arxiv: 2605.02679 · v1 · submitted 2026-05-04 · ❄️ cond-mat.dis-nn · cond-mat.soft· cond-mat.stat-mech

Recognition: unknown

Computational Methods towards Ultrastable Glasses

Fabio Leoni , Misaki Ozawa , John Russo , Taiki Yanagishima , Andrea Ninarello

Authors on Pith no claims yet

Pith reviewed 2026-05-08 01:55 UTC · model grok-4.3

classification ❄️ cond-mat.dis-nn cond-mat.softcond-mat.stat-mech

keywords ultrastable glassescomputational methodsglass transitionnonequilibrium statesamorphous solidsunphysical movesstability analysissimulation algorithms

0 comments

The pith

Numerical methods using unphysical moves reach glassy states with stabilities beyond any conventional cooling protocol.

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

The paper reviews computational algorithms that deliberately incorporate unphysical particle moves to prepare ultrastable glasses, amorphous solids that sit in unusually deep energy minima. These techniques bypass the severe slowing down that limits ordinary molecular-dynamics cooling and thereby produce configurations whose kinetic, thermodynamic and mechanical properties exceed those of glasses made by any realistic thermal quench. The authors describe the algorithmic steps, efficiency limits and physical meaning of each method, then compare the stabilities achieved across the literature so that readers can see which routes currently deliver the deepest states.

Core claim

By cataloguing the main families of unphysical-move algorithms and performing a comparative stability analysis, the review establishes that multiple independent protocols can generate nonequilibrium amorphous configurations whose energies and relaxation times lie well below those accessible by physical cooling at any computationally feasible rate, thereby furnishing new benchmarks for the glass transition and for the design of mechanically robust glassy materials.

What carries the argument

The central mechanism is the controlled insertion of unphysical moves, such as particle swaps or landscape-hopping steps, inside Monte Carlo or molecular-dynamics schemes to escape kinetic traps and sample deeper regions of the potential-energy landscape.

If this is right

These protocols supply numerical data on glassy states closer to the ideal glass than conventional simulations allow.
Direct stability comparisons let researchers choose the most efficient method for a given question or material.
Physical interpretations of the moves suggest new experimental protocols for preparing ultrastable glasses.
The resulting low-energy configurations provide sharper tests for theories of the glass transition.

Where Pith is reading between the lines

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

If the same ranking of methods holds across different interaction potentials, the stability gains are largely universal rather than model-specific.
Coupling the unphysical moves to machine-learned potentials could extend the approach to chemically complex glasses.
Measuring the mechanical response of the simulated glasses at larger system sizes would show whether the stability improvements translate to macroscopic toughness.

Load-bearing premise

The algorithms and stability metrics drawn from the cited literature are accurately represented and can be compared directly without hidden differences in model or protocol.

What would settle it

A single simulation in the same model system showing that an extremely slow, physically realistic cooling run reaches an inherent-structure energy or yield stress equal to or better than the lowest value reported for any unphysical-move protocol.

Figures

Figures reproduced from arXiv: 2605.02679 by Andrea Ninarello, Fabio Leoni, John Russo, Misaki Ozawa, Taiki Yanagishima.

**Figure 1.** Figure 1: FIG. 1. (a) Kinetic ( view at source ↗

**Figure 2.** Figure 2: FIG. 2. Schematic representation of PVD. Blue and red are view at source ↗

**Figure 3.** Figure 3: FIG. 3. Schematic illustration of cyclic shear applied to a view at source ↗

**Figure 4.** Figure 4: FIG. 4. Schematic representation of the Swap Monte-Carlo view at source ↗

**Figure 5.** Figure 5: FIG. 5. Structural optimization procedure combining stress view at source ↗

**Figure 6.** Figure 6: FIG. 6. Schematic representation of lifted Monte-Carlo dy view at source ↗

**Figure 7.** Figure 7: FIG. 7. Starting from an equilibrium liquid configuration view at source ↗

**Figure 8.** Figure 8: FIG. 8. Starting from an equilibrium liquid configuration view at source ↗

**Figure 9.** Figure 9: FIG. 9. Schematic representation of the Parallel Tempering view at source ↗

**Figure 10.** Figure 10: FIG. 10. Schematic representation of trajectory sampling in view at source ↗

**Figure 11.** Figure 11: FIG. 11. Three schematic examples of ML algorithms used view at source ↗

read the original abstract

Ultrastable glasses, amorphous solids with exceptionally low-energy states and enhanced kinetic, thermodynamic and mechanical stability, have long been a subject of intense experimental interest. Over the past decade, their computational realization has emerged as a major goal in condensed matter physics, as numerical methods can exploit unphysical moves to access deeply supercooled and nonequilibrium glassy states far beyond the reach of conventional cooling protocols, thereby providing key insights into the nature of the glass transition and amorphous states and enabling the design of mechanically robust glassy materials. In this review, we outline the key steps underlying the most effective algorithms developed across the field. For each approach, we discuss its efficiency, limitations, and physical interpretation. We finally present a comparative analysis of the stability achieved across these methods, with the aim of equipping both newcomers and experts with an intuitive and comprehensive understanding of the field's current state and the opportunities it presents.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note private letter to a colleague

This review organizes existing computational methods for ultrastable glasses and gives a decent map of the options, but the comparative stability analysis risks inconsistency from unnormalized metrics across cited studies.

read the letter

The main thing to know is that this is a review paper summarizing computational approaches to ultrastable glasses rather than introducing new algorithms or data. It walks through the key steps of the main methods, their efficiency, limitations, and physical meaning, then ends with a side-by-side look at achieved stabilities. That structure can save time for someone entering the area or trying to choose a technique. The authors clearly put effort into pulling the literature together and framing what each approach offers. The comparative section is the part that aims to show which routes reach deeper states than ordinary cooling. The soft spot is exactly the one the stress test flags. Stability gets measured in different ways across the original papers, from fictive temperature to inherent-structure energy to yield stress, and the underlying models, sizes, and protocols differ too. If the review simply reports the published numbers without converting them to a common reference or explicitly discussing those differences, the claim that certain methods go far beyond conventional protocols rests on weaker footing than it appears. The abstract presents the comparison as a strength, but the execution needs to be checked for how thoroughly it handles the inconsistencies. This paper is for condensed-matter researchers and materials people who work on glasses and want a consolidated view of the computational toolkit. Newcomers would get the most immediate value. It deserves peer review because a careful review that accurately summarizes the methods and flags the comparison issues can still help coordinate work in the field, even without new results.

Referee Report

1 major / 1 minor

Summary. This review outlines computational methods for realizing ultrastable glasses by exploiting unphysical moves to access deeply supercooled and nonequilibrium states beyond conventional cooling protocols. For each algorithm, it discusses efficiency, limitations, and physical interpretation, and concludes with a comparative analysis of achieved stabilities to provide insights into the glass transition and material design.

Significance. If the comparative analysis is made rigorous, the review would offer significant value by synthesizing algorithmic advances in the field, equipping researchers with guidance on method selection, and highlighting opportunities for understanding amorphous states and designing robust glasses. The manuscript's strength is its broad coverage of the literature on these specialized simulation techniques.

major comments (1)

[Comparative analysis section] Comparative analysis section: the review tabulates stability values (fictive temperature, inherent-structure energy, mechanical yield stress) drawn from cited works that employ non-equivalent interaction potentials, system sizes, and cooling rates, without conversion to a common reference scale or explicit discussion of these differences. This prevents a rigorous demonstration that the reviewed methods reach states 'far beyond' conventional cooling and weakens support for the claims of key insights and design enablement.

minor comments (1)

[Physical interpretation subsections] Ensure that the physical interpretation subsections consistently distinguish between equilibrium and nonequilibrium aspects of the accessed states.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive feedback on the comparative analysis. We address the concern below and will revise the manuscript to strengthen the discussion of comparability across studies.

read point-by-point responses

Referee: [Comparative analysis section] Comparative analysis section: the review tabulates stability values (fictive temperature, inherent-structure energy, mechanical yield stress) drawn from cited works that employ non-equivalent interaction potentials, system sizes, and cooling rates, without conversion to a common reference scale or explicit discussion of these differences. This prevents a rigorous demonstration that the reviewed methods reach states 'far beyond' conventional cooling and weakens support for the claims of key insights and design enablement.

Authors: We acknowledge that the tabulated values come from studies using different interaction potentials, system sizes, and protocols, which complicates direct quantitative comparison. The current manuscript notes some of these variations in the text surrounding the table but does not provide an extended dedicated discussion or attempt normalization. In the revised version we will add an explicit subsection on the challenges of cross-study comparison, including the difficulty of mapping between potentials, and we will emphasize relative improvements (e.g., stability gain versus each study's own conventional-cooling reference) rather than absolute values. Where literature data permit simple rescaling (for example, reporting fictive temperatures in units of the model's own Tg), we will include such normalized entries. Full conversion to a single universal scale is not feasible within a review without performing new simulations, but the added discussion will make the limitations transparent and better support the claims of methods reaching states beyond conventional cooling. revision: partial

Circularity Check

0 steps flagged

Review paper summarizes external literature with no internal derivations or self-referential reductions.

full rationale

The manuscript is a review article that outlines algorithms from prior works, discusses their efficiency and limitations, and presents a comparative analysis of stability metrics drawn from the cited literature. No original derivations, fitted parameters renamed as predictions, self-definitional steps, or load-bearing self-citations that reduce the central claims to the paper's own inputs are present. The comparative stability discussion relies on external sources without introducing circular equivalences by construction. This matches the default expectation for non-circular review papers.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

As a review, the paper draws on prior literature for all technical content and introduces no new free parameters, axioms, or invented entities of its own.

pith-pipeline@v0.9.0 · 5464 in / 970 out tokens · 26239 ms · 2026-05-08T01:55:01.679927+00:00 · methodology

discussion (0)

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

Identifying the relevant parameters in design strategies for stable glasses
cond-mat.stat-mech 2026-05 unverdicted novelty 7.0

Optimizing hyperuniformity and local ordering without changing particle diameters produces no stability gain, showing that diameter dynamics drives ultrastability rather than the optimized quantities.

Reference graph

Works this paper leans on

300 extracted references · 29 canonical work pages · cited by 1 Pith paper

[1]

Cooling-rate dependence of kinetic and mechanical stabilities of simulated glasses , author=. J. Chem. Phys. , volume=. 2015 , publisher=

2015
[2]

Hybrid Monte Carlo simulation of a glass-forming binary mixture , author=. Phys. Rev. E , volume=. 2006 , publisher=

2006
[3]

Evaluation of the Disorder Temperature and Free-Volume Formalisms<? format?> via Simulations of Shear Banding in Amorphous Solids , author=. Phys. Rev. Lett. , volume=. 2007 , publisher=

2007
[4]

Crossover in dynamics in the Kob-Andersen binary mixture glass-forming liquid , author=. J. Non-Cryst. Solids X , volume=. 2022 , publisher=

2022
[5]

Inherent structure entropy of supercooled liquids , author=. Phys. Rev. Lett. , volume=. 1999 , publisher=

1999
[6]

Front-mediated melting of isotropic ultrastable glasses , author=. Phys. Rev. Lett. , volume=. 2019 , publisher=

2019
[7]

Replica Monte Carlo simulation of spin glasses , author=. Phys. Rev. Lett. , volume=
[8]

Replica-exchange molecular dynamics method for protein folding , author=. Chem. Phys. Lett. , volume=. 1999 , publisher=

1999
[9]

Spin Glasses and Random Fields , pages=

Numerical simulations of spin glass systems , author=. Spin Glasses and Random Fields , pages=. 1998 , publisher=

1998
[10]

On the approach to the equilibrium and the equilibrium properties of a glass-forming model , author=. J. Phys. A: Math. Gen. , volume=. 1998 , publisher=

1998
[11]

Replica-exchange molecular dynamics simulation for supercooled liquids , author=. Phys. Rev. E , volume=. 2000 , publisher=

2000
[12]

Equilibration times in numerical simulation of structural glasses: Comparing parallel tempering and conventional molecular dynamics , author=. Phys. Rev. E , volume=. 2002 , publisher=

2002
[13]

Point-to-set lengths, local structure, and glassiness , author=. Phys. Rev. E , volume=. 2016 , publisher=

2016
[14]

Efficient measurement of point-to-set correlations and overlap fluctuations in glass-forming liquids , author=. J. Chem. Phys. , volume=. 2016 , publisher=

2016
[15]

Efficient swap algorithms for molecular dynamics simulations of equilibrium supercooled liquids , author=. J. Stat. Mech.: Theory Exp. , volume=. 2019 , publisher=

2019
[16]

Structural order as a genuine control parameter of dynamics in simple glass formers , volume =

Hua Tong and Hajime Tanaka , doi =. Structural order as a genuine control parameter of dynamics in simple glass formers , volume =. Nat. Commun. , month =
[17]

From ultra-fast growth to avalanche growth in devitrifying glasses , volume =

T Yanagishima and J Russo and R P A Dullens and H Tanaka , doi =. From ultra-fast growth to avalanche growth in devitrifying glasses , volume =. J. Chem. Phys. , month =
[18]

Common mechanism of thermodynamic and mechanical origin for ageing and crystallization of glasses , volume =

T Yanagishima and J Russo and H Tanaka , doi =. Common mechanism of thermodynamic and mechanical origin for ageing and crystallization of glasses , volume =. Nat. Commun. , pages =
[19]

Dale and James D

Jack R. Dale and James D. Sartor and R. Cameron Dennis and Eric I. Corwin , doi =. Hyperuniform jammed sphere packings have anomalous material properties , volume =. Phys. Rev. E , month =
[20]

Methodology to construct large realizations of perfectly hyperuniform disordered packings , volume =

J Kim and S Torquato , doi =. Methodology to construct large realizations of perfectly hyperuniform disordered packings , volume =. Phys. Rev. E , keywords =
[21]

Torquato, S. , doi =. Hyperuniform states of matter , volume =. Phys. Rep. , month =
[22]

B. D. Lubachevsky and F. H. Stillinger , doi =. Geometric properties of random disk packings , volume =. J. Stat. Phys. , keywords =
[23]

and Nöhring, W

Guénolé, J. and Nöhring, W. G. and Vaid, A. and Houllé, F. and Xie, Z. and Prakash, A. and Bitzek, E. , doi =. Comp. Mater. Sci. , title =
[24]

Hagh, V. F. and Nagel, S. R. and Liu, A. J. and Manning, M. L. and Corwin, E. I. , journal =. Transient learning degrees of freedom for introducing function in materials , year =. doi:10.1073/pnas , isbn =

work page doi:10.1073/pnas
[25]

Crystal nucleation in a vapor deposited Lennard-Jones mixture , author=. J. Mol. Liq. , volume=. 2023 , publisher=

2023
[26]

The physics of empty liquids: From patchy particles to water , author=. Rep. Prog. Phys. , volume=. 2022 , publisher=

2022
[27]

and Martelli, F

Leoni, F. and Martelli, F. and Russo, J. , title =. J. Phys. Chem. Lett. , volume =
[28]

Towards glasses with permanent stability , author=. Phys. Rev. Lett. , volume=. 2021 , publisher=

2021
[29]

Low-frequency vibrational modes of stable glasses , author=. Nat. Commun. , volume=. 2019 , publisher=

2019
[30]

Modern computational studies of the glass transition , author=. Nat. Rev. Phys. , volume=. 2023 , publisher=

2023
[31]

Equilibrium phase diagram of a randomly pinned glass-former , author=. Proc. Natl. Acad. Sci. U.S.A. , volume=. 2015 , publisher=

2015
[32]

