⚛️ quantum physics

Millimeter Wave Readout of a Superconducting Qubit

This paper demonstrates high-fidelity (>99%) readout of a superconducting transmon qubit using a highly detuned millimeter-wave cavity at 34.7 GHz, which suppresses unwanted state transitions and enables strong-drive readout without requiring a quantum-limited amplifier.

Original authors: Akash V. Dixit, Zachary L. Parrott, Dennis Chunikhin, Bradley Hauer, Trevyn F. Q. Larson, John D. Teufel

Published 2026-03-17

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Akash V. Dixit, Zachary L. Parrott, Dennis Chunikhin, Bradley Hauer, Trevyn F. Q. Larson, John D. Teufel

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

The Big Picture: A New Way to "Listen" to a Quantum Computer

Imagine you are trying to listen to a very shy, delicate bird (the qubit, which is the basic unit of a quantum computer) singing a specific note. To hear it, you need to shine a light or send a sound wave toward it.

In the past, scientists used "sound waves" (microwaves) that were very similar in pitch to the bird's song. The problem? When you shout too loudly at the bird to make sure you hear it, you accidentally scare it, and it changes its song or flies away. This is called a state transition error. It ruins the measurement.

The Solution: This paper introduces a clever trick. Instead of using a sound wave similar to the bird's song, they use a sound wave that is extremely high-pitched (millimeter waves), completely different from the bird's voice.

Because the "shout" is so different from the bird's natural song, the bird doesn't get confused or scared, even if you shout very loudly. This allows the scientists to listen to the bird very clearly without messing up its state.

The Characters and the Stage

To understand how this works, let's break down the cast of characters:

The Qubit (The Bird): A superconducting circuit that acts like a quantum bit. It usually "sings" at a low frequency (around 3 GHz).
The Readout Resonator (The Megaphone): A 3D metal box (cavity) that holds the "shout" (the photons used to measure the qubit).
The Millimeter Wave (The High-Pitched Shout): Instead of using a standard shout (6–10 GHz), they use a super high-pitched shout (34.7 GHz).

The Problem: The "Scary Shout"

In traditional quantum computers, the "megaphone" and the "bird" sing at similar pitches.

The Analogy: Imagine trying to talk to a friend in a noisy room. If you shout to be heard, you might accidentally start a conversation with a stranger standing next to them, or you might accidentally knock over a vase.
In Physics: When you send a strong signal to read the qubit, if the signal is too close in frequency to the qubit's own energy levels, it can accidentally kick the qubit into a higher energy state (like kicking the bird out of its nest). This ruins the data.

The Innovation: The "High-Pitched" Strategy

The researchers realized that if they make the "megaphone" sing at a frequency 10 times higher than the bird, the bird simply doesn't care about the shout, even if it's very loud.

The Analogy: Imagine the bird only reacts to sounds in the range of a human voice. If you scream at it using a frequency that sounds like a dog whistle (which the bird can't hear), the bird stays calm. You can scream as loud as you want, and the bird won't jump.
The Result: They were able to use 1,000 times more power (more photons) to read the qubit without accidentally changing its state.

The Experiment: What Did They Do?

Building the Box: They built a tiny, 3D aluminum box (about the size of a large coin) to act as the megaphone. Inside, they placed a superconducting chip with the qubit.
The Test: They prepared the qubit in a specific state (like "Sleeping" or "Awake"). Then, they blasted it with their high-pitched millimeter waves.
The Observation:
- They cranked up the volume (the number of photons) until it was huge.
- Result: The qubit stayed exactly where they put it. No accidental jumps.
- The Payoff: Because they could use such a loud signal, they could tell the difference between "Sleeping" and "Awake" with 99% accuracy (fidelity), even without using fancy, expensive amplifiers to boost the signal.

Why Does This Matter?

Think of this as upgrading the microphone for the next generation of quantum computers.

Scalability: As quantum computers grow to have thousands of qubits, you need to read them all very fast and very accurately. If you use the old method, the "shouts" might accidentally mess up neighboring qubits. This new method keeps the qubits safe.
Hybrid Systems: Millimeter waves are a "universal translator." They can connect different types of quantum technologies (like atoms, mechanical parts, and superconducting circuits) that usually can't talk to each other. This experiment proves we can use these high-frequency waves to control superconducting qubits, opening the door for a "hybrid" quantum internet.

The Bottom Line

The scientists found a way to turn up the volume on reading quantum bits without breaking them. By using a frequency that is "out of tune" with the qubit's natural song, they created a safe, high-power environment for reading quantum information. It's like being able to shout at a shy animal to get its attention without ever making it run away.

1. Problem Statement

In conventional circuit quantum electrodynamics (cQED), superconducting transmon qubits (typically operating at 3–6 GHz) are read out using resonators in the 6–10 GHz range. A significant limitation of this architecture is drive-induced resonant state transitions. When the readout drive frequency is close to the qubit frequency, strong drive tones required for high-fidelity, fast readout can inadvertently couple the qubit's computational states ( $|0\rangle, |1\rangle$ ) to higher energy levels ( $|2\rangle, |3\rangle, \dots$ ).

These transitions are problematic because:

They compromise the Quantum Non-Demolition (QND) nature of the measurement.
They introduce readout errors and limit the maximum drive power that can be applied, thereby capping measurement fidelity and speed.
They are difficult to engineer around because the charge matrix elements connecting low-lying states to higher states within the Josephson potential well are relatively large.

