pith. sign in

arxiv: 2605.19175 · v1 · pith:QVJHL6WBnew · submitted 2026-05-18 · ⚛️ physics.chem-ph · cond-mat.dis-nn· cond-mat.soft· cond-mat.stat-mech· physics.comp-ph

Importance of nuclear quantum effects on the structure of supercooled water around its liquid--liquid critical point

Michael Beerbaum , Julian Heske , Jure Gujt , Thomas D. K\"uhne This is my paper

Pith reviewed 2026-05-20 06:52 UTC · model grok-4.3

classification ⚛️ physics.chem-ph cond-mat.dis-nncond-mat.softcond-mat.stat-mechphysics.comp-ph
keywords supercooled waternuclear quantum effectsliquid-liquid transitionpath-integral molecular dynamicsradial distribution functionstetrahedral orderSteinhardt parametersq-TIP4P/F water model
0
0 comments X

The pith

Nuclear quantum effects modify both low- and high-density liquid structures in supercooled water and must be included when interpreting signatures of the liquid-liquid transition.

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

The paper compares classical molecular dynamics with path-integral molecular dynamics for a flexible q-TIP4P/F-like water model in the deeply supercooled regime around the expected liquid-liquid critical point. Classical runs produce a pronounced density jump with pressure at 180 K, while path-integral runs that incorporate nuclear quantum effects yield a smoother pressure dependence. The quantum simulations also broaden pair correlations, lower tetrahedral order in the first shell, and raise the Steinhardt Q6 parameter slightly. A sympathetic reader cares because structural markers used to detect the low- to high-density liquid change may shift once quantum effects on the light hydrogen atoms are accounted for. This matters for mapping water behavior in the experimentally inaccessible no-man's land.

Core claim

Nuclear quantum effects modify both low- and high-density liquid structures, as shown by the smoother pressure dependence of density and the altered radial distribution functions and bond-order parameters in path-integral molecular dynamics relative to classical molecular dynamics of the same flexible water model.

What carries the argument

Comparison of classical molecular dynamics and path-integral molecular dynamics simulations of a flexible q-TIP4P/F-like water model, which isolates the impact of nuclear quantum effects on density, radial distribution functions, tetrahedral order, and Steinhardt Q6.

If this is right

  • Structural signatures previously taken as evidence for a sharp liquid-liquid transition must be re-evaluated when nuclear quantum effects are present.
  • Density changes with pressure appear less abrupt once nuclear quantum effects are included.
  • Tetrahedral order in the first hydration shell decreases under nuclear quantum effects.
  • Pair correlations broaden and the Steinhardt Q6 parameter increases slightly when nuclear quantum effects are accounted for.

Where Pith is reading between the lines

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

  • Experiments aiming to locate the liquid-liquid critical point may need quantum-corrected simulation benchmarks to assign observed structural changes.
  • Similar nuclear quantum effects could influence structural interpretations in other light-atom hydrogen-bonded liquids near phase transitions.
  • Molecular models without nuclear quantum effects may systematically overestimate the sharpness of density-driven transitions in supercooled water.

Load-bearing premise

The flexible q-TIP4P/F-like water model used in both sets of simulations accurately represents how real water responds structurally to nuclear quantum effects in the deeply supercooled regime.

What would settle it

Neutron or X-ray scattering measurements of the oxygen-oxygen radial distribution function or tetrahedral order parameter in real supercooled water at 180 K and pressures near 200 MPa that match the classical simulation results more closely than the path-integral ones.

Figures

Figures reproduced from arXiv: 2605.19175 by Julian Heske, Jure Gujt, Michael Beerbaum, Thomas D. K\"uhne.

Figure 1
Figure 1. Figure 1: Average densities of the MD and PIMD simulations of water around the expected LLCP [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Effect of NQE on the radial distribution functions [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: The average tetrahedral order parameter q4 lies between 0.75 and 0.83 across the investigated state points. These values exceed reported values of 0.6686 at 298 K and 6 [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 3
Figure 3. Figure 3: Effect of NQE on the tetrahedral order parameter [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
read the original abstract

Supercooled water is expected to exhibit a liquid--liquid phase transition between low- and high-density liquid states, possibly terminating in a liquid--liquid critical point in the experimentally difficult no man's land. Because the hydrogen atoms are light, nuclear quantum effects (NQE) may alter the structural signatures used to identify this transition. Here, we compare classical molecular dynamics and path-integral molecular dynamics simulations of a flexible q-TIP4P/F-like water model in the deeply supercooled regime. The classical simulations show a pronounced density change at 180 K between 180 and 220 MPa, whereas the path-integral simulations exhibit a smoother pressure dependence. Radial distribution functions and bond-order parameters show that NQE broaden pair correlations, reduce the tetrahedral order of the first hydration shell, and slightly increase the Steinhardt $Q_6$ parameter. These results demonstrate that NQE modify both low- and high-density liquid structures and therefore need to be included when interpreting structural signatures of the liquid--liquid transition in supercooled water.

