⚛️ quantum physics

Microwave-driven same-species sympathetic cooling for trapped ions

This paper proposes and experimentally demonstrates a simplified sympathetic cooling scheme for trapped-ion quantum computing that uses the same ion species and integrated microwave control, achieving near-ground-state cooling of a two-ion gate mode with a low induced error rate of $1.7(4)\times 10^{-4}$ per cycle.

Original authors: M. C. Smith, E. Vandrey, A. D. Leu, N. Drotleff, A. Agrawal, K. Miyanishi, D. M. Lucas, M. F. Gely

Published 2026-07-23

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: M. C. Smith, E. Vandrey, A. D. Leu, N. Drotleff, A. Agrawal, K. Miyanishi, D. M. Lucas, M. F. Gely

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 computers don't just calculate numbers but solve mysteries that would take the universe a lifetime to crack. This is the promise of quantum computing, a field where tiny particles called ions act as the brain's neurons. But here's the catch: these ions are incredibly sensitive, like a house of cards in a hurricane. To do their magic, they need to be perfectly still, vibrating as little as possible. If they jiggle too much, the information they hold gets scrambled. Scientists have a clever trick to keep them calm: they bring in a "coolant" friend. Think of it like a dance partner who absorbs all the clumsy steps so the main dancer can stay graceful. Usually, this requires two different types of dancers (different ion species) and a complex setup of lasers to keep them separate. But what if you could use the exact same type of dancer for both roles? That's the big question this paper tackles, aiming to simplify the dance floor and make quantum computers more practical.

The researchers, working with a team at the University of Oxford, have come up with a new way to cool these quantum ions using the same species for both the data and the cooling, driven by microwaves instead of just lasers. They tested this idea using Calcium-43 ions and found that they could successfully chill the shared motion of the ions down to a near-perfect stillness, reaching an average vibration level of just 0.16 quanta. Even more impressively, they proved that this cooling process didn't mess up the quantum information stored in the data ion, introducing a tiny error rate of only 1.7(4) × 10⁻⁴ per cooling cycle. This suggests that we can build simpler, more efficient quantum computers without needing a zoo of different ion types or a maze of lasers, provided we can manage the heat generated by the microwave controls.

The Quantum Dance Floor

To understand why this matters, let's picture a quantum computer as a high-stakes dance floor. The "data ions" are the star performers, holding the secret recipes for solving complex problems. But these stars are fragile; if the floor beneath them shakes (a phenomenon called "heating"), their moves get sloppy, and the recipe gets ruined. To fix this, scientists use "sympathetic cooling." Imagine a second dancer, the "coolant ion," who is really good at absorbing energy. When the floor shakes, the coolant ion grabs the energy and dumps it out, leaving the star performer calm and collected.

Traditionally, this required two different species of ions—like a human and a robot dancing together. The robot (coolant) would be cooled by lasers that didn't bother the human (data). While effective, this is like trying to build a dance studio that can handle both ballet and breakdancing; you need different equipment, different lights, and a lot of extra space. It gets complicated, expensive, and hard to scale up.

The Microwave Magic Trick

The team in this paper asked a bold question: What if we used two humans for the dance? Could we use the same type of ion for both the data and the cooling? The problem is that if you shine a laser on one, you might accidentally zap the other. To solve this, the researchers used a clever trick involving magnetic fields and microwaves.

They placed the ions in a magnetic field that acted like a prism, splitting the energy levels of the ions so that the "data" ion and the "coolant" ion sang at slightly different pitches. Even though they were the same species, the magnetic field made them distinct enough to be addressed separately. Instead of using complex, tightly focused lasers to target just the coolant, they used microwaves—the same kind of waves that heat your popcorn, but tuned to a very specific frequency.

Here's how the process worked, step-by-step:

Preparation: They got the ions ready, putting the data ion in one state and the coolant ion in another.
The Cooling Pulse: They fired a microwave pulse at the coolant ion. This pulse was tuned to grab energy from the shared motion of the ions (the shaking floor) and dump it into the coolant ion's internal state.
Resetting: Once the coolant ion had absorbed the heat, they used a quick sequence of lasers and microwaves to reset it, readying it to absorb more heat.
Protection: Crucially, they had to make sure the microwave pulses didn't accidentally disturb the data ion. They did this by using a "compensation" pulse—a counter-movement that canceled out any unwanted shifts, much like noise-canceling headphones blocking out background chatter.

The Results: A Near-Perfect Chill

The team tested this on a pair of Calcium-43 ions trapped 40 micrometers above a chip. They started with the ions vibrating at a level of about 1.7 quanta (a unit of vibration energy). After running their microwave-driven cooling cycle 16 times, they brought the vibration down to 0.16(2) quanta. This is incredibly close to the "ground state," the absolute coldest, stillest state possible in the quantum world.

But the real test was whether this process hurt the data. They ran a series of random logic gates on the data ion while interleaving the cooling cycles. The result? The cooling introduced an error of only 1.7(4) × 10⁻⁴ per cycle. To put that in perspective, if you cooled the ions a million times, you'd only expect about 170 errors. This is a very small price to pay for keeping the system stable.

Why This Matters (and What's Still Hard)

The paper explicitly rules out the idea that you need different ion species to do this job. They showed that using the same species is not only possible but can be done with simpler equipment (microwaves instead of complex laser arrays). However, they also highlighted that the current error rate isn't perfect yet. The main culprit for the errors wasn't the cooling itself, but the heat generated by the microwave equipment. The microwaves caused tiny thermal drifts that shifted the frequencies slightly, leading to small mistakes.

