🔬 optics

Quantum storage with flat bands

This paper proposes and experimentally validates a method for creating robust, spatially localized quantum states in flat-band lattices by hybridizing compact localized states with resonant dispersive waves through edge injection and localized potentials, demonstrating the technique's effectiveness in photonic waveguide arrays for quantum memory applications.

Original authors: Carlo Danieli, Jie Liu, Rudolf A. Römer, Rodrigo A. Vicencio

Published 2026-02-12

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Carlo Danieli, Jie Liu, Rudolf A. Römer, Rodrigo A. Vicencio

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 very busy highway where cars (representing light or quantum information) are zooming around at high speeds. Usually, if you want to park a car in a specific spot, you have to drive right up to that spot and stop. But what if the highway is designed so that cars never stop naturally? They just keep flowing.

This is the problem scientists face with quantum memory. They want to store information (like a single bit of data) in a specific, tiny spot so it doesn't disappear or get scrambled. But in many quantum systems, the "cars" (quantum states) are too restless to stay put.

This paper introduces a clever new way to park these "cars" using a special kind of road called a Flat Band Lattice.

The Analogy: The "Flat" Highway and the "Trap"

1. The Flat Band (The Restless Highway)
Imagine a highway where the road is perfectly flat. On a normal road, cars speed up or slow down depending on hills. But on this "flat" road, every car moves at the exact same speed, no matter what. In physics, this creates a situation where the cars don't want to move anywhere specific; they are spread out and restless. This is called a "Flat Band."

2. The Compact Localized State (The Perfect Parking Spot)
Even though the highway is flat, there are special "parking spots" (called Compact Localized States or CLS) where a car can sit perfectly still. The problem is, these spots are everywhere, and they all look the same. If you try to drive a car onto the highway, it's like trying to park in a sea of identical spots—you might accidentally park in the wrong one, or the car might just keep driving past.

3. The Old Way (The Hard Way)
Previously, to park a car in a specific spot, scientists had to build a special ramp right next to that spot (out-of-plane). It was like building a separate, tiny driveway just to drop a car into one specific parking space. It was complicated and hard to scale up.

4. The New Method (The "Side-Door" Trick)
This paper proposes a much smarter way. Instead of building a ramp next to the spot, they use two tricks:

The "Siren" (The Plane Wave): They send a wave of cars down the highway from the edge of the system. Think of this as a siren or a specific rhythm that matches the "hum" of the parking spots.
The "Magnet" (The Local Potential): At the exact spot where they want to park the car, they turn on a temporary "magnet" (a localized potential). This magnet doesn't just sit there; it changes the rules of the road only at that spot.

How it works together:
When the "siren" wave hits the "magnet" spot, something magical happens. The restless cars on the highway suddenly "lock in" with the magnet. They stop flowing and get trapped in that specific parking spot. It's like the wave and the magnet shake hands, and the car decides, "Okay, I'll stay right here."

Once the car is parked and the "magnet" is turned off (or adjusted), the car stays perfectly still because the road is flat. It's now stored!

The Experiment: Building a Light Highway

The researchers didn't just do this on a computer; they built it using light.

They used a laser to carve tiny glass "tubes" (waveguides) into a piece of glass. These tubes act like the highway lanes.
They made two specific shapes of highways: a Diamond Chain and a Lieb Ladder (think of these as different patterns of parking lots).
They shot a beam of light into the side of the glass (the edge).
By slightly changing the speed at which they carved the glass at specific spots, they created the "magnets" (the asymmetric potentials).

The Result:
The light didn't just pass through. It got caught in the specific spots they chose. They could even prove the light was "parked" by looking at the phase (the timing of the light waves), which showed a perfect pattern, like a fingerprint of a successful storage.

Why Does This Matter?

Think of this as a breakthrough for Quantum Computers and the Quantum Internet.

Storage: We need to store quantum information safely. This method shows we can "park" that information in a specific spot without needing complex, custom-built ramps for every single piece of data.
Simplicity: You can inject the information from the edge of the device (like plugging a USB drive into a port) rather than having to access the middle of the chip directly.
Scalability: Because this works with light, sound, or even electrons, we could potentially build massive storage systems where we can write and read data from the edges, making future quantum devices much easier to build.

In a nutshell:
The scientists found a way to use a "wave" from the edge of a system to gently guide a quantum state into a "trap" at a specific location, where it stays perfectly still. It's like using a gentle breeze to guide a leaf into a specific cup, rather than trying to grab the leaf with tweezers. This makes building future quantum memory devices much more feasible.

1. Problem Statement

The realization of robust quantum storage devices requires the ability to generate long-lived, spatially localized quantum states. While flat-band (FB) lattices offer a promising solution due to their macroscopically degenerate number of spatially compact eigenstates (Compact Localized States or CLS) and long lifetimes, a critical limitation exists: selective excitation.