Creating bulk ultrastable glasses by random particle bonding , author=. Nat. Commun. , volume=. 2023 , publisher=

2023
[33]

Europhys

Density controls the kinetic stability of ultrastable glasses , author=. Europhys. Lett. , volume=. 2017 , publisher=

2017
[34]

Theory for swap acceleration near the glass and jamming transitions for continuously polydisperse particles , author=. Phys. Rev. X , volume=. 2018 , publisher=

2018
[35]

Fast generation of ultrastable computer glasses by minimization of an augmented potential energy , author=. Phys. Rev. E , volume=. 2019 , publisher=

2019
[36]

Ideal Glass and Ideal Disk Packing in Two Dimensions , author =. Phys. Rev. Lett. , volume =. 2026 , month =. doi:10.1103/vldy-r77w , url =

work page doi:10.1103/vldy-r77w 2026
[37]

Ideal non-crystals as a distinct form of ordered states without symmetry breaking , author=. Nat. Mater. , pages=. 2026 , publisher=

2026
[38]

Inverse design of glass structure with deep graph neural networks , author=. Nat. Commun. , volume=. 2021 , publisher=

2021
[39]

Policy-guided Monte Carlo on general state spaces: Application to glass-forming mixtures , author=. J. Chem. Phys. , volume=. 2024 , publisher=

2024
[40]

Predicting dynamic heterogeneity in glass-forming liquids by physics-inspired machine learning , author=. Phys. Rev. Lett. , volume=. 2023 , publisher=

2023
[41]

Normalizing flows as an enhanced sampling method for atomistic supercooled liquids , author=. Mach. Learn.: Sci. Technol. , volume=. 2024 , publisher=

2024
[42]

Science , volume=

Boltzmann generators: Sampling equilibrium states of many-body systems with deep learning , author=. Science , volume=. 2019 , publisher=

2019
[43]

MRS bulletin , volume=

Topological engineering of glasses using temperature-dependent constraints , author=. MRS bulletin , volume=. 2017 , publisher=

2017
[44]

Accessing ultrastable glass via a bulk transformation , author=. Nat. Commun. , volume=. 2025 , publisher=

2025
[45]

Reentrant glass transition leading to ultrastable metallic glass , author=. Mater. Today , volume=. 2020 , publisher=

2020
[46]

Models and algorithms for the next generation of glass transition studies , author=. Phys. Rev. X , volume=. 2017 , publisher=

2017
[47]

Physical Review , volume=

Calculation of order parameters in a binary alloy by the Monte Carlo method , author=. Physical Review , volume=. 1959 , publisher=

1959
[48]

Surface Science , volume=

The structure and thermodynamics of binary microclusters: A Monte Carlo simulation , author=. Surface Science , volume=. 1978 , publisher=

1978
[49]

Equation of state for symmetric non-additive hard-sphere fluids: An approximate analytic expression and new Monte Carlo results , author=. Chem. Phys. Lett. , volume=. 1989 , publisher=

1989
[50]

Ultrastable metallic glasses in silico , author=. Phys. Rev. Lett. , volume=. 2020 , publisher=

2020
[51]

Stable glassy configurations of the Kob--Andersen model using swap Monte Carlo , author=. J. Chem. Phys. , volume=. 2020 , publisher=

2020
[52]

Fast Monte Carlo algorithm for supercooled soft spheres , author=. Phys. Rev. E , volume=. 2001 , publisher=

2001
[53]

The Journal of Physical Chemistry B , volume=

Numerical investigation of the entropy crisis in model glass formers , author=. The Journal of Physical Chemistry B , volume=. 2004 , publisher=

2004
[54]

Nature , volume=

Phonon interpretation of the ‘boson peak’in supercooled liquids , author=. Nature , volume=. 2003 , publisher=

2003
[55]

Nature Physics , volume=

Thermodynamic signature of growing amorphous order in glass-forming liquids , author=. Nature Physics , volume=. 2008 , publisher=

2008
[56]

