Pith. sign in

REVIEW 2 major objections 5 minor 294 references

VASP can now run user Python plugins that edit structure, forces, potential and occupancies through shared-memory NumPy views, without rewriting the Fortran core.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · grok-4.5

2026-07-11 16:16 UTC pith:SLC3XQ34

load-bearing objection Solid, officially integrated zero-copy Python hooks for VASP; useful engineering, moderate novelty, ordinary caveats on mid-SCF safety and missing overhead numbers. the 2 major comments →

arxiv 2607.04622 v1 pith:SLC3XQ34 submitted 2026-07-06 cond-mat.mtrl-sci

VASP Plugins: Linking the Vienna ab-initio Simulation Package with Python

Sudarshan Vijay , Martijn Marsman , Georg Kresse , Martin Schlipf This is my paper
classification cond-mat.mtrl-sci
keywords VASPPython pluginsDFTshared memorypybind11NumPySCF hooksmaterials modelling
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

Production DFT codes such as VASP are written in Fortran and tuned for speed on large computers, so adding a new physical model or algorithm usually means deep surgery on legacy source. This paper shows a practical way around that barrier: a thin C++/pybind11 layer that hands selected VASP arrays to Python as live NumPy views. Users write ordinary Python functions that receive two dataclasses—one of read-only constants and one of quantities they are allowed to change—and can therefore inject structure updates, force/stress corrections, local-potential shifts or occupancy changes at defined points in the SCF or ionic cycle. Three working examples (scipy-driven ionic relaxation, an implicit solvent model, and DFT-D4 dispersion) demonstrate that the same infrastructure can host both generic numerical libraries and specialised physical corrections while the heavy electronic-structure work stays inside the optimised Fortran engine. The result is a bridge that lets the scientific Python ecosystem extend a production DFT code without file I/O round-trips or source-code forks.

Core claim

A plugin architecture that exposes VASP data as shared-memory NumPy arrays through a C++ intermediate layer lets users modify structure, forces and stress, local potential and occupancies from Python at fixed SCF and ionic hooks, without data copies and without editing the Fortran source.

What carries the argument

The constants/additions split plus shared-memory NumPy views: unmodifiable data are frozen, modifiable data are changed only by in-place +=/-= operations on buffers that Fortran and Python already share, so there is no copy overhead and the Fortran side never sees an inconsistent state.

Load-bearing premise

The chosen hooks and the constants-versus-additions split are enough to support the intended class of extensions without breaking SCF convergence or shared-memory consistency under parallel runs.

What would settle it

A mid-SCF local_potential or occupancies plugin that systematically prevents SCF convergence, or a parallel run in which the shared-memory NumPy views become inconsistent with the Fortran arrays, would falsify the claim that the architecture is both safe and sufficient.

Watch this falsifier — get emailed when new claim-graph text bears on it.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

2 major / 5 minor

Summary. The manuscript introduces a Python plugin infrastructure for VASP that inserts C++/pybind11 hooks into the Fortran code, exposing selected quantities (structure, forces/stress, local potential, occupancies, charge density, etc.) as zero-copy NumPy views via shared-memory buffers and ISO_C_BINDING. Users supply a vasp_plugin.py file containing named functions that receive frozen Constants and mutable Additions dataclasses; modifications are applied only through +=/-= at defined end-of-SCF or mid-SCF points. Three illustrations demonstrate the interface: scipy-based ionic relaxation via the structure plugin, a self-consistent continuum solvent model via local_potential, and DFT-D4 dispersion corrections via force_and_stress. The feature is stated to be available in VASP 6.5.0.

Significance. If the interface is robust, it meaningfully lowers the barrier to rapid prototyping of new algorithms inside a production DFT code that is otherwise difficult to modify, while preserving the performance of the optimized Fortran kernels. The zero-copy shared-memory design, the clean constants/additions split, and the three working end-to-end examples (including an external ASE calculator wrapper) are concrete engineering strengths that make the contribution immediately usable by the materials-modeling community. The work therefore has clear practical value for method development that sits between pure black-box wrappers (ASE/AiiDA) and full source-code forks.

