Dynamic heterogeneity in sodium silicate melts via machine-learning potential

Kumpei Shiraishi, Rikuta Nozawa, Emi Minamitani · 2026

Read Dynamic heterogeneity in sodium silicate melts via machine-learning potential by Kumpei Shiraishi, Rikuta Nozawa, Emi Minamitani (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-26646.txt.

Introduction

Kumpei Shiraishi, Rikuta Nozawa, and Emi Minamitani characterise uneven motion inside molten sodium silicate glass. In a melt, some atoms move far while neighbours barely stir. That unevenness is called dynamic heterogeneity. The team simulated three sodium silicate recipes with machine-learning potentials, accurate learned models of atomic forces. Temperatures varied across the runs.

Motion was tracked for up to a billionth of a second in fine detail. Sodium ions moved in discrete hops rather than smooth drift. That hopping lets sodium escape the slow relaxation of the surrounding silicate framework. A standard unevenness measure was computed for every species. Oxygen atoms, part of the framework itself, showed the strongest uneven behaviour. Their motion comes in rare, sudden rearrangements. Mobile ions, by contrast, displace frequently in small steps.

Catalog record

CREATORKumpei Shiraishi, Rikuta Nozawa, Emi Minamitani
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE41 KB
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2606.26646

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Dynamic heterogeneity in sodium silicate melts via machine-learning potential Kumpei Shiraishi, Rikuta Nozawa, Emi Minamitani Abstract We present a comprehensive characterisation of dynamic heterogeneity in sodium silicate melts using molecular dynamics simulation with machine-learning potentials. By studying sodium disilicate, tetrasilicate, and hexasilicate melts across a range of temperatures, mean squared displacement and a time-correlation function computed up to the nanosecond timescale provide a detailed account of how spatial mobility disparities emerge in a realistic multicomponent oxide glass. Within these timescales, the self-part of the van Hove function for sodium displays a bimodality, demonstrating that alkali transport is mediated by discrete displacement events consistent with a hopping mechanism. This distinct hopping allows sodium ions to decouple from the sluggish relaxation of the silicate matrix. Furthermore, evaluation of the non-Gaussian parameter reveals that, although all constituent species exhibit dynamic heterogeneity, the non-Gaussian behaviour is most pronounced for oxygen atoms. This trend reflects the intermittency of structural rearrangements, where framework atoms undergo rare and stochastic events compared to the frequent displacements of mobile ions. Our findings elucidate the microscopic mechanism of ion transport and its connection to dynamic heterogeneity in silicate melts, offering a new avenue to study fundamental glassy physics in realistic vitreous materials. \usephysicsmodule ab I Introduction Ion diffusion in alkali-doped silica glasses is a fundamental problem of broad relevance, spanning chemical physics, materials science, and geology Angell et al. 2000 ; Greaves and Sen 2007 ; Micoulaut and Bauchy 2022 . Its importance has been further highlighted by applications in advanced industrial products such as all-solid-state secondary batteries, making it a timely and active area of research Grady et al. 2020 , where precise control of ion transport is paramount. The mechanism by which alkali ions relax within the disordered silicate network is a key factor governing the macroscopic properties of these materials. At the heart of this context is the preferential pathway picture, which posits that alkali ions do not diffuse uniformly through the \text{SiO}{\vphantom{\text{X}}}_{\smash[t]{\text{2}}} matrix but instead migrate through spatially distinct, interconnected channels that form within the network Greaves 1985 . To date, numerous studies have focused on verifying the static structural aspects of this framework, observing structural changes that support this depiction as sodium density increases Greaves and Ngai 1995 ; Meyer et al. 2002 ; Meyer et al. 2004 . However, since ion transport is inherently a dynamical phenomenon, there is a pressing need to characterise these processes from the perspective of relaxation dynamics through direct observation via computer simulations Horbach et al. 2001 ; Jund et al. 2001 ; Horbach et al. 2002 ; Sørensen et al. 2023 . Such a characterisation is most naturally pursued within the broader framework of glass physics, where relaxation dynamics have long been a central concern Ediger et al. 1996 ; Debenedetti and Stillinger 2001 ; Berthier and Biroli 2011 . In supercooled liquids approaching the glass transition, particles of high and low mobility coexist and fluctuate spatially, a phenomenon known as dynamic heterogeneity Ediger 2000 . These dynamical fluctuations