Apparent bistability from weak long-range interactions
Achilleas Lazarides, Andrea Pizzi · 2025
Read Apparent bistability from weak long-range interactions by Achilleas Lazarides, Andrea Pizzi (2025) 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-2506-10068.txt.
Introduction
Achilleas Lazarides and Andrea Pizzi study a subtle form of bistability in a 2025 physics paper. A system is bistable when two stable states can both last. A bias field can kill one state by seeding defects that grow and take over.
Long-range forces can block that takeover when they are strong. The weak range was murky. The authors show it gives apparent bistability: one state is doomed in theory but lasts for all practical purposes. The seed size needed to kill it grows so fast that such seeds almost never form. Math arguments and a cell model support the claim, which reframes stability in long-range systems.
Catalog record
| CREATOR | Achilleas Lazarides, Andrea Pizzi |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 25 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2506-10068
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Apparent bistability from weak long-range interactions Achilleas Lazarides, Andrea Pizzi Abstract Bistability, or the coexistence of two stable phases, can be broken by a bias field h h destabilising one of the phases via the nucleation and growth of defects. Strong long-range interactions, 1 / r α 1/r^{\alpha} with α \alpha less than the system’s dimensionality d d , can suppress the proliferation of defects and restore bistability. The case of weak long-range interactions d < α < d + 1 d<\alpha<d+1 remains instead poorly understood. Here, we show that it supports apparent bistability: While the system has in principle a unique stable phase, it appears bistable for all practical purposes for α < α c \alpha<\alpha_{c} , with α c > d \alpha_{c}>d behaving like a genuine critical point. At the core of this is an exponential scaling of the critical droplet size R c ∼ h − 1 / ( α − d ) R_{c}\sim h^{-1/(\alpha-d)} , which makes nucleating destabilizing droplets extremely unlikely for α < α c \alpha<\alpha_{c} , and such that α c \alpha_{c} is mostly independent of system size. In support of these conclusions we provide field-theoretical arguments and numerics on a probabilistic cellular automaton. Overall, our results offer a way to rethink phase stability in systems with long-range interactions as well as a new route to achieve practical bistability. Many physical systems can settle into one of two stable states, a property known as bistability Binder 1987 . This underlies familiar phenomena such as the coexistence of ice and water, the two magnetic orientations in a ferromagnet, filling and wetting phenomena at surfaces Bonn et al. 2009 , and the triple phase coexistence of ouzo Sibley et al. 2025 . Often, however, a small external bias field or perturbation renders one of the two states less favorable. The system may remain in such a metastable state for a long time, but will eventually transition to the more stable one Langer 1969 . This transition is usually understood in terms of droplet theory Fisher 1967 ; Langer 1967 ; Langer 1969 ; Gunther et al. 1980 ; Tomita and Miyashita 1992 ; Rikvold et al. 1994 . Droplets of the stable phase can spontaneously nucleate out of a metastable sea due to fluctuations, or form via coarsening from an initial mixture of the two phases Bray 1993 ; Bray 2010 , Fig. 1 (a). The former takes an exponentially long time t\sim e^{\mathcal{O}(R^{d})} , with R the droplet radius and d the dimensionality, whereas the latter only takes t\sim R^{z} with z a dynamical exponent. Small droplets then shrink due to surface tension and disappear, whereas large droplets grow until the entire system is stable, Fig. 1 (b). A critical droplet radius R_{c} separating these scenarios stems from the competition between bulk and boundary droplet energies. For example, in a two-dimensional Ising model at low temperature, H=-J\sum_{\langle i,j\rangle}\sigma_{i}\sigma_{j}-h\sum_{i}\sigma_{i} , a small field h destabilizes one of the two ferromagnetic phases, the bulk and surface energies of a droplet read \sim\pi R^{2}|h| and \sim 2\pi RJ , respectively, and balancing them yields R_{c}\sim J/|h| Rikvold et al. 1994 . Figure 1: Phenomenology of apparent bistability. (a) Droplets of