major comments (2)
  1. [II Architecture / III.B] Section II and the mid-SCF plugins (local_potential, occupancies): the architecture description and the three illustrations establish that the hooks fire and produce qualitatively expected results, but the manuscript does not discuss or test whether arbitrary modifications to the potential or occupancies preserve SCF convergence, charge neutrality, or consistency under VASP’s MPI domain decomposition. Because the central claim is that the infrastructure is safe and useful for experimental algorithms, a short paragraph on known limitations, recommended usage patterns, and any internal consistency checks would strengthen the paper without requiring new theory.
  2. [III.A / V] Section III.A and Computational Methods: the scipy CG/BFGS runs require substantially more ionic steps than VASP’s native optimizers (Fig. 2), yet no wall-time or per-call overhead measurements are reported for the plugin layer itself. A brief quantification of the Python-call cost (even for a single SCF cycle) would let readers judge whether the convenience is free or carries a measurable price, which is load-bearing for the “high performance without data duplication” claim.
minor comments (5)
  1. [global] Throughout: several formatting artifacts remain from the arXiv conversion (e.g., “V ASP”, “ILLUSTRA TIONS”, “form EuroHPC”, spaced identifiers in Listings 1–4). These should be cleaned for the journal version.
  2. [II] Figure 1 caption and surrounding text: the schematic is helpful but does not indicate which quantities are pointers versus copies, nor how the buffer is synchronized across MPI ranks. A one-sentence clarification would improve reproducibility.
  3. [III.A] Listing 2: the StopIteration control-flow pattern for driving scipy.minimize is clever but non-obvious; a short comment or reference to the intended usage pattern would help new users.
  4. [I / IV] References: related hybrid Fortran–Python efforts in other electronic-structure packages (e.g., GPAW’s Python core, Quantum ESPRESSO’s Python bindings, or ASE calculators that already wrap VASP) are only lightly cited; a brief comparison paragraph would better situate the contribution.
  5. [V] Section V: the Open Catalyst structure index and the precise SCCS parameters (rho_min/max) are given, but the graphite interlayer-distance curve (Fig. 4) would benefit from an explicit statement of the energy zero and whether zero-point or thermal corrections are included.

Circularity Check

0 steps flagged

No circularity: software-architecture paper with external demos, not a fitted or self-definitional derivation chain.

full rationale

The manuscript claims a C++/pybind11 shared-memory plugin layer that exposes selected VASP quantities as NumPy views so user Python functions can modify structure, forces/stress, local potential, and occupancies at defined SCF/ionic hooks without copies or Fortran edits. That claim is established by the architecture description (ISO_C_BINDING buffers, constants/additions split, frozen dataclasses, +=/-= mutation) and by three working illustrations that simply call established external packages (scipy.optimize, an SCCS dielectric/Poisson implementation, DFT-D4 via ASE). No parameter is fitted to data and then re-presented as a prediction; no uniqueness theorem or ansatz is imported from the authors’ prior work to force the result; no known empirical pattern is renamed as a new first-principles derivation. The graphite interlayer spacing, force-norm curves, and solvent potential plots are demonstrations that the hooks fire and return plausible numbers, not circular predictions. Circularity burden is therefore zero.

Axiom & Free-Parameter Ledger

0 free parameters · 3 axioms · 1 invented entities

The central claim is an engineering claim about an interface, not a physical derivation. It rests on standard language-interop assumptions and on the design choice that a small set of hooks plus a constants/additions split is adequate. No free parameters are fitted to produce a scientific prediction; the only “parameters” are ordinary DFT settings used in the demos.

axioms (3)
  • domain assumption ISO_C_BINDING and pybind11 correctly expose Fortran arrays as contiguous NumPy views without hidden copies or lifetime issues under the MPI/OpenMP decompositions used by VASP.
    Stated as the foundation of the shared-memory design in §II; correctness of the binding layer is assumed rather than formally verified.
  • ad hoc to paper Modifying only the declared “additions” quantities at the published hooks leaves the rest of the VASP SCF and ionic algorithms numerically well-behaved.
    Implicit in the plugin contract (§II and the three illustrations); no general proof of SCF stability under arbitrary mid-cycle potential or occupancy changes is given.
  • domain assumption Standard DFT approximations (PBE, PAW pseudopotentials, Gaussian smearing) remain valid when external Python corrections are added.
    Used throughout the computational methods and demos; inherited from the broader DFT literature.
invented entities (1)
  • constants / additions dataclass pair and the four named plugin hooks (structure, force_and_stress, local_potential, occupancies) independent evidence
    purpose: Provide a controlled, type-safe surface for Python to read immutable data and write only intended modifications back into VASP.
    These are software abstractions introduced by the paper; they have no physical content beyond the interface contract.

pith-pipeline@v1.1.0-grok45 · 14188 in / 2683 out tokens · 20179 ms · 2026-07-11T16:16:39.480880+00:00 · methodology

0 comments
read the original abstract

Implementing novel features and experimental algorithms into widely adopted density functional theory (DFT) codes is frequently hindered by complex legacy architectures and the use of compiled languages such as Fortran. These production codes, while optimised for high-performance computing clusters, present significant hurdles for software development and rapid prototyping, often requiring deep expertise in the code's internal structure to modify. To address this challenge, we present a Python plugin infrastructure for the Vienna ab-initio Simulation Package (VASP) that combines computational efficiency with the flexibility of high-level scripting. Our architecture uses a C++ intermediate layer and pybind11 to expose VASP data as NumPy arrays via shared memory buffers, ensuring high performance without data duplication. We implement two categories of plugins: those that modify quantities at the end of each converged self-consistent field (SCF) cycle, such as structure and force_and_stress, and those that operate during the SCF cycle, such as local_potential and occupancies. We demonstrate the utility of our implementation through three applications, structure relaxation using the scipy library, implementing an implicit solvent model, and adding the DFT-D4 dispersion corrections. This infrastructure effectively bridges the gap between high-performance electronic structure routines and the widespread scientific Python ecosystem.

Figures

Figures reproduced from arXiv: 2607.04622 by Georg Kresse, Martijn Marsman, Martin Schlipf, Sudarshan Vijay.

Figure 1
Figure 1. Figure 1: FIG. 1. Architecture of the Fortran-Python interface. Data is loaded into a buffer in the Fortran layer and exposed to [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. Comparison of the norm of the forces between existing [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. SCCS generated dielectric (orange), charge density (green) and computed vacuum (dashed blue) and solvent (solid [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4. Change in the total energy with graphite interlayer [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

294 extracted references · 208 canonical work pages · 30 internal anchors

  1. [1]

    Schaipp, Fabian , month = jul, year =. How to. doi:10.48550/arXiv.2607.01487 , abstract =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2607.01487

  2. [2]

    and Bernstein, Noam and Csányi, Gábor , month = jun, year =

    Tompa, Tamás Lajos and Varga-Umbrich, Eszter and Batatia, Ilyes and Elena, Alin M. and Bernstein, Noam and Csányi, Gábor , month = jun, year =. Fine-tuning. doi:10.48550/arXiv.2606.12704 , abstract =

    Pith/arXiv arXiv doi:10.48550/arxiv.2606.12704

  3. [3]

    and Van Voorhis, Troy , month = jul, year =

    Abraham, Ethan and Bazant, Martin Z. and Van Voorhis, Troy , month = jul, year =. Ab. Physical Review Letters , publisher =. doi:10.1103/dct7-q652 , abstract =

    work page doi:10.1103/dct7-q652

  4. [4]

    and Transtrum, Mark K

    Kurniawan, Yonatan and Wen, Mingjian and Tadmor, Ellad B. and Transtrum, Mark K. , month = jun, year =. Comparative study of ensemble-based uncertainty quantification methods for neural network interatomic potentials , url =. doi:10.48550/arXiv.2508.06456 , abstract =

    doi:10.48550/arxiv.2508.06456

  5. [5]

    and Guo, Xiaoxi and Zhang, Guangjie and Zhang, Yanning and Wang, Zhiming and Song, Li and Liu, Zheng and Xiong, Jie and Honkala, Karoliina , month = jun, year =

    Wu, Tongwei and Melander, Marko M. and Guo, Xiaoxi and Zhang, Guangjie and Zhang, Yanning and Wang, Zhiming and Song, Li and Liu, Zheng and Xiong, Jie and Honkala, Karoliina , month = jun, year =. Potential-dependent polaron formation activates. doi:10.26434/chemrxiv-2024-j7thd , abstract =

    work page doi:10.26434/chemrxiv-2024-j7thd 2024

  6. [6]

    and Huang, Jun , month = jun, year =

    Zhang, Mengke and Chen, Yanxia and Melander, Marko M. and Huang, Jun , month = jun, year =. Electrochemical. Chemical Reviews , publisher =. doi:10.1021/acs.chemrev.5c00926 , abstract =

    work page doi:10.1021/acs.chemrev.5c00926

  7. [7]

    Nature , author =

    Fluctuation-induced quantum friction in nanoscale water flows , volume =. Nature , author =. 2022 , note =. doi:10.1038/s41586-021-04284-7 , abstract =

    doi:10.1038/s41586-021-04284-7 2022

  8. [8]

    Solé, Àlex and Suárez-Dou, Sergio and Mosella-Montoro, Albert and Gómez-Coca, Silvia and Ruiz, Eliseo and Tkatchenko, Alexandre and Ruiz-Hidalgo, Javier , month = may, year =. Machine. doi:10.48550/arXiv.2605.25710 , abstract =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2605.25710

  9. [9]

    Communications Chemistry , author =

    Reaction-conditioned generative model for catalyst design and optimization with. Communications Chemistry , author =. 2025 , keywords =. doi:10.1038/s42004-025-01732-7 , abstract =

    work page doi:10.1038/s42004-025-01732-7 2025

  10. [10]

    The Journal of Physical Chemistry Letters , author =

    Measuring the. The Journal of Physical Chemistry Letters , author =. 2024 , keywords =. doi:10.1021/acs.jpclett.4c01303 , abstract =

    work page doi:10.1021/acs.jpclett.4c01303 2024

  11. [11]

    , month = mar, year =

    Sahu, Sumit and Delibas, Berk and Dawlaty, Jahan M. , month = mar, year =. Ions as. The Journal of Physical Chemistry Letters , publisher =. doi:10.1021/acs.jpclett.6c00561 , abstract =

    work page doi:10.1021/acs.jpclett.6c00561

  12. [12]

    and Dawlaty, Jahan M

    Maitra, Anwesha and Das, Pratyusha and Thompson, Barry C. and Dawlaty, Jahan M. , month = mar, year =. Distinguishing between the. The Journal of Physical Chemistry B , publisher =. doi:10.1021/acs.jpcb.2c08030 , abstract =

    doi:10.1021/acs.jpcb.2c08030

  13. [13]

    Physical Review E , author =

    Uncertainty in. Physical Review E , author =. 2026 , note =. doi:10.1103/6dxn-p9nt , abstract =

    work page doi:10.1103/6dxn-p9nt 2026

  14. [14]

    Hoellmer, Philipp and Martiniani, Stefano , month = jun, year =. Open. doi:10.48550/arXiv.2602.00424 , abstract =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2602.00424

  15. [15]

    Six Open Questions in Machine-Learned Interatomic Potential Foundation Models

    Creed, Isabel and Rein, Tim and Vitenburgs, Ingvars and Stark, Wojciech G. and Ellingsson, Viktor and Ismail, Ahmed Y. and Liu, Guangyu and Lou, Yuchen and Martin, Bradley A. A. and Bone, Cyprien and Walker, Matthew A. H. and Taj, Mueen and Wang, Shirui and Wong, Kelvin and Wu, Ruiqi and Kayastha, Prakriti and Cheng, Bingqing and Krishnapriyan, Aditi and ...

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2606.07327

  16. [16]

    Parkar, Poonam and Vijay, Sudarshan and Chatterjee, Abhijit , month = dec, year =. On-the-. 2025. doi:10.1109/HiPCW66559.2025.00033 , abstract =

    work page doi:10.1109/hipcw66559.2025.00033 2025

  17. [17]

    Kim, Heemoon and Park, Kwangheon and Kim, Bong Goo and Choo, Yong Sun and Kim, Keon Sik and Song, Kun Woo and Hong, Kwon Pyo and Kang, Young Hwan and Ho, Kwangil , month = jul, year =. Xenon. Nuclear Technology , publisher =. doi:10.13182/NT04-A3521 , abstract =

    work page doi:10.13182/nt04-a3521

  18. [18]

    Journal of Nuclear Materials , author =

    Transport properties of. Journal of Nuclear Materials , author =. 2009 , pages =. doi:10.1016/j.jnucmat.2009.03.038 , abstract =

    work page doi:10.1016/j.jnucmat.2009.03.038 2009

  19. [19]

    Enthalpy and heat capacity of the actinide oxides

  20. [20]

    Journal of Nuclear Materials , author =

    Coated particle fuel:. Journal of Nuclear Materials , author =. 2019 , keywords =. doi:10.1016/j.jnucmat.2018.09.044 , abstract =

    work page doi:10.1016/j.jnucmat.2018.09.044 2019

  21. [21]

    , month = jul, year =

    Reutler, Herbert and Lohnert, Günter H. , month = jul, year =. The. Nuclear Technology , publisher =. doi:10.13182/NT83-A33228 , abstract =

    work page doi:10.13182/nt83-a33228

  22. [22]

    Chemical Reviews , author =

    Thermal. Chemical Reviews , author =. 2022 , pages =. doi:10.1021/acs.chemrev.1c00262 , language =

    work page doi:10.1021/acs.chemrev.1c00262 2022

  23. [23]

    and Basu, M

    Das, D. and Basu, M. and Kolay, S. and Shirsat, A. N. , editor =. Transport. Thoria-based. 2013 , keywords =. doi:10.1007/978-1-4471-5589-8_5 , abstract =

    work page doi:10.1007/978-1-4471-5589-8_5 2013

  24. [24]

    Journal of Alloys and Compounds , author =

    The solidification behaviour of the. Journal of Alloys and Compounds , author =. 2014 , keywords =. doi:10.1016/j.jallcom.2014.07.055 , abstract =

    work page doi:10.1016/j.jallcom.2014.07.055 2014

  25. [25]

    Journal of Nuclear Materials , author =

    Melting behavior of (. Journal of Nuclear Materials , author =. 2016 , pages =. doi:10.1016/j.jnucmat.2016.06.037 , abstract =

    work page doi:10.1016/j.jnucmat.2016.06.037 2016

  26. [26]

    Energy Procedia , author =

    Advanced. Energy Procedia , author =. 2011 , keywords =. doi:10.1016/j.egypro.2011.06.005 , abstract =

    work page doi:10.1016/j.egypro.2011.06.005 2011

  27. [27]

    Vijayan, P. K. and Dulera, I. V. and Krishnani, P. D. and Vaze, K. K. and Basu, S. and Sinha, R. K. , editor =. Overview of the. Thorium. 2016 , keywords =. doi:10.1007/978-3-319-26542-1_10 , abstract =

    work page doi:10.1007/978-3-319-26542-1_10 2016

  28. [28]

    International Journal of Energy Research , author =

    Review of the microheterogeneous thoria-urania fuel for micro-sized high temperature reactors , volume =. International Journal of Energy Research , author =. 2021 , note =. doi:10.1002/er.5923 , abstract =

    work page doi:10.1002/er.5923 2021

  29. [29]

    Energy Policy , author =

    Abundant thorium as an alternative nuclear fuel:. Energy Policy , author =. 2013 , keywords =. doi:10.1016/j.enpol.2013.04.062 , abstract =

    work page doi:10.1016/j.enpol.2013.04.062 2013

  30. [31]

    Nature Catalysis , author =

    Promoting. Nature Catalysis , author =. 2023 , pages =. doi:10.1038/s41929-023-01010-6 , language =

    work page doi:10.1038/s41929-023-01010-6 2023

  31. [32]

    Küngas, Rainer , month = feb, year =. Review—. Journal of The Electrochemical Society , publisher =. doi:10.1149/1945-7111/ab7099 , abstract =

    work page doi:10.1149/1945-7111/ab7099 1945

  32. [33]

    Journal of the American Chemical Society , author =

    Step. Journal of the American Chemical Society , author =. 2025 , pages =. doi:10.1021/jacs.4c17263 , language =

    work page doi:10.1021/jacs.4c17263 2025

  33. [34]

    H., Mohith and Vijay, Sudarshan , year =. opt-. doi:10.48550/ARXIV.2604.10984 , abstract =

    Pith/arXiv arXiv doi:10.48550/arxiv.2604.10984

  34. [35]

    Swenson, David W. H. and Prinz, Jan-Hendrik and Noe, Frank and Chodera, John D. and Bolhuis, Peter G. , month = feb, year =. Journal of Chemical Theory and Computation , publisher =. doi:10.1021/acs.jctc.8b00626 , abstract =

    doi:10.1021/acs.jctc.8b00626

  35. [36]

    Journal of Chemical Theory and Computation , author =

    Replica. Journal of Chemical Theory and Computation , author =. 2025 , pages =. doi:10.1021/acs.jctc.5c00588 , language =

    work page doi:10.1021/acs.jctc.5c00588 2025

  36. [37]

    Boltzmann generators:

    Noé, Frank and Olsson, Simon and Köhler, Jonas and Wu, Hao , month = sep, year =. Boltzmann generators:. Science , publisher =. doi:10.1126/science.aaw1147 , abstract =

    doi:10.1126/science.aaw1147

  37. [38]

    Kam, Lucas B. T. de and Zhu, Jia-Xin and Mathanker, Ankit and Doblhoff-Dier, Katharina and Govindarajan, Nitish , month = feb, year =. Benchmarking short-range machine learning potentials for atomistic simulations of metal/electrolyte interfaces , url =. doi:10.48550/arXiv.2602.22931 , abstract =

    doi:10.48550/arxiv.2602.22931

  38. [39]

    Science , author =

    Ionic. Science , author =. 2011 , pages =. doi:10.1126/science.1209786 , abstract =

    work page doi:10.1126/science.1209786 2011

  39. [40]

    and Vinson, John and Stafford, Gery R

    Raciti, David and Schwarz, Kathleen A. and Vinson, John and Stafford, Gery R. , month = mar, year =. Compressive. The Journal of Physical Chemistry C , publisher =. doi:10.1021/acs.jpcc.2c00134 , abstract =

    work page doi:10.1021/acs.jpcc.2c00134

  40. [41]

    , month = feb, year =

    Li, Lang and Reuter, Karsten and Hörmann, Nicolas G. , month = feb, year =. Deciphering the. ACS Electrochemistry , publisher =. doi:10.1021/acselectrochem.4c00062 , abstract =

    work page doi:10.1021/acselectrochem.4c00062

  41. [42]

    Entropic contributions to the stability of electrochemically adsorbed anion layers on

    Schönig, Marco and Schuster, Rolf , year =. Entropic contributions to the stability of electrochemically adsorbed anion layers on. Physical Chemistry Chemical Physics , publisher =. doi:10.1039/D2CP04680F , language =

    work page doi:10.1039/d2cp04680f

  42. [43]

    Microcalorimetric

    Schönig, Marco and Schuster, Rolf , month = sep, year =. Microcalorimetric. The Journal of Physical Chemistry C , publisher =. doi:10.1021/acs.jpcc.4c03723 , abstract =

    work page doi:10.1021/acs.jpcc.4c03723

  43. [44]

    Zeng, Liang and Huang, Nan and Feng, Guang , month = oct, year =. Water-. Physical Review Letters , publisher =. doi:10.1103/79nl-455r , abstract =

    work page doi:10.1103/79nl-455r

  44. [46]

    and Pillai, Hemanth S

    Heenen, Hendrik H. and Pillai, Hemanth S. and Reuter, Karsten and Bukas, Vanessa J. , month = jul, year =. Exploring mesoscopic mass transport effects on electrocatalytic selectivity , volume =. Nature Catalysis , publisher =. doi:10.1038/s41929-024-01177-6 , abstract =

    work page doi:10.1038/s41929-024-01177-6

  45. [47]

    Current Opinion in Electrochemistry , author =

    Coupled cation and. Current Opinion in Electrochemistry , author =. 2026 , keywords =. doi:10.1016/j.coelec.2026.101852 , language =

    work page doi:10.1016/j.coelec.2026.101852 2026

  46. [48]

    , month = feb, year =

    Zhang, Yufan and Binninger, Tobias and Huang, Jun and Eikerling, Michael H. , month = feb, year =. Theory of. Physical Review Letters , publisher =. doi:10.1103/PhysRevLett.134.066201 , abstract =

    work page doi:10.1103/physrevlett.134.066201

  47. [49]

    Physical Review Letters , author =

    Machine. Physical Review Letters , author =. 2025 , keywords =. doi:10.1103/48ct-3jxm , language =

    work page doi:10.1103/48ct-3jxm 2025

  48. [50]

    Electrochimica Acta , author =

    On the origin of capacitance dispersion of rough electrodes , volume =. Electrochimica Acta , author =. 2000 , keywords =. doi:10.1016/S0013-4686(00)00574-0 , abstract =

    work page doi:10.1016/s0013-4686(00)00574-0 2000

  49. [51]

    PRX Energy , author =

    Classical. PRX Energy , author =. 2026 , keywords =. doi:10.1103/65xq-svmf , abstract =

    work page doi:10.1103/65xq-svmf 2026

  50. [52]

    , month = jul, year =

    Weaver, Michael J. , month = jul, year =. Potentials of. Langmuir , publisher =. doi:10.1021/la9801054 , abstract =

    work page doi:10.1021/la9801054

  51. [53]

    Electrochimica Acta , author =

    Potentials of zero total and zero free charge of platinum group metals , volume =. Electrochimica Acta , author =. 1975 , keywords =. doi:10.1016/0013-4686(75)90017-1 , abstract =

    work page doi:10.1016/0013-4686(75)90017-1 1975

  52. [54]

    Errors that matter:

    Kellner, Matthias and Hansen, Teitur and Bligaard, Thomas and Jacobsen, Karsten Wedel and Ceriotti, Michele , month = apr, year =. Errors that matter:. doi:10.48550/arXiv.2604.24607 , abstract =

    Pith/arXiv arXiv doi:10.48550/arxiv.2604.24607

  53. [55]

    doi:10.48550/arXiv.2604.27685 , abstract =

    Gouvêa, Rogério Almeida and Rignanese, Gian-Marco , month = apr, year =. doi:10.48550/arXiv.2604.27685 , abstract =

    Pith/arXiv arXiv doi:10.48550/arxiv.2604.27685

  54. [56]

    Revealing the

    Zhang, Riyi and Yan, Yong and Yi, Zhihao and Wang, Dong , month = apr, year =. Revealing the. Chemical Communications , publisher =. doi:10.1039/D6CC01805J , abstract =

    work page doi:10.1039/d6cc01805j

  55. [57]

    Flexible uncertainty calibration for machine-learned interatomic potentials

  56. [58]

    Allam, Omar and Wander, Brook and Kim, SungYeon and Plesch, Rudi and Sours, Tyler and Chu, Jia-Min and Ludwig, Thomas and Kim, Jiyoon and Wang, Rodrigo and Agarwal, Shivang and Rask, Alan and Fleury, Alexandre and Wang, Chuhong and Wildman, Andrew and Mustard, Thomas and Ryczko, Kevin and Abruzzo, Paul and Nish, A. J. and Singh, Aayush R. , month = apr, y...

    work page doi:10.1038/s41524-026-02099-6

  57. [59]

    and Gupta, Ankur and Govind Rajan, Ananth , month = nov, year =

    Sudhakar, Hari R. and Gupta, Ankur and Govind Rajan, Ananth , month = nov, year =. Near-electrode anion dehydration and field-dependent dielectric response govern double-layer capacitance , volume =. Physical Review E , publisher =. doi:10.1103/l12q-f95m , abstract =

    work page doi:10.1103/l12q-f95m

  58. [60]

    The Journal of Chemical Physics , author =

    Double-layer capacitance peaks:. The Journal of Chemical Physics , author =. 2025 , keywords =. doi:10.1063/5.0251548 , abstract =

    work page doi:10.1063/5.0251548 2025

  59. [61]

    Assessing the

    Hagopian, Arthur and Koper, Marc T M and Doblhoff-Dier, Katharina , month = mar, year =. Assessing the. doi:10.26434/chemrxiv.15001131/v1 , language =

    work page doi:10.26434/chemrxiv.15001131/v1

  60. [62]

    Manoj, Uday Sankar and Drew, Nicole and Dabo, Ismaila and Muechler, Lukas , month = apr, year =. Nudged. doi:10.48550/arXiv.2604.19964 , abstract =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2604.19964

  61. [63]

    Ríos-García, Martiño and Alampara, Nawaf and Gupta, Chandan and Mandal, Indrajeet and Mannan, Sajid and Aghajani, Ali Asghar and Krishnan, N. M. Anoop and Jablonka, Kevin Maik , month = apr, year =. doi:10.48550/arXiv.2604.18805 , abstract =

    Pith/arXiv arXiv doi:10.48550/arxiv.2604.18805

  62. [64]

    Chemical Engineering Science , author =

    Ageing of trisiloxane solutions , volume =. Chemical Engineering Science , author =. 2010 , keywords =. doi:10.1016/j.ces.2010.06.024 , abstract =

    work page doi:10.1016/j.ces.2010.06.024 2010

  63. [65]

    Journal of Agricultural and Food Chemistry , author =

    Performance and stability of the organosilicon surfactant. Journal of Agricultural and Food Chemistry , author =. 1991 , pages =. doi:10.1021/jf00001a041 , language =

    work page doi:10.1021/jf00001a041 1991

  64. [66]

    Etz, Brian and K

    Mifkovic, Maleigh and D. Etz, Brian and K. Shukla, Manoj and Vyas, Shubham , year =. Aqueous solution degradation pathways of trimethylsiloxane surfactants , volume =. Environmental Science: Advances , publisher =. doi:10.1039/D4VA00256C , language =

    work page doi:10.1039/d4va00256c

  65. [67]

    Getting the basics right: capacitance, entropy and stress of electrified solid/liquid interfaces , language =

  66. [68]

    Ciacchi, Massimo and Saitov, Ilnur and Fonte, Nico Di and Daidone, Isabella and Pierleoni, Carlo , month = apr, year =. Ion-. doi:10.48550/arXiv.2604.13659 , abstract =

    Pith/arXiv arXiv doi:10.48550/arxiv.2604.13659

  67. [69]

    The data-only illusion in materials discovery

  68. [70]

    Physical Review Letters , author =

    Constant-. Physical Review Letters , author =. 2025 , pages =. doi:10.1103/PhysRevLett.134.046201 , language =

    work page doi:10.1103/physrevlett.134.046201 2025

  69. [71]

    Sechi, Renata and Kastlunger, Georg and Bhowmik, Arghya and Hansen, Heine Anton , month = may, year =. First-. Journal of Chemical Theory and Computation , publisher =. doi:10.1021/acs.jctc.5c00406 , abstract =

    work page doi:10.1021/acs.jctc.5c00406

  70. [72]

    Langmuir , author =

    Probing the. Langmuir , author =. 2016 , pages =. doi:10.1021/acs.langmuir.6b01697 , language =

    work page doi:10.1021/acs.langmuir.6b01697 2016

  71. [73]

    Physical Chemistry Chemical Physics , author =

    Tracking the surface structure and the influence of cations and anions on the double-layer region of a. Physical Chemistry Chemical Physics , author =. 2024 , pages =. doi:10.1039/D4CP02133A , abstract =

    work page doi:10.1039/d4cp02133a 2024

  72. [74]

    and Keane, Theo and Benson, Raz L

    Ho, Cheuk Hin and Oord, Cas van der and Darby, James P. and Keane, Theo and Benson, Raz L. and Espinoza, Cristian Rebolledo and Kulkarni, Rutvij and Spinu, Elina and Papanikolaou, Michail and Tomsett, Richard and Forrest, Robert M. and Bean, Jonathan J. and Csányi, Gábor and Ortner, Christoph , month = apr, year =. Equivariant. doi:10.48550/arXiv.2604.081...

    Pith/arXiv arXiv doi:10.48550/arxiv.2604.08143

  73. [75]

    and Pretzie, Luke N

    Grizzi, Vitor F. and Pretzie, Luke N. and Xu, Jiayi and Liu, Cong , month = apr, year =. doi:10.48550/arXiv.2604.12140 , abstract =

    Pith/arXiv arXiv doi:10.48550/arxiv.2604.12140

  74. [76]

    , month = apr, year =

    Zhu, Andrew and Komonvasee, Poramas and Nejad, Arman and Tew, David P. , month = apr, year =. Surface. Journal of Chemical Theory and Computation , publisher =. doi:10.1021/acs.jctc.5c02179 , abstract =

    work page doi:10.1021/acs.jctc.5c02179

  75. [77]

    Machine Learning: Science and Technology , author =

    Machine learning the derivative discontinuity of density-functional theory , volume =. Machine Learning: Science and Technology , author =. 2022 , note =. doi:10.1088/2632-2153/ac3149 , abstract =

    work page doi:10.1088/2632-2153/ac3149 2022

  76. [78]

    Journal of Chemical Theory and Computation , author =

    The. Journal of Chemical Theory and Computation , author =. 2026 , note =. doi:10.1021/acs.jctc.6c00034 , abstract =

    work page doi:10.1021/acs.jctc.6c00034 2026

  77. [79]

    Influence of

    Vernekar, Dnyanesh and Gupta, Soumyajit and Lee, Gi-Hyeok and Devi, Reshma and Rafiuddin, Mohamed Ruwaid and Sai Gautam, Gopalakrishnan and Yang, Wanli and Ramakrishnan, Srinivasan , month = mar, year =. Influence of. Chemistry of Materials , publisher =. doi:10.1021/acs.chemmater.6c00507 , abstract =

    work page doi:10.1021/acs.chemmater.6c00507

  78. [80]

    Electrochemical

    Low, Jensie and Ou, Pengfei , month = apr, year =. Electrochemical. Chemical Communications , publisher =. doi:10.1039/D6CC00216A , abstract =

    work page doi:10.1039/d6cc00216a

  79. [81]

    and Csányi, Gábor and Hermansson, Kersti and Kullgren, Jolla , month = mar, year =

    Svahn, Viktor and Magdău, Ioan-Bogdan and Niblett, Samuel P. and Csányi, Gábor and Hermansson, Kersti and Kullgren, Jolla , month = mar, year =. Limitations of. Journal of Chemical Theory and Computation , publisher =. doi:10.1021/acs.jctc.5c02043 , abstract =

    work page doi:10.1021/acs.jctc.5c02043

  80. [82]

    and Markland, Thomas E

    Ceriotti, Michele and Manolopoulos, David E. and Markland, Thomas E. and Rossi, Mariana , month = mar, year =. Path. doi:10.48550/arXiv.2603.28588 , abstract =

    work page doi:10.48550/arxiv.2603.28588

Showing first 80 references.