liquid experiment tension

Hi, what are you looking for?

Metal Injection

• New Albums Out This Week
• Top Tracks Of The Week
• Top Stories Of The Week

Upcoming Releases

• Funny Stories
• Photo Gallery
• Latest Videos Listing

Music Videos

• Metal Injection Exclusives
• Live Footage
• Funny Videos
• Slay At Home
• ASK THE ARTIST
• Kids In Metal
• Mashups & Covers
• Rockumentary
• View All Channels
• Apple Playlists
• Spotify Playlists
• Los Angeles

Album Review: LIQUID TENSION EXPERIMENT Liquid Tension Experiment 3

liquid tension experiment

Sponsored links from around the internet.

You May Also Like

10 Records We Waited A LONG Time For

Was it worth the wait?

LIQUID TENSION EXPERIMENT Plans On Making More Music Together

"...although there's no date right now when it'll happen."

Weekly Injection

The weekly injection: new releases from cannibal corpse, liquid tension experiment, and more out today – 4/16.

Plus releases from The Armed, Amigo The Devil, Greta Van Fleet, and August Burns Red.

LIQUID TENSION EXPERIMENT Streams Shreddy New Song "Hypersonic"

We wait over 20 years and it was worth it.

Bummer Alert

Saint vitus has closed for good.

DEZ FAFARA Suffers Health Emergency, COAL CHAMBER Postpones Headlining Tour To 2025

Latest News

Avenged sevenfold's johnny christ on gojira playing the olympics: "i was in the other room and i heard metal being played, i was, like, 'what the f*ck'".

PIG DESTROYER & THE RED CHORD Announce More Shows Together

MARILYN MANSON Streams Second New Single "Raise The Red Flag"

Metal Merch

Type o negative's peter steele gets his own action figure, bulge may be inaccurate.

Breakups & Shakeups

Galder quits dimmu borgir after nearly 25 years, revives old man's child.

• PROG ROCK GUIDES
• PROG RADIOS
• PROG SUB-GENRES:
• Canterbury Scene
• Crossover Prog
• Eclectic Prog
• Experimental/Post Metal
• Indo-Prog/Raga Rock
• Jazz Rock/Fusion
• Post Rock/Math Rock
• Progressive Electronic
• Progressive Metal
• Psychedelic/Space Rock
• RIO/Avant-Prog
• Rock Progressivo Italiano
• Symphonic Prog
• Tech/Extreme Prog Metal
• Various Genres/Artists
• Prog Related
• VA: SAMPLERS
• VA: CONCEPT ALBUMS
• VA: TRIBUTES
• Top Prog Albums (all-time)
• Top 2024 Albums
• Top 2023 Albums
• Gigs Reviews
• Active Topics
• Review , Video
• PROG SHOPPING:

LIQUID TENSION EXPERIMENT

Progressive metal • united states, from progarchives.com, the ultimate progressive rock music website.

• Tour & shows updates
• Press & news updates
• Forum discussions
• Videos (Youtube)
• Discography
• Official website

LIQUID TENSION EXPERIMENT Videos (YouTube and more)

Buy liquid tension experiment music.

• DOUG LARSON IMPORTS — Buy prog rock music and rarities (Free shipping on orders over 10 CD's)

LIQUID TENSION EXPERIMENT discography

Liquid tension experiment top albums (cd, lp, mc, sacd, dvd-a, digital media download).

LIQUID TENSION EXPERIMENT Live Albums (CD, LP, MC, SACD, DVD-A, Digital Media Download)

LIQUID TENSION EXPERIMENT Videos (DVD, Blu-ray, VHS etc)

LIQUID TENSION EXPERIMENT Boxset & Compilations (CD, LP, MC, SACD, DVD-A, Digital Media Download)

LIQUID TENSION EXPERIMENT Official Singles, EPs, Fan Club & Promo (CD, EP/LP, MC, Digital Media Download)

LIQUID TENSION EXPERIMENT Reviews

Review by Warthur Prog Reviewer

Review by Dapper~Blueberries Prog Reviewer

Review by kev rowland Special Collaborator Prog Reviewer / Special Collaborator

Review by Wicket Prog Reviewer

Review by The Crow Prog Reviewer

Review by alainPP

Review by DiversionConVinilos

Review by thesameoldfears

Review by Grumpyprogfan

Review by luisM.

Other sites in the MAC network: JazzMusicArchives.com — jazz music reviews and archives | MetalMusicArchives.com — metal music reviews and archives

Liquid Tension Experiment - LTE3 Review

The gang’s back together: mike portnoy, john petrucci, jordan rudess and tony levin reunite the dream team for lte’s first new album of instrumental prog genius in 22 years..

You can trust Louder Our experienced team has worked for some of the biggest brands in music. From testing headphones to reviewing albums, our experts aim to create reviews you can trust. Find out more about how we review.

The rebirth of all-instrumental prog rock supergroup Liquid Tension Experiment feels like an unexpected visit from an old friend – and it comes as a welcome surprise. Last seen on a 10th anniversary tour in 2008, much has happened in the interim years, chiefly founder Mike Portnoy ’s exit from his main band Dream Theater in 2010 (something that made the prospect of a reunion with fellow DT members John Petrucci and Jordan Rudess under the LTE banner unlikely).

Liquid Tension Experiment’s history is inextricably entwined with that of the prog metal giants. Formed in 1997 by Portnoy, the two albums LTE released at the end of that decade, LTE1 and LTE2 , featured Petrucci and Rudess alongside bassist and Chapman Stick maestro Tony Levin. The chemistry generated by the quartet during the recording of the second album in 1999 prompted Rudess to finally accept Portnoy and Petrucci’s overtures to join Dream Theater, despite the keyboard player having previously rebuffed them.

Assorted permutations followed the initial incarnation: Liquid Trio Experiment (Portnoy, Rudess and Levin, minus Petrucci) and Liquid Trio Experiment 2 (Portnoy, Petrucci and Levin, mostly minus Rudess), as well as that anniversary tour from the original line-up. But then Portnoy’s split with his Dream Theater bandmates seemingly put an end to it all.

Until now. The aftershocks of his departure have finally settled. Even before Portnoy added drums to Petrucci’s 2020 solo album, Terminal Velocity , all four LTE members had expressed a wish to renew their collaboration. But it took Covid and a wiping-out of everybody’s schedules to get them together in a studio again – socially distanced, naturally – for the first time in more than two decades.

And so here we are, with LTE3 , an album that sticks to the blueprint of its predecessors with four fully composed tracks, a pair of duets and an on-the-fly jam, plus a retooled cover version.

Like all great prizefighters, LTE land a knockout punch in the first round and never look back. Opening number Hypersonic delivers a similar sensation to a deep glug of super-strong coffee. Its core ingredients – Petrucci’s fluent skylark melodies, Rudess’ classical keyboard embellishments, Levin’s low Chapman Stick frequencies and Portnoy’s everything-including-the-kitchen-sink drumming – are separable from each other yet all part of one gigantic, multitempoed, earthquake-inducing whole.

The ‘Wow!’ that concludes Hypersonic is the sole human voice throughout, but don’t let that serve as an obstacle. These songs are daubed in melodies handed down by the instrumental players. Beating The Odds sees Petrucci and Rudess run riot through painting a triumphant tidal wave of colour and notes.

Bass players tend to sit in the background and follow the drummer. Not Levin, who adds a supple, fleet-fingered spine to the jazz-flavoured Liquid Evolution , and also excels with Portnoy via the brooding, stripped-down duet Chris & Kevin’s Amazing Odyssey (a continuing in-joke that extends back to the first album when a photographer got the pair’s names wrong). It’s downright insidious.

The band’s avant-garde side is further revealed by their reinterpretation of George Gershwin’s Rhapsody In Blue . Performed on past tours, here they tear it apart over the course of 13 minutes, throwing away the instruction manual and proceeding to rebuild the parts with the carefree disrespect of such past progressive giants as The Nice and ELP.

The sentimental Shades Of Hope provides Petrucci with a showcase of his own, another reminder that he can play with real passion and finesse (incredibly, this song was captured in a single take). Those same qualities are further rammed home as the guitarist and Rudess barrel extravagantly through another 13-minute epic, Key To The Imagination .

Elsewhere, LTE go up through the gears together on the band’s first ever entirely group-written track, Passage Of Time , Rudess dominating the song’s latter stages with a section of true orchestral-flavoured beauty before everyone combines to drive the song past the finish line.

The artistic success of LTE3 will doubtless prompt Dream Theater fans to wonder if a reunion with the band’s original drummer is on the cards. Speaking to this correspondent recently, Petrucci headed the idea off at the pass. “Please don’t speculate beyond that. I feel spoiled because I get to play with both Mikes,” referring to Portnoy and current DT sticksman Mike Mangini. And in truth, this isn’t an album to encourage ‘what ifs’. Rather, LTE3 is a deeply satisfying listening experience that should be filed in an extremely small folder that’s marked: ‘Positive outcomes of the coronavirus.’

Buy from Amazon .

Dave Ling was a co-founder of Classic Rock magazine. His words have appeared in a variety of music publications, including RAW, Kerrang!, Metal Hammer, Prog, Rock Candy, Fireworks and Sounds. Dave’s life was shaped in 1974 through the purchase of a copy of Sweet’s album ‘Sweet Fanny Adams’, along with early gig experiences from Status Quo, Rush, Iron Maiden, AC/DC, Yes and Queen. As a lifelong season ticket holder of Crystal Palace FC, he is completely incapable of uttering the word ‘Br***ton’.

New proggy sounds from Nightwish, i Häxa, Souls Of Ambience and more in Prog's Tracks Of The Week

Tiger Moth Tales preview brand new single Return To Chigwick

"It was an unfortunate incident which has been exaggerated and turned into folklore": How contractual differences between Aerosmith and Metallica at Woodstock 1994 came to blows

Most Popular

Liquid Tension Experiment

Occasional prog rock supergroup comprising drummer Mike Portnoy, keyboardist Jordan Rudess, guitarist John Petrucci, and bassist Tony Levin.

Read Full Biography

STREAM OR BUY:

Group Members

Sign Up or Log In to Set Your Streaming Service

Related Artists

Discography, moods and themes.

• listening party
• existing artist
• See all results

No matching results

Try a different filter or a new search keyword.

Search all Bandcamp artists, tracks, and albums

• artists PRO view site
• edit profile
• subscription subscription
• view collection
• showLinkedBands(!showLinkedBands())" data-test="linked-accounts-header">

Spontaneous Combustion

By liquid trio experiment.

Digital Album Streaming + Download

Buy digital album   $10 usd  or more, send as gift  , full digital discography.

Get all 3 Liquid Tension Experiment releases available on Bandcamp and save 10% .

Buy Digital Discography   $27 USD or more ( 10% OFF )

Share / embed.

