pith. sign in

arxiv: 2604.15852 · v1 · submitted 2026-04-17 · ❄️ cond-mat.soft

Voids in liquids: peculiarities of molecular dynamics simulation of fluid systems

Yu. D. Fomin This is my paper

Pith reviewed 2026-05-10 07:35 UTC · model grok-4.3

classification ❄️ cond-mat.soft
keywords molecular dynamicsliquid voidscritical temperaturetwo-phase regionfluid simulationsphase separationcavities
0
0 comments X

The pith

Large voids in molecular dynamics simulations of liquids appear only when the system is above its critical temperature or in the two-phase liquid-gas region.

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

The paper shows that large cavities observed in simulations of liquids are not anomalies but follow directly from the system's location on the phase diagram. Cavities form when temperature exceeds the critical temperature of the liquid-gas transition or when the simulation is in the coexistence region where liquid and gas phases separate. A reader would care because this explains why simulations sometimes contradict the everyday picture of liquids as uniformly dense without big empty spaces. It emphasizes that correct interpretation requires checking the thermodynamic state rather than assuming a stable single-phase liquid.

Core claim

The cavities appear either if the temperature of the system is above the critical temperature of liquid-gas transition or if the system is in two-phase liquid-gas region. These conclusions are illustrated by several examples from literature and our own simulations.

What carries the argument

The thermodynamic state of the fluid relative to the critical point and the liquid-gas coexistence line, which controls whether voids from phase separation or supercritical fluctuations appear.

If this is right

  • Simulations reporting large voids in liquids must be checked against the phase diagram to confirm whether they are above the critical temperature or inside the two-phase region.
  • Below the critical temperature and outside coexistence, a correct simulation of a dense liquid should not produce large stable cavities.
  • Interpretation of structural features like voids requires explicit verification that the simulated state is a single-phase liquid.
  • Examples from prior literature that show voids can be reclassified as supercritical or two-phase cases once their temperature and density are mapped.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Simulators of soft-matter fluids may need to report the reduced temperature relative to the model's critical point to avoid misidentifying phase-separated states as uniform liquids.
  • This distinction could clarify results in other simulation studies of liquids near their critical points where apparent voids are routinely observed.
  • A direct test would be to quench a two-phase simulation below the critical temperature at fixed overall density and observe whether the voids shrink and disappear as the system equilibrates into a single dense phase.

Load-bearing premise

The large voids seen in the simulations result solely from the thermodynamic state being supercritical or two-phase rather than from finite-size effects, truncation, or equilibration problems.

What would settle it

A well-equilibrated simulation of a model liquid at a temperature well below its critical temperature, with average density corresponding to the single-phase liquid region and large enough system size, that still develops persistent macroscopic voids would falsify the claim.

Figures

Figures reproduced from arXiv: 2604.15852 by Yu. D. Fomin.

Figure 1
Figure 1. Figure 1: FIG. 1: Snapshots of 2d LJ system at [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2: Equation of state of GAP-20 model of carbon at [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3: Snapshots of GAP-20 model of carbon at [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4: Snapshots of configurations of 4000 LJ particles at [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5: Snapshots of configurations of 4000 LJ particles at [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6: Snapshots of configurations of 10 [PITH_FULL_IMAGE:figures/full_fig_p008_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7: Comparison of simulation of water in (a) NVT and [PITH_FULL_IMAGE:figures/full_fig_p009_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: FIG. 8: Relaxation of density of water starting from different [PITH_FULL_IMAGE:figures/full_fig_p009_8.png] view at source ↗
read the original abstract

Molecular dynamics is a powerful tool to investigate the properties of fluid systems. However, a correct interpretation of the results of simulations is required. In particular, some simulations show appearance of large voids in liquids, which contradicts our common sense on what is liquid. In the present paper we discuss the origin of large cavities liquids in molecular dynamics simulations. We demonstrate that the cavities appear either if the temperature of the system is above the critical temperature of liquid-gas transition or if the system is in two-phase liquid-gas region. These conclusions are illustrated by several examples from literature and our own simulations.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript claims that large voids observed in molecular dynamics simulations of liquids are not unphysical but arise precisely when the system temperature exceeds the critical temperature of the liquid-gas transition or when the simulation is in the two-phase coexistence region. The argument is illustrated by selected literature examples and the authors' own simulations.

