EAGER: Quantum Manufacturing: Machine learning-powered deterministic nanoassembly of ultrafast quantum photonic devices

点击翻译按钮获取中文摘要

Single photons carry quantum information faithfully and quickly, making quantum photonic networks a key component of all quantum information technologies. Realizing quantum photonic systems is not currently possible on a single material platform, and devices must be assembled from dissimilar materials. Quantum light can be produced by atom-like systems in nanoscale particles, which can supply the required quantum functionality to the established classical photonic platforms. Nanoparticles are also compatible with the nanophotonic enhancement of single-photon emission, helping to boost front-end quantum bitrates, a long-standing problem for quantum photonic networks. However, there is no established manufacturing process for quantum devices realized with nanoparticles. The team will produce a blueprint for such a process by addressing key bottlenecks in nanoparticle post-processing, selection, and manipulation using neural network-assisted control. This research will enable unique workforce development opportunities, including a crowdsourcing initiative at the undergraduate level to train neural networks controlling the manufacturing process.Quantum photonic networks promise to power future distributed quantum computers, secure communication links and distributed quantum sensors. The diverse and stringent requirements on quantum photonic devices for quantum networks cannot be currently satisfied on a single material platform, and a hybrid manufacturing approach is strongly desirable. Nanoparticle-based quantum emitters, such as color centers in nanodiamonds are compatible with all photonic material platforms. Additionally, nanoparticles allow the quantum emitters to be interfaced with nanoscale plasmonic modes, offering a multi-order speedup in spontaneous emission, and increasing the front-end quantum bitrates. However, the small size and heterogeneity hinder the use of nanoparticles in the manufacturing of quantum devices. A combination of supervised colloidal synthesis, large-scale screening, and deterministic manipulation, driven by neural networks, can yield a scalable quantum device manufacturing process. The team will investigate an in-situ monitored synthesis of low-loss plasmonic shell nanodiamond core structures for enhanced single-photon emission, develop rapid all-optical selection of color centers, and a neural-network driven atomic force microscope-based pick-and-place procedure. Recognizing the expected imminent impact of machine learning in nanotechnology, the team will pursue an educational activity aimed at achieving the participation and training of undergraduate students. A portion of the proposed process will be made available to the undergraduate community for crowdsourced online data generation and neural network training.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.