[DFTB-Plus-User] Problem with geometry optimization

[Pasan Henadeera]

Dear developers and users,

I have been trying to use DFTB+ to optimise a structure made of Si
(consisting of around 750 atoms). However, even though I reduce the SCC
tolerance and increase the Max steps, I still am unable to figure out how
convergence can be obtained. Would it please be possible to give me some
guidance regarding this matter? I have attached the input and log files as
well. The optimization and Hamiltonian sections of the input are as follows,


Driver = GeometryOptimization {
Optimizer = Rational {DiagLimit = 1.0e-2}
LatticeOpt = Yes
MaxSteps = 500
}

Hamiltonian = DFTB {
  Scc = Yes
  SccTolerance = 0.2
  SlaterKosterFiles = Type2FileNames {
    Prefix = "../../../../slakos/wfc/"
    Separator = "-"
    Suffix = ".skf"
  }
    MaxAngularMomentum {
    Si = "d"
  }
  KPointsAndWeights = SupercellFolding {
    1 0 0
    0 2 0
    0 0 2
    0.5 0.5 0.5
  }
}

ParserOptions {
  ParserVersion = 7
}



I have only 16GB of memory, which is why I tried reducing the number of
k-points. However I got the same issue when using more k-points.


Thank you
-- 
Best regards
Pasan Henadeera
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.zfn.uni-bremen.de/pipermail/dftb-plus-user/attachments/20220116/b825711f/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: log.dftb
Type: application/octet-stream
Size: 67348 bytes
Desc: not available
URL: <http://mailman.zfn.uni-bremen.de/pipermail/dftb-plus-user/attachments/20220116/b825711f/attachment-0002.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: dftb_in.hsd
Type: application/octet-stream
Size: 34030 bytes
Desc: not available
URL: <http://mailman.zfn.uni-bremen.de/pipermail/dftb-plus-user/attachments/20220116/b825711f/attachment-0003.obj>