High-pressure behavior of the low-frequency mode of water

arXiv:2611.17096 · cond-mat.other, cond-mat.mtrl-sci · Submitted 2026-11-12 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Condensed Matter. The day in condensed matter: materials, phases and the measurements under them.

Kai: I'm Kai.

Mira: And I'm Mira, here with Lev.

Kai: We're looking at the paper: "High-pressure behavior of the low-frequency mode of water".

Mira: Here's the gist of the paper. The low-frequency (LF) mode observed in the THz spectrum of liquid water near 10 meV has been known for decades, yet its microscopic origin remains debated.

Kai: The LF mode remains heavily damped, close to critical damping, throughout the explored range, while its frequency hardens markedly with pressure, much more than expected for a transverse acoustic mode.

Mira: Comparison with IXS measurements and ab initio simulations of crystalline Ice VI reveals a cluster of weakly dispersive modes in the same energy range, suggesting that the low-frequency mode in liquid water reflects a mixed character arising from the superposition of several branches rather than a single well-defined excitation.

Kai: Experimental Methodology The study addresses the question of whether the LF mode is a transverse acoustic-like excitation, a transverse optic-like mode, or a mode-mode coupling mechanism by probing the collective dynamics of H2O and D2O at room temperature and pressures from 0.1 to 1 GPa using high-resolution (1.6 meV) inelastic Xray scattering (IXS).

Title and authors: Kai: Let's start with the title and authors of this paper, "High-pressure behavior of the low-frequency mode of water." It sounds very specific, focusing on how pressure alters that particular low frequency mode we’ve been tracking in the THz spectrum for decades.

Mira: Right. And it’s interesting to see who is behind this work; we have Maria Rescigno as one of the authors, and Guillaume Radtke, Umbertoluca Ranieri, Richard Gaal, Luigi Paolasini, Andrea Piovano as others.

Lev: From a quantum hardware perspective, I look at how they set up their measurements; they used a large volume piston cylinder cell to get into the gigapascal range where you can really push those liquid water samples.

Kai: And that’s important because it lets them approach the ice VI crystallization boundary, which is where things get structurally interesting for water.

Mira: That boundary is crucial because it’s where we see how these excitations might transition from the liquid phase behavior to the solid phase behavior they study on a solid sample too.

Lev: What I don't know immediately is how they handle the complexity of fitting those IXS spectra, especially when trying to disentangle different modes.

Kai: That’s where their methodology gets pretty deep, because they use a Bayesian approach where the number of damped harmonic oscillator components, which we call k, is treated as a parameter in their fit.

Mira: So instead of assuming there are two modes and fitting them, they let the data tell them how many components there actually are.

Lev: That’s powerful because it lets you test different numbers of modes against the data itself rather than forcing a pre-defined number onto the system.

Kai: It moves us away from just guessing what's there and towards an AI that determines the structure of the excitation spectrum based on evidence.

The paper's summary: Mira: So, to summarize what they found in "High-pressure behavior of the low-frequency mode of water," they are confirming that this low frequency mode isn't a single thing.

Kai: They found that the IXS spectra are not consistent with an assignment to a single purely acoustic or optic branch across all measured pressures and wave vectors.

Lev: They fitted the data using a Bayesian method, and they found that there’s a sum of two central Lorentzian peaks plus k damped harmonic oscillator functions convolved with the experimental resolution.

Kai: And crucially, this method lets them determine that the number of DHO components, which is k, is actually a parameter of the fit.

Mira: So what this means for us is that they are confirming a superposition of excitations rather than a single well-defined excitation.

Lev: This supports the idea that we’re looking at something more complex than just one simple wave propagating through the liquid.

Kai: And they show this complexity persists even when they compare their liquid results with ab initio simulations of crystalline ice VI, where there is a cluster of weakly dispersive optical branches near ten meV.

Mira: That comparison with the crystal structure really solidifies the idea that these excitations in the liquid map onto that same manifold of excitations found in the solid.

Lev: This suggests that this mixed nature is an intrinsic feature of how water’s tetrahedral hydrogen-bond network organizes itself under pressure, and it survives structural reorganization.

