Doubts about phosphorene nanoribbon

[Giamm Street]

Hi everyone,
I created a monolayer of phosphorene (pic 1), as shown in your tutorial: https://docs.quantumwise.com/tutorials/crystal_builder/crystal_builder.html
and I made some simulations. The density of states shows a results not expected: the orbital "d" has energetic density (pic 2), but the hybridation of black phosphorus and phosphorene is sp3. Instead, it seems an sp3d hybridation, how can it be possible?
After that, I built the phosphorene device (pic 3). The calculate of I(V) plot shows a value of zero current from -0.8 eV to 0.8 eV, but the measured band gap of phosphorene is only about 0.8 eV, not the double, so I don't undestand this result.
Thank you for your attention
G. C:

[Petr Khomyakov]

I would suggest first reproducing the band structure as in the tutorial to see that you have the right structure and band gap. Regarding DOS, it is not clear what you are actually showing in that picture. Also, why do you think there is no contribution of d-orbitals, e.g., to some of the states that are far from the Fermi energy?

Regarding IV calculations for a perfect ballistic system with the ballistic leads made of the same material as that of the central ("scattering") region, the only quantity that makes sense in this case is the Transmission Spectrum and related ballistic conductance of a perfect crystal at V_dd=0. So, I would calculate the transmission spectrum at V_dd=0 to see if the gap in the transmission is consistent with that in the band structure. Note that you should use sufficient number of k-points to converge the transmission spectrum, i.e., you have to increase the k-point grid density until reaching the convergence.       

[Giamm Street]

First of all, thanks so much Mr. Khomyakov for your answer.
I suppose that the sp3 hybridation of phosphonere doesn't allow the 15 electrons of each P atom to reach the 3d orbitals (according to the last level of phosphorus 3s^2, 3p^3). In pic 2 I show the specific contribution of d orbitals to total density of states (I expected contribution=0, like f orbitals).
As you requested, I attach the bandstructure of my phosphorene monolayer. The band gap measured at Γ point is 0,82 eV.
Thanks for your attention
G. C.

[Petr Khomyakov]

If you calculate the Transmission Spectrum for the nanoribbon at zero bias voltage, you will be able to see the transmission gap of 0.82 eV in consistency with the energy gap.

Regarding d-orbitals contribution to the total DOS, it is small compared to s and p-orbital contributions, being consistent with your argument, see a png-file (with projected density of states) enclosed. But as I said the DOS does not need to be 100% of sp-character, because a complete basis set would actually include all harmonics given by the orbital quantum number (l). For some k-points in the Brillouin zone, I guess some harmonics may not contribute to the wave function because of the particular symmetry of the corresponding Bloch state at these k-points.

[Giamm Street]

Thank you so much again Mr. Khomyakov,
Can I ask you how to get a graph like yours, i.e. with different overlaid curves? And if you can attach the phosphorene nanoribbon or device with which you got this graph?
Best regards
G. C.

[Petr Khomyakov]

You can use PDOS (projected density-of-states) analysis object to project DOS on elements, elements and shells and so on. This version is available in both VNL-ATK 2017.2 and a recent version of QuantumATK O-2018.06.

The structure is posted at https://quantumwise.com/forum/index.php?topic=5757.0.

[Giamm Street]

Thank you Mr. Khomyakov,
I have one last question to ask you. I made this device with Au electrodes (Figure 1) and I plotted the I/V curve. The result shows that the band gap of phosphorene (ca 0,8 eV) has disappared (Figure 2) and I don't undestand why. Is it so because the central region is too short? Or because the gold electrodes act as doping elements?
Thank you in advance for your attention sir.                                                                                                                                                                                                                   Best regards
G. C.

[Petr Khomyakov]

I guess answering your questions requires explicit calculations. You may, for example, calculate PLDOS, e.g., see https://docs.quantumwise.com/casestudies/ag_si_interface/ag_si_interface.html, to obtain the band diagram across your device.