Monolayer to Bulk

[ams_nanolab]

Dear Sir,
I would like to simulate the effect of adding successive layers of MoS2 from monolayer to Bulk, as described in

Appl. Phys. Lett. vol. 99, pp. 261908 (2011). dx.doi.org/10.1063/1.3672219

and

P. Lu, X. Wu, W. Guo, and X.C. Zeng, Physical Chemistry Chemical Physics : PCCP 14, 13035 (2012).

But if I simulate the structures by adding layers successively with the interface tool, bandstructures seem to horribly wrong.
Upto bilayer I'm doing it by simply adding unitcell from materials database. Still it is giving wrong results.

For more layers I am using the interface tool. Also for more than two layers the G-M-K-G path is no longer available so I have to override it with python scripts.Could you suggest the right procedure to reproduce the results in the mentioned papers.

[zh]

There is a video for building MoS2 nanotube on youtube (http://www.youtube.com/watch?v=t2_nlrW3_Qk). In the former part of this video, the steps for the construction of MoS2 monolayer are demonstrated.

If you build the MoS2 bilayer with hexagonal lattice, the G-M-K-G path may be already available on the scripter generator.

[ams_nanolab]

I know how to construct monolayer. That is not the issue, but the problem is with bilayer and multi-layers. That's where the results don't match with other published results. If u want I could mail the nc files to you. Kindly provide ur e-mail.

[Anders Blom]

It's extremely critical to optimize the geometry in MoS2 to get reasonable results. If you just take database structures (which are experimental geometries) and stack them, the results will not be correct, even for the monolayer (in fact, even for bulk).

[ams_nanolab]

For monolayer I'm getting ok. But for bilayer the CBmin-VBmax are no longer at K point but shifted which should not be the case as in the papers Appl. Phys. Lett. vol. 99, pp. 261908 (2011) etc.

The bilayer is just the mos2 unit cell from the database without deleting the second group of atoms (S-Mo-S) in it. So shouldn't that be sufficient to at-least give correct results for bilayer???

By optimization do you mean I should include the optimize geometry block or just simply adjust the interlayer spacings myself?

Help me out here.

[Anders Blom]

First of all, the reference you quote uses screened exchange, so it's not possible to compare the results to LDA/GGA directly. Second, without knowing more about the details of your calculation, I can't say much. The bulk MoS2 in the database is not a bilayer unless you also add vacuum, but perhaps you did. However, even so, minimizing the forces on the atoms (by using an OptimizeGeometry block, yes) will be crucial to getting good results (even for bulk, as I noted above). Whether they agree with screened exchange I don't know, however.

[ams_nanolab]

I did not add vacuum in my calculations, maybe that's the reason. Could you give me some pointers as how to include vacuum in such a structure?

[Anders Blom]

Make the unit cell larger.