Conversation
Companion page for the simulation notebook added in api-examples: what the manuscript (Dahal & Batzill, Nanoscale 6, 2548 (2014)) reports about the high-symmetry registries of graphene on Ni(111), which quantities are comparable across methods and which are not, the two-tier recipe (MACE-MP + D3 survey, then Total Energy jobs per registry), and the computational settings next to the manuscript's. Registered in both navs and linked from the index guide's Dahal row. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
✅ Deploy Preview for mat3ra-documentation ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
The settings table documented the combination that inverts the result: MACE medium where the notebook uses large, and vdw_corr = "d3_grimme" where the notebook sets "grimme-d3". The medium model at float32 misses the shallow chemisorbed minimum entirely and reports every registry as physisorbed, so the page now says which model is required and why. Covers all four of the review's Fig. 1 registries under its own names, with the figure shown, and states plainly what is not claimed: which of the two atop registries is lowest is below the resolution of either method here. Documents the two same-cell reference jobs and the adsorption energy they produce. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The bridge registry is a C-C bond straddling a first-layer Ni, not a carbon on the Ni-Ni midpoint. Cutoffs follow the notebook at 40/320 Ry. The adsorption-energy references are described as off by default, and what their same-cell arrangement does and does not cancel is stated plainly. Job counts now match the notebook: four registries, and the two reference jobs add to whatever is active rather than being part of a default run. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Cutoffs corrected to the 40 / 200 Ry pair GBRV publishes for its ultrasoft set. The settings section now says, for every value, whether it is a platform default, the pseudopotential set's published value, or something this system's physics requires — including what the SCF defaults did on a first attempt and which three changes address it. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The page now leads with Lahiri et al. Table 1 as the quantitative target, states the published recipe (LDA, spin-polarized, relaxation with the bottom substrate layers fixed) and why relaxation is not optional, explains that the fast tier is expected to fail the energetic targets and why that is the physics working as documented, and lists the divergences from the paper in one place. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
VsevolodX
marked this pull request as ready for review
September 3, 2026 00:40
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The page claimed the fast tier "runs MACE-MP + D3 in the browser" and is "good for the geometry survey"; measured, it buckles the atop carbon the wrong way, sits 0.2 A too close, and the browser embed has no torch-dftd, so the dispersion-bound hollow reports itself unbound there. The page now states the MACE numbers per environment beside the paper's and what the tier is for: the registry set, the two-branch landscape, and starting geometries. The precise tier is described as it is now built — one fixed-cell relaxation per registry with the final structure read back, graphene reference unpolarized — and the verdict section shows the line's form rather than a "yes" no run has produced. Lahiri et al. (2011), whose Table 1 is the target, gets a bibliography entry and citations. The artifact carries one name everywhere: "Gr/Ni(111) Registry and Work of Adhesion" — the headline quantity — in the page, both navs and the index. Nav padding realigned with its neighbours; the tier list is prose, per the repo's flat-list rule. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
VsevolodX
changed the title
The fast tier relaxes positions along z only, with the deepest substrate layers fixed, and the two same-cell references are relaxed under the same constraint. This equals a full relaxation at the paper's symmetric registries because in-plane forces vanish there by symmetry; for the bridge registry, an in-plane saddle, it is what keeps the point defined. The page previously said the fast tier was relaxed "with the bottom substrate layers fixed — the paper's scheme," which reads as unconstrained in-plane, letting the film slide out of registry. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The registry re-check cannot fire under z-only relaxation, since every atom's xy is fixed; say where it actually matters (the precise tier, which relaxes all coordinates) instead of describing a guard the fast tier can't exercise. Carried over from the api-examples notebook fix. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Aligns with the notebook's change: the final cell no longer prints a
pass/fail verdict per tier, just the computed values beside the
published ones. Drops the "[MACE tier] verdict" mention in 5.3 and
rewrites 5.5 ("Read the verdict" -> "Read the final table").
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
timurbazhirov
approved these changes
Sep 13, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Tutorial page for the simulation notebook in mat3ra/api-examples#364 — SOF-8043, epic SOF-7994.
Follows the shape established by the twisted-MoS₂ simulation page: front matter with the M-CODE tag, the manuscript in a
!!!noteblock citing[@Dahal2014](key already present inreferences.bib), the structure tutorial linked as a prerequisite, what is compared and what deliberately is not, a settings table against the manuscript's, step-by-step instructions, and the JupyterLite embed.Registered in
mkdocs.ymlandmkdocs-guide.ymlnext to the structure page, and the index guide's Dahal row now links the simulation page instead of reading "Coming Soon".Two points the page makes explicitly, because both are easy to get wrong:
Draft alongside the api-examples PR, whose DFT tier is blocked on a local messaging fault (see that PR).
🤖 Generated with Claude Code