Pith. sign in

REVIEW 3 major objections 4 minor 110 references

Towards Large-Scale Condensed Phase Simulations using Machine Learned Energy Functions

T0 review · 3 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read A modular machine-learned energy function reproduces bulk water properties in large simulations.

desk verdict A competent, honest integration of prior ML/electrostatics pieces into a water model with an excellent monomer PES; the bulk validation is partly selection, so the transferability claim is untested. read the letter →

arxiv 2506.23272 v1 pith:LYM3UORV submitted 2025-06-29 physics.chem-ph

Eric D. Boittier , Silvan Käser , Markus Meuwly This is my paper · ORCID
classification physics.chem-ph
keywords machine-learnedenergyfunctionswaterpotentialneuralnetworksurfacekernel-basedminimaldistributedchargesLennard-Jonesparametrizationcondensed-phasemoleculardynamicsclustersthermodynamicmodelselection
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper tries to establish that a modular, machine-learned energy function can serve as a practical force field for liquid simulations: a neural network represents the CCSD(T)-F12 monomer surface, a kernel-based fluctuating charge model handles electrostatics, and Lennard-Jones (12-6) terms fitted to ab initio cluster interaction energies capture the remaining van der Waals physics. Applying this workflow to water produces two models—M-DFT, with Lennard-Jones terms fitted to density-functional cluster energies, and M-CC, fitted to coupled-cluster energies—that reproduce the liquid density and heat of vaporization within about 0.01 g/ml and 0.2 kcal/mol of experiment after selection among many equivalent parameter sets. The same models give gas-phase vibrational frequencies within a few reciprocal centimeters, correctly identify the prism isomer as the most stable water hexamer (M-CC), and yield reasonable dielectric, compressibility, diffusion, and reorientation properties in multi-nanosecond simulations of 2000 to 8000 molecules. The wider point is that such a workflow could be recycled for other condensed phases where experimental reference data are scarce, although the paper also documents a clear failure mode: neither model produces the density maximum of liquid water near 277 K, which it attributes to the neglect of three- and four-body interactions.

What carries the argument

The machinery is a modular energy decomposition: $E = E_\mathrm{NN}(\mathrm{monomer}) + E_\mathrm{kMDCM} + E_\mathrm{LJ}$, where $E_\mathrm{NN}$ is a small feed-forward neural network mapping the three interatomic distances of a water monomer to its energy, $E_\mathrm{kMDCM}$ is the kernel-based minimal distributed charge model whose six geometry-dependent charges capture the fluctuating electrostatics (including lone-pair-like sites), and $E_\mathrm{LJ}$ is a 12-6 Lennard-Jones pair potential. The Lennard-Jones parameters are fitted by subtracting the neural-network monomer energies and the kMDCM electrostatics from reference cluster interaction energies, so the pair potential absorbs whatever remains. Model selection is then driven by a loss based on bulk density and heat of vaporization, which chooses one working model out of many equally good cluster fits. This same three-part decomposition is the object the paper argues is generic and reusable.

What would settle it

Compute explicit three-body CCSD(T)-F12 interaction energies for a set of water trimers and compare them with the residual that the fitted Lennard-Jones pair term absorbs: if the three-body contribution is comparable to or larger than that residual, the pairwise-additive core of the model cannot be the true source of its bulk accuracy.

Watch

Extended reading notes

Core claim

The paper's core claim is that decomposing the total energy into an intramolecular neural-network term, a geometry-dependent distributed-charge electrostatic term, and a fitted Lennard-Jones pair term yields an energy function that is both accurate enough for CCSD(T)-quality water and fast enough for boxes of about $10^{4}$ molecules. For water, reference monomer energies and forces came from CCSD(T)-F12B/aug-cc-pVTZ-F12 calculations; reference interaction energies for the Lennard-Jones fit came from ωB97X-V/def2-QZVP for clusters of 2 to 60 monomers (M-DFT) and from CCSD(T)-F12 for clusters of 2 to 4 monomers (M-CC). Because cluster interaction energies alone do not determine a unique Lennard-Jones parametrization, the authors generated hundreds of fits and selected the ones agreeing best with experimental liquid density and heat of vaporization. The resulting models reproduce the bulk density and vaporization enthalpy, are within the right range for dielectric constant, compressibility, self-diffusion, and orientational relaxation, and reproduce the spectroscopy of the monomer and the structure of small clusters. The paper is explicit that the omission of three-body and higher interactions shows up as the missing density maximum and a too-weak temperature dependence of the vaporization enthalpy.

Load-bearing premise

The load-bearing premise is that pairwise-additive Lennard-Jones interactions plus geometry-dependent distributed charges, without explicit three-body terms, can represent the effective interactions of bulk water.

Editorial extensions

If this is right

  • The workflow provides a route to CCSD(T)-level condensed-phase simulations for liquids that lack extensive experimental force-field data, because the only experimental inputs used for parameter selection are density and heat of vaporization.
  • The M-CC model is efficient enough for routine 10 ns molecular dynamics with 2000 to 8000 water molecules on commodity hardware, and remains stable when O-H bonds are constrained with SHAKE at a 1 fs time step.
  • Because dimers dominate the cluster training set, the models capture two-body CCSD(T) energetics well, and the paper identifies explicit three-body terms as the next step for fixing the missing density maximum and the weak temperature dependence of the vaporization enthalpy.
  • Adding more experimental observables to the model-selection loss, such as the temperature dependence of density or radial distribution functions, is expected to refine the Lennard-Jones parameters within the same workflow.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The strong correlation between fitted Lennard-Jones σ and ε values means cluster interaction energies underdetermine the pair potential; the choice that wins on density and heat of vaporization may not be the one that transfers to solutes, surfaces, or ice, so those properties are genuine tests of the workflow rather than guarantees.
  • Because the paper's own diagnostic points to missing three- and four-body terms, a practical extension is to fit the same workflow with explicit trimer three-body energies added to the reference data and check whether the density maximum reappears without sacrificing the already-good bulk properties.
  • For liquids with stronger many-body electrostatics, such as ionic or deep-eutectic systems, the pairwise-additive Lennard-Jones term would have to absorb more physics, so the density and heat-of-vaporization selection could mask a deficiency; testing temperature dependence and structure would expose this earlier than the two thermodynamic observables used here.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 4 minor

Summary. The manuscript proposes a generic, modular workflow for building condensed-phase energy functions: a neural-network monomer PES trained to CCSD(T)-F12 data, kMDCM conformationally dependent electrostatics, and Lennard-Jones 12-6 parameters fitted to cluster interaction energies from either DFT (M-DFT) or CCSD(T)-F12 (M-CC) reference data. For water as the test case, the authors select one of 300 near-degenerate LJ parameter sets by matching the experimental bulk density and heat of vaporization, then evaluate a wide range of cluster, thermodynamic, structural, dielectric, transport, and vibrational properties from MD simulations of 2000-8000 water molecules. The monomer PES is very accurate (energy MAE 0.007 kcal/mol, VPT2 frequency MAE 3.3 cm^-1), and several out-of-sample bulk properties (e.g., hydration free energy, self-diffusion, reorientation time) are in reasonable agreement with experiment, while others (density maximum, dielectric constant, compressibility, third-shell gOO structure) show systematic deviations that the authors attribute mainly to neglected many-body interactions.

