Alex Granovsky
gran@classic.chem.msu.su
with Firefly, there are several variables called maxdii belonging
to different input groups. The one suggested by Masoud affects the
behavior of DIIS code used to converge SCF equations. The another
one you found in the output affects the behavior of the geometric
DIIS (gdiis) geometry optimization code.
Kind regards,
Alex Granovsky
On Sat Jun 1 '13 2:20am, Jia Huo wrote
--------------------------------------
>Dear Masoud,
>thank you very much for your suggestion.
>After change MAXDII to 40, i can successfully optimize the structure. But i found MAXDII=20 in the output file even after change.
> PARAMETERS CONTROLLING GEOMETRY SEARCH ARE
> METHOD =GDIIS UPHESS =BFGS
> NNEG = 0 NFRZ = 0
> NSTEP = 200 IFOLOW = 1
> HESS =GUESS RESTAR = F
> IHREP = 0 HSSEND = F
> NPRT = 0 NPUN = 0
> OPTTOL = 1.000E-05 RMIN = 1.500E-03
> RMAX = 1.000E-01 RLIM = 7.000E-02
> DXMAX = 3.000E-01 PURIFY = F
> MOVIE = F TRUPD = T
> TRMAX = 5.000E-01 TRMIN = 5.000E-02
> ITBMAT = 100 STPT = F
> STSTEP = 1.000E-02 PROJCT= T
> MAXDII = 20 NSKIP = 2
>Many thanks again!
>Jia
>On Mon May 27 '13 5:03am, Masoud Nahali wrote
>---------------------------------------------
>>Dear Jia
>>Please rise the MAXIT to 300 and add "MAXDII=40" to the $SCF. Then let us know you can figure the convergence issue out or not ? Ciao, m
>>Sincerely Yours
>>Masoud Nahali
>>SUT
>>
>>
>>
>>On Mon May 27 '13 2:44am, Jia Huo wrote
>>---------------------------------------
>>>Dear All,
>>>i have tried to optimize a CoO cluster using B3LYP_TZVPP. even though i have tried the suggestion in the output file, i still couldn't make SCF converged: SCF converge for first 3 steps but failed after that.
>>>the input file is
>>>! PCGAMESS STYLE B3LYP OPTIMIZATION CALCULATION OF COBALT OXIDE COMPLEX
>>> $CONTRL SCFTYP=UHF DFTTYP=B3LYP maxit=200 runtyp=optimize ICHARG=0 MULT=3 $END
>>> $CONTRL INTTYP=HONDO ICUT=11 ITOL=30 $END
>>> $SYSTEM TIMLIM=10000 MEMORY=100000000 $END
>>> $BASIS EXTFIL=.T. GBASIS=TZVPP $END
>>> $SCF DIRSCF=.True. FDIFF=.FALSE. DIIS=.T. SOSCF=.F. $END
>>> $GUESS GUESS=HUCKEL KDIAG=0 $END
>>> $STATPT NSTEP=200 OPTTOL=1e-05 $END
>>> $DATA
>>> UHF/B3LYP/TZVPP C4H4O8Co2
>>> C1
>>>OXYGEN 8.0 14.190477831 13.531629284 11.184729037
>>>OXYGEN 8.0 14.118975084 10.850079641 11.260893205
>>>CARBON 6.0 14.605933600 13.943005986 12.300263974
>>>HYDROGEN 1.0 15.400965861 14.696206805 12.267837538
>>>OXYGEN 8.0 11.507982408 13.603735483 11.218087165
>>>OXYGEN 8.0 11.436999908 10.922597992 11.294029467
>>>CARBON 6.0 11.144325716 14.036447002 12.343562832
>>>HYDROGEN 1.0 10.390516134 14.831479021 12.331058759
>>>OXYGEN 8.0 14.149980345 10.915251093 13.524442405
>>>OXYGEN 8.0 14.221301758 13.596328867 13.448273469
>>>CARBON 6.0 14.513791536 10.482622041 12.398984237
>>>HYDROGEN 1.0 15.267592776 9.687590339 12.411583205
>>>OXYGEN 8.0 11.467526545 10.987491299 13.557565990
>>>OXYGEN 8.0 11.539298334 13.669078574 13.481621545
>>>CARBON 6.0 11.052280192 10.575966507 12.442012400
>>>HYDROGEN 1.0 10.257360187 9.822632151 12.474346671
>>>COBALT 27.0 12.812527956 12.224553743 11.144743301
>>>COBALT 27.0 12.845592368 12.294513594 13.597688767
>>> $END
>>>the output file is shown in attachment.
>>>Could you please give me some suggestion what should i do to converge SCF?
>>>Many thanks,
>>>Jia