[Pw_forum] geometry optiiztion of metal clusters

[Martin Andersson]

Hi,

You don't need k-points with a cluster, so that should speed up things 
by a factor of 64. If you also use k-points gamma instead of automatic 1 
1 1 0 0 0 you can gain some more as well. If you want further speed up 
you can put tqr=.true. under &electrons because you are using ultrasoft 
pseudopotentials. If you are able to reduce the cell size (which you 
have to check anyway) you can also gain significant time.

Cheers,
Martin Andersson
University of Copenhagen

On 3/28/2011 10:19 PM, Wohlwend, Jennifer L CTR USAF AFMC AFRL/RXBT wrote:
>
> I'm trying to perform geometry optimization of someclusters andthey 
> arerunning extremely slow. Forexamplea TiO2 anatase cluster of only 12 
> atoms has completedonly a fewself-consistentiterationsafter ~3 hours. 
> IknowI must be doing something wrong but I've not been able to spot it 
> yet.For the TiO2 system, myinput is as follows:
>
> &CONTROL
>
>                        title = Ti_relax ,
>
>                  calculation = 'relax' ,
>
>                 restart_mode = 'restart' ,
>
>                       outdir = './' ,
>
>                       wfcdir = './' ,
>
>                   pseudo_dir = '/hafs15/jen7182/espresso/pseudo/' ,
>
>                  lkpoint_dir = .true. ,
>
>                etot_conv_thr = 1.0D-6 ,
>
>                forc_conv_thr = 1.0D-6 ,
>
>                        nstep = 120 ,
>
>                      tstress = .true. ,
>
>                      tprnfor = .true. ,
>
>                       lberry = .false. ,
>
>  /
>
> &SYSTEM
>
>                        ibrav = 1,
>
>                    celldm(1) = 50.0000000,
>
>                          nat = 12,
>
>                         ntyp = 2,
>
>                      ecutwfc = 55 ,
>
>                      ecutrho = 500 ,
>
>                  occupations = 'smearing' ,
>
>                      degauss = 0.02D0 ,
>
>                     smearing = 'methfessel-paxton' ,
>
>  /
>
> &ELECTRONS
>
>                     conv_thr = 1.D-8 ,
>
> /
>
> &IONS
>
>                 ion_dynamics = 'bfgs' ,
>
>   pot_extrapolation = "second_order",
>
>   wfc_extrapolation = "second_order",
>
>  /
>
> ATOMIC_SPECIES
>
>     Ti   47.87000  Ti.pw91-nsp-van.UPF
>
> 15.99960  O.pw91-van_ak.UPF
>
> ATOMIC_POSITIONS
>
> Ti      0.019646921155  -0.032104589149 0.072982942110
>
> Ti      -0.053021298671 0.015633269982  0.132549510485
>
> Ti      0.057125188587  0.058771047647  0.164020158511
>
> Ti      0.110388124153  0.105884441223  0.244068937730
>
> O       -0.023076017443 -0.044401500581 0.131365372724
>
> O       0.035579202732  -0.080351647885 0.029496520030
>
> O       -0.037172544158 0.018329722652  0.066778689936
>
> O       0.057973293750  0.014983392566  0.111654528302
>
> O       -0.014261394120 0.057693305910  0.168967680139
>
> O       0.116729745829  0.144794684381  0.294813978638
>
> O       0.083638150059  0.040575025164  0.225871612978
>
> O       0.091668444256  0.117426432280  0.178170567176
>
> K_POINTS automatic
>
> 4 4 4   1 1 1
>
> Any help would be greatly appreciated!!
>
> Thank you!
>
> Jennifer
>
> ***Jennifer L. Wohlwend, Ph.D.*
>
> Thermal Sciences and Materials Branch
>
> Materials and Manufacturing Directorate
>
> Air Force Research Laboratory
>
> 2941 Hobson Way B654 R015
>
> Wright Patterson Air Force Base, OH 45433-7750
>
> Ctr: Universal Technology Corporation
>
> Phone: (937) 656-4918
>
> Email:jennifer.wohlwend.ctr at wpafb.af.mil<mailto:jennifer.wohlwend.ctr at wpafb.af.mil>
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20110328/3721cdf7/attachment.html>