Nanoscrolls

[ahsan_javed]

Can we create POSCAR for TMDCs nanoscrolls and study them in VASP for properties?

[martin.schlipf]

I principle yes but I fear the smallest system size that is interesting might be to big for practical purposes. What is the smallest system of interest? Please be aware that VASP uses periodic boundary conditions so for a nanotube the smallest system would be a ring that is then periodically repeated to form the tube. What would a similar system be for nanoscrolls?

[ahsan_javed]

I wish to do it for a TMDCs monolayer, to turn that monolayer into a nano-scroll. But i am confused in how to make a POSCAR for it.

[martin.schlipf]

I have no experience with nanoscrolls so I cannot tell you what there structure would be. More generally, you would look for experimental reference data and then translate this to a POSCAR. If you have found such a structure in the literature, I can give you advice on how to turn it into a POSCAR.

If no experimental data on the structure is available, you need to come up with a guess yourself. This structure will then be likely very far away from the true groundstate, so you need to be careful when you relax it to avoid getting stuck in a local minimum perhaps using simulated annealing. If you want to go down this route, I would probably start with translating the positions of the 2d sheet into a scroll by using the equation for a spiral.

[ahsan_javed]

This is from literature review:

"Different from other 1D nanomaterials, the nanoscrolls have spiral tubular structure with weak van der Waals (vdW) interaction between adjacent layers, which are transformed from 2D nanosheets. The MoS2 nanoscroll is in an energy-favorable state when the interlayer spacing is in the range of 4.7~6.5 Å"

This is a paper which discusses TMDCS nanoscrolls experimentally: https://www.mdpi.com/2079-4991/13/17/2433

[martin.schlipf]

If that is the best experimental data that is available then you need to go with the second route. This is likely only to give you some qualitative insights because of the problems with finding the exact ground state mentioned earlier. You will probably also need some vdW functional to account for the interlayer interaction.

With this caveats in mind, here is how you can proceed if you want to use the approach I suggested before:

• Find a reference structure of a sheet in plane and decide on an origin where you want to cut the sheet and start rolling it up. Ideally you would want to cut in a way that the structure is periodic along what will be the length of the scroll. Otherwise the unit cell will get unreasonably large.
• Every atom should now be completely determined by a distance L perpendicular to the cut, a height z within the periodic repetition, and a distance d from the plane of the sheet.
• From the site I linked before, you will find that L = b/2 (θ sqrt(1 + θ^2) + ln(θ + sqrt(1 + θ^2)), where 2 pi b is the interlayer distance of the scroll and θ is the angle. You can use this equation to numerically map the position of each atom onto a radius r = θ b.
• Accounting for the distance d, you get the following positions of the atoms in the nanoscroll (r + d, θ, z) in cylindrical coordinates. You can transform them to Cartesian coordinates and then you should be able to construct a POSCAR.

[hszhao.cn@gmail.com]

Dear ahsan_javed,

You can try atomsk with option "-roll".

See here for the related discussion.

[ahsan_javed]

Okay, and i was looking around this software too. Will compare its result with atomsk.

https://www.sciencedirect.com/science/a ... 7016303066