Significance. If the workflow is ultimately validated, it would be a useful contribution because it combines high-level monomer energies, a flexible electrostatic model, and a cluster-based parametrization into a computationally efficient force field suitable for large-scale simulations. The paper ships its data and analysis code, and the monomer PES validation is a clear strength: the NN reproduces CCSD(T)-F12 energies and forces closely, conserves energy in NVE runs, and yields VPT2 frequencies within 3.3 cm^-1 of experiment. The explicit comparison of two reference levels (DFT and CCSD(T)) and the systematic evaluation of many observables are also commendable. However, the central validation is weakened because the two headline bulk properties, density and heat of vaporization, are used as selection targets rather than as predictions, and the genuinely out-of-sample bulk properties deviate in ways that the paper itself traces to missing many-body physics. The transferability claim for other liquids is therefore not yet established.

major comments (3)
  1. [Sections 3.2 and 3.4, Table 3] The agreement on density and heat of vaporization is a selection outcome, not an independent prediction. Table 3 explicitly labels these two quantities as 'training reference data,' and Section 3.2 states that the loss L = 0.5 MAE ρ + MAE ΔH was used to select one model from the 300 fitted LJ parameter sets. The statement in Section 3.4 that 'all these results correspond to performance on a test set of observables' is therefore misleading for the 300 K ρ and ΔHvap entries; only quantities not appearing in the selection loss (ρ(T), ΔHvap(T), κ, ε, D0, τ2, gOO(r)) are genuine predictions. Among those, the density maximum is absent, the dielectric constant is 63 (M-DFT) and 72 (M-CC) versus 78.2, the compressibility is 35.2/38.5 versus 45.8, and the third solvation shell of gOO is over-structured. The paper should either re-frame ρ and ΔHvap as calibration targets or supply a bulk observable that was not used in parameter selection and is reproduced quantitatively.
  2. [Section 3.3] The hexamer benchmarks show total interaction energies underbound by 2-4 kcal/mol (M-DFT) and 4-8 kcal/mol (M-CC) relative to CCSD(T)/CBS, even though the relative energies between isomers are better. The text describes this as 'rather successful' in capturing two-body contributions, but because the LJ parameters are fitted to cluster interaction energies and are intended to absorb all non-electrostatic, non-intramolecular physics, the magnitude of the underbinding is a warning that the pair term may not be capturing the effective many-body interactions present in bulk water. This is directly relevant to the transferability claim: if the pair-fit compensates for missing three- and four-body terms in clusters, nothing guarantees the compensation transfers to bulk densities. The authors should discuss this compensation explicitly and, ideally, test transferability on a second liquid or on a bulk property far outside the fitting set.
  3. [Section 3.2, Section 3.4] The model selection uses only two observables (ρ and ΔHvap) to choose among 300 candidate LJ parameter sets, and Section 3.4 notes that only one of these 300 models was subsequently assessed. Because all 300 sets fit the cluster interaction energies with RMSE better than 1 kcal/mol, the out-of-sample properties reported in Tables 3 and 4 could depend substantially on which near-degenerate fit is selected. The authors should report the spread of predicted bulk properties across the candidate parameter sets, or at least a sensitivity analysis, to establish that the chosen set is representative rather than a fortuitous choice.
minor comments (4)
  1. [Section 3.4, Figure 6F] There is a unit inconsistency: Section 3.4 reports 'ρcalc = 0.999 kg/m3' while the rest of the paper uses g/cm3 or g/ml, and Figure 6F labels ΔHvap in 'g/cm3', which should be kcal/mol.
  2. [Table S1] The sign convention for the Lennard-Jones ε values should be stated explicitly; the tabulated values are negative, whereas the usual convention for a well depth is positive, and this may confuse readers who are not familiar with the CHARMM convention.
  3. [Section 2.2] The sentence 'This leads to a total of 36000 parameter combinations' follows from '180 sets of structures' times 200 optimizations, but the preceding text says 'Three repeats of this procedure were carried out, resulting in 180 sets of structures'; please clarify how 180 arises from the 200 random initial values and three repeats.
  4. [Figure 2 caption] The caption lists panel labels A1, A2, A3, B, and C1 but the text refers to 'Panels A1-3' without defining the layout of these subpanels; a short explanation of the Schlegel projection and its axes would improve readability.

Circularity Check

2 steps flagged · score 6.0 of 10

Bulk ρ and ΔHvap are used to select the LJ parameters and then presented as test-set results; M-DFT hexamer benchmarks also lie inside the cluster fitting range, though genuine extrapolation (M-CC hexamers, κ, D0, ε, τ2, IR) retains independent content.

  1. fitted input called prediction [Abstract; Section 3.2 'Fitting and Performance of the Intermolecular Model'; Table 3; Section 3.4 'Density and Structural Properties']
    "Based on the bulk liquid density and heat of vaporization, the best-performing set of LJ(12-6) parameters was selected and a wide range of condensed phase properties were determined and compared with experiment. ... For further model selection out of the 300 fitted M-DFT and M-CC models, the pure liquid density (ρ) and enthalpy of vaporization (ΔHvap) were determined from MD simulations using each of the 300 models and comparing with measured [ρ, ΔHvap] values ... To determine the preferred (“best performing”) model, the loss function L = 0.5 MAEρ + MAEΔH was used."

    The two observables used to choose the final LJ parameters among the 300 degenerate cluster fits are the same two thermodynamic quantities later presented as evidence of 'good bulk properties'. Table 3 itself labels ρ and ΔHvap as 'Training reference data', and Section 3.4 calls the same results part of a 'test set of observables' while admitting the models 'were only based on ab initio electronic energies and [ρ, ΔHvap] at 300 K'. Agreement on ρ and ΔHvap is therefore a selection outcome, not an independent prediction; the genuinely unselected evidence must come from κ, D0, ε, τ2, g(r), and the IR spectrum, several of which deviate.

  2. fitted input called prediction [Section 2.2 'Intermolecular Model'; Section 3.3 'Gas-Phase Benchmarks for Water Hexamer Clusters']
    "For each cluster size N ∈ [2, 60], 1000 configurations were extracted yielding a total of 59000 snapshots. ... As DFT calculations are computationally much more efficient and require less memory, reference data for cluster sizes (H2O)2 up to (H2O)60 were computed. ... Next, the fitted models M-DFT and M-CC were used to probe finer details of the intermolecular interactions. This is, for example, encoded in the structures, relative interaction energies and normal mode frequencies of water hexamers (H2O)6."

