PLDOS analysis

[EchoLee]

Dear sir,
    Resently，I computed the IV curves of γ-graphyne devices, the results display that I(down)>I(up) at Vb=0.5V, that is to say, we can get I(down) at Vb=0.5V. However, when I plot the PLDOS figures to explain the micromechanism of this phenomenon, I found that the PLDOS given opposite results compared with IV curves. I cannot get the answer of this issue.
Sincerely,
Echo

[Petr Khomyakov]

The electric current certainly depends on what  PLDOS looks like for a given device, but the PLDOS does not give a direct answer to the actual current values for the two spin channels. Please explain how you then came to this conclusion that the current for one spin channel is greater than for the other, using the PLDOS analysis.

[EchoLee]

The delocalized of PLDOS indicates the larger current value, we observe that the delocalization of PLDOS for up lager than that for down, but the IV curves show that the current value for up smaller than that for down. That to say, PLDOS given opposite results compared with IV curves.

[Petr Khomyakov]

Just from a general point of view, I still do not see how you can draw a quantitative picture out of this. You have also not posted any pictures and scripts, related to these calculations, so it is impossible to give you any concrete advice on what might need to be verified in your calculations to validate their reliability.  For example, it might be that you have to compute your I-V and/or PLDOS with stricter computational settings  (more k-points, energy points etc).

[EchoLee]

Dear Sir,
  In my opinion, the more delocalized of the PLDOS is, the larger value of current will be, this is a way to qualitativelly explain the micromechanism of quantum transport phenomenon. In my IV curves, the spin-down current is dominated (especially the situation of B-doping structure for AP configuration), that is to say, the delocalization of spin-down PLDOS is larger than that of spin-up PLDOS. However, when I calculated the spin-PLDOS, I got the opposite results.

[Petr Khomyakov]

I assume that you have calculated PLDOS for the bias voltages where you suspect the problem. Otherwise, using the zero-bias PLDOS is not relevant.

You should also make sure that your high bias calculations are converged. From the IV curve, it looks like a convergence problem.

In general, calculating IV for a pristine material, which is a ballistic conductor, do not make much sense if you use the same material for the electrodes.  The Transmission Spectrum and related conductance at zero bias are the parameters describing the electron transport for ballistic conductors.