Universal finite-size scaling in high-dimensional critical phenomena

Basic Information

• Paper ID: 2412.08814
• Title: Universal finite-size scaling in high-dimensional critical phenomena
• Authors: Yucheng Liu (University of British Columbia), Jiwoon Park (Republic of Korea Air Force Academy), Gordon Slade (University of British Columbia)
• Classification: cond-mat.stat-mech math-ph math.MP math.PR
• Publication Date: December 2024 (arXiv preprint)
• Paper Link: https://arxiv.org/abs/2412.08814

Abstract

This paper proposes a unified theory describing critical finite-size scaling (FSS) of lattice statistical mechanics models above the upper critical dimension under periodic boundary conditions. The theory is based on recent mathematically rigorous results for linear and branched polymers, multi-component spin systems, and percolation. The theory encompasses both short-range and long-range interactions. Universal finite-size scaling is inherited from the scaling properties of the system expanded to the infinite lattice. The authors also conjecture universal scaling profiles for susceptibility and two-point function plateaus within the critical window. For free boundary conditions, universal scaling has been proven to apply at the pseudocritical point for hierarchical spins, and the authors conjecture this result is universal.

Research Background and Motivation

Problem Background

1. Importance of finite-size effects: Laboratory samples and numerical simulations involve finite systems; therefore, finite-size scaling (FSS) of phase transitions is an essential component of theory.
2. Specificity of the upper critical dimension: Above the upper critical dimension $d_c$, the role of boundary conditions in FSS has been a subject of controversy and discussion, lacking unified mathematically rigorous results.
3. Diversity of models: Different statistical mechanics models (Ising model, self-avoiding walks, percolation, branched polymers, etc.) exhibit similar behavior above their respective upper critical dimensions, requiring a unified theoretical framework.

Research Motivation

1. Need for unified theory: Existing research lacks a unified mathematically rigorous framework describing finite-size scaling behavior of different models at high dimensions.
2. Controversy over boundary conditions: Significant differences exist in scaling behavior under periodic boundary conditions (PBC) and free boundary conditions (FBC), requiring theoretical explanation.
3. Extension to long-range interactions: Extending the theory from short-range to long-range interactions provides broader applicability.

Core Contributions

1. Proposed unified finite-size scaling theory: Based on the "unwrapping" concept, connecting finite toroidal systems with infinite lattice systems.
2. Established two key hypotheses: Hypothesis 1 describes two-point function behavior on infinite lattices; Hypothesis 2 establishes a comparison principle between toroidal and unwrapped systems.
3. Proved main theorem: Theorem 1 provides rigorous lower bounds for susceptibility and two-point functions, determining exponents for the critical window, susceptibility scaling, and plateau scaling.
4. Verified multiple models: Validated the theory's applicability to long-range self-avoiding walks, short-range Ising models, percolation, and branched polymers.
5. Proposed universal profile conjecture: Conjectured universal scaling profile functions across different models within the critical window.
6. Extended to free boundary conditions: Conjectured identical universal scaling behavior at pseudocritical points under free boundary conditions.

Methodology Details

Core Concept: Unwrapping Operation

Unwrapped two-point function is defined as: $\Gamma_{R,\beta}(x) = \sum_{u \in \mathbb{Z}^d} G_\beta(x + Ru) \quad (x \in T_R^d)$

This concept "unwraps" configurations on finite tori to infinite lattices, as illustrated in Figure 1 showing lattice tree unwrapping.

Two Key Hypotheses

Hypothesis 1 (Infinite lattice behavior):

• Susceptibility satisfies $\chi(\beta) \propto t^{-\gamma}$, correlation length satisfies $\xi(\beta) \propto t^{-\nu}$, where $\nu = \gamma/\alpha$
• Two-point function maintains critical decay within correlation length scale: $G_\beta(x) \gtrsim \frac{1}{|||x|||^{d-\alpha}} \quad (|x| \leq s_1\xi(\beta))$
• Rapid decay beyond correlation length

Hypothesis 2 (Comparison principle): $\left(1 - c_0\frac{\chi(\beta)^{d_c/2}}{V}\right)\Gamma_{R,\beta}(x) \leq G_{R,\beta}(x) \leq \Gamma_{R,\beta}(x)$

The upper bound reflects weaker interactions in the unwrapped model; the lower bound quantifies error in the comparison.

