Honeycomb flakes

[pfebrer]

Could a honeycomb/graphene flake geometry be useful? I'm using one currently and I think they are useful in general to create finite bilayers with arbitrary twist.

This PR is a proposal to add them.

Examples

import sisl
import sisl.viz

sisl.geom.graphene_flake(shells=3).plot(axes="xy")

import sisl
import sisl.viz

sisl.geom.honeycomb_flake(shells=3, bond=1.6, atoms=["B", "N"]).plot(axes="xy")

• Closes #xxxx
• Tests added
• Ranned isort . at top--level
• Documentation for functionality, docs/
• Changes documented, CHANGELOG.md

[zerothi]

How about an API of graphene_flakes that resembles the heteroribbon? In that way attachments of various segments could be done?

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (89dcede) 87.71% compared to head (86fd8c0) 87.73%.
Report is 1 commits behind head on main.

❗ Current head 86fd8c0 differs from pull request most recent head 5c72a0a. Consider uploading reports for the commit 5c72a0a to get more accurate results

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #636      +/-   ##
==========================================
+ Coverage   87.71%   87.73%   +0.01%     
==========================================
  Files         362      362              
  Lines       48354    48403      +49     
==========================================
+ Hits        42415    42464      +49     
  Misses       5939     5939              

Files	Coverage Δ
src/sisl/geom/flat.py	100.00% <100.00%> (ø)
src/sisl/geom/tests/test_geom.py	100.00% <100.00%> (ø)

... and 2 files with indirect coverage changes

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[pfebrer]

What would you attach to what in this case? 😅

[zerothi]

What would you attach to what in this case? 😅

A new flake/ribbon, connecting ribbons and flakes could be quite interesting ;)

[pfebrer]

Hmm yes, but the construction of the flake itself is not a composite geometry, no?

So I would imagine you can take graphene_flake and implement a section that knows how to connect to ribbons, which determines the arguments for graphene_flake and maybe rotates it, etc...

[zerothi]

yes, that sounds cool :)
Ok, but anyways, I think this is fitting

[zerothi]

great!

If adding 2 atoms, would it then alternate in each hexagon? That would be my ideal usecase, no?

Sorry, haven't tested, but looks good!

[pfebrer]

Ok, but anyways, I think this is fitting

Good 👍 I will then finish the PR some time this week probably.

To do the section stuff, it would be useful that the graphene_flake stored metadata on the Geometry, like edge_axes (the axes perpendicular to edges), edge_R (how far is the edge from the center) or edge_W (the number of C in the edge). But this requires the transformable attributes that we've discussed many times already 😅

[pfebrer]

If adding 2 atoms, would it then alternate in each hexagon? That would be my ideal usecase, no?

See the picture in the initial PR comment, is this how you expect it? Is there any other way?

[zerothi]

If adding 2 atoms, would it then alternate in each hexagon? That would be my ideal usecase, no?

See the picture in the initial PR comment, is this how you expect it? Is there any other way?

Sorry, too fast ;) Beautiful! :)

[pfebrer]

Ok, this PR is ready 👍

[pfebrer]

Should I set the boundary conditions to open?

[zerothi]

Should I set the boundary conditions to open?

All you need to do is to call geometry_define_nsc(geom, [False, False, False]) then return geom.

This will set the boundary conditions correctly, to UNKNOWN.

[pfebrer]

Why do you say the correct bc is UNKNOWN and not OPEN? It is a 0D flake.

[zerothi]

Why do you say the correct bc is UNKNOWN and not OPEN? It is a 0D flake.

For this it would be better to use Dirichlet, OPEN is a reference to NEGF, perhaps OPEN should be bulk, or something. Typically OPEN is used in the nomenclature of NEGF BC.

[pfebrer]

Ok, it should be fine now 👍