The coherent tunnelling of Cooper pairs across Josephson junctions (JJs) generates a nonlinear inductance that is used extensively in quantum information processors based on superconducting circuits, from setting qubit transition frequencies and interqubit coupling strengths to the gain of parametric amplifiers for quantum-limited readout. The inductance is either set by tailoring the metal oxide dimensions of single JJs, or magnetically tuned by parallelizing multiple JJs in superconducting quantum interference devices with local current-biased flux lines. JJs based on superconductor-semiconductor hybrids represent a tantalizing all-electric alternative. The gatemon is a recently developed transmon variant that employs locally gated nanowire superconductor-semiconductor JJs for qubit control. Here we go beyond proof-of-concept and demonstrate that semiconducting channels etched from a wafer-scale two-dimensional electron gas (2DEG) are a suitable platform for building a scalable gatemon-based quantum computer. We show that 2DEG gatemons meet the requirements by performing voltage-controlled single qubit rotations and two-qubit swap operations. We measure qubit coherence times up to ~2 μs, limited by dielectric loss in the 2DEG substrate.
Box 1: Example Platforms for Analogue Quantum Simulation Analogue Quantum Simulations are today performed on a variety of platforms [15][16][17][18][19][20], each of which offer distinct features that make them more suitable for specific simulation tasks. Ultracold atoms in optical lattices -Currently up to 3000 atoms in optical potentials with single atom detection and control via so-called quantum gas microscopes. These uniquely implement models of interacting fermionic particles (such as the Hubbard model, see Box 2) or its bosonic variant using bosonic atoms. Spin models can also be engineered, with tailored local parameters and potentials shaped by spatial light modulators [21,22]. Trapped Ions -Currently ca. 50 ions in linear chains (Paul traps) or 2D arrays (Penning traps), with local addressing. Using laser excitations and the collective motion of the array to generate effective interactions, a range of magnetic spin models can be engineered, with tailored long-range interactions. They benefit from readout techniques and manipulation developed for digital quantum computing in ion traps [17]. Atom arrays with Rydberg interactions -Currently ca. 256 atoms held in optical tweezers, with spin models generated by excitation to high-lying electronic states. They benefit from arrangements in arbitrary geometries, and tailored magnetic spin models from different choices in laser excitation to excited states [20]. Superconducting circuits -Currently ca. 50-130 superconducting resonators, which can be used as qubits (spins) or anharmonic oscillators (interacting bosons), with couplings adjusted on the level of individual resonators. As with trapped ions, they benefit directly from architectures and readout techniques in superconducing quantum computers [19,23]. Photonic waveguides or beamsplitter arrays -Currently up to ca. 50-100 channels and photons, either chip-based or with large-scale beamsplitter arrays. They generally implement models for non-interacting bosons, including boson sampling, which is exponentially difficult to reproduce using classical calculations [18,24].Box 2: The Hubbard model A particularly good example of the complex problems addressed in quantum simulation is the determination of low-energy properties and dynamics in a Hubbard model, which is a prototypical model for describing strongly interacting electrons in solids [25]. Although in one dimension the Hubbard model is exactly solvable [26], in two dimensions, it has been a long-term challenge even to find the lowest energy states of this model, although a lot of recent progress has been made [27,28].
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.