Description
Living organisms use ions flowing and accumulating within (sub)nanoscale aqueous protein channels to process information. This process occurs at energy costs orders of magnitude lower than those of man-made computers. The contrast is striking: Lee Sedol defeated AlphaGo in one game while consuming only ~20 W, about 50,000 times less power than the compute infrastructure supporting AlphaGo. This observation highlights the remarkable capacities of biological brains and motivates the development of brain-inspired ionic computing. This nascent field aims to implement data storage and processing using artificial nanoscale fluidic channels.
In this presentation I will display experimental results illustrating the birth of this field. I will present Nanofluidic memory devices: 2D channels, highly asymmetric channels and protein-based channels. Through direct imaging approaches and analytical modeling, we will elucidate electromechanical mechanisms at the root of ionic memory. Finally, we will see how these devices can be integrated into logic circuit, opening the path for brain-inspired liquid hardware.
References
- Robin, P. et al. Long-term memory and synapse-like dynamics in two-dimensional nanofluidic channels. Science 379, 161–167 (2023).
- Saurav, K. V. et al. Direct imaging reveals electromechanical ionic memory in 2D nanochannels. Preprint at https://doi.org/10.48550/arXiv.2509.11637 (2025).
- Emmerich, T. et al. Nanofluidic logic with mechano–ionic memristive switches. Nat. Electron. 7, 271–278 (2024).
- Mayer, S. F. et al. Lumen charge governs gated ion transport in β-barrel nanopores. Nat. Nanotechnol. 1–9 (2025) doi:10.1038/s41565-025-02052-6.