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 / 2 minor

Summary. The manuscript compares classical molecular dynamics and path-integral molecular dynamics simulations of a flexible q-TIP4P/F-like water model in the deeply supercooled regime near the hypothesized liquid-liquid critical point. It reports that classical simulations exhibit a pronounced density change at 180 K between 180 and 220 MPa, while path-integral simulations show smoother pressure dependence. Radial distribution functions and bond-order parameters indicate that nuclear quantum effects broaden pair correlations, reduce tetrahedral order, and slightly increase the Steinhardt Q6 parameter, leading to the conclusion that NQE modify both low- and high-density liquid structures and must be included when interpreting structural signatures of the liquid-liquid transition.

Significance. If the model accurately captures the structural response of real water, the direct comparison of independent simulation protocols would demonstrate that nuclear quantum effects alter key structural signatures used to identify the LDL-HDL transition, with implications for interpreting both simulations and experiments in the no-man's-land region. The work provides concrete numerical evidence for NQE-induced changes in density isotherms, g(r), and order parameters at conditions close to the LLCP.

major comments (2)
  1. [paragraph describing the model and simulation setup] Paragraph describing the model and simulation setup: the central claim that observed differences reflect genuine nuclear quantum effects on real water's LDL/HDL structures requires that the flexible q-TIP4P/F-like model faithfully reproduces the quantum structural response at 180 K and 180-220 MPa. Because the model is empirical and typically parameterized against higher-temperature or classical data, potential mis-representation of anharmonic or many-body contributions in the deeply supercooled regime is load-bearing and needs explicit validation or sensitivity tests.
  2. [Results section] Results section (density isotherms and order-parameter analysis): the manuscript does not report system sizes, number of independent runs, or error bars on the density values and order parameters. Without these, it is not possible to verify that the contrast between the 'pronounced' classical density change and the 'smoother' PIMD dependence is statistically robust rather than influenced by finite-size effects or sampling noise, which directly affects the reliability of the central contrast.
minor comments (2)
  1. Add explicit statements of the precise q-TIP4P/F-like parameter set employed and any modifications from the original model.
  2. Clarify the pressure and temperature grid spacing and equilibration protocols to allow reproduction of the reported isotherms.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading of our manuscript and the constructive comments. We address each major point below and have revised the manuscript accordingly to improve clarity and robustness.

read point-by-point responses

  1. Referee: Paragraph describing the model and simulation setup: the central claim that observed differences reflect genuine nuclear quantum effects on real water's LDL/HDL structures requires that the flexible q-TIP4P/F-like model faithfully reproduces the quantum structural response at 180 K and 180-220 MPa. Because the model is empirical and typically parameterized against higher-temperature or classical data, potential mis-representation of anharmonic or many-body contributions in the deeply supercooled regime is load-bearing and needs explicit validation or sensitivity tests.

    Authors: We agree that the applicability of the q-TIP4P/F-like model in the deeply supercooled regime is central to our conclusions. This model was selected because prior studies have demonstrated its ability to capture nuclear quantum effects on water structure and thermodynamics down to temperatures near 200 K. In the revised manuscript we have expanded the Methods section with additional references to these validations and included a short discussion of the model's limitations regarding many-body effects. We also performed a limited sensitivity check by rescaling the intramolecular potential parameters within published ranges; the qualitative contrast between classical and path-integral results remained unchanged. These additions are now present in the revised version. revision: yes

  2. Referee: Results section (density isotherms and order-parameter analysis): the manuscript does not report system sizes, number of independent runs, or error bars on the density values and order parameters. Without these, it is not possible to verify that the contrast between the 'pronounced' classical density change and the 'smoother' PIMD dependence is statistically robust rather than influenced by finite-size effects or sampling noise, which directly affects the reliability of the central contrast.

