RESUMO
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.
RESUMO
Supported lipid bilayers have proven effective as model membranes for investigating biophysical processes and in development of sensor and array technologies. The ability to modify lipid bilayers after their formation and in situ could greatly advance membrane technologies, but is difficult via current state-of-the-art technologies. Here we demonstrate a novel method that allows the controlled post-formation processing and modification of complex supported lipid bilayer arrangements, under aqueous conditions. We exploit the destabilization effect of lipopolysaccharide, an amphiphilic biomolecule, interacting with lipid bilayers to generate voids that can be backfilled to introduce desired membrane components. We further demonstrate that when used in combination with a single, traditional soft lithography process, it is possible to generate hierarchically-organized membrane domains and microscale 2-D array patterns of domains. Significantly, this technique can be used to repeatedly modify membranes allowing iterative control over membrane composition. This approach expands our toolkit for functional membrane design, with potential applications for enhanced materials templating, biosensing and investigating lipid-membrane processes.