🔬 optics

A high-Q Split Cavity Enabling Independent Electrical Tuning of two Quantum Dots

This paper presents the design, fabrication, and characterization of a high-quality factor ( $Q \geq 20,000$ ) GaAs photonic crystal split cavity that enables independent electrical tuning of two InAs quantum dots into resonance with a single cavity mode, thereby providing a viable experimental pathway for generating multi-dimensional photon graph states via strong coupling.

Original authors: Lan Hoang Mai, Nazifa Tasnim Arony, R S Joshya, Tyler Manfre, Joshua M. O. Zide, Matthew F. Doty

Published 2026-08-04

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Lan Hoang Mai, Nazifa Tasnim Arony, R S Joshya, Tyler Manfre, Joshua M. O. Zide, Matthew F. Doty

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a world where light isn't just something that helps us see, but a building block for a super-powerful kind of computer. This is the realm of quantum photonics, a field where scientists try to trap tiny particles of light (photons) and matter (like electrons) together to do complex math. To make this work, you need a "dance floor" for these particles: a tiny, perfect mirror box called a cavity. Inside this box, a single atom or a "quantum dot" (a speck of semiconductor material that acts like an artificial atom) can bounce light back and forth so fast that they become best friends, sharing energy in a dance called "strong coupling."

But here's the tricky part: nature is messy. When scientists grow these quantum dots, they don't all come out exactly the same size or shape. It's like baking a batch of cookies where some are slightly bigger or smaller than the others. Because of these tiny differences, each cookie (quantum dot) wants to dance to a slightly different song (a different color of light). If you put two different cookies in the same box, they usually can't hear each other because they are tuned to different frequencies. To fix this, scientists need a way to "tune" each cookie individually, like adjusting the radio dial on two different cars so they can both listen to the same station, without the tuning of one car messing up the other.

This paper introduces a clever new design for that radio dial. The researchers built a special "split cavity"—a light-trapping box that has been sliced right down the middle. This cut creates two separate electrical rooms within the same box. Now, they can apply a unique voltage to the top half and a different voltage to the bottom half. This allows them to independently tune two different quantum dots, forcing them to sing in perfect harmony with the same light mode. The big question was: if you slice a perfect mirror box in half, does it fall apart and lose its magic? The authors set out to find out if this split design could still hold the light tightly enough to make the quantum dance happen.

The Split Cavity: A Box with Two Doors

The researchers designed a photonic crystal, which is essentially a slab of material (Gallium Arsenide) punched with a precise pattern of tiny holes. Think of it like a Swiss cheese where the holes are arranged in a perfect grid to trap light. In their design, they cut a narrow 90-nanometer-wide trench right through the middle of this cheese. This cut separates the structure into two electrically distinct halves, but the light can still "leak" across the gap, keeping the two halves connected as a single system.

They placed two quantum dots in this setup, one on the top half and one on the bottom. By applying different electric fields to each side, they could stretch or squeeze the energy levels of the dots, tuning them to match the light inside the cavity. The goal was to see if this "split" design could still act as a high-quality trap for light, or if the cut would scatter the light away and ruin the experiment.

The Results: High Quality, But Not Perfect

The team built these split cavities and tested them. They found that despite the cut, the devices worked surprisingly well. They measured a "quality factor" (Q), which is a score telling us how long a photon can bounce around inside the box before escaping. A higher score means a better box. The best device they made had a Q of 62,704, and on average, the devices scored around 21,886.

Why does this number matter? To get the quantum dots and the light to "strongly couple" (that best-friend dance mentioned earlier), you need a Q of at least about 12,728. The average device in this study hit that mark, and six out of the thirteen devices they tested even scored higher than this threshold. This suggests that the split cavity is a viable tool for making quantum computers.

The Mystery of the Imperfections

The researchers then asked: "Why aren't all our scores perfect? Why do some devices score 62,000 and others only 6,000?" They suspected it might be because the walls of the holes in the cavity were rough, scattering light like a bumpy road scatters a car. To test this, they ran computer simulations. They created a virtual version of their cavity and intentionally added the kind of tiny errors that happen when you build things with electron-beam lithography (the tool used to draw the tiny patterns).

The simulation showed that the variations in size and position of the holes—caused by the limits of their drawing tool—were enough to explain the drop in scores. They found that the Q was limited primarily by the precision of the tool used to make the device, not by rough walls or bad materials. In fact, they showed that if they had optimized the design for a different type of light wave (called the "odd mode"), the average score could have been even higher, around 32,261, because that specific wave pattern is less sensitive to the tiny errors in the hole placement.

