A tri-linear quantum dot architecture for semiconductor spin qubits

Abstract

Semiconductor quantum dot spin qubits hold significant potential for scaling to millions of qubits for practical quantum computing applications, as their structure highly resembles the structure of conventional transistors. Since classical semiconductor manufacturing technology has reached an unprecedented level of maturity, reliably mass-producing CMOS chips with hundreds of billions or even trillions of components, conventional wisdom dictates that leveraging CMOS technologies for quantum dot qubits can result in upscaled quantum processors with thousands or even millions of interconnected qubits. However, the interconnect requirements for quantum circuits are very different from those for classical circuits, where for each qubit individual control and readout wiring could be needed. Although significant developments have been demonstrated on small scale spin qubit systems, qubit numbers remain limited, to a large extent due to the lack of scalable qubit interconnect schemes. Here, we present a tri-linear quantum dot array that is simple in physical layout while allowing individual wiring to each quantum dot. By means of electron shuttling, the tri-linear qubit architecture provides qubit connectivity that is equivalent to or even surpasses that of 2D square lattice qubit arrays. Assuming the current qubit fidelities of small-scale qubit devices can be extrapolated to large-scale arrays, even medium-length shuttling arrays on the order of tens of microns in length would already allow the realization of million-scale qubit systems, while maintaining manageable overheads. Beyond the qubit architecture, we also present a scalable control scheme, where the qubit chip is 3D-integrated with a low-power switch-based cryoCMOS circuit for parallel qubit operation with limited control inputs. As our tri-linear quantum dot array is fully compatible with existing semiconductor technologies, this qubit architecture represents one possible framework for future research and development of large-scale spin qubit systems.