The Challenge: In standard flat-band systems, the exact location of a CLS to be excited is largely determined by chance when using non-local excitation methods. There is no simple mechanism to target a specific CLS at an arbitrary position within the bulk of the lattice without complex, out-of-plane arrangements.
The Goal: Develop a method to selectively excite and trap CLS at desired positions using in-plane injection, enabling the storage of classical bits (or quantum states) in specific bulk plaquettes for quantum memory applications.

2. Methodology

The authors propose a mechanism based on hybridization between flat-band CLS and resonant dispersive plane waves (PW), controlled by localized on-site potentials.

Theoretical Framework:
- Models: Two 1D tight-binding models were studied: the Diamond chain and the Lieb ladder. Both possess a flat band at energy $E=0$ and chiral sublattices.
- Mechanism: The system is perturbed by introducing asymmetric on-site potentials ( $\pm \gamma$ ) at specific target unit-cells. This breaks the degeneracy and hybridizes the target CLS with extended zero-energy states confined to the minority sublattices.
- Excitation Scheme: Instead of local injection, an in-plane plane wave generator is attached to the edge of the lattice. This generator (a defect waveguide) launches a wave packet with a tailored quasi-momentum ( $k_0 = \pi$ ) that resonates with the flat band energy ( $E=0$ ).
- Trapping: When the resonant wave reaches a perturbed unit-cell, the hybridization allows energy to transfer from the propagating wave into the localized CLS. Once the potential is "switched off" (simulated via a sigmoid function), the state remains trapped in the CLS due to the flat band's dispersionless nature.
Experimental Implementation:
- Platform: Photonic waveguide arrays fabricated in borosilicate glass using femtosecond (fs) laser writing.
- Mapping: Propagation distance ( $z$ ) maps to time ( $t$ ) in the Schrödinger equation. The refractive index change maps to the on-site potential.
- Control:
  - Asymmetric Potential: Created by varying the writing velocity of the laser for specific waveguides (faster velocity = shallower guide = lower potential; slower = deeper = higher potential).
  - Plane Wave Generator: A side-coupled defect waveguide attached to the lattice edge.
  - Optimization: The authors optimized the coupling strength between the generator and the lattice, the distance of the generator, and the strength of the asymmetric potential ( $\Delta v$ ) to maximize trapping efficiency.

3. Key Contributions

Targeted Excitation Mechanism: The paper introduces a novel method to selectively excite CLS at arbitrary bulk positions using in-plane injection, overcoming the limitation of random excitation or complex out-of-plane setups.
Hybridization Strategy: It demonstrates how local potential engineering can hybridize compact states with dispersive waves, creating "defect eigenstates" that act as traps for incoming radiation.
Experimental Validation: The theoretical mechanism was successfully validated experimentally in two distinct geometries (Diamond and Lieb ladders), proving the feasibility of the approach in photonic systems.
Scalability: The approach is shown to be applicable to any platform supporting flat-band physics (acoustics, electronics, etc.) and can be extended to 2D and 3D systems.

4. Results

Numerical Simulations:
- Simulations confirmed that a single-site excitation at the edge defect generates a radiating wave packet that travels through the lattice.
- Upon reaching perturbed unit-cells (e.g., $n=3$ and $n=7$ ), the energy is efficiently trapped in the CLS.
- Diamond Chain: The trapped state showed high localization with orthogonal CLS, resulting in clean trapping at target sites.
- Lieb Ladder: Due to non-orthogonal CLS, the excitation induced some exponential decay into neighboring plaquettes. However, strong localization was still achieved, with the main intensity concentrated in the target sites ( $a, c, e$ ).
- Potential Switching: Applying a sigmoid function to turn off the potential after the wavefront passed successfully trapped the light, preventing leakage.
Experimental Observations:
- Diamond Chain: A clear, localized FB state was observed at the third plaquette. Interferograms confirmed the expected "out-of-phase" configuration between the upper and lower sites of the plaquette.
- Lieb Ladder: Despite the complexity of the geometry and non-orthogonal basis, a symmetric localized state was successfully excited at the third plaquette. Interferograms revealed the characteristic staggered phase structure (X-like profile) of the Lieb FB state.
- Optimization: The study identified optimal parameters for the writing velocity gradient ( $\Delta v \approx 0.08$ mm/s) and the defect distance ( $d \approx 18\text{--}19$ $\mu$ m) to maximize the purity of the trapped state.

5. Significance

Quantum Memory Architecture: This work provides a practical blueprint for quantum memory devices. By allowing the "slotting" of a quantum state into a specific location via a local "gate" (the asymmetric potential), it simplifies the engineering of quantum storage.
Read/Write Capability: The mechanism inherently supports both writing (trapping via potential) and reading (releasing by re-activating the potential to break resonance), offering a complete storage cycle.
Technological Advancement: The shift from out-of-plane to in-plane injection significantly simplifies the fabrication and control requirements for future quantum devices.
Broad Applicability: While demonstrated in photonics, the principles apply to any linear wave system, suggesting potential applications in acoustic metamaterials and electronic circuits for robust state storage.

In conclusion, the paper successfully bridges the gap between theoretical flat-band physics and practical device engineering, demonstrating a controllable, robust method for spatially localized quantum state storage.