Significance. If the central claim is substantiated with appropriate controls, the work would provide a useful reminder that apparent voids in liquid simulations are expected thermodynamic features rather than simulation pathologies, helping researchers correctly interpret results near critical points or in coexistence. It draws on standard phase-equilibrium knowledge without introducing new parameters or entities.

major comments (2)
  1. [Abstract] Abstract: the central claim that voids appear 'either if the temperature of the system is above the critical temperature ... or if the system is in two-phase liquid-gas region' is not accompanied by quantitative checks, error bars, or explicit tests that rule out alternative causes such as finite-size effects, cutoff truncation, or insufficient equilibration.
  2. [Simulation details] Simulation section: no system-size scaling (e.g., N = 500 vs. 5000), cutoff-radius variation, or extended equilibration diagnostics are reported to confirm that large voids are absent in single-phase subcritical liquids and appear only when the thermodynamic state crosses Tc or enters coexistence.
minor comments (1)
  1. [Abstract] Abstract: 'large cavities liquids' is missing the preposition 'in'.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments. We agree that quantitative controls are needed to strengthen the central claim and will add them in revision. Below we address each point.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claim that voids appear 'either if the temperature of the system is above the critical temperature ... or if the system is in two-phase liquid-gas region' is not accompanied by quantitative checks, error bars, or explicit tests that rule out alternative causes such as finite-size effects, cutoff truncation, or insufficient equilibration.

    Authors: We accept this criticism. In the revised manuscript we will add quantitative diagnostics: (i) void-size histograms with error bars obtained from independent runs, (ii) explicit comparison of void statistics for the same state point simulated with different cutoffs (1.0 nm vs 1.5 nm), and (iii) extended equilibration runs (10 ns vs 50 ns) showing that void statistics stabilize after ~5 ns. These additions will be placed in a new subsection of Results and will directly test that large voids are absent below Tc in the single-phase region. revision: yes

  2. Referee: [Simulation details] Simulation section: no system-size scaling (e.g., N = 500 vs. 5000), cutoff-radius variation, or extended equilibration diagnostics are reported to confirm that large voids are absent in single-phase subcritical liquids and appear only when the thermodynamic state crosses Tc or enters coexistence.

    Authors: We agree that system-size and cutoff controls are missing. In revision we will perform and report additional simulations at N = 500, 2000 and 5000 particles for the same subcritical single-phase state point, demonstrating that the probability of large voids remains negligible (<0.1 %) independent of N. We will also repeat the near-critical and two-phase runs at two different cutoffs and show that the appearance of macroscopic voids is insensitive to cutoff once the thermodynamic state is fixed. These new data will be presented as supplementary figures and referenced in the Simulation section. revision: yes

Circularity Check

0 steps flagged

No circularity: thermodynamic attribution of voids rests on independent phase-transition knowledge

full rationale

The paper's derivation chain consists of (1) recalling the standard thermodynamic definition of the critical temperature Tc and the liquid-gas coexistence region, (2) running or citing MD trajectories at various state points, and (3) observing that large voids appear precisely when the simulated state point lies above Tc or inside the two-phase dome. None of these steps reduces to a self-definition, a fitted parameter renamed as a prediction, or a load-bearing self-citation. Tc is an externally known quantity from independent equations of state or experiments; the simulation observations are presented as empirical illustrations rather than as the sole justification for the thermodynamic boundary. No ansatz is smuggled in, no uniqueness theorem is invoked, and no known empirical pattern is merely relabeled. The central claim therefore remains logically independent of its own inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

Only abstract available; no explicit free parameters, axioms, or invented entities are stated. The claim implicitly rests on the existence of a well-defined critical temperature and two-phase region for the model fluids used.

axioms (2)
  • domain assumption Every fluid model possesses a critical temperature separating liquid-like and gas-like regimes.
    Invoked when attributing voids to T > Tc.
  • domain assumption Molecular dynamics with periodic boundaries can stably represent two-phase coexistence when average density lies inside the binodal.
    Required to interpret voids as gas bubbles rather than artifacts.

pith-pipeline@v0.9.0 · 5383 in / 1301 out tokens · 34247 ms · 2026-05-10T07:35:03.780485+00:00 · methodology

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

46 extracted references · 9 canonical work pages

  1. [1]

    Then we use the obtained structure as an initial one for the simulations at lower temperatures

    we firstly equilibrate the system at very high temper- atureT= 3000 K for 5 ps. Then we use the obtained structure as an initial one for the simulations at lower temperatures. These simulations also take 5 ps. The timestep is 1 fs. The pressure is set to zero. The box is allowed to change its dimension in all directions, while the angles of the box are fixe...

