🔬 physics

Experimental zero-added-loss multiplexing Bell-pair source for long-haul quantum networks

This paper demonstrates a scalable, high-rate quantum network source by successfully implementing zero-added-loss multiplexing (ZALM) across 16 frequency modes and 3.0 GHz time slots, achieving high-fidelity entanglement swapping without additional narrowband filtering.

Original authors: Yoshiaki Tsujimoto, Daiki Ichii, Rikizo Ikuta, Mikio Fujiwara, Masahiro Takeoka, Go Kato, Kentaro Wakui

Published 2026-10-27

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Yoshiaki Tsujimoto, Daiki Ichii, Rikizo Ikuta, Mikio Fujiwara, Masahiro Takeoka, Go Kato, Kentaro Wakui

Original paper licensed under CC BY 4.0 (https://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

The dream of a quantum internet is to connect computers and sensors in a way that is fundamentally secure and capable of solving problems no ordinary machine ever could. To make this happen, scientists need to send delicate packets of information, called quantum states, over long distances without losing their special properties. A major hurdle has been that these signals are incredibly fragile; as they travel through fiber-optic cables, they fade away, and if you try to boost them like a regular radio signal, you destroy the very information they carry. The solution lies in a process called entanglement swapping, which acts like a relay race for quantum information. Imagine two pairs of linked particles: if you bring one member from each pair together and perform a specific measurement, the two remaining particles, which have never met, instantly become linked. This allows a network to extend its reach step by step. However, doing this efficiently has been difficult because the process is probabilistic, meaning it often fails, and previous attempts to speed it up by running many channels at once have introduced too much noise or required complex, custom-built equipment that doesn't fit with existing telecommunications infrastructure.

A team of researchers has now demonstrated a practical way to overcome these bottlenecks, creating a source of entangled pairs that is both fast and compatible with standard fiber networks. They built a system that uses a technique called zero-added-loss multiplexing, which essentially runs many parallel quantum channels at the same time without the usual penalties of signal loss or confusion. Instead of trying to force a single, perfect channel to work harder, they utilized the natural ability of light to carry many different colors, or frequencies, simultaneously. By carefully shaping the laser pulses that create the particles and using standard commercial filters to separate these colors, they managed to run sixteen distinct channels in parallel. The researchers operated their system with a laser that pulses three billion times per second, a speed that is incredibly fast even by modern standards, allowing them to generate entangled pairs at a rate that was previously out of reach for this type of architecture.

The core of their experiment involved taking these sixteen parallel streams of light and performing the entanglement swapping process on all of them at once. They sent the particles through a setup where they could be measured and swapped, effectively linking distant parts of the network. The results were striking: across all sixteen channels, the system maintained a high average fidelity of 93.9 percent, meaning the quantum link remained strong and accurate despite the complexity of the process. In total, the system successfully swapped entangled pairs at a rate of 5.38 times per second. When the researchers accounted for the efficiency of their detectors and the specific way the system was designed, this translated to a potential generation rate of over 800 entangled pairs per second. This is a significant leap forward, showing that it is possible to scale up quantum networks using existing technology rather than waiting for entirely new, unproven inventions.

What makes this achievement particularly notable is how it balances speed with quality. In many previous attempts, increasing the speed of the process led to a drop in the quality of the connection, or required such narrow filtering that most of the signal was lost. Here, the researchers managed to keep the particles spectrally pure, meaning each color of light was distinct and clean, without needing extra, lossy filters. They achieved this by precisely engineering the length of the crystal that generates the light and shaping the laser pulses to fit perfectly within the bandwidth of standard filters used in telecommunications. This approach ensures that the system can be integrated into the global fiber-optic network without needing a complete overhaul of the hardware. The team also showed that their method works well even when the laser pulses are packed very closely together in time, a condition that usually causes errors in other systems.

Looking ahead, the researchers used their experimental data to model what might be possible if they could reduce the remaining sources of loss and increase the number of channels. Their simulations suggest that by using even denser frequency spacing and improving the efficiency of the detectors, the system could eventually reach a rate of one billion entangled pairs per second while maintaining high fidelity. This would be a critical threshold for building practical, long-distance quantum repeaters, the devices needed to extend quantum networks across continents. While the current experiment is a proof of concept, it establishes the key capabilities required to move from theory to reality. The work demonstrates that the path to a high-rate quantum internet does not require abandoning current infrastructure but rather refining how we use the vast capacity already available in our fiber-optic cables. By proving that high-speed, high-fidelity entanglement swapping is feasible with off-the-shelf components, this research provides a scalable route toward the practical realization of a global quantum network.

