Energy-Efficient Neuromorphic Architectures for Nuclear Radiation Detection Applications.

A comprehensive analysis and simulation of two memristor-based neuromorphic architectures for nuclear radiation detection is presented. Both scalable architectures retrofit a locally competitive algorithm to solve overcomplete sparse approximation problems by harnessing memristor crossbar execution of vector-matrix multiplications. The proposed systems demonstrate excellent accuracy and throughput while consuming minimal energy for radionuclide detection. To ensure that the simulation results of our proposed hardware are realistic, the memristor parameters are chosen from our own fabricated memristor devices. Based on these results, we conclude that memristor-based computing is the preeminent technology for a radiation detection platform.

Array of Graphene Variable Capacitors on 100 mm Silicon Wafers for Vibration-Based Applications.

Highly flexible, electrically conductive freestanding graphene membranes hold great promise for vibration-based applications. This study focuses on their integration into mainstream semiconductor manufacturing methods. We designed a two-mask lithography process that creates an array of freestanding graphene-based variable capacitors on 100 mm silicon wafers. The first mask forms long trenches terminated by square wells featuring cone-shaped tips at their centers. The second mask fabricates metal traces from each tip to its contact pad along the trench and a second contact pad opposite the square well. A graphene membrane is then suspended over the square well to form a variable capacitor. The same capacitor structures were also built on 5 mm by 5 mm bare dies containing an integrated circuit underneath. We used atomic force microscopy, optical microscopy, and capacitance measurements in time to characterize the samples.

Terahertz Detection with Perfectly-Absorbing Photoconductive Metasurface.

Terahertz (THz) photoconductive devices are used for generation, detection, and modulation of THz waves, and they rely on the ability to switch electrical conductivity on a subpicosecond time scale using optical pulses. However, fast and efficient conductivity switching with high contrast has been a challenge, because the majority of photoexcited charge carriers in the switch do not contribute to the photocurrent due to fast recombination. Here, we improve efficiency of electrical conductivity switching using a network of electrically connected nanoscale GaAs resonators, which form a perfectly absorbing photoconductive metasurface. We achieve perfect absorption without incorporating metallic elements, by breaking the symmetry of cubic Mie resonators. As a result, the metasurface can be switched between conductive and resistive states with extremely high contrast using an unprecedentedly low level of optical excitation. We integrate this metasurface with a THz antenna to produce an efficient photoconductive THz detector. The perfectly absorbing photoconductive metasurface opens paths for developing a wide range of efficient optoelectronic devices, where required optical and electronic properties are achieved through nanostructuring the resonator network.