⚛️ quantum physics

Photonic CZ gates based on few-photon scattering from emitter-cavity systems

This paper analyzes photonic CZ gates based on N waveguide-coupled cavities with two-level emitters, revealing that a fundamental trade-off between preserving single-photon wave packets and acquiring the necessary two-photon phase shift limits the gate's average fidelity to approximately 60%.

Original authors: Mateusz Duda, Pieter Kok

Published 2026-09-23

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Mateusz Duda, Pieter Kok

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

In the quest to build a quantum computer, scientists are constantly searching for the most reliable way to process information. One of the most promising candidates for carrying this information is the photon, a single particle of light. Photons are excellent messengers because they travel at incredible speeds and can hold their state for a long time without getting scrambled by their environment. This makes them ideal for connecting different parts of a quantum network, acting as "flying qubits" that shuttle data between stationary processors. However, while it is relatively easy to manipulate a single photon, making two photons interact with each other is a profound challenge. Unlike atoms or electrons, photons do not naturally bump into one another; they simply pass through like ghosts. To perform the complex logic required for a quantum computer, researchers need a way to force these light particles to influence each other, specifically to create a gate that flips the state of one photon based on the presence of another.

To solve this, physicists often turn to a strategy that mimics a strong interaction by using a mediator. Imagine trying to get two people to talk to each other by having them both shout into the same echo chamber; the room itself transmits the influence. In the quantum world, this "echo chamber" is a tiny cavity, a microscopic box that traps light, containing a single atom-like object called an emitter. When a photon enters this box, it interacts with the emitter, which then affects any subsequent photon. The goal is to engineer this system so that if two photons arrive together, they leave with a specific change in their wave-like rhythm, a shift known as a phase shift, while a single photon passes through unchanged. This specific interaction is the heart of a logic gate called a controlled-Z gate, a fundamental building block for quantum computing.

A team of researchers at the University of Sheffield recently investigated how well this concept works when scaled up. They focused on a specific design where photons travel through a series of these tiny cavities, each containing an emitter, rather than just a single box. Their work involved simulating the behavior of light as it scatters off these systems to see if they could create the necessary logic gate. The researchers were particularly interested in a trade-off that arises when trying to tune the system. They found that to keep a single photon's shape intact as it passes through, the connection between the emitter and the cavity needs to be very strong. However, to get the two photons to interact strongly enough to produce the required phase shift, that same connection needs to be much weaker.

This conflict creates a significant bottleneck. The team discovered that while adding more cavities to the chain helps reduce a type of quantum noise called spectral entanglement—which would otherwise distort the photons—the benefit is canceled out by the distortion of the single photons themselves. Because the system cannot simultaneously preserve the shape of a single photon and generate the correct interaction for two photons, the gate does not work perfectly. Through their calculations, the researchers determined that the best possible performance for this specific, passive setup is an average fidelity of approximately 60 percent. This means that for every time the gate is used, there is a substantial chance it will fail to perform the correct logical operation.

The study does not suggest that this approach is a dead end, but rather that it sets a clear baseline for what can be achieved with a simple, static arrangement of cavities and emitters. The authors note that their results apply to a system where the components are fixed and not actively controlled during the process. They suggest that future work might need to explore more complex setups, perhaps using emitters with more energy levels or systems where the coupling strength can be changed dynamically, to overcome this 60 percent limit. For now, their findings provide a crucial reality check: while scattering photons off emitter-cavity systems is a viable path toward quantum gates, the inherent struggle between preserving the photon's shape and forcing it to interact imposes a strict ceiling on performance unless more sophisticated engineering is introduced.