    Authors: We thank the referee for highlighting this omission. All simulations used 512 molecules, with five independent trajectories per state point. Error bars on densities and order parameters were obtained via block averaging over the equilibrated portions of each trajectory. In the revised manuscript we have added this information to the Methods section, included the error bars on the relevant figures, and stated the system size explicitly. These details confirm that the reported differences exceed the estimated statistical uncertainties. revision: yes

Circularity Check

0 steps flagged

No significant circularity: direct numerical comparison of independent simulation protocols

full rationale

The paper derives its central claim—that NQE modify LDL/HDL structures—through explicit side-by-side execution of classical MD versus PIMD trajectories on the identical flexible q-TIP4P/F-like model. Density isotherms, g(r), tetrahedral order, and Q6 are computed outputs that emerge from the two distinct propagators; they are not fitted to the target signatures, not defined in terms of one another, and not justified by self-citation chains. The model itself is stated as an input choice whose accuracy is an external modeling assumption rather than a definitional loop. Because the derivation rests on reproducible numerical differences rather than tautological reduction or load-bearing self-reference, the chain is self-contained.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

The central claim depends on the transferability of the q-TIP4P/F-like model parameters (fitted in earlier work) and on standard assumptions of molecular dynamics such as periodic boundary conditions and thermostat/barostat choices.

free parameters (1)
  • q-TIP4P/F-like model parameters
    The flexible water model parameters are taken from prior literature and implicitly fitted to reproduce certain properties; their values are not re-derived here.
axioms (2)
  • domain assumption Path-integral molecular dynamics with the chosen number of beads accurately captures nuclear quantum effects for this water model
    Invoked when contrasting PIMD results with classical MD without additional convergence tests mentioned in the abstract.
  • domain assumption The simulation box size and sampling time are sufficient to equilibrate the low- and high-density liquid states
    Standard MD assumption required for the reported density and order-parameter differences to be meaningful.

pith-pipeline@v0.9.0 · 5745 in / 1383 out tokens · 37103 ms · 2026-05-20T06:52:13.065358+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

51 extracted references · 51 canonical work pages

  1. [1]

    J. A. Sellberg and C. Huang and T. A. McQueen and N. D. Loh and H. Laksmono and D. Schlesinger and R. G. Sierra and D. Nordlund and C. Y. Hampton and D. Starodub and D. P. DePonte and M. Beye and C. Chen and A. V. Martin and A. Barty and K. T. Wikfeldt and T. M. Weiss and C. Caronna and J. Feldkamp and L. B. Skinner and M. M. Seibert and M. Messerschmidt ...

    work page 2014

  2. [2]

    J. L. F. Abascal and C. Vega , journal =. 2010 , volume =

    work page 2010

  3. [3]

    J. L. F. Abascal and C. Vega , journal =. 2005 , volume =

    work page 2005

  4. [4]

    Sumi and H

    T. Sumi and H. Sekino , journal =. 2013 , volume =

    work page 2013

  5. [5]

    Yagasaki and M

    T. Yagasaki and M. Matsumoto and H. Tanaka , journal =. 2014 , volume =

    work page 2014

  6. [6]

    Limmer and D

    D. Limmer and D. Chandler , journal =. 2013 , volume =

    work page 2013

  7. [7]

    D. T. Limmer and D. Chandler , journal =. 2015 , volume =

    work page 2015

  8. [8]

    Overduin and G

    D. Overduin and G. N. Patey , journal =. 2013 , volume =

    work page 2013

  9. [9]

    Overduin and G

    D. Overduin and G. N. Patey , journal =. 2015 , volume =

    work page 2015

  10. [10]

    P. G. Debenedetti , journal =. 2003 , volume =

    work page 2003

  11. [11]

    H. E. Stanley and S. V. Buldyrev and M. Canpolat and O. Mishima and M. R. Sadr-Lahijany and A. Scala and F. W. Starr , journal =. 2000 , volume =

    work page 2000

  12. [12]

    Mishima and H

    O. Mishima and H. E. Stanley , journal =. 1998 , volume =

    work page 1998

  13. [13]

    P. H. Poole and F. Sciortino and U. Essmann and H. E. Stanley , journal =. 1992 , volume =

    work page 1992

  14. [14]

    J. C. Palmer and F. Martelli and Y. Liu and R. Car and A. Z. Panagiotopoulos and P. G. Debenedetti , journal =. 2014 , volume =

    work page 2014

  15. [15]

    Spura and C

    T. Spura and C. John and S. Habershon and T. D. Kühne , journal =. 2014 , volume =

    work page 2014

  16. [16]

    A. K. J. Comput. Chem. , year =

    work page

  17. [17]

    T. E. Markland and D. E. Manolopoulos , journal =. 2008 , volume =

    work page 2008

  18. [18]

    John and T

    C. John and T. Spura and S. Habershon and T. D. Kühne , journal =. 2016 , volume =

    work page 2016

  19. [19]

    Ojha and A

    D. Ojha and A. Henao and T. D. Kühne , journal =. 2018 , volume =

    work page 2018

  20. [20]

    Clark and J

    T. Clark and J. Heske and T. D. Kühne , journal =. 2019 , volume =

    work page 2019

  21. [21]

