Short answer: Spin qubits in silicon sit in small quantum dots, so linking distant ones means either coupling them at a distance or physically moving the electron. Conveyor-mode shuttling carries an electron in a moving potential wave using only a few control signals. Experiments report single-electron shuttling fidelity of about 99.4% over a 420 nm channel, and a 2025 result moving an electron over an effective 10 micrometres in under 200 ns with an average spin fidelity of 99.5%. These are laboratory demonstrations.
Why shuttle electrons
A 2025 Nature Nanotechnology paper says the power and fault tolerance of future quantum processors come in large part from qubit connectivity. One way to raise it is to engineer interactions at a distance, and another is to displace the qubits. It notes that for semiconductor spin qubits, several studies have looked at spin-coherent shuttling of single electrons, but high-fidelity transport over extended distances had not yet been shown. A 2024 Nature Communications paper calls long-ranged coherent coupling a missing function block for scaling up, and says conveyor-mode shuttling could enable sparse, scalable qubit architectures with few tunable input terminals and compatibility with industrial gate fabrication.
Proof of principle, 2022
A 2022 npj Quantum Information paper reports shuttling of a single electron by a propagating wave potential in an electrostatically defined 420 nm quantum channel in Si/SiGe. The approach needs only four sinusoidal control signals, independent of channel length. The authors discuss tuning the signal parameters, detect the smoothness of the electron motion, which lets them map potential disorder, and observe a single-electron shuttling fidelity of 99.42 +/- 0.02% including a reversal of direction. They say the device could be embedded in industrial fabrication of Si/SiGe chips and address the signal-fanout problem.
Spin coherence during shuttling, 2024
The 2024 paper studies spin coherence in conveyor-mode shuttling by separating and rejoining an Einstein-Podolsky-Rosen spin pair. Compared with earlier work, the authors raise the shuttle velocity by a factor of 10,000. They observe the spin dephasing time rising with longer shuttle distance, owing to motional narrowing, and estimate the spin-shuttle infidelity due to dephasing at 0.7% for a nominal total distance of 560 nm. After several loops, up to an accumulated 3.36 micrometres, spin entanglement of the pair is still detectable.
Longer distances, 2025
The 2025 paper reports shuttling in an isotopically purified Si/SiGe heterostructure with electric gate potentials. In bucket-brigade shuttling, an electron is moved repeatedly between up to five static dots. In conveyor mode, a travelling-wave potential formed with one or two sets of sine waves carries the electron in a moving dot, and the authors report spin coherence an order of magnitude better than bucket-brigade. They report displacing an electron over an effective distance of 10 micrometres in under 200 ns while preserving the spin state with an average fidelity of 99.5%.
Limits
The papers differ in channel length, isotopic purity and how fidelity is defined, so the numbers are not directly comparable. They move single electrons between prepared locations, and none reports shuttling inside a working multi-qubit processor.
FAQ
What is conveyor-mode shuttling?
Carrying an electron in a moving quantum dot created by a travelling wave potential.
Why only four control signals?
The 2022 paper says the approach needs four sinusoidal signals regardless of channel length.
How far has an electron been shuttled?
An effective 10 micrometres in the 2025 paper, in under 200 ns.
What fidelity was reported?
About 99.4% in 2022 and an average of 99.5% in 2025, with each paper defining its own measure.
Is it used in a processor yet?
These papers report single-electron experiments, not a processor.
All field notes