CASTEP simulation for graphene band structure

[muonneutrino91]

Homework Statement

I am running CASTEP DFT software to simulate pristine graphene, and am unsuccessful in obtaining the trademark zero band gap at the k-point as reported in numerous papers to date.

Any assistance would be greatly appreciated as the subject is very new but I am keen to push on.
It seems likely to me that I probably haven;t got the ground state energy of the system ; how to I go about interrogating this?

Homework Equations

Software makes use of cell and parameter files.

The Attempt at a Solution

I have modified the cell file of graphite available at (http://www.castep.org/Tutorials/BandStructureAndDOS) ;
I set the second graphite plane 10 angstroms away so as to reduce the effect on the monolayer being observed.
The two atoms have fractional coordinate (0,0,0.25) and (0,0,0.75).
Symmetry has been generated, and used point_mp_grid : 9x9x1.
band structure point path is G(0,0,0) ; K(1/3,1/3,0) ; M(0,1/2,0) ; G(0,0,0)

Parameter file uses PBE method, cut off energy has been set at 500eV, occupancy of atoms is fixed, grid scale is 2.0, no dump cycles and max SCF cycles set as 100.

I have mainly tried changing the point sampling (common error). Also tried using basis_precision instead of cut off energy at fine and precise settings. All to no avail thus far.

Re: I have uploaded a screenshot capturing the cell and parameter files and the band structure calculation for guidance.

 

[mark726]

Thank you for sharing your concerns regarding your DFT simulation of pristine graphene. It is understandable that you are keen to push on and obtain the expected results. Based on the information you have provided, I have a few suggestions that may help you troubleshoot and improve your simulation.

Firstly, it is important to note that the zero band gap at the K-point in graphene is a theoretical prediction and may not always be observed in simulations due to various factors such as numerical errors and approximations in the DFT method. Therefore, it is possible that your simulation may not show the exact same results as reported in literature.

That being said, I would recommend checking the convergence of your simulation by increasing the grid scale and/or the number of SCF cycles. This can help in obtaining more accurate results. Additionally, it may be useful to try using a different exchange-correlation functional, such as the LDA or hybrid functionals, to see if it affects the band gap at the K-point.

In terms of interrogating the ground state energy of your system, you can compare it to the results reported in literature for pristine graphene. If there is a significant discrepancy, it may indicate that your simulation has not yet converged or that there are other issues with your setup.

I hope these suggestions are helpful to you. If you continue to face difficulties, I would recommend reaching out to the CASTEP support team for further assistance. They may be able to provide more specific recommendations based on your simulation setup.

Best of luck with your simulation!