    2005

  2. [2]

    Debenedetti, Th.M

    P.G. Debenedetti, Th.M. Truskett, Fluid Phase Equilibria 158-160 (1999) 549-556

    1999

  3. [3]

    Akola, R.O

    J. Akola, R.O. Jones, S. Kohara, T. Usuki, E. Bychkov, Phys. Rev. B 81 (2010) 094202

    2010

  4. [4]

    Logunov, N

    M. Logunov, N. Orekhov, Carbon 192 (2022) 179-186

    2022

  5. [5]

    Cai, H.A

    Y. Cai, H.A. Wu, S.N. Luo, J. Chem. Phys. 140 (2014) 214317. doi:10.1063/1.4880960

    work page doi:10.1063/1.4880960 2014

  6. [6]

    Kuksin, G

    A. Kuksin, G. Norman, V. Stegailov, A. Yanilkin, P. Zhilyaev, Int. J. Fract. 162 (2010) 127-136

    2010

  7. [7]

    Fomin, E.N

    Yu.D. Fomin, E.N. Tsiok, V.N. Ryzhov, J. Chem. Phys. 163 (2025) 174501. https://doi.org/10.1063/5.0298557

    work page doi:10.1063/5.0298557 2025

  8. [8]

    Rowe, V.L

    P. Rowe, V.L. Deringer, P. Gasparotto, G. Csanyi, A. Michaelides, J. Chem. Phys. 153 (2020) 034702

    2020

  9. [9]

    Fomin, arXiv:2602.23786 (2026)

    Yu.D. Fomin, arXiv:2602.23786 (2026). https://doi.org/10.48550/arXiv.2602.23786

    work page doi:10.48550/arxiv.2602.23786 2026

  10. [10]

    Mastny, J.J

    E.A. Mastny, J.J. de Pablo, J. Chem. Phys. 127 (2007) 104504

    2007

  11. [11]

    Stephan, M

    S. Stephan, M. Thol, J. Vrabec, H. Hasse, J. Chem. Inf. Model. 59 (2019) 4248-4265

    2019

  12. [12]

    Abascal, C

    J.L. Abascal, C. Vega, J. Chem. Phys. 123 (2005) 234505. doi:10.1063/1.2121687

    work page doi:10.1063/1.2121687 2005

  13. [13]

    Vega, J.L.F

    C. Vega, J.L.F. Abascal, I. Nezbeda, J. Chem. Phys. 125 (2006) 034503

    2006

  14. [14]

    Thompson et al., Comput

    A. Thompson et al., Comput. Phys. Commun. 271 (2022) 108171

    2022

  15. [15]

    Kresse, J

    G. Kresse, J. Hafner, Phys. Rev. B 47 (1993) 558

    1993

  16. [16]

    Kresse, J

    G. Kresse, J. Hafner, Phys. Rev. B 49 (1994) 14251

    1994

  17. [17]

    Kresse, J

    G. Kresse, J. Furthmuller, Mater. Sci. 6 (1996) 15

    1996

  18. [18]

    Kresse, J

    G. Kresse, J. Furthmuller, Phys. Rev. B 54 (1996) 11169

    1996

  19. [19]

    Kresse, D

    G. Kresse, D. Joubert, Phys. Rev. B 59 (1999) 1758

    1999

  20. [20]

    Perdew, A

    J.P. Perdew, A. Zunger, Phys. Rev. B 23 (1981) 5048

    1981

  21. [21]

    Stukowski, Modelling Simul

    A. Stukowski, Modelling Simul. Mater. Sci. Eng. 18 (2010) 015012

    2010

  22. [22]

    Debenedetti, J

    P.G. Debenedetti, J. Peters, Supercritical flu- ids: fundamentals and applications, Kluwer Academic Publishers, 2000

    2000

  23. [23]

    L. Xu, P. Kumar, S.V. Buldyrev, S.-H. Chen, P.H. Poole, F. Sciortino, H.E. Stanley, Proc. Natl. Acad. Sci. U.S.A. 102 (2005) 16558- 16562

    2005

  24. [24]

    Brazhkin, Yu.D

    V.V. Brazhkin, Yu.D. Fomin, A.G. Lyapin, V.N. Ryzhov, E.N. Tsiok, J. Phys. Chem. B 115 (2011) 14112-14115

    2011

  25. [25]

    Brazhkin, A.G

    V.V. Brazhkin, A.G. Lyapin, V.N. Ryzhov, K. Trachenko, Yu.D. Fomin, E.N. Tsiok, Phys. Usp. 55 (2012) 1061-1079

    2012

  26. [26]

