Dear Matter Modeling Stack Exchange community, I am currently working on calculating the electronic and structural properties of Transition Metal Dichalcogenide (TMD) nanotubes using DFT, with a specific focus on the 1T'-MoTe2. I would like to inquire about the best approach and tools available for preparing an input file, such as a POSCAR file, for a nanotube structure derived from the 2D structure of 1T'-MoTe2. I have attempted to use the c2x tool https://www.c2x.org.uk/nanotube.html to generate the nanotube structure. However, I encountered a problem with this approach because the c-axis in 1T'-MoTe2 is not perpendicular to the a-b plane, which is due to the distorted nature of the structure. Considering this challenge, I am seeking your guidance on alternative tools or methods that can be employed to construct an accurate nanotube structure starting from the distorted 2D structure of 1T'-MoTe2.

Thank you in advance for your expertise and assistance!

  • $\begingroup$ In the past, I used these two sites to generate nanotubes: TubeGen and TubeASP. $\endgroup$
    – Camps
    Oct 30, 2023 at 12:21
  • 1
    $\begingroup$ I am aware of those websites, but they don't support the thing I am looking for. Firstly, they are mainly for carbon nanotubes and can only support hexagonal or cubic cells. Besides, as I mentioned in the question the lattice I am trying to roll into a nanotube is distorted octahedral, so the c plane is not perpendicular to the ab plane. I have tried c2x and vaspkit and both didn't work as well. $\endgroup$ Oct 30, 2023 at 23:56

1 Answer 1


I'd recommend ChiralTube: http://chiraltube.com

The code (and simulations) were covered in a recent publication: "Chiraltube, rolling 2D materials into chiral nanotubes"

The code is also available on GitHub: https://github.com/Chema-dac/chiraltube

The paper covers, carbon nanotubes, phosphorene, but also $\ce{MoS2}$ and $\ce{Ti3C2}$. I haven't had much time to explore it, but the UnitCellExamples include QuantumEspresso input, "special XYZ", and VASP.

Here's $\ce{MoS2}$ in "special xyz" format:

S 1.5905260315876495465 0.9182342920325894154  6.565586833
S 1.5905260315876495465 0.9182342920325894154  3.434413167
Mo 0.0000142401699290626 1.8365313040729291692 5.000000000000
mass S 28.09
mass Mo 95.95
 3.181079696    0.000000000   0.000000000
-1.590540237    2.754889038   0.000000000
 0.000000000    0.000000000  10.000000000
cartesian coordinates

The website writes the "special xyz" by default, but you can use VASP input and output from the Python script:

python chiraltube.py <inputfile> -indexn5m10 -VASPin -VASPout

It also gives help with:

python chiraltube.py -h

For example, a supercell of $\ce{MoS2}$ 5,5 nanotube in POSCAR rendered by Avogadro:

enter image description here

  • $\begingroup$ Thank you for sharing the article. For people who might be interested, Chiraltube is the fastest approach. You can create a tubular structure using c2x tool in a few steps: put atoms into absolute coordinates, adjust lattice vectors for perpendicular planes, then make the tube. Chiraltube automates this process with distortion cells, requiring only chiral or achiral vectors. However, I am just concerned that by doing so (1T'-MoTe2 in my case)we are chaning the phase of the material from 1T'-MoTe2 to Td-MoTe2 by forcing the planes to be perpendicular. $\endgroup$ Nov 3, 2023 at 3:13
  • $\begingroup$ @JaafarMehrez - interesting question about the phase change. That sounds like a paper topic - run some calculations and compare the results from the two programs / geometries. Are the results substantially different? Unclear until someone (you?) runs the calculations. $\endgroup$ Nov 4, 2023 at 15:30
  • 1
    $\begingroup$ Well that's actually what I am working on. My intuition is that as far as it is single layer and we are constructing a single-wall tube then there will be no difference between the two phases (yet still I need to confirm with the calculation). However, when we talk about a multi-wall tube, then I think the phase shall be taken into consideration, and in this regard I was not able to build a mult-wall tube using Chiraltube. The produced structure is not reasonable. $\endgroup$ Nov 6, 2023 at 0:20

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .