Distributed Quantum Computing in Silicon

TL;DR

This work addresses the challenge of scaling quantum computation beyond a single module by enabling high-fidelity entanglement distribution across modular silicon-based processors. It develops and validates a silicon T-centre platform with a spin–photon interface integrated into nanophotonic cavities, demonstrating remote entanglement between two modules via the Barrett–Kok protocol and a teleported CNOT gate. Key results include cavity-enhanced optical transitions, long spin coherence times (electron and nuclear spins), and measurable remote Bell-pair generation with prospects for near-term improvements toward fault-tolerant distributed computation. The findings establish a telecom-band, silicon-based pathway for Phase 3 distributed quantum computing and inform architecture designs for scalable quantum networks.

Abstract

Commercially impactful quantum algorithms such as quantum chemistry and Shor's algorithm require a number of qubits and gates far beyond the capacity of any existing quantum processor. Distributed architectures, which scale horizontally by networking modules, provide a route to commercial utility and will eventually surpass the capability of any single quantum computing module. Such processors consume remote entanglement distributed between modules to realize distributed quantum logic. Networked quantum computers will therefore require the capability to rapidly distribute high fidelity entanglement between modules. Here we present preliminary demonstrations of some key distributed quantum computing protocols on silicon T centres in isotopically-enriched silicon. We demonstrate the distribution of entanglement between modules and consume it to apply a teleported gate sequence, establishing a proof-of-concept for T centres as a distributed quantum computing and networking platform.

Figures (13)

• Figure 1: Phases of quantum computing. A. In phase 1 Quantum computing (the NISQ phase), quantum computers are single modules containing a small number of physical qubits (orange) too noisy to implement QEC. B. In Phase 2 Quantum computing, quantum computers are still confined to a single module, however the module contains enough physical qubits with low enough noise to encode logical qubits (blue). C. In Phase 3 Quantum computing, quantum computers grow beyond a single module, and can implement large-scale quantum algorithms fault-tolerantly.
• Figure 2: Schematic of the demonstration.A. Schematic of a T centre in the silicon lattice embedded in an optical cavity in a photonic chip. The T centre is excited with light coupled in via a grating coupler and the spin transitions are driven by microwave and radio-frequency drives delivered from a metal antenna. Magnetic field $B_0$ is applied in-plane, perpendicular to the waveguide and $B_1$ is generated from on-chip antennae in the out-of-plane direction. B. Schematic of the two T centre qubit modules separated by approximately 40 meters of fibre. Optical fibre is depicted in red and electrical and microwave lines in grey. The Optical Modulators are an acousto-optic modulator, electro-optic modulator, and semiconductor optical modulator on the excitation path and an acousto-optic modulator on the collection path.
• Figure 3: Individual T centre optical performance.A. Fine structure of a T centre under a magnetic field showing electron ($\ket{\uparrow_e}$), excited state hole ($\ket{\uparrow_h}$), and hydrogen ($\ket{\Uparrow_H}$) spins. Splittings are (from left to right): ground state (GS) to excited state (TX0), electron/hole Zeeman splitting, and electron-nuclear hyperfine. B. PLE spectra (left axis) of the two T centres TC1 and TC2 at a magnetic field of $122.3$ mT, with their corresponding cavity resonances (right axis) in dashed lines. C. Purcell-enhanced lifetimes of the $C$ and $B$ transitions of TC1 and TC2. D. Initialization of the T centres from optical pumping. The dashed vertical line corresponds to the point where the initialization fidelity is $98.3(2)\%$.
• Figure 4: Individual T centre spin performance.A. ODMR showing the PLE signal as a function of MW frequency. The two resonant frequencies corresponding to nuclear-spin-selective MW transitions. B. Coherent Rabi oscillations driving the MW$_{\Downarrow}$ nuclear-spin-selective transition. A $\pi$ rotation constitutes a C$_{\rm H}$NOT$_{\rm e}$ gate. Inset: ground state energy level diagram showing the driven transition. C. ODMR spectrum showing the resonant frequency corresponding to a nuclear transition. D. Coherent Rabi oscillations driving a electron-spin-selective nuclear transition. A $\pi$ rotation constitutes a C$_{\rm e}$NOT$_{\rm H}$. Inset: ground state energy level diagram showing the driven transition. E. Nuclear spin preparation and measurement: number of photons measured for a nuclear spin in the up or down state. Middle inset: the pulse sequence used to initialize and prepare the nuclear spin and readout the spin-state. Right inset: state preparation and measurement fidelity versus photon number threshold.
• Figure 5: Spin coherence times. A. Electron Rabi oscillations, driven without nuclear selectivity, the inset denotes which transition is driven. Ramsey interference fringes showing a $T_2^\ast$ decay for the electron spin, B, and the nuclear spin, C. Inset shows the pulse sequence used. Hahn-echo $T_2$ decay curves for the electron spin, D, and the nuclear spin, E. With further improvements to materials and fabrication, it is likely that the coherence times of device-integrated T centres can be improved significantly toward the coherence times observed in bulk $^{28}$Si (0.28(1) s for nuclear spins and 2.1(1) ms for electron spins Simmons.2020). Inset shows the pulse sequence used.