World’s First Demonstration of Entanglement Swapping Using Sum-Frequency Generation Between Single Photons

November 6, 2025
(Japanese version released on October 8, 2025)

National Institute of Information and Communications Technology

Highlights

  • Demonstrated the world’s first entanglement swapping using sum-frequency generation between single photons, one of the fundamental quantum communication protocols.
  • Successfully observed sum-frequency generation between single photons with a high signal-to-noise ratio, made possible by NICT’s state-of-the-art technologies.
  • Expected to contribute to the miniaturization and efficiency improvement of photonic quantum information processing circuit, as well as the extension of transmission distance in device independent quantum key distribution.
The National Institute of Information and Communications Technology (NICT, President: TOKUDA Hideyuki, Ph.D.) has successfully demonstrated entanglement swapping (one of the key quantum communication protocols) using sum-frequency generation (SFG) between single photons for the first time.
Although nonlinear optical effects of single photons have long been theoretically recognized as powerful tools for advancing quantum communication protocols, such effects are extremely weak at the single-photon level and had never been applied for quantum operations. By combining NICT’s state-of-the-art technologies including high-speed-clocked entangled photon-pair sources, low-noise superconducting nanowire single-photon detectors, and a high-efficiency nonlinear optical crystal, the research team succeeded in observing SFG between single photons with an unprecedented signal-to-noise ratio. Using this effect, they achieved the first experimental demonstration of entanglement swapping via single-photon SFG.
This achievement is expected to pave the way for miniaturized and efficient photonic quantum information processing circuit, as well as long-distance device independent quantum key distribution.
The results were published in Nature Communications on October 7, 2025 (Tuesday).

Background

Figure 1 (a) Conventional entanglement swapping and (b) SFG-based entanglement swapping
a When photon-pair generation is probabilistic, a two-photon interference measurement between A2 and B1 alone cannot distinguish successful events from unsuccessful ones. Therefore, additional measurements are required to verify the presence of one photon each in modes A1 and B2.
b When the SFG photon is detected, it indicates that there is one photon in each of modes A2 and B1, allowing the successful entanglement swapping event to be identified.
In the field of quantum information processing such as quantum communication and quantum computing, two-qubit gate operations are fundamental building blocks. In optical implementations, two-photon interference has been used to realize such operations. While this method allows for a relatively simple experimental setup using only a standard beam splitter and photon detectors, it suffers from a major limitation: Unless the existence of a photon pair obtained through entanglement swapping is confirmed by a measurement (and thus destroyed), the fidelity becomes low (see Figure 1a), limiting the range of applications.
To overcome this limitation, a theoretical scheme based on entanglement swapping using sum-frequency generation (SFG) between single photons has been proposed (see Figure 1b) [1]. In this approach, by detecting the photon generated via SFG between two single photons (the SFG photon), it becomes possible to perform high-fidelity entanglement swapping without destroying the resulting entangled photon pair. This feature offers significant advantages for loophole-free Bell tests and long-distance device independent quantum key distribution.
However, although SFG between single photons was first reported in 2014 [2], the detected signal at that time was extremely weak and buried in noise. Therefore, to apply this effect to entanglement swapping, it was essential to dramatically improve the signal-to-noise ratio (SNR) of the detected SFG signal.

Achievements

In this study, the research team constructed an experimental setup by combining NICT’s state-of-the-art technologies including high-speed-clocked entangled photon-pair sources [3,4], low-noise superconducting nanowire single-photon detectors (SNSPDs)  [5,6], and a high-efficiency nonlinear optical crystal [7] (see Figure 2 and Appendix for details).
As a result, the SFG photons were detected with a high SNR (see Figure 3a), achieving nearly an order of magnitude improvement compared with the previous study [2]. Furthermore, the researchers confirmed the presence of strong entanglement in the final state (see Figure 3b), estimating a lower bound of the fidelity to the maximally entangled state as 0.770 ± 0.076.
These results represent the world’s first experimental demonstration of entanglement swapping via sum-frequency generation between single photons. This achievement marks a significant step forward in photonic quantum information processing and is expected to serve as an important guideline for the development of next-generation nonlinear optical devices.
Figure 2 Experimental setup for the SFG-based entanglement swapping
One entangled photon pair is generated from each of the two sources, EPS I and EPS II, and a gate operation based on sum-frequency generation between single photons is performed using the SFG-based Bell-state analyzer (SFG-BSA).
Figure 3 Experimental results
a Detection signal of the SFG photon.
b Two-photon polarization correlation of the swapped state. H, V, D and A represent horizontal, vertical, diagonal and anti-diagonal polarizations, respectively.