Technical Summary: Photonic CZ Gates Based on Few-Photon Scattering from Emitter–Cavity Systems

Problem Statement
The realization of deterministic two-photon gates is a central challenge in photonic quantum computing due to the absence of direct photon–photon interactions. While linear optics can achieve universal quantum computation, it relies on probabilistic post-selection, which is inefficient for scalable architectures. Deterministic gates require nonlinear media to mediate effective interactions, such as cross-phase modulation via Kerr nonlinearities or scattering from quantum emitters. However, Kerr nonlinearities are typically weak at the single-photon level and suffer from phase noise that limits gate fidelity. Alternatively, scattering from emitter–cavity systems offers a route to strong conditional phase shifts, but this approach introduces spectral entanglement. This entanglement distorts the photon wave packets, preventing the ideal realization of a Controlled-Z (CZ) gate where the single-photon state remains unchanged while the two-photon state acquires a $\pi$ phase shift. Previous attempts to mitigate this distortion using arrays of emitters or complex active control mechanisms often introduce experimental difficulties, such as requiring counter-propagating photons or time-dependent coupling rates.

Methodology
The authors analyze the performance of a photonic CZ gate architecture based on the Knill-Laflamme-Milburn (KLM) scheme, which utilizes two dual-rail encoded qubits, two 50:50 beam splitters, and two nonlinear sign (NS) gates. The core of their implementation involves the NS gates, realized by $N$ waveguide-coupled cavities, each containing a two-level emitter. The system is modeled as a passive setup with no active control over emitter or cavity parameters, and the photons are co-propagating.

To quantify performance, the authors derive the single- and two-photon scattering matrices for the system:

Single Cavity ( $N=1$ ): They utilize the input–output formalism of quantum optics to derive the single-photon transmission coefficient $t(\omega)$ and the nonlinear two-photon term $C_{p_1 p_2 k_1 k_2}$ .
Multiple Cavities ( $N>1$ ): They extend the analysis to an array of $N$ cavities using the SLH (Scattering-Lindblad-Hamiltonian) formalism. This allows them to construct the total scattering matrix for the cascaded system, accounting for the phase shifts acquired during propagation between cavities.

The gate fidelity is calculated by evaluating the overlap between the ideal output state and the actual output state derived from the scattering matrices. The authors consider Gaussian input wave packets centered on a two-photon transition frequency ( $\omega_0 = \Omega + g/\sqrt{2}$ ), which is detuned from the single-photon transitions ( $\Omega \pm g$ ). They compute both the fidelity for specific input states and the average fidelity over the Haar measure.

Key Contributions and Results
The primary contribution of the paper is the rigorous derivation of the scattering matrices for $N$ coupled emitter-cavity systems and the subsequent analysis of the fidelity trade-offs inherent in this passive architecture.

The Trade-off: The analysis reveals a fundamental trade-off between optimizing the single-photon and two-photon components of the scattering process.
- To preserve the single-photon wave packet (minimizing distortion), the emitter–cavity coupling rate ( $g$ ) must be large, placing the photons off-resonant with the single-photon transitions.
- To achieve the necessary $\pi$ phase shift for the two-photon state, the coupling rate must be smaller, comparable to the cavity decay rate ( $\Gamma$ ) multiplied by the number of cavities ( $N\Gamma$ ).
Fidelity Limit: Due to this trade-off, the average CZ gate fidelity is limited to approximately 60% when averaging over all input states for Gaussian wave packets.
Effect of $N$ : While increasing the number of cavities ( $N$ ) reduces spectral entanglement (a known benefit of translational invariance in such arrays), the fidelity does not significantly improve. This is because the distortion of the single-photon wave packet, which persists even with large $N$ , prevents the fidelity from increasing substantially.
Spectral Profiles: The authors present spectral profiles of the output wave packets, demonstrating how the single-photon component is distorted when parameters are tuned to maximize the two-photon phase shift, and vice versa.

Significance
The paper establishes a baseline figure of approximately 60% fidelity for photonic CZ gates implemented using simple, passive emitter–cavity systems without active control. The authors explicitly state that this result highlights the limitations of such simple setups. They suggest that future work to improve upon this fidelity would require more complex setups, such as utilizing multi-level emitters or implementing time-dependent coupling rates. The study serves as a theoretical benchmark for what can be achieved with current passive nanophotonic architectures and underscores the difficulty of simultaneously preserving wave packet shape and achieving strong conditional phase shifts in localized or arrayed scattering systems.

More like this