Kai: So, to put it plainly, the paper shows that this low frequency mode is not a single excitation but a superposition of transverse acoustic and transverse optic branches whose mixed nature reflects the topological complexity of the tetrahedral hydrogen bond network.

The paper's improvements: Mira: Now let’s talk about what they suggest as improvements or extensions in their findings from this study on "High-pressure behavior of the low-frequency mode of water."

Kai: One major suggestion is that the AI can be trained to use pressure as a decisive discriminator between competing interpretations by comparing the frequency evolution of modes with changes in the shear modulus.

Lev: So, they propose testing if the low frequency mode’s frequency increases by about fifty five percent between one kbar and seven kbar, while simultaneously checking that the shear modulus of water, which sets a transverse acoustic mode frequency, only changes by less than twenty percent.

Kai: That comparison directly tests whether the hardening behavior we see is consistent with a purely acoustic character or something else entirely.

Mira: They also suggest incorporating isotopic effects into their predictions for collective dynamics by calculating the ratio of the D frequency to the H frequency, which is about equal to that molar mass ratio between H2O and D2O.

Lev: This would allow them to assess whether a mode is associated with the movement of whole molecules rather than just protons alone, which rules out a purely optic nature.

Kai: And another way they suggest using AI is to recognize structural reorganization by matching experimental IXS spectra at high pressure to ab initio simulations of crystalline ice VI.

Mira: This confirms that the excitations observed in liquid water are part of a cluster of weakly dispersing modes in the same energy range when compared to the ice structure.

Lev: This helps ground the abstract findings by linking them back to concrete structural models, which is always important for validating any complex theoretical model.

Conclusion: Kai: So, wrapping up this discussion on "High-pressure behavior of the low-frequency mode of water," the paper strongly supports that this low frequency mode isn't a well defined elementary excitation of purely acoustic or optic character.

Mira: It’s a superposition of transverse acoustic and transverse optic branches whose mixed nature reflects the topological complexity of the tetrahedral hydrogen bond network, which is robust up to the crystallization boundary of ice VI.

Lev: From an engineering standpoint, this means that if you were trying to build error correction for something related to these modes, you’d need to account for this mixed nature because treating it as just one thing wouldn't capture the physics.

Kai: We should keep exploring how these different branches might interact as pressure increases toward one gigapascal and see if that reveals any new physics.

Kai: That would be interesting because it links the spectral features directly to the collective dynamics we’re trying to probe in liquid water.

Maria Rescigno, * Guillaume Radtke, Umbertoluca Ranieri, Richard Gaal, Luigi Paolasini, Andrea Piovano, Ferdinando Formisano, Paola Giura and Livia Eleonora Bove

Institut Laue-Langevin Institute of Minéralogie de Physique des Matériaux et de Cosmochimie Sorbonne University CNRS UMR 7590 MNHN · Centro de Física de Materiales CSIC-UPV/EHU · Laboratory of Quantum Magnetism Institute of Physics Ecole Polytechnique Federale Lausanne European Synchrotron Radiation Facility

cond-mat.other, cond-mat.mtrl-sci

Submitted: 2026-11-12

Updated: 2026-11-12

Importance score: 76/100

The gist: The low-frequency (LF) mode observed in the THz spectrum of liquid water near 10 meV has been known for decades, yet its microscopic origin remains debated.

Key concepts

Low-Frequency (LF) Mode
A specific vibrational mode observed in liquid water around 10 meV in the THz spectrum. Its origin is debated, but it appears to be a complex collective excitation rather than a simple single vibration.
Bayesian Approach with k components
A statistical method used to analyze IXS data where the number of damped harmonic oscillator components (k) is treated as a parameter. The analysis strongly suggested that two modes are present, supporting the idea of a superposition.
Mode Superposition
The finding that the LF mode arises from combining several different vibrational branches (acoustic and optic). This complexity indicates that water's dynamics are governed by the interplay of multiple structural motions rather than one simple type of excitation.

