Practical Strategies for Enhancing the Valley Splitting in Si/SiGe Quantum Wells

Merritt P. Losert,M. A. Eriksson,Robert Joynt,Rajib Rahman,Giordano Scappucci,Susan N. Coppersmith,Mark Friesen
DOI: https://doi.org/10.1103/PhysRevB.108.125405
2024-01-11
Abstract:Silicon/silicon-germanium heterostructures have many important advantages for hosting spin qubits. However, controlling the valley splitting (the energy splitting between the two low-lying conduction-band valleys) remains a critical challenge for ensuring qubit reliability. Broad distributions of valley splittings are commonplace, even among quantum dots formed on the same chip. In this work, we theoretically explore the interplay between quantum-well imperfections that suppress the valley splitting and cause variability, such as broadened interfaces and atomic steps at the interface, while self-consistently accounting for germanium concentration fluctuations. We consider both conventional and unconventional approaches for controlling the valley splitting, and present concrete strategies for implementing them. Our results provide a clear path for achieving qubit uniformity in a scalable silicon quantum computer.
Mesoscale and Nanoscale Physics
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to enhance valley splitting in Si/SiGe quantum wells. Specifically, the researchers focus on how to control and increase the valley - splitting energy (\(E_v\)) in quantum dots to ensure the reliability of spin qubits. In silicon - based materials, due to the degeneracy of the low - energy states in the conduction band (i.e., "valleys"), especially in quantum dots, this degeneracy can lead to leakage paths outside the logical space, thus threatening the operation of qubits. ### Core challenges of the problem include: 1. **Wide range of valley - splitting energy**: Even among quantum dots fabricated on the same chip, the valley - splitting energy (\(E_v\)) can vary from 20 to 300 micro - electron volts (µeV), which leads to significant differences in device performance. 2. **Influence of interface defects and disorder**: In traditional SiGe/Si/SiGe heterostructures, valley splitting is mainly determined by the confinement potential of the quantum well. However, defects such as the width, inclination, and single - atom steps at the interface will introduce additional variability. In addition, random disorder in SiGe alloys (such as fluctuations in Ge concentration) can also significantly affect valley splitting. 3. **Difficulty in achieving high and consistent valley splitting**: So far, no reliable method has been found to design devices with high and uniform valley splitting, which is crucial for large - scale quantum computing. ### Research objectives: - **Understand and quantify the influence of various factors on valley splitting**: including interface width, atomic steps, alloy disorder, etc. - **Propose specific strategies to enhance valley splitting**: Through theoretical simulation and analysis, explore how to use sharp interfaces, special geometric structures (such as Wiggle Well), etc. to enhance valley splitting. - **Provide guidance for experiments**: Through theoretical model prediction and optimization, provide a basis for future experimental designs to achieve more reliable qubit performance. ### Main contributions: - Proposed a unified theoretical framework (2k0 theory), which explains how all key features (from deterministic to random) in heterostructures work together to determine valley splitting. - Compared the effectiveness of traditional and non - traditional heterostructure designs and proposed the optimal strategy to enhance valley splitting, especially under different operating conditions. Through these studies, the authors hope to provide important theoretical support and technical guidance for building scalable silicon - based quantum computers.