Walking on Archimedean Lattices: Insights from Bloch Band Theory

Basic Information

• Paper ID: 2507.12662
• Title: Walking on Archimedean Lattices: Insights from Bloch Band Theory
• Authors: Davidson Noby Joseph, Igor Boettcher (University of Alberta)
• Classification: cond-mat.stat-mech, cond-mat.mes-hall, cond-mat.str-el, math-ph, math.MP
• Publication Date: January 2025 (arXiv preprint)
• Paper Link: https://arxiv.org/abs/2507.12662

Abstract

This paper investigates the problem of returning walks on lattices, namely the sequence of paths that return to the origin after n steps from a given lattice point. By establishing connections with Bloch band theory, the authors derive analytical expressions for the number of returning walks on all eleven two-dimensional Archimedean lattices. The results are verified through an alternative method of computing powers of adjacency matrices for large graphs. As an application to condensed matter physics, the authors utilize these formulas to calculate the density of states (DOS) for tight-binding models on Archimedean lattices and analytically determine the asymptotic behavior of return probabilities.

Research Background and Motivation

Problem Definition and Significance

1. Core Problem: Determining the total number of returning walks of length n on a lattice, a typical graph theory problem with important applications in statistical physics and condensed matter lattice models.
2. Physical Significance:
   • The number of returning walks is directly related to the density of states (DOS), a fundamental quantity describing electronic properties of materials
   • Return probability plays a crucial role in transient behavior of statistical models and Anderson localization in disordered solids
   • Can be used to define discrete path integrals for quantum lattice models
3. Limitations of Existing Methods:
   • Simple combinatorial methods are difficult to apply to complex lattice structures
   • Traditional direct computation methods have excessive computational complexity for high dimensions or large unit cells
   • Lack of unified theoretical framework for handling different types of periodic tilings
4. Research Motivation:
   • Archimedean lattices encompass the most commonly studied planar lattices (square, triangular, honeycomb, kagome, etc.)
   • Establish a bridge between graph theory problems and physical band theory, providing new theoretical tools for computation
   • Develop generalizable techniques for more general two-dimensional or higher-dimensional Euclidean lattices

Core Contributions

1. Established fundamental connections between returning walk counts and Bloch band theory, deriving the core formula: $S_n = \frac{1}{N_u}\int_k \text{Tr}[A(k)^n]$
2. Systematically calculated returning walk counts for all eleven Archimedean lattices, including explicit formulas or generating functions
3. Developed universal methods for constructing large finite graphs (flakes and clusters) supporting both open and periodic boundary conditions
4. Calculated analytical expressions for the density of states of seven Archimedean lattices, with some reported for the first time
5. Determined the asymptotic behavior of return probabilities, obtaining the universal form $p_n \sim \alpha/n$

Detailed Methodology

Task Definition

Given a periodic lattice, define the number of returning walks as $S_n^{(i)} = (A^n)_{ii}$, where $A$ is the adjacency matrix. For vertex-transitive Archimedean lattices, $S_n = S_n^{(i)}$ is independent of the starting point.

Core Theoretical Framework

1. Bloch Adjacency Matrix Construction

For periodic tilings with $N_u$ atoms in the unit cell, the Bloch adjacency matrix $A(k)$ is an $N_u \times N_u$ matrix with elements: $A(k)_{i_u j_u} = \sum_v \langle i_u|\hat{A}\hat{T}_v|j_u\rangle e^{-iv \cdot k}$

2. Key Identity Derivation

Through rigorous mathematical derivation, the following identity is proven: $S_n^{(i_u)} = \int_k [A(k)^n]_{i_u i_u}$

This identity transforms the combinatorial problem on infinite lattices into an integral of finite-dimensional matrices over the Brillouin zone.

3. Generating Function Method

Define the Bloch generating function: $G(z,k) = \frac{1}{N_u}\text{Tr}\left(\frac{1}{1-zA(k)}\right)$

Then: $G(z) = \int_k G(z,k) = \sum_{n \geq 0} S_n z^n$

Technical Innovations

1. Theoretical Breakthrough: First systematic establishment of deep connections between returning walk problems in graph theory and Bloch band theory in condensed matter physics
2. Computational Method Innovation:
   • Developed pure algebraic techniques for evaluating momentum integrals using constant term extraction methods
   • For complex lattices, avoided explicit eigenvalue calculations by computing diagonal elements of matrix inverses
3. Construction Method: Proposed a universal master formula (56) for constructing finite graphs of arbitrary size: $A_{p,q} = (1_p \otimes 1_q) \otimes A_u + \sum_d [(R_p^{(d)} \otimes R_q^{(d)}) \otimes \Gamma^{(d)} + \text{h.c.}]$

Experimental Setup

Research Objects

Eleven Archimedean Lattices:

• Bipartite: Square, Honeycomb, CaVO, SHD
• Non-bipartite: Triangular, Kagome, Trellis, Star, SrCuBO, Ruby, Maple-Leaf

Verification Methods

1. Theoretical Calculation: Using Bloch adjacency matrices and generating function methods
2. Numerical Verification: Constructing large finite graphs and computing powers of adjacency matrices
3. Convergence Testing: Examining convergence of $S_n$ by increasing parameters $p, q$