Terminology

Summary

The low-frequency (LF) mode observed in the THz spectrum of liquid water near 10 meV has been known for decades, yet its microscopic origin remains debated. p. 0 The LF mode remains heavily damped, close to critical damping, throughout the explored range, while its frequency hardens markedly with pressure, much more than expected for a transverse acoustic mode. p. 0 Comparison with IXS measurements and ab initio simulations of crystalline Ice VI reveals a cluster of weakly dispersive modes in the same energy range, suggesting that the low-frequency mode in liquid water reflects a mixed character arising from the superposition of several branches rather than a single well-defined excitation. p. 0

Experimental Methodology

The study addresses the question of whether the LF mode is a transverse acoustic-like excitation, a transverse optic-like mode, or a mode-mode coupling mechanism by probing the collective dynamics of H2O and D2O at room temperature and pressures from 0.1 to 1 GPa using high-resolution (1. p. 06 meV) inelastic Xray scattering (IXS). p. 0 The researchers exploited a specifically developed large-volume piston-cylinder cell that overcomes the sample-size limitation of earlier work. p. 0 They performed these measurements on H2O and D2O at room temperature and pressures from 0. p. 01 to 1 GPa, approaching the ice VI crystallization boundary. p. 0

Mode Characterization

The analysis of the IXS spectra utilized a Bayesian approach where the number of DHO components (k) is a parameter of the fit, yielding a posterior probability P(k=2y) > 0.999. Two modes are observed throughout the liquid phase for Q≥4 nm−1: a high-frequency LA mode with sound velocities consistent with prior literature [3, 42, 43], and a weakly dispersive, heavily damped LF mode near 10 meV. p. 3 The LF mode's damping ratio is pressure independent, ΓLF /omegaLF = 2.05 ± 0.12 (plateau value for Q ≥ 10 nm−1), while its frequency increases by 55 ± 12% between 1 and 7 kbar. p. 1

Distinguishing Competing Interpretations

The observed hardening of the LF mode is incompatible with a purely transverse acoustic character because its frequency evolution is inconsistent with the shear modulus of water, which would set the frequency of a transverse acoustic mode. Furthermore, the isotopic frequency ratio omegaD LF /omega H LF ≈ p mH2O/mD2O is consistent with a mode involving the motion of whole molecules rather than protons alone, ruling out a purely optic nature. p. 3 The comparison with IXS measurements and ab initio simulations of ice VI reveals a cluster of weakly dispersive optical branches near 10 meV, which suggests the liquid LF mode maps onto this manifold of overlapping excitations rather than any single branch. p. 3

Conclusion on Mode Nature

Taken together, these results strongly support the conclusion that the LF mode of liquid water is not a well-defined elementary excitation of purely acoustic or optic character, but rather a superposition of transverse acoustic and transverse optic branches whose mixed nature reflects the topological complexity of the tetrahedral hydrogen-bond network. This complexity is robust, surviving structural reorganization up to the crystallization boundary of ice VI. p. 4 The anomalous THz dynamics of water are an intrinsic consequence of the versatility of the hydrogen bond, whose tetrahedral coordination persists under pressure through the interpenetration of open ring motifs rather than their disruption. p. 4

Acknowledgments

The authors acknowledge the European Synchrotron Radiation Facility (ESRF) for provision of synchrotron radiation facilities under proposal number HC3412. They thank the Institut Laue-Langevin (ILL) for provision of beamtime on the IN8 triple axis spectrometer (experiment number DIR-317), and Claude Payre for assistance with the high pressure sample environment. p. 4 L.E.B., F.F. and M.R. acknowledge the financial support by the European Union - NextGenerationEU (PRIN N. p. 4 2022NRBLPT). p. 4 L.E.B and M.R acknowledge the ANR-23-CE30-0034 EXOTIC-ICE. p. 4 L.E.B and R.G acknowledge the financial support by the Swiss National Fund (FNS) under Grant No. p. 4 212889. U.R acknowledges support from the Diputaci´on Foral de Gipuzkoa through Grant 2025-FELL-000013-01. p. 4 This work was granted access to the HPC resources of IDRIS under the allocation AD010910820R3 made by Grand Equipement National de Calcul Intensif (GENCI). p. 4

