Deep-Learning-Designed AlGaAs Interface Linking Trapped Ions to Telecom Quantum Networks

Abstract

The realization of a scalable quantum internet requires efficient light-matter interfaces that map stationary qubits onto photonic carriers for long-distance transmission. A central challenge is the generation of entangled photons simultaneously compatible with single-emitter transitions and low-loss telecom fiber infrastructure. Spontaneous parametric down-conversion in integrated photonic platforms offers a promising route toward this goal. Among available material systems, AlGaAs is particularly attractive due to its large second-order nonlinearity and strong potential for monolithic integration. However, engineering the spectral and spatial properties of the generated quantum states requires the simultaneous optimization of numerous geometric and material parameters, a task remaining computationally demanding for conventional numerical approaches. To address this challenge and enable rapid and high-fidelity modeling of complex nonlinear photonic devices, we develop an inverse-design framework based on neural network surrogate models. Using this readily extendable method, we design a transversely pumped AlGaAs waveguide microcavity that produces polarization-entangled photon pairs in distinct spatial modes and frequency channels, one at 1092 nm, resonant with a $^{88}\text{Sr}^{+}$ transition, and the other at 1550 nm in the telecom C-band. This device establishes a direct photonic interface between trapped-ion qubits and long-haul fiber networks, providing a scalable pathway toward hybrid quantum network architectures.

Figures (3)

• Figure 1: Experimental setup of the quantum interface between non-degenerate SPDC source and trapped strontium ions. (a) Simplified level scheme of the $^{88}Sr^+$ ion. A 422 nm $\sigma^-$ - polarized laser pulse coherently drives the transition from $S_{1/2}$ to $P_{1/2}$. Ion photon entanglement is generated through the 1092 nm spontaneous decay. (b) Ion-trap apparatus. A single ion confined above a surface trap spontaneously decays. The $\pi$-polarized emission is not collected, while the $\sigma$-polarized photons are coupled into optical fibers and routed to the Bell-state measurement setup. (c) Schematic view of the non-degenerate SPDC source. A pump beam impinges on the waveguide surface, generating polarization-entangled photon pairs with one photon at a telecom wavelength and the other matched to 1092 nm, corresponding to the considered ion transition.(d) Bell-state measurement setup. Entanglement swapping is implemented using a 50:50 fiber beam splitter (BS) followed by two fibered polarizing beam splitters (PBS).
• Figure 2: Ion-photon interface design: (a) Workflow for source optimization (see text). (b) Evolution of the figure of merit (FOM) throughout the optimization process. The yellow star indicates the optimal configuration corresponding to the highest FOM. (c) Relative discrepancy between our neural-network-based surrogate model and a conventional transfer-matrix-method. (d)-(e): Comparison between the initial (d) and optimized (e) designs: the black curve shows the refractive-index profile of the multilayer stack. The blue and purple curves correspond to the spatial distributions of the fundamental TE and TM guided modes of the generated photons, respectively, while the red curve represents the transverse profile of the pump field.
• Figure 3: Counterpropagating ion-photon interface characteristics. (a) Schematic of the device and the pump-incidence geometry for the generation of non-degenerate polarization-entangled photons. (b) Tunability curves showing the dependence of the photon pair wavelengths on the pumping angle. (c) Joint Spectral Intensity (JSI) of the biphoton state emitted by the device when pumped simultaneously at $\theta_1 \approx 33.66^\circ$ and $\theta_2 \approx 33.95^\circ$. (d) Marginal Joint Spectral Intensity for the signal photon, highlighting the peak at the ion transition wavelength.