very slow electronic convergence for FeCrNi

[peterklaver]

Hello people,

I've recently started looking at some FeCrNi calculations and I'm encountering pretty bad convergence trouble.

The system I'm looking at is 4x4x4 fcc unit cells, 256 atoms, 180 Fe, 50 Cr, 26 Ni, in and SQS ordering. With IALGO = 38 and otherwise also fairly standard settings, convergence required a restart from WAVECAR after 60 electronic iterations wasn't enough. Input files and OSZICARs and OUTCARs are at
http://www.ph.ed.ac.uk/~pklaver/FeCrNi.tar.gz , part of the OSZICAR for the restart run is at the end of this post

I tried various mixing parameter combinations to improve things:

AMIX = 0.3, BMIX = 0.001
AMIX = 0.3, BMIX = 0.0001
AMIX = 0.3, BMIX = 0.0001, AMIX_MAG = 0.8 BMIX_MAG = 0.0001

Unfortunately, as far as these had any effect, they made it worse, sometimes killing convergence altogether, even if I set NELM to 150.

I also tried different setups for the initial magnetic moments through the MAGMOM line. The configuration for which the in/output files are linked above has alternating anti-ferromagnetic <100> monolayers. Apart from that I also tried the double <100> layer afm setup and single layer <111> setup. The double <100> layer and single <111> configurations didn't converge, instead they fluctuated around energy values higher than the value to which the single <100> layer system converged. Some of these attempts did come close to convergence, so I guess they might well be some non-ground state solutions that VASP was getting close to.

Given that the system is certainly not your average clean lab system with just one or two alloy atoms in otherwise pure Fe, the existence of several magnetic states, etc I guess this would be a difficult task for VASP to do. Is the large number of iterations unavoidable and should I just pay the high computational price for picking such a system, or is there a way to make them converge better, like some mixing parameter combination for anti-ferromagnetic systems? Or some other solution?

If you have any idea then please let me know.

greets,
Peter


N E dE d eps ncg rms rms(c)
DAV: 1 -0.206706489284E+04 -0.20671E+04 -0.10630E+00 9600 0.348E+00 0.941E+00
DAV: 2 -0.213388741284E+04 -0.66823E+02 -0.11571E+02 9264 0.420E+01 0.111E+02
DAV: 3 -0.206873297501E+04 0.65154E+02 -0.88095E+01 8672 0.395E+01 0.261E+01
DAV: 4 -0.206653276652E+04 0.22002E+01 -0.20226E+01 9520 0.103E+01 0.575E+00
DAV: 5 -0.206668985489E+04 -0.15709E+00 -0.33615E+00 8656 0.491E+00 0.593E+00
DAV: 6 -0.206675785737E+04 -0.68002E-01 -0.12497E+00 7584 0.404E+00 0.611E+00
DAV: 7 -0.206658412833E+04 0.17373E+00 -0.65329E-01 7360 0.313E+00 0.299E+00
DAV: 8 -0.206654938818E+04 0.34740E-01 -0.22631E-01 8416 0.150E+00 0.130E+00
DAV: 9 -0.206655310797E+04 -0.37198E-02 -0.63108E-02 9648 0.713E-01 0.134E+00
DAV: 10 -0.206655116010E+04 0.19479E-02 -0.28399E-02 8592 0.595E-01 0.109E+00
DAV: 11 -0.206654870951E+04 0.24506E-02 -0.14585E-02 8624 0.432E-01 0.811E-01
DAV: 12 -0.206654900415E+04 -0.29464E-03 -0.73765E-03 9184 0.251E-01 0.723E-01
DAV: 13 -0.206655040163E+04 -0.13975E-02 -0.42621E-03 9472 0.187E-01 0.682E-01



DAV: 37 -0.206659254047E+04 0.16412E-03 -0.15001E-04 3424 0.283E-02 0.529E-01
DAV: 38 -0.206659636869E+04 -0.38282E-02 -0.84483E-03 9648 0.229E-01 0.535E-01
DAV: 39 -0.206659598920E+04 0.37949E-03 -0.12100E-03 7168 0.885E-02 0.516E-01
DAV: 40 -0.206659653451E+04 -0.54531E-03 -0.10441E-03 7072 0.987E-02 0.517E-01
DAV: 41 -0.206660012793E+04 -0.35934E-02 -0.79694E-03 9616 0.213E-01 0.504E-01
DAV: 42 -0.206660202676E+04 -0.18988E-02 -0.34382E-03 9232 0.126E-01 0.501E-01
DAV: 43 -0.206660325196E+04 -0.12252E-02 -0.12567E-03 7600 0.764E-02 0.512E-01
DAV: 44 -0.206660373183E+04 -0.47987E-03 -0.38913E-04 4592 0.501E-02 0.503E-01



DAV: 63 -0.206662059758E+04 -0.61930E-03 -0.96080E-04 6832 0.747E-02 0.479E-01
DAV: 64 -0.206661954203E+04 0.10555E-02 -0.76666E-04 6640 0.831E-02 0.495E-01
DAV: 65 -0.206661988786E+04 -0.34583E-03 -0.13980E-04 2992 0.370E-02 0.482E-01
DAV: 66 -0.206662029946E+04 -0.41160E-03 -0.11743E-04 2848 0.254E-02 0.470E-01
DAV: 67 -0.206662978622E+04 -0.94868E-02 -0.53840E-02 9648 0.545E-01 0.433E-01
DAV: 68 -0.206663096032E+04 -0.11741E-02 -0.53102E-03 6128 0.133E-01 0.460E-01
DAV: 69 -0.206663102178E+04 -0.61458E-04 -0.64174E-04 5696 0.358E-02
1 F= -.20666310E+04 E0= -.20667959E+04 d E =0.494724E+00 mag= 7.1929

[support_vasp]

Hi,

We're sorry that we didn’t answer your question. This does not live up to the quality of support that we aim to provide. The team has since expanded. If we can still help with your problem, please ask again in a new post, linking to this one, and we will answer as quickly as possible.

Best wishes,

VASP