Future prospects

To apply the current system to the more advanced quantum information protocols beyond entanglement swapping, further improvement in the SNR will be required. In the future, the research team aims to enhance a nonlinear optical efficiency, leading to the miniaturization and efficiency improvement of photonic quantum information processing circuits and the extension of transmission distance in device independent quantum key distribution.

Researchers

TSUJIMOTO Yoshiaki
Quantum ICT Laboratory, Advanced ICT Research Institute
WAKUI Kentaro
Space-Time Standards Laboratory, Radio Research Institute
KISHIMOTO Tadashi
Terahertz Technology Research Center, Beyond 5G Research and Development Promotion Unit
MIKI Shigehito
Superconductive ICT Device Laboratory, Advanced ICT Research Institute
YABUNO Masahiro
Superconductive ICT Device Laboratory, Advanced ICT Research Institute
TERAI Hirotaka
Superconductive ICT Device Laboratory, Advanced ICT Research Institute
FUJIWARA Mikio
Quantum ICT Collaboration Center
KATO Go
Quantum ICT Laboratory, Advanced ICT Research Institute

Article information

Authors: Yoshiaki Tsujimoto*, Kentaro Wakui, Tadashi Kishimoto, Shigehito Miki, Masahiro Yabuno, Hirotaka Terai, Mikio Fujiwara, Go Kato
(*Corresponding author)
Title: Experimental entanglement swapping through single-photon χ(2) nonlinearity
Journal: Nature Communications
DOI: 10.1038/s41467-025-63785-5

This work was supported by the Japan Society for the Promotion of Science (JP18K13487, JP20K14393, JP22K03490) and R&D of ICT Priority Technology Project (JPMI00316).
 

References

[1] N. Sangouard et al., “Faithful entanglement swapping based on sum-frequency generation”, Phys. Rev. Lett. 106, 120403 (2011). 
[2] T. Guerreiro et al., “Nonlinear interaction between single photons”, Phys. Rev. Lett. 113, 173601 (2014).
[3] K. Wakui et al., “Ultra-high-rate non-classical light source with 50 GHz-repetition-rate mode-locked pump pulses and multiplexed single-photon detectors”, Opt. Exp. 28, 22399 (2020). 
[4] Y. Tsujimoto et al., “Ultra-fast Hong-Ou-Mandel interferometry via temporal filtering,” Opt. Exp. 29, 37150 (2021). 
[5] T. Yamashita et al., “Superconducting nanowire single-photon detectors with non-periodic dielectric multilayers”, Sci. Rep. 6, 35240 (2016).
[6] S. Miki et al., “Stable, high-performance operation of a fiber-coupled superconducting nanowire avalanche photon detector”, Opt. Exp. 25, 6796 (2017).
[7] T. Kishimoto et al., “Highly efficient phase-sensitive parametric gain in periodically poled LiNbO3 ridge waveguide”, Opt. Lett. 41, 1905 (2016).

Appendix

Figure 2 Experimental setup for the SFG-based entanglement swapping (Reprinted)
One entangled photon pair is generated from each of the two sources, EPS I and EPS II, and a gate operation based on sum-frequency generation between single photons is performed using the SFG-based Bell-state analyzer (SFG-BSA).
In this study, the research team combined NICT’s state-of-the-art technologies including a high-speed-clocked entangled photon-pair sources [3,4], low-noise SNSPDs [5,6], and a high-efficiency nonlinear optical crystal [7] to construct an experimental setup (Figure 2) and achieved the world’s first demonstration of entanglement swapping via SFG between single photons.
In the experiment, two entangled photon pairs were first generated. One photon from each pair was then input into PPLN/W to perform sum-frequency generation. Finally, detecting the SFG photon with the SNSPD heralds the formation of entanglement between the remaining photons, thus completing the entanglement swapping.
Several novel technologies were developed for this experiment. To maximize nonlinear optical efficiency, a long 63-mm PPLN/W was fabricated and placed inside a Sagnac interferometer to perform the SFG-based gate operation. The generated SFG photons were detected with the low-noise SNSPD having a dark count of only 0.15 Hz. In addition, an electro-optic comb was employed to excite the entangled photon pairs at a high-speed clock of 1.0 GHz, optimized for this experiment. These advancements resulted in a nearly one order-of-magnitude improvement in the SNR compared with previous studies [2].
Figure 3 shows the detection signal of the SFG photon and the polarization correlations of the photon pairs obtained by entanglement swapping. From this experimental data, the lower bound of fidelity to the maximally entangled state was estimated as 0.770 ± 0.076, confirming that the swapped state is strongly entangled.
Figure 3 Experimental results (Reprinted)
a Detection signal of the SFG photon.
b The two-photon polarization correlation of the swapped state. H, V, D and A represent horizontal, vertical, diagonal and anti-diagonal polarizations, respectively.
These results represent a major step forward in photonic quantum information processing and are expected to provide important guidance for the development of next-generation nonlinear optical devices. 
To apply this approach to more advanced quantum information protocols beyond entanglement swapping, further improvement in the SNR will be required. Specifically, simulations conducted in this study indicate that at least a threefold enhancement of the nonlinear efficiency would be needed to observe loophole-free violation of Bell’s inequality, and approximately a 50-fold improvement would be necessary for obtaining positive key rate by device independent QKD. Such enhancements could be achieved, for example, by incorporating an optical cavity structure into the PPLN waveguide to strengthen the electric field.

