[Pw_forum] vc-relax doesn't seem to converged.
Ari P Seitsonen
Ari.P.Seitsonen at iki.fi
Tue May 24 11:02:19 CEST 2016
Dear Amir,
Well, rather the other way around: 1 2 2 in the k points, as the longer
the cell in real space, the fewer the k points needed in the corresponding
accuracy in the reci-procal space, or, the integration over the Brillouin
zone? Why? Exactly because it is the _reci-procal_ space (the longer the
real-space dimension, the shorter the one in reci-procal space, so one
needs less sampling points == k points to achieve the same
sampling distance in the integrand).
Well, I just mentioned LDA, because my first thought was that it is so
old, why not going for the GGA's plus possibly a correction for the
missing London dispersions, but at the second thought LDA is known to give
quite good results in the bulk, and it is less of the art of "fitting" the
"correct" GGA to obtain good results. Some results on the LDA versus GGA
versus other schemes were recently performed by Colleagues (my own
contribution was small but hopefully positive, ie non-destructive) in
Henri Hay et alia, Phys. Rev. B 92, 144111 DOI: 10.1103/PhysRevB.92.144111
Maybe this gives you some thoughts. I guess that if going for bulk-like
properties the GGA functionals such as PBEsol ("PBE-for-solids") might be
quite good results. It is up to you, indeed LDA might give reasonable
results, without all this fit-to-be-ab-initio-considerations of choosing
the functional to use. :) (wasntme) - just my personal opinion.
Greetings,
apsi
-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-
Ari Paavo Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/
Ecole Normale Supérieure (ENS), Département de Chimie, Paris
Mobile (F) : +33 789 37 24 25 (CH) : +41 79 71 90 935
On Mon, 23 May 2016, Mofrad, Amir Mehdi (MU-Student) wrote:
> Thank you Ari for your reply. Yes you were right about the number of atoms. It should be 72. I figured that out too right after I sent out the email. However, I wanted to ask about what you said at the end of your reply. So do you think 3 2 2 or even 4 2 2 would be good enough to do the optimization? Also, would you please clarify a bit on the LDA part you mentioned?
>
>
> Best,
>
> Amir M. Mofrad
> Graduate Research Assistant
> Chemical Engineering Department
> University of Missouri
>
> ________________________________________
> From: pw_forum-bounces at pwscf.org <pw_forum-bounces at pwscf.org> on behalf of Ari P Seitsonen <Ari.P.Seitsonen at iki.fi>
> Sent: Monday, May 23, 2016 7:18:16 AM
> To: PWSCF Forum
> Subject: Re: [Pw_forum] vc-relax doesn't seem to converged.
>
> Dear Amir,
>
> Did you check your geometry, for example visually (XCrysDen, VESTA,
> ...)? To me it looks as if there are some Si atoms missing in the
> structure, and thus there are dangling bonds on oxygens. Was this your
> purpose?
>
> In principle you can later change your k point sampling in the direction
> corresponding to the long axis of the unit cells. LDA, hmm, well if you
> want. :)
>
> Greetings from Paris,
>
> apsi
>
> -=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-
> Ari Paavo Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/
> Ecole Normale Supérieure (ENS), Département de Chimie, Paris
> Mobile (F) : +33 789 37 24 25 (CH) : +41 79 71 90 935
>
>
> On Thu, 19 May 2016, Mofrad, Amir Mehdi (MU-Student) wrote:
>
>>
>> Dear all QE users and developers,
>>
>>
>> So far I have been doing scf calculations on one unit cell of a specific type of zeolite (which is called SOD) and I have been
>> getting some good results. However, recently I wanted to extend my calculations on more than one unit cell (2x1x1 for
>> instance). The problem is when I do vc-relax it does not optimize the cell nor the internal coordinates of atoms. I put my
>> input file in case you might need to take a look at. Any help would be thoroughly appreciated.
