Capturing carbon and mimicking neurons with water, salt and nanochannels

Abstract

Ionic transport through nanometre scale channels is of importance for a wide array of different applications. In reflection of this diverse range of nanochannel applications, we research nanofluidic systems in the context of two different topics, that of brain-inspired iontronic circuitry and that of capturing CO2.

Electrolyte-filled conical channels are memristors, a device with a history-dependent electrical resistance, i.e. a resistor with “a memory”. Memristors are the fundamental building blocks of brain-inspired computer circuits, which promise excellent energy efficiency compared to conventional computers. With a theoretical model originating from the Poisson-Nernst-Planck-Stokes equations we can attribute the memristive effects of conical channels to dynamic salt accumulation or depletion in the channel due to a time-dependent electric potential drop over the channel. This model predicts time-dependent currents which match well with finite element simulations. Using this theoretical model, we design iontronic circuits containing these conical channels, which exhibit neuromorphic behaviour such as all-or-none action potentials and spike trains. Moreover, by leveraging the mathematical framework behind memristors, we can make general statements about a wide class of volatile memristors, which we validate against simulations of nanochannel memristors.

The study of nanochannels is also applicable in the investigation of the CO2 capture method Supercapacitive Swing Adsorption (SSA). This is a relatively new and poorly understood method of capturing CO2, which relies on cheap non-toxic materials and promises excellent energy efficiency. SSA relies on charging and discharging porous electrodes, soaked in an electrolyte. By numerically solving the General Modified Poisson-Nernst-Planck equations we show that the local increase or decrease in CO2 derived ions in the electrodes can account for observations made in experimental work.