Shopping cart

Liquid Tension Experiment New York, New York

• magnacarta.net

discography

contact / help

Contact Liquid Tension Experiment

Streaming and Download help

Redeem code

Report this album or account

If you like Spontaneous Combustion, you may also like:

Handmade Cities by Plini

supported by 32 fans who also own “Spontaneous Combustion”

This is the ultimate compilation, of going on a nature walk, with your lover, to literally making massive constructs, with progressive funk mekazael

Sunhead by Plini

supported by 29 fans who also own “Spontaneous Combustion”

so uplifting, like a perfect smile. mekazael

Impulse Voices by Plini

supported by 28 fans who also own “Spontaneous Combustion”

One of my all time favourite albums. nathan_dudeman

Blood Maker by Wild Throne

featured on Bandcamp Radio Nov 25, 2014

Blue Record (Deluxe Version) by Baroness

featured on Bandcamp Radio Sep 15, 2023

Red Album by Baroness

Sweet Nothings by Plini

Plini can do no wrong. Plini has done no wrong. 8/10 RJ

Bandcamp Daily    your guide to the world of Bandcamp

Where to Begin With Ripple Music’s Heavy, Freaky Fuzz Rock

Ten German Hard Rock Bands Bridging Present and Past

The Places That Inspired Planes Mistaken For Stars’ First Album in a Decade

On Bandcamp Radio

Lord Kaiju from Oxygen Destroyer chats death/thrash and giant monster movies.

Liquid Tension Experiment

Aliases: Liquid Trio Experiment .

• Best albums
• Your feedback

The best album credited to Liquid Tension Experiment is Liquid Tension Experiment which is ranked number 9,743 in the overall greatest album chart with a total rank score of 122.

Liquid Tension Experiment is ranked number 3,773 in the overall artist rankings with a total rank score of 244.

• Members: N.B. The member list may not contain a full list of members (and there may be notable omissions). If a member has never released a solo album, or their solo albums do not appear on BEA, they will not be able to be listed as a member. These omissions are not meant to cause offense to anybody. The member list is solely meant as an additional navigation aid to browse between the items listed on the site.
• John Petrucci
• Jordan Rudess
• Steve Gorn / Tony Levin / Jerry Marotta
• ProjeKct Four
• Spin 1ne 2wo
• King Crimson
• John Petrucci & Jordan Rudess
• Bruford Levin Upper Extremities
• Tony Levin / David Torn / Alan White
• Bozzio Levin Stevens
• Joe Satriani / Steve Vai / John Petrucci

Upcoming concerts

Listen to Liquid Tension Experiment on YouTube

Liquid tension experiment best albums.

The following albums by Liquid Tension Experiment are ranked highest in the greatest album charts:

This may not be a complete discography for Liquid Tension Experiment.This listing only shows those albums by this artist that appear in at least one chart on this site. If an album is 'missing' that you think deserves to be here, you can include it in your own chart from the My Charts page!

Liquid Tension Experiment bestography composition

Liquid Tension Experiment best tracks

Include tracks from compilations & live albums | Exclude tracks from compilations & live albums The same track can appear on multiple albums, so excluding tracks from compilations and live albums helps to remove duplicates from this list.

• Related links:
• Top tracks of all time

Liquid Tension Experiment ratings

Showing latest 5 ratings for this artist. | Show all 9 ratings for this artist.

Please log in or register if you want to be able to leave a rating

Liquid Tension Experiment favourites

Showing all 3 members who have added this artist as a favourite

Liquid Tension Experiment comments

Showing all 3 comments | Most Helpful First | Newest First | Maximum Rated First | Longest Comments First (Only showing comments with -2 votes or higher. You can alter this threshold from your profile page. Manage Profile )

They make me fly away

Virtuoso musicianship of the finest caliber.

These musicians blew me away; they were one of my first listens into prog music in general. I became hooked.

Please log in or register if you want to be able to add a comment

Your feedback for Liquid Tension Experiment

• Advanced Search

.css-az2b2f{-webkit-text-decoration:underline;text-decoration:underline;text-decoration-color:rgba(41, 117, 133, 0.4);-webkit-text-decoration:none;text-decoration:none;color:#2653D9;}.css-az2b2f:hover{text-decoration-color:inherit;}.css-az2b2f:visited{color:#98158B;}.css-az2b2f:hover{color:#07b;}.css-az2b2f:active{color:#07b;} .css-1yze6f2{margin:0;font-family:inherit;font-weight:inherit;font-size:inherit;line-height:inherit;letter-spacing:inherit;color:#297585;-webkit-text-decoration:underline;text-decoration:underline;text-decoration-color:rgba(41, 117, 133, 0.4);-webkit-text-decoration:none;text-decoration:none;color:#2653D9;}.css-1yze6f2:hover{text-decoration-color:inherit;}.css-1yze6f2:visited{color:#98158B;}.css-1yze6f2:hover{color:#07b;}.css-1yze6f2:active{color:#07b;} Liquid Tension Experiment – Liquid Tension Experiment

More images

Companies, etc.

• Engineered At – Millbrook Sound Studios
• Mixed At – Avatar Studios
• Mastered At – Absolute Audio
• Phonographic Copyright ℗ – Magna Carta
• Copyright © – Magna Carta
• Glass Mastered At – CMCA – R80114-13

Barcode and Other Identifiers

• Barcode (Text) : 0 26245-9023-2 9
• Barcode (Scanned) : 026245902329
• Matrix / Runout (Variant 1 to 9, mirrored) : L808 4174 MA90232 R80114-13 A @
• Mastering SID Code (Variant 1 to 9 mirrored) : L808
• Mould SID Code (Variant 1) : IFPI 2F89
• Mould SID Code (Variant 2) : IFPI 2F82
• Mould SID Code (Variant 3) : IFPI 2F81
• Mould SID Code (Variant 4) : IFPI 2F87
• Mould SID Code (Variant 5) : IFPI 2F62
• Mould SID Code (Variant 6) : none
• Mould SID Code (Variant 7) : IFPI 2F80
• Mould SID Code (Variant 8) : IFPI 2F88
• Mould SID Code (Variant 9) : IFPI 2F86
• Pressing Plant ID (Variant 6, etched into inner hub) : CMCA

Other Versions (5 of 28)

Recommendations

For sale on Discogs

• Have : 1026
• Avg Rating : 4.26 / 5
• Ratings : 115
• Last Sold : Aug 1, 2024
• Low : $6.00
• Median : $11.98
• High : $40.00

• Owned Pre-Marriage by robruff

Contributors

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 13 August 2024

On the temperature dependence of the density of states of liquids at low energies

• Sha Jin 1 , 2 , 3   na1 ,
• Xue Fan 4 , 5 , 6   na1 ,
• Caleb Stamper 7 , 8 ,
• Richard A. Mole 8 ,
• Yuanxi Yu 1 ,
• Liang Hong 1 , 4 , 9 ,
• Dehong Yu 8 &
• Matteo Baggioli 1 , 2 , 3  

Scientific Reports volume  14 , Article number:  18805 ( 2024 ) Cite this article

164 Accesses

1 Altmetric

Metrics details

• Materials science

We report neutron-scattering measurements of the density of states (DOS) of water and liquid Fomblin in a wide range of temperatures. In the liquid phase, we confirm the presence of a universal low-energy linear scaling of the experimental DOS as a function of the frequency, \(g(\omega )= a(T) \omega \) , which persists at all temperatures. The low-frequency scaling of the DOS exhibits a sharp jump at the melting point of water, below which the standard Debye’s law, \(g(\omega ) \propto \omega ^2\) , is recovered. On the contrary, in Fomblin, we observe a continuous transition between the two exponents reflecting its glassy dynamics, which is confirmed by structure measurements. More importantly, in both systems, we find that the slope a ( T ) grows with temperature following an exponential Arrhenius-like form, \(a(T) \propto \exp (-\langle E \rangle /T)\) . We confirm this experimental trend using molecular dynamics simulations and show that the prediction of instantaneous normal mode (INM) theory for a ( T ) is in qualitative agreement with the experimental data.

Similar content being viewed by others

Manifestations of metastable criticality in the long-range structure of model water glasses

Curie-Weiss behavior of liquid structure and ideal glass state

Modern computational studies of the glass transition

Introduction.

The density of states (DOS) is a fundamental concept in solid state physics, and plays a key role in determining vibrational, thermodynamic and transport properties of a given material ( e.g. , heat capacity, thermal conductivity, superconductivity, etc.). In crystalline solids with long-range order, the definition and determination of the DOS can be achieved using a normal mode analysis, which ultimately leads to the construction of the Debye model 1 . The Debye model predicts that the DOS of 3D crystalline solids, in the low frequency regime, displays a universal quadratic scaling known as Debye’s law,

with \(v_{T,L}\) respectively the speed of propagation of transverse and longitudinal phonons. The Debye model has been very successful in describing the dynamics of crystalline solids 1 .

Despite a lot of effort having been dedicated to investigate and explain deviations from Debye’s law in the DOS of amorphous systems 2 , much less is known for the case of classical liquids. In particular, the properties of the DOS of liquids at low energies, in the range where Debye’s law holds in crystalline solids, remains unclear.

The experimental investigation of the density of states of liquids at low energies is much less explored due to the fact that many techniques used to study liquid properties, such as infrared, Raman and nonlinear IR, do not have sufficient energy resolution to probe the details at very low energies. To the best of our knowledge, the first experimental investigation in this direction was presented by Phillips et al. in 1989 3 . A comparison of the low-frequency experimental density of states on an absolute scale for glassy, liquid, and polycrystalline selenium suggests that the DOS of liquids does not obey Debye’s law. Eleven years later, Dawidowski et al. 4 performed an experimental study of the density of states of heavy water by comparing two temperatures slightly above and slightly below the freezing transition. Once again, their experimental data indicate the disappearance of Debye scaling in the liquid phase.

Along with the development of modern cold neutron spectrometers and intense inelastic X-ray spectrometers based on synchrotron radiation facilities, this field has become more active. Recent experimental results of the density of states, using inelastic neutron scattering (INS), in several liquid systems including water, liquid metal and polymer liquids have been reported in Ref. 5 (see also Ref. 6 and Ref. 7 for previous studies using neutrons). These more recent results confirmed the existence of a universal linear in frequency law in the DOS of liquids, \(g(\omega )\propto \omega \) , compatible with the experimental data of 3 , 4 .

To some degree, there is experimental evidence that the low-energy density of states of liquids follows a universal linear scaling in frequency of the form:

where a ( T ) is an unkown function of the temperature T which we will refer to in the rest of this manuscript as the “slope”. Nevertheless,

It is not clear how universal the form in Eq. ( 2 ) is, in which range of energies it emerges, and if it holds at any temperature;

It is not known how this linear scaling transitions into the more standard Debye law, Eq. ( 1 ), by approaching the freezing transition;

It remains unclear how the slope a ( T ) depends on temperature and whether the resulting functional dependence could be rationalized with any of the existing theoretical paradigms.

The main objective of this work is to answer these three questions.

In order to achieve this task, we perform a full scan of the experimental DOS of water and Fomblin oil in a wide range of temperatures, below and above the melting transition, and with particular emphasis in the liquid phase. Additionally, we combine molecular dynamics simulations with theoretical methods to compare their predictions with our experimental results.

As a theoretical method, we use the concept of instantaneous normal modes (INMs) that emerges within a solid-like description of liquid dynamics. The difficulties of using an harmonic interpretation of liquids are evident (see more details about this point below); nevertheless, this harmonic picture has achieved reasonable success in various directions as it seems to capture certain aspects of the physics of liquids. Some concrete examples of this success are the reconstruction of the self-diffusion constant of liquids in terms of “normal modes” 8 , the prediction of gapped collective shear waves in classical liquids 9 , 10 and the description of the heat capacity of liquids 11 , 12 , 13 . For more details on what INMs are and are not, and their applications, we refer the Readers to Refs. 14 , 15 . In our work, we will limit ourselves to describe in detail the predictions of INMs for the DOS of liquids and to compare them to the experimental data. In other words, we will provide a direct experimental test of what INMs can tell us about liquid dynamics. We will see that important aspects of the experimental DOS, for which to the best of our knowledge no alternative theoretical description exists, can indeed be rationalized using INMs.

Experimental results

We performed INS measurements on two liquids with drastically different elemental compositions, molecular weights, and, therefore, dynamic properties. Water, “the universal solvent”, is undoubtedly the most well-studied chemical system and so makes a great reference. Nonetheless, water is a scientifically interesting system with complex dynamics that lead to unique chemical properties. To compare with water, we chose to measure Fomblin, a perfluorinated polyether fluid most commonly used as a lubricant for vacuum pumps 16 , but also technologically relevant for cold neutron storage 17 . Furthermore, Fomblin is also identified as a low-background pressure-medium for high-pressure neutron scattering experiments 18 . In this work, we measured Fomblin-Y, 25/6, with the chemical formula CF \(_3\) O[–CF(CF \(_3\) )CF \(_2\) O–] \(_x\) (–CF \(_2\) O–) \(_y\) CF \(_3\) where x/y gives an average molecular weight of 3300. The thermodynamic properties of both systems are technologically important and distinct from one another. A summary of some chemical properties relevant to their dynamics is given in Table 1 .

In Fig.  1 A, we present the experimental DOS for water as a function of temperature from 250 to 360 K measured with inelastic neutron scattering (see method for DOS derivation). Two major peaks are observed in the DOS spectrum. The first peak appears around \(\approx 6.5\) meV and corresponds to the hydrogen-bond bending, perpendicular to the hydrogen-bond (O–H–O) 22 . The large peak at around 65 meV is attributed to librational motion due to intermolecular coupling 23 . In the region between 20 and 35 meV, the DOS is characterized by a flat band which represents the weak hydrogen-bond stretching modes in line with the hydrogen-bond 24 .

The experimental density of states (DOS), measured by INS for different temperatures for ( A ) water ( B ) and Fomblin oil. The DOS curves have been normalized by the total area. The insets zoom on the low frequency region below 15 meV.

As shown in the inset of Fig.  1 A, the DOS exhibits a universal linear behavior at low frequencies in the liquid phase, with temperature above 273 K. This observation confirms the presence of a low-frequency linear scaling regime, \(g(\omega )\propto \omega \) , in the whole liquid phase, independently of the temperature, and it expands the results of Ref. 5 . Let us notice that this linear behavior is not expected to extend all the way down to zero frequency, since the zero frequency value of the DOS is finite due to the diffusive processes active in the liquid phase 25 .

Below the melting temperature, in the solid phase, the low-energy DOS is drastically modified. First, the zero frequency value disappears, as a confirmation that the self-diffusion constant vanishes in the solid phase. Second, the low-frequency scaling is suddenly modified to recover the standard quadratic Debye law, \(g(\omega ) \propto \omega ^2\) . For temperatures below 273 K, the system is in the solid ice phase and the linewidth of the peak corresponding to the hydrogen-bond bending is significantly smaller. Moreover, the amplitude of the two major peaks is larger than in the liquid phase. As shown in the inset of Fig.  1 A, upon the phase transition from solid to liquid, the well-defined peak around 6.5 meV becomes significantly broadened with much high-intensity signal shifted to low energies. Around 330 K, in the liquid phase, the lowest peak \(\approx 6.5\) meV becomes completely overdamped and the DOS becomes flat up to the next broadened and red-shifted peak at \(\approx 65\) meV. This tells us that the dynamics become very unstable with a wide distribution in frequency.

In Fig.  1 B, we show the DOS measured by INS for the Fomblin oil. The lowest band in the DOS has been suggested to correspond to the low-frequency modes caused by the torsion of fluoromethyl (O–CF \(_3\) ) groups at the end of the chain, and C–C torsions of the chain. In the range of 25–45 meV, the deformational in-plane modes ( \(\delta \) [O–C–C], \(\delta \) [O–C–F] and \(\delta \) [C–O–C]) represent the dominant contribution to the DOS spectra 17 . As shown in the inset of Fig.  1 B, and analyzed in more detail in Fig.  2 B later, the transition between a low-frequency quadratic scaling and a linear law in Fomblin is more gradual, revealing the absence of a proper first-order melting transition. Nonetheless, also for Fomblin the DOS is linear in the low energy region for all temperatures above 260 K.

A zoom of the experimental data in the range of energies of interest is provided in panels A and B of Fig. 2 , and confirms the marked difference between the linear scaling in the liquid state and the quadratic Debye law in the solid phase. Moreover, in the Supplementary Information (SI), we provide another representation of the experimental DOS reduced by the frequency \(\omega \) to make these two scalings even more evident.

The low-energy experimental DOS data for ( A ) water and ( B ) Fomblin fit using the function given in Eq. ( 3 ). More details regarding this fit and other tests to confirm the extracted power-law are presented in the Supplementary Information . The power-law b ( T ) of the low-frequency experimental DOS as a function of the temperature for ( C ) liquid water and ( D ) Fomblin oil. The horizontal gray dashed lines indicate the linear power, \(b(T)=1\) . The dashed red line is the result of a fit to a function \(b(T) \propto (T^* -T)^{0.648}\) , with \(T^*=260\) K.

The experimental data for water and Fomblin presented in Fig.  1 are taken at different temperatures, and go beyond the solid-liquid phase transition for the two systems. In order to understand how the universal linear scaling characteristic of the liquid phase is modified as a function of temperature entering into the solid phase, we fit the low-frequency regime of the experimental DOS with the following expression:

The parameter c ( T ) relates to the self-diffusion constant in the liquid phase and vanishes in solids.

The results of the fits using Eq. ( 3 ) are shown with dashed lines in the top panels of Fig.  2 . Here, we are mostly interested in the power-law b ( T ). In the liquid phase, as already discussed, we always find that \(b(T)=1\) . On the contrary, in a crystalline solid with long-range order, we have \(b(T)=2\) , as predicted by Debye’s law. By simple extrapolation, we then expect such a power-law to interpolate between these two values by varying the temperature.

In the bottom panels of Fig.  2 , we show the behavior of the power-law b ( T ) as a function of the temperature for water and Fomblin. For water, we observe a sharp jump of the power-law between the liquid value 1 to the solid value 2 at around 273 K, which coincides exactly with the solidification temperature. This sharp change is already evident in the data presented in panel A of Fig. 2 where the dashed lines guide the eyes of the Reader and it is consistent with the structure changes. As shown in the SI, water (using D \(_2\) O, as H \(_2\) O does not give diffraction peaks due to the dominant incoherent neutron scattering cross sections) has a first order phase transition from a liquid state at 286 K, represented by a broad peak in the structure factor, to a crystallized structure at 260 K with well defined sharp peaks in the structure factor. Therefore, the drastic change in b ( T ) simply reflects the first-order nature of the liquid-solid phase transition in water.

Differently from the water-ice case, the transition to the Debye scaling in Fomblin is not sharp, but rather continuous. As confirmed by the measured structure factor S ( K ), shown in the SI, Fomblin does not display an ordered crystalline phase at low temperature but it presents a short-range glassy structure with high viscosity. This implies that for Fomblin no sharp first order phase transition between a liquid and a crystalline phase exists. More precisely, a continuous glass transition connects the low and high temperature phases. As a consequence, the scaling of the low-frequency DOS exhibits a continuous crossover as well. Wee observe a deviation in b ( T ) from the liquid-like scaling below \(\approx 260\) K, which is a much higher temperature than the reported pour temperature for Fomblin, 238 K. Moreover, we find that the data are well fitted by function \(b(T) \propto (T^* -T)^{0.648}\) , with \(T^* \approx 260\) K. This behavior is reminiscent of the dynamics of the order parameter across a continuous (second-order) phase transition, e.g. , Curie’s law for magnetic materials. While we only present two systems here, our results suggest that the scaling of the low-frequency DOS is able to capture some important properties regarding the nature of the phase transition to a solid state at low temperature.

In order to further understand these experimental results, we turn to an analysis based on computational methods using instantaneous normal mode theory.

Instantaneous normal mode theory in a nutshell

The dynamics of liquids are profoundly different from solids as they do not display any translational order, and are more complex than gases due to their high density and strong particle interactions. In solids, atomic motion is entirely characterized by stable oscillations around well-defined potential minima, which are well approximated by a quadratic function of the coordinates. In liquids, or more in general disordered systems, there are many minima contributing to the thermodynamics, together with negative curvature regions and saddles connecting those minima (see Fig.  3 for a cartoon). In the end, these hopping processes across energy barriers are the responsible for macroscopic diffusion, and cannot be neglected.

Schematic illustration of the typical potential energy landscape of a liquid and the corresponding stable and unstable normal modes. Regions with negative local curvature correspond to unstable modes, while minima with positive curvature correspond to solid-like stable modes performing a quasi-harmonic motion. The color-map indicates the local value of the potential energy.

From a computational point of view, in order to derive the density of states of liquids, a common approach is to calculate the DOS using the Hessian matrix, by extending the concept of normal modes in solids to instantaneous normal modes (INMs) (see 15 for a review on the topic), and proceed with a normal-mode analysis for the liquid state 26 , 27 . The main idea behind the INM approach is that, for short time-scales, a liquid is not so different from a solid, or that in more technical words, beyond the hydrodynamic regime, solids and liquids are alike ( e.g. , they both support propagating shear waves 9 , 10 ). Maxwell 28 suggested that the time-scale separating the liquid-like from the solid-like regime corresponds to the stress relaxation time, derived from linear viscoelasticity. As a concrete manifestation of that idea, Zwanzig 29 showed that a normal mode analysis for liquids is still meaningful at short time-scales, i.e. , at each instantaneous snapshot.

For each liquid configuration, the Hessian matrix is a \(3N\times 3N\) matrix, evaluating the second derivatives of the potential energy. Its elements are constructed as follows:

where \(i,j=1,...,N, \mu ,\nu =x,y,z\) . \(\textbf{R}\equiv {\textbf{r}_1,...,\textbf{r}_N}\) represents each liquid configuration and \({\textbf {r}}_i\) is the position of the ith atom. V is the potential energy and \(r_{i,\mu }\) represents the \(\mu \) -coordinate of the ith atom. The instantaneous normal mode frequencies \(\omega _i\) are the square roots of the eigenvalues of the dynamical matrix. The INM spectrum is then obtained,

by averaging on several instantaneous configurations, indicated by the \(\langle \cdot \rangle \) symbol.

