Disagreement between the result I got and the result in publication

[mldavidhuang]

Why the optimized interlayer distance of bi-layer Graphene is 3.08 Angstrom, but the value in other publication is around 3.3 Angstrom(reference for http://apl.aip.org/resource/1/applab/v92/i22/p223106_s1 also using ATK), is there anything wrong with my calculation ?

[zh]

You had better check which exchange-correlation functional is used exactly in your calculations. It is well-known that the GGA exchange functional would overestimate the layer distance between atomic layer in graphite, and the ver der Wales force between atomic layers in graphite is very important for the prediction of reasonable layer distance in graphite (or bilayer graphene).

For other parameters, you had better carefully refer to them used in the paper mentioned by you.

[mldavidhuang]

Actually，the optimized interlayer distance I got using LDA and GGA both are around 3.1 Angstrom……and the paper does not mention the parameter they use… Many other calculation all got the similar optimized interlayer distance of 3.34 Angstrom…So，I found rather puzzled

[zh]

The layer distance between bilayer graphene cannot be optimized by calling "OptimizeGeometry()".  You had better do a serials of calculations for total energy versus the layer distance and then find out the energy minimum, which will correspond to the equilibrium layer distance.

[mldavidhuang]

Howerver, the result I got from serials of calculations for total energy versus interlayer distance also turn out to be around 3.1 Angstrom. It's similar to the result I got from  "OptimizeGeometry()"……

[zh]

In your calculations, how about the optimized C-C bond length in monolayer of graphene? Is it very close to the result reported in literature?

In your script file, the lattice constant a is set as 2.5 angstrom, which corresponds to the C-C bond length of 1.44 angstrom. This is slightly larger than the experiment value.   

[mldavidhuang]

Oh, I change the C-C bond and the lattice constant to the experiment value, and this time equilibrium interlayer distance turn out to be around 3.36 Angstrom, Maybe this time it seems right ! Thank you for your help, I did not expect the result relate so closely to lattice constant, sorry for my negligence!

[mldavidhuang]

Another question I think out is that if I have to calculate the electronic structure of Graphene on Boron nitride for example, or Graphene on metal surface, there always seem to be the lattice mismatch, then which lattice constant I should chose? The graphene lattice or boron nitride lattice, since I also want to get the equilibrium distance between two layers……

[zh]

The lattice constants in plane  of graphene sheet can be fixed as those of the substrate. Actually, such problems can be referred to the literature. There are too many papers published for such interface system of graphene/metal or graphene/BN sheet.