A peer reviewed experiment shows electron qubits on a helium film can survive a billion shuttling cycles without losing charge, a physics proof of routing qubits by movement rather than by wire.
A quantum computer with a thousand qubits needs a thousand control wires. A peer-reviewed paper this month in Physical Review Applied describes one way to skip that requirement: moving qubits across a film of superfluid helium rather than wiring to each one.
The result comes from EeroQ, a startup that builds spin qubits from single electrons trapped on top of a thin film of helium condensed directly onto a silicon chip. The paper, "Selective shuttling of electrons on helium using a CMOS control platform," reports that electrons driven across the helium surface lost no detectable charge across roughly 10^9 repetitive shuttling cycles arXiv preprint. The transport aggregated to kilometers of total electron movement, and the device held up. "Over the course of this experiment, we observed zero detectable loss of charge from any electron shuttle event," the company wrote in its announcement.
The mechanism is what makes the result architectural rather than incremental. EeroQ's control layer was fabricated on SkyWater Technology's 130-nm commercial silicon process, a standard foundry node rather than a bespoke run, so the chip can in principle scale on ordinary semiconductor tooling. Fourteen external control lines drive 128 independently addressable channels on the helium surface, meaning one wire does the work of several. In most current qubit architectures, moving a qubit from one gate operation to the next means coordinating an entangled choreography of fixed gates and neighbors; in the EeroQ geometry, the qubit itself moves. Trade-press coverage frames the paper as validation of EeroQ's transport layer for "all-to-all qubit routing," an architectural property the company has branded under the internal name "Wonder Lake".
Several caveats matter. The 10^9-cycle figure measures charge survival, not the properties that turn a transport mechanism into a useful quantum computer. Spin coherence times (T2), single- and two-qubit gate fidelities, and crosstalk at scale are not in the published excerpt. All co-authors are EeroQ-affiliated, so the result is a peer-reviewed physics demonstration of one path past the wiring wall, not an independent replication of a working qubit array. The trade-press framing of the work as a hardware milestone is curation rather than third-party measurement.
The peer-reviewed result establishes one concrete claim: an electron can be moved across a helium surface using commercial CMOS control without losing it, surviving roughly a billion round trips. If that holds up in independent labs, the implication is concrete. Floating the qubit on an atomically smooth superfluid film and shuttling it past control gates turns per-qubit wiring into shared control infrastructure. A denser, more routable array becomes physically possible without waiting on exotic cryogenic hardware.
The remaining questions live in the next papers. How long an electron spin stays coherent while crossing the helium film, the T2 figure that competing qubit platforms measure in microseconds to milliseconds, will determine whether shuttling introduces its own error budget. Parallel operation across all 128 channels and full-wafer yield on the foundry process remain unanswered. EeroQ has not published a timeline for a product chip. The result is a physics-engineering validation that one of the field's hardest scaling problems has at least one credible experimental route past it, not a quantum computer on a shelf.