In short, INMs are the eigenvalues of the force constant matrix at an instant of time. Rhaman and collaborators 30 early realized that the diagonalization of such a matrix in disordered systems would give rise not only to positive eigenvalues ( \(\lambda >0\) ), but also to negative ones which correspond to purely imaginary frequencies, \(\lambda _\alpha \equiv \omega _\alpha ^2<0\) , and which are labelled as unstable INMs. The INM density of states can be therefore split into a stable part \(g^{\text {INM}}_s(\omega )\) , corresponding to the positive eigenvalues, and an unstable one \(g^{\text {INM}}_u\) , corresponding to the negative ones. For the unstable part, it is conventional to redefine a positive definite “frequency” \(\tilde{\omega } =-i \sqrt{\lambda }\) and plot the corresponding “density of states” along the negative frequency axes by identifying \(\tilde{\omega }=-\omega \) . In this way, one can define a generalized INM density of states \(g^{\text {INM}}(\omega )\) , which corresponds for \(\omega >0\) to \(g^{\text {INM}}_s(\omega )\) and for \(\omega <0\) to \(g^{\text {INM}}_u(\omega )\) . We will follow this convention. Importantly, only the stable part of the INM spectrum can be directly interpreted as a physical density of states function of real eigenfrequencies. Therefore, in the rest of this work, when we will compare the experimental DOS to the INM one, we will always refer to the stable part of the INM spectrum, \(g^{\text {INM}}_s(\omega )\) , or equivalently \(g^{\text {INM}}(\omega )\) for \(\omega >0\) . In this direction, a comparison between the stable part of the INM spectrum and the experimental DOS of heavy water at 290 K has been already shown in Ref. 4 . We also notice that previous studies 12 , 31 , 32 have discussed the density of states of liquids in terms of the interplay of a gas-like and a solid-like component. In that picture, the solid-like contribution relates to the stable part of the INM spectrum.

In order to understand the distinction between positive and negative eigenvalues further, it is illustrative to think of a liquid as a collection of relatively stable local minima, around which the dynamics are harmonic and solid-like, accompanied by structural relaxation in the form of barrier crossing to neighbor wells, with a certain hopping frequency 33 , 34 (see Fig.  3 ). The imaginary frequency modes relate to these relaxational dynamics and, from a potential landscape picture, they correspond to visiting regions of the potential with locally negative curvature. In simpler words, unstable modes are a measure of fluidity, as suggested by several authors 35 , 36 . It then comes as no surprise that unstable INMs bear a close relation to diffusion 8 , 37 , 38 , 39 , 40 , 41 , 42 and several other properties of liquids 13 , 43 , 44 , 45 , 46 , 47 .

The INM density of states has been investigated in several works using molecular dynamics simulations ( e.g. , water 48 , 49 , 50 , 51 , CS \(_2\) 52 , glass-forming liquids 53 , 54 , and even proteins 55 ). Importantly, it has been corroborated by many simulations that at low frequency both the stable and unstable branches of the INM DOS follow a linear scaling in frequency such that the generalized INM DOS can be written as

where the “ \(\dots \) ” indicate higher order corrections which are different for the stable and unstable branches and not relevant for the present discussion. To avoid clutter, we will indicate the slope of the linear regime in Eq. ( 6 ) with the same symbol a ( T ) used for the experimental density of states in Eq. ( 2 ). Nevertheless, a priori, the two slopes are not necessarily the same, as we will explicitly confirm. We emphasize that we are interested in the generalized INM density of states in Eq. ( 6 ) evaluated on the positive frequency axes, corresponding to the stable frequencies and positive eigenvalues.

From a theoretical perspective, several explanations for the linear scaling in Eq. ( 6 ) have been proposed in the past, based on different degrees of simplification 8 , 56 , 57 , 58 , 59 , 60 , 61 , 62 , 63 , 64 . In one way or another, all of them attribute this scaling to the presence of unstable modes. On the contrary, not much is known about the pre-factor a ( T ) in Eq. ( 6 ). Simulations show that the temperature dependence of the linear coefficient a ( T ) is not universal, and theory suggests that it does not depend on a simple physical parameter (as for the Debye’s coefficient in solids), but rather on both the details of the topology of the potential landscape and the associate complex thermodynamic structure. In the literature, one can find simulated systems in which such a coefficient decreases with temperature, and systems in which it increases. In this respect, the emblematic examples are Lennard Jones liquid 8 for the former, and CS \(_2\) for the latter 52 .

Within the framework of INM theory, Keyes 8 proposed a theoretical framework, in good agreement with the data from simulations 8 , 65 , to predict the whole frequency behavior of the unstable INM density of states. In this series of works 8 , 52 (see 64 for a review), Keyes and collaborators derived also a semi-analytical expression for the temperature dependence of the linear slope a ( T ).

The theory predicts that the dominant contribution to the slope in Eq. ( 6 ) is of the form

where \(\langle E\rangle \) is the average barrier height at zero frequency in the potential landscape. Equation ( 7 ) was derived for unstable modes. Nevertheless, as already explained above, the stable and unstable branches are symmetric at low frequency. Therefore, we here assume that Eq. ( 7 ) holds also for the stable part of the spectrum, that can be directly compared to the experimental results.

The mean activation energy \(\langle E\rangle \) plays a fundamental role in relating the INM properties to the temperature dependence of the self-diffusion constant, and it is ultimately connected to the hopping rate along the potential barriers. The concrete definition of \(\langle E \rangle \) is subtle and not very meaningful when the distribution of the energy barriers is broad. Given the existing theories 8 , 52 , the best definition of \(\langle E \rangle \) that one could provide is that of an average inflection point energy, which serves as a natural zero for the barrier energy as a function of the frequency \(\omega \) , and which can be estimated directly in the soft potential model 57 , 66 . Because of the absence of a robust theoretical definition for the activation energy, in the rest of this manuscript we will take \(\langle E \rangle \) as a phenomenological fitting parameter.

In summary, Eq. ( 7 ) is a simple formula (it neglects several microscopic details about the topology of the potential landscape) that nevertheless provides a sharp prediction. The rest of the manuscript will be devoted to validate this prediction against the experimental data.

Before proceeding, we notice that another important quantity in the study of liquid dynamics is given by the Fourier transform of the velocity auto-correlation function, that is sometimes referred to as the “density of state function” \(\Xi (\omega )\) . It is well known (see for example Ref. 67 ) that the density of state function \(\Xi (\omega )\) and the INM spectrum \(g(\omega )\) exhibit crucial differences, and they agree only in the low-temperature harmonic limit. For example, \(\Xi (\omega )\) shows a finite value at \(\omega =0\) , while the INM spectrum shows imaginary frequencies and vanishes at \(\omega =0\) . The origin of these discrepancies lies in the instantaneous and harmonic approximations assumed in the INM analysis, that cannot entirely capture the inherent anharmonicity in the long-time dynamics of the liquid state 68 . To the best of our knowledge, a detailed comparison between \(\Xi (\omega )\) and \(g(\omega )\) for different systems and different conditions has never been performed. For our water system, a comparison between the experimental data, the density of state function \(\Xi (\omega )\) and the INM spectrum \(g(\omega )\) is provided in the SI for two benchmark values of temperature. A more detailed and complete study, involving more liquid systems, is left for future work. In this manuscript, we will mainly focus on the comparison between the experimental data and the normal mode spectrum. We also notice that some anharmonic extensions of the INM analysis already exist in the literature, e.g. , Ref. 69 . We plan to explore these methods to check whether some of the differences mentioned above disappear when anharmonicities are properly accounted for.

Instantaneous normal mode analysis

At each temperature, we performed INM analyses for 100 different liquid configurations generated at 5 ps intervals during the last 500 ps of the full simulation using GROMACS. The INM spectrum was averaged over the analyzed liquid configurations.

Unfortunately, for Fomblin we have not been able to carry out an INM analysis since the force potential is unknown. For liquid water, we carried out a 1 ns molecular dynamics simulation of the flexible TIP3P model. Previous normal mode analyses for supercooled water can be found in 48 , 49 , 50 , 51 . The resulting INM density of states after average, including both the stable branch \(g^{\text {INM}}_s(\omega )\) and the unstable one \(g^{\text {INM}}_u(\omega )\) , is shown in Fig.  4 A. We have followed the standard notation and plotted the imaginary frequencies, corresponding to the unstable modes, on the negative frequency axis. As evident from Fig.  4 A, both the stable and unstable parts display a clear linear-in-frequency behavior at low frequency. Moreover, the slope is the same for the two parts 64 , as anticipated. In other words, both the stable and unstable branches of the INM DOS follow the linear behavior presented in Eq. ( 6 ). A further proof of this behavior is provided in the SI using a zoom of the simulation data around the origin, \(\omega =0\) .

Importantly, only the stable branch of the INM DOS, \(g^{\text {INM}}_s(\omega )\) , can be compared to the experimentally measured DOS \(g(\omega )\) . The density of states for stable INMs, \(g^{\text {INM}}_s(\omega )\) can be decomposed into two regimes. The region below 50 meV involves mostly modes related to the translational motion. On the contrary, the region above 50 meV is governed by rotational modes 51 . By increasing temperature, the number of unstable modes increases. Additionally, despite both parts of the INM DOS displaying a clear peak at low frequency, the one for stable modes appears to be insensitive to the temperature T , while the one for the unstable modes grows in intensity with temperature. This is a concrete proof that the two branches are symmetric only in the low-frequency linear regime presented in Eq. ( 6 ) but not for larger frequencies.

For completeness, in the SI, we show the comparison between our results for the INM DOS of water at 300K and the existing data in the literature that used TIP4P/2005 rigid water potential 70 and flexible SPC water potential 71 . Evidently, the low-frequency behavior of the INM DOS, and in particular the linear scaling and its slope, do not depend on the specific potential used. Nonetheless, the higher energy dynamics do show strong dependency on different potentials. This will be a topic for further studies.

In Fig.  4 B, we show the comparison of the experimental DOS for water at 310 K and the DOS of stable INMs at the same temperature. The two curves have been normalized to the first peak. As evident from Fig.  4 B, there is a crucial difference between the two curves since the INM density of states \(g^{\text {INM}}_s(\omega )\) does not contain the diffusion component, and therefore \(g^{\text {INM}}_s(0)=0\) , while the experimental curve clearly shows a finite value at zero frequency. As explained above, the zero frequency value is given by the self-diffusion coefficient D and increases with temperature. In the SI, we show a more in depth comparison between the two curves at different temperatures and also a comparison with the density of state function obtained from the velocity auto-correlation function for two concrete values of temperature. Since g (0) diminishes when the temperature is lowered, the two curves resemble each other more near the solid-phase. Indeed, in a low-temperature solid, we expect the two curves to be identical (see 45 for a similar observation). Moving towards higher frequency, we observe that in the low-energy regime, below approximately 20 meV, the two curves are similar to each other. Both of them show a clear linear-in-frequency regime, and then a sharp peak with a similar linewidth. At larger frequencies, the two curves differ considerably. It is plausible that this mismatch is caused by the relative intensity of the features in the neutron scattering DOS which is skewed by the larger scattering cross section of hydrogen atoms compared to oxygen (see Methods for details). Another possible reason may be that the INM approach itself, or the potential chosen, can not reproduce the high energy modes properly. In order to avoid potential misunderstandings, we clarify that the normalization in Fig.  4 B is arbitrary. The apparent quantitative agreement at low energy is influenced by the normalization scheme and should not be interpreted as evidence of a quantitative match between the experimental DOS and the INM DOS, as the rest of our manuscript demonstrates otherwise. We also notice that the overall normalization factor certainly affects quantitative comparisons but not qualitative features, as for example the linear in frequency behavior at small frequency discussed in our manuscript.

The instantaneous normal mode (INM) analysis for liquid water. ( A ) The INM density of states (plotted with the imaginary frequency branch on the negative \(\omega \) axis) for different temperatures. ( B ) Experimental data for the DOS obtained by INS (red), and INM density of states (blue) at 310 K. The data are normalized using the height of the first peak. ( C ) The fraction of unstable modes \(f_u(T)\) as a function of temperature. The dashed line indicates the result of a linear fit. ( D ) The slope of the INM DOS a ( T ) as a function of the temperature in a linear logarithmic scale. The line indicates the fit to the prediction from INM theory, Eq. ( 8 ). The background colored regions illustrate the uncertainties of the numerical data.

A crucial quantity in the INM analysis is given by the fraction of unstable modes \(f_u\) , which is defined as the ratio of the number of unstable modes to the total 3N modes. Figure  4 C displays the fraction of unstable modes \(f_u\) as a function of temperature. As expected, the fraction increases with increasing temperature and for the case of water exhibits an evident linear in T dependence. Interestingly, this functional form coincides exactly with the prediction of a specific theoretical model discussed in the literature, the “random energy model” 63 ( e.g. , Eq. 13.33 in 64 ).

Let us now focus on the linear slope a ( T ), which is plotted in Fig.  4 D as a function of the inverse temperature. First, the behavior of the linear slope is consistent with the result from the experiments, as it increases monotonically with the temperature. In order to provide a more quantitative understanding of the slope, we resort to Eq. ( 7 ) based on the INM analysis. As shown in Fig.  4 D, the theoretical formula is in good qualitative agreement with data from simulations. In particular, by plotting the simulation data in a linear logarithmic scale in Fig.  4 D, we unequivocally show that the slope follows an Arrhenius-like exponential form \(\exp \left( -\langle E\rangle /k_B T\right) \) , as predicted by INM theory, i.e. , Eq. ( 7 ). Given the complexity of liquid dynamics, it is remarkable that INM theory is able to capture the qualitative temperature dependence of a ( T ). In order to provide a more quantitative analysis, we fit our data for the slope a ( T ) using Eq. ( 7 ) with an additional constant pre-factor A ,

According to our best fit, we obtain the following values for the various parameters

Next, we will extend the analysis to the experimental data and compare the outcomes with the results from the simulations performed in this section.

Comparing the experimental DOS to the stable INM DOS

We return to the experimental data for the DOS of water and Fomblin displayed in Fig.  1 . Here, we are mostly concerned with the temperature dependence of the slope for the linear scaling which has been observed to be universal in the whole liquid phase. The data, obtained by fitting the experimental results with Eq. ( 8 ), are shown in Fig.  5 . For Fomblin, the slope is extracted by fitting the experimental data from INS normalized by the total area. This normalization is justified by the fact that around \(\approx 120\) meV, where our numerical data stops, the spectral weight is already very small (see Fig.  1 B), and the remaining tail is negligible. For water (see Fig.  1 A), this is not the case, as around \(\approx 140\) meV, the DOS is still large. As a consequence, a normalization of the DOS by the area of the curves up to that experimental cutoff would lead to uncontrollable results. Therefore, for water, we have normalized all the curves by their value at zero frequency, using the experimental and simulation data for the self-diffusion constant D ( T ) 72 , which is known precisely at all temperatures. This problem with the normalization of the experimental data for water will not affect qualitatively our findings and our analysis, but it will make impossible to reliably study quantitative features such as the value of \(\langle E \rangle \) . On top of that, the experimental DOS for water is also affected by the different scattering cross sections of different atoms. This implies that certain vibrational modes can be over- or under-represented if there is a large difference in scattering power between the elements involved in different modes as it happens for hydrogen and oxygen (see Methods).

The temperature dependence of the linear slope a ( T ) for the experimental DOS of water ( A ) and Fomblin ( B ) in a linear-log plot. The background shaded regions indicate the uncertainties of the data. The colored dashed lines indicate the fit with Eq. ( 8 ). For the case of water, panel A, we present also the slope of the low-frequency DOS a ( T ) extracted from the INM analysis (red color).

Let us first focus on the qualitative functional dependence of a ( T ) that can be analyzed using the data presented in Fig.  5 . For both systems, the slope of the experimental DOS increases monotonically with temperature. Additionally, we find that both liquids display an exponential Arrhenius-like behavior as in Eq. ( 7 ), which is derived from INM theory 8 , 52 based on the dynamics of unstable modes.

Taking into account the difficulties previously discussed, we attempt a more quantitative analysis by fitting the slope of the experimental DOS of water using Eq. ( 8 ). We obtain the following results:

By comparing these results with those obtained from the INM DOS, and presented in Eq. ( 9 ), the discrepancy between the two set of values is evident. First, we notice that the fitted parameters concerning the experimental data are inevitably affected by much larger errors. Then, the overall prefactor A is almost two orders of magnitude different. This is not a surprise as the latter is highly sensitive to the normalization of the DOS, and therefore its value not reliable for the reasons explained above. Additionally, we notice that the energy scale \(\langle E\rangle \) extracted from the experimental data is \(\approx 5\) times larger than that from the simulation data. This is made more evident by presenting together the results for the slope a ( T ) from the experimental data and the INM theoretical analysis in Fig.  5 A. In summary, both the experimental and simulation data show a clear exponential behavior, \(\exp \left( -\langle E\rangle /k_B T\right) \) , but with a different activation energy. We believe that this discrepancy is due to the strong anharmonic effects that are inevitably present in liquids and that are not entirely captured by analysis of the Hessian eigenvalues on which INM theory is based. It would be interesting to consider anharmonic extensions of the normal mode analysis, such as that proposed in Ref. 69 , and see whether this quantitative discrepancy will be resolved. We leave this task for future research.

In order to conclude our quantitative analysis, we analyzed the experimental data for the slope a ( T ) for Fomblin following the same method. We obtain the following values for the fitting parameters:

These results suggest that the energy scale \(\langle E\rangle \) for fomblin is about a factor 2-3 smaller than for water. At this point, without performing a more comprehensive analysis involving more liquid systems, it is hard to make any claim about this energy scale and its physical meaning. We plan to extend our investigation in the near future by performing large-scale MD simulations and INM analysis of several liquid substances.

Without further analysis, we can conclude that our experimental results confirm the validity of the universal linear law for the DOS of liquids \(g(\omega )\propto \omega \) in a wide range of temperature in the liquid phase. On top of that, and as the main novelty of our work, our results reveal that the temperature dependence of the slope associated to this universal scaling law exhibits an exponential \(\exp (-\langle E\rangle /T)\) form. Importantly, this exponential behavior can be derived using INM theory that successfully reproduces the qualitative trend of the experimental data.

However, we find that the INM analysis is not sufficient from a quantitative point of view. In particular, as evident from Fig.  5 A, it cannot correctly capture the precise value of the energy scale \(\langle E\rangle \) , whose precise meaning remains unclear. We leave this important point as an open question for the future.

In this work, we studied the DOS of water and Fomblin oil, combining experimental neutron scattering techniques (INS), MD simulations and INM theory. Our focus is the low-energy regime of the DOS and in particular (I) its scaling with frequency, and (II) its temperature dependence. In such a regime, and independently of the value of the temperature (from the melting temperature to \(\approx 340\) K), we have experimentally verified that both liquids display a universal linear in frequency scaling \(g(\omega )= a(T) \omega +\dots \) , which was previously experimentally observed for only one value of temperature in Ref. 5 , and predicted before in Ref. 56 .

By analyzing the temperature dependence of the linear slope a ( T ), we observed that a ( T ) grows monotonically with temperature over a wide range of temperatures in the liquid phase, and that the functional behavior is compatible with an exponential Arrhenius-like form typical of thermally activated dynamics. In order to rationalize this behavior, we have resorted to INM theory 8 . Our INM simulations show a good qualitative agreement with the experimental data and predict the same temperature dependence of the slope values as observed in the INS experiments. In particular, the experimental data confirmed that the temperature dependence of the linear slope is dominated by an exponential factor \(\exp (-\langle E \rangle /k_B T)\) , as predicted by INM theory 8 , 52 , 64 . Nevertheless, the INM analysis fails in reproducing quantitatively the experimental results (see Fig.  5 A), and more work has to be done to establish the role and the physical meaning of the energy scale appearing in the exponential behavior of the slope as a function of temperature. We propose that this discrepancy originates from the anharmonic effects that are neglected in the INM analysis, based on the instantaneous Hessian matrix.

Finally, we have tracked the power-law scaling of the low-frequency portion of the experimental DOS across the melting temperature. For water, we have consistently observed a sharp transition between the liquid-like linear scaling to the quadratic Debye law, which appears around the expected melting temperature. On the contrary, for Fomblin, the transition from the liquid behavior to Debye’s law is continuous, as expected from its polymeric glassy structure. These findings imply that the low-frequency temperature dependent behavior of the experimental DOS is able not only to capture the solid-liquid transition but also its nature, whether a sharp first-order thermodynamic transition or a continuous glassy-like transition.

In conclusion, we hope that our analysis will motivate future studies on the low-frequency behavior of the liquid DOS, a fundamental but rather unexplored topic that deserves further attention. The important lesson from our work is that (as probably one might have expected) the DOS of liquids cannot be explained solely based on the instantaneous normal modes but that, nevertheless, important physical properties of liquids can be correctly captured and explained by INMs. For the future, we can envisage two immediate questions that need to be resolved. First, it is fundamental to compare in detail the density of state function \(\Xi (\omega )\) and the INM spectrum \(g(\omega )\) under different conditions. These two quantities are identical in the harmonic limit but they consistently differ in the liquid phase at finite temperature. It would be very fruitful to investigate this point further in relation to the available experimental data as well. Moreover, it would be interesting to extend the INM analysis by incorporating anharmonic effects, as done for example in 69 , and verify whether the agreement between the experimental data and the INM spectrum is resolved in doing so. We leave these two questions for the near future.