Computational Tools

• Mathematica and other computer algebra software for matrix inversion calculations
• Generating functions computed from first few moments using Cayley-Hamilton theorem
• Brillouin zone integration using 2500 uniformly sampled points

Experimental Results

Main Results

1. Returning Walk Counts

Successfully calculated returning walk sequences for all eleven lattices, for example:

• Square Lattice: $S_{2n} = \binom{2n}{n}^2$
• Honeycomb Lattice: $S_{2n} = \sum_{l=0}^n \binom{2l}{l}\binom{n}{l}^2$
• Triangular Lattice: $S_n = \sum_{l=0}^n \binom{n}{l}(-3)^{n-l}S_{2l}^{(H)}$

2. Density of States Calculation

Obtained analytical expressions for density of states of seven lattices, including:

• Square Lattice: $D_\square(E) = \frac{1}{2\pi^2}K(1-\frac{E^2}{16})$
• Honeycomb Lattice: $D_H(E) = \frac{2|E|}{\pi^2\sqrt{Z_0(E)}}K(\frac{Z_1(E)}{Z_0(E)})$

3. Asymptotic Behavior

Determined asymptotic coefficients $\alpha$ for return probabilities:

• Bipartite Lattices: $S_{2n} \sim \alpha \frac{q^{2n}}{2n}$
• Non-bipartite Lattices: $S_n \sim \alpha \frac{q^n}{n}$

Specific values include $\alpha = \frac{2}{\pi}$ for square lattice and $\alpha = \frac{\sqrt{3}}{2\pi}$ for triangular lattice.

Verification Results

1. Numerical Consistency: Theoretical calculations completely agree with numerical computations on large finite graphs
2. Reproduction of Known Results: Successfully reproduced known sequences for square, honeycomb, triangular, and kagome lattices
3. New Sequences: First complete calculation of returning walk sequences for the remaining seven lattices

Related Work

Traditional Methods

1. Combinatorial Methods: Only applicable to simple lattices such as square lattices
2. Adjacency Matrix Method: Direct computation of $A^n$, but computationally difficult for large systems
3. Continued Fraction Method: Construction via Lanczos algorithm, but subject to numerical rounding errors

Theoretical Development

1. Spectral Moment Theorem: Results can be viewed as special cases of universal spectral moment theorems
2. Path Integrals: Related to discrete path integral theory
3. Hypergeometric Functions: Closely related to elliptic integrals and hypergeometric function theory

Advantages of This Work

1. Unified Framework: Provides a unified method for handling all Archimedean lattices
2. Analytical Results: Obtains exact analytical expressions rather than numerical approximations
3. Physical Insights: Reveals deep connections between graph theory and band theory

Conclusions and Discussion

Main Conclusions

1. Theoretical Contribution: Established fundamental connections between returning walk problems and Bloch band theory
2. Computational Achievement: Complete calculation of returning walk counts for eleven Archimedean lattices
3. Applied Value: Provides new tools for density of states calculation and asymptotic analysis

Limitations

1. Scope of Applicability: Primarily applicable to symmorphic crystallographic tilings; not suitable for non-periodic structures
2. Computational Complexity: For large unit cells (e.g., SHD with 12 atoms), analytical calculations remain complex
3. Dimensional Constraints: While theoretically generalizable to higher dimensions, computational complexity increases significantly

Future Directions

1. Extended Applications: Extension to Laves lattices and other periodic tilings
2. Non-Euclidean Geometry: Application to hyperbolic lattices and other non-Euclidean tilings
3. Related Problems: Extension to self-avoiding walks and walks with area constraints
4. Physical Applications: Application to critical temperature and free energy calculations in two-dimensional Ising models

In-Depth Evaluation

Strengths

1. Theoretical Innovation: First systematic establishment of connections between graph theory and band theory with significant theoretical value
2. Computational Completeness: Covers all Archimedean lattices, providing a complete dataset
3. Method Generalizability: Developed techniques generalizable to other periodic tilings
4. Physical Significance: Results have direct applications in condensed matter physics problems such as DOS calculation
5. Mathematical Rigor: Derivations are rigorous with detailed mathematical proofs

Weaknesses

1. Limited Novelty: Core identity (68) can be viewed to some extent as an application of known spectral moment theorems
2. Computational Efficiency: For certain complex lattices, computation still relies on symbolic computation software
3. Experimental Verification: Lacks comparison with actual physical experiments
4. Application Depth: While providing tools, in-depth applications to specific physical problems are limited

Impact

1. Academic Value: Provides a new paradigm for interdisciplinary research between graph theory, mathematical physics, and condensed matter physics
2. Practical Value: Provides new tools for density of states calculations in materials science
3. Reproducibility: Detailed computational methods and results facilitate verification and application
4. Inspirational Significance: Opens new perspectives for research on related combinatorial problems

Applicable Scenarios

1. Theoretical Research: Lattice models, statistical physics, graph theory research
2. Materials Computation: Electronic structure calculations for novel two-dimensional materials
3. Algorithm Development: New algorithms for large sparse matrix eigenvalue problems
4. Educational Applications: Typical case study for interdisciplinary mathematical physics

References

The paper cites 75 relevant references, spanning from classical Pólya random walk theory to recent hyperbolic lattice research, reflecting the historical development and current frontiers of the field. Important references include Wallace's graphene band theory, Kitaev's quantum spin liquid models, and recent experimental work on hyperbolic lattices.