radius R can nucleate from a clean metastable phase due to fluctuations, a slow process requiring exponentially long times t\sim e^{\mathcal{O}(R^{d})} , or from an initial mixture of phases via coarsening dynamics, in a shorter time t\sim R^{1/z} . (b) Once nucleated, subcritical droplets shrink and disappear, while supercritical droplets grow. (c) The critical droplet radius R_{c} scales with the bias field h as h^{-1} in the short-range regime where interactions decay with power \alpha>d+1 and as R_{c}\sim h^{-1/(\alpha-d)} in the weak long-range regime d<\alpha<d+1 , diverging rapidly for \alpha\to d^{+} . (d) The critical radius R_{c} becomes comparable to the system size L at \alpha=\alpha_{c} . For d<\alpha<\alpha_{c} , nucleation of a supercritical droplet is extremely unlikely and the system apparently bistable. The exponential divergence of R_{c} makes \alpha_{c} depend weakly on L and behave like a genuine transition point. The idea of surface tension at the core of standard droplet theory assumes short-range interactions. If the interactions are strong long-range, 1/r^{\alpha} with \alpha<d , the droplet picture breaks and the critical droplet size R_{c} can diverge with system size, restoring bistability in the thermodynamic limit. This scenario has long been studied for classical systems Griffiths et al. 1966 ; Penrose and Lebowitz 1971 ; Chandra 1989 ; Antoni and Ruffo 1995 ; Yamaguchi et al. 2004 ; Mukamel et al. 2005 ; Campa et al. 2009 ; Gabrielli et al. 2010 ; Nishino et al. 2011 ; Gupta and Ruffo 2017 , and more recently for quantum ones Kastner 2011 ; Defenu 2021 ; Defenu et al. 2023 ; Defenu et al. 2024 ; Mattes et al. 2025 . Excluding cases that ‘‘piggyback’’ on a symmetry 1 1 1 For instance, for the two-dimensional Ising model, the critical droplet radius R_{c} and metastable lifetime \tau diverge for h\to 0 . The existence of more stable phases in this case is unsurprising, indeed unavoidable given the Z_{2} symmetry for h=0 . By contrast, we want here to focus on bistability in spite of a bias field h , that is, not directly relying on an underlying symmetry. , bistability can also be restored for irreversible dynamics Bennett and Grinstein 1985 , as famously realized in Toom’s rule Toom 1980 ; Grinstein 2004 and as recently exploited in connection to time crystals Gambetta et al. 2019a ; Gambetta et al. 2019b ; Pizzi et al. 2021a ; McGinley et al. 2022 ; Machado et al. 2023 . Much less is instead known about the regime \alpha>d , for which the droplet picture should apply but can be substantially modified. In dimension one, it was shown for an Ising model that R_{c}\sim h^{-1/(\alpha-1)} for 1<\alpha<2 McCraw 1980 ; van Enter et al. 2019 . In dimension two, numerics shows that R_{c} diverges for \alpha\to 2^{+} Horikx 2020 , and metastability was investigated for long-range biaxial interactions Jacquier et al. 2024 . Despite this progress, a clear understanding of metastability and nucleation for \alpha>d is missing Jacquier 2025 . Here, we investigate metastability in systems with power-law interactions. We develop a field-theoretical description of coarsening and droplet formation, predicting that in the weak long-range regime, d<\alpha<d+1 , the critical droplet size scales as R_{c}\sim h^{-1/(\alpha-d)} . Such scaling has a remarkable consequence: Even for moderately large values of the bias h , the system can be apparently bistable , namely appear bistable for all practical purposes, while strictly admitting a single stable phase. Namely, beyond \alpha<d bistability extends in an apparent fashion to d<\alpha<\alpha_{c} , see Fig. 1 (c). By numerically implementing a probabilistic cellular automaton, we show that the apparent phase boundary \alpha_{c} displays the hallmarks of a bona fide phase transition and is in practice indistinguishable from it. Field theoretical insights — We begin by elucidating the functional dependence of the critical droplet size R_{c} on power-law exponent \alpha and bias field h using two complementary field theoretical approaches. The first assumes a Landau-Ginzburg-Wilson-type Hamiltonian for a scalar field \phi(\mathbf{x}) with power-law interactions Bray 1993 , H[\phi]=H_{SR}+\int\mathrm{d}^{d}\mathbf{x}\int\mathrm{d}^{d}\mathbf{x}^{\prime}\,\left(\phi(\mathbf{x})-\phi(\mathbf{x}^{\prime})\right)^{2}/|\mathbf{x}-\mathbf{x}^{\prime}|^{\alpha}