[QE-users] 回复: scf not converge in neb.x caluculation
LEUNG Clarence
liangxy123 at hotmail.com
Thu Mar 21 12:43:17 CET 2019
Dear Giuseppe,
Thanks very much for your suggestion.
________________________________
发件人: users <users-bounces at lists.quantum-espresso.org> 代表 Giuseppe Mattioli <giuseppe.mattioli at ism.cnr.it>
发送时间: 2019年3月21日 18:00
收件人: users at lists.quantum-espresso.org
主题: Re: [QE-users] scf not converge in neb.x caluculation
Dear Liang Xiongyi
There are a lot of things you can try, but neb.x can be somewhat whimsical...
1) You can change the total number of images. This will change the
starting points and might ensure a smoother convergence.
2) path_thr = 0.05D0
This does not help convergence, but it is a bit tight in my
experience. 0.1 is generally tight enough (in reasonably complex
systems) to provide meaningful results.
3) degauss = 0.005
This is very tight and can hinder convergence. You should try to use
0.01~0.02 (check how much it affects total energies of images by
looking at the -TS contribution in the PW.out files)
4) mixing_beta=0.7d0
You can lower this even down to 0.01 (but 0.1 is generally a good
value for not extremely stubborn convergence).
5) you are using a default electron_maxstep=100 value. You can raise
this value, or you can raise (a littele bit) conv_thr. You may also
carefully use scf_must_converge=.false., specially if scf cycles dance
close to convergence and you want to force the system to step across a
problematic atomic configuration.
6) K_POINTS {automatic}
5 5 1 0 0 0
This might be too much for calculation of barriers in a quite large
supercell having no symmetry. For example, I would check if the total
energy difference between the converged final and initial images was
(almost) constant by using K_POINTS {gamma}. If this was true, then I
would calculate the whole path by using gamma_only tricks, which
should also provide better convergence for the system.
HTH
Giuseppe
LEUNG Clarence <liangxy123 at hotmail.com> ha scritto:
> Dear QE users,
>
> I want to calculate the mobility of metal atom on surface bu neb.x.
> However, the image 9 is very difficult to converge. Could you please
> give me some advice. The input file as follows:
>
> Thanks.
> LIANG XIongyi
> City University of Hong Kong
>
> BEGIN
> BEGIN_PATH_INPUT
> &PATH
> restart_mode = 'from_scratch'
> string_method = 'neb',
> nstep_path = 30,
> ds = 1.D0,
> opt_scheme = "broyden",
> num_of_images = 10,
> k_max = 0.3D0,
> k_min = 0.2D0,
> path_thr = 0.05D0,
>
> /
> END_PATH_INPUT
>
> BEGIN_ENGINE_INPUT
> &CONTROL
> prefix='av-y',
> outdir='/home/qeuser/Clarence/phosphorene/v/d3/av/neb',
> pseudo_dir='/home/qeuser/SSSP_acc_PBE' ,
> verbosity='low',
> etot_conv_thr = 1.0D-5 ,
> forc_conv_thr = 6.0D-4 ,
> nstep = 200 ,
> wf_collect = .true. ,
> verbosity='low',
> /
> &SYSTEM
> ibrav=8,
> celldm(1)=25.0298004651d0, celldm(2)=1.3810663486d0,
> celldm(3)=1.8874762178d0,
> nat=65,
> ntyp=2,
> ecutwfc=50,
> ecutrho=500,
> input_dft='PBE',
> occupations='smearing',
> smearing = 'gaussian' ,
> degauss = 0.005 ,
> nspin = 2 ,
