RESUMEN
Component errors limit the scaling of programmable coherent photonic circuits. These errors arise because the standard tunable photonic coupler-the Mach-Zehnder interferometer (MZI)-cannot be perfectly programmed to the cross state. Here, we introduce two modified circuit architectures that overcome this limitation: (1) a 3-splitter MZI mesh for generic errors, and (2) a broadband MZI+Crossing design for correlated errors. Because these designs allow for perfect realization of the cross state, the matrix fidelity no longer degrades with increased mesh size, allowing scaling to arbitrarily large meshes. The proposed architectures support progressive self-configuration, are more compact than previous MZI-doubling schemes, and do not require additional phase shifters. This removes a key limitation to the development of very-large-scale programmable photonic circuits.
RESUMEN
Advanced machine learning models are currently impossible to run on edge devices such as smart sensors and unmanned aerial vehicles owing to constraints on power, processing, and memory. We introduce an approach to machine learning inference based on delocalized analog processing across networks. In this approach, named Netcast, cloud-based "smart transceivers" stream weight data to edge devices, enabling ultraefficient photonic inference. We demonstrate image recognition at ultralow optical energy of 40 attojoules per multiply (<1 photon per multiply) at 98.8% (93%) classification accuracy. We reproduce this performance in a Boston-area field trial over 86 kilometers of deployed optical fiber, wavelength multiplexed over 3 terahertz of optical bandwidth. Netcast allows milliwatt-class edge devices with minimal memory and processing to compute at teraFLOPS rates reserved for high-power (>100 watts) cloud computers.
RESUMEN
Spintronic devices usually rely on long spin relaxation times and/or long spin relaxation lengths for optimum performance. Therefore, the ability to modulate these quantities with an external agent offers unique possibilities. The dominant spin relaxation mechanism in most technologically important semiconductors is the D'yakonov-Perel' (DP) mechanism which may vanish if the spin carriers (electrons) are confined to a single conduction subband in a quantum wire. Here, we report modulating the DP spin relaxation rate (and hence the spin relaxation length) in self assembled 50 nm diameter InSb nanowires with infrared (IR) light at room temperature. In the dark, almost all the electrons in the nanowires are in the lowest conduction subband, resulting in near-complete absence of DP relaxation. This allows observation of spin-sensitive effects in the magnetoresistance. Under IR illumination, higher subbands get populated and the DP spin relaxation mechanism is revived, leading to a three-fold decrease in the spin relaxation length. Consequently, the spin sensitive effects disappear under illumination. This phenomenon may have applications in spintronic room-temperature IR photodetection.
RESUMEN
A longstanding goal of spintronics is to inject, then coherently transport, and finally detect electron spins in a semiconductor nanowire in which a single quantized subband is occupied by the electrons at room temperature. Here, the achieving of this goal in electrochemically self-assembled 50-nm diameter InSb nanowires is reported and substantiated by demonstrating both the spin-valve effect and the Hanle effect. Observing both effects in the same sample allows one to estimate the electron mobility and the spin relaxation time in the nanowires. It is found that despite four orders of magnitude degradation in the mobility compared to bulk or quantum wells and a resulting four orders of magnitude increase in the Elliott-Yafet spin relaxation rate, the spin relaxation time in the nanowires is still about an order of magnitude longer than what has been reported in bulk and quantum wells. This is caused by the elimination or suppression of the D'yakonov-Perel' spin relaxation through single subband occupancy. These experiments shed light on the nature of spin transport in a true quantum wire and raise hopes for the realization of a room-temperature Datta-Das spin transistor, where single subband occupancy is critical for optimum performance.
RESUMEN
We show that nanoporous anodic alumina films, with pore diameters in the range 10-80 nm, can be transformed from being very hydrophilic (or super-hydrophilic) to very hydrophobic (or super-hydrophobic) by coating the surface with a thin (2-3 nm) layer of a hydrophobic polymer. This dramatic transformation happens as a result of the interplay between surface morphology and surface chemistry. The coated surfaces exhibit 'sticky' hydrophobicity as a result of ingress of water into the pores by capillary action. The wetting parameters (contact angle and contact angle hysteresis) exhibit qualitatively different dependences on pore diameters in coated and uncoated films, which are explained by invoking appropriate models for wetting.