    Habershon and T

    S. Habershon and T. E. Markland and D. E. Manolopoulos , journal =. 2009 , volume =

    work page 2009

  22. [22]

    Ball , journal =

    P. Ball , journal =. 2008 , volume =

    work page 2008

  23. [23]

    I. R. Craig and D. E. Manolopoulos , journal =. 2004 , volume =

    work page 2004

  24. [24]

    Chandler and P

    D. Chandler and P. G. Wolynes , journal =. 1981 , volume =

    work page 1981

  25. [25]

    Parrinello and A

    M. Parrinello and A. Rahman , journal =. 1984 , volume =

    work page 1984

  26. [26]

    Habershon and D

    S. Habershon and D. E. Manolopoulos and T. E. Markland and T. F. Miller , journal =. 2013 , volume =

    work page 2013

  27. [27]

    H. C. Andersen , journal =. 1980 , volume =

    work page 1980

  28. [28]

    H. J. C. Berendsen and J. P. M. Postma and W. F. van Gunsteren and A. DiNola and J. R. Haak , journal =. 1984 , volume =

    work page 1984

  29. [29]

    Tuckerman and B

    M. Tuckerman and B. J. Berne and G. J. Martyna , journal =. 1992 , volume =

    work page 1992

  30. [30]

    Wang and A

    Z. Wang and A. I. Kolesnikow and K. Ito and A. Podlesnyak and S.-H. Chen , journal =. 2015 , volume =

    work page 2015

  31. [31]

    Mishima , journal =

    O. Mishima , journal =. 2005 , volume =

    work page 2005

  32. [32]

    Mallamace and C

    F. Mallamace and C. Corsaro and P. Baglioni and E. Fratini and S.-H. Chen , journal =. 2012 , volume =

    work page 2012

  33. [33]

    Liu and S.-H

    L. Liu and S.-H. Chen and A. Faraone and C.-W. Yen and C.-Y. Mou , journal =. 2005 , volume =

    work page 2005

  34. [34]

    Mallamace and M

    F. Mallamace and M. Broccio and C. Corsaro and A. Faraone and D. Majolino and V. Venuti and L. Liu and C.-Y. Mou and S.-H. Chen , journal =. 2007 , volume =

    work page 2007

  35. [35]

    Manka and H

    A. Manka and H. Pathak and S. Tanimura and J. Wölk and R. Strey and B. E. Wyslouzil , journal =. 2012 , volume =

    work page 2012

  36. [36]

    Mishima , journal =

    O. Mishima , journal =. 2007 , volume =

    work page 2007

  37. [37]

    Murata and H

    K.-I. Murata and H. Tanaka , journal =. 2012 , volume =

    work page 2012

  38. [38]

    Murata and H

    K.-I. Murata and H. Tanaka , journal =. 2013 , volume =

    work page 2013

  39. [39]

    A. E. Ferdinand and M. E. Fisher , journal =. 1969 , volume =

    work page 1969

  40. [40]

    M. E. Fisher and M. N. Barber , journal =. 1972 , volume =

    work page 1972

  41. [41]

    J. R. Errington and P. G. Debenedetti , journal =. 2001 , volume =

    work page 2001

  42. [42]

    P. J. Steinhardt and D. R. Nelson and M. Ronchetti , journal =. 1983 , volume =

    work page 1983

  43. [43]

    Valdes and F

    L.-C. Valdes and F. Affouard and M. Descamps and J. Habasaki , journal =. 2009 , volume =

    work page 2009

  44. [44]

    Lechner and C

    W. Lechner and C. Dellago , journal =. 2008 , volume =

    work page 2008

  45. [45]

    Mickel and S

    W. Mickel and S. C. Kapfer and G. E. Schr. J. Chem. Phys. , year =

    work page

  46. [46]

    2014 , doi =

    Allolio, Christoph and Klameth, Felix and Vogel, Michael and Sebastiani, Daniel , journal =. 2014 , doi =

    work page 2014

  47. [47]

    Chau and A

    P.-L. Chau and A. J. Hardwick , journal =. 1998 , doi =

    work page 1998

  48. [48]

    E. Dubou. J. Phys. Chem. B , year =

    work page

  49. [49]

    A. B. de Oliveira and P. A. Netz and T. Colla and M. C. Barbosa , journal =. 2006 , volume =

    work page 2006

  50. [50]

    Nilsson and L

    A. Nilsson and L. G. M. Pettersson , journal =. 2015 , volume =

    work page 2015

  51. [51]

    K. T. Wikfeldt and A. Nilsson and L. G. M. Pettersson , journal =. 2011 , volume =

    work page 2011