search

nomad-coe / nomad

NOMAD lets you manage and share your materials science data in a way that makes it truly useful to you, your group, and the community.
https://nomad-lab.eu
Apache License 2.0
71 stars 16 forks source link

Incorrect geometry visualization, FHI-aims #33

Closed jmjarvi closed 2 years ago

jmjarvi commented 2 years ago

I noticed an incorrect geometry visualization with some FHI-aims calculations. As shown in the attached image, the topmost atomic layer of a 3-layer Ir(111) slab is misplaced. The structure in question is graphene/O/Ir(111) and I have used a non-standard species name "Ir1" for the topmost Ir layer for easier postprocessing (partial atomic charges etc.). Maybe the non-standard name is the reason for incorrect parsing of the geometry file.

As far as I know, using non-standard names is a common trick to separate same-species atoms in the FHI-aims output, so this would be nice to get fixed.

Link to this structure: https://nomad-lab.eu/prod/rae/gui/entry/id/cpU_o7sHRti8RMO6DNgynQ/Pq-SBeTml5xBC36EkEYoML5b42RG

All structures in this dataset have the same problem: https://nomad-lab.eu/prod/rae/gui/dataset/id/y1zh1Hs4R--vU6KHZJoHIA

Gr_O_Ir111

markus1978 commented 2 years ago

@ladinesa @lauri-codes, do you have any comments on this?

ladinesa commented 2 years ago

The parser seems to parse the correct coordinates.

lauri-codes commented 2 years ago

Hi @jmjarvi! Hope you are doing fine :)

This issue is gone with version 1.1.0, which produces the following structure:

structure_1 1 0

We are in the middle of transitioning into this new version, and we can let you know once your dataset has been reprocessed and is available in the beta release. The new version also fixes an issue with the overview page showing an empty plot for "Geometry Optimization".

Let us know if you encounter any other issues with your dataset.

jmjarvi commented 2 years ago

Hi @lauri-codes and thanks for the quick reply! That's excellent, the visualization looks good to me. Also very nice that the unconventionally named "Ir1" species are recognized correctly as Ir. Great!