Long-distance distribution of atom-photon entanglement based on a cavity-free cold atomic ensemble

TL;DR

This work tackles long-distance atom–photon entanglement distribution by building a cavity-free, cold-atom quantum node that generates atom–photon entanglement in the 780 nm band and converts the photon to telecom wavelength via a polarization-independent quantum frequency converter. The system achieves a high initial retrieval efficiency (~$50\%$) and a memory lifetime of $160\,\mu\mathrm{s}$, enabling entanglement to survive fiber transmission up to $20\,\mathrm{km}$ with fidelity exceeding $0.80$; the telecom photons are transmitted to long fibers with polarization compensation and low-noise conversion, yielding a signal-to-noise ratio of up to $6.9$ for path lengths up to $100\,\mathrm{km}$. The demonstrated platform offers a scalable route toward hundred-kilometer quantum networks and provides a foundation for distant memory–memory entanglement, contingent on improvements in memory efficiency and coherence to reach unitary performance. Overall, the cavity-free atomic ensemble + PIQFC approach presents a practical, integrable path to long-distance quantum networking with solid-state and atomic systems.

Abstract

Constructing a quantum memory node with the ability of long-distance atom-photon distribution is the essential task for future quantum networks, enabling distributed quantum computing, quantum cryptography and remote sensing. Here we report the demonstration of a quantum-network node with a simple cavity-free cold atomic ensemble. This node gives an initial retrieval efficiency of approximately 50\% and memory lifetime of 160 $μ$s for atomic qubits. With the aid of a high-efficiency and polarization-independent quantum frequency conversion (QFC) module, the generated entangled photon in the node at 780-nm wavelength is converted to telecom S band at 1522 nm, enabling atom-photon distribution over long distance. We observe an entanglement fidelity between the atoms and telecom photon exceeding 80\% after photon transmission over 20-km fiber, the remaining infidelity being dominated by atomic decoherence. The low-noise QFC with an external efficiency up to 48.5\% gives a signal-to-noise-ratio of 6.9 for transmitted photons with fiber length up to 100 km, laying the cornerstone for entanglement distribution at a hundred-km level. This result provides a new platform towards the realization of a long-distance quantum network.

Figures (5)

• Figure 1: Simplified experimental layout. The $^{87}\mathrm{Rb}$ atomic-ensemble module employs a counter-propagating write-read geometry whose beams have a waist of $\sim$240 $\mu$m, the write-out and read-out modes adopt the same layout but with a reduced beam diameter of $\sim$70 $\mu$m. The two optical axes are separated by 0.9$^\circ$. After spontaneous Raman scattering (SRS), the entangled write-out photon at 780-nm wavelength is coupled into a QFC module where it is mixed with 1601-nm pump light in a bulk periodically poled lithium niobat (PPLN) crystal within a Sagnac interferometer, and converted into 1522-mm photons for long fiber transmission up to 20 km. The polarization drift in long fibers is automatically compensated with the aid of a liquid crystal variable retarder, two electrically controlled flip mirrors, a home-made polarimeter and a polarization controller, see more details in SM part 5. After long fiber transmission, the telecom single photon is first conducted to a filtering module containing a etalon and a volume Bragg grating (VBG) and then projected on a PBS and detected by SNSPDs. The successful detection of 1522-nm photons triggers a read process on atoms, later on the emitted read-out single photon is projected on a PBS and detected by an APD. Here, the atomic ensemble and the QFC module with SNSPDs are housed in separate rooms, a communicating system including 20-m communicating fibers and three pairs of electro-optic converter and photoelectric converter, and a 15-m fiber for 780-nm photon transmission are used. INSET: A 50-ns left-circular write pulse, 20-MHz blue-detuned, couples the state $\ket{g}$ and $\ket{e}$, generating atom-photon entanglement with a low probability ($\sim$1% as measured on local APDs). PBS: Polarizing Beam Splitter, QWP: Quarter-Wave Plate, HWP: Half-Wave Plate, APD: Avalanche Photodiode, DM: dichroic mirror.
• Figure 2: Read scheme and performance. (a) For read in atomic $\sigma_z$ base, the $\sigma^-$ polarized read pulse, with a duration of approximately 250 ns, couples states $\left | 5S_{1/2}, F=2, m_{F} \right \rangle$ and $\left | 5P_{3/2}, F=2, m_{F}-1 \right \rangle$ to convert the atomic spin wave into a single photon for detection. (b) For read in superposition base, a $\pi /2$ Raman pulse with a duration $\sim$400 ns containing simultaneous $\sigma^+$ and $\sigma^-$ polarization components is applied before the read pulse. The Raman pulse couples the $\left | 5S_{1/2}, F=2 \right \rangle$$\to$$\left | 5P_{1/2}, F=1 \right \rangle$ and $\left | 5P_{1/2}, F=2 \right \rangle$ transitions with an intermediate detuning of 407 MHz to map the states in $\sigma_x$ base to $\sigma_z$ base, namely $\left | \Downarrow \right \rangle_x = \left | \Downarrow \right \rangle_z + \left | \Uparrow \right \rangle_z$$\to$$\left | \Downarrow \right \rangle_z$ and $\left | \Uparrow \right \rangle_x = \left | \Downarrow \right \rangle_z - \left | \Uparrow \right \rangle_z$$\to$$\left | \Uparrow \right \rangle_z$, achieving measurement on the superposition basis of atoms. (c-d) The change of internal retrieval efficiency and correlation with memory time when read-out photon fixed to $\sigma^-$ and write-out projected on $\sigma^-/\sigma^+$.
• Figure 3: The QFC module's external quantum efficiency versus pump power. The data points are fitted with $\eta(P)=\eta_{\max} \sin^2\left(\sqrt{\alpha_{\text{nor}} P} L\right)$Wong-2004-Opt.Lett.. The pump power is set to 1.749 W, with maximized efficiency of 47.2% and 48.5% for the H- and V-polarization arms, respectively.
• Figure 4: (a-b) The time histogram of 1552-nm write-out photons and 780-nm read-out photons. In both figures, red vertical lines demarcate the full width signal window. Each bin of the histogram occupies 1 ns. (c-d) The normalized coincidence counts for changing the angle of a HWP under eigenbasis and changing the read-out delay (offset $\sim$100$\mu$s) under superposition basis. The results are fitted by a sinusoidal function.
• Figure 5: Fidelity and SNR dependence on transmission distance. Red squares denote atom-photon entanglement fidelity at different transmission distances for the telecom measurement, while blue circles represent fidelity for local atom-photon entanglement but with same readout delay (the red and blue lines are plot for visual reference). The green triangles are measured SNR with same exc efficiency (1% on local APDs), the fitted curve is from the SNR model.