Two samples have been studied with INS. General-purpose laboratory grade of deionized water with a mass of 0.5 g is filled inside an annular aluminum can having 0.1 mm gap. A similar can with 0.5 mm gap is filled with 4 g of Fomblin oil (YL 25/6, Fomblin ® Y LVAC 25/6 Solvay) 73 ; CF \(_3\) O[–CF(CF \(_3\) )CF \(_2\) O–] \(_x\) (–CF \(_2\) O–) \(_y\) CF \(_3\) where x / y gives an average molecular weight of 3300. These aluminum cans have been specially designed with high strength aluminum to maintain the integrity of the sample can under high vapor pressure of water at high temperatures. A half dozen of sample cans have been destroyed during the test. The estimated transmission is \(90\%\) for these samples and this ensures the \(10\%\) level of scattering to minimize multiple scatterings.

Inelastic neutron scattering.

The inelastic neutron scattering measurements were conducted using Pelican - the time-of-flight cold neutron spectrometer at the Australian Nuclear Science and Technology Organisation. A neutron energy of 3.7 meV (wavelength of 4.69 Å) was employed, yielding an energy resolution of 0.135 meV at the elastic line. Sample temperature control, ranging from 1.5 to 800 K, was achieved via a top-load cryo-furnace. Background subtraction involved measuring the corresponding empty can under identical conditions as the samples. Additionally, a standard vanadium sample was measured for detector efficiency normalization and energy resolution determination. The experimental dynamic scattering function \(S(Q,\omega )\) was derived using the Large Array Manipulation Program (LAMP) 74 , following background subtraction and detector normalization. For density of states (DOS) determination, the \(S(Q,\omega )\) corresponding to the neutron energy gain the SIde was utilized. The DOS for the solid phase of materials was determined through a standard procedure outlined in various textbooks. In the incoherent one-phonon approximation, the relationship between DOS and the experimental scattering function for a Bravais powder sample (isotropic system) can be expressed as 75 :

Here, C is a factor containing the atomic mass and Debye-Waller factor, \(\exp \left( -2W\right) \) , which is taken as unity for all samples. Furthermore, \(k_B\) is the Boltzmann constant. Experimentally, the DOS is averaged over the Q range covered, typically from 0.2 to 9Å −1 , depending on the energy range considered.

For non-Bravais samples, the measured phonon density of states (DOS) is determined by the neutron-weighted phonon density of states, known as the generalized phonon density of states (GDOS) 75 :

where the sum over i includes all elements in the sample, \(f_i\) is the \(i_{th}\) atomic concentration, \(\sigma _i\) is the total neutron bound cross section accounting for both coherent and incoherent scattering processes, \(M_i\) is the atomic mass, and \(g_i\left( \omega \right) \) is the partial DOS of the element i . In the case of water, the measured DOS is predominantly influenced by hydrogen due to the higher \(\sigma /M\) ratio for H compared to O.

The experimental determination of DOS is subject to weighted averaging effects and potential errors introduced by multiple scatterings and multi-phonon scattering. To mitigate multiple scattering effects, sample thicknesses were chosen to yield approximately \(10\%\) neutron scattering. For instance, sample thicknesses of 0.2 mm and 1.0 mm were utilized for water and Fomblin, respectively. Multi-phonon scattering effects are generally small, particularly within the low Q and low energy ranges interested in this study. The Q range covered for the low energy ( \(< 3\) meV) DOS is from 0.2 Å −1 to about 2 Å −1 . Consequently, the reported GDOS here has no correction for multiple scatterings and multi-phonon scattering effects. Furthermore, the correct Debye \(\omega ^2\) law obtained for solid phase of the materials has indicated that these effects are at the negligible level.

For the liquid phase, a similar procedure was followed to derive the DOS. In the low-energy limit, Eq. ( 14 ) simplifies to:

It is noteworthy that Eq. ( 16 ) is essentially equivalent to the formula for determining the frequency spectrum 76 or velocity frequency function 77 for liquids in the limit of \(Q \rightarrow 0\) . This same formulation was also utilized to derive the proton density of states of water 78 .

Molecular dynamics (MD) simulations

The MD simulations to simulate liquid water at different temperatures were performed with GROMACS 79 . We used the flexible TIP3P water model defined in GROMACS, which allows the stretching of the O–H bond and the bending of the H–O–H angle. The flexible water was applied into the initial topology and the SETTLE algorithm was not applied. We simulated 392 water molecules in our liquid system which was equilibrated in NVT and NPT ensembles, using the Nosé-Hoover thermostat and Parrinello-Rahman barostat to control the temperature and pressure. We carried a 1 ns production MD simulation for data collection at atmospheric pressure and constant temperature. During the whole simulation, the periodic boundary conditions were employed. Using the Leapfrog-Verlet algorithm, the time step was set to 0.1 fs. The long-range electric interactions were calculated by using the particle mesh Ewald (PME) method. The self-diffusion coefficient of the system can be determined through the mean squared displacement (MSD) via the Einstein relation:

where \({\textbf {r}}_i(t)\) is the position vector of the ith particle at time t.

Data availability

The datasets generated and analysed during the current study are available upon reasonable request by contacting the corresponding authors.

Kittel, C. Introduction to solid state physics Eighth edition (2021).

Ramos, M. A. Low-Temperature Thermal and Vibrational Properties of Disordered Solids (World Scientific, 2022).

Book   Google Scholar  

Phillips, W. A., Buchenau, U., Nücker, N., Dianoux, A.-J. & Petry, W. Dynamics of glassy and liquid selenium. Phys. Rev. Lett. 63 , 2381–2384. https://doi.org/10.1103/PhysRevLett.63.2381 (1989).

Article   ADS   CAS   PubMed   Google Scholar  

Dawidowski, J., Bermejo, F., Cabrillo, C. & Bennington, S. Generalized frequency spectra of water at both sides of the freezing transition. Chem. Phys. 258 , 247–255. https://doi.org/10.1016/S0301-0104(00)00128-2 (2000).

Article   CAS   Google Scholar  

Stamper, C., Cortie, D., Yue, Z., Wang, X. & Yu, D. Experimental confirmation of the universal law for the vibrational density of states of liquids. J. Phys. Chem. Lett. 13 , 3105–3111. https://doi.org/10.1021/acs.jpclett.2c00297 (2022) ( PMID: 35362320 ).

Article   CAS   PubMed   Google Scholar  

Buchsteiner, A., Lerf, A. & Pieper, J. Water dynamics in graphite oxide investigated with neutron scattering. J. Phys. Chem. B 110 , 22328–22338. https://doi.org/10.1021/jp0641132 (2006) ( PMID: 17091972 ).

Trouw, F. R. & Price, D. L. Chemical applications of neutron scattering. Ann. Rev. Phys. Chem. 50 , 571–601. https://doi.org/10.1146/annurev.physchem.50.1.571 (1999) ( PMID: 15012422 ).

Article   ADS   CAS   Google Scholar  

Keyes, T. Unstable modes in supercooled and normal liquids: Density of states, energy barriers, and self-diffusion. J. Chem. Phys. 101 , 5081–5092. https://doi.org/10.1063/1.468407 (1994).

Trachenko, K. & Brazhkin, V. Collective modes and thermodynamics of the liquid state. Rep. Prog. Phys. 79 , 016502 (2015).

Article   ADS   MathSciNet   PubMed   Google Scholar  

Baggioli, M., Vasin, M., Brazhkin, V. & Trachenko, K. Gapped momentum states. Phys. Rep. 865 , 1–44. https://doi.org/10.1016/j.physrep.2020.04.002 (2020).

Article   ADS   MathSciNet   Google Scholar  

Bolmatov, D., Brazhkin, V. V. & Trachenko, K. The phonon theory of liquid thermodynamics. Sci. Rep. 2 , 421. https://doi.org/10.1038/srep00421 (2012).

Article   ADS   CAS   PubMed   PubMed Central   Google Scholar  

Moon, J., Thébaud, S., Lindsay, L. & Egami, T. Normal mode description of phases of matter: Application to heat capacity. Phys. Rev. Res. 6 , 013206. https://doi.org/10.1103/PhysRevResearch.6.013206 (2024).

Baggioli, M. & Zaccone, A. Explaining the specific heat of liquids based on instantaneous normal modes. Phys. Rev. E 104 , 014103. https://doi.org/10.1103/PhysRevE.104.014103 (2021).

Article   ADS   MathSciNet   CAS   PubMed   Google Scholar  

Stratt, R. M. The instantaneous normal modes of liquids. Acc. Chem. Res. 28 , 201–207 (1995).

Keyes, T. Instantaneous normal mode approach to liquid state dynamics. J. Phys. Chem. A 101 , 2921–2930. https://doi.org/10.1021/jp963706h (1997).

Fomblin, P. F. P. E. Vacuum Pump Oils (Product Data Sheet, Solvay Solexis, 2002).

Pokotilovski, Y. N., Natkaniec, I. & Holderna-Natkaniec, K. The experimental and calculated density of states and UCN loss coefficients of perfluoropolyether oils at low temperatures. Physica B: Condensed Matter 403 , 1942–1948 (2008).

Yuan, Bao et al. Two pressure cells for quasielastic and inelastic neutron scatterings. EPJ Web Conf. 272 , 02009. https://doi.org/10.1051/epjconf/202227202009 (2022).

O’Neil, M. J. The Merck index : an encyclopedia of chemicals, drugs, and biologicals . Encyclopedia of chemicals, drugs, and biologicals (Royal Society of Chemistry, Cambridge, UK, 2013), 15th ed. / maryadele j. o’neil, editor-in-chief ... [et al.]. edn.

Wagner, W. & Pruß, A. The IAPWS formulation 1995 for the thermodynamic properties of ordinary water substance for general and scientific use. J. Phys. Chem. Ref. Data 31 , 387–535. https://doi.org/10.1063/1.1461829 (2002).

van der Leeden, F. The Water Encyclopedia, Second Edition . Geraghty & Miller ground-water series (Taylor & Francis, 1990).

Franks, F. Water a comprehensive treatise: volume 4: aqueous solutions of amphiphiles and macromolecules (Springer Science & Business Media, 2013).

Toukan, K. et al. Neutron-scattering measurements of wave-vector-dependent hydrogen density of states in liquid water. Phys. Rev. A 37 , 2580 (1988).

Amann-Winkel, K. et al. X-ray and neutron scattering of water. Chem. Rev. 116 , 7570–7589 (2016).

Hansen, J.-P. & McDonald, I. R. Theory of simple liquids: with applications to soft matter (Academic press, 2013).

Google Scholar  

Seeley, G. & Keyes, T. Normal-mode analysis of liquid?state dynamics. J. Chem. Phys. 91 , 5581–5586. https://doi.org/10.1063/1.457664 (1989).

Krämer, N., Buchner, M. & Dorfmüller, T. Normal mode dynamics in simple liquids. J. Chem. Phys. 109 , 1912–1919. https://doi.org/10.1063/1.476768 (1998).

Article   ADS   Google Scholar  

Maxwell, J. C. IV. On the dynamical theory of gases. Philosophical transactions of the Royal Society of London 49–88 (1867).

Zwanzig, R. Elementary excitations in classical liquids. Phys. Rev. 156 , 190–195. https://doi.org/10.1103/PhysRev.156.190 (1967).

Mandell, M., McTague, J. & Rahman, A. Crystal nucleation in a three-dimensional Lennard-Jones system: A molecular dynamics study. J. Chem. Phys. 64 , 3699–3702 (1976).

