Tiling by Near Coincidence

✨

This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you have two transparent sheets of graph paper. On the first sheet, you draw a perfect grid of squares. On the second sheet, you draw the exact same grid, but you rotate it slightly (like turning a steering wheel) or stretch it out a bit.

Now, if you stack these two sheets on top of each other and shine a light through them, you don't just see two messy grids. You see a beautiful, complex, swirling pattern where the lines almost, but not quite, line up. This is called a Moiré pattern, and it's the same kind of pattern you see when you hold two fine mesh nets together or look at the ripples in a pond.

This paper introduces a clever new way to turn those "almost matching" points into a perfect, non-repeating puzzle (a tiling). The authors call this the "Near-Coincidence Method."

Here is the breakdown of their idea using simple analogies:

1. The Problem: The "Perfect Match" is Boring

Usually, if you try to make a pattern that never repeats (like a quasicrystal), you have to use very complicated math. If you just take two grids and rotate them, the points where they line up perfectly are either non-existent or they form a boring, repeating pattern.

2. The Solution: The "Almost There" Club

The authors realized that while the points might not line up perfectly, many of them get very close.

The Analogy: Imagine a dance floor with two groups of dancers. Group A is standing in a grid. Group B is standing in the same grid, but rotated.
Most dancers are far apart. But occasionally, a dancer from Group A and a dancer from Group B stand right next to each other, almost touching.
The authors say: "Let's ignore the dancers who are far apart. Let's only care about the pairs that are almost touching."

3. How the Method Works (The Recipe)

They propose a simple three-step recipe to turn these "almost-touching" pairs into a new, beautiful pattern:

Step 1: The "Buddy System" (Merging): Whenever a red dot (from the first layer) and a blue dot (from the second layer) are closer than a certain distance (like holding hands), you merge them into a single new point. You place this new point exactly in the middle of the two.
Step 2: The "Handshake" (Connecting): Once you have all these new middle points, you draw lines between the ones that are a specific distance apart.
Step 3: The "Cleanup Crew" (Refining): Sometimes, you get two new points that are too close together, which makes the pattern look messy. The authors suggest a simple rule: keep the point that came from the "best match" (the closest pair) and throw away the other one.

4. Why This is a Big Deal

It's Intuitive: Instead of using abstract, high-dimensional math (which is like trying to solve a puzzle in 4D space), this method uses simple 2D geometry that anyone can visualize. It's like finding a shortcut through a forest instead of climbing a mountain.
It Recreates Classics: When they tried this with specific angles (like rotating square grids by 45 degrees), they accidentally recreated famous, complex mathematical puzzles like the Ammann–Beenker tiling (an 8-sided pattern) and the Penrose tiling (a 10-sided pattern).
It Discovers New Things: Because the method is so flexible, they found brand-new patterns that mathematicians hadn't seen before. For example, by using a circular "acceptance zone" instead of a square one, they found a new 12-sided pattern featuring a "three-arm star" shape.

5. The "Real World" Connection

Why do we care? This isn't just about drawing pretty pictures.

Twisted Graphene: Scientists are currently obsessed with "twisted bilayer graphene" (two layers of carbon atoms twisted slightly). These materials can conduct electricity in weird ways, become superconductors, or act like magnets.
The Bridge: This paper provides a simple map to understand the atomic structure of these twisted materials. It explains how the "near misses" between the two layers create the giant, complex patterns that give these materials their special powers.

Summary

Think of this paper as a new way to build a mosaic. Instead of carefully calculating where every tile goes using complex formulas, you just take two layers of tiles, rotate them, and glue together any two tiles that are almost touching. The result is a stunning, non-repeating masterpiece that nature seems to love, and now, thanks to this method, we can build it with our eyes wide open.

The authors even built a free website where you can play with this yourself: you can rotate the layers, change the "closeness" threshold, and watch new patterns appear in real-time.

1. Problem Statement and Physical Motivation

The paper addresses the challenge of constructing quasiperiodic tilings of the plane that satisfy the Delone set conditions: the vertices must be uniformly discrete (no arbitrarily close points) and relatively dense (no arbitrarily large voids), while maintaining specific rotational symmetries (e.g., 8-fold, 10-fold, 12-fold).

The Limitation of Naive Constructions: Simply taking the closure of an $n$ -fold symmetric star of vectors under addition yields a valid tiling only for periodic cases ( $n=3, 4, 6$ ). For other $n$ , the point set becomes dense, violating uniform discreteness.
Existing Methods: Standard rigorous methods include the dual-grid and cut-and-project formalisms. While mathematically robust, these methods often rely on abstract higher-dimensional embeddings or dual spaces, which can lack physical intuition.
The Gap: There is a need for a more intuitive construction method inspired by physical systems, specifically twisted bilayer graphene and other 2D materials, where long-range aperiodic order (Moiré patterns) emerges from the superposition of periodic layers.

2. Methodology: The Near-Coincidence Method

The authors introduce a new, intuitive, yet rigorous algorithm based on the superposition of two periodic layers (bilayers).

