Skip to content

Combine the GPAW and VASP ab initio potential tutorials into one page - #33

Open
TomaSusi wants to merge 1 commit into
mainfrom
claude/ab-initio-potentials-vasp
Open

TomaSusi wants to merge 1 commit into
mainfrom
claude/ab-initio-potentials-vasp

Conversation

@TomaSusi

@TomaSusi TomaSusi commented Oct 6, 2026

Copy link
Copy Markdown
Member

Combines the VASP documentation contributed by a tester with the existing GPAW tutorial, and renames the page from ab initio potentials with GPAW to ab initio potentials.

The page now covers all three routes to a DFT potential in one place:

takes core electrons from
GPAWPotential a converged GPAW calculator the GPAW setups
VASPPotential AECCAR2 + POTCAR the POTCAR radial core density
ChargeDensityPotential any valence density as a NumPy array an analytic Slater model

Structure

  • ab initio potentials — intro, pointing at the three routes
  • Potentials from a GPAW calculation — the existing hBN walkthrough, unchanged
  • Potentials from a VASP calculation — the contributed SrTiO₃ walkthrough: setting up the VASP run, loading the files, building and comparing the four potentials, diffraction, a thickness series
  • How accurate are the core models? — the contributor's validation table across SrTiO₃, hBN and MoS₂
  • Further reading

What changed in the GPAW half

Nothing, apart from the headings being demoted one level under the new section — 21 of its 28 cells are byte-identical — and one fix. The citation of the experimental intensities was written ​`cite`{gpawpotential}, with the braces and backticks the wrong way round and the key missing its final s, so it rendered as literal text instead of a citation. It now resolves.

What changed in the contributed half

The code is as the contributor wrote it. The edits are editorial:

  • headings demoted to sit under the VASP section
  • references to "the chapter ab initio potentials with GPAW" rewritten, since it is now a section of the same page
  • abTEM spelled *ab*TEM, as elsewhere in the docs
  • the four wide figures tagged full-width
  • a dead material_name assignment dropped
  • the import cell reduced to the VASP-specific imports and left visible, since numpy/matplotlib/abtem are already imported in the hidden setup cell at the top

Notes

  • The file name stays charge_density.ipynb, so the published URL and the (walkthrough:charge_density) label referenced from the potentials walkthrough keep working. Only the title changes.
  • Notebooks are not executed at build time (execute_notebooks: 'off'). That is what makes the VASP half possible: POTCAR files are under the VASP license, so nothing executable can be shipped. Its cells carry the contributor's stored outputs and read from path/to/... placeholders, which the text says explicitly.
  • jupyter-book build docs/ is clean for this page — no new warnings, and the three new cross-references resolve.
  • VASPPotential and ChargeDensityPotential(valence_electrons=...) are not released yet; they are in abTEM#395 and abTEM#422. This page should land after those.
  • The VASP material is someone else's work — you may want to add them as a co-author on the commit, or credit them on the page, before merging. I did not guess at a name or address.

🤖 Posted by Claude Code on behalf of Toma

The tutorial is renamed from "*ab initio* potentials with GPAW" to
"*ab initio* potentials", and the VASP material is added as a second half,
from a notebook contributed by a user who tested VASPPotential and
ChargeDensityPotential.

The page now covers all three routes to a DFT potential -- GPAWPotential,
VASPPotential, and ChargeDensityPotential for any other code -- and closes
with a section comparing their accuracy against converged GPAW references on
SrTiO3, hBN and MoS2.

The GPAW half is unchanged apart from being demoted one heading level under a
new "Potentials from a GPAW calculation" section, and one fix: the citation of
the experimental intensities was written `cite`{gpawpotential}, with the
braces and backticks the wrong way round and the key missing its final s, so
it rendered as literal text rather than a citation.

The VASP half keeps the contributor's code as written. The changes are
editorial: headings demoted to fit under "Potentials from a VASP calculation",
cross-references to "the chapter *ab initio potentials with GPAW*" rewritten
now that it is a section of the same page, abTEM spelled *ab*TEM as elsewhere
in the docs, the wide figures tagged full-width, a dead material_name
assignment dropped, and the import cell reduced to the VASP-specific imports,
since numpy/matplotlib/abtem are already imported in the hidden setup cell at
the top.

The file name stays charge_density.ipynb so the published URL and the
(walkthrough:charge_density) label referenced from the potentials walkthrough
keep working.

The notebook is not executed at build time (execute_notebooks: 'off'), which
is what makes the VASP half possible at all: POTCAR files are under the VASP
license, so its cells carry the contributor's stored outputs and read from
path/to/... placeholders.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

This branch has not been deployed

No deployments
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

Successfully merging this pull request may close these issues.

1 participant