Monolithic Integration of Quantum Resonant Tunneling Gate on a 22nm FD-SOI CMOS Process

Abstract

The proliferation of quantum computing technologies has fueled the race to build a practical quantum computer. The spectrum of the innovation is wide and encompasses many aspects of this technology, such as the qubit, control and detection mechanism, cryogenic electronics, and system integration. A few of those emerging technologies are poised for successful monolithic integration of cryogenic electronics with the quantum structure where the qubits reside. In this work, we present a fully integrated Quantum Processor Unit in which the quantum core is co-located with control and detection circuits on the same die in a commercial 22-nm FD-SOI process from GlobalFoundries. The system described in this work comprises a two dimensional (2D) 240 qubits array integrated with 8 detectors and 32 injectors operating at 3 K and inside a two-stage Gifford -McMahon cryo-cooler. The power consumption of each detector and injector is 1 mW and 0.27 mW, respectively. The control sequence is programmed into an on-chip pattern generator that acts as a command and control block for all hardware in the Quantum Processor Unit. Using the aforementioned apparatus, we performed a quantum resonant tunneling experiment on two qubits inside the 2D qubit array. With supporting lab measurements, we demonstrate the feasibility of the proposed architecture in scaling-up the existing quantum core to thousands of qubits.

Cryogenic Controller for Electrostatically Controlled Quantum Dots in 22-nm Quantum SoC

“…The quantum dot (QD)-based qubits can be constructed based on the spin (e.g., up/down polarization) [17], [20], [21], [26] or charge (e.g., electron's position within a double QD) [14], [15], [31] of the carriers. When looked in isolation, the spin qubits appear more advantageous than the charge qubits, mainly due to longer decoherence times.…”

“…There are several approaches toward the physical realization of quantum computers, with those most notable based on superconducting [11], photonic [12], ion traps [13], and semiconductor qubits [14]- [31] (see our recent discussion on state-of-the-art in qubits in [7]). The superconducting quantum computing is the most prevalent in industry, as a result of quality of qubits and maturity of the technology, and is currently deployed for commercial applications [32].…”

“…In this article, we offer a vision of a position-based charge qubit structure intended to operate at 4 K and fully integrated with its interfacing circuitry using a volume-production 22-nm fully depleted silicon-on-insulator (FD-SOI) CMOS process that has unique benefits for quantum operation [31], [47]- [49]. In contrast to bulk CMOS, FD-SOI provides a thin semiconductor layer isolated vertically from the substrate by a 20-nm buried oxide (BOX) layer.…”