Multiple ionic memories in asymmetric nanochannels revealed by mem-spectrometry

Recently discovered nanofluidic memristors, have raised promises for the development of iontronics and neuromorphic computing with ions. Ionic memory effects are related to ion dynamics inside nanochannels, with timescales associated with the manifold physicochemical phenomena occurring at confined interfaces. Here, we explore experimentally the frequency-dependent current–voltage response of model nanochannels—namely glass nanopipettes—to investigate memory effects in ion transport. This characterisation, which we refer to as mem-spectrometry, highlights two characteristic frequencies, associated with short and long timescales of the order of 50 ms and 50 s in the present system. Whereas the former can be associated with ionic diffusion, very long timescales are difficult to explain with conventional transport phenomena. We develop a minimal model accounting for these mem-spectrometry results, pointing to surface charge regulation and ionic adsorption-desorption as possible origins for the long-term memory. Our work demonstrates the relevance of mem-spectrometry to highlight subtle ion transport properties in nanochannels, giving hereby new insights on the mechanisms governing ion transport and current rectification in charged conical nanopores.