ORCHID data: umbrella record of the ORCHID datasets

ORCHID data (Open Repository for CALPHAD Hypotheses, Intelligence, and Discovery), release v2026.10: umbrella record of eight citable datasets, each a separate Zenodo record with its own DOI (the ESPEI datasets are derived from ORCHID-Values and deposited apart because they are used apart):

• ORCHID-Values: thermodynamic and phase-equilibrium values read from the literature (doi:10.5281/zenodo.23735569): 73,542 values from 1,321 papers
• ORCHID ESPEI datasets: literature and database-derived thermodynamic fitting inputs (doi:10.5281/zenodo.23735788): 6,453 ESPEI datasets
• ORCHID-Diagrams: calculated binary phase diagrams (doi:10.5281/zenodo.23735859): 2,722 calculated SGTE binary diagrams
• ORCHID-Hulls: binary convex hulls (Materials Project, OQMD, assessed CALPHAD) (doi:10.5281/zenodo.23735891): 3,828 MP and 3,916 OQMD binary hulls, 2,695 SGTE descriptions
• ORCHID-Parity: DFT-vs-experiment formation-enthalpy benchmark (doi:10.5281/zenodo.23735943): 7,646 DFT-vs-experiment points
• ORCHID-Registry: phase registry of binary systems (doi:10.5281/zenodo.23735970): 10,430 phases in 1,241 systems
• ORCHID-Assemblages: assessed phase assemblages of binary systems (doi:10.5281/zenodo.23735996): 23,129 points in 2,178 assessed diagrams
• ORCHID-Bibliography: papers and typed citation relations (doi:10.5281/zenodo.23736021): 9,570 works, 60,235 typed edges

The runnable workflows that use the release are archived as a software record: ORCHID examples: runnable workflows on the ORCHID data (PhaseForge+, ESPEI, one-shot MLIP phase diagrams, MLIP formation enthalpies, quickstart) (doi:10.5281/zenodo.23736043).

This umbrella links eight dataset records and the examples software archive within one versioned ORCHID release; several records are derived from the same observations and should not be treated as independent evidence.

Status. Version 1, machine-extracted: values were read by language-model readers constrained to quote their source and checked automatically, and the release was then audited on the printed pages by adversarial audits, an independent second review of fit-facing records and a preliminary human review of a frozen sample. Flags such as accepted and ESPEI-ready record screening results, not certification; known issues are listed in the release notes.

This record holds the release documentation: README, CHANGELOG, DATASET_CARD (datasheet with loading examples for pandas, DuckDB and Hugging Face datasets), licences and attributions, the Parquet manifest, and the Zenodo metadata of all ten records with the list of every number in it (zenodo/metadata_numbers.json). The data files are in the eight dataset records.

Limitations

See each dataset record and DATASET_CARD.md (known limitations). Literature values are machine-extracted (Version 1); calculated and DFT values are model results, not measurements.

Licence

ORCHID's data are released under CC BY 4.0, as are the SGTE (B. Hallstedt, Calphad 89 (2025) 102833), Materials Project and OQMD data from which they are derived; attributions and statements of changes are in NOTICE.md, the licence of each folder in LICENSES.md (both in the bundle). The quotations from papers that accompany values are short excerpts kept for verification; they remain under the publishers' copyright and are not relicensed. Source papers, page images and full texts are not redistributed. Kim et al. 2017 cross-check values: MIT (NOTICE.md); Crossref metadata in bibliography/: CC0.

The stated licence covers the ORCHID-produced content in this deposit; upstream data and software retain their documented terms and attribution requirements, and no publisher pages, handbook drawings or restricted TDB source files are included.

Files: one zip (orchid-data-v2026.10.zip) holding the files at their repository paths, a MANIFEST.sha256, the README, CHANGELOG and licence files. Its file list with sizes and SHA-256 is zenodo/orchid-data.files.json in the data repository.