    For M-DFT, the hexamer (N=6) lies inside the cluster-size range used to fit the Lennard-Jones parameters to ωB97X-V reference interaction energies (N=2..60). The hexamer interaction-energy RMSE reported in Figure 4 is therefore a training-range consistency check rather than an extrapolative test. The genuinely out-of-range case is M-CC, which was fitted only to CCSD(T) clusters of size N=2..4 and which honestly shows larger 4-8 kcal/mol errors. The paper's own explanation that the improved two-body quality is 'explained by the over representation of dimers in the training set' confirms that this benchmark tracks training composition for M-DFT.

full rationale

The monomer neural network PES is independently strong: it is validated against a held-out CCSD(T)-F12 test set and against experimental VPT2 frequencies, so that part of the workflow is self-contained. The bulk observables not used in parameter selection—isothermal compressibility, self-diffusion, dielectric constant, orientational relaxation, g(r), and the IR spectrum—are genuine predictions, and the paper reports their deviations candidly. However, the central 'good bulk properties' claim partly reduces to a fit: ρ and ΔHvap are explicitly used to select the final set of LJ parameters out of 300 degenerate fits, Table 3 labels them 'Training reference data', and Section 3.4 still calls the results a 'test set of observables'. That is a direct instance of a fitted input being presented as a prediction. The M-DFT hexamer benchmark is also contaminated because hexamers fall within the DFT cluster-size fitting range, although M-CC provides an honest extrapolation test and performs worse. No load-bearing self-citation or imported uniqueness theorem was found; the self-references to prior kMDCM and cluster-fitting work are normal methodological provenance with external benchmarks. Overall the circularity is partial, not total: the workflow has independent content, but the specific bulk-density and heat-of-vaporization validation is double-counted.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The central claim rests on the quality of the quantum-chemical reference data, the additivity of the energy decomposition, and the transferability of gas-phase and cluster fits to the bulk liquid. The main fitted quantities are the four Lennard-Jones parameters per model, selected among 300 candidates using experimental density and heat of vaporization. No new physical entities are introduced; the kMDCM charge model and the NN monomer PES are existing components from prior work.

free parameters (4)
  • Lennard-Jones parameters, M-DFT (epsilon_O, sigma_O, epsilon_H, sigma_H) = -0.2275 kcal/mol, 1.7653 Å, -0.2470 kcal/mol, 0.0230 Å (as listed in Table S1)
    Fitted to omegaB97X-V cluster interaction energies; final values selected from 300 candidates by matching experimental rho and DeltaHvap.
  • Lennard-Jones parameters, M-CC (epsilon_O, sigma_O, epsilon_H, sigma_H) = -0.1125 kcal/mol, 1.8194 Å, -0.0434 kcal/mol, 0.4265 Å (as listed in Table S1)
    Fitted to CCSD(T)-F12 cluster interaction energies; selected by rho and DeltaHvap.
  • Monomer neural network weights = not listed; trained on 2566 CCSD(T)-F12 geometries
    Trained on CCSD(T)-F12B/aug-cc-pVTZ-F12 energies and forces; hyperparameters include force weight omega_F = 10 (Section 2.1).
  • kMDCM distributed charge parameters = not listed; fitted to 180 monomer geometries
    Fitted to CCSD(T)/aug-cc-pVTZ electrostatic potentials in prior work (Ref 35); used here without re-fitting.
assumptions (5)
  • domain assumption CCSD(T)-F12B/aug-cc-pVTZ-F12 and omegaB97X-V/def2-QZVP reference energies are accurate enough for water cluster interaction energies.
    Section 2.2 uses these as ground truth for all fits and benchmark comparisons; no uncertainty quantification on the references is given.
  • ad hoc to paper The total interaction energy decomposes exactly into monomer NN energies, kMDCM electrostatics, and pairwise LJ terms, with no explicit many-body terms.
    Section 2.2: the residual after subtracting monomer and electrostatic energies is assigned to LJ; many-body effects are implicitly absorbed into effective pairwise parameters.
  • domain assumption Clusters sampled from TIP3P MD are representative of the configurations relevant to bulk liquid water.
    Section 2.2: 1000 configurations per cluster size N in [2,60] extracted from 5 ns NpT TIP3P simulations.
  • domain assumption The gas-phase monomer NN PES transfers to monomers in the condensed phase.
    Section 2.1: NN trained on gas-phase CCSD(T) data is used in bulk MD without re-fitting.
  • domain assumption Classical mechanics without nuclear quantum effects is adequate for the computed bulk properties.
    Sections 2.3 and 3.1: classical MD used for spectra and thermodynamics; the paper notes zero-point energy leakage as a limitation for vibrational frequencies.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Towards Large-Scale Condensed Phase Simulations using Machine Learned Energy Functions." pith.science (2026). https://pith.science/paper/LYM3UORV

@misc{pith2026250623272,
  author       = {Pith},
  title        = {Pith review of: Towards Large-Scale Condensed Phase Simulations using Machine Learned Energy Functions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LYM3UORV}},
  note         = {Machine review of arXiv:2506.23272}
}
abstract

Accurate, yet computationally efficient energy functions are essential for state-of-the art molecular dynamics (MD) studies of condensed phase systems. Here, a generic workflow based on a combination of machine learning-based and empirical representations of intra- and intermolecular interactions is presented. The total energy is decomposed into internal contributions, and electrostatic and van der Waals interactions between monomers. The monomer potential energy surface is described using a neural network, whereas for the electrostatics the flexible minimally distributed charge model is employed. Remaining contributions between reference energies from electronic structure calculations and the model are fitted to standard Lennard-Jones (12-6) terms. For water as a topical example, reference energies for the monomers are determined from CCSD(T)-F12 calculations whereas for an ensemble of cluster structures containing $[2,60]$ and $[2,4]$ monomers DFT and CCSD(T) energies, respectively, were used to best match the van der Waals contributions. Based on the bulk liquid density and heat of vaporization, the best-performing set of LJ(12-6) parameters was selected and a wide range of condensed phase properties were determined and compared with experiment. MD Simulations on the multiple-nanosecond time scale were carried out for water boxes containing 2000 to 8000 monomers, depending on the property considered. The performance of such a generic ML-inspired parametrization scheme is very promising and future improvements and extensions are discussed, also in view of recent advances for water in particular in the literature.

Figures

Figures reproduced from arXiv: 2506.23272 by the authors.

