Advanced synaptic devices and their applications in biomimetic sensory neural system

An Organic Optoelectronic Synapse with Multilevel Memory Enabled by Gate Modulation

“…5−9 However, mimicking these complex functionalities�which relies on separate imaging, memory, and processing units�presents significant challenges in terms of device integration and energy consumption. 10 Optoelectronic devices, such as two-terminal memristors 11,12 and three-terminal transistors, 13−15 have been extensively studied toward integration of light sensing and memory functions in a single device. Particularly, three-terminal fieldeffect transistors (FETs) 16 are renowned for their capability in facilitating tunable memory behaviors.…”

An electric-field-driven ferroelectric nanodomain structure and its multilevel data storage application

“…9–15 Ferroelectric-based solid-state synapses show promise for achieving a highly efficient biomimetic neural network. 6–20 The ferroelectric nanodomain structure is adjusted to control spike-timing-dependent plasticity or multilevel data storage in ferroelectric field effect transistors, memristors, and tunnel junctions. 6–20 As future biomimetic neural synaptic networks will consist of billions of ferroelectric material-based synapses, a clear understanding of the electric-field-driven ferroelectric nanodomain structure is necessary for their application.…”

“…[8][9][10][11][12] These synapses reconfigure to facilitate learning through variable connection strength. [9][10][11][12][13][14][15] Ferroelectric-based solid-state synapses show promise for achieving a highly efficient biomimetic neural network. [6][7][8][9][10][11][12][13][14][15][16][17][18][19][20] The ferroelectric nanodomain structure is adjusted to control spike-timing-dependent plasticity or multilevel data storage in ferroelectric field effect transistors, memristors, and tunnel junctions.…”

“…[9][10][11][12][13][14][15] Ferroelectric-based solid-state synapses show promise for achieving a highly efficient biomimetic neural network. [6][7][8][9][10][11][12][13][14][15][16][17][18][19][20] The ferroelectric nanodomain structure is adjusted to control spike-timing-dependent plasticity or multilevel data storage in ferroelectric field effect transistors, memristors, and tunnel junctions. [6][7][8][9][10][11][12][13][14][15][16][17][18][19][20] As future biomimetic neural synaptic networks will consist of billions of ferroelectric material-based synapses, a clear understanding of the electric-field-driven ferroelectric nanodomain structure is necessary for their application.…”

Sub-10 nm HfZrO ferroelectric synapse with multiple layers and different ratios for neuromorphic computing

“…With precisely controlling the partial domain switching process, multi-state storage, and synapse emulation are achieved in HZO-based ferroelectric devices [12,13]. There are many efforts to emulate typical synapse behaviours including long-term potentiation (LTP), long-term depression (LTD), spike-timedependent plasticity (STDP), spike-rate-dependent plasticity (SRDP) and so on, by using a designed external electrical stimulate [14].…”