⚛️ quantum physics

Integrated Photon-Memory Entanglement Generation using Dual Photonic Resonators

This paper demonstrates the first integrated photonic platform using dual silicon-carbide microring resonators to generate and verify high-dimensional entanglement between telecom photons and a quantum memory, achieving a peak on-chip entanglement rate of 5.6 kEbits s $^{-1}$ and establishing a scalable route toward chip-scale quantum repeaters.

Original authors: Alexander Kolar, Ian Chin, Conner Fong, Daniil M. Lukin, Melissa A. Guidry, Milan Palei, Jelena Vučković, Tian Zhong

Published 2026-07-03

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Alexander Kolar, Ian Chin, Conner Fong, Daniil M. Lukin, Melissa A. Guidry, Milan Palei, Jelena Vučković, Tian Zhong

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 build a super-fast, ultra-secure internet that uses light instead of electricity. To make this work over long distances, you need two special ingredients: a machine that creates pairs of "entangled" light particles (photons) and a safe place to store them temporarily while they travel.

The problem with current technology is like trying to fit a square peg into a round hole. The machines that create these light particles usually speak a different "language" (frequency) than the machines that store them. To make them work together, scientists usually have to force them to match using complex filters or converters, which slows everything down and loses valuable information.

The New Solution: A Perfectly Matched Set
This paper describes a breakthrough where researchers built a single, integrated chip that solves this mismatch problem naturally. Think of it like building a custom-made lock and key from the exact same mold, rather than trying to file down a key to fit a lock.

Here is how they did it, using simple analogies:

1. The Twin Resonators (The Factory and the Vault)

The researchers used a material called Silicon Carbide to build two tiny, circular tracks called "microring resonators."

The Factory (Source Resonator): One ring acts as a factory. When they shine a laser into it, it splits the light into pairs of entangled photons. It's like a machine that stamps out two matching tickets at the exact same time.
The Vault (Memory Resonator): The second ring is a vault. It is made of the exact same design as the factory but is coated with a special crystal containing Erbium ions. This crystal acts like a sponge that can catch and hold the light particles.

Because both rings are built from the same blueprint, they naturally "speak the same language." The light coming out of the factory fits perfectly into the vault without needing any awkward adapters or filters.

2. The "Sponge" Effect

Usually, a sponge (the memory) is too rough or the wrong size to catch delicate light particles without damaging them. In this experiment, the researchers engineered the "sponge" (the Erbium crystal) to be perfectly smooth and sized to catch the light.

They managed to get the "sponge" to grab the light very efficiently. They call this a "cooperativity" of 1.9, which is like saying the sponge is almost twice as good at holding the light as the light is at escaping it.
They proved they could store the light and retrieve it later without losing the special "entanglement" connection between the two light particles. It's like putting a pair of magic dice in a box, closing it, and later opening it to find they still always roll the same number, even though they were separated.

3. The High-Speed Data Highway

The most exciting part is how much information they can send at once.

The Qubit vs. The Qudit: Usually, these systems send one bit of information at a time (like a light switch being on or off). This new system can send a whole "handshake" of information at once.
The Analogy: Imagine a standard internet cable sends one letter at a time. This new system can send a whole book, or even a library, in a single flash of light.
They demonstrated this by using up to 63 different time slots (like 63 different lanes on a highway) simultaneously.
The Result: They achieved a record-breaking speed for this type of technology, generating entanglement at a rate of 5,600 "entanglement bits" per second. Furthermore, for every single light particle they detected, they got back about 5.1 bits of information. This is a huge leap in efficiency, meaning they get much more value out of every photon they use.

Why This Matters

The paper claims this is a major step toward building a "Quantum Internet." By combining the light generator and the memory storage into one seamless, chip-sized package that works with standard fiber-optic cables (telecom wavelengths), they have created a practical building block.

Instead of trying to force different technologies to work together, they built a unified system where the source and the storage are naturally compatible. This makes the system faster, more efficient, and ready to be scaled up into a real network that could eventually connect quantum computers over long distances.

Technical Summary: Integrated Photon-Memory Entanglement Generation using Dual Photonic Resonators

Problem Statement
Scalable quantum networks require the efficient generation and storage of entanglement between photonic qubits and quantum memories, a prerequisite for quantum repeaters. While significant progress has been made in integrated photonic pair sources and integrated quantum memories, these technologies have largely developed independently. A primary barrier to integration is spectral matching: quantum memories typically operate over narrow optical bandwidths, whereas photon-pair sources often produce photons with broader spectra. Bridging this gap traditionally necessitates spectral filtering, cavity engineering, or frequency conversion, which reduce entanglement generation rates and increase system complexity. Furthermore, existing unified platforms that combine high-brightness entangled-photon sources with quantum memories operating at telecommunications wavelengths have not yet been realized.