Main Theorem

Theorem 1: For dimension $d > d_{c,\alpha}$ and sufficiently large $R$, at $\beta^* = \beta_c - c_1V^{-\frac{2}{\gamma d_c}}$:

$\chi_R(\beta^*) \asymp V^{\frac{2}{d_c}}, \quad G_{R,\beta^*}(x) \asymp \frac{1}{|||x|||^{d-\alpha}} + \frac{1}{V^{1-\frac{2}{d_c}}}$

Technical Innovations

1. Unwrapping mechanism: Establishes connection between finite and infinite systems through unwrapping operations, allowing infinite volume scaling to directly produce finite-size scaling exponents.
2. Unified framework: Simultaneously applicable to short-range ($\alpha = 2$) and long-range ($\alpha \in (0,2)$) interactions, as well as different model types.
3. Rigorous mathematical proof: Based on lace expansion and other rigorous mathematical methods, rather than physical intuition or numerical observation.

Experimental Setup

Model Verification

The paper validates the theory's applicability to the following models:

1. Self-avoiding walks ($d_c = 4$):
   • Short-range nearest-neighbor model (d > 4)
   • Expanded long-range model (d > 2α)
2. Branched polymers ($d_c = 8$):
   • Expanded short-range lattice trees and lattice animals (d > 8)
3. Percolation ($d_c = 6$):
   • Short-range percolation (d ≥ 11)
   • Expanded short-range percolation (d > 6)
4. Spin systems ($d_c = 4$):
   • Short-range Ising model (d > 4)
   • Expanded long-range Ising model

Verification Methods

• Lace expansion technique to verify Hypothesis 1
• Graph-theoretic analysis and coupling processes to verify Hypothesis 2
• Random current representation (Ising model) and exploration processes (percolation)

Experimental Results

Finite-size Scaling Exponents

Long-range Self-avoiding Walk Verification

Detailed calculations verify that long-range self-avoiding walk models satisfy both hypotheses:

• Lace expansion proves near-critical estimates
• Differential inequalities and convolution estimates establish comparison principle

Universal Profile Functions

Universal profile functions proposed for different models:

where $I_k(s) = \int_0^{\infty} x^k e^{-\frac{1}{4}x^4 - \frac{1}{2}sx^2} dx$.

Related Work

Historical Development

1. Early controversy: Long-standing debate over the role of boundary conditions above the upper critical dimension
2. Numerical studies: Numerical investigations by Jones & Young, Wittmann & Young, and others
3. Theoretical development: Renormalization group calculations by Brézin & Zinn-Justin (1985)

Advantages of This Work

1. Mathematical rigor: Based on rigorous mathematical proofs rather than physical intuition
2. Universality: Simultaneously addresses multiple models and interaction types
3. Completeness: Complete theory from scaling exponents to precise profiles

Conclusions and Discussion

Main Conclusions

1. Above the upper critical dimension, finite-size scaling under periodic boundary conditions is universal
2. Scaling behavior is inherited from systems expanded to infinite lattices
3. Different models share identical scaling exponents but possess different profile functions
4. Behavior under free boundary conditions can be described by identical profiles at pseudocritical points

Limitations

1. Dimensional restriction: Theory applies only above the upper critical dimension
2. Model specificity: Verification for certain models (e.g., long-range percolation) remains incomplete
3. Profile functions: Most profile functions remain conjectural, lacking rigorous proof

Future Directions

1. Extension to more long-range models
2. Inclusion of logarithmic corrections at critical dimension
3. Development of rigorous theory for free boundary conditions
4. Verification of profile function universality

In-depth Evaluation

Strengths

1. Theoretical unity: First unified rigorous theoretical framework describing finite-size scaling of multiple models
2. Mathematical rigor: Based on mature mathematical techniques such as lace expansion, results are rigorous
3. Innovation: Unwrapping mechanism provides new perspective for understanding finite-size effects
4. Completeness: Complete theoretical system from fundamental assumptions to concrete applications

Weaknesses

1. Limited scope: Restricted to above the upper critical dimension; direct application to real physical systems (typically low-dimensional) is limited
2. Technical threshold: Requires complex mathematical techniques, limiting accessibility of results
3. Numerical verification: Lacks direct verification through large-scale numerical simulations

Impact

1. Theoretical contribution: Provides important mathematical foundation for finite-size scaling theory in statistical mechanics
2. Methodology: Unwrapping techniques may inspire research on related problems
3. Interdisciplinary: Connects mathematical physics, probability theory, and statistical mechanics

Applicable Scenarios

1. Theoretical analysis of high-dimensional statistical mechanics models
2. Mathematical research on critical phenomena
3. Theoretical prediction of finite-size effects
4. Theoretical interpretation of numerical simulation results

References

The paper cites over 50 related references, primarily including:

• Pioneering work by Hara & Slade on self-avoiding walks
• Renormalization group theory by Brézin & Zinn-Justin
• Recent series of papers on rigorous results for high-dimensional models
• Numerical and theoretical studies of finite-size scaling

This paper represents significant progress in finite-size scaling theory in statistical mechanics, unifying the behavior of multiple models through rigorous mathematical methods and providing a solid theoretical foundation for the field.