Polyamorphism in Glassy Network Materials Preprint: AIP/123-QED
M. H. Brown, P. G. Wolynes · 2026
Read Polyamorphism in Glassy Network Materials Preprint: AIP/123-QED by M. H. Brown, P. G. Wolynes (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-2605-04273.txt.
Introduction
Max Hall-Brown and Peter Guy Wolynes model polyamorphism, the ability of one material to form more than one distinct liquid phase. In network liquids such as water, the effect comes from competition between bonded and non-bonded local arrangements.
Their simple microscopic model calculates both glassy and thermodynamic behaviour directly from the underlying interactions. It contains a liquid-liquid transition that can be placed above, near, or below the glass transition by tuning the parameters. The work aims to explain related oddities such as water's density peak, while avoiding the freezing problems that make supercooled water hard to study directly.
Catalog record
| CREATOR | M. H. Brown, P. G. Wolynes |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 108 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.04273
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Polyamorphism in Glassy Network Materials Preprint: AIP/123-QED M. H. Brown, P. G. Wolynes Abstract One dramatic feature of network liquids is the emergence at low temperatures and high pressures of polyamorphism, where multiple distinct liquid phases are accessed in a single material. Polyamorphism can arise from the competition between distinct local inherent structures corresponding to bonded and nonbonded ordering. Thermal bond breaking thus can lead to a phase transition often accompanied by thermodynamic anomalies away from the transition itself, such as the familiar density maximum in water at atmospheric pressure and 4 ∘ 4^{\circ} C. Water exhibits network interactions in the form of hydrogen bonding between water molecules. The polyamorphic transition in water, however, is difficult to study due to the rapid crystallization of supercooled water and due to glassy effects at low temperatures. In the present work, we propose a simple microscopic model where the glassy and thermodynamic properties are both calculated directly from the microscopic potentials. The model contains a liquid-liquid phase transition, which, after tuning the microscopic parameters, may be located either above, near, or below the glass transition. By applying the Random First Order Transition theory of the glass transition to this simple microscopic model, we shine light on the interplay of polyamorphism and glassy properties in network liquids. We show the connection between the thermodynamic water-like anomalies and corresponding anomalies in the glassy kinetics. The analysis unveils key details on the way glassy dynamics modifies the phase transition kinetics. When the parameters of the model are tuned to produce a phase diagram resembling that of water, the liquid-liquid phase transformation near T g T_{g} occurs via “nanonucleation”, resulting in extremely small domains sizes and nonclassical nucleation kinetics which are predicted from the RFOT theory. I Introduction Network forming materials display a rich variety of macroscopic forms. Much of minerology’s complexity, for example, arises from the many distinct crystalline forms, called polymorphs, of silicates. Technology relies on the many solid yet nonperiodic forms of glasses that arise when network liquids are rapidly cooled, avoiding the nucleation and growth of crystals. Unlike crystalline polymorphs, the detailed properties of such glasses depend on their history of preparation. Network formers in their fully liquid forms also display sometimes puzzling thermodynamic patterns that suggest they may be able to form structurally and thermodynamically distinct dense fluid phases. Such fluid phases have been called “polyamorphs”. The most notorious example of such a possibility is water 36 ; 19 . The well known density maximum of water at 4 ∘ C early on encouraged the idea of there being two distinct forms of liquid water which could coexist or interconvert 49 , although other scenarios were also proposed 57 ; 54 . Extrapolating these data indicated that thermodynamic polyamorphism might occur in a deeply supercooled regime. Unfortunately, in the laboratory, crystallization is hard to avoid when liquid water is cooled to the suggested temperatures. Glassy amorphous solid forms of water are, however, easily accessed by other routes, both in the laboratory 11 ; 44 and the cosmos 26 . The relation of these glassy forms to a thermodynamic polyamorphic transition raises interesting questions, that, in this paper, we will try to address in the context of a tunable model of network materials that we introduced recently to describe network glasses within the random first order transition theory framework 10 . Indeed, our own interest in the question of polyamorphism arose from our noting that there seemed to be modest deviations in the experimental data from our simple model predictions of the way the fragility and glass transition temperature vary with composition in doped silica and germania glasses. We ascribed these deviations, in part, to thermodynamic variations of the configurational entropy like those known in water. We note, of course, that since many glasses used in technology are mixtures, possibilities of distinct multiphase glass behavior abound, simply through compositional phase separation. This possibility was exploited in the invention of Vycor glass, decades ago 77 . Both discrete polyamorphism of the type imputed to water and the glass transition are encouraged by the low configurational entropy of network materials which through frustration between different sources of stability allows distinct local structures to compete with each other 65 . This propinquity between polyamorphs and glasses encouraged the frustration limited domain theory put forward by Kivelson et al. 32 . To a certain extent one can imagine the RFOT’s random energy landscape, described globally by one-step replica symmetry breaking in the mean field limit 31 ; 30 , as corresponding to the extreme limit of there being a near infinite, or at least very large, number of polyamorphs. In this paper we will explore the interplay of a single discrete polyamorphic transition and the glass transition in a simple microscopic model of a one component network material. In addition to exploring the phase diagrams of the models, we will explore various regimes of nucleation mechanisms of the polyamorphs which are produced when they occur in the proximity to the glassy dynamics. As noted earlier by Stevenson and Wolynes 61 , the closeness of a first order transition to a glass transition allows discrete polyamorphs to interconvert through a non-classical mechanism called “nanopercolative nucleation” and we see this is true for the present model when it is tuned to resemble water. The organization of the paper is as follows. We first propose in section II a simple model for a fluid with two distinct liquid phases which may be analyzed both in terms of a liquid-liquid phase transition and in terms of the glass transition based on the same microscopic potentials. We review the approximation methods used to find the equilibrium liquid phase diagram, along with the physics governing the glass transition in the RFOT theory, where the behavior of individual glassy states is computed in the self-consistent phonon theory. In section III several of the main features which characterize the behavior of the model, including the presence of liquid-gas and liquid-liquid phase transitions, water-like anomalies, and glass transitions at low temperatures are described. In section IV , we tune several of the key parameters of the model corresponding to the strengths of certain microscopic interactions in order to study how the phase diagram varies depending on the microscopic potentials. We then study the phase diagrams and glassy landmarks for several of these parameter choices in more detail in section V , in order to see the behavior of three representative liquids with varying proximity between the liquid-liquid critical point and the glass transition. Last, in VI , we discuss nucleation processes for the transformation of one liquid to another within our model. II Methods II.1 Model We begin by considering a liquid made up of N identical particles which may freely transform between two states: a “bonding" state, and a “non-bonding" state. These two states are analogous to the “up" and “down" spin states of the model of polyamorphs studied by Lee and Swendsen 35 . A bonding particle has a larger diameter d_{b} than does a non-bonding particle which has a smaller diameter d_{u} which allows the nonbonding particle to enter the solvation shell of a bonded molecule. At any moment in time, some fraction