Hybridization of Delta FAPbI₃ and Atomic-Layer-Deposited SnO₂ Enhances Artificial Synaptic Behavior

Resistive switching characteristics can be extended to artificial synapse for neuromorphic computing. Organic-inorganic hybrid halide perovskites are emerging materials for nonvolatile resistive switching memory devices. 1-dimensional hexagonal FAPbI₃ (δ-FAPbI₃) was found to demonstrate resistive switching behavior. However, δ-FAPbI₃ cannot be used for artificial synapse due to the absence of analog switching characteristics. Here, we report a bilayer memristor hybridizing δ-FAPbI₃ with atomic-layer-deposited (ALD) SnO₂ for artificial synapse. The activation energy required for vacancy migration was increased from 0.42 eV for δ-FAPbI₃ to 0.53 eV after introduction of ALD-SnO₂, associated with the Schottky barrier at the heterojunction structure, which led to an analog switching behavior at applied voltage. Consequently, this significantly reduced the current level in the high resistance state by restricting the orientation of migration channels. Artificial synaptic characteristics were confirmed by linear potentiation/depression, long-term potentiation (LTP), long-term depression (LTD) and spike-timing-dependent plasticity. The non-linearity in LTP and LTD was considerably reduced from 12.26 to 0.60 and from -8.79 to -3.47, respectively. Moreover, the δ-FAPbI₃/ALD-SnO₂ bilayer system achieved a recognition rate of up to 94.04% based on the modified National Institute of Standards and Technology database using deep neural networks. This study highlights the critical role of the interface layers, such as ALD-SnO₂, in controlling ion migration and modifying resistive switching behavior.