Lin, S.-T., Blanco, M., Goddard, I. & William, A. The two-phase model for calculating thermodynamic properties of liquids from molecular dynamics: Validation for the phase diagram of Lennard-Jones fluids. J. Chem. Phys. 119 , 11792–11805. https://doi.org/10.1063/1.1624057 (2003).

Moon, J. Continuity of the Solid, Liquid, and Gas Phases of Matter 61–91 (Springer Nature Switzerland, 2024).

Goldstein, M. Viscous liquids and the glass transition: A potential energy barrier picture. J. Chem. Phys. 51 , 3728–3739. https://doi.org/10.1063/1.1672587 (1969).

Stillinger, F. H. & Weber, T. A. Dynamics of structural transitions in liquids. Phys. Rev. A 28 , 2408–2416. https://doi.org/10.1103/PhysRevA.28.2408 (1983).

LaViolette, R. A. & Stillinger, F. H. Multidimensional geometric aspects of the solid-liquid transition in simple substances. J. Chem. Phys. 83 , 4079–4085 (1985).

Cotterill, R. M. J. & Madsen, J. U. Localized fluidity modes and the topology of the constant-potential-energy hypersurfaces of Lennard-Jones matter. Phys. Rev. B 33 , 262–268. https://doi.org/10.1103/PhysRevB.33.262 (1986).

Madan, B., Keyes, T. & Seeley, G. Diffusion in supercooled liquids via normal mode analysis. J. Chem. Phys. 92 , 7565–7569. https://doi.org/10.1063/1.458192 (1990).

Keyes, T. Normal mode theory of diffusion in liquids for a broad temperature range. J. Chem. Phys. 103 , 9810–9812. https://doi.org/10.1063/1.469947 (1995).

Clapa, V. I., Kottos, T. & Starr, F. W. Localization transition of instantaneous normal modes and liquid diffusion. J. Chem. Phys. 136 , 144504. https://doi.org/10.1063/1.3701564 (2012).

Gezelter, J. D., Rabani, E. & Berne, B. J. Calculating the hopping rate for diffusion in molecular liquids: CS2. J. Chem. Phys. 110 , 3444–3452. https://doi.org/10.1063/1.478211 (1999).

Li, W.-X., Keyes, T. & Sciortino, F. Three-flavor instantaneous normal mode formalism: Diffusion, harmonicity, and the potential energy landscape of liquid cs2. J. Chem. Phys. 108 , 252–260. https://doi.org/10.1063/1.475376 (1998).

Chowdhary, J. & Keyes, T. Conjugate gradient filtering of instantaneous normal modes, saddles on the energy landscape, and diffusion in liquids. Phys. Rev. E 65 , 026125. https://doi.org/10.1103/PhysRevE.65.026125 (2002).

Madan, B., Keyes, T. & Seeley, G. Normal mode analysis of the velocity correlation function in supercooled liquids. J. Chem. Phys. 94 , 6762–6769. https://doi.org/10.1063/1.460252 (1991).

Madan, B. & Keyes, T. Unstable modes in liquids density of states, potential energy, and heat capacity. J. Chem. Phys. 98 , 3342–3350. https://doi.org/10.1063/1.464106 (1993).

Moon, J., Thébaud, S., Lindsay, L. & Egami, T. Microscopic view of heat capacity of matter: solid, liquid, and gas. arXiv preprint arXiv:2210.06218 (2022).

Zhang, W., Douglas, J. F. & Starr, F. W. What does the instantaneous normal mode spectrum tell us about dynamical heterogeneity in glass-forming fluids?. J. Chem. Phys. 151 , 184904. https://doi.org/10.1063/1.5127821 (2019).

Donati, C., Sciortino, F. & Tartaglia, P. Role of unstable directions in the equilibrium and aging dynamics of supercooled liquids. Phys. Rev. Lett. 85 , 1464–1467. https://doi.org/10.1103/PhysRevLett.85.1464 (2000).

La Nave, E., Scala, A., Starr, F. W., Sciortino, F. & Stanley, H. E. Instantaneous normal mode analysis of supercooled water. Phys. Rev. Lett. 84 , 4605–4608. https://doi.org/10.1103/PhysRevLett.84.4605 (2000).

La Nave, E., Scala, A., Starr, F. W., Stanley, H. E. & Sciortino, F. Dynamics of supercooled water in configuration space. Phys. Rev. E 64 , 036102. https://doi.org/10.1103/PhysRevE.64.036102 (2001).

Sciortino, F. & Tartaglia, P. Harmonic dynamics in supercooled liquids: The case of water. Phys. Rev. Lett. 78 , 2385–2388. https://doi.org/10.1103/PhysRevLett.78.2385 (1997).

Cho, M., Fleming, G. R., Saito, S., Ohmine, I. & Stratt, R. M. Instantaneous normal mode analysis of liquid water. J. Chem. Phys. 100 , 6672–6683 (1994).

Li, W.-X. & Keyes, T. Instantaneous normal mode theory of diffusion and the potential energy landscape: Application to supercooled liquid cs2. J. Chem. Phys. 111 , 5503–5513. https://doi.org/10.1063/1.479810 (1999).

Bembenek, S. D. & Laird, B. B. Instantaneous normal modes and the glass transition. Phys. Rev. Lett. 74 , 936–939. https://doi.org/10.1103/PhysRevLett.74.936 (1995).

Shimada, M., Shiraishi, K., Mizuno, H. & Ikeda, A. Instantaneous normal modes of glass-forming liquids during the athermal relaxation process of the steepest descent algorithm (2023). arXiv:2111.11681 .

Schulz, R., Krishnan, M., Daidone, I. & Smith, J. C. Instantaneous normal modes and the protein glass transition. Biophys. J. 96 , 476–484. https://doi.org/10.1016/j.bpj.2008.10.007 (2009).

Zaccone, A. & Baggioli, M. Universal law for the vibrational density of states of liquids. Proc. Natl. Acad. Sci. 118 , e2022303118. https://doi.org/10.1073/pnas.2022303118 (2021).

Article   MathSciNet   CAS   PubMed   PubMed Central   Google Scholar  

Zürcher, U. & Keyes, T. Anharmonic potentials in supercooled liquids: The soft-potential model. Phys. Rev. E 55 , 6917–6927. https://doi.org/10.1103/PhysRevE.55.6917 (1997).

Wu, T. & Loring, R. F. Phonons in liquids: A random walk approach. J. Chem. Phys. 97 , 8568–8575. https://doi.org/10.1063/1.463375 (1992).

Wu, T. & Loring, R. F. Collective motions in liquids with a normal mode approach. J. Chem. Phys. 99 , 8936–8947. https://doi.org/10.1063/1.465563 (1993).

Wan, Y. & Stratt, R. M. Liquid theory for the instantaneous normal modes of a liquid. J. Chem. Phys. 100 , 5123–5138. https://doi.org/10.1063/1.467178 (1994).

Schirmacher, W., Bryk, T. & Ruocco, G. Modeling the instantaneous normal mode spectra of liquids as that of unstable elastic media. Proc. Natl. Acad. Sci. 119 , e2119288119. https://doi.org/10.1073/pnas.2119288119 (2022).

Xu, B. & Stratt, R. M. Liquid theory for band structure in a liquid. J. Chem. Phys. 91 , 5613–5627. https://doi.org/10.1063/1.457564 (1989).

Keyes, T., Chowdhary, J. & Kim, J. Random energy model for dynamics in supercooled liquids: \(n\) dependence. Phys. Rev. E 66 , 051110. https://doi.org/10.1103/PhysRevE.66.051110 (2002).

Keyes, T. Normal Mode Analysis: Theory and Applications to Biological and Chemical Systems , chap. The Relation Between Unstable Instantaneous Normal Modes and Diffusion (Chapman and Hall/CRC, 2005).

Keyes, T., Vijayadamodar, G. V. & Zurcher, U. An instantaneous normal mode description of relaxation in supercooled liquids. J. Chem. Phys. 106 , 4651–4657. https://doi.org/10.1063/1.473481 (1997).

Xu, H., Baggioli, M. & Keyes, T. A fresh look at the vibrational and thermodynamic properties of liquids within the soft potential model. J. Chem. Phys. 159 , 154501. https://doi.org/10.1063/5.0158089 (2023).

Seeley, G., Keyes, T. & Madan, B. From the density of states to the velocity correlation function in liquids. J. Chem. Phys. 96 , 4074–4076. https://doi.org/10.1021/j100189a029 (1992).

David, E. F. & Stratt, R. M. The anharmonic features of the short-time dynamics of fluids: The time evolution and mixing of instantaneous normal modes. J. Chem. Phys. 109 , 1375–1390. https://doi.org/10.1063/1.476690 (1998).

Kuo, Y.-W., Wang, C.-W., Tang, P.-H. & Wu, T.-M. Layer structure and intermolecular vibrations of water confined within graphite nanoslits. Chem. Phys. Lett. 825 , 140612 (2023).

Huang, B.-C. & Chang, C.-H. Localization-delocalization transition of the instantaneous normal modes of liquid water. Phys. Rev. E 88 , 042116 (2013).

Tsimpanogiannis, I. N. et al. Self-diffusion coefficient of bulk and confined water: a critical review of classical molecular simulation studies. Mol. Simulat. 45 , 425–453. https://doi.org/10.1080/08927022.2018.1511903 (2019).

Technical Data Sheet Fomblin ® Y LVAC \(25/6\) (2020).

Richard, D., Ferrand, M. & Kearley, G. Lamp, the large array manipulation program. J. Neutron Res. 4 , 33–39 (1996).

Article   Google Scholar  

Boothroyd, A. T. Principles of Neutron Scattering from Condensed Matter (Oxford University Press, 2020).

Powles, J. G. Atomic and molecular motion in liquids by thermal neutron scattering. In Chemical Applications of Thermal Neutron Scattering (ed. Willis, B. T. M.) 130–132 (Oxford University Press, 1973).

Squires, G. Introduction to the Theory of Thermal Neutron Scattering . Dover books on physics (Dover Publications, 1996).

Chen, S.-H., Toukan, K., Loong, C.-K., Price, D. L. & Teixeira, J. Hydrogen-bond spectroscopy of water by neutron scattering. Phys. Rev. Lett. 53 , 1360 (1984).

Van Der Spoel, D. et al. Gromacs: fast, flexible, and free. J. Comput. Chem. 26 , 1701–1718 (2005).

Article   PubMed   Google Scholar  

Download references

Acknowledgements

We would like to thank H. Xu, J. Douglas, Y. Feng and especially T. Keyes for fruitful discussions and related collaborations on the topic of liquids. We are grateful to Tom Keyes for comments and suggestions on a preliminary version of this manuscript. M.B. acknowledges the support of the Shanghai Municipal Science and Technology Major Project (Grant No.2019SHZDZX01) and the sponsorship from the Yangyang Development Fund. D. Y., C. S. and R. M. acknowledge the beam time awarded from ANSTO for the access to Pelican instrument (P13964).

