[Wien] Boltztrap- working with nanowires

[Christopher W. Twombly]

Wein users,

I am trying to use BoltzTraP to calculate conductivity for a 001 Si
nanowire. Looking in the ***.outputtrans file I see the following error:

WARNING - kcomp: large error in re-interpolation of elec bands

The error is outputted from line 64 of the kcomp.F90 routine in BoltzTraP.
I think this indicates that the Fourier interpolation of the bands has
failed, since BoltzTraP cannot find the Fermi level after this happens. I
think I may have something wrong with my ***.struct file used by BoltzTraP
based on the k mesh I am using. The ***.struct file is short:

BoltzTraP geometry file
    0.6000000000E+02    0.0000000000E+00    0.0000000000E+00
    0.0000000000E+00    0.6000000000E+02    0.0000000000E+00
    0.0000000000E+00    0.0000000000E+00    0.1026121700E+02
      4
    1    0    0      0    1    0      0    0    1   ! symmetry rotation
matrix isym =     1
   -1    0    0      0   -1    0      0    0    1   ! symmetry rotation
matrix isym =     2
    0    1    0      1    0    0      0    0    1   ! symmetry rotation
matrix isym =     3
    0   -1    0     -1    0    0      0    0    1   ! symmetry rotation
matrix isym =     4

The nanowire is periodic along the z direction. I have tried several
different k grid combinations that all produce the same error. The only
test k grid that worked with this structure has completely flat energy
bands. I have been able to calculate quantities accurately for bulk Si,
diamond, and graphite. Any suggestions?
Thanks,
Chris
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20141016/648223d2/attachment.html>