Heterogeneous entanglement between a trapped ion and a solid-state quantum memory

Abstract

Hybrid quantum networks offer a promising architecture for scalable quantum information processing and a future quantum internet, as they can combine the complementary strengths of disparate physical platforms. While single-atom systems provide deterministic quantum logic gates, atomic ensembles enable large-capacity quantum storage. However, generating entanglement between such heterogeneous systems has remained an open challenge, primarily due to fundamental spectral mismatches and system complexity. Here, we demonstrate a hybrid quantum network that entangles a single trapped $\mathrm{^{171}Yb^{+}}$ ion and a quantum memory based on $\rm ^{153}Eu^{3+}\colon\!Y_2SiO_5$ crystal over a 75-m separation. Using polarization-maintaining quantum frequency conversion, we map spin-photon entanglement onto a hybrid entanglement between a single spin qubit and a collective excitation of the quantum memory. The resulting entangled state achieves a fidelity of $(89.21 \pm 2.23)\%$ and violates the CHSH-Bell inequality by 6 standard deviations ($S = 2.328 \pm 0.055$), confirming nonlocality between two heterogeneous nodes. This work establishes entanglement between a quantum processing module with a multiplexed quantum memory node, representing a key step toward a scalable, multifunctional quantum internet.

Figures (6)

• Figure 1: A prototype hybrid quantum network.a, Schematic of the envisioned hybrid architecture, where heterogeneous quantum nodes are interconnected via flying photonic qubits. This work focuses on establishing entanglement between a quantum processor node based on trapped ion (TI) and a quantum memory (QM) node based on rare-earth ion ensemble, via a quantum frequency conversion (QFC) module. In this experiment, the two nodes are housed in separate laboratories spaced 75 m apart and linked by a 90-m single-mode fiber (SMF). b, The trapped-ion (TI) processor node. A single $\mathrm{^{171}Yb^{+}}$ ion is excited by a pulsed laser. Photons emitted along the quantization axis (defined by an external magnetic field) are collected by a high-numerical-aperture objective (NA = 0.64) and coupled into an SMF. c, The QFC module. Photons at 369 nm are converted to 580 nm via difference-frequency generation in a periodically poled stoichiometric lithium tantalate (PPSLT) waveguide, pumped at 1018 nm (see panel h). A Sagnac interferometer configuration, comprising a PBS, OAPMs, and an intra-loop HWP, enables robust conversion independent of the input polarization. d, The QM node is based on a laser-written waveguide fabricated in an $\rm ^{153}Eu^{3+}\colon\!Y_2SiO_5$ crystal. An atomic-frequency comb (AFC) is prepared via optical pumping, and on-chip electrodes enable electric-field control for on-demand readout of the $n^\text{th}$ AFC echo (Inset). e--h, Sequence detailing the ion-photon entanglement generation and the QFC process. i, Relevant energy-level diagram of the QM. The quantum storage utilizes the polarization-independent ${}^{7}F_0\leftrightarrow{}^{5}D_0$ optical transition of site-2 $\mathrm{^{153}Eu^{3+}}$ ions in the $\mathrm{Y_2SiO_5}$ crystal. Abbreviations: MW, microwave; SNSPD, superconducting nanowire single-photon detector; BS, beam splitter; HWP, half-wave plate; PBS, polarizing beam splitter; OAPM, off-axis parabolic mirror.
• Figure 1: Fabrication and characterization of the integrated quantum memory device.a, Front and c, top view of the laser-written depressed-cladding waveguide. The waveguide (visible between the two black electrodes) is aligned along the crystal's $D_2$ axis. Dark spots are due to surface contamination. Scale bars: 200 µ m (a) and 20 µ m (c). b, Beam profile at the waveguide output for the $|H\rangle$ and $|V\rangle$ polarization states; scale bar: 20µm. d, Stark splitting of the doped $\mathrm{^{153}Eu^{3+}}$ ions as a function of the applied electric field. The average Stark splitting is $5.80 \pm 0.09$ kHz/(V/cm).
• Figure 2: Reconstructed density matrices of the ion–photon entangled state. Matrices are shown before (a) and after the QFC (b), as obtained via quantum state tomography. The corresponding fidelities are $(95.46 \pm 0.02)\%$ (a) and $(89.12 \pm 1.83)\%$ (b) to the target Bell state. The dashed lines denote the expected results for ideal Bell states.
• Figure 2: The reconstructed process matrix of the QFC module and the 90-m SMF. The process matrix fidelity is $(97.32\pm0.14)$% to the ideal identity matrix, with a measurement time of 3 minutes.
• Figure 3: Timed sequence for entangling the TI node and the QM node.a, Overall time sequence. Each experimental cycle comprises an 8-s QM preparation phase followed by a 250-s entanglement storage phase. During QM preparation, the trapped ion is maintained under continuous laser cooling. The storage phase is divided into multiple frames, each containing 100µs of ion-laser cooling followed by 100 sequential storage attempts. b, QM preparation. This process involves three stages: initialization, antihole burning, and atomic‑frequency‑comb (AFC) preparation. c, Single entanglement‑generation attempt. The sequence begins with ion‑qubit initialization (2.5µs, Fig. \ref{['fig:network_arch']}e). A 369-nm laser pulse then excites the ion to generate ion-photon entanglement, producing a flying qubit at 369 nm (Fig. \ref{['fig:network_arch']}f). The emitted photon is converted to 580 nm by the QFC module and transmitted through the optical fiber connecting the two labs, introducing a delay of $\tau_\mathrm{fly}=484~\mathrm{ns}$. Finally, the 580-nm photon is stored and on-demand retrieved from the QM node. The photon-detection signal in the QM node is sent back to the TI node to herald whether a ion‑state readout should be performed.