[Wien] Why Wien2k does not provide proper energy degeneracy for graphene?

[Martin Gmitra]

Dear prof. Blaha,

We have observed using Wien2k_08.3 an artificial energy difference for free
standing graphene at the K-point, where bands have to cross. The degenerated

energies at the K-point are separated of about 4.15 micro eV when full
symmetry (NSYM=24) of the system is considered. Interestingly, when
2 inequivalent C atoms are considered without inversion symmetry (NSYM=6)
the energies at the K-point differ of about 0.28 micro eV only. For both the
cases the Rkmax=7, 33x33 k-mesh and Gmax=20 is considered. Why are
the results more accurate in the less symmetric case?

I am giving here more hints from my test runs. The following does not reduce

the energy difference:
1. increasing k-mesh
2. increasing Rkmax
3. increasing number of LMs
4. increasing IFFT parameter

With many thanks in advance,
Martin Gmitra

p.s.: I am attaching a figure of the energy dispersions
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20090125/4babb9db/attachment.html
-------------- next part --------------
A non-text attachment was scrubbed...
Name: accuracy_test.ps
Type: application/postscript
Size: 20599 bytes
Desc: not available
Url : http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20090125/4babb9db/accuracy_test.ps