In standard statistical mechanics, we often assume that particles or agents only care about their immediate surroundings. In these short-range systems, a spin on a lattice only feels its nearest neighbors, and a fluid molecule only collides with those touching it. This localized behavior gives rise to well-known universality classes—broad categories where vastly different physical systems share identical critical exponents near phase transitions.
But what happens when the reach of an interaction extends across the entire system? When forces decay slowly with distance, such as power-law interactions $V(r) \sim r^{-\alpha}$, the traditional rules of criticality, domain growth, and thermodynamic stability break down. By tuning the spatial range of an interaction, researchers can systematically alter how a system transitions between order and disorder, effectively shifting it between entirely different universality classes.
The Dial Between Local and Mean-Field Behavior
To understand why the range of an interaction matters, imagine a one-dimensional chain of elements. If each element only interacts with its nearest neighbor, the system is highly constrained by its geometry; for instance, a 1D Ising model cannot sustain long-range order at any finite temperature.
However, if we introduce long-range interactions that decay according to a power law ($1/r^\alpha$), we gain a mathematical "dial" to control the system's effective dimensionality. When $\alpha$ is very large, the interaction drops off so fast that it mimics a localized, short-range system. When $\alpha = 0$, the interaction strength is uniform regardless of distance, meaning every element interacts with every other element identically—a scenario known as the infinite-range or mean-field limit.
Helmut G. Katzgraber leveraged this exact mechanism in Analysis of evolutionary algorithms on the one-dimensional spin glass with power-law interactions to evaluate optimization algorithms, demonstrating that as $\alpha$ increases and interactions become localized, preserving the structural hierarchy of the problem becomes critical.
This power-law dial fundamentally changes whether a system can achieve global order. For example, M. C. Cross showed in Synchronization of oscillators with long-range power law interactions that a 1D chain of coupled oscillators can overcome the disruptive effects of low spatial dimensionality and achieve macroscopic synchronization, but only if the interaction decays slowly enough (specifically, when $\alpha < 1.5$).
Mean-Field Limit Long-Range Regime Short-Range Limit
(All-to-All) (Power-Law Decay) (Nearest-Neighbor)
α = 0 ───────────────────────► Low α ────────────────────────► High α (α → ∞)
Global order easy; Non-universal exponents; Standard universality;
No spatial structure. Range-dependent kinetics. Geometry-constrained.
As we tune between these limits, the system must cross a threshold where mean-field behavior gives way to short-range spatial dynamics. In Extreme Value Statistics Distributions in Spin Glasses, Michele Castellana and Elia Zarinelli highlighted that tracking how critical temperature fluctuations shift between these regimes remains essential for bridging idealized mean-field models with real-world, finite-range materials.
Reshaping Phase Transitions and Universality
When the range of an interaction is extended, it does not just tweak the critical temperatures of a system; it can fundamentally rewrite the nature of the phase transition itself.
In a standard two-dimensional Potts model (a generalization of the Ising model), transitions are typically continuous for a small number of states. However, Youjin Deng et al. demonstrated in Equivalent-neighbor Potts models in two dimensions that increasing the interaction range can force a continuous phase transition to become discontinuous. This structural shift introduces tricritical points where different types of transitions meet, proving that long-range forces can induce phenomena analogous to dilution or impurities in short-range models.
Similarly, altering the interaction range can give rise to exotic, inverted thermodynamic phases. In Inverted Berezinskii-Kosterlitz-Thouless singularity and high-temperature algebraic order in an Ising model on a scale-free hierarchical-lattice small-world network, A. Nihat Berker showed that distance-dependent interaction strengths on hierarchical networks can cause algebraic order to appear in the high-temperature phase, narrowing the critical regions and upending conventional expectations of thermal disorder.
Breaking the Kinetic Speed Limits of Coarsening
The consequences of long-range interactions are perhaps most vividly seen out of equilibrium. When a disordered system is suddenly quenched to a low temperature, it undergoes phase separation, forming ordered domains that grow over time—a process known as coarsening.
