[Wien] [SPAM?] Covergence Problems of Slab Calculation

[Laurence Marks]

Many things:
1) By the look of it you are using the default "riter" of 20, which
removes the Broyden history after 20 cycles. I don't like this (others
use it, so there is no overall concensus). I override this by setting
riter to 999 in all run_XXX scripts and also use 999  999
as the last line in case.inm, e.g.
BROYD  0.000 YES  (BROYD/PRATT, extra charge (+1 for additional e), norm)
0.10            mixing FACTOR for BROYD/PRATT scheme
0.03  1.50      PW and CLM-scaling factors
999  999

2) Your calculation is not converging. Search the maillist for how to
change the PW scaling terms, the mixing amplitude and use :PLA to
understand this.

3) See the FAQ on minimization.

On 12/9/06, 张旺 <wzhang4 at ustc.edu> wrote:
> Dear wien-users,
>         When I used my Fe adsorption model on Zn4O4 with 40Bohr Vacumm between Slab Layers, executed the program "run_lapw -p -i 80 -cc 0.0001  >out&", and the covergence results What i get  from the case are like the things show @last.
>         For the Fe reasons I choose -7.0Ry to seperate the core & Valence.RKmax 7.0 .
>         The question that puzzled me should be "IS there any problems with my structure,especially the Fe-O bond-length?and the bond-length of the Fe-O I chose is 2.13870 angstrom, and What should I do to change the length, Increase or decrease?or maybe not the bond-legnth reason?
>
>          And What Should I do to check the reason?How?
>          Thanks a lot!
>
> in cycle 2    ETEST: 0   CTEST: 0
> in cycle 3    ETEST: 0   CTEST: 0
> in cycle 4    ETEST: .1509470000000000   CTEST: .3806849
> in cycle 5    ETEST: .1211920000000000   CTEST: .3092707
> in cycle 6    ETEST: .0811985000000000   CTEST: .2151343
> in cycle 7    ETEST: .0994920000000000   CTEST: .2085661
> in cycle 8    ETEST: .0747730000000000   CTEST: .1156605
> in cycle 9    ETEST: .0187465000000000   CTEST: .1451649
> in cycle 10    ETEST: .0159515000000000   CTEST: .0956633
> in cycle 11    ETEST: .0151900000000000   CTEST: .0950830
> in cycle 12    ETEST: .0077710000000000   CTEST: .0735470
> in cycle 13    ETEST: .0059290000000000   CTEST: .0743307
> in cycle 14    ETEST: .0046345000000000   CTEST: .0717344
> in cycle 15    ETEST: .0035450000000000   CTEST: .0736045
> in cycle 16    ETEST: .0041075000000000   CTEST: .0641029
> in cycle 17    ETEST: .0045775000000000   CTEST: .0539203
> in cycle 18    ETEST: .0061410000000000   CTEST: .0615848
> in cycle 19    ETEST: .0063015000000000   CTEST: .0606164
> in cycle 20    ETEST: .0582760000000000   CTEST: .0997554
> in cycle 21    ETEST: .2043690000000000   CTEST: .2449669
> in cycle 22    ETEST: .1272270000000000   CTEST: .0727216
> in cycle 23    ETEST: .5653710000000000   CTEST: .4965958
> in cycle 24    ETEST: .3614260000000000   CTEST: .0465985
> in cycle 25    ETEST: .3572905000000000   CTEST: .0306658
> in cycle 26    ETEST: .0041515000000000   CTEST: .0503465
> in cycle 27    ETEST: .0028735000000000   CTEST: .0262064
> in cycle 28    ETEST: .3609530000000000   CTEST: .2912574
> in cycle 29    ETEST: .1817600000000000   CTEST: .1325841
> in cycle 30    ETEST: .5887710000000000   CTEST: .5096802
> in cycle 31    ETEST: .3734855000000000   CTEST: .1718762
>
> --
> USTC Alumni Email System
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>


-- 
Laurence Marks
Department of Materials Science and Engineering
MSE Rm 2036 Cook Hall
2220 N Campus Drive
Northwestern University
Evanston, IL 60208, USA
Tel: (847) 491-3996 Fax: (847) 491-7820
email: L-marks at northwestern dot edu
http://www.numis.northwestern.edu