Quantum-classical neurons upgraded with optical spin qubits for advanced computing and networking architectures

Research Overview

This paper presents a novel hardware architecture that bridges the gap between neuromorphic computing and quantum information processing. By integrating an opto-electronic hardware neuron—equipped with a memristive memory layer doped with Neodymium rare-earth ions—with optical spin qubits formed from silicon vacancy states in silicon carbide, the author creates a system capable of hybrid quantum-classical operations. The research explores how the spiking patterns generated by the artificial neuron can be used to drive and modulate the quantum spin transitions of the integrated qubits.

The researcher fabricated a device stack consisting of a superconducting Niobium/Niobium-oxide memory layer gated by indium-tin-oxide, which is then interfaced with silicon carbide substrates containing negatively charged silicon vacancies. The system was characterized in a cryo-magneto-optical probe station. The team applied electrical, microwave, and optical stimuli to the neuron to generate specific spiking sequences. These sequences were then used to modulate the photoluminescence and spin transition resonances of the silicon vacancy qubits. A quantum model was developed to describe the interaction between the neuron and the spin qubit, utilizing a spin Hamiltonian that accounts for electronic and nuclear interactions, as well as the effects of applied electric and magnetic fields.

Findings and Implications

The study successfully demonstrates that neuronal spiking sequences can effectively drive and modulate the quantum states of the integrated spin qubits. Experimental measurements confirmed that the non-volatility of the memristive memory can be adjusted via optical excitation, which in turn influences the neuronal output and the resulting quantum spin transitions. Theoretical simulations further suggest that this QN-SQ (Quantum Neuron-Spin Qubit) system can perform distributed quantum information processing, with the ability to adjust interaction strengths dynamically. This work provides a foundational step toward developing hardware neurons that mimic biological biophoton-based information processing while offering the enhanced computational power of quantum systems.