Methodology
The authors demonstrate a unified on-chip platform based on dual silicon carbide (SiC) microring resonators fabricated on a single 4H-SiC-on-insulator wafer. The architecture employs two self-similar resonators with identical design and fabrication processes to ensure intrinsic spectral compatibility:

Source Resonator: Functions as a cavity-enhanced spontaneous four-wave mixing (SFWM) source. A quasi-continuous-wave pump drives the generation of non-degenerate signal and idler photon pairs. The signal photon is spectrally aligned to the memory resonator.
Memory Resonator: Functions as a cavity-enhanced Atomic Frequency Comb (AFC) quantum memory. It is evanescently coupled to an ensemble of $^{167}\text{Er}^{3+}$ ions doped in a $\text{Y}_2\text{SiO}_5$ crystal.
Integration Strategy: A polished $^{167}\text{Er}^{3+}:\text{Y}_2\text{SiO}_5$ crystal is bonded directly onto the SiC chip via a thin hydrogen silsesquioxane (HSQ) layer. This bonding increases the loss of the memory resonator intentionally to broaden its linewidth (exceeding the photon bandwidth) while enabling strong collective light-matter coupling.
Spectral Alignment: The system achieves spectral matching through temperature tuning of the source resonator and magnetic field tuning of the erbium ensemble at millikelvin temperatures. No external spectral filtering or frequency conversion is required.
Experimental Setup: The device operates in a dilution refrigerator (7 mK) with a 1.238 T magnetic field. Hyperfine initialization prepares the erbium ensemble in the $|m = 7/2\rangle$ state. Photon pairs are generated using a pulsed pump synchronized to the AFC storage cycle.

Key Results

Spectral Matching and Cooperativity: The dual-resonator design successfully aligns the source signal resonance with the memory cavity and the erbium absorption feature. The memory resonator achieves an ensemble cooperativity of 1.9 with hyperfine initialization, representing a 3.4-fold increase over non-initialized states.
Photon-Pair Generation: The source resonator demonstrates a pair generation rate of $6.0 \times 10^5 \text{ s}^{-1} \text{ mW}^{-2}$ with a coincidence-to-accidental ratio (CAR) of 60. Two-photon Franson interference yields a visibility of $96.2 \pm 3.5\%$ , confirming high-quality entanglement.
Quantum Storage: The memory demonstrates a total storage-and-retrieval efficiency of 7.7% for weak coherent pulses (storage time 1 $\mu$ s), compared to 2.0% without hyperfine initialization. Time-bin qubit storage shows an interference visibility of $96.9 \pm 2.0\%$ , and decoy-state analysis estimates a single-photon storage fidelity lower bound exceeding 98.5%.
Photon-Memory Entanglement: The system generates and verifies entanglement between a propagating telecom signal photon and the stationary memory. Franson interference between the idler photon and the memory-retrieved signal photon yields a single-pair visibility of $88.1 \pm 10.6\%$ , confirming the faithful preservation of entanglement.
High-Dimensional Qudit Entanglement: By exploiting the multimode capacity of the AFC memory, the authors demonstrate high-dimensional entanglement spanning up to 63 temporal modes.
- Photon Information Efficiency: Up to 5.1 Ebits per detected photon (at $d=63$ ).
- Entanglement Rate: A peak on-chip photon–memory entanglement rate of 5.6 kEbits s $^{-1}$ (optimized at $d=12$ ).

Significance and Claims
The paper establishes an integrated platform for photon-memory entanglement that addresses the critical challenge of spectral matching without the need for external filtering or frequency conversion. The authors claim this work provides:

A scalable route toward chip-scale quantum networking hardware compatible with existing telecommunications infrastructure.
A practical demonstration of the building block required for absorptive quantum repeaters: high-fidelity entanglement between a telecom photon and a quantum memory.
A versatile resource for memory-enabled networks, capable of generating high-dimensional qudit entanglement optimized for either maximum information capacity or peak throughput.
A foundation for future multi-node quantum interconnects, where the integrated architecture naturally extends to interference between independent source-memory modules.

The authors note that the current entanglement rate is limited by coupling losses between the integrated devices and fibers rather than intrinsic device performance, suggesting that improved packaging could yield rates two orders of magnitude higher. The work also highlights the potential for using this dual-ring architecture to study fundamental light-matter interactions, such as quantum-enhanced spectroscopy and squeezing-enhanced coupling.

1. The Twin Resonators (The Factory and the Vault)

2. The "Sponge" Effect

3. The High-Speed Data Highway

Why This Matters

More like this