Figure 1
Figure 1. Performance of the NN PES A: Prediction error ∆E = ECCSD(T)−F12 − ENN on a hold-out test set containing 366 random water structures. The predictions are all within ∼ 0.1 kcal/mol and have a high coefficient of determination (1 − R2 = 5.3E − 06). The potential energy of a water molecule as a function of the bending angle Θ (B) and of the O– H bond length (C). For B and C the energy is given with respect to the minimi… view at source ↗
Figure 2
Figure 2. The LJ-Fit for M-CC. Panel A1: Projections of the four fitted LJ-parameters using Schlegel’s hypercube representation (see text), with vertices annotated as 0 (minima) or 1 (maxima) in the transformed (σO, σH, ϵO, ϵH) coordinate system. The three color scales are for σOH = (σO + σH)/2, the values of the loss L = ∆ρ/2 + ∆(∆H), and ϵOH = ϵOϵH, respectively. Panel B: Correlations among the fitted LJ-parameters: σO, ϵOH… view at source ↗
Figure 3
Figure 3. Simulated density and enthalpy of vaporization of ambient water using the [PITH_FULL_IMAGE:figures/full_fig_p018_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Total interaction and differential stabilization energies for water hexamer from us [PITH_FULL_IMAGE:figures/full_fig_p021_4.png]
Figure 5
Figure 5. Figure 5: Comparisons of the harmonic frequencies of four low-lying hexamer minima (from [PITH_FULL_IMAGE:figures/full_fig_p023_5.png]
Figure 6
Figure 6. Figure 6: Structural features of the bulk liquid water: (A) radial distribution functions [PITH_FULL_IMAGE:figures/full_fig_p025_6.png]
Figure 7
Figure 7. Figure 7: To account for finite-size effects, the self-diffusion coefficient obtained from length [PITH_FULL_IMAGE:figures/full_fig_p030_7.png]
Figure 8
Figure 8. Figure 8: Infrared spectrum of liquid water between 0 and 4000 cm [PITH_FULL_IMAGE:figures/full_fig_p031_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

110 extracted references · 73 canonical work pages

  1. [1]

    F.; Berendsen, H

    Van Gunsteren, W. F.; Berendsen, H. J. Computer simulation of molecular dynamics: methodology, applications, and perspectives in chemistry. Angew. Chem., Int. Ed. Engl. 1990, 29, 992--1023

    1990

  2. [2]

    Karplus, M.; Petsko, G. A. Molecular dynamics simulations in biology. Nature 1990, 347, 631--639

    1990

  3. [3]

    D.; McCammon, J

    Durrant, J. D.; McCammon, J. A. Molecular dynamics simulations and drug discovery. BMC biology 2011, 9, 1--9

    2011

  4. [4]

    CHARMM general force field: A force field for drug-like molecules compatible with the CHARMM all-atom additive biological force fields

    Vanommeslaeghe, K.; Hatcher, E.; Acharya, C.; Kundu, S.; Zhong, S.; Shim, J.; Darian, E.; Guvench, O.; Lopes, P.; Vorobyov, I., et al. CHARMM general force field: A force field for drug-like molecules compatible with the CHARMM all-atom additive biological force fields. J. Comput. Chem. 2010, 31, 671--690

    2010

  5. [5]

    M.; Caldwell, J

    Wang, J.; Wolf, R. M.; Caldwell, J. W.; Kollman, P. A.; Case, D. A. Development and testing of a general amber force field. J. Comput. Chem. 2004, 25, 1157--1174

    2004

  6. [6]

    L.; Tirado-Rives, J

    Jorgensen, W. L.; Tirado-Rives, J. The OPLS potential functions for proteins - energy minimizations for crystals of cyclic-peptides and crambin. J. Am. Chem. Soc. 1988, 110, 1657--1666

    1988

  7. [7]

    E.; Van Gunsteren, W

    Oostenbrink, C.; Villa, A.; Mark, A. E.; Van Gunsteren, W. F. A biomolecular force field based on the free enthalpy of hydration and solvation: the GROMOS force-field parameter sets 53A5 and 53A6. J. Comput. Chem. 2004, 25, 1656--1676

    2004

  8. [8]

    M.; Koner, D.; Meuwly, M

    Salehi, S. M.; Koner, D.; Meuwly, M. Vibrational Spectroscopy of N _3^- in the Gas and Condensed Phase. J. Phys. Chem. B 2019, 123, 3282--3290

    2019

Show all 110 references
  1. [9]

    M.; Mondal, P.; Meuwly, M

    Koner, D.; Salehi, S. M.; Mondal, P.; Meuwly, M. Non-conventional Force Fields for Applications in Spectroscopy and Chemical Reaction Dynamics. J. Chem. Phys. 2020, 153, 010901

    2020

  2. [10]

    Permutationally invariant, reproducing kernel-based potential energy surfaces for polyatomic molecules: From formaldehyde to acetone

    Koner, D.; Meuwly, M. Permutationally invariant, reproducing kernel-based potential energy surfaces for polyatomic molecules: From formaldehyde to acetone. J. Chem. Theory Comput. 2020, 16, 5474--5484

    2020

  3. [11]

    Nandi, A.; Qu, C.; Bowman, J. M. Full and fragmented permutationally invariant polynomial potential energy surfaces for trans and cis N-methyl acetamide and isomerization saddle points. J. Chem. Phys. 2019, 151, 084306

    2019

  4. [12]

    M.; Dawes, R.; Xie, D.; Guo, H

    Li, J.; Carter, S.; Bowman, J. M.; Dawes, R.; Xie, D.; Guo, H. High-level, first-principles, full-dimensional quantum calculation of the ro-vibrational spectrum of the simplest criegee intermediate (CH 2 OO). J. Phys. Chem. Lett. 2014, 5, 2364--2369

    2014

  5. [13]

    Mackerell, A. D. Empirical Force Fields for Biological Macromolecules: Overview and Issues. J. Comput. Chem. 2004, 25, 1584--1604

    2004

  6. [14]

    The Theory of Intermolecular Forces; Oxford University Press: Cambridge, 2013

    Stone, A. The Theory of Intermolecular Forces; Oxford University Press: Cambridge, 2013

    2013

  7. [15]

    M.; Hawe, G

    Handley, C. M.; Hawe, G. I.; Kell, D. B.; Popelier, P. L. A. Optimal Construction of a Fast and Accurate Polarisable Water Potential Based on Multipole Moments Trained by Machine Learning. Phys. Chem. Chem. Phys. 2009, 11, 6365

    2009

  8. [16]

    Leveraging Symmetries of Static Atomic Multipole Electrostatics in Molecular Dynamics Simulations

    Bereau, T.; Kramer, C.; Meuwly, M. Leveraging Symmetries of Static Atomic Multipole Electrostatics in Molecular Dynamics Simulations. J. Chem. Theo. Comput. 2013, 9, 5450--5459

    2013

  9. [17]

    G.; Meuwly, M

    Devereux, M.; Raghunathan, S.; Fedorov, D. G.; Meuwly, M. A Novel, Computationally Efficient Multipolar Model Employing Distributed Charges for Molecular Dynamics Simulations. J. Chem. Theory Comput. 2014, 10, 4229

    2014

  10. [18]

    Many-Body Effects and Electrostatics in Biomolecules; Jenny Stanford Publishing, 2016; pp 251--286

    Bereau, T.; Meuwly, M. Many-Body Effects and Electrostatics in Biomolecules; Jenny Stanford Publishing, 2016; pp 251--286

    2016

  11. [19]

    Y.; Qi, R.; Walker, B

    Jing, Z.; Liu, C.; Cheng, S. Y.; Qi, R.; Walker, B. D.; Piquemal, J.-P.; Ren, P. Polarizable Force Fields for Biomolecular Simulations: Recent Advances and Applications. Annu. Rev. Biochem. 2019, 48, 371--394

    2019

  12. [20]

    An efficient water force field calibrated against intermolecular THz and Raman spectra

    Sidler, D.; Meuwly, M.; Hamm, P. An efficient water force field calibrated against intermolecular THz and Raman spectra. J. Chem. Phys. 2018, 148, 244504

    2018

  13. [21]

    Halgren, T. A. The Representation of Van Der Waals (vdw) Interactions in Molecular Mechanics Force Fields: Potential Form, Combination Rules, and Vdw Parameters. J. Am. Chem. Soc. 1992, 114, 7827--7843

    1992

  14. [22]

    On Combination Rules for Molecular Van Der Waals Potential-well Parameters

    Bzowski, J.; Mason, E.; Kestin, J. On Combination Rules for Molecular Van Der Waals Potential-well Parameters. Int. J. Therm.. 1988, 9, 131--143

    1988

  15. [23]

    Inadequacy of the Lorentz-berthelot Combining Rules for Accurate Predictions of Equilibrium Properties by Molecular Simulation

    Delhommelle, J.; Milli \'e , P. Inadequacy of the Lorentz-berthelot Combining Rules for Accurate Predictions of Equilibrium Properties by Molecular Simulation. Mol. Phys. 2001, 99, 619--625

    2001

  16. [24]

    A.; Pineda, L

    Lemkul, J. A.; Pineda, L. I. G.; Miller, C. J.; MacKerell, A. D. BPS2025 - Development of a polarizable force field for phosphorylated proteins and polypeptides. Biophys. J. 2025, 124, 380a

    2025

  17. [25]

    The Q-AMOEBA (CF) Polarizable Potential

    Mauger, N.; Plé, T.; Lagardère, L.; Huppert, S.; Piquemal, J.-P. The Q-AMOEBA (CF) Polarizable Potential. J. Phys. Chem. Lett. 2025, 16, 5723–5731

    2025

  18. [26]

    P.; Kim, L.; Head-Gordon, M.; Head-Gordon, T

    Heindel, J. P.; Kim, L.; Head-Gordon, M.; Head-Gordon, T. Completely Multipolar Model for Many-Body Water–Ion and Ion–Ion Interactions. J. Phys. Chem. Lett. 2025, 16, 975–984

    2025

  19. [27]

    Teng, X.; Yu, W.; MacKerell, A. D. J. Computationally Efficient Polarizable MD Simulations: A Simple Water Model for the Classical Drude Oscillator Polarizable Force Field. J. Phys. Chem. Lett. 2025, 16, 1016–1023

    2025

  20. [28]

    T.; Meuwly, M

    Unke, O. T.; Meuwly, M. Toolkit for the construction of reproducing kernel-based representations of data: Application to multidimensional potential energy surfaces. J. Chem. Inf. Model. 2017, 57, 1923--1931

    2017

  21. [29]

    T.; Meuwly, M

    Unke, O. T.; Meuwly, M. PhysNet: A neural network for predicting energies, forces, dipole moments, and partial charges. J. Chem. Theory Comput. 2019, 15, 3678--3693

    2019

  22. [30]

    The Bigger the Better? Accurate Molecular Potential Energy Surfaces from Minimalist Neural Networks

    K \"a ser , S.; Koner , D.; Meuwly , M. The Bigger the Better? Accurate Molecular Potential Energy Surfaces from Minimalist Neural Networks . arXiv e-prints 2024, arXiv:2411.18121

    2024 arXiv

  23. [31]

    D.; Upadhyay, M.; Meuwly, M

    K \"a ser, S.; Boittier, E. D.; Upadhyay, M.; Meuwly, M. Transfer Learning to CCSD(T): Accurate Anharmonic Frequencies from Machine Learning Models. J. Chem. Theory Comput. 2021, 17, 3687--3699

    2021

  24. [32]

    O.; Meuwly, M

    K \"a ser, S.; Richardson, J. O.; Meuwly, M. Transfer Learning for Affordable and High-Quality Tunneling Splittings from Instanton Calculations. J. Chem. Theory Comput. 2022, 18, 6840--6850

    2022

  25. [33]

    O.; Meuwly, M

    K \"a ser, S.; Richardson, J. O.; Meuwly, M. Transfer Learning for Predictive Molecular Simulations: Data-Efficient Potential Energy Surfaces at CCSD(T) Accuracy. J. Chem. Theory Comput. 2025, arXiv preprint arXiv:2407.21366, in print

    2025 arXiv

  26. [34]

    Bereau, T.; Andrienko, D.; von Lilienfeld, O. A. Transferable Atomic Multipole Machine Learning Models for Small Organic Molecules. J. Chem. Theory Comput. 2015, 11, 3225--3233

    2015

  27. [35]

    Kernel-Based Minimal Distributed Charges: A Conformationally Dependent ESP-Model for Molecular Simulations

    Boittier, E.; Töpfer, K.; Devereux, M.; Meuwly, M. Kernel-Based Minimal Distributed Charges: A Conformationally Dependent ESP-Model for Molecular Simulations. J. Chem. Theory Comput. 2024, 20, 8088–8099

    2024

  28. [36]

    A.; Goedecker, S.; Behler, J

    Ko, T.; Finkler, J. A.; Goedecker, S.; Behler, J. A fourth-generation high-dimensional neural network potential with accurate electrostatics including non-local charge transfer. Nat. Comm. 2021, 12, 398

    2021

  29. [37]

    L.; Kaminski, G

    Banks, J. L.; Kaminski, G. A.; Zhou, R.; Mainz, D. T.; Berne, B. J.; Friesner, R. A. Parametrizing a Polarizable Force Field from Ab Initio Data. I. The Fluctuating Point Charge Model. J. Chem. Phys. 1999, 110, 741--754

    1999

  30. [38]

    D.; Meuwly, M

    Devereux, M.; Boittier, E. D.; Meuwly, M. Systematic improvement of empirical energy functions in the era of machine learning. J. Comput. Chem. 2024, 45, 1899–1913

    2024

  31. [39]

    Force Fields for Deep Eutectic Mixtures: Application to Structure, Thermodynamics and 2D-Infrared Spectroscopy

    T\"opfer, K.; Boittier, E.; Devereux, M.; Pasti, A.; Hamm, P.; Meuwly, M. Force Fields for Deep Eutectic Mixtures: Application to Structure, Thermodynamics and 2D-Infrared Spectroscopy. J. Phys. Chem. B 2024, 128, 10937--10949

    2024

  32. [40]

    Structure and dynamics of deep eutectic systems from cluster-optimized energy functions

    T\"opfer, K.; Wang, J.; Patel, S.; Meuwly, M. Structure and dynamics of deep eutectic systems from cluster-optimized energy functions. Adv. Quant. Chem. 2025, arXiv preprint arXiv:2502.21233, in print

    2025 arXiv

  33. [41]

    L.; Conte, R.; Nandi, A.; Bowman, J

    Yu, Q.; Qu, C.; Houston, P. L.; Conte, R.; Nandi, A.; Bowman, J. M. q-AQUA: A many-body CCSD(T) water potential, including four-body interactions, demonstrates the quantum nature of water from clusters to the liquid phase. J. Phys. Chem. Lett. 2022, 13, 5068--5074

    2022

  34. [42]

    F.; Paesani, F

    Zhu, X.; Riera, M.; Bull-Vulpe, E. F.; Paesani, F. MB-pol(2023): Sub-chemical Accuracy for Water Simulations from the Gas to the Liquid Phase. J. Chem. Theory Comput. 2023, 19, 3551--3566

    2023

  35. [43]

    L.; Chandrasekhar, J.; Madura, J

    Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D.; Impey, R. W.; Klein, M. L. Comparison of simple potential functions for simulating liquid water. J. Chem. Phys. 1983, 79, 926--935

    1983

  36. [44]

    GFN2-xTB—An accurate and broadly parametrized self-consistent tight-binding quantum chemical method with multipole electrostatics and density-dependent dispersion contributions

    Bannwarth, C.; Ehlert, S.; Grimme, S. GFN2-xTB—An accurate and broadly parametrized self-consistent tight-binding quantum chemical method with multipole electrostatics and density-dependent dispersion contributions. J. Chem. Theory Comput. 2019, 15, 1652--1671

    2019

  37. [45]

    Werner, H.-J. et al. MOLPRO, version 2019, a package of ab initio programs. 2019

    2019

  38. [46]

    B.; Knizia, G.; Werner, H.-J

    Adler, T. B.; Knizia, G.; Werner, H.-J. A simple and efficient CCSD(T)-F12 approximation. J. Chem. Phys. 2007, 127, 221106

    2007

  39. [47]

    G.; Pearlmutter, B

    Baydin, A. G.; Pearlmutter, B. A.; Radul, A. A.; Siskind, J. M. Automatic differentiation in machine learning: a survey. J. Mach. Learn. Res. 2017, 18, 5595--5637

    2017

  40. [48]

    J.; Kale , S.; Kumar , S

    Reddi , S. J.; Kale , S.; Kumar , S. On the Convergence of Adam and Beyond . arXiv e-prints 2019, arXiv:1904.09237

    2019 arXiv

  41. [49]

    Shao, K.; Chen, J.; Zhao, Z.; Zhang, D. H. Communication: Fitting potential energy surfaces with fundamental invariant neural network. J. Chem. Phys. 2016, 145, 071101

    2016

  42. [50]

    L.; Blondel, A.; Boittier, E

    Hwang, W.; Austin, S. L.; Blondel, A.; Boittier, E. D.; Boresch, S.; Buck, M.; Buckner, J.; Caflisch, A.; Chang, H.-T.; Cheng, X., et al. CHARMM at 45: Enhancements in Accessibility, Functionality, and Speed. J. Phys. Chem. B 2024, 128, 9976--10042

    2024

  43. [51]

    Constructing multidimensional molecular potential energy surfaces from ab initio data

    Hollebeek, T.; Ho, T.-S.; Rabitz, H. Constructing multidimensional molecular potential energy surfaces from ab initio data. Ann. Rev. Phys. Chem. 1999, 50, 537--570

    1999

  44. [52]

    K.; Straight, S

    Reddy, S. K.; Straight, S. C.; Bajaj, P.; Huy Pham, C.; Riera, M.; Moberg, D. R.; Morales, M. A.; Knight, C.; G \"o tz, A. W.; Paesani, F. On the accuracy of the MB-pol many-body potential for water: Interaction energies, vibrational frequencies, and classical thermodynamic an...

    2016

  45. [53]

    Partridge, H.; Schwenke, D. W. The determination of an accurate isotope dependent potential energy surface for water from extensive ab initio calculations and experimental data. J. Chem. Phys. 1997, 106, 4618--4639

    1997

  46. [54]

    Brooks, B. R. et al. CHARMM: The Biomolecular Simulation Program. J. Comput. Chem. 2009, 30, 1545--1614

    2009

  47. [55]

    o decker, M.; Schweer, S. M.; Dupont, J.; Lep \`e re, V.; Zehnacker-Rentien, A.; Suhm, M. A.; Schr \

    Fischer, T. L.; B \"o decker, M.; Schweer, S. M.; Dupont, J.; Lep \`e re, V.; Zehnacker-Rentien, A.; Suhm, M. A.; Schr \"o der, B.; Henkes, T.; Andrada, D. M., et al. The first HyDRA challenge for computational vibrational spectroscopy. Phys. Chem. Chem. Phys. 2023, 25, 22089--22102

    2023

  48. [56]

    D.; Devereux, M.; Meuwly, M

    Boittier, E. D.; Devereux, M.; Meuwly, M. Molecular dynamics with conformationally dependent, distributed charges. J. Chem. Theory Comput. 2022, 18, 7544--7554

    2022

  49. [57]

    B97X-V: A 10-parameter, range-separated hybrid, generalized gradient approximation density functional with nonlocal correlation, designed by a survival-of-the-fittest strategy

    Mardirossian, N.; Head-Gordon, M. B97X-V: A 10-parameter, range-separated hybrid, generalized gradient approximation density functional with nonlocal correlation, designed by a survival-of-the-fittest strategy. Phys. Chem. Chem. Phys. 2014, 16, 9904--9924

    2014

  50. [58]

    A look at the density functional theory zoo with the advanced GMTKN55 database for general main group thermochemistry, kinetics and noncovalent interactions

    Goerigk, L.; Hansen, A.; Bauer, C.; Ehrlich, S.; Najibi, A.; Grimme, S. A look at the density functional theory zoo with the advanced GMTKN55 database for general main group thermochemistry, kinetics and noncovalent interactions. Phys. Chem. Chem. Phys. 2017, 19, 32184--32215

    2017

  51. [59]

    A unified formulation of the constant temperature molecular dynamics methods

    Nosé, S. A unified formulation of the constant temperature molecular dynamics methods. J. Chem. Phys. 1984, 81, 511–519

    1984

  52. [60]

    VMD -- V isual M olecular D ynamics

    Humphrey, W.; Dalke, A.; Schulten, K. VMD -- V isual M olecular D ynamics. J. Mol. Graph. 1996, 14, 33--38

    1996

  53. [61]

    Characterization of the Local Structure in Liquid Water by Various Order Parameters

    Duboué-Dijon, E.; Laage, D. Characterization of the Local Structure in Liquid Water by Various Order Parameters. J. Phys. Chem. B 2015, 119, 8406–8418

    2015

  54. [62]

    L.; Faires, J

    Burden, R. L.; Faires, J. D. Numerical analysis, brooks. 1997

    1997

  55. [63]

    L.; Jenson, C

    Jorgensen, W. L.; Jenson, C. Temperature dependence of TIP3P, SPC, and TIP4P water from NPT Monte Carlo simulations: Seeking temperatures of maximum density. J. Comput. Chem. 1998, 19, 1179--1186

    1998

  56. [64]

    Dudek, A. E. Circular block bootstrap for coefficients of autocovariance function of almost periodically correlated time series. Metrika 2015, 78, 313–335

    2015

  57. [65]

    Dipole moment fluctuation formulas in computer simulations of polar systems

    Neumann, M. Dipole moment fluctuation formulas in computer simulations of polar systems. Mol. Phys. 1983, 50, 841–858

    1983

  58. [66]

    Molecular dynamics simulation of a polymer chain in solution

    D \"u nweg, B.; Kremer, K. Molecular dynamics simulation of a polymer chain in solution. J. Chem. Phys. 1993, 99, 6983--6997

    1993

  59. [67]

    System-Size Dependence of Diffusion Coefficients and Viscosities from Molecular Dynamics Simulations with Periodic Boundary Conditions

    Yeh, I.-C.; Hummer, G. System-Size Dependence of Diffusion Coefficients and Viscosities from Molecular Dynamics Simulations with Periodic Boundary Conditions. J. Phys. Chem. B 2004, 108, 15873–15879

    2004

  60. [68]

    K.; Meuwly, M.; Karplus, M

    El Hage, K.; H\'edin, F.; Gupta, P. K.; Meuwly, M.; Karplus, M. Valid molecular dynamics simulations of human hemoglobin require a surprisingly large box size. eLife 2018, 7, e35560

    2018

  61. [69]

    S.; Wieder, M.; Xu, Y.; Zhang, J

    Wang, Y.; Takaba, K.; Chen, M. S.; Wieder, M.; Xu, Y.; Zhang, J. Z.; Yu, K.; Wang, X.; Zhang, L.; Cole, D. J., et al. On the design space between molecular mechanics and machine learning force fields. arXiv preprint arXiv:2409.01931 2024,

    2024 arXiv

  62. [70]

    Nejad, A.; Suhm, M. A. Concerted pair motion due to double hydrogen bonding: The formic acid dimer case. J. Indian Inst. Sci. 2020, 100, 5--19

    2020

  63. [71]

    o pfer, K.; K \

    T \"o pfer, K.; K \"a ser, S.; Meuwly, M. Double proton transfer in hydrated formic acid dimer: Interplay of spatial symmetry and solvent-generated force on reactivity. Phys. Chem. Chem. Phys. 2022, 24, 13869--13882

    2022

  64. [72]

    L.; Sewell, T

    Guo, Y.; Thompson, D. L.; Sewell, T. D. Analysis of the zero-point energy problem in classical trajectory simulations. J. Chem. Phys. 1996, 104, 576--582

    1996

  65. [73]

    Piccardo, M.; Bloino, J.; Barone, V. Generalized vibrational perturbation theory for rotovibrational energies of linear, symmetric and asymmetric tops: Theory, approximations, and automated approaches to deal with medium-to-large molecular systems. Int. J. Quantum Chem. 2015, ...

    2015

  66. [74]

    E.; Rao, K

    Fraley, P. E.; Rao, K. N. High resolution infrared spectra of water vapor: _1 and _3 band of H _2^ 16 O. J. Mol. Spectrosc. 1969, 29, 348--364

    1969

  67. [75]

    Benedict, W.; Gailar, N.; Plyler, E. K. Rotation-vibration spectra of deuterated water vapor. J. Chem. Phys. 1956, 24, 1139--1165

    1956

  68. [76]

    E.; Lan, Z

    Bertie, J. E.; Lan, Z. Infrared intensities of liquids XX: The intensity of the OH stretching band of liquid water revisited, and the best current values of the optical constants of H _2 O (l) at 25 C between 15,000 and 1 cm ^ -1 . Appl. Spectrosc. 1996, 50, 1047--1057

    1996

  69. [77]

    Frisch, M. J. et al. Gaussian 16 R evision B .01. 2016; Gaussian Inc. Wallingford CT

    2016

  70. [78]

    J.; Brooks III, C

    Price, D. J.; Brooks III, C. L. A modified TIP3P water potential for simulation with Ewald summation. J. Chem. Phys. 2004, 121, 10096--10103

    2004

  71. [79]

    A.; Mead, R

    Nelder, J. A.; Mead, R. A simplex method for function minimization. Comp. J. 1965, 7, 308--313

    1965

  72. [80]

    Ueber Projectionsmodelle der regelmässigen vier-dimensionalen Körper

    Schlegel, V. Ueber Projectionsmodelle der regelmässigen vier-dimensionalen Körper . Zeitschrift für Mathematik und Physik 1886, 31, 297--321

  73. [81]

    Polarizable multipolar molecular dynamics using distributed point charges

    Devereux, M.; Pezzella, M.; Raghunathan, S.; Meuwly, M. Polarizable multipolar molecular dynamics using distributed point charges. J. Chem. Theory Comput. 2020, 16, 7267--7280

    2020

  74. [82]

    Fadda, E.; Woods, R. J. On the Role of Water Models in Quantifying the Binding Free Energy of Highly Conserved Water Molecules in Proteins: The Case of Concanavalin A. Journal of Chemical Theory and Computation 2011, 7, 3391–3398

    2011

  75. [83]

    Abascal, J. L. F.; Vega, C. A general purpose model for the condensed phases of water: TIP4P/2005. J. Chem. Phys. 2005, 123, 234505

    2005

  76. [84]

    M.; Artacho, E.; Fernández-Serra, M.-V

    Wang, J.; Román-Pérez, G.; Soler, J. M.; Artacho, E.; Fernández-Serra, M.-V. Density, structure, and dynamics of water: The effect of van der Waals interactions. J. Chem. Phys. 2011, 134, 024516

    2011

  77. [85]

    Systematic parametrization of polarizable force fields from quantum chemistry data

    Wang, L.-P.; Chen, J.; Van Voorhis, T. Systematic parametrization of polarizable force fields from quantum chemistry data. J. Chem. Theory Comput. 2013, 9, 452--460

    2013

  78. [86]

    F.; Paesani, F

    Zhu, X.; Riera, M.; Bull-Vulpe, E. F.; Paesani, F. MB-pol(2023): Sub-chemical Accuracy for Water Simulations from the Gas to the Liquid Phase. J. Chem. Theory Comput. 2023, 19, 3551–3566

    2023

  79. [87]

    C.; Tschumper, G

    Howard, J. C.; Tschumper, G. S. Benchmark structures and harmonic vibrational frequencies near the CCSD(T) complete basis set limit for small water clusters:(H _2 O) _ n=2, 3, 4, 5, 6 . J. Chem. Theory Comput. 2015, 11, 2126--2136

    2015

  80. [88]

    Sublimation pressure and sublimation enthalpy of H _2 O ice Ih between 0 and 273.16 K

