Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 2611.06109

Subtle Energy Corrections Reshape Condensed-Phase Property Predictions from a Machine-Learning Interatomic Potential

Haihui Wang·Xinyan Wang·Jiashu Liang·Yuanheng Wang·Chang Han·Wen Yan·Tianze Zheng·Kuang Yu
View original source ↗Read source document ↗Export metadata ↓

Abstract

Accurate reproduction of electronic-structure labels by machine-learning interatomic potentials does not necessarily ensure reliable condensed-phase properties. We investigate the effects of basis-set superposition error (BSSE) and an explicit Axilrod-Teller-Muto (ATM) three-body dispersion contribution on molecular-liquid predictions. Starting from MACELES-OFF, we fine-tune a baseline potential at the $ω$B97M-V/def2-TZVPD level. A separate branch learns counterpoise (CP) energy and force corrections with the baseline frozen, while a D4-parametrized ATM term is added during molecular-dynamics inference. Mean atom-normalized CP and ATM contributions are 0.222 and 0.205 meV/atom, respectively, for the analyzed SPICE2 configurations from OMol25. We compare baseline, CP-only, ATM-only, and CP + ATM variants for 39 organic liquids. Relative to baseline, the mean system-wise density changes are -1.17%, -10.38%, and -11.93% for CP-only, ATM-only, and CP + ATM, respectively; the corresponding enthalpy-of-vaporization changes are -2.79%, -14.74%, and -18.32%. For the 11 liquids with experimental self-diffusion data, the mean self-diffusion changes are +17.66%, +203.02%, and +230.36%. Both corrections thus lower density and enthalpy of vaporization while increasing molecular mobility on average. ATM dominates these changes, while CP provides a smaller contribution; their combination yields the largest average shifts. These results show that corrections with sub-meV/atom mean values on the analyzed configurations can substantially reshape the predicted structural, thermodynamic, and transport properties of molecular liquids.

Explore related subjects

physics.chem-ph

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Haihui Wang, Xinyan Wang, Jiashu Liang, Yuanheng Wang, Chang Han, Wen Yan, Tianze Zheng, Kuang Yu. 2026-11-04. Subtle Energy Corrections Reshape Condensed-Phase Property Predictions from a Machine-Learning Interatomic Potential. https://arxiv.org/abs/2611.06109

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Estimating Irving-Kirkwood pressure profiles from per-atom virial and density profiles: a closure for the contour-dependent term in planar liquid films

The local pressure tensor of an inhomogeneous fluid is not unique. Two tangential profiles are in common use in planar simulations: the Irving-Kirkwood (IK) profile, which requires a dedicated pass over every pair of atoms for every slab, and the per-atom virial (PAV) profile, which molecular-dynamics codes produce at negligible cost and coincides with the Harasima contour. The two differ by the derivative of a pair quantity, P_T^IK - P_T^PAV = dQ/dz. We show that, for Lennard-Jones free-standing films, Q(z) can be predicted from the density profile and the bulk radial distribution function of the coexisting liquid: under the closure rho2(r1,r2) ~ rho(z1) rho(z2) g(r12), with a pair-local interpolation of g between its liquid and dilute limits, Q is a double integral of rho(z) against a kernel built from u'(r) g(r). The estimator P_T^IK,est = P_T^PAV + dQ/dz is tested against IK profiles measured on the same trajectories for 20 films (thicknesses 7-25 sigma, three cutoff radii, two temperatures): the root-mean-square residual is 0.001-0.002 epsilon/sigma^3, against 0.010-0.021 for the uncorrected profile, and Q, the amount that the contour choice transfers between the two interfaces of a film with non-equivalent surfaces, is recovered to within 0.6% of the surface tension. Applied to the normal component, the same closure removes 83-92% of the spurious structure of the per-atom zz profile, a test that needs no IK run. Its mixture form is validated on three two-component films with an adsorbate unequally distributed between their faces, where it resolves the tension of each face; a clean face recovers the tension of the pure film.

physics.chem-ph↗
arXiv2026

OPRD-Imp: a dataset of experimentally observed impurities and side products in multi-step pharmaceutical syntheses

Predicting reaction impurities is a critical yet persistently data-limited challenge in pharmaceutical process chemistry. While machine learning models for reaction outcome prediction have advanced rapidly, benchmarking their ability to predict impurities remains impractical due to the absence of curated, publicly accessible datasets. We present OPRD-Imp, a dataset of 229 organic reactions drawn from 62 publications in Organic Process Research & Development, each annotated with experimentally observed impurities and their mechanistic origins. A total of 784 impurity entries is provided, classified into four origin categories. Reactions are represented as reaction SMILES and impurities as individual SMILES strings, with multi-step synthetic route connectivity preserved. All data were manually extracted from the source publications by trained chemists following a standardized extraction protocol, with expert chemical validation and manual structure drawing. OPRD-Imp is intended to serve as a benchmark resource for evaluating models that predict reaction impurities, providing experimentally reported impurity annotations and mechanistic origin labels that are not systematically represented in conventional reaction datasets.

physics.chem-ph↗
arXiv2026

Drift-React: One-step Generation of Reaction Pathways via SE(3) Drifting Fields

Mapping reaction pathways and transition states (TS) is fundamental to chemistry but computationally expensive at scale. The minimum energy pathway (MEP) dictates reaction rates and mechanisms, yet recovering it via electronic-structure methods requires thousands of costly force evaluations. Recent generative models accelerate TS identification but rely on iterative inference and only predict isolated saddle-point snapshots, missing the continuous reaction trajectory. We introduce Drift-React, an $\mathrm{SE}(3)$-equivariant generative framework that predicts complete reaction pathways in a single forward pass from only reactant and product geometries. By shifting distribution evolution to training via a Sinkhorn-weighted drifting field, Drift-React eliminates both the iterative force evaluations of NEB-style methods and the sequential ODE/SDE integration of diffusion and flow matching models. Evaluated on the Transition1x and Halo8 datasets, our one-step model generates physically consistent MEPs that accurately capture energetic bottlenecks and enable arbitrary-resolution sampling along the reaction coordinate. For isolated TS prediction, Drift-React matches the sub-Ångström accuracy of state-of-the-art iterative models while delivering orders-of-magnitude acceleration, clearing a major computational bottleneck for large-scale reaction network exploration.

physics.chem-ph↗