The Centre for Nanosciences and Nanotechnologies (Centre de Nanosciences et de Nanotechnologies de l'université Paris-Saclay) or C2N, is a nanotechnology laboratory created as a collaboration between the University of Paris-Saclay and the French National Centre for Scientific Research (CNRS.)CNRS and the university announced this collaboration in 2013, with the goal of uniting two existing laboratories of Ile-de-France: the Institute for Fundamental Electronics (Institut d'Electronique Fondamentale, IEF) and the Laboratory for Photonics Nanostructures (Laboratoire de Photonique et de Nanostructures, LPN)Facility construction began in April, 2015, the first stone was laid on June 28, 2016, and the C2N facility began operations in September, 2017. It is located on the Paris-Saclay campus in Palaiseau, 20 miles south of Paris, France.According to the European Union's MIR-Bose project, "The Centre for Nanoscience and Nanotechnology (C2N) is one of the largest laboratories of University Paris-Sud with 283 members including 83 permanent researchers, 160 PhD students and post-doctoral researchers and 40 technical and administrative staff members.
The deployment of AI on edge computing devices faces significant challenges related to energy consumption and functionality. These devices could greatly benefit from brain-inspired learning mechanisms, allowing for real-time adaptation while using low-power. In-memory computing with nanoscale resistive memories may play a crucial role in enabling the execution of AI workloads on these edge devices. In this study, we introduce voltage-dependent synaptic plasticity (VDSP) as an efficient approach for unsupervised and local learning in memristive synapses based on Hebbian principles. This method enables online learning without requiring complex pulse-shaping circuits typically necessary for spike-timing-dependent plasticity (STDP). We show how VDSP can be advantageously adapted to three types of memristive devices (TiO2, HfO2-based metal-oxide filamentary synapses, and HfZrO4-based ferroelectric tunnel junctions (FTJ)) with disctinctive switching characteristics. System-level simulations of spiking neural networks incorporating these devices were conducted to validate unsupervised learning on MNIST-based pattern recognition tasks, achieving state-of-the-art performance. The results demonstrated over 83
Efficient generation of radiation in the mid- and far-infrared relies primarily on lasers and coherent nonlinear optical phenomena driven by lasers. This wavelength range lacks practical and efficient luminescent devices, particularly, because the spontaneous emission rate becomes much longer than the nonradiative energy relaxation processes and therefore emitters have to count on stimulated emission produced by linear or nonlinear optical gain. However, spontaneous emission is not a fundamental property of the emitter. By engineering metamaterials composed of arrays of nanoemitters into microcavities coupled to patch antennas, we have demonstrated midinfrared electroluminescent devices emitting a collimated beam with excellent spatial properties and a factor of 100 increase in the collected power, compared to standard devices. Our results illustrate that by reshaping the photonic environment around emitting dipoles, as in the Purcell effect, it is possible to enhance the spontaneous emission and conceive efficient optoelectronic light-emitting devices that operate close to the thermodynamical equilibrium as light-emitting diodes in the visible range.
Light-matter interfaces are pivotal for quantum computation and communication. While typically analyzed using single-mode or open-quantum-system approximations, these models often neglect multi-mode field states and light-matter entanglement, hindering exact protocol modeling. Here, we solve the full Hamiltonian dynamics of a solid-state spin-photon interface for three key protocols: the generation of photon-number superpositions, a controlled photon-photon gate, and the production of photonic cluster states. By deriving exact fidelities, we identify fundamental performance limits. Our results reveal that while realistic imperfections severely limit photon-photon gates, they only slightly affect linear photonic clusters and are nearly harmless for photon-number state superpositions.