    Feistel, R.; Wagner, W. Sublimation pressure and sublimation enthalpy of H _2 O ice Ih between 0 and 273.16 K. Geochim. Cosmochim. Acta 2007, 71, 36–45

    2007

  81. [89]

    Can We Learn the Energy of Sublimation of Ice from Water Clusters? arXiv e-prints 2024, arXiv:2408.05234

    Bowman , J.; Yu , Q.; Qu , C.; Houston , P.; Conte , R. Can We Learn the Energy of Sublimation of Ice from Water Clusters? arXiv e-prints 2024, arXiv:2408.05234

    2024 arXiv

  82. [90]

    Ryckaert, J.-P.; Ciccotti, G.; Berendsen, H. J. Numerical integration of the cartesian equations of motion of a system with constraints: molecular dynamics of n-alkanes. J. Comput. Phys. 1977, 23, 327--341

    1977

  83. [91]

    D.; Wang, L.-P.; Huggins, D

    Wade, A. D.; Wang, L.-P.; Huggins, D. J. Assimilating Radial Distribution Functions To Build Water Models with Improved Structural Properties. J. Chem. Inf. and Mod. 2018, 58, 1766–1778

    2018

  84. [92]

    V.; Roux, B.; MacKerell Jr, A

    Lamoureux, G.; Harder, E.; Vorobyov, I. V.; Roux, B.; MacKerell Jr, A. D. A polarizable model of water for molecular dynamics simulations of biomolecules. Chem. Phys. Lett. 2006, 418, 245--249

    2006

  85. [93]

    T.; Pettersson, L

    Camisasca, G.; Pathak, H.; Wikfeldt, K. T.; Pettersson, L. G. M. Radial distribution functions of water: Models vs experiments. J. Chem. Phys. 2019, 151, 044502

    2019

  86. [94]

    L.; Conte, R.; Nandi, A.; Bowman, J

    Qu, C.; Yu, Q.; Houston, P. L.; Conte, R.; Nandi, A.; Bowman, J. M. Interfacing q-AQUA with a polarizable force field: The best of both worlds. J. Chem. Theory Comput. 2023, 19, 3446--3459

    2023

  87. [95]

    R.; Debenedetti, P

    Errington, J. R.; Debenedetti, P. G. Relationship between structural order and the anomalies of liquid water. Nature 2001, 409, 318--321

    2001

  88. [96]

    H.; Tsironi, I.; Mariedahl, D.; Blanco, M.; Huotari, S.; Honkimäki, V.; Nilsson, A

    Pathak, H.; Späh, A.; Kim, K. H.; Tsironi, I.; Mariedahl, D.; Blanco, M.; Huotari, S.; Honkimäki, V.; Nilsson, A. Intermediate range O–O correlations in supercooled water down to 235 K. The Journal of Chemical Physics 2019, 150, 224506

    2019

  89. [97]

    E.; Head-Gordon, T

    Ruiz Pestana, L.; Marsalek, O.; Markland, T. E.; Head-Gordon, T. The Quest for Accurate Liquid Water Properties from First Principles. The Journal of Physical Chemistry Letters 2018, 9, 5009–5016

    2018

  90. [98]

    Teng, X.; Yu, W.; MacKerell, A. D. J. Revised 4‑Point Water Model for the Classical Drude Oscillator Polarizable Force Field: SWM4‑HLJ. J. Chem. Theory Comput. 2024, 20, xxxx--xxxx

    2024

  91. [99]

    The Q‑AMOEBA (CF) Polarizable Potential

    Mauger, N.; Pl \'e , T.; Lagard \`e re, L.; Huppert, S.; Piquemal, J. The Q‑AMOEBA (CF) Polarizable Potential . arXiv preprint arXiv:2502.12708 2025,

    2025 arXiv

  92. [100]

    Thermal conductivity, shear viscosity and specific heat of rigid water models

    Mao, Y.; Zhang, Y. Thermal conductivity, shear viscosity and specific heat of rigid water models. Chem. Phys. Lett. 2012, 542, 37–41

    2012

  93. [101]

    Lankhorst, D.; Schriever, J.; Leyte, J. C. Determination of the Rotational Correlation Time of Water by Proton NMR Relaxation in H217O and Some Related Results. Berichte der Bunsengesellschaft für physikalische Chemie 1982, 86, 215–221

    1982

  94. [102]

    Rezus, Y. L. A.; Bakker, H. J. On the orientational relaxation of HDO in liquid water. The Journal of Chemical Physics 2005, 123, 114502

    2005

  95. [103]

    The IAPWS formulation 1995 for the thermodynamic properties of ordinary water substance for general and scientific use

    Wagner, W.; Pru , A. The IAPWS formulation 1995 for the thermodynamic properties of ordinary water substance for general and scientific use. J. Phys. Chem. Ref. Data 2002, 31, 387--535

    1995

  96. [104]

    Communication: Thermodynamics of water modeled using ab initio simulations

    Weber, V.; Asthagiri, D. Communication: Thermodynamics of water modeled using ab initio simulations. J. Chem. Phys. 2010, 133

    2010

  97. [105]

    R.; Sacco, A

    Holz, M.; Heil, S. R.; Sacco, A. Temperature-dependent self-diffusion coefficients of water and six selected molecular liquids for calibration in accurate 1H NMR PFG measurements. Physical Chemistry Chemical Physics 2000, 2, 4740–4742

    2000

  98. [106]

    Polarizable Water Model with Ab Initio Neural Network Dynamic Charges and Spontaneous Charge Transfer

    Liang , Q.; Yang , J. Polarizable Water Model with Ab Initio Neural Network Dynamic Charges and Spontaneous Charge Transfer . arXiv e-prints 2024, arXiv:2410.19445

    2024 arXiv

  99. [107]

    M.; Shi, L

    Han, B.; Isborn, C. M.; Shi, L. Incorporating polarization and charge transfer into a point-charge model for water using machine learning. J. Phys. Chem. Lett. 2023, 14, 3869--3877

    2023

  100. [108]

    Isotope effects in liquid water by infrared spectroscopy

    Max, J.-J.; Chapados, C. Isotope effects in liquid water by infrared spectroscopy. III. H _2 O and D _2 O spectra from 6000 to 0 cm ^ -1 . J. Chem. Phys. 2009, 131, 184505

    2009

  101. [109]

    Rumble, J. R., Ed. CRC Handbook of Chemistry and Physics, 104th ed.; CRC Press: Boca Raton, FL, 2023

    2023

  102. [110]

    H :ܼt \(**?>M>^|> []]] ͑Jmnn b׮]p[

    Lemmon, E. W.; Huber, M. L.; McLinden, M. O. REFPROP: Reference Fluid Thermodynamic and Transport Properties Database. National Institute of Standards and Technology (NIST), 2022; https://www.nist.gov/srd/refprop mcitethebibliography arxiv.tex0000664000000000000000000023004515...

    2023

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.