Brain-Inspired Chip Runs Near Absolute Zero, Pointing to Cooler, Smarter Quantum Computers

Engineers in Hong Kong have built a brain-inspired chip that can operate at temperatures just above absolute zero, an advance that could help solve one of the most stubborn bottlenecks in quantum computing. The device, described this week in the journal Nature Communications, coaxes a single ordinary transistor into behaving like an energy-efficient biological neuron.

The work, led by Professor Yuhao Zhang and PhD student Xin Yang of the University of Hong Kong's Department of Electrical and Computer Engineering, exploits an unusual electrical behavior in silicon carbide, or SiC. By driving the transistor into a regime known as negative differential resistance, the team got it to generate sharp electrical "spikes" — the same signaling currency the human brain uses to process information — at temperatures as low as 10 millikelvin.

That temperature is no arbitrary milestone. The sensitive quantum bits, or qubits, at the heart of a quantum computer must be chilled to within thousandths of a degree of absolute zero to function. The control electronics that read and steer those qubits, however, typically sit far away at room temperature, connected by bulky wiring that generates heat and limits how large a quantum machine can grow. A chip that can compute right next to the qubits, without overheating them, could dramatically simplify that architecture.

The behavior arises from a phenomenon called electron-donor impact ionization, an atomic-level cascade that the researchers found to be stable and repeatable across different manufacturing batches. Crucially, the chip remained reliable below 2 Kelvin and is compatible with standard 300-millimeter silicon wafers, meaning it could in principle be mass-produced using existing semiconductor fabrication lines rather than exotic one-off processes.

The team reported that the resulting circuits are thousands of times more energy-efficient than conventional electronics, a quality that matters enormously in cryogenic systems where every stray watt of heat is a liability. "Robust and scalable" was how Yang characterized the approach, emphasizing that its compatibility with industrial manufacturing sets it apart from many laboratory curiosities.

Beyond quantum computing, the researchers suggested the technology could find a home in deep-space exploration, where instruments must endure the brutal cold of the outer solar system on minimal power budgets. Neuromorphic chips that compute the way brains do — efficiently, and in the cold — could one day run autonomous probes far from the sun. For now, the demonstration offers a tangible step toward quantum machines that are easier to scale, and a reminder that some of the field's biggest gains may come not from new qubits, but from rethinking the humble transistor.