Speaker
Description
Nanofluidic memristors have emerged as promising substrates in neuromorphic computing applications due to their programable conductance memory states that are reminiscent of the brain synapses. We explore different options for the modulation of the conductance of track-etched conical nanopores by programming series of rectangular voltage pulses of different characteristics: amplitude, duration, and frequency. The resulting conductance states enables short-term memory and allow the implementation of learning procedures through potentiation (connection strengthening) and depression (connection weakening) effects, which can be further characterized by impedance spectroscopy measurements. Also, the nanofluidic system can exhibit negative differential resistance (NDR) phenomena due to salt precipitation, providing rich nonlinear dynamics in signal processing tasks. The memory features, together with nonlinear dynamics, allow the design of electrochemical networks of nanofluidic memristors that can provide operational procedures to implement both logical responses and reservoir computing algorithms using different voltage signals as inputs.