Nanowire networks are promising memristive architectures for neuromorphic applications due to their connectivity and neurosynaptic-like behaviours. Here, we demonstrate a self-similar scaling of the conductance of networks and the junctions that comprise them. We show this behavior is an emergent property of any junction-dominated network. A particular class of junctions naturally leads to the emergence of conductance plateaus and a “winner-takes-all” conducting path that spans the entire network, and which we show corresponds to the lowest-energy connectivity path. The memory stored in the conductance state is distributed across the network but encoded in specific connectivity pathways, similar to that found in biological systems. These results are expected to have important implications for development of neuromorphic devices based on reservoir computing.

Time-controlled plasma treatment of MoS2 FETs improves carrier transport due to the presence of a two-dimensional oxide phase.

In this work, we introduce a combined experimental and computational approach to describe the conductivity of metallic nanowire networks. Due to their highly disordered nature, these materials are typically described by simplified models in which network junctions control the overall conductivity. Here, we introduce a combined experimental and simulation approach that involves a wire-by-wire junction-byjunction simulation of an actual network. Rather than dealing with computer-generated networks, we use a computational approach that captures the precise spatial distribution of wires from an SEM analysis of a real network. In this way, we fully account for all geometric aspects of the network, i.e. for the properties of the junctions and wire segments. Our model predicts characteristic junction resistances that are smaller than those found by earlier simplified models. The model outputs characteristic values that depend on the detailed connectivity of the network, which can be used to compare the performance of different networks and to predict the optimum performance of any network and its scope for improvement.