In systems governed by local, short-range interactions, domain growth is bound by a universal kinetic speed limit: the characteristic length scale of the domains grows over time as $L(t) \sim t^{1/2}$.
Short-Range Coarsening: L(t) ~ t^(1/2) [Rugged, irregular boundaries]
Long-Range Coarsening: L(t) ~ t^(3/4) [Smoother, force-stabilized boundaries]
Long-range interactions shatter this speed limit. Because an element can feel the influence of a domain far away, the rate of domain growth accelerates. Suman Majumder and Wolfhard Janke investigated this in Phase ordering kinetics of the long-range Ising model, revealing that the growth law becomes non-universal and depends directly on the interaction decay exponent.
Surprisingly, this transition from short-range to long-range kinetic regimes can be discrete rather than continuous. In Zero-temperature coarsening in the two-dimensional long-range Ising model, the authors discovered that during a zero-temperature quench, even a weak long-range component dominates the late-time evolution, causing domains to grow with a universal exponent of approximately $3/4$ instead of the standard $1/2$.
This accelerated kinetics also changes the physical morphology of the phases and how the system "remembers" its past:
- Boundary Smoothing: In Coarsening in the long-range Ising model: Metropolis versus Glauber criterion, Wolfhard Janke and Suman Majumder showed that long-range forces penalize rugged, irregular boundaries, producing significantly smoother domain interfaces than short-range models.
- Altered Aging Dynamics: In Aging in the Long-Range Ising Model, Suman Majumder et al. proved that the long-range spatial coupling shifts the system into a new universality class for aging, dropping the characteristic autocorrelation exponent from $1.25$ down to $1.0$.
Geometric Reshaping and Nonequilibrium Clustering
Beyond spin lattices, the spatial extent of a potential acts as a critical factor in geometric and transport properties across fluids, solids, and electronic systems.
In quantum materials, the geometry of the Fermi surface dictates electronic behavior. Jia Ning Leaw et al. demonstrated in Universal Fermi-surface anisotropy renormalization for interacting Dirac fermions with long-range interactions that long-range Coulomb interactions systematically reduce distortions in the Fermi surface, driving the system toward an isotropic state. Crucially, this geometric reshaping is entirely absent in models containing only short-range contact forces, proving that the spatial reach of the potential is required to trigger the renormalization.
The range of interaction also dictates macroscopic stability and relaxation pathways:
- Fluid Dynamics: In Violent relaxation in two-dimensional flows with varying interaction range, Antoine Venaille et al. showed that varying the interaction range parameter determines the final state of a 2D fluid. "Weak" long-range systems relax toward statistical equilibrium via standard coarsening, whereas "strong" long-range systems get trapped in stable, long-lived dipolar structures through filamentation.
- Pattern Formation: In Stability Analysis of Line Patterns of an Anisotropic Interaction Model, José A. Carrillo et al. established that whether emergent line patterns (like those in fingerprints) remain stable depends explicitly on whether the underlying force coefficients decay algebraically or exponentially.
- Nonequilibrium Transport: In Anomalous current fluctuations and mobility-driven clustering, Punyabrata Pradhan identified a novel phase transition mechanism where infinite-range particle hopping triggers a singularity in mobility. This "mobility-driven clustering" causes particles to condense into aggregates and is completely absent in models restricted to finite-range hopping.
Ultimately, whether looking at the cooling of a magnetic spin glass, the synchronization of nanomechanical oscillators, or the clustering of a lattice gas, the message is clear: the spatial range of an interaction is not a minor detail to be swept away by universality. Instead, it is a fundamental parameter that dictates how a system organizes itself across both space and time.
Key Foundational Work
Inverted Berezinskii-Kosterlitz-Thouless singularity and high-temperature algebraic order in an Ising model on a scale-free hierarchical-lattice small-world network (2006) — contributed 1 concept elements to this direction.
Analysis of evolutionary algorithms on the one-dimensional spin glass with power-law interactions (2009) — contributed 1 concept elements to this direction.