In biological neural networks, chemical communication follows the reversible integrate-and-fire (I&F) dynamics model, enabling efficient, anti-interference signal transport. However, existing artificial neurons fail to follow the I&F model in chemical communication, causing irreversible potential accumulation and neural system dysfunction. Herein, we develop a supercapacitively gated artificial neuron that mimics the reversible I&F dynamics model. Upon upstream neurotransmitters, an electrochemical reaction occurs on a graphene nanowall (GNW) gate electrode of artificial neurons. Charging and discharging the supercapacitive GNWs mimic membrane potential accumulation and recovery, realizing highly efficient chemical communication upon use of acetylcholine down to 2 × 10-10 M. By combining artificial chemical synapses with axon-hillock circuits, the output of neural spikes is realized. With the same neurotransmitter and I&F dynamics, the artificial neuron establishes chemical communication with other artificial neurons and living cells, holding promise as a basic unit to construct a neural network with compatibility to organisms for artificial intelligence and deep human-machine fusion.
Keywords: artificial neurons; chemical neural communication; field-effect transistor; neurotransmitter; supercapacitor.