I-V curve of DTB device

[OMG123]

I redid the calculation of transmission spectrum and I-V curve illustrated in Fig2 and Fig4 in ' Comp. Mat. Sci. 27, 151 (2003) '
The parameter setting is followed by tutorial 'Molecular Devices' except for k-point sampling. I set (1,1,50) instead of (3,3,50).
As we calculated I-V curve of DTB device, our result couldn't match the result in paper at voltage higher than 4 Volt.
I guessed it due to iteration tolerance because I reduced iteration tolerance to 0.05 at higher bias. However, I didn't found
the exact problem, so I need some advises from you.
thanks you

[kstokbro]

Difficult to determine what is the exact reason. The geometries are also not exactly the same, and I do not have the old geometry.
It is  important that you have a fine mesh for:
DoubleContourIntegralParameters(
real_axis_point_density )
thus, you might try lowering this value, and check convergence.

However, I would not be concerned about what is going on at 4 V bias, this is a very high bias of no experimental importance.