What This Means

The paper concludes that this split cavity design is a promising path forward. It proves that you can slice a high-quality optical cavity in half to independently control two quantum dots without destroying the box's ability to trap light. While the current devices are just good enough to enter the "strong coupling" zone, the research suggests that with slightly better manufacturing precision or by tuning the design for the "odd mode," scientists could build even better devices. This opens the door to creating complex networks of quantum dots that can talk to each other, a crucial step toward building powerful quantum computers that process information using light.

Technical Summary: A High-Q Split Cavity Enabling Independent Electrical Tuning of Two Quantum Dots

Problem Statement
Cavity-mediated strong coupling is a fundamental requirement for quantum photonics, particularly for generating entanglement between spin qubits and creating photonic graph states. However, the practical implementation of coupling two distinct quantum dots (QDs) to a single cavity mode is hindered by the inherent spectral inhomogeneity of QD ensembles, which arises from the random diffusion in Stranski–Krastanov growth. While electrical tuning via the Stark effect is the preferred method for correcting these spectral mismatches due to its speed and integration ease, existing methods struggle to apply unique electric fields to individual QDs within the same device without compromising the cavity's optical quality. Previous approaches, such as proton implantation or shallow etching, often introduce significant scattering losses or lack the necessary isolation. There is a need for a device architecture that allows for independent electrical tuning of two QDs while maintaining a high quality factor ( $Q$ ) sufficient to enter the strong coupling regime.

Methodology
The authors present the design, fabrication, and characterization of a GaAs-based photonic crystal split cavity. The device consists of two evanescently coupled L3 cavities separated by a monolithic cut, creating two electrically distinct diode structures (top and bottom).

Design and Simulation: The authors utilized Finite-Difference Time-Domain (FDTD) simulations to optimize the cavity geometry. They adjusted the horizontal position ( $\Delta x$ ) and radius ( $\Delta r$ ) of the end holes adjacent to the L3 cavities to maximize the quality factor while maintaining structural symmetry. Both symmetric (even) and anti-symmetric (odd) modes were analyzed, though the even mode was prioritized for initial fabrication due to the sensitivity of the characterization technique used.
Fabrication: The device was fabricated on a $p-i-p-i-n$ diode structure embedded in a 150-nm thick GaAs membrane. The process involved electron-beam lithography, chlorine-based inductively coupled plasma (Cl-ICP) etching, and wet chemical release using hydrofluoric acid. A specific 90-nm wide cut was designed to electrically isolate the two halves while remaining above the minimum feature size of the fabrication process.
Characterization: The quality factor was measured using a cross-polarized resonant scattering technique. This method allows for the measurement of intrinsic cavity $Q$ without the need for coupled waveguides, which would introduce additional loss. The setup utilized a tunable laser and a high-numerical-aperture objective to detect $p$ -polarized light scattered by the cavity resonance.
Statistical Analysis: To understand the sources of $Q$ variation, the authors performed Monte Carlo FDTD simulations incorporating experimentally measured fabrication disorders (hole radius variation, placement errors, and cut width deviations) derived from Scanning Electron Microscope (SEM) images.

Key Contributions and Results

Device Performance: The fabricated split cavity achieved an experimental quality factor as high as $6.2 \times 10^4$ . Across 13 characterized device instances, the mean $Q$ was $21,886 \pm 600$ , with a median of $12,417$.
Strong Coupling Threshold: The authors calculated that a minimum $Q$ of approximately $12,728$ is required to reach the strong coupling regime for this specific split cavity geometry (accounting for its doubled mode volume compared to a standard L3 cavity). The median experimental $Q$ sits right at this threshold, and six of the thirteen measured devices exceeded it.
Limiting Factors: By comparing experimental data with Monte Carlo simulations, the authors determined that the $Q$ factor is limited primarily by the precision and accuracy of the e-beam lithography tool and wet etch bias (positional and feature width disorders), rather than by sidewall scattering. Simulations indicated that the hole sidewalls are sufficiently vertical and smooth that scattering losses are negligible compared to lithographic errors.
Mode Analysis: While the initial devices were optimized for the even mode (due to characterization constraints), simulations predict that devices optimized for the odd mode would achieve a median $Q$ of $32,261$ under realistic fabrication disorders. This is because the odd mode's electric field node at the cut minimizes overlap with the disordered lattice near the monolithic cut.

Significance
The paper demonstrates a viable experimental path toward devices capable of mediating controlled interactions between two InAs quantum dots via strong coupling to a single cavity mode. By successfully electrically isolating two distinct regions of a high- $Q$ cavity, the device enables the independent Stark tuning of two QDs into resonance. This capability is a critical step toward generating multi-dimensional photon graph states and facilitating fully on-chip quantum information processing. The work establishes that high-quality factors compatible with strong coupling are achievable in split-cavity architectures, provided that fabrication precision is maintained, and suggests that future optimization targeting the odd mode could further enhance performance margins.

Technical Summary: A High-Q Split Cavity Enabling Independent Electrical Tuning of Two Quantum Dots

More like this