    Brazhkin, Yu.D

    V.V. Brazhkin, Yu.D. Fomin, A.G. Lyapin, V.N. Ryzhov, K. Trachenko, Phys. Rev. E 85 (2012) 031203

    2012

  27. [27]

    Brazhkin, Yu.D

    V.V. Brazhkin, Yu.D. Fomin, V.N. Ryzhov, E.E. Tareyeva, E.N. Tsiok, Phys. Rev. E 89 (2014) 042136

    2014

  28. [28]

    Fomin, V.N

    Yu.D. Fomin, V.N. Ryzhov, E.N. Tsiok, V.V. Brazhkin, Phys. Rev. E 91 (2015) 022111

    2015

  29. [29]

    Fomin, E.N

    Yu.D. Fomin, E.N. Tsiok, V.N. Ryzhov, J. Chem. Phys. 163 (2025) 174501. 11

    2025

  30. [30]

    Sedunov, J

    B.I. Sedunov, J. Thermodyn. 2011 (2011) 194353

    2011

  31. [31]

    Mareeva, V.A

    E.I. Mareeva, V.A. Aleshkevich, F.V. Potemkin, N.V. Minaev, V.M. Gordienko, Russ. J. Phys. Chem. B 13 (2019) 1214-1219

    2019

  32. [32]

    Mareev, V

    E. Mareev, V. Aleshkevich, F. Potemkin, V. Bagratashvili, N. Minaev, V. Gordienko, Opt. Express 26 (2018) 13229-13238

    2018

  33. [33]

    Abramo et al., J

    M.C. Abramo et al., J. Chem. Phys. 142 (2015) 214502

    2015

  34. [34]

    Billion atom molecular dynamics simulations of carbon at extreme conditions and experimental time and length scales,

    K. Nguyen-Cong, J.T. Willman, S.G. Moore, A.B. Belonoshko, R. Gayatri, E. Wein- berg, M.A. Wood, A.P. Thompson, I.I. Oleynik, in: SC ’21: Proceedings of the International Conference for High Perfor- mance Computing, Networking, Storage and Analysis, Article No. 4, 2021, pp. 1-12. https://doi.org/10.1145/3458817.3487400

    work page doi:10.1145/3458817.3487400 2021

  35. [35]

    D. Lu, H. Wang, M. Chen, L. Lin, R. Car, W. E, W. Jia, L. Zhang, Comput. Phys. Com- mun. 259 (2021) 107624

    2021

  36. [36]

    L. Xu, Z. Li, W. Xia, npj Soft Matter 1 (2025) 12. https://doi.org/10.1038/s44431- 025-00010-9

    work page doi:10.1038/s44431- 2025

  37. [37]

    Frenkel, B

    D. Frenkel, B. Smit, Understanding Molecu- lar Simulation, Cambridge University Press, 2000

    2000

  38. [38]

    Stadler, M.J

    R. Stadler, M.J. Gillan, J. Phys.: Condens. Matter 12 (2000) 6053-6061

    2000

  39. [39]

    Thurn, J

    H. Thurn, J. Ruska, J. Non-Cryst. Solids 22 (1976) 331-343

    1976

  40. [40]

    Tsuchiya, J

    Y. Tsuchiya, J. Phys.: Condens. Matter 3 (1991) 3163-3172

    1991

  41. [41]

    Takeda, H

    S. Takeda, H. Okazaki, S. Tamaki, J. Phys. Soc. Jpn. 54 (1985) 1890-1898

    1985

  42. [42]

    Tsuchiya, J

    Y. Tsuchiya, J. Phys. Soc. Jpn. 57 (1988) 3851-3857

    1988

  43. [43]

    Tsuchiya, J

    Y. Tsuchiya, J. Phys. Soc. Jpn. 60 (1991) 960- 967

    1991

  44. [44]

    Brazhkin, I.V

    V.V. Brazhkin, I.V. Danilov, O.B. Tsiok, JETP Lett. 117 (2023) 834-848. https://doi.org/10.1134/S0021364023601318

    work page doi:10.1134/s0021364023601318 2023

  45. [45]

    Anta, B.J

    J.A. Anta, B.J. Jesson, P.A. Mad- den, Phys. Rev. B 58 (1998) 6124. https://doi.org/10.1103/PhysRevB.58.6124

    work page doi:10.1103/physrevb.58.6124 1998

  46. [46]

    Nagata, T

    Yu. Nagata, T. Ohto, M. Bonn, Th.D. Kuhne, J. Chem. Phys. 144 (2016) 204705. https://doi.org/10.1063/1.4951710

    work page doi:10.1063/1.4951710 2016