Memory-Assisted Nonlocal Interferometer Towards Long-Baseline Telescopes

TL;DR

This work demonstrates a memory-assisted nonlocal interferometer with a long optical baseline by integrating two quantum memories via the DLCZ protocol, quantum frequency conversion, and a simulated thermal stellar field. A triple-interferometer phase-stabilization scheme provides a stable phase reference for interferometric measurements over a fiber link up to 20 km, achieving a geometric-delay compensation of $\approx$1.5 km and enabling an angular-resolution potential of $\lambda/B \approx 8~\mu\mathrm{as}$ in future fully separated implementations. The approach leverages auxiliary entanglement as a tunable phase reference and shows a Fisher-information-based advantage over coherent-light schemes, highlighting practical routes toward high-precision astronomical imaging and extended-baseline quantum networks. With improvements in memory lifetimes and multiplexing, this memory-assisted nonlocal interferometry could scale to hundreds of kilometers, significantly advancing optical VLBI concepts and quantum-enhanced sensing.

Abstract

Quantum networks and remote quantum entanglement serve as vital future quantum communication resources with broad applicability. A key direction lies in extending the baseline of optical interferometers to enhance angular resolution in interferometric imaging. Here, by measuring a simulated thermal light field, we report the demonstration of a memory-assisted nonlocal interferometer achieving a fiber-link baseline up to 20 km while simultaneously showing its capability to compensate for a geometric delay equivalent to 1.5 km. This result demonstrates potential for enhancing the angular resolution of interferometric imaging in the optical band with delocalized single-photon entanglement, and paves the way for future application of quantum memories in astronomical observation.

Figures (5)

• Figure 1: The memory-assisted nonlocal interferometer. Entanglement between the memories (QM) at two nodes can be established via the DLCZ protocol. The two nodes, separated by 2 m, are housed within a laboratory at the University of Science and Technology of China (USTC). The quantum frequency conversion (QFC) and two 10.1-km deployed fibers are implemented to send the write-out photon to Node C at Hefei Software Park (located 6.5 km from Node A and B) for interference, which extends the equivalent baseline length up to 20 km. The subsequently retrieved entangled optical fields can then be separately interfered with the thermal signals locally received at Node A and B respectively. The thermal light generated through Raman scattering is used to mimic the stellar light in the experiment. By analyzing the coincidence counts between the detectors, the complex interference visibility can be determined.
• Figure 2: Details about the thermal field generation. (a) The experiment setup for thermal field generation. (b) The measured second-order correlation function $g^{(2)}(\tau)$ with a 2.5 ns integration window width. (c) Comparison of experimental photon number distribution with theoretical models for ideal thermal and coherent light fields. (d) The measured second-order correlation function $g^{(2)}(0)$ with different integration window width. Error bars indicate one standard deviation of the photon-counting statistics.
• Figure 3: Theory and experimental results. (a) Verification of two-memory entanglement. (b) $V_H$ versus the intensity ratio $x$ in the main text. The solid lines correspond to the theoretical values under different auxiliary field conditions in the ideal case. The dotted (dashed) line represents the calculation results within a 20-ns (60-ns) interference window, accounting for various experimental imperfections. The orange hollow points show the experimental results presented in the subsequent figures. The green inverted triangle is the experiment result from tang2025phase with a visibility of 0.35. (c) Local experiment result with a 20 ns interference window. (d) The experiment result for a baseline of 20 km with a 20-ns interference window. (e) As (d) with a 60-ns interference window. (f) Interference results simulating a 5-$\upmu$s arrival time delay. All fitted visibilities are shown at the corner with the same color. Error bars indicate one standard deviation of the photon-counting statistics.
• Figure S1: Measurement protocol.
• Figure S2: Phase locking scheme details. (a) The local write-read interferometer (named WRI) covers paths of write and read paths. The relative locking phase between two arms can be adjusted by a sandwich configuration of wave plates. It's also the way we introduce a controllable phase shift in auxiliary entanglement verification and GJC scheme demonstration. (b) Interferometer configuration covering the write-out, read-out, and thermal light paths. (c) The two locking-beam interference loops used to stabilize the optical paths in (b). The first loop (named THI, schematically indicated within blue box) is similar in configuration to a Michelson interferometer. The locking beam is reflected by the mirror at the bottom, passing twice through the transmission path of the thermal field and a short extra optical path. This arrangement is primarily used to stabilize the relative phase of the thermal field transmission path. A portion of the reflected beam from the mirror is reflected via a PBS (with relative intensity adjusted by the QWP) to cover both the read-out and write-out paths, and subsequently interferes at the detection node for feedback, thereby stabilizing the phase of the overall read-out-write-out path (named WOI). Similar to Ref. luo2025, we also implement dual-wavelength locking for write-out path, which includes the field-deployed fiber.