[DFTB-Plus-User] Dispersion model (SlaterKirkwood)

[Benjamin Hourahine]

Dear Chao,

UFF dispersion is much easier to use, you might want to try it instead. The covalent radius
in the Slater-Kirkwood model depends on the hybridization state of the atoms, so
sp2 carbon has a different value from sp3. Your nanotube structures probably  remain 
sp2 during the calculation,so use the appropriate values, but check for re-hybridization during
the calculation.
 
Regards

Ben

________________________________________
From: dftb-plus-user-bounces at dftb-plus.info [dftb-plus-user-bounces at dftb-plus.info] On Behalf Of 张超 [201031220001 at mail.bnu.edu.cn]
Sent: 01 December 2012 01:28
To: dftb-plus-user at dftb-plus.info
Subject: [DFTB-Plus-User] Dispersion model (SlaterKirkwood)

Dear all,

I want to calculate the dynamics process of the Muti-walled carbon nanotubes.
So, the dispersion block is used to empirically correct the DFTB interactions for van der Waals interactions.

The SlaterKirkwood dispersion model is implemented in my calculation, but i do not know whether the code is right,
especially the CovalentRadius of the carbon.
Below is my input file.

Besides, in the user manual (page 34) of version 1.2 (DFTB+), it says that if atoms are able to move during your
calculation (geometry relaxation or molecular dynamics) you should always check whether the coordination of your
atoms has changed during the run. what does it mean?

Geometry = GenFormat {
  <<< "geom.gen"
}
Driver = VelocityVerlet {
  MovedAtoms = 1:-1
  Steps = 10000
  KeepStationary = Yes
  TimeStep [fs] = 0.1
  OutputPrefix = "geom.out"
  Thermostat = None {}
  Velocities [AA/ps] = {
   <<< "velocity.txt"
   }
   ConvergentForcesOnly = Yes
}
Hamiltonian = DFTB {
  SCC = Yes
  SCCTolerance = 1.0e-5
  MaxSCCIterations = 10000
  OrbitalResolvedSCC = Yes
  Mixer = Broyden {
    MixingParameter = 0.2
    CachedIterations = -1
    InverseJacobiWeight = 1.000000000000000E-002
    MinimalWeight = 1.00000000000000
    MaximalWeight = 100000.000000000
    WeightFactor = 1.000000000000000E-002
  }
  MaxAngularMomentum = {
    C = "p"
  }
  SpinPolarisation = {}
  Eigensolver = RelativelyRobust{}
  Filling = Fermi {
    Temperature [Kelvin] = 1000.0
  }
  SlaterKosterFiles = {
    C-C = "C-C.skf"
  }
  KPointsAndWeights = {
  0.0 0.0 0.0 1.0
  }
  OldSKInterpolation = No
  OldRepulsiveSum = No
  ReadInitialCharges = No
  Dispersion = SlaterKirkwood{
 PolarRadiusCharge =HybridDependentPol{
  C={
    CovalentRadius[Angstrom]=0.8
    HybridPolarisations [Angstrom^3, Angstrom,]={
    1.382 1.382 1.382 1.064 1.064 1.064 3.8 3.8 3.8 3.8 3.8 3.8 2.5
}
}
}
}
}
Options = {
  WriteDetailedOut = Yes
  WriteBandOut = Yes
}
ParserOptions = {
  ParserVersion = 4
  StopAfterParsing = Yes
}

Thank you
chao