Modeling of molecular dynamics of filaments and mesoscopic conductance in a neuromorphic ensemble of nanoparticles

Recently, self-organizing networks of nanoparticles and nanowires have shown promise as systems for neuromorphic reservoir computing. The advantages of such systems are high energy efficiency, absence of necessity for connections between elements within networks, scalability, availability of synthesis technologies, etc. In this work, we calculate the conductance of tunneling contacts between nanoparticles and study the molecular dynamics of the destruction of filamentous compounds. Experimental studies of a percolation memristive network of silver nanoparticles are presented. Using cyclic voltammetry, switching between the capacitive and memristive states of the network is demonstrated and its features are considered.