🔬 optics

Light Storage in Light Cages: A Scalable Platform for Multiplexed Quantum Memories

This paper demonstrates a scalable platform for multiplexed quantum memories by integrating multiple 3D-nanoprinted "light cage" waveguides into a single cesium vapor cell to achieve efficient, spatially distributed photon storage.

Original authors: Esteban Gómez-López, Dominik Ritter, Jisoo Kim, Harald Kübler, Markus A. Schmidt, Oliver Benson

Published 2026-02-11

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Esteban Gómez-López, Dominik Ritter, Jisoo Kim, Harald Kübler, Markus A. Schmidt, Oliver Benson

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 are trying to run a high-speed digital highway, but there’s a problem: the cars (which represent quantum information) are moving so fast that they often arrive at their destination at the wrong time, causing massive traffic jams. To make a quantum computer or a quantum internet work, we need "parking garages" where we can pull a car off the road, hold it for a few nanoseconds, and then release it exactly when the rest of the traffic is ready.

In the world of physics, these parking garages are called Quantum Memories.

This paper describes a breakthrough in building these "parking garages" using a technology they call "Light Cages." Here is the breakdown of how it works.

1. The Problem: The "Giant, Slow Pipe" Dilemma

Until now, scientists have tried to build these memories using long, hollow glass fibers. Think of these like massive, miles-long tunnels. They work, but they have two big problems:

The Filling Problem: It’s like trying to fill a massive tunnel with smoke; it takes forever for the "stuff" (the atoms) to drift inside and fill the space.
The Size Problem: They are huge and clunky. You can't exactly fit a mile-long tunnel onto a tiny computer chip.

2. The Solution: The "Light Cage"

Instead of a giant tunnel, the researchers used a high-tech 3D printer (using a process called two-photon polymerization) to print microscopic, intricate structures called Light Cages.

The Analogy: Imagine instead of a massive tunnel, you have a tiny, delicate birdcage made of microscopic wires.

Because the cage is so small and has "side doors" (openings in the structure), the atoms (the "smoke") can drift inside almost instantly.
Because it’s printed on a silicon chip, you can print hundreds of these tiny cages side-by-side.

3. How the "Parking" Works (EIT)

The researchers use a trick called Electromagnetically Induced Transparency (EIT).

The Analogy: Imagine a crowd of people (atoms) standing in a room, making it impossible to walk through. If you try to send a light pulse through, it hits the crowd and stops.
However, if you shine a special "control laser" into the room, it acts like a choreographer. The choreographer tells the crowd to move in a specific way that creates a clear path. As the light pulse enters this path, the "choreographer" can suddenly change the instructions, causing the light to slow down and eventually "freeze" into a wave of atomic motion. When you want the light back, you just give a new command, and the light "melts" out of the atoms and continues on its way.

4. The Big Discovery: Multiplexing

The most exciting part of this paper is Multiplexing. Because these cages are so small and easy to print, the researchers didn't just make one memory; they made a whole "parking lot" on a single chip.

They proved that they could print multiple cages and they all worked almost exactly the same way. This is like being able to mass-produce identical, tiny parking garages on a single microchip. This allows us to handle many different pieces of quantum information at the same time, rather than just one.

Summary: Why does this matter?

If we want to build a Quantum Internet (a way to send unhackable messages across the world) or a Quantum Computer (a machine millions of times faster than your laptop), we need to be able to synchronize information.

By moving from "giant, slow tunnels" to "tiny, high-speed light cages on a chip," these scientists have provided a blueprint for a scalable, compact, and efficient way to manage the lightning-fast world of quantum data.

Technical Summary: Light Storage in Light Cages: A Scalable Platform for Multiplexed Quantum Memories

1. Problem Statement

Quantum memories are critical components for photonic quantum technologies, specifically for long-distance quantum communication (via quantum repeaters) and photonic quantum computing (for photon synchronization and feed-forward operations). While hot atomic vapors using Electromagnetically Induced Transparency (EIT) offer a scalable and low-complexity platform, integrating them with light-guiding structures presents significant engineering hurdles.

