🔬 optics

Bounded frequency lattices in integrated lithium niobate coupled ring cavities

This paper demonstrates an integrated thin-film lithium niobate coupled ring system that successfully simulates a one-dimensional frequency crystal lattice with sharp boundaries by using a single auxiliary cavity to suppress specific coupling terms, thereby enabling the study of high-dimensional topological physics and optical information processing on a photonic chip.

Original authors: Hiep X. Dinh, Armandas Balčytis, Guanghui Ren, Mei Xian Low, Arnan Mitchell, Thach G. Nguyen

Published 2026-03-03

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Hiep X. Dinh, Armandas Balčytis, Guanghui Ren, Mei Xian Low, Arnan Mitchell, Thach G. Nguyen

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ 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 a giant, endless hallway made of mirrors. If you walk down this hallway and clap your hands, the sound bounces back and forth forever, creating a continuous echo. In the world of light and physics, this is similar to how light behaves in a standard ring-shaped laser cavity: it can bounce between different "frequencies" (colors of light) endlessly in both directions.

But what if you wanted to build a wall in the middle of that hallway? What if you wanted the light to hit a boundary and stop, creating a finite, contained space? This is exactly what the researchers in this paper achieved, but instead of a hallway, they built a "frequency crystal" on a tiny chip.

Here is a simple breakdown of their discovery using everyday analogies:

1. The "Synthetic Dimension" (The Magic Elevator)

Usually, to study complex physics, scientists need huge, multi-dimensional structures. But this team used a clever trick called a synthetic dimension.

The Analogy: Imagine a single elevator shaft. Instead of moving up and down in space, the elevator moves between different "floors" based on the pitch of the sound inside it.
In the Lab: They used a ring of light (a micro-ring resonator). By shaking the ring with a radio signal (like a DJ scratching a record), they made the light "hop" from one frequency to the next.
- Frequency 100 = Floor 1
- Frequency 101 = Floor 2
- Frequency 102 = Floor 3
- And so on.
- To the light, this looks like a 1D lattice (a line of connected rooms), even though it's all happening inside one tiny ring.

2. The Problem: The Infinite Hallway

In a normal setup, this "elevator" has no top or bottom. The light can hop up to infinite high frequencies or down to infinite low frequencies. It's an endless random walk.

Why this matters: In physics, "boundaries" are crucial. They create special "edge states" (like a surfer riding a wave only at the shore). Without a wall, you can't study these edge effects.

3. The Solution: The "Ghost Wall"

The researchers wanted to build a wall in this frequency hallway to stop the light from hopping forever. They couldn't just paint a wall on a frequency; they had to trick the light.

The Setup: They built a main ring (the hallway) and attached a smaller, slightly different "auxiliary" ring next to it.
The Trick: Think of the main ring as a row of perfectly spaced stepping stones. The auxiliary ring is like a second row of stones that occasionally overlaps with the first.
- Where they overlap, the stones get "glued" together and split apart.
- This creates a gap in the stepping stones.
- When the light tries to hop to the next frequency, it hits this gap. The "elevator" breaks. The light cannot jump over the gap because the rhythm is wrong.
The Result: The light is trapped in a small, finite section of the hallway (specifically, 7 "floors" or frequencies). It hits the "wall" and bounces back.

4. The "Lithium Niobate" Chip (The Super Highway)

They built this on a material called Lithium Niobate (LNOI).

The Analogy: Think of this material as a super-high-speed, ultra-smooth highway for light. It's so good that they could shrink a system that used to require 38 meters of fiber-optic cable (a huge loop) down to a tiny chip only 1.6 centimeters long.
Why it's cool: This makes the device tiny, fast, and efficient enough to be used in future computers or communication devices.

5. What Did They Prove?

They proved that they could:

Create a Boundary: They successfully stopped the light from hopping past a certain point, creating a "bounded" lattice.
Control the Wall: They could tune the wall to stop the light from hopping just one step away (nearest neighbor) or even two steps away (next-nearest neighbor).
See the "Edge States": By measuring the light, they saw the "band structure" (the energy map). In the infinite hallway, the energy map is a smooth curve. In their bounded hallway, the curve broke into distinct, separate dots, proving the light was trapped in a finite box.

Why Should You Care?

This isn't just about playing with light; it's about building the future of optical computing.

Topological Protection: Just like a knot in a string is hard to untie, light trapped at these "edges" is very robust. It won't scatter or get lost easily, even if the chip has tiny defects.
New Computers: This technology could lead to optical chips that process information using these "edge states," making them faster and more energy-efficient than current silicon chips.
Security: It could help create better optical isolators (one-way mirrors for light) to protect sensitive laser equipment.

In a nutshell: The researchers built a tiny, high-tech "frequency cage" on a chip. They used a clever trick with a secondary ring to create invisible walls that trap light in a specific range of colors, proving that we can now control the "boundaries" of light in ways that were previously impossible. This is a major step toward building the next generation of super-fast, light-based computers.

1. Problem Statement

Synthetic dimensions allow researchers to simulate high-dimensional topological physics using low-dimensional photonic structures by utilizing degrees of freedom such as frequency, time, or polarization. A critical component in topological physics is the existence of boundaries, which give rise to edge states and bulk-edge correspondence.

