Negative exchange interaction in Si quantum dot arrays via valley-phase induced $\mathbb{Z}_2$ gauge field
Benjamin D. Woods
TL;DR
The paper shows that valley-phase differences in Si quantum dot arrays generate a $\mathbb{Z}_2$ gauge field on interdot links, enabling a negative exchange $J<0$ in the two-electron sector without an external magnetic field. By deriving a low-energy ground-valley Hamiltonian with $t'_{i,j} = t_{i,j} cos(phi_{i,j}/2)$ and introducing a gauge field on links, the authors connect valley physics to $\pi$-fluxes threading plaquettes, which modify exchange and can even break Nagaoka ferromagnetism. They explore triangular and square plaquettes to demonstrate negative exchange and flux-driven ferromagnetism breaking, and show that valley-phase configurations can be engineered on demand by selectively occupying excited valleys or using ancillary dots, enabling programmable exchange across 2D dot arrays and 1D chains. The work suggests a path to dynamically corrected exchange-based quantum gates and provides a framework for studying $\mathbb{Z}_2$ gauge-field effects and valley-driven quantum magnetism in semiconductor quantum-dot systems.
Abstract
The exchange interaction $J$ offers a powerful tool for quantum computation based on semiconductor spin qubits. However, the exchange interaction in two-electron systems in the absence of a magnetic field is usually constrained to be non-negative $J \geq 0$, which inhibits the construction of various dynamically corrected exchange-based gates. In this work, we show that negative exchange $J < 0$ can be realized in two-electron Si quantum dot arrays in the absence of a magnetic field due to the presence of the valley degree of freedom. Here, valley phase differences between dots produce a non-trivial $\mathbb{Z}_2$ gauge field in the low-energy effective theory, which in turn can lead to a negative exchange interaction. In addition, we show that this $\mathbb{Z}_2$ gauge field can break Nagaoka ferromagnetism and be engineered by altering the occupancy of the dot array. Therefore, our work uncovers new tools for exchange-based quantum computing and a novel setting for studying quantum magnetism.