2. Methodology

The authors propose and experimentally demonstrate a solution by shifting the readout frequency from the microwave regime into the millimeter wave (mmWave) regime while keeping the transmon qubit in its conventional frequency range.

System Architecture:
- Qubit: A transmon qubit operating at $\omega_q/2\pi \approx 3.083$ GHz with an anharmonicity of $\alpha_q/2\pi \approx -141$ MHz.
- Readout Resonator: A 3D aluminum cavity operating in its fundamental mode at $\omega_r/2\pi \approx 34.670$ GHz.
- Coupling: The qubit and cavity are coupled via a capacitive interaction with a rate $g/2\pi \approx 1.3$ GHz. This results in a large frequency detuning ratio of $\omega_r/\omega_q > 10$ .
Theoretical Approach:
- The team utilized Floquet theory to simulate the dynamics of a charge-driven transmon. They analyzed the hybridization parameter $\Theta$ , which quantifies the deviation of the driven state from an ideal displaced state.
- Simulations predicted that by operating with a large detuning ( $\omega_r/\omega_q > 10$ ), the drive frequency would fall outside the range of resonant transitions within the Josephson potential well, effectively suppressing unwanted state transitions.
Experimental Protocol:
- The device was cooled to 10 mK in a dilution refrigerator.
- State Transitions Test: The qubit was prepared in $|0\rangle$ or $|1\rangle$ , subjected to a strong drive near the readout frequency with varying photon numbers ( $\bar{n}_d$ ), and then measured to detect any transitions to higher states.
- Readout Fidelity Test: The qubit was read out using varying numbers of photons in the resonator ( $\bar{n}_r$ ) to determine the measurement fidelity without a quantum-limited amplifier.

3. Key Contributions

Extension of cQED to mmWave: This work successfully extends the cQED toolbox into the 30+ GHz range, demonstrating a hybrid system where a microwave qubit interacts with mmWave photons.
Suppression of Resonant Transitions: The paper provides experimental proof that a large frequency detuning ( $\omega_r/\omega_q > 10$ ) effectively eliminates drive-induced resonant transitions that plague conventional readout schemes.
High-Power Readout: The system allows for the application of extremely strong readout drives (up to $\sim 1,000$ photons) without causing the qubit to escape its computational subspace, a regime previously inaccessible due to transition errors.
High-Fidelity Readout without Amplification: The authors achieved a measurement fidelity of >99% using more than 100 photons in the resonator, all without the aid of a quantum-limited amplifier (JPA/JTWPA), relying instead on the suppression of transitions to boost signal-to-noise ratio.

4. Results

Suppression of Transitions:
- In the transition probability tests, no resonant state transitions were observed for drive photon numbers up to $\bar{n}_d \approx 1,000$ when the qubit was initialized in $|0\rangle$ or $|1\rangle$ .
- The probability of remaining in the initial state remained near unity ( $P_{0\to0} \approx 0.998$ ), consistent with the residual thermal population and $T_1$ decay, rather than drive-induced errors.
- Transitions were only observed at very high drive powers ( $\bar{n}_d \gtrsim 1,000$ ), corresponding to the quantum-to-classical transition where the drive is strong enough to push the qubit out of the potential well entirely, rather than a resonant transition.
Readout Fidelity:
- Using a fixed readout time of $\tau_r = 780$ $τ_{r} = 780$ ns, the measurement fidelity increased with photon number:
  - $\bar{n}_r = 1$ : Fidelity $\approx 53.5\%$
  - $\bar{n}_r = 10$ : Fidelity $\approx 79.0\%$
  - $\bar{n}_r = 109$ : Fidelity $\approx 99.2\%$
System Parameters:
- The system achieved a cross-Kerr interaction rate of $\chi/2\pi \approx -1.515$ MHz, enabled by the high coupling rate $g$ .
- The measurement efficiency was measured at $\eta = 1.74\%$ , corresponding to a system added noise of $\bar{n}_{sys} \approx 28$ photons. The authors note that adding a mmWave quantum-limited amplifier could significantly improve this.

5. Significance and Future Outlook

Scalability: As quantum processors scale, the number of control lines increases. This architecture minimizes "spectral crowding" and accidental collisions between qubit frequencies and parametric processes used for inter-qubit coupling, as the readout band is spectrally isolated.
Hybrid Quantum Systems: This work establishes mmWave photons as a critical resource for connecting disparate quantum platforms (e.g., Rydberg atoms, optomechanics, and superconducting qubits), facilitating hybrid quantum networks.
New Frontiers: The ability to operate in the ultra-strong coupling regime ( $g/\omega_q \approx 0.4$ ) opens new avenues for studying light-matter interactions, quantum memories, and error correction protocols at mmWave frequencies.
Dark Matter Search: The platform is directly applicable to dark matter searches (e.g., axion detection) using superconducting qubits, where mmWave frequencies are often the target range.

In conclusion, this paper demonstrates that moving the readout frequency to the millimeter wave regime is a viable and powerful strategy to overcome the fundamental limits of drive-induced errors in superconducting qubit readout, enabling high-fidelity measurements with high photon numbers and paving the way for more complex hybrid quantum systems.