- Accepted Paper
Robust valley splitting against interface atomic disorder towards scalable silicon quantum computing: Computational design
Phys. Rev. B - Accepted 6 November, 2026
DOI: https://doi.org/10.1103/f66n-c6sm
Phys. Rev. B - Accepted 6 November, 2026
DOI: https://doi.org/10.1103/f66n-c6sm
Inevitableatomic-scale disorderatSi/SiGe quantum-wellinterfaces suppressesvalley splittingand causes substantial variations in its magnitude, hindering uniform operation of silicon spin qubits. Many proposed strategies for enhancing valley splitting rely on idealized heterostructures and can lose their effectiveness in the presence of interface disorder. Here, we propose using in-plane shear strain to achieve large valley splitting with reduced relative variability in conventional Si/SiGe quantum wells. Atomistic pseudopotential calculations predict that just 0.2% shear strain raises the valley splitting above the 0.2 meV target for reliable spin-qubit operation. Crucially, this enhancement does not proportionally amplify disorder-induced fluctuations, allowing the relative standard deviation to decrease from about 34% without shear to below 5% at 0.5% shear strain. We show that this robustness originates from a strain-activated coupling channel between bulk valleys, whose small momentum transfer renders it largely insensitive to atomic-scale interface fluctuations. Our results offer a CMOS-compatible route to overcoming the valley-splitting challenge of Si qubits and establish a microscopic framework for understanding and designing valley physics in semiconductor quantum devices.
If the author has provided any supplemental materials with this article they will be available upon publication of the version of record.