Large-Area Polymorphic In₂Se₃ Ferroelectric Transistor Array for Stable Nonvolatile Storage and High-Precision...

Area of Science:

• Materials Science
• Condensed Matter Physics
• Electrical Engineering

Background:

• Emerging 2D materials are crucial for overcoming the von Neumann bottleneck in computing.
• Indium Selenide (In2Se3) exhibits unique properties, including room-temperature ferroelectricity and high carrier mobility.
• Chemical Vapor Deposition (CVD) enables the synthesis of large-area 2D materials for device fabrication.

Purpose of the Study:

• To investigate the performance of Indium Selenide (In2Se3) thin film ferroelectric field-effect transistor (FeFET) arrays.
• To evaluate the potential of these 2D FeFETs for in-memory and neuromorphic computing applications.
• To demonstrate the stability and non-volatile characteristics of the In2Se3 FeFET array.

Main Methods:

• Fabrication of In2Se3 thin film FeFET arrays using Chemical Vapor Deposition (CVD).
• Characterization of device performance, including field-effect mobility (µFE) and on-off ratio.
• Testing the stability and non-volatile memory characteristics of the FeFET array.
• Application of the In2Se3 FeFET array in machine learning tasks for neuromorphic computing.

Main Results:

• The In2Se3 FeFET array achieved a high field-effect mobility (µFE) of 151.7 cm² V⁻¹ s⁻¹.
• An impressive on-off ratio of up to 10⁶ was recorded.
• Stable, non-volatile characteristics were observed, with a resistance state window > 10² maintained for over 1800 seconds.
• Recognition accuracy exceeding 90% was consistently achieved in machine learning tasks.

Conclusions:

• In2Se3 thin film FeFET arrays are a promising solution for next-generation in-memory computing.
• The demonstrated high performance and stability highlight their potential for neuromorphic computing.
• CVD synthesis facilitates the development of large-area 2D FeFET arrays for practical applications.