References

[1] P. G. Debenedetti, Supercooled and glassy water, Journal of Physics: Condensed Matter 15, R1669 (2003). p. 4

[2] L. E. Bove, S. Klotz, T. Str¨assle, M. p. 4 Koza, J. Teixeira, and A. p. 4 M. Saitta, Translational and Rotational Diffusion in Water in the Gigapascal Range, Physical Review Letters 111, 185901 (2013). p. 4

[3] U. Ranieri, P. p. 0 Giura, F. A. Gorelli, M. p. 4 Santoro, S. p. 4 Klotz, P. Gillet, L. Paolasini, M. p. 0 M. Koza, and L E Bove Dynamical crossover in hot dense water: The hydrogen bond role, The Journal of Physical Chemistry B 120, 9051 (2016). p. 4

[4] K. Amann-Winkel, M. p. 4-C. Bellissent-Funel, L E Bove, T Loerting, A Nilsson, A Paciaroni, D Schlesinger, and L Skinner X-ray and Neutron Scattering of Water, Chemical Reviews 116, 7570 (2016). p. 4

[5] M. Rescigno, A. p. 0 Toffano, U Ranieri, L Andriambariarijaona, R Gaal S Klotz M M Koza J Ollivier F Martelli J Russo F Sciortino J Teixeira and L E Bove Observation of plastic ice VII by quasi-elastic neutron scattering, Nature 640, 662 (2025). p. 4

[6] Y.-H. Lee, J. K. Kim, Y.-J. Kim, M Kim, Y C Cho R J Husband C Strohm E Ehrenreich-Petersen K Glazyrin T Laurus H Graafsma R P C Bauer F Lehmk¨uhler K Appel Z Konˆopkov´a M Tang A P Dwivedi J Sztuck-Dambietz L Randolph K Buakor O Humphries C Baehtz T Eklund L Mohrbach A Mondal H Marquardt E F O’Bannon K AmannWinkel C.-S Yoo U Zastrau H. p. 4-P Liermann H Nada and G W Lee Multiple freezing–melting pathways of high-density ice through ice xxi phase at room temperature, Nature Materials 25, 302 (2026). p. 4

[7] G. Ruocco and F. p. 4 Sette The high-frequency dynamics of liquid water, Journal of Physics: Condensed Matter 11, R259 (1999). p. 4

[8] L E Bove, F Sacchetti C Petrillo B Dorner F Formisano and F Barocchi Neutron investigation of the ion dynamics in liquid mercury: Evidence for collective excitations, Phys. p. 4 Rev. Lett. 87, 215504 (2001). p. 4

[9] L E Bove, B Dorner C Petrillo F Sacchetti and J-B Suck Neutron scattering investigation of low-momentum collective ion dynamics in liquid potassium, Phys. p. 4 Rev. B 68, 024208 (2003). p. 4

[10] F H Stillinger and A Rahman Improved simulation of liquid water by molecular dynamics, The Journal of Chemical Physics 60, 1545 (1974). p. 4

[11] P Bosi, F Dupre F Menzinger F Sacchetti and M Spinelli Observation of collective excitations in heavy water in the 108 cm−1 momentum range, Lettere al Nuovo Cimento (1971-1985) 21, 436 (1978). p. 4

[12] J Teixeira, M Bellissent-Funel S Chen and B Dorner Observation of new short-wavelength collective excitations in heavy water by coherent inelastic neutron scattering, Physical Review Letters 54, 2681 (1985). p. 4

[13] G Ruocco F Sette U Bergmann M Krisch C Masciovecchio V Mazzacurati G Signorelli and R Verbeni Equivalence of the sound velocity in water and ice at mesoscopic wavelengths, Nature 379, 521 (1996). p. 4

[14] H E Smorenburg R M Crevecoeur and I M de Schepper Fast sound in a dense helium argon gas mixture, Physics Letters A 211, 118 (1996). p. 4

