[DFTB-Plus-User] issues with diamond, free atom energy and DOS

[Adt D Bhatu]

Dear Users,

I am practically novice in terms of DFTB+ usage. I have a few questions
regarding the usage, they might be very naive but
would be a lot of help to me when answered.

1. I was trying to calculate optimize geometry of diamond structure.
Using the mio-0-1 set for the .skf files. The bulk modulus value what I got
is 1.5x10^2 (while the experimental value
is 4.42x10^2. We varied the lattice parameter manually in order to fit it to
the Equation of State. Please find one of the
input files below.
**********************
Geometry = {
  Periodic = Yes
  LatticeVectors [Angstrom] = {
    1.70  1.70  0.00
    1.70  0.00  1.70
    0.00  1.70  1.70
}
 TypeNames = { "C" }
 TypesAndCoordinates [relative] = {
   1  0.00  0.00  0.00
   1  0.25  0.25  0.25
 }
}

Hamiltonian = DFTB {
  SCC = Yes
  SCCTolerance = 1.0e-6
  MaxSCCIterations = 1000
  Mixer = Broyden {
    MixingParameter = 0.2
  }
  MaxAngularMomentum = {
    C="p"
  }
  SlaterKosterFiles = {
    C-C = "C-C.skf"
  }
 KPointsAndWeights = SupercellFolding {
    8   0   0
    0   8   0
    0   0   8
    0.5 0.5 0.5
  }

 SpinPolarisation = {}

 Filling = Fermi {
    Temperature [Kelvin] = 0.0
  }
}

Driver = ConjugateGradient {
  MovedAtoms = Range { 1 -1 }
  MaxForceComponent = 1.0e-5
  MaxSteps = 100
  OutputPrefix = "geo_dia.out"
}

Options = {
#  CalculateForces = Yes
  WriteAutotestTag = Yes
  AtomResolvedEnergies = Yes
}


ParserOptions = {
  ParserVersion = 3
}
***************************************


2. In order to calculate the energy of a free C (2s^2, 2p^2) atom, I set up
the spin polarisation as below
**************************************
SpinPolarisation = Colinear{
  UnpairedElectrons = 2
  InitialSpin = {
   AtomSpin = {
    Atoms = {1}
    SpinPerAtom = 2.0
  }
 }
}

 SpinConstants = {
 C = {
 #ss     sp     ps     pp
 -0.031 -0.025 -0.025 -0.023
 }
***********************************
The detailed.out gives occupations as this:

Net atomic charges
  Atom       Net charge
     1       0.00000000

 SPIN =            1

 Eigenvalues /H
   -0.55489172
   -0.24035511
   -0.24035511
   -0.24035511

 Eigenvalues /eV
  -15.09937195
   -6.54039531
   -6.54039531
   -6.54039531

 Fillings
     1.00000
     0.66667
     0.66667
     0.66667

 Nr. of electrons (spin: 1):      3.00000000
 Atom populations (spin: 1)
  Atom       Population
     1       3.00000000

 l-shell populations (spin: 1)
  Atom Sh.   l       Population
     1   1   0       1.00000000
     1   2   1       2.00000000

 Orbital populations (spin: 1)
  Atom Sh.   l   m       Population
     1   1   0   0       1.00000000
     1   2   1  -1       0.66666667
     1   2   1   0       0.66666667
     1   2   1   1       0.66666667

and
SPIN =            2

 Eigenvalues /H
   -0.45489172
   -0.14835511
   -0.14835511
   -0.14835511

 Eigenvalues /eV
  -12.37823350
   -4.03694794
   -4.03694794
   -4.03694794

 Fillings
     1.00000
     0.00000
     0.00000
     0.00000

 Nr. of electrons (spin: 2):      1.00000000
 Atom populations (spin: 2)
  Atom       Population
     1       1.00000000

 l-shell populations (spin: 2)
  Atom Sh.   l       Population
     1   1   0       1.00000000
     1   2   1       0.00000000

 Orbital populations (spin: 2)
  Atom Sh.   l   m       Population
     1   1   0   0       1.00000000
     1   2   1  -1       0.00000000
     1   2   1   0       0.00000000
     1   2   1   1       0.00000000
-------------------------------

are there any trick or flag that sets the populations /fillings properly for
a free C atom i.e.,
SPIN 1Atom Sh.   l   m       Population
     1   1   0   0       1.00000000
     1   2   1  -1       1.00000000
     1   2   1   0       1.00000000
     1   2   1   1       0.00000000

3. I plotted DOS for a C60 atom following the example described in
band_struct.pdf, I have used the coordinates of the autotest scc file.
I found finite DOS for C60 at E_F (assuming that the script normalizes the
energy wrt fermi energy). Is this normal? Please help.
Also the C and alpha in dosplot.in has to be set by trial and error for my
system?
************************************
I am sorry for this long e-mail. Hope I stated the problem clearly.

thank you,
Aditi Datta
University of Pittsburgh
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.zfn.uni-bremen.de/pipermail/dftb-plus-user/attachments/20090331/a4020a24/attachment.htm>