| 초록 |
Artificial synaptic devices are key building blocks for neuromorphic computing systems that can go beyond the constraints of von Neumann architecture. The intercalation of Lithium ion-based two-terminal synaptic device is developed for improved weight updating in hardware-based neuromorphic systems. Herein, all-solid-state battery-inspired artificial synaptic devices are demonstrated. The developed ion synapse consists of the most traditional cathode material, lithium cobalt oxide (LCO) thin film layers and a centrally inserted solid electrolyte lithium titanate (LLTO) thin film. The resistance states of LixCoO2 thin film can be adjusted continuously by sequential input voltage pulses train, and that these resistance states depend on the pulses input conditions. Lithium cations in Lithium cobalt oxide layer can migrate from one side of LCO layer to the other side of LCO layer through LLTO solid electrolyte under programming voltage pulse or bias sweep, resulting in change of total conductance correspond to synaptic weight, hence emulating biological synapse plasticity. Therefore, these LixCoO2-based synaptic devices allow for a combination of possibilities, offering new perspectives of usage in nanoelectronics and bio-inspired neuromorphic circuits. |