Core Algorithm:

Superposition: Two identical periodic point sets (lattices or tilings) are superimposed. One layer is rotated by an angle $\theta$ or scaled by a factor $\lambda$ relative to the other.
Near-Coincidence Identification: Pairs of points $(p_1, p_2)$ —one from each layer—are identified if their Euclidean distance $|p_1 - p_2|$ is less than a defined threshold $r$ .
Vertex Generation: Each accepted pair is merged into a single vertex located at the midpoint: $V = (p_1 + p_2)/2$ .
Edge Connection: Vertices are connected by edges based on permitted distances (determined by the threshold $r$ ).
Local Cleaning: To ensure a proper tiling (removing overlapping tiles or crossing edges), local rules are applied. If two vertices are too close, the one with the higher degree of coincidence (smaller $|p_1 - p_2|$ ) is retained, and the other is discarded. This mimics "phason flips" observed in physical quasicrystals.

Mathematical Rigor:
The authors prove that this method maps directly to the cut-and-project formalism.

The bilayer is treated as a direct product of two lattices in a higher-dimensional space.
The physical space coordinate is the midpoint $(p_1+p_2)/2$ .
The internal space (perpendicular space) coordinate is the difference vector $p_1 - p_2$ .
The coincidence window (the threshold $r$ ) acts as the acceptance domain (or atomic surface) in the internal space.
By choosing specific shapes for the coincidence window (e.g., circular, polygonal), the method reproduces standard cut-and-project tilings exactly.

3. Key Contributions and Results

A. Reproduction of Classical Tilings

The method successfully reproduces well-known quasiperiodic tilings by adjusting the bilayer parameters and the coincidence window shape:

Octagonal (8-fold) Tiling: Generated by superimposing two square lattices rotated by $45^\circ$ $4 5^{\circ}$ .
- Using a circular coincidence window yields a tiling with squares, rhombs, kites, and trapezoids.
- Using an octagonal coincidence window (inscribed in the circle) yields the classic Ammann–Beenker tiling (squares and $45^\circ$ rhombs).
Dodecagonal (12-fold) Tiling: Generated by superimposing two triangular lattices rotated by $30^\circ$ $3 0^{\circ}$ .
- By varying the window shape (dodecagon, 12-fold star, or circle), the method reproduces the Niizeki–Gähler, Stampfli, and Shield tilings.
- The authors also discover new variants using circular windows, including a tiling containing a "tristar" (three-arm star) tile, which was previously unobserved in this context.
Fibonacci Tilings:
- 1D: Generated by scaling a 1D lattice by the golden mean $\tau$ . The method reveals that centered windows lead to symmetric substitution rules, while shifted windows recover canonical Fibonacci rules.
- 2D Square: Generated by scaling a square bilayer by $\tau$ .
- 2D Hexagonal: Generated by scaling a honeycomb bilayer by $\tau$ . This case requires four distinct coincidence windows corresponding to different vertex parity classes, reflecting the higher-dimensional nature of the hexagonal Fibonacci tiling.

B. Discovery of New Tilings

The flexibility of the coincidence window shape allows for the generation of tilings that are unlikely to arise from standard cut-and-project constructions using polygonal windows.

Circular Windows: Using circular windows (the most intuitive choice in this framework) produces valid Delone sets with unique prototile sets (e.g., the tristar tile in the dodecagonal family).
Substitution Rules: The method naturally reveals substitution rules (inflation/deflation) by varying the threshold $r$ . For example, scaling the coincidence window by the silver mean $\tau_S = 1+\sqrt{2}$ generates self-similar copies of the Ammann–Beenker tiling.

C. Extension to Trilayers and Small Angles

Trilayer Systems: Preliminary results show that three square layers rotated by $30^\circ$ and $60^\circ$ can generate dodecagonal order.
Pseudo-Dodecagonal Order: If only two layers are used in a trilayer setup, the symmetry breaks, resulting in "pseudo-dodecagonal" tilings with distorted rhombs and reduced symmetry (tetragonal), observable in Fourier space.
Graphene Applications: The method is proposed as a tool for modeling giant Moiré patterns in twisted bilayer and trilayer graphene at very small twist angles.

4. Significance and Future Directions

Intuitive vs. Rigorous: The primary significance is bridging the gap between physical intuition (Moiré patterns in materials) and mathematical rigor (cut-and-project). It provides a physical mechanism for how quasiperiodic order emerges from periodic layers.
Algorithmic Simplicity: The method is computationally efficient. It requires only defining layers, a threshold, and checking distances, making it ideal for numerical implementation.
Practical Tool: The authors have developed a web-based application (Near-Coincidence Tiling App) that allows users to generate these tilings interactively by specifying layer parameters and coincidence conditions.
Diffraction Spectrum: The paper notes that while the raw superposition is not a reciprocal lattice, the process of accepting near-coincident pairs and shifting them to midpoints generates additional harmonics, resulting in a well-behaved reciprocal lattice with the correct $N$ -fold symmetry.

Conclusion:
The "Tiling by Near Coincidence" method offers a powerful, physically motivated alternative to traditional quasiperiodic construction techniques. It not only reproduces established mathematical models but also facilitates the discovery of new quasiperiodic structures, providing a unified framework for understanding the aperiodic order found in twisted 2D materials like graphene.