Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

robocrys mp-24850 #4

Open
computron opened this issue Dec 19, 2018 · 3 comments
Open

robocrys mp-24850 #4

computron opened this issue Dec 19, 2018 · 3 comments

Comments

@computron
Copy link
Contributor

LiCoO2 is described in the literature as a layered (2d) material

See also the description of the corresponding mineral name: https://pdfs.semanticscholar.org/fc95/e3cd55b7c4760b2d1296e64bd5180e9df84f.pdf

(although I never heard of LiCoO2 referred to as Caswellsilverite, just a defect rocksalt)

@utf
Copy link
Member

utf commented Dec 19, 2018

When Robocrystallographer determines the wrong dimensionality it will be due to the bonding identified by CrystalNN. In this case it seem like CrystalNN is finding bonds from the Li to the surrounding O.

From the python API you can choose a different nearest neighbor algorithm which might correct the bonding.

I'm unsure of any other alternatives here.

@computron
Copy link
Contributor Author

The only thing I can think of is try a few different algorithms (or CrystalNN settings) and if some of them come up as layered, you could say "possibly layered". But this will most likely add more problems than it solves, better if one bonding algorithm just got it right ...

@computron
Copy link
Contributor Author

(note that LiCoO2 is both described as being layered, e.g. Li in separate layers, as well as Li being octahedrally coordinated by O. So usually it's described both ways ...)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants