Implementation of an electrically modifiable artificial synapse based on ferroelectric field-effect transistors using...

Area of Science:

• Neuromorphic Engineering
• Materials Science
• Solid-State Electronics

Background:

• Ferroelectric field-effect transistors (FeFETs) are promising for neuromorphic computing.
• Emulating synaptic behavior in artificial devices is crucial for brain-inspired computing.

Purpose of the Study:

• To emulate human brain-like synaptic behaviors using metal-ferroelectric-metal-insulator-semiconductor (MFMIS) gate stacks.
• To investigate the synaptic plasticity modulation in FeFETs based on ferroelectric thin films.

Main Methods:

• Fabrication of MFMIS gate stacks using Al-doped HfO2 (Al:HfO2) ferroelectric thin films.
• Characterization of synaptic plasticity through pulse modulation of ferroelectric polarization.
• Demonstration of biological synaptic functions including EPSC, PPF, and STDP.

Main Results:

• FeFETs successfully emulated synaptic behaviors at low operation voltages.
• Synaptic plasticity was modulated by partial polarization switching of Al:HfO2 films.
• A 3x3 array demonstrated electrically modifiable weighted-sum operations.

Conclusions:

• Al:HfO2 based MFMIS-FETs can effectively emulate biological synaptic functions.
• The devices offer a pathway for low-voltage, high-performance neuromorphic computing hardware.
• Integrated synapse arrays show potential for complex computational tasks.