>
More information about the users
mailing list