The authors suggest that if they moved the experiment to a cryogenic (super-cold) environment, they could reduce these thermal errors significantly, potentially lowering the error rate by two orders of magnitude. They also noted that the fundamental limit of this method would eventually be the scattering of photons during the laser reset steps, but that limit is far lower than the errors they are seeing now.

In short, this paper demonstrates a promising new path for quantum computing: a simpler, more unified way to keep the quantum dance floor still, using the same dancers for both the performance and the cleanup crew. While there are still technical hurdles to clear, the proof of concept is solid, suggesting that the future of quantum computing might be less about managing a zoo of different ions and more about mastering the rhythm of a single, well-coordinated team.

Technical Summary: Microwave-driven Same-Species Sympathetic Cooling for Trapped Ions

Problem Statement
Sympathetic cooling is a critical requirement for trapped-ion quantum computing to mitigate motional heating caused by electric field noise and ion transport, which degrades entangling gate fidelities. Conventional approaches utilize a second ion species as a "coolant" to cool shared motional modes without disturbing the "data" qubits. However, this necessitates complex laser infrastructure to address distinct wavelengths and introduces mass-mismatch issues that can reduce cooling efficiency. While using isotopes of the same element mitigates mass mismatch, it often requires spatially addressing only the coolant ion or mapping qubits to metastable states to avoid off-resonant scattering errors. Existing methods to protect data qubits during same-species cooling have relied on tightly focused lasers or local magnetic field gradients, limiting scalability and increasing experimental overhead.

Methodology
The authors propose and demonstrate a sympathetic cooling scheme using ions of the same species and isotope ( $^{43}\text{Ca}^+$ ) with data qubits encoded in the ground-state manifold. The key innovation is the use of a homogeneous static magnetic field to create large Zeeman splittings, separating the data and cooling transitions in frequency space. This allows for global addressing via integrated near-field microwaves (MWs) without the need for tightly focused lasers.

The experimental setup involves two $^{43}\text{Ca}^+$ ions (one data, one coolant) trapped 40 $\mu$ m above a surface-electrode Paul trap. The data qubit is defined by the clock states $|4, 1\rangle$ and $|3, 1\rangle$ (connected by a 3.123 GHz transition), while the coolant ion utilizes the $|4, 2\rangle \leftrightarrow |3, 2\rangle$ transition (2.908 GHz) for resolved sideband cooling.

The cooling cycle proceeds as follows:

State Preparation: Ions are initialized in specific hyperfine states ( $|3, 1\rangle$ for data, $|3, 2\rangle$ for coolant) using Doppler cooling and microwave-enhanced optical pumping.
Cooling: A global microwave pulse drives the red motional sideband of the coolant ion's $|3, 2\rangle \leftrightarrow |4, 2\rangle$ transition. This removes energy from the shared motional mode.
Re-initialization: The "cooled" population in $|4, 2\rangle$ is optically pumped via a 729 nm $\pi$ -pulse to a metastable state ( $|D_{5/2}, 6, 6\rangle$ ) and repumped back to the ground state using an 854 nm pulse. Simultaneously, the remaining population is transferred via microwaves to recombine the states.
Data Protection: To protect the data qubit from ac Zeeman shifts induced by the cooling microwaves, the authors employ a "compensation" microwave tone detuned from the data transition. Furthermore, cooling cycles are executed in pairs interleaved with spin-echo pulses on the data qubit to cancel residual frequency drifts and ac Zeeman shifts.

Key Results

Cooling Performance: The scheme successfully cooled the in-plane radial rocking mode of the two-ion crystal to an average phonon occupation of $\bar{n} \approx 0.16(2)$ . The cooling rate was measured at $0.27(1)$ quanta/cycle (or 290(10) quanta/s) starting from $\bar{n}=1$ .
Qubit Fidelity: Interleaved Randomised Benchmarking (IRB) was used to measure the error induced on the data qubit. The total error per cooling cycle was found to be $1.7(4) \times 10^{-4}$ .
Error Breakdown: The majority of the error ( $1.2(3) \times 10^{-4}$ ) was attributed to the microwave sideband pulse and associated ac Zeeman shift compensation, likely due to thermal transients in the microwave drive chain. The contribution from laser pulses was indistinguishable from the "idle error" of $0.7(2) \times 10^{-4}$ .
Scalability: The authors estimate that maintaining $\bar{n}=1$ would require approximately 50 cooling cycles per second. The contribution of this cooling scheme to the error of a two-qubit gate is estimated to be $\sim 10^{-6}$ , significantly below current state-of-the-art gate errors.

Significance and Claims
The paper claims to demonstrate the first same-isotope sympathetic cooling method where both logical and cooling operations reside in the ground-state manifold and are driven entirely by microwaves. This approach eliminates the need for additional laser species and tightly focused beams, simplifying the optical setup. The authors assert that the demonstrated error rate is sufficiently low to support long sequences of gates and that the technically dominated error rate can be further reduced through cryogenic operation and improved microwave power management. The fundamental limit of the error is estimated to be two orders of magnitude lower than the measured value, primarily limited by photon scattering during the repumping step. This work provides a pathway toward scalable trapped-ion quantum computing by integrating cooling and control into a unified microwave-driven framework.

The Quantum Dance Floor

The Microwave Magic Trick

The Results: A Near-Perfect Chill

Why This Matters (and What's Still Hard)

Technical Summary: Microwave-driven Same-Species Sympathetic Cooling for Trapped Ions

More like this