Single- and double-gate synaptic transistor with TaO _x gate insulator and IGZO channel layer

Area of Science:

• Materials Science
• Neuroscience Engineering
• Electronics

Background:

• Artificial synaptic devices are crucial for developing neuromorphic computing systems.
• Existing synaptic devices face challenges in achieving high performance and analog weight modulation.

Purpose of the Study:

• To demonstrate single- and double-gate synaptic operations in a novel thin-film transistor (TFT).
• To investigate the potential of TaO_x/IGZO TFT for artificial synaptic applications.

Main Methods:

• Fabrication of a TFT with a double-gate stack: Al-top-gate/SiO_x/TaO_x/n-IGZO on a SiO₂/n⁺-Si-bottom-gate substrate.
• Application of repeated gate and drain voltages to modulate drain current, mimicking synaptic weight.
• Analysis of drain current modulation characteristics, including analog behavior, voltage-polarity dependence, and dynamic range.

Main Results:

• The synaptic TFT demonstrated tunable drain current, mimicking biological synaptic weight modulation with a dynamic range of ~10⁴.
• Analog and reversible drain current modulation was achieved, attributed to changes in IGZO channel mobility, gate insulator capacitance, and threshold voltage.
• Emulation of synaptic potentiation, depression, paired-pulse facilitation, and memory transition behaviors through precise voltage pulse control.
• Successful realization of synaptic operations using a double-gate configuration for independent channel conductance tuning and sensing.

Conclusions:

• The TaO_x/IGZO TFT exhibits promising characteristics for artificial synaptic devices.
• Single- and double-gate operations offer versatile control over synaptic weight modulation.
• The device's performance validates its potential for integration into next-generation neuromorphic computing architectures.