Current solutions, primarily Hollow-Core Fibers (HCFs), suffer from several drawbacks:

Slow Filling Times: Loading atomic vapors into HCF cores can take months.
Complexity: Using laser-cooled atoms to load fibers adds significant experimental overhead.
Design Constraints: The fiber-pulling fabrication process limits geometric versatility.
Integration Issues: HCFs are difficult to integrate into compact, on-chip photonic architectures.

2. Methodology

The researchers propose and demonstrate a platform based on 3D-nanoprinted "Light Cages" (LCs).

Fabrication: LCs are anti-resonant hollow-core waveguides fabricated on a silicon substrate using two-photon polymerization (TPP). This allows for arbitrary 3D geometries and high reproducibility. To protect the polymer from chemically reactive Cesium (Cs) vapor, the structures are coated with 100 nm of alumina ( $\text{Al}_2\text{O}_3$ ) via Atomic Layer Deposition (ALD).
Quantum Memory Protocol: The team utilized the EIT effect in a Cs vapor cell. They implemented a storage protocol where a signal pulse is mapped into a spin-wave excitation within the atomic ensemble using a control laser pulse, and subsequently retrieved by a third control pulse.
Multiplexing Approach: To demonstrate scalability, multiple LCs with identical geometries were printed on a single chip, spaced only hundreds of micrometers apart, and tested for consistent performance.

3. Key Contributions

Demonstration of On-Chip EIT Storage: The first successful storage of faint coherent light pulses in a nanoprinted LC structure for several hundred nanoseconds.
Scalable Architecture: Proven ability to integrate multiple, highly reproducible quantum memories on a single chip.
Design Versatility: Showcased a platform that bypasses the filling-time and geometric limitations of traditional hollow-core fibers.
Stability: Demonstrated that alumina-coated LCs remain stable in a reactive Cs atmosphere for over five years.

4. Results

Storage Performance: The system achieved storage times of several hundred nanoseconds, with a measured $1/e$ -lifetime of 83(2) ns.
Fractional Delay: The memory achieved a fractional delay ( $F = t_{\text{storage}}/\Delta t_s$ ) of approximately 4, which is comparable to performance seen in hollow-core photonic crystal fibers.
Bandwidth: The memory demonstrated a bandwidth of 35.2(6) MHz.
Spin Precession: The researchers observed damped oscillatory behavior in the storage efficiency. This was attributed to spin precession caused by residual magnetic fields (Earth's field and heating wires), which provides a potential mechanism for magnetic polarization control of stored light.
Efficiency vs. Control Power: A counterintuitive finding showed that increasing control power ( $P_c$ ) actually decreased efficiency for short storage times. This is due to spatial pulse compression: higher $P_c$ reduces the group velocity so significantly that the signal pulse becomes longer than the physical LC, meaning only a fraction of the pulse is captured.
Reproducibility: Two different LCs (LC-A and LC-B) on the same chip showed nearly identical storage lifetimes (86 ns vs 87 ns), confirming the high precision of the nanoprinting process.

5. Significance and Future Outlook

This work represents a significant advancement toward spatially multiplexed quantum memories. By moving from bulky fibers to integrated, nanoprinted light cages, the researchers have provided a path toward:

Quantum Repeaters: Enabling parallel single-photon synchronization.
Photonic Quantum Computing: Providing scalable delay units for measurement-based quantum computing.
Enhanced Integration: The ability to combine these memories with other on-chip photonic components (like phase plates or fiber-coupled V-grooves) positions this platform as a "plug-and-play" solution for future quantum networks.

Future improvements will focus on increasing optical depth (via longer LCs or Light-Induced Atomic Desorption) and reducing waveguide roughness to minimize propagation losses.