grow as the system cools and relaxation slows, giving rise to hallmark features of the glass transition such as stretched exponential decay of time correlation functions Kob and Andersen 1994 ; Kob and Andersen 1995a ; Kob and Andersen 1995b , exponential distribution tails of single-particle displacements Chaudhuri et al. 2007 , and four-point susceptibility peaks Lačević et al. 2003 ; Berthier et al. 2005 ; Toninelli et al. 2005 . Moreover, dynamic heterogeneity is not merely of theoretical interest; it governs transport properties, including macroscopic diffusion Cicerone and Ediger 1996 ; Sengupta et al. 2013 ; Kawasaki and Kim 2017 and ionic conductivity Horbach and Kob 1999 ; Habasaki and Ngai 2007 , and is therefore of direct relevance to the ion migration picture outlined above. Nevertheless, detailed insights into dynamic heterogeneity have been predominantly obtained from packing-type glass models, in which interparticle interactions are isotropic and short-ranged Kob and Andersen 1995a ; Kob et al. 1997 ; Donati et al. 1998 ; Weeks et al. 2000 . The correlation between this phenomenon and ion diffusion in network glasses such as sodium silicate, where directional covalent bonding gives rise to qualitatively richer local environments, remains to be fully elucidated. It is therefore essential to accurately capture individual particle dynamics within the diverse local environments that are far more pronounced in network glasses, typified by sodium silicate, than in simple packing-type systems. From this perspective, it is crucial to go beyond simple empirical potentials Horbach et al. 2001 ; Horbach et al. 2002 ; Meyer et al. 2004 and track relaxation with the precision of first-principles calculations that incorporate the underlying electronic states. However, as is well known, first-principles calculations are computationally intensive, making it increasingly impractical to cover the growing relaxation times near the glass transition as the temperature decreases. More critically, the computational cost of first-principles methods scales as O(N^{3}) with respect to the number of particles. This scaling is fundamentally at odds with the requirement for large system sizes necessary to capture dynamic heterogeneity in glass transition research. In this study, we overcome this computational barrier by utilising machine learning potentials, which have undergone remarkable methodological advances Behler and Parrinello 2007 ; Deringer et al. 2019 . Such potentials have also enabled detailed and transferable atomistic modelling of the silicon-oxygen system, spanning crystalline, amorphous, and nanostructured environments Erhard et al. 2022 ; Erhard et al. 2024 , including alkali silicate glasses closely related to the compositions studied here Bertani et al. 2024 ; Pedone et al. 2025 ; Ganisetti et al. 2025 . Whereas these prior works on alkali silicate glasses have focused primarily on establishing the accuracy and transferability of machine-learning force fields through structural validation, the present study extends this approach to the long-time dynamical regime. Specifically, through the construction of a machine learning potential for sodium silicate and a molecular dynamics (MD) code developed in-house and optimised for graphics processing unit execution, we have succeeded in dynamically tracking the relaxation of both the frozen silica network and the diffusive sodium ions over extended spatial and temporal scales previously inaccessible to first-principles methods. This computational strategy, which leverages the architectural characteristics of modern hardware, enables the application of analytical methods developed in fundamental glass transition research to realistic sodium silicate systems. We provide compelling evidence for the preferential pathway picture from a dynamical perspective, supported by complementary structural evidence presented in the Appendix, establishing a new foundation for the study of realistic glass systems. II Methods The machine-learning interatomic potential (MLIP) was constructed using a PaiNN-type equivariant graph neural network Schütt et al. 2021 . Atomic features were embedded in a 60-dimensional latent space, and interatomic interactions were evaluated within a cutoff radius of { 5.0 } using 40 Gaussian radial basis functions combined with a smoothstep cutoff envelope. The network architecture comprised two interaction blocks. The model parameters were optimised by minimising a weighted loss function incorporating total energies, atomic forces, and cell virials: \displaystyle\mathcal{L}=w_{E}\mathcal{L}_{E}+w_{F}\mathcal{L}_{F}+w_{\Xi}\mathcal{L}_{\Xi}, (1) where the weighting factors were set to w_{E}=0.01 , w_{F}=0.99 , and w_{\Xi}=0.01 . The energy, force, and cell-virial contributions were evaluated as mean squared errors relative to the reference first-principles values. To ensure high-fidelity MD trajectories,

Read the complete text