5
$\begingroup$

I am trying to produce a band structure for TiO2, with the cell file shown below:

%BLOCK Lattice_ABC
  4.853556 4.853556 3.018297
  90.0 90.0 89.989556
%ENDBLOCK Lattice_ABC

%BLOCK Positions_Frac
  Ti 0.0 0.0 0.0
  Ti 0.5 0.5 0.5
  O  0.194689 0.805311 0.5
  O  0.805311 0.194689 0.5
  O  0.694707 0.694707 0.0
  O  0.305293 0.305293 0.0
%ENDBLOCK Positions_Frac

SPECTRAL_KPOINT_PATH_SPACING: 0.1

%BLOCK SPECTRAL_KPOINT_PATH
    0.0000000000     0.0000000000     0.0000000000 !    GAMMA
    0.0000000000     0.5000000000     0.0000000000 !    X
    0.5000000000     0.5000000000     0.0000000000 !    M
    0.0000000000     0.0000000000     0.0000000000 !    GAMMA
    0.0000000000     0.0000000000     0.5000000000 !    Z
    0.0000000000     0.5000000000     0.5000000000 !    R
    0.5000000000     0.5000000000     0.5000000000 !    A
    0.0000000000     0.0000000000     0.5000000000 !    Z
    BREAK
    0.0000000000     0.5000000000     0.0000000000 !    X
    0.0000000000     0.5000000000     0.5000000000 !    R
    BREAK
    0.5000000000     0.5000000000     0.0000000000 !    M
    0.5000000000     0.5000000000     0.5000000000 !    A
%ENDBLOCK SPECTRAL_KPOINT_PATH

Thanks in advance!

The band structure plot produced does not contain breaks between the K points, as can be seen below.

enter image description here

Any help on why they are connected, and how to separate them to achieve something more like this:

enter image description here

$\endgroup$
1
  • $\begingroup$ +1 but can you please provide that .cell file in a code block rather than a screenshot? The screenshot does not appear visible on browsers that have image-blockers, screen-readers (for blind people), and or slower computers. $\endgroup$ Mar 6, 2023 at 17:56

1 Answer 1

2
$\begingroup$

Your .cell file is correct, and CASTEP has added a break between the k-point path for the spectral kpoints; for example, there are no spectral k-points between the Z and X points.

What you are seeing in your band-structure plot is that the plotting software is joining the bands directly between the points either side of the break. Your first plot looks like xmgrace, perhaps because you're using the simple dispersion.pl script we distribute with CASTEP; I don't think that you can stop the points being joined by lines on a case-by-case basis, you'd have to either remove the lines entirely or leave it as it is. The CASTEP output is just a set of eigenvalues at a set of k-points, so there's no way for a plotting program to know whether they are to be "joined up" or not, you'd have to instruct it directly.

For xmgrace, I think the only way would be to split the data sets around the "break" and plot them separately, and then move the x-coordinate of the different k-points together. Other plotting software might be more flexible in that respect.

$\endgroup$

You must log in to answer this question.

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