6
$\begingroup$

I would like to understand how to calculate the band structure of the Graphene monolayer system using DFT. I am using VASP for material simulation. My question is how to write the crystal structure information file (POSCAR) for a 2D system such as Graphene. I think I can write the POSCAR file for many 3D systems, but I would like to understand how to write the POSCAR file if I am dealing with a 2D material system.

$\endgroup$
1
  • $\begingroup$ Why don't you register your account? $\endgroup$ Jun 22 at 19:52
6
$\begingroup$

You generally cannot calculate a 2D system per se with a plane-wave program, but rather you should do a vacuum slab calculation. As for graphene, you would essentially do a calculation on graphite, but with the interlayer distance set at a large value, say 20 Angstrom (and the lattice constant c enlarged accordingly), so that the different layers of graphene virtually do not interact with each other. Make sure to freeze the lattice constant c, and use only one k-point along the c direction, although you still need an appropriate number of k-points along the a and b directions.

$\endgroup$
5
  • $\begingroup$ Thank you very much. But why is it not possible to calculate a 2D system from first principles? Is it a limitation of the current computer tools or is there some physical reason related to DFT theory? $\endgroup$
    – phdstudent
    Jun 22 at 20:10
  • 1
    $\begingroup$ You can definitely carry out DFT calculations on 2D systems. It just requires a code that supports 2D periodic boundary conditions rather than the conventional 3D. For instance, the code Crystal can handle 2D systems without relying on artificial vacuum space. $\endgroup$ Jun 23 at 0:05
  • $\begingroup$ You have to be aware that if the atomic coordinates are in fractional format, increasing c will do nothing. Also if you have a unit cell for graphite, v and increase c you will end not with only one layer of graphene, but with two, three... $\endgroup$
    – Camps
    Jun 23 at 0:37
  • $\begingroup$ If the system is 2D need also 2D k-points, not just one. $\endgroup$
    – Camps
    Jun 23 at 0:38
  • 1
    $\begingroup$ Thank you for all your comments - I edited my answer to reflect some of them. As @AndrewRosen said, some programs do support 2D calculations, but they must use atomic-centered basis functions instead of plane waves. As the OP is asking about VASP, I gave a conclusion that is valid for plane waves, but failed to point out that it is only valid for plane waves. $\endgroup$
    – wzkchem5
    Jun 23 at 8:27
4
$\begingroup$

The VASP package solving the KS-equation with periodic boundary conditions (PBC). For two-dimensional materials such as graphene, you only have two periodic directions along $\vec{a}$ and $\vec{b}$ but the VASP assumes you still have three directions. To avoid this, you need to add enough vacuum (in general large than 15 angstroms) along the $\vec{c}$ dirction.

Here is a POSCAR template:

C
1.0
    2.4600000381         0.0000000000         0.0000000000
   -1.2300000191         2.1304225263         0.0000000000
    0.0000000000         0.0000000000        15.0000000000  ! add vacuum along the assumed PBC direction.
C
2
Direct
    0.000000000         0.000000000         0.500000000
    0.333333359         0.666666706         0.500000000
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.