> starting_magnetization(1) = 0.5 ,
> vdw_corr = 'DFT-D3' ,
> /
>
> &ELECTRONS
> conv_thr=1d-06,
> mixing_beta=0.7d0,
> mixing_mode ='local-TF',
> electron_maxstep = 1000 ,
> /
> &IONS
> /
>
> ATOMIC_SPECIES
> P 30.973800d0 P.pbe-n-rrkjus_psl.1.0.0.UPF
> V 50.941500d0 V_pbe_v1.uspp.F.UPF
>
> K_POINTS {automatic}
> 5 5 1 0 0 0
>
> BEGIN_POSITIONS
> FIRST_IMAGE
> ATOMIC_POSITIONS {crystal}
> P 0.062364056 0.146835680 0.461004179
> P 0.187979850 0.227571721 0.458486567
> P 0.187359072 0.023967396 0.547448145
> P 0.062390271 0.105486489 0.546314111
> P 0.062391562 0.397230966 0.452233415
> P 0.187453676 0.479039381 0.452606421
> P 0.189364962 0.278117380 0.541172586
> P 0.062407749 0.357692419 0.537875656
> P 0.062401956 0.644403307 0.461059945
> P 0.188082333 0.725080248 0.462700469
> P 0.188395810 0.514757114 0.539192335
> P 0.062460702 0.595965919 0.543895751
> P 0.062381001 0.895525242 0.463867650
> P 0.186790459 0.977279845 0.463560403
> P 0.187798718 0.768448197 0.547468271
> P 0.062423049 0.849704828 0.548098847
> P 0.312659868 0.146168383 0.460453461
> P 0.437781276 0.227744221 0.458426404
> P 0.437455901 0.023776880 0.547310817
> P 0.312549613 0.105320407 0.545904997
> P 0.312139598 0.397030501 0.450854221
> P 0.438074098 0.477737365 0.445066729
> P 0.439912044 0.279485733 0.540862256
> P 0.314542445 0.359645859 0.537231792
> P 0.314544837 0.644905805 0.460878636
> P 0.434594009 0.730077067 0.459781958
> P 0.448827043 0.514927416 0.532500900
> P 0.317525398 0.592780735 0.543968174
> P 0.311728367 0.896097058 0.463720553
> P 0.436893865 0.977260819 0.463347579
> P 0.437860024 0.769032959 0.545891816
> P 0.312518709 0.850247622 0.547849195
> P 0.562393332 0.146370379 0.460376779
> P 0.686969347 0.227774704 0.458451805
> P 0.687291897 0.023755188 0.547362489
> P 0.562382204 0.105350763 0.545849427
> P 0.562473269 0.396448431 0.443507308
> P 0.686748477 0.477847993 0.445052781
> P 0.684946773 0.279498604 0.540893439
> P 0.562452036 0.365431435 0.533574680
> P 0.562388294 0.651473527 0.453514755
> P 0.690184909 0.730079028 0.459780328
> P 0.676015132 0.514980165 0.532495771
> P 0.562386005 0.604919173 0.537701074
> P 0.562374845 0.896356486 0.463525401
> P 0.687848329 0.977264329 0.463389247
> P 0.686894465 0.769048044 0.545881597
> P 0.562366018 0.850830890 0.547709669
> P 0.812039406 0.146172639 0.460500457
> P 0.936731753 0.227556339 0.458506445
> P 0.937413894 0.023975028 0.547450256
> P 0.812195810 0.105311249 0.545951465
> P 0.812689390 0.397151038 0.450847120
> P 0.937418545 0.479112680 0.452648391
> P 0.935397084 0.278168840 0.541173905
> P 0.810251826 0.359711909 0.537223718
> P 0.810287195 0.644943491 0.460914925
> P 0.936753166 0.725104783 0.462754832
> P 0.936455552 0.514807673 0.539237217
> P 0.807300389 0.592829495 0.544001475
> P 0.813022989 0.896104987 0.463696223
> P 0.937973444 0.977284048 0.463557949
> P 0.937000660 0.768483060 0.547514207
> P 0.812240625 0.850260753 0.547823269
> V 0.562457425 0.460038104 0.589861636
> LAST_IMAGE
> ATOMIC_POSITIONS {crystal}
> P 0.062445615 0.146456673 0.451477017
> P 0.187505780 0.228300342 0.450996126
> P 0.189411819 0.029049537 0.541552681
> P 0.062449177 0.108536309 0.537537957
> P 0.062447760 0.395196377 0.458733376
> P 0.188131570 0.475920562 0.460828460
> P 0.188374349 0.265401889 0.537271427
> P 0.062442541 0.346561452 0.541532408
> P 0.062445852 0.646720019 0.463330906
> P 0.186881361 0.728456886 0.463846544
> P 0.187786345 0.518231310 0.545890007
> P 0.062453053 0.599463807 0.547113520
> P 0.062449884 0.896772437 0.462088203
> P 0.188068030 0.977403357 0.459227449
> P 0.187407760 0.773919686 0.548121428
> P 0.062451045 0.855509144 0.547403030
> P 0.312157931 0.146197729 0.450100263
> P 0.437962144 0.226895792 0.443334809
> P 0.439781032 0.030364618 0.541401226
> P 0.314537903 0.110579019 0.536930990
> P 0.314516246 0.395673292 0.458770827
> P 0.434514870 0.480945643 0.457958047
> P 0.448849945 0.265612968 0.530384087
> P 0.317545829 0.343415876 0.541814557
> P 0.311882120 0.647337595 0.463336957
> P 0.437025970 0.728486497 0.463886605
> P 0.437893522 0.518734801 0.544383856
> P 0.312565523 0.599936087 0.546984686
> P 0.312792214 0.896123190 0.461534507
> P 0.437703418 0.977802230 0.459342638
> P 0.437537090 0.773854879 0.548208275
> P 0.312529860 0.855396994 0.547005515
> P 0.562445549 0.145582412 0.442645546
> P 0.686938850 0.226884972 0.443352934
> P 0.685108766 0.030376879 0.541401741
> P 0.562439931 0.116211878 0.533161816
> P 0.562454842 0.402446010 0.451051275
> P 0.690383198 0.480958602 0.457954855
> P 0.676029527 0.265601905 0.530394460
> P 0.562442494 0.355543724 0.535081296
> P 0.562448326 0.647583315 0.463280787
> P 0.687873726 0.728484011 0.463890075
> P 0.687007439 0.518737405 0.544382783
> P 0.562448794 0.600455024 0.546972980
> P 0.562448881 0.896623478 0.461579252
> P 0.687191466 0.977803753 0.459347311
> P 0.687350603 0.773851566 0.548211631
> P 0.562445632 0.855549232 0.547048125
> P 0.812742443 0.146187679 0.450102692
> P 0.937382360 0.228298381 0.450991288
> P 0.935497478 0.029039399 0.541554721
> P 0.810375354 0.110570352 0.536929886
> P 0.810378355 0.395682293 0.458771630
> P 0.936769992 0.475925052 0.460830172
> P 0.936505251 0.265408800 0.537265660
> P 0.807335019 0.343422437 0.541813266
> P 0.813017687 0.647336128 0.463349568
> P 0.938011871 0.728458147 0.463853218
> P 0.937122704 0.518229776 0.545892176
> P 0.812342101 0.599934329 0.546995984
> P 0.812096929 0.896122158 0.461548112
> P 0.936825737 0.977397364 0.459230552
> P 0.937491966 0.773921441 0.548127880
> P 0.812360758 0.855393535 0.547018412
> V 0.562433776 0.211933054 0.588349784
> END_POSITIONS
> END_ENGINE_INPUT
> END
GIUSEPPE MATTIOLI
CNR - ISTITUTO DI STRUTTURA DELLA MATERIA
Via Salaria Km 29,300 - C.P. 10
I-00015 - Monterotondo Scalo (RM)
Mob (*preferred*) +39 373 7305625
Tel + 39 06 90672342 - Fax +39 06 90672316
E-mail: <giuseppe.mattioli at ism.cnr.it>
_______________________________________________
users mailing list
users at lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20190321/9ca3be6e/attachment.html>
More information about the users
mailing list