>>
>>
>> &CONTROL
>> calculation = 'vc-relax' ,
>> restart_mode = 'from_scratch' ,
>> wf_collect = .true. ,
>> outdir = './scratch' ,
>> wfcdir = './scratch' ,
>> pseudo_dir = '/global/espresso/pseudo' ,
>> prefix = 'SOD' ,
>> verbosity = 'high' ,
>> etot_conv_thr = 1e-5 ,
>> forc_conv_thr = 1e-4 ,
>> nstep = 50 ,
>> tstress = .true. ,
>> tprnfor = .true. ,
>>
>> /
>>
>> &SYSTEM
>> ibrav = 0,
>> nat = 70,
>> ntyp = 2,
>> ecutwfc = 31 ,
>> ecutrho = 310 ,
>>
>>
>> /
>> &ELECTRONS
>> electron_maxstep = 100,
>> conv_thr = 3e-8 ,
>> mixing_mode = 'plain' ,
>> mixing_beta = 0.7 ,
>> diagonalization = 'david' ,
>> /
>> &IONS
>> ion_dynamics = 'bfgs' ,
>> trust_radius_ini = 0.5 ,
>> /
>> &CELL
>> cell_dynamics='bfgs',
>> cell_factor=5,
>> /
>>
>> CELL_PARAMETERS angstrom
>> 18.0000 0.00000 0.00000
>> 0.00000 9.00000 0.00000
>> 0.00000 0.00000 9.00000
>> ATOMIC_SPECIES
>> Si 28.08600 Si.pz-n-rrkjus_psl.0.1.UPF
>> O 15.99940 O.pz-n-rrkjus_psl.0.1.UPF
>>
>> ATOMIC_POSITIONS angstrom
>> O 1.31516 4.48250 7.64983
>> Si 2.24125 4.48250 0.00000
>> O 4.48250 1.31517 7.64983
>> O 7.64983 4.48250 7.64983
>> O 4.48250 7.64983 7.64983
>> O 1.31517 4.48250 1.31517
>> O 4.48250 1.31517 1.31517
>> O 7.64983 4.48250 1.31517
>> O 4.48250 7.64983 1.31517
>> O 7.64983 1.31517 4.48250
>> O 7.64983 7.64983 4.48250
>> O 1.31516 7.64983 4.48250
>> O 1.31517 1.31517 4.48250
>> O 5.79767 -0.00000 3.16733
>> O -0.00000 5.79767 3.16733
>> O 3.16733 -0.00000 3.16733
>> O -0.00000 3.16733 3.16733
>> O 5.79767 -0.00000 5.79767
>> O -0.00000 5.79767 5.79767
>> O 3.16733 -0.00000 5.79767
>> O -0.00000 3.16733 5.79767
>> O 3.16733 5.79767 0.00000
>> O 3.16733 3.16733 0.00000
>> O 5.79767 3.16733 0.00000
>> O 5.79767 5.79767 0.00000
>> Si 4.48250 2.24125 0.00000
>> Si 6.72375 4.48250 0.00000
>> Si 4.48250 6.72375 0.00000
>> Si -0.00000 2.24125 4.48250
>> Si 6.72375 -0.00000 4.48250
>> Si -0.00000 6.72375 4.48250
>> Si 2.24125 -0.00000 4.48250
>> Si 4.48250 -0.00000 2.24125
>> Si -0.00000 4.48250 2.24125
>> O 10.28017 4.48250 7.64983
>> Si 11.20625 4.48250 0.00000
>> O 13.44750 1.31517 7.64983
>> O 16.61483 4.48250 7.64983
>> O 13.44750 7.64983 7.64983
>> O 10.28017 4.48250 1.31517
>> O 13.44750 1.31517 1.31517
>> O 16.61483 4.48250 1.31517
>> O 13.44750 7.64983 1.31517
>> O 16.61483 1.31517 4.48250
>> O 16.61483 7.64983 4.48250
>> O 10.28017 7.64983 4.48250
>> O 10.28017 1.31517 4.48250
>> O 14.76267 -0.00000 3.16733
>> O 8.96500 5.79767 3.16733
>> O 12.13233 -0.00000 3.16733
>> O 8.96500 3.16733 3.16733
>> O 14.76267 -0.00000 5.79767
>> O 8.96500 5.79767 5.79767
>> O 12.13233 -0.00000 5.79767
>> O 8.96500 3.16733 5.79767
>> O 12.13233 5.79767 0.00000
>> O 12.13233 3.16733 0.00000
>> O 14.76267 3.16733 0.00000
>> O 14.76267 5.79767 0.00000
>> Si 13.44750 2.24125 0.00000
>> Si 15.68875 4.48250 0.00000
>> Si 13.44750 6.72375 0.00000
>> Si 8.96500 2.24125 4.48250
>> Si 15.68875 -0.00000 4.48250
>> Si 8.96500 6.72375 4.48250
>> Si 11.20625 -0.00000 4.48250
>> Si 13.44750 -0.00000 6.72375
>> Si 8.96500 4.48250 6.72375
>> Si 13.44750 -0.00000 2.24125
>> Si 8.96500 4.48250 2.24125
>>
>> K_POINTS automatic
>> 2 2 2 1 1 1
>>
>> Best,
>>
>>
>> Amir M. Mofrad
>>
>> University of Missouri
>>
>>
>>
>
More information about the users
mailing list