Nanofluidic computing makes a splash

Neuromorphic computing aims to mimic the energy-efficient information processing of the human brain. To enable brain like processing, a new device architecture that forgoes the rigid computing language of zeros and ones is needed. This brings the question, what if devices used ions moving in fluids to carry and store information?

MSD’s Alex Noy co-authored a perspective article, published in Science, which describes and compares nanofluidic synaptic devices (highlighted in two related research articles published in Science, see Xiong and Robin) that operate at voltage and power consumption close to those of biological systems, providing a glimpse into the possibilities of neuromorphic ionic computing.

The authors express the need for a robust scale-up strategy to build compact, interconnected nanofluidic circuits that can support the complexity required for meaningful computing capabilities, mimic various brain functions, and endure billions of information processing events. While the road to nanofluidic computing is still riddled with formidable challenges, nature gave humans an amazingly powerful and compact computer that uses an amount of energy that would be insufficient to light a desk lamp. Researchers are now tasked with bringing this level of efficiency to the computing devices that power civilization.

[A. Noy, S.B. Darling, Nanofluidic computing makes a splash, Science (2022), doi: 10.1126/science.adf6400.]