Scalable Quantum Photonic Platform Based on Site-Controlled Quantum Dots Coupled to Circular Bragg Grating Resonators

TL;DR

This work tackles the challenge of scalable, high-quality single-photon sources by combining a buried-stressor site-controlled quantum dot platform with circular Bragg gratings in a marker-free fabrication workflow. The authors demonstrate a 6×6 SCQD-CBG array with 100% device yield, achieving a best-case photon extraction efficiency of $$(47.1\pm3.8)\%$$, a linewidth of $$(1.41\pm0.22)\ { m GHz}$$, a radiative lifetime of $$(0.80\pm0.02)\ { m ns}$$, near-perfect single-photon purity of $$(99.58\pm0.18)\%$$, and Hong-Ou-Mandel visibility of $$(81\pm5)\%$$ under quasi-resonant excitation. The study shows offset-dependent improvements in PEE, linewidth, and indistinguishability, consistent with a charge-noise model where emitter position relative to the etched surface modulates dephasing. By eliminating the need for emitter localization and complex deterministic lithography, this platform enables scalable, wafer-scale integration of high-performance SPS, with strong potential for on-chip quantum networks and photonic processors. The results establish quantitative benchmarks for emitter–cavity alignment tolerances and provide a unified framework linking growth, nanofabrication, and quantum-optical properties in scalable quantum photonic devices.

Abstract

The scalable integration of solid-state quantum emitters into photonic nanostructures remains a central challenge for quantum photonic technologies. Here, we demonstrate a robust and streamlined integration strategy that tackles the long-standing issue of deterministic fabrication on randomly positioned self-assembled quantum dots (QDs), leveraging a buried-stressor-based site-controlled InGaAs QD platform. We show that this deterministic growth approach enables precise spatial alignment with circular Bragg grating (CBG) resonators for enhanced emission, eliminating the need for complex and time-consuming deterministic lithography techniques. We fabricated a $6\times6$ SCQD-CBG array with 100\% device yield, with 35 devices falling within the radial-offset range where the simulated photon-extraction efficiency (PEE) exceeds 20\%, underscoring the spatial precision and scalability of our fabrication concept. A systematically selected subset of five devices with varying radial displacements reveals clear offset-dependent trends in extraction efficiency, spectral linewidth, and photon indistinguishability, thereby establishing quantitative bounds on spatial alignment tolerances. In the best-aligned QD-CBG device, we achieve a PEE of $(47.1\pm3.8)\%$, a linewidth of $(1.41\pm0.22)$ GHz, a radiative decay lifetime of $(0.80\pm0.02)$~ns, a single-photon purity of $(99.58\pm0.18)\%$, and a Hong-Ou-Mandel two-photon interference visibility of $(81\pm5)\%$ under quasi-resonant excitation at saturation power. We confirm our conceptual understanding of the effect of emitter-position dependent charge-noise fluctuations in terms of a quantum-optical model for the (quantum-)emission properties.

Figures (4)

• Figure 1: Schematic process flow for SCQD-CBG device fabrication. The fabrication pathway begins with MOCVD growth of a DBR template incorporating a buried AlAs stressor layer (a), followed by SCG-mesa definition through UV-lithography and etching (b) and lateral oxidation to form apertures at the center of the mesas (c). SCQDs are subsequently overgrown within a GaAs cavity by MOCVD (d). Prior to patterning, the CBG parameters were optimized through FEM simulations, as illustrated by the schematic device representation and the corresponding simulated electric field distribution (e). The CBGs with numerically optimized geometry are subsequently patterned on the SCG-mesas using EBL and etching (f), also shown in the representative SEM image of a completed SCQD-CBG device. This process yields an array of SCQD-CBG structures (g), ready for optical characterization.
• Figure 2: (a) Low-temperature ($20$ K) CL map (overlaid with SEM image) of a $6\times6$ SCQD-CBG device array (spectral range: $(934\pm6)$ nm), showing emission confined to the CBG-mesa centers with a 100% integration yield. (b) Statistical analysis of alignment accuracy from high-resolution CL scans (50 nm pixel size) of all 36 devices. QD positions were localized by 2D Gaussian fitting and compared with CBG-mesa centers extracted from SEM data. The resulting radial offset distribution, with mean displacements of $\mu_x = (18.3\pm202.4)$ nm and $\mu_y = (-11.0\pm156.3)$ nm, confirms the high reproducibility of the SCQD-CBG alignment across the array and validates the robustness of our marker-free fabrication strategy.
• Figure 3: (a-d) Optical and quantum optical characterization of the best-performing SCQD-CBG5 device (radial offset: 54 nm), measured at 4 K. (a) $\mu$PL spectra under pulsed excitation at 870 nm for varying pump powers (vertically offset for clarity), revealing a max PEE of $(47.1\pm3.8)\%$. All subsequent measurements were performed under p-shell excitation at 911 nm and at the QD saturation power. (b) Time-resolved $\mu$PL indicating a QD radiative decay lifetime of $(0.80\pm0.02)$ ns. (c) Second-order autocorrelation measurements demonstrating $g^{(2)}(0)=(0.0042\pm0.0018)$, corresponding to a single-photon purity of $(99.58\pm0.18)$%. (d) HOM TPI measurements exhibiting a raw visibility of $(81\pm5)$%, evidencing very high photon indistinguishability.
• Figure 4: Assessment of optical and quantum optical device performance as a function of radial offset. (a) QD radial-offset dependence of PEE, with FEM simulations (black dashed line) in agreement with experimental results (red circles). (b) Dependence of raw HOM-TPI visibility (%, left axis) and emission linewidth (GHz, right axis) on the QD radial offset. Experimental data are represented by black (visibility) and red (linewidth) circles, while the corresponding theoretical modeling is depicted by black and red dashed lines.