However, existing methods for creating synthetic frequency lattices using modulated ring resonators typically produce unbounded lattices. In these systems, photon hopping occurs symmetrically in both directions across the frequency spectrum without inherent termination. While some approaches attempt to create boundaries by engineering mode splitting or using coupled fiber loops, they suffer from significant drawbacks:

Fiber-based systems: Susceptible to environmental decoherence and lack integration.
Mode-splitting approaches: Often sensitive to fabrication imperfections and lack reconfigurability.
General limitation: There is a lack of robust, integrated, and tunable methods to create sharp frequency-space boundaries that can confine photon hopping to a finite lattice segment.

2. Methodology

The authors propose and demonstrate a solution using an integrated thin-film Lithium Niobate on Insulator (LNOI) platform.

Device Architecture: The device consists of two coupled asymmetric micro-ring resonators:
- Primary Ring (Ring A): Acts as the synthetic lattice site. It has a Free Spectral Range (FSR) of $FSR_A \approx 9.5$ GHz.
- Auxiliary Ring (Ring B): Has a larger FSR ( $FSR_B \approx 76$ GHz), which is exactly $8 \times FSR_A$ .
Boundary Mechanism:
- The auxiliary ring is coupled to the primary ring via a directional coupler.
- Due to the $8:1$ FSR ratio, the modes of the two rings align periodically. At these alignment points, mode hybridization occurs, causing mode splitting.
- This splitting creates a local defect where the uniform mode spacing is disrupted. The frequency shift and dispersion induced by this coupling create a "frequency mirror" that suppresses photon hopping beyond a specific point.
- This effectively partitions the infinite frequency lattice into a finite segment of $N=7$ sites bounded by these splitting-induced defects.
Modulation: A sinusoidal RF signal ( $\approx 9.2$ GHz) is applied to an integrated phase modulator on the primary ring. This matches the FSR to induce nearest-neighbor ( $J_1$ ) and next-nearest-neighbor ( $J_2$ ) coupling, simulating a tight-binding model.
Fabrication: The device uses a strip-loaded waveguide design (SiN ridge on LiNbO $_3$ ) to ensure vertical sidewalls and reproducible coupling, with gold electrodes for electro-optic modulation.

3. Key Contributions

Realization of Bounded Synthetic Lattices: The first demonstration of a robust, tunable frequency-space boundary in an integrated photonic chip that confines a synthetic lattice to a finite number of sites ( $N=7$ ).
Suppression of Coupling Terms: The device successfully suppresses both nearest-neighbor ( $1\times FSR$ ) and next-nearest-neighbor ( $2\times FSR$ ) coupling terms using a single auxiliary cavity.
High-Performance Integration: Achieved a three-order-of-magnitude reduction in optical path length compared to previous fiber-loop implementations (15.9 mm vs. 38.6 m) while maintaining high quality factors ( $Q \approx 3.1 \times 10^5$ ).
Experimental Band Structure Spectroscopy: The authors successfully mapped the discretized band structures of the finite lattice, confirming the transition from continuous bands (unbounded) to discrete energy states (bounded).

4. Results

Spectral Confinement:
- Unbounded Case (Single Ring): Under modulation, the spectrum showed up to 40 sidebands, indicating continuous random-walk hopping.
- Bounded Case (Coupled Rings): The spectrum was strictly confined to 7 sidebands.
- Boundary Effectiveness: A sharp drop in comb line power of approximately 16 dB was observed at the boundaries, indicating a reflectivity ( $R$ ) of 0.9988.
Coupling Range Verification:
- The boundary was effective for $1\times FSR$ ( $\approx 9.2$ GHz) and $2\times FSR$ ( $\approx 18.4$ GHz) driving signals.
- For $2\times FSR$ driving, the $N=7$ lattice was observed to split into two sublattices ( $N'=3$ and $N''=4$ ), demonstrating the ability to engineer complex lattice topologies.
- The device failed to suppress $3\times FSR$ coupling because the modulation frequency coincided with the boundary span, highlighting a design constraint for higher-order terms.
Band Structure Measurements:
- Time-resolved transmittance measurements revealed the transition from continuous tight-binding bands (single ring) to discretized energy states (coupled ring).
- The number of observed discrete states (7 for $1\times FSR$ , 3 for $2\times FSR$ ) matched the theoretical number of sites in the bounded lattice.
- Slight dispersion in the energy states was observed near the boundaries due to local FSR variations, which was consistent with simulations.

5. Significance

This work represents a pivotal step toward practical topological photonics and optical information processing on a chip.

Topological Physics: It enables the study of bulk-edge correspondence and edge states in synthetic dimensions, which were previously difficult to realize in integrated systems.
Optical Processing: The ability to create robust boundaries allows for applications such as topological lasing, optical isolation, and photonic delay lines.
Computational Capabilities: The system can perform arbitrary linear transformations and convolution processing, essential for photonic computing.
Scalability: The LNOI platform offers low loss, high power handling, and strong nonlinearity, making this architecture a default choice for future complex electro-optic circuits. The method of using a single auxiliary ring to create boundaries is highly scalable and reconfigurable via thermal or electro-optic tuning.