[15] P H K de Jong P Verkerk C F de Vroege L A de Graaf W S Howells and S M Bennington Observation of fast sound in liquid Li4Tl and liquid Li4Pb by inelastic neutron scattering, Journal of Physics: Condensed Matter 6, L681 (1994). p. 4

[16] F Sette G Ruocco M Krisch C Masciovecchio R Verbeni and U Bergmann Transition from Normal to Fast sound in liquid water, Phys. p. 4 Rev. Lett. 77, 83 (1996). p. 4

[17] F Sciortino and S Sastry Sound propagation in liquid water: The puzzle continues, The Journal of Chemical Physics 100, 3881 (1994). p. 4

[18] A Cunsolo and M Nardone Velocity dispersion and viscous relaxation in supercooled water, The Journal of Chemical Physics 105, 3911 (1996). p. 4

[19] F Bencivenga A Cimatoribus A Gessini M G Izzo and C Masciovecchio Temperature and density dependence of the structural relaxation time in water by inelastic ultraviolet scattering, The Journal of Chemical Physics 131, 144502 (2009). p. 4

[20] S C Santucci D Fioretto L Comez A Gessini and C Masciovecchio Is there any fast sound in water? p. 5 Phys. Rev. Lett. 97, 225701 (2006). p. 5

[21] S Santucci S Di Fonzo and C Masciovecchio Infinite frequency sound velocity in liquid water by inelastic UV scattering, Journal of Electron Spectroscopy and Related Phenomena 144-147, 941 (2005). p. 5

[22] C Petrillo F Sacchetti B Dorner and J-B Suck High-resolution neutron scattering measurement of the dynamic structure factor of heavy water, Phys. p. 5 Rev. E 62, 3611 (2000). p. 5

[23] F Sacchetti J-B Suck C Petrillo and B Dorner Brillouin neutron scattering in heavy water: Evidence for two-mode collective dynamics, Phys. p. 5 Rev. E 69, 061203 (2004). p. 5

[24] A Cunsolo C N Kodituwakku F Bencivenga M Frontzek B M Leu and A H Said Transverse dynamics of water across the melting point: A parallel neutron and x-ray inelastic scattering study, Phys. p. 5 Rev. B 85, 174305 (2012). p. 5

[25] J Fink and A Cunsolo The thz spectrum of density fluctuations of water: The viscoelastic regime, Advances in Condensed Matter Physics 2015, 137435 (2015). p. 5

[26] U Balucani J P Brodholt and R Vallauri Dynamical properties of liquid water, Journal of Physics: Condensed Matter 8, 9269 (1996). p. 5

[27] A Cimatoribus S Saccani F Bencivenga A Gessini M G Izzo and C Masciovecchio The mixed longitudinal–transverse nature of collective modes in water, New Journal of Physics 12, 053008 (2010). p. 5

[28] M.-C Bellissent-Funel and J Teixeira Dynamics of water studied by coherent and incoherent inelastic neutron scattering, Journal of Molecular Structure 250, 213 (1991). p. 5

[29] G E Walrafen Raman spectral studies of the effects of temperature on water and electrolyte solutions, The Journal of Chemical Physics 44, 1546 (1966). p. 5

[30] M M Koza H Schober B Geil M Lorenzen and H Requardt Crystalline inelastic response of highdensity amorphous ice, Phys. p. 5 Rev. B 69, 024204 (2004). p. 5

[31] M M Koza H Schober S F Parker and J Peters Vibrational dynamics and phonon dispersion of polycrystalline ice xii and of high-density amorphous ice, Phys. p. 5 Rev. B 77, 104306 (2008). p. 5

[32] M M Koza B Geil M Scheuermann H Schober G Monaco and H Requardt Vibrational dynamics of very high density amorphous ice studied by highresolution x-ray spectroscopy, Phys. p. 5 Rev. B 78, 224301 (2008). p. 5

[33] H Schober M M Koza A T¨olle C Masciovecchio F Sette and F Fujara Crystal-like high frequency phonons in the amorphous phases of solid water, Phys. p. 5 Rev. Lett. 85, 4100 (2000). p. 5