Problem Statement
The realization of high-rate, high-fidelity quantum networks is a central challenge in quantum information science. While frequency multiplexing offers a promising route to scale entanglement distribution by leveraging the broad spectral bandwidth of spontaneous parametric down-conversion (SPDC) and existing dense wavelength-division multiplexing (DWDM) infrastructure, previous experimental demonstrations have been limited. Specifically, implementing "zero-added-loss multiplexing" (ZALM)—a scheme proposed to enable quasi-deterministic entanglement generation via local frequency-multiplexed entanglement swapping—has remained an experimental hurdle. The primary challenges involve achieving high-fidelity entanglement swapping across densely multiplexed time-frequency modes without introducing switching losses or degrading interference visibility. Furthermore, previous multiplexing demonstrations were often restricted to direct distribution of entangled pairs or utilized architectures and repetition rates (tens of MHz) incompatible with the GHz-level requirements of the ZALM architecture.

Methodology
The authors experimentally demonstrated a ZALM Bell-pair source based on time-frequency-multiplexed entanglement swapping in the telecom C-band. The experimental setup involved two entangled photon pair sources (EPS I and II) utilizing type-0 periodically poled lithium niobate waveguides (PPLN/Ws) within Sagnac interferometers.

Key methodological components included:

Pump Source: A pump laser operating at a repetition frequency of 3.0 GHz, generated via second harmonic generation (SHG) of fundamental telecom pulses. The pulse bandwidth was engineered (87 GHz FWHM) to fully utilize the phase-matching bandwidth of the SPDC crystals while minimizing pulse duration.
Frequency Multiplexing: The system employed off-the-shelf 50-GHz DWDM filters to extract 16 parallel frequency modes in both signal and idler arms. To ensure spectral purity without additional narrowband filtering, the SPDC crystal length (2.0 cm) was carefully selected to match the pump bandwidth and filter specifications, ensuring each frequency channel was spectrally single-mode.
Entanglement Swapping: Idler photons from the two sources were interfered on a 50:50 beamsplitter to perform a time-frequency-resolved Bell-state measurement (BSM). Coincidence detection of idler photons in specific frequency modes heralded the generation of entangled signal photon pairs.
Detection: Superconducting single-photon detectors (SSPDs) with an average timing jitter of 39 ps were used. The 3.0 GHz repetition rate (333 ps pulse interval) was chosen to satisfy the condition $1/f_{rep} \gg \Delta\tau_j$ , preventing temporal overlap of pulses that would degrade Hong-Ou-Mandel (HOM) interference visibility.

Key Contributions

High-Fidelity Multiplexed Swapping: The study achieved entanglement swapping across 16 parallel frequency modes with a high average fidelity of $93.9 \pm 1.4\%$ .
Spectral Purity via Off-the-Shelf Components: The authors demonstrated that polarization-entangled photon pairs in each frequency mode could be spectrally single-mode using only standard 50-GHz DWDM filters, eliminating the need for complex, custom narrowband filtering.
GHz-Level Operation: By operating the pump at 3.0 GHz, the experiment fully exploited the temporal degree of freedom, a significant step toward the 30 GHz repetition rates proposed for ideal ZALM architectures.
Performance Regime: The combination of frequency and time multiplexing resulted in a total swapping rate of $5.38 \pm 0.17$ pairs s $^{-1}$ , corresponding to a ZALM Bell-pair rate of $8.2 \times 10^2$ pairs s $^{-1}$ (after correcting for detection losses in signal modes).

Results

Interference Visibility: The average HOM interference visibility across all 16 modes was measured at $95.0 \pm 0.7\%$ , indicating nearly perfect indistinguishability of the heralded idler photons.
State Fidelity: The reconstructed density matrices of the initial states showed fidelities of $\sim98.8\%$ to the maximally entangled Bell state. The swapped states exhibited an average fidelity of $93.9 \pm 1.4\%$ .
Rate-Fidelity Trade-off: The experimental data points lie above the theoretical rate-fidelity trade-off curves of previous broadband entanglement swapping experiments, demonstrating a superior performance regime enabled by the ultrafast repetition rate and multiplexing.
Simulation Projections: Theoretical simulations based on the experimental parameters suggest that by fully exploiting the 8.7-THz SPDC spectrum and utilizing a BSM capable of discriminating two Bell states, the ZALM Bell-pair rate could reach $10^5$ pairs s $^{-1}$ while maintaining 91% fidelity. Further improvements in system loss and spectral purity (e.g., using photon-number-resolving detectors) could theoretically push rates toward $10^9$ pairs s $^{-1}$ .

Significance
The paper establishes the key experimental capabilities required for the ZALM architecture, specifically the ability to perform high-fidelity entanglement swapping across densely multiplexed time-frequency modes. By demonstrating a scalable route using commercially available DWDM components and high-repetition-rate operation, the work constitutes a critical enabling step toward practical, high-rate quantum repeaters and long-haul quantum networks. The results place the experiment in a performance regime previously unaccessed, bridging the gap between theoretical proposals for ZALM and practical implementation.

More like this