How to cite

R. Arróyave, J. Saal, D. Shin, ORCHID data: umbrella record of the ORCHID datasets (version 2026.10; Version 1, machine-extracted), Zenodo (2026). https://doi.org/10.5281/zenodo.23736065

Also cite the original source of every value you use (paper_citation / paper_doi, and cited_dois for values the paper cites from earlier work); B. Hallstedt, The SGTE collection of binary datasets, Calphad 89 (2025) 102833, https://doi.org/10.1016/j.calphad.2025.102833 (CC BY 4.0), and the original assessment named in each file; A. Jain et al., The Materials Project: A materials genome approach to accelerating materials innovation, APL Materials 1 (2013) 011002, https://doi.org/10.1063/1.4812323 (CC BY 4.0); J. E. Saal et al., JOM 65 (2013) 1501, https://doi.org/10.1007/s11837-013-0755-4 and S. Kirklin et al., npj Comput. Mater. 1 (2015) 15010, https://doi.org/10.1038/npjcompumats.2015.10 (CC BY 4.0).

@misc{orchid_data_2026,
  author    = {Arr{\'o}yave, Raymundo and Saal, James and Shin, Dongwon},
  title     = {{ORCHID data: umbrella record of the ORCHID datasets}},
  version   = {2026.10},
  publisher = {Zenodo},
  year      = {2026},
  doi       = {10.5281/zenodo.23736065},
  url       = {https://doi.org/10.5281/zenodo.23736065},
  note      = {Version 1, machine-extracted}
}

Software and data citations

These credits describe the ORCHID project and its examples. Cite the papers and software used in your analysis, together with the ORCHID release and original source of each observation. The list does not imply that every dataset was generated with every package. Preserve the versions and commit hashes recorded by each example; a repository link is not a reproducibility pin. Materials Project and OQMD energies retain separate references.

• pycalphad: Thermodynamic calculations and phase equilibria. R. Otis and Z.-K. Liu (2017), pycalphad: CALPHAD-based Computational Thermodynamics in Python (10.5334/jors.140). Repository.
• ESPEI: Thermodynamic parameter fitting and uncertainty quantification. B. Bocklund et al. (2019), ESPEI for efficient thermodynamic database development, modification, and uncertainty quantification: application to Cu–Mg (10.1557/mrc.2019.59). Repository.
• PhaseForge: MLIP-informed CALPHAD construction in the examples. S. Zhu, D. Sarıtürk, and R. Arróyave (2025), Machine learning potentials for alloys: a detailed workflow to predict phase diagrams and benchmark accuracy (10.1038/s41524-025-01814-z). Repository; archived software.
• PhaseForge+: Construction and experimental tuning of CALPHAD models in the examples; repository name PhaseForgePlus. C. Kunselman, S. Zhu, D. Sarıtürk, and R. Arróyave (2025), Construction and Tuning of CALPHAD Models Using Machine-Learned Interatomic Potentials and Experimental Data: A Case Study of the Pt–W System (10.1007/s11669-025-01222-2). Repository; archived software.
• MaterialsFramework: Uniform interface to machine-learning interatomic potentials (single-point energies, relaxations, molecular dynamics) used by PhaseForge and directly in the ORCHID examples and MLIP benchmarks. D. Sarıtürk (2026), MaterialsFramework (v1.0.0), Zenodo (10.5281/zenodo.15731044). Repository; archived software.
• Materials Project: DFT structures and energies; repository link is the API client, not the database itself. A. Jain et al. (2013), Commentary: The Materials Project: A materials genome approach to accelerating materials innovation (10.1063/1.4812323). Repository; data resource.
• OQMD: DFT structures and energies; qmpy is the supporting software. J. E. Saal et al. (2013), Materials Design and Discovery with High-Throughput Density Functional Theory: The Open Quantum Materials Database (OQMD) (10.1007/s11837-013-0755-4); S. Kirklin et al. (2015), The Open Quantum Materials Database (OQMD): assessing the accuracy of DFT formation energies (10.1038/npjcompumats.2015.10). Repository; data resource.

Browse: orchid-project.org · data site data.orchid-project.org · source github.com/orchid-calphad/orchid-data.

This work was supported by the U.S. National Science Foundation under awards OAC-2608825 and 2534344.