🔬 mesoscale physics

g-factor theory of Si/SiGe quantum dots: spin-valley and giant renormalization effects

This paper presents a comprehensive theory for calculating the $g$ -factor in Si/SiGe quantum dots applicable to general heterostructures, revealing significant renormalization effects in "Wiggle Well" structures and a giant $g$ -factor suppression driven by spin-valley coupling at low valley splitting locations.

Original authors: Benjamin D. Woods, Merritt P. Losert, Robert Joynt, Mark Friesen

Published 2026-09-25

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Benjamin D. Woods, Merritt P. Losert, Robert Joynt, Mark Friesen

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

In the quest to build a quantum computer, scientists are looking for tiny, isolated islands of electricity called quantum dots, where they can trap a single electron to act as a bit of information. These bits, or qubits, rely on a property of the electron called spin, which behaves like a tiny internal compass needle. To control this needle, researchers apply a magnetic field, which causes the electron's energy to split into two distinct levels, much like a radio tuning to a specific frequency. The strength of this split depends on a number known as the g-factor. In a perfect, empty world, this number is a fixed value for all electrons, but in the real, messy world of a silicon chip, the local environment can nudge this number slightly. For quantum computers to work reliably, these nudges must be understood and controlled, because even tiny, unpredictable shifts can cause the computer to lose its memory or make errors.

A team of researchers at the University of Wisconsin-Madison has developed a new way to calculate exactly how this g-factor behaves in silicon chips made from layers of silicon and silicon-germanium. While previous theories could explain the behavior in simple, idealized shapes, they failed when faced with the complex, wavy structures that are currently being built to improve performance. The researchers created a universal mathematical framework that can predict the g-factor for any shape of these chips, accounting for the messy reality of atomic disorder and the specific way electrons move between different energy valleys within the silicon crystal. Their work reveals that in certain engineered structures, the g-factor can be dramatically altered in ways that were previously invisible to scientists.

The team focused their attention on a specific design called a "Wiggle Well," a quantum dot where the concentration of germanium atoms oscillates up and down like a wave inside the silicon layer. This structure was designed to boost the interaction between the electron's spin and its motion, a phenomenon known as spin-orbit coupling. Using their new theory, the researchers found that this oscillation does more than just tweak the g-factor; it can amplify the effect by a thousand times compared to standard silicon chips. In these Wiggle Wells, the g-factor can shift by a massive amount, changing the electron's response to magnetic fields in a way that is far more significant than anything observed in previous experiments.

Perhaps even more surprising is the discovery of a "giant suppression" effect. In specific spots within these quantum dots, where the energy difference between the electron's two possible valley states becomes very small, the g-factor can drop to nearly zero. This happens because the electron's spin becomes locked to its valley state, a condition where the internal compass needle refuses to turn in response to the external magnetic field unless the field is perfectly aligned with a specific axis. The researchers found that this suppression is not a rare fluke but a robust feature that occurs in many locations across the chip. Crucially, they showed that this effect is stable even when the chip is subjected to electrical noise, which typically scrambles quantum states. The suppression remains intact because the physical region where it occurs is large enough to resist the jitter of random electrical fluctuations.

This work overturns the idea that the g-factor in silicon is a static, predictable number that only changes by tiny fractions. Instead, the researchers demonstrate that it is a dynamic property that can be engineered to be either highly sensitive or completely inert, depending on the local atomic landscape. They also clarified a long-standing confusion about how different types of spin-orbit coupling contribute to the g-factor versus other control methods, showing that they depend on different physical mechanisms. By providing a clear map of how these factors interact, the study offers a new toolkit for engineers. They can now design quantum dots with specific g-factor properties, potentially using the giant suppression effect to solve problems like frequency crowding, where too many qubits try to operate at the same frequency, or to create more stable qubits that are less likely to be disturbed by their environment. The findings suggest that by carefully shaping the atomic structure of these chips, scientists can gain precise control over the fundamental physics that drives the next generation of quantum computers.