[34] T Bryk and I Mryglod Longitudinal optical-like excitations in binary liquid mixtures, Journal of Physics: Condensed Matter 14, L445 (2002). p. 5

[35] L E Bove F Formisano E Guarini A Ivanov C Petrillo and F Sacchetti Evidence for the coexistence of two density fluctuation modes in molten li30bi70 as probed by neutron scattering, Europhysics Letters 79, 16002 (2007). p. 5

[36] J-F Wax M R Johnson L E Bove and M Mihalkoviˇc Multiscale study of the influence of chemical order on the properties of liquid Li-Bi alloys, Phys. p. 5 Rev. B 83, 144203 (2011). p. 5

[37] E Pontecorvo M Krisch A Cunsolo G Monaco A Mermet R Verbeni F Sette and G Ruocco Highfrequency longitudinal and transverse dynamics in water, Phys. p. 5 Rev. E 71, 011501 (2005). p. 5

[38] M Sampoli G Ruocco and F Sette Mixing of longitudinal and transverse dynamics in liquid water, Phys. p. 5 Rev. Lett. 79, 1678 (1997). p. 5

[39] L E Bove E Fabiani A Fontana F Paoletti C Petrillo O Pilla I C V Bento Brillouin neutron scattering of v-GeO2, Europhysics Letters 71, 563 (2005). p. 5

[40] O Pilla S Caponi A Fontana J R Gon¸calves M Montagna F Rossi G Viliani L Angelani G Ruocco G Monaco and F Sette The low energy excess of vibrational states in v-SiO2: the role of transverse dynamics, Journal of Physics: Condensed Matter 16, 8519 (2004). p. 5

[41] G Monaco A Cunsolo G Ruocco and F Sette Viscoelastic behavior of water in the terahertz-frequency range: An inelastic x-ray scattering study, Phys. p. 5 Rev. E 60, 5505 (1999). p. 5

[42] M Krisch P Loubeyre G Ruocco F Sette A Cunsolo M D’Astuto R LeToullec M Lorenzen A Mermet G Monaco and R Verbeni Pressure evolution of the high-frequency sound velocity in liquid water, Phys. p. 5 Rev. Lett. 89, 125502 (2002). p. 5

[43] M Santoro F A Gorelli T Scopigno M Krisch F Sette and G Ruocco Pressure behavior of the sound velocity of liquid water at room temperature in the terahertz regime, Phys. p. 5 Rev. B 84, 092301 (2011) <ref:2611. p. 5

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved AI systems can distinguish between a single, well-defined excitation and a superposition of multiple branches by analyzing IXS spectra using a Bayesian approach that allows for an unknown number of damped harmonic oscillator components (k). This enables the system to determine if the low-frequency mode is a single longitudinal acoustic mode or a superposition of transverse acoustic and transverse optic branches.

  2. The AI can be trained to use pressure as a decisive discriminator between competing interpretations by comparing the frequency evolution of modes with shear modulus changes. Specifically, it can test the hypothesis that the LF frequency increases by about 55% between 1 and 7 kbar, whereas the shear modulus of water, which would set the frequency of a transverse acoustic mode, changes by less than 20%.

  3. The improved system can incorporate isotopic effects into its predictions for collective dynamics. By calculating the ratio omegaD LF /omega H LF ≈ p mH2O/mD2O, the AI can assess whether a mode is associated with the movement of whole molecules rather than protons alone, ruling out a purely optic nature.

  4. The AI can recognize structural reorganization through comparison with crystalline simulations. It can match experimental IXS spectra at high pressure to ab initio simulations of crystalline Ice VI to confirm that the observed excitations are part of a cluster of weakly dispersing modes in the same energy range.

  5. The system can rule out specific theoretical models based on spectral features. For instance, it can determine that the mode-mode interaction interpretation "does not satisfactorily describe the low Q (Q<4 nm−1) behavior of liquid D2O as no indication for a low Q flattening of the high frequency mode is observed."

Related papers