RESUMEN
Vertical two-terminal synaptic devices based on resistive switching have shown great potential for emulating biological signal processing and implementing artificial intelligence learning circuitries. To mimic heterosynaptic behaviors in vertical two-terminal synaptic devices, an additional terminal is required for neuromodulator activity. However, adding an extra terminal, such as a gate of the field-effect transistor, may lead to low scalability. In this study, a vertical two-terminal Pt/bilayer Sr1.8Ag0.2Nb3O10 (SANO) nanosheet/Nb:SrTiO3 (Nb:STO) device emulates heterosynaptic plasticity by controlling the number of trap sites in the SANO nanosheet via modulation of the tunneling current. Similar to biological neuromodulation, we modulated the synaptic plasticity, pulsed pair facilitation, and cutoff frequency of a simple two-terminal device. Therefore, our synaptic device can add high-level learning such as associative learning to a neuromorphic system with a simple cross-bar array structure.
RESUMEN
In the era of "big data," the cognitive system of the human brain is being mimicked through hardware implementation of highly accurate neuromorphic computing by progressive weight update in synaptic electronics. Low-energy synaptic operation requires both low reading current and short operation time to be applicable to large-scale neuromorphic computing systems. In this study, an energy-efficient synaptic device is implemented comprising a Ni/Pb(Zr0.52 Ti0.48 )O3 (PZT)/0.5 wt.% Nb-doped SrTiO3 (Nb:STO) heterojunction with a low reading current of 10 nA and short operation time of 20-100 ns. Ultralow femtojoule operation below 9 fJ at a synaptic event, which is comparable to the energy required for synaptic events in the human brain (10 fJ), is achieved by adjusting the Schottky barrier between the top electrode and ferroelectric film. Moreover, progressive domain switching in ferroelectric PZT successfully induces both low nonlinearity/asymmetry and good stability of the weight update. The synaptic device developed here can facilitate the development of large-scale neuromorphic arrays for artificial neural networks with low energy consumption and high accuracy.
