[DFTB-Plus-User] surface slab calculation don't converge

[Sascha Thinius]

Hi malahmer,
Convergence is a general problem, dealing with surfaces having an 
dipole moment, since there are mostly not stable (at all).
You should use a surface model with equivalent surfaces on both sides 
of the crystal.
And adsorb your Hydrogen also on both sides symmetrically.

All the best.

Sascha




On Tue, 3 Dec 2013 09:54:57 +0100 (CET)
  malahmer at umbb.dz wrote:
> dear dftb user's
> i have tried to use the dftb code to study the properties of polar 
>ZnO surface. however the scc calculations don't converge
> i have used the next input file:
> 
> Geometry = GenFormat {
> 66  S
> O  Zn H 
>   1  3    .00000   .00000   .25000
>   2  1    .00000  1.89763  2.68800
>   3  2    .00000   .00000  2.01062
>   4  2    .00000  1.89763  4.69862
>   5  3   3.28680   .00000   .25000
>   6  1   3.28680  1.89763  2.68800
>   7  2   3.28680   .00000  2.01062
>   8  2   3.28680  1.89763  4.69862
>   9  3   6.57360   .00000   .25000
>  10  1   6.57360  1.89763  2.68800
>  11  2   6.57360   .00000  2.01062
>  12  2   6.57360  1.89763  4.69862
>  13  3  -1.64340  2.84645   .25000
>  14  1  -1.64340  4.74409  2.68800
>  15  2  -1.64340  2.84645  2.01062
>  16  2  -1.64340  4.74409  4.69862
>  17  3   1.64340  2.84645   .25000
>  18  1   1.64340  4.74409  2.68800
>  19  2   1.64340  2.84645  2.01062
>  20  2   1.64340  4.74409  4.69862
>  21  3   4.93020  2.84645   .25000
>  22  1   4.93020  4.74409  2.68800
>  23  2   4.93020  2.84645  2.01062
>  24  2   4.93020  4.74409  4.69862
>  25  1    .00000   .00000  5.37600
>  26  1    .00000  1.89763  8.06400
>  27  2    .00000   .00000  7.38662
>  28  2    .00000  1.89763 10.07462
>  29  1   3.28680   .00000  5.37600
>  30  1   3.28680  1.89763  8.06400
>  31  2   3.28680   .00000  7.38662
>  32  2   3.28680  1.89763 10.07462
>  33  1   6.57360   .00000  5.37600
>  34  1   6.57360  1.89763  8.06400
>  35  2   6.57360   .00000  7.38662
>  36  2   6.57360  1.89763 10.07462
>  37  1  -1.64340  2.84645  5.37600
>  38  1  -1.64340  4.74409  8.06400
>  39  2  -1.64340  2.84645  7.38662
>  40  2  -1.64340  4.74409 10.07462
>  41  1   1.64340  2.84645  5.37600
>  42  1   1.64340  4.74409  8.06400
>  43  2   1.64340  2.84645  7.38662
>  44  2   1.64340  4.74409 10.07462
>  45  1   4.93020  2.84645  5.37600
>  46  1   4.93020  4.74409  8.06400
>  47  2   4.93020  2.84645  7.38662
>  48  2   4.93020  4.74409 10.07462
>  49  1    .00000   .00000 10.75200
>  50  1    .00000  1.89763 13.44000
>  51  2    .00000   .00000 12.76262
>  52  1   3.28680   .00000 10.75200
>  53  1   3.28680  1.89763 13.44000
>  54  2   3.28680   .00000 12.76262
>  55  1   6.57360   .00000 10.75200
>  56  1   6.57360  1.89763 13.44000
>  57  2   6.57360   .00000 12.76262
>  58  1  -1.64340  2.84645 10.75200
>  59  1  -1.64340  4.74409 13.44000
>  60  2  -1.64340  2.84645 12.76262
>  61  1   1.64340  2.84645 10.75200
>  62  1   1.64340  4.74409 13.44000
>  63  2   1.64340  2.84645 12.76262
>  64  1   4.93020  2.84645 10.75200
>  65  1   4.93020  4.74409 13.44000
>  66  2   4.93020  2.84645 12.76262
>    0.0000000000E+00    0.0000000000E+00    0.0000000000E+00
>    9.86064505577087    0.0000000000E+00    0.0000000000E+00
>   -3.28688168528543    5.69304607688769    0.0000000000E+00
>    0.0000000000E+00    0.0000000000E+00   80.0000000000
> }
> Driver = ConjugateGradient{
>   LatticeOpt = No
>   AppendGeometries = Yes
>   MaxForceComponent [eV/AA]= 0.06
>   MaxSteps = 0
> }
> 
> Hamiltonian = DFTB {
>  SCC = Yes
>  SCCTolerance = 1.000000000000000E-003
>  MaxSCCIterations = 30
>  EwaldParameter = 0.02
>    Mixer = Broyden {
>  MixingParameter = 0.1
>  }
>  MaxAngularMomentum = {
>   Zn = "d"
>    O = "p"
>    H = "s"
>  }
>  Charge = 0
> 
>  Eigensolver = DivideAndConquer {}
> 
>  Filling = Fermi {
>    Temperature [Kelvin] = 100.000000000000000E-000
>  }
>  SlaterKosterFiles = {
>    O-O = "O-O.skf"
>    O-Zn = "O-Zn.skf"
>    Zn-O = "Zn-O.skf"
>    Zn-Zn = "Zn-Zn.skf"
>    H-H = "H-H.skf"
>    O-H = "O-H.skf"
>    H-O = "H-O.skf"
>    H-Zn = "H-Zn.skf"
>    Zn-H = "Zn-H.skf"
>  }
>  KPointsAndWeights = SupercellFolding{
>   5    0   0
>   0    7   0
>   0    0   1
>  0.0  0.0 0.0
>  }
>  }
>  Options = {
>  WriteAutotestTag = Yes
> }
> ParserOptions = {
>  ParserVersion = 4
> }
> 
> i have changed the ewald parameter as well as k point mesh and 
>vaccum regionlenght bur without results
> if any one has an idea please let me know
> best regards
> _______________________________________________
> DFTB-Plus-User mailing list
> DFTB-Plus-User at dftb-plus.info
> http://www.dftb-plus.info/mailman/listinfo/dftb-plus-user