Problem Statement
Silicon/silicon-germanium (Si/SiGe) quantum dots are promising candidates for scalable spin qubits, where qubit operations rely on the Zeeman splitting ( $E_z$ ) of a single-electron spin in a magnetic field. While the free-electron g-factor ( $g_0 \approx 2$ ) provides a baseline, spin-orbit coupling (SOC) introduces corrections that are critical for understanding local g-factor fluctuations, which impact qubit coherence and control. Existing theoretical frameworks for g-factor physics in Si/SiGe are limited: they are largely specialized to triangular well geometries, rely on vaguely defined SOC parameters, and fail to fully account for valley physics. Consequently, these models cannot explain g-factor behavior in "Wiggle Well" (WW) structures—heterostructures featuring oscillating Germanium concentrations—and have overlooked potential regimes of significant g-factor renormalization.

Methodology
The authors develop a general theory for g-factor physics applicable to all Si/SiGe heterostructures of current interest. The approach involves:

Effective-Mass Modeling: Constructing a Hamiltonian that treats the two conduction band valleys near the Z-point as a pseudospin-1/2 degree of freedom. The model incorporates in-plane magnetic fields, arbitrary Ge concentration profiles (including WW oscillations), and random-alloy disorder.
Derivation from Tight-Binding: The effective-mass model is rigorously derived from a minimal tight-binding model augmented with Dresselhaus SOC, ensuring the correct momentum selection rules and valley coupling mechanisms are captured.
Schrieffer-Wolff Transformation: A low-energy effective Hamiltonian is derived for the ground orbital state by projecting out high-energy subbands. This transformation yields an effective Hamiltonian containing terms for valley splitting, Zeeman coupling, and a novel spin-valley coupling term.
Matrix Element Formulation: The g-factor is expressed in terms of readily computable matrix elements, including SOC coefficients ( $\beta_{\mu,\nu}$ ), momentum matrix elements ( $k_{z}^{\mu,\nu}$ ), subband energies ( $\epsilon_\nu$ ), and valley-orbit matrix elements ( $\tilde{\Delta}_{m,n}^{\mu,\nu}$ ).

Key Contributions and Results
The paper presents two primary findings regarding g-factor physics in Si/SiGe quantum dots:

Giant Renormalization in Wiggle Wells:
In the "normal" valley splitting regime (where valley splitting $E_v$ is large), the authors demonstrate that WW structures can induce g-factor variations ( $|g - g_0|$ ) of order $O(10^{-2})$ . This is 1–2 orders of magnitude larger than in conventional Si/SiGe quantum wells without Ge oscillations. This enhancement arises because the Ge concentration oscillations significantly boost the intersubband SOC coefficients, which drive g-factor renormalization. The theory distinguishes between intersubband SOC (driving g-factor variations) and intrasubband SOC (driving electric dipole spin resonance), noting their different dependencies on the oscillation wavelength $\lambda$ .
Giant g-Factor Suppression via Spin-Valley Locking:
The most significant finding is a regime where g-factor variations reach order $O(1)$ , representing a suppression three orders of magnitude larger than previously observed. This occurs in regions where the valley splitting is locally suppressed ( $E_v \lesssim 2 \mu eV$ ) due to random-alloy disorder, leading to a condition where the spin-valley energy ( $\epsilon_{sv}$ ) dominates.
- Mechanism: In these regions, the spin and valley degrees of freedom become "locked." The spin quantization axis is determined by the spin-valley field ( $B_{sv}$ ) rather than the external magnetic field ( $B$ ).
- Consequence: If the external magnetic field is orthogonal to the spin-valley field, the g-factor can effectively vanish ( $g \approx 0$ ).
- Robustness: The authors show that while the location of these suppression regions depends on local disorder, the suppression effect itself is robust against charge noise. Even when charge noise displaces the quantum dot from the exact $E_v=0$ point, the ratio of valley coupling to spin-valley energy remains small enough to maintain the giant suppression.

Significance
The authors claim this work provides a universal theoretical framework for g-factor engineering in Si/SiGe quantum dots. By expressing the g-factor in terms of computable matrix elements, the theory allows for the prediction and design of g-factor behavior in complex geometries, including those with disorder.

The discovery of the giant g-factor suppression regime offers new avenues for qubit control. Specifically, the ability to suppress or extinguish magnetic field couplings could help overcome scaling challenges such as frequency crowding in large qubit arrays or facilitate singlet-triplet oscillations in double quantum dots. Furthermore, the theory clarifies the relationship between g-factor physics, valley phases, and disorder, suggesting that g-factor measurements could serve as a tool to probe the valley-phase landscape and characterize valley-coupling statistics in future experiments.

More like this