Glossary

Entanglement Swapping
A quantum communication protocol in which two pairs of entangled photons are connected to produce a single entangled photon pair. A Bell-state measurement is performed on one photon from each pair, resulting in creation of entanglement between the remaining photons. In this study, the Bell-state measurement is implemented using sum-frequency generation between single photons.
Sum-Frequency Generation (SFG)
When two light fields of frequencies ω1 and ω2 are simultaneously input into a second-order nonlinear optical crystal, light at the sum frequency ω3 ( = ω1 + ω2) is generated with an intensity proportional to the product of the input intensities. For extremely weak input light, such as single photons, the sum-frequency photon is generated with an efficiency proportional to the product of the average photon numbers of the inputs.
Nonlinear Optical Effect
When light is incident on a nonlinear optical medium, a nonlinear polarization is induced inside the medium. The nonlinear polarization can be expressed as a power series of the input electric field. For example, effects arising from the second-order term of the electric field are called second-order nonlinear optical effects, in which the output light frequency differs from the input. Typical examples include second-harmonic generation, sum-frequency generation, and difference-frequency generation. 
Device Independent Quantum Key Distribution
In conventional quantum key distribution (QKD), security relies on the assumption that the devices operate as intended by users. In device independent QKD, this assumption is unnecessary. Instead, security is guaranteed by verifying loophole-free violations of Bell’s inequalities. In practice, photon loss in optical fibers causes a detection loophole. By generating an additional entangled photon pair and performing SFG-based Bell-state measurement (as in Figure 1b), events where photons are lost during transmission can be excluded, closing the detection loophole. 
Two-Qubit Gate Operation
A quantum gate operation involving two or more qubits, such as a CNOT gate. This operation enables quantum operations such as entanglement generation. Realizing a quantum computer requires not only single-qubit gates but also two-qubit gate operations.
Two-photon interference
Consider two input photons entering a beam splitter (transmittance = reflectance = 1/2) as in the center of Figure 1a. If the photons are distinguishable, the probability of both photons exiting the same output port is 1/2, and the probability of one photon exiting each port is also 1/2. If the photons are indistinguishable, the latter event (one photon per output port)  disappears due to the destructive interference.
Fidelity
A measure ranging from 0 to 1 that indicates how close a state is to the ideal maximally entangled state. Fidelity greater than 0.5 implies the presence of entanglement.
Loophole-Free Bell test
Entangled photon pairs can violate Bell’s inequality. However, in real experiments, various loopholes exist, especially the detection loophole, arising from imperfect photon detection. To close this loophole, at least two-thirds of the generated entangled photon pairs must be detected; otherwise, It cannot be ruled out that the inequality is not violated when considering photons that were not detected.
Sagnac interferometer
In a Mach-Zehnder interferometer, photons travel along two different paths, requiring active feedback stabilization to maintain path and phase differences. In contrast, a Sagnac interferometer has a folded structure in which photons travel along the same path in opposite directions. This ensures that each photon experiences identical phase changes, allowing the interferometer to remain stable without active feedback.

Technical Contact

TSUJIMOTO Yoshiaki
Quantum ICT Laboratory
Koganei Frontier Research Center
Advanced ICT Research Institute

Media Contact

Press Office
Public Relations Department