A Van Der Waals Photo‐Ferroelectric Synapse

For hardware artificial intelligence, the central task is to design and develop synapses with needed characteristics. Here, experimental demonstration of a van der Waals (vdW) photo-ferroelectric synapse are reported. In synapse, synaptic memory extracted by reading photocurrent, written or edited electrical pulses. The semiconducting vdW organic-inorganic halide perovskite ((R)-(–)-1-cyclohexylethylammonium)PbI3 (R-CYHEAPbI3) serves as model channel. organic layer provides ferroelectric dipole PbI6 octahedron responsible for photon absorption charge transport. R-CYHEAPbI3 show writing/reading dynamics >200 states, close 103 on/off ratio, reasonable endurance retention With experimentally measured weight (parallel through photovoltaic effect writing pulses) synapses, feasibility using crossbar circuit implement classic training inference hand-written digits presented. An image recognition accuracy up 90% obtained. such opens window in advanced devices intelligence.