Author information

These authors contributed equally: Sha Jin and Xue Fan.

Authors and Affiliations

School of Physics and Astronomy, Shanghai Jiao Tong University, Shanghai, 200240, China

Sha Jin, Yuanxi Yu, Liang Hong & Matteo Baggioli

Wilczek Quantum Center, Shanghai Jiao Tong University, Shanghai, 200240, China

Sha Jin & Matteo Baggioli

Shanghai Research Center for Quantum Sciences, Shanghai, 201315, China

Shanghai National Center for Applied Mathematics, Shanghai Jiao Tong University, Shanghai, 200240, China

Xue Fan & Liang Hong

Materials Genome Institute, Shanghai University, Shanghai, 200444, China

School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, GA, 30332, USA

Institute for Superconducting and Electronic Materials, University of Wollongong, Wollongong, NSW, 2500, Australia

Caleb Stamper

The Australian Nuclear Science and Technology Organisation, Lucas Heights, NSW, 2232, Australia

Caleb Stamper, Richard A. Mole & Dehong Yu

Institute of Natural Sciences, Shanghai Jiao Tong University, Shanghai, 200240, China

You can also search for this author in PubMed   Google Scholar

Contributions

D.Y., C.S. and R.M. performed the experimental measurements; M.B., D.Y conceived the idea of this work with the help of L.H.; S.J, X.F. implemented the MD simulations and the INM analysis; S.J., X.F., Y.Y. and C.S. performed the analysis of the experimental and simulation data; M.B. and S.J. wrote the manuscript with the help of D.Y.

Corresponding authors

Correspondence to Dehong Yu or Matteo Baggioli .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Information.

Rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Jin, S., Fan, X., Stamper, C. et al. On the temperature dependence of the density of states of liquids at low energies. Sci Rep 14 , 18805 (2024). https://doi.org/10.1038/s41598-024-69504-2

Download citation

Received : 24 March 2024

Accepted : 06 August 2024

Published : 13 August 2024

DOI : https://doi.org/10.1038/s41598-024-69504-2

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

share this!

August 14, 2024

This article has been reviewed according to Science X's editorial process and policies . Editors have highlighted the following attributes while ensuring the content's credibility:

fact-checked

trusted source

Researchers discover new way to purify liquid argon for neutrino experiments

by Emily Driehaus, Fermi National Accelerator Laboratory

Construction workers have finished the excavation of the huge caverns that will house the international Deep Underground Neutrino Experiment. While engineers and technicians are preparing for the installation of the gigantic neutrino detectors into these caverns a mile underground, scientists around the world are working to optimize DUNE's particle detector technology.

From new photon detection systems to improved detector designs, researchers are refining devices and technologies to make DUNE's neutrino detectors—liquid- argon time projection chambers—the best they can be for the most precise neutrino measurements.

The purity of the liquid argon inside the DUNE cryostats, which is crucial to observing particles and light created by neutrino interactions in the liquid, might get an upgrade too. An interdisciplinary team of researchers in Brazil discovered that a filter media typically used in industrial applications can filter out nitrogen contamination in liquid argon. Future large-scale tests will help determine whether this promising method might be applicable for DUNE.

"We started with the goal of finding new materials that could capture oxygen and water in a more efficient way," said lead researcher Pascoal Pagliuso, a physics professor at "Gleb Wataghin" Institute of Physics, Unicamp in Campinas-São Paulo. "We decided to try and find a way to capture nitrogen, too. And we succeeded."

Neutrinos are the most abundant matter particles in the universe, yet they are hard to detect. For DUNE scientists to observe some of the neutrinos flying through their gigantic particle detectors, the neutrinos must interact with something.

In the case of DUNE, hosted by the U.S. Department of Energy's Fermi National Accelerator Laboratory, neutrinos will collide with argon atoms. The process creates secondary particles that knock loose electrons and emit brief flashes of scintillation light.

Instruments can record the electrons and light so that scientists can identify and reconstruct the particles and tracks that the collisions produce. However, for this measurement to happen, the particles need to have a clear, unobstructed path through the liquid argon all the way to the detection devices. In particular, the liquid argon has to be ultra-pure and contain few atoms of other elements that could absorb electrons or light.

A tricky challenge

Researchers and engineers ensure that the liquid argon in the detectors is as pure as it can be by filtering out contaminants like water and oxygen. These are two of the most common impurities. However, there's a third contaminant that is common: nitrogen. While neutrino researchers have well-established methods to filter out water and oxygen, reducing nitrogen levels below the levels provided by commercial providers has been a challenge.

Nitrogen can have a significant impact on the results of experiments—up to 20% of scintillation light can be lost even with just one part per million of nitrogen present in liquid argon. With ambitious experiments such as DUNE, ensuring the quality of all detector components and materials to produce the best results is critical to finding out more about neutrinos and their role in the subatomic world.

Fermilab currently uses a molecular sieve and a copper material to filter out water and oxygen, respectively, but neither can capture nitrogen from liquid argon.

Led by Pagliuso, researchers in Brazil discovered a way to reduce even small contamination levels of nitrogen in liquid argon. His interdisciplinary team of physicists and engineers found a material that removes both nitrogen and water.

Combined with a filter media like the copper material used by Fermilab, the media can remove the three most common liquid-argon contaminants, ensuring the argon is as pure as possible for neutrino experiments.

The material is known as Lithium-FAU, a Faujasite LiX zeolite. This type of aluminosilicate material has industrial applications in refining petroleum and purifying air. The Brazilian team discovered it also has the ability to remove nitrogen from liquid argon through adsorption. "It's like when you have a medicine for one disease and discover that it also works for another disease," Pagliuso said.

When zeolites are used as adsorbents to refine or purify liquids and gases, they attract particular particles that will stick to their surface while allowing others to pass through the crystalline structure. It's the concept that applies when putting packets of silica gel in new shoes to capture moisture: water clings to the surface of the silica gel beads so that the humidity won't damage the shoes.

In this case, the nitrogen molecules interact with the positively charged ions in the zeolite; the size of the lithium molecules is small enough to leave room for the nitrogen to be captured and for the liquid argon to flow free of contaminants.

The chemical engineering branch of the research team at Unicamp developed simulations to predict how nitrogen would be adsorbed by Li-FAU, giving DUNE experimenters the necessary framework for testing how the media performs in the specific environment of liquid argon in a cryostat.

"Predicting the behavior of the filter is important to determine the capacity of the filter," said Dirceu Noriler, a professor and Director of the School of Chemical Engineering at Unicamp. "We helped the engineers design the filter by specifying the saturation time and number of cycles needed to reach the required purity."

Successful tests

Researchers and engineers initially tested the media in the relatively small Liquid Argon Purification Cryostat at Unicamp with approximately 90 liters of liquid argon. Their successful results matched the simulations Noriler and his team had developed.

Further testing took place this past fall at the 3,000-liter ICEBERG test stand in Fermilab's Noble Liquid Test Facility, which was filled with 2,625 liters of liquid argon for this particular test. The results confirmed Li-FAU's ability to remove nitrogen from liquid argon at a larger scale.

"The Noble Liquid Test Facility here at Fermilab supports all sorts of liquid argon detector R&D, and we were very happy with the results. We intend to add Li-FAU to the facility as soon as possible so that, in addition to DUNE, all projects working on improving light collection systems can benefit," said Flor de María Blaszczyk, R&D coordinator and test facility manager at Fermilab.

Not only will the purification media improve the quality of the liquid argon that Fermilab uses for experiments, but it would also allow for the removal of nitrogen contaminants if air would accidentally be introduced into the cryostat due to the malfunction of equipment. Nitrogen is the most abundant element in air, so knowing how to filter it out will be crucial for ensuring DUNE and other experiments won't be compromised.

The next step is working to scale up the testing with larger volumes of liquid argon to ensure the media performs as well as it has so far. Scientists hope that ultimately the method will be capable of removing nitrogen at the large scale required for DUNE, which will feature detector modules that each contain 17,500 tons of liquid argon.

Provided by Fermi National Accelerator Laboratory

Explore further

Feedback to editors

Evidence stacks up for poisonous books containing toxic dyes

17 hours ago

Researchers develop an instant version of trendy, golden turmeric milk

Saturday Citations: Citizen scientists observe fast thing; controlling rat populations; clearing nanoplastic from water

Aug 17, 2024

New AI tool captures how proteins behave in context

Scientists discover phenomenon impacting Earth's radiation belts

Geophysicists find link between seismic waves called PKP precursors and strange anomalies in Earth's mantle

New twist on synthesis technique promises sustainable manufacturing

Researchers discover smarter way to recycle polyurethane

Aug 16, 2024

DNA study challenges thinking on ancestry of people in Japan

A visionary approach: How a team developed accessible maps for colorblind scientists

Relevant physicsforums posts, why is the speed of light the same everywhere in the universe.

4 hours ago

Magnetic Resonance physics question: RF pulse role and 'in-phase'

8 hours ago

Bruker TopSpin import function in pulse program

Aug 8, 2024

Energy production that converts Hydrogen to Iron?

Aug 6, 2024

Free carrier density and the electron momentum relaxation time

Aug 5, 2024

What materials, if any, are transparent to Long Wave IR?

Aug 3, 2024

More from Other Physics Topics

Related Stories

ProtoDUNE's argon filling underway, a significant step toward next era of neutrino research

Apr 12, 2024

Scientists observe first neutrinos with prototype detector

Aug 12, 2024

Tests start at CERN for large-scale prototype of new technology to detect neutrinos

Oct 10, 2019

CERN to provide second DUNE cryostat

Aug 18, 2021

Identifying lower-energy neutrinos with a liquid-argon particle detector

Jan 9, 2019

They are there and they are gone: ICARUS chases a fourth neutrino

Mar 3, 2020

Recommended for you

Photon entanglement could explain the rapid brain signals behind consciousness

Large Hadron Collider pipe brings search for elusive magnetic monopole closer than ever

Crashes, blackouts and climate tipping points: How can we tell when a system is close to the edge?

NASA demonstrates 'ultra-cool' quantum sensor for first time in space

Aug 13, 2024

'Mirror' nuclei help connect nuclear theory and neutron stars

Align or die: Revealing unknown mechanism essential for bacterial cell division

Let us know if there is a problem with our content.

Use this form if you have come across a typo, inaccuracy or would like to send an edit request for the content on this page. For general inquiries, please use our contact form . For general feedback, use the public comments section below (please adhere to guidelines ).

Please select the most appropriate category to facilitate processing of your request

Thank you for taking time to provide your feedback to the editors.

Your feedback is important to us. However, we do not guarantee individual replies due to the high volume of messages.

E-mail the story

Your email address is used only to let the recipient know who sent the email. Neither your address nor the recipient's address will be used for any other purpose. The information you enter will appear in your e-mail message and is not retained by Phys.org in any form.

Newsletter sign up

Get weekly and/or daily updates delivered to your inbox. You can unsubscribe at any time and we'll never share your details to third parties.

More information Privacy policy

Donate and enjoy an ad-free experience

We keep our content available to everyone. Consider supporting Science X's mission by getting a premium account.

E-mail newsletter