Configurational entropy measurements in extremely supercooled liquids that break the glass ceiling , author=. Proc. Natl. Acad. Sci. U.S.A. , volume=. 2017 , publisher=

2017
[57]

Zero-temperature glass transition in two dimensions , author=. Nat. Commun. , volume=. 2019 , publisher=

2019
[58]

Absence of marginal stability in a structural glass , author=. Phys. Rev. Lett. , volume=. 2017 , publisher=

2017
[59]

Nature of excitations and defects in structural glasses , author=. Nat. Commun. , volume=. 2019 , publisher=

2019
[60]

Physical Review X , volume=

Hierarchical landscape of hard disk glasses , author=. Physical Review X , volume=. 2019 , publisher=

2019
[61]

Finite dimensional vestige of spinodal criticality above the dynamical glass transition , author=. Phys. Rev. Lett. , volume=. 2020 , publisher=

2020
[62]

Physical Review Research , volume=

Role of fluctuations in the yielding transition of two-dimensional glasses , author=. Physical Review Research , volume=. 2020 , publisher=

2020
[63]

Finite-size study of the athermal quasistatic yielding transition in structural glasses , author=. J. Chem. Phys. , volume=. 2021 , publisher=

2021
[64]

Rare events and disorder control the brittle yielding of well-annealed amorphous solids , author=. Phys. Rev. Res. , volume=. 2022 , publisher=

2022
[65]

Science advances , volume=

A stability-reversibility map unifies elasticity, plasticity, yielding, and jamming in hard sphere glasses , author=. Science advances , volume=. 2018 , publisher=

2018
[66]

Soft Matter , volume=

Spin-glass-like aging in colloidal and granular glasses , author=. Soft Matter , volume=. 2018 , publisher=

2018
[67]

Role of Short-Range Order and Hyperuniformity in the Formation of Band Gaps in Disordered Photonic Materials , author =. Phys. Rev. Lett. , volume =
[68]

Physical Review B , volume=

Kauzmann paradox: A possible crossover due to diminishing local excitations , author=. Physical Review B , volume=. 2023 , publisher=

2023
[69]

Depletion of two-level systems in ultrastable computer-generated glasses , author=. Phys. Rev. Lett. , volume=. 2020 , publisher=

2020
[70]

Soft matter , volume=

Sound attenuation in stable glasses , author=. Soft matter , volume=. 2019 , publisher=

2019
[71]

Microscopic analysis of sound attenuation in low-temperature amorphous solids reveals quantitative importance of non-affine effects , author=. J. Chem. Phys. , volume=. 2022 , publisher=

2022
[72]

Low-frequency excess vibrational modes in two-dimensional glasses , author=. Phys. Rev. Lett. , volume=. 2021 , publisher=

2021
[73]

Relation between heterogeneous frozen regions in supercooled liquids and non-Debye spectrum in the corresponding glasses , author=. Phys. Rev. Lett. , volume=. 2019 , publisher=

2019
[74]

Density of states below the first sound mode in 3D glasses , author=. J. Chem. Phys. , volume=. 2022 , publisher=

2022
[75]

The nature of non-phononic excitations in disordered systems , author=. Nat. Commun. , volume=. 2024 , publisher=

2024
[76]

Physical Review Materials , volume=

Relationship between two-level systems and quasilocalized normal modes in glasses , author=. Physical Review Materials , volume=. 2021 , publisher=

2021
[77]

SciPost Physics , volume=

Spatial structure of unstable normal modes in a glass-forming liquid , author=. SciPost Physics , volume=
[78]

Pinching a glass reveals key properties of its soft spots , author=. Proc. Natl. Acad. Sci. U.S.A. , volume=. 2020 , publisher=

2020
[79]

Nonlinear quasilocalized excitations in glasses: True representatives of soft spots , author=. Phys. Rev. E , volume=. 2020 , publisher=

2020
[80]

Shear transformations in metallic glasses without excessive and predefinable defects , author=. Proc. Natl. Acad. Sci. U.S.A. , volume=. 2022 , publisher=

2022

Showing first 80 references.