[DFTB-Plus-User] Segmentation fault in REKS

[Crespi, Stefano]

Hello,

I wanted to ask if it is possible to optimize molecules using reks.
I am very new to the code and I am now starting some test runs.
I am modifying the si-sa-reks examples from the recipes, adding Driver to
try to perform the optimization
No matter which driver I am adding, the code just ends after one cycle with

Program received signal SIGSEGV: Segmentation fault - invalid memory
reference.

without entering in the optimization step.

I am using the precompiled binaries of DFTB+20.1

An example of the code I am using follows:

Geometry = GenFormat {
 6  C
 C H
    1  1    0.4682053268E-01   -0.4157402516E-10   -0.2089753853E-11
    2  1   -0.1373995990E+01    0.2843708212E-10   -0.2577001194E-10
    3  2    0.6337363363E+00    0.9349896145E+00   -0.4289412964E-09
    4  2    0.6337363362E+00   -0.9349896144E+00    0.4283618438E-09
    5  2   -0.1960911747E+01    0.9349896174E+00   -0.4253969181E-09
    6  2   -0.1960911747E+01   -0.9349896173E+00    0.4314535043E-09
}

Driver = ConjugateGradient {
  MovedAtoms = 1:-1
  MaxForceComponent = 1E-4
  MaxSteps = 1000
  OutputPrefix = "geom.out"
}

Hamiltonian = DFTB {
  SCC = Yes
  SCCTolerance = 1e-6
  MaxSCCIterations = 10000
  Charge = 0.0
  MaxAngularMomentum = {
    H = "s"
    C = "p"
  }
  SlaterKosterFiles = Type2FileNames {
    Prefix = "../../slakos/mio-ext/"
    Separator = "-"
    Suffix = ".skf"
  }
  SpinConstants = {
    ShellResolvedSpin = Yes
    H = { -0.072 }
    C = { -0.031 -0.025 -0.025 -0.023 }
  }
}

REKS = SSR22 {
  Energy = {
    Functional = { "PPS" "OSS" }
    StateInteractions = Yes
  }
  TargetState = 1
  FonMaxIter = 100
  Shift = 1.0
  Gradient = ConjugateGradient {
    CGmaxIter = 100
    Tolerance = 1.0E-8
    Preconditioner = Yes
    SaveMemory = Yes
  }
  VerbosityLevel = 1
}

Analysis {
  CalculateForces = Yes
}

Options {
  WriteAutotestTag = Yes
}

ParserOptions {
  ParserVersion = 8
}

Thanks so much for the hep!

Stefano



-- 
Dr. Stefano Crespi, Ph.D.

Feringa Research Group
Stratingh Institute for Chemistry
University of Groningen
Nijenborgh 4
9747 AG Groningen
The Netherlands

E-mail stecrespi01 at gmail.com;  s.crespi at rug.nl

http://www.benferinga.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.zfn.uni-bremen.de/pipermail/dftb-plus-user/attachments/20201002/c8dfafab/attachment.htm>