Mind the Gap: Mapping Valley Splitting in Silicon Qubits

Silicon is the bedrock of classical computing, and physicists are keen to ensure it underpins the quantum revolution as well. However, electron spin qubits—the fundamental units of these potential machines—suffer from a specific ailment known as near-degenerate valley states. These states create low-lying excited energy levels that effectively muddy the waters for qubit readout and control fidelity.

The primary culprit appears to be microscopic disorder within the Silicon-Germanium (SiGe) alloy and at the material interfaces. While silicon devices fit neatly into existing manufacturing pipelines, achieving a uniformly large 'valley splitting' energy across a device spanning mesoscopic distances has proven to be an outstanding challenge. It is akin to trying to build a perfectly level railway track over terrain that shifts unpredictably at the microscopic level.

In a recent analysis of a one-dimensional quantum dot array fashioned from a Si_0.972Ge_0.028 quantum well by Intel, researchers mapped these variations with precision. They observed distinct correlations in valley splitting at both the single-gate level (sub-100 nm) and across the wider device (over 1 μm). These findings align perfectly with theories attributing the variance to alloy disorder. By characterising how these splits behave over larger distances, engineers can better design the architecture needed for scalable quantum processors.