Forum Navigation
You need to log in to create posts and topics.

Large Coordination numbers in Voronoi

Hi,

I am trying to make Voronoi analysis on amorphous tantala. This system is made up with a large number of TaO6 octahedra and there is hardly any O-O bonding. While doing the Voronoi analysis, I get a CN of ~6 for Tantalum atoms as expected but the CN for O are really large. Could you please explain why this is happening.

My .ovito is attached.

Can I create a Voronoi cell of an atom with its neighbor lying below certain user defined cut-off distance?

Thanks,

Rajendra Thapa.

Uploaded files:
  • You need to login to have access to uploads.

Hi,

this is the expected behavior of the Voronoi Analysis modifier. If you activate the option "Generate neighbor bonds" in the Voronoi Analysis modifier, you can see the actual Voronoi neighbor bonds. You will notice that they are not the same bonds you calculated with the Create bonds modifier (especially since you specified a pair-wise cutoff for O and Ta pairs only).

If you would like to calculate the coordination number according to those bonds instead, you can do that with a short python script modifier in the GUI. Let me know if you would like to know more about this.

Can I create a Voronoi cell of an atom with its neighbor lying below certain user defined cut-off distance?

No, the only thing you have control over is to specify a face area or edge length threshold, that allows you to exclude degenerate faces or edges from the analysis, please see section "Face area and edge length tresholds" in the manual.

-Constanze

 

Uploaded files:
  • You need to login to have access to uploads.

That makes sense. A couple of questions:

  1. How does the particle radii come into play during the Voronoi analysis?
  2. What is the appropriate edge length threshold?

Thanks,

Rajen.

Hi,

only when you activate the option "use particle radii" (and only if you have defined them in the particle settings) a polydisperse Voronoi tesselation will be performed. Otherwise a mono-disperse Voronoi tessellation is computed, which is independent of the particle sizes.
For a comparison, see e.g. http://math.lbl.gov/voro++/examples/radical/.

That very much depends on your system and what you would like to analyze. If you are unsure, if that's something you need, you should set it to zero. In any case, to make sure your results are reproducible it is important that you report this value if it's non-zero.

-Constanze

 

That helps. Thank you.
Also I am looking at your example code: https://www.ovito.org/docs/current/python/introduction/examples/batch_scripts/voronoi_indices.php

What radius are you referring to when you say atomic radius? Could you please let me know the source you get the atomic radius from?

Also, does this script make the use of the Periodic boundary conditions?

Thanks.

As stated in the example code, the model to be analyzed consists of  Cu and Zr atoms. In this example, the empirical radii have been used.

Yes, as explained in the manual,  https://www.ovito.org/docs/current/particles.modifiers.voronoi_analysis.php, the Voronoi analysis modifier takes into account the periodic boundary conditions (if they are active). If that information is not in your input data, you should make sure to manually set the periodic boundary conditions under Data Source -> Simulation Cell in the Modifications tab.

 

I want to check what the Voronoi polyhedra of some atom (let's say atom number 200) looks like. Can I do that after I make the Voronoi polyhedra for all the atoms. I tried expression selection but did not succeed.