Recent Advances in In-Memory Computing: Exploring Memristor and Memtransistor Arrays with 2D Materials

Area of Science:

• Materials Science
• Computer Engineering
• Nanotechnology

Background:

• Conventional computing faces significant latency and energy challenges due to the separation of memory and processing units.
• In-memory computing architectures perform operations within memory arrays, offering a potential solution to these limitations.
• Memristive devices, particularly those based on two-dimensional (2D) materials, are key components for high-density, low-power in-memory computing and neuromorphic applications.

Purpose of the Study:

• To review the current state-of-the-art research on 2D material-based memristive arrays for in-memory computing.
• To cover critical aspects including material selection, device performance, array architectures, and applications.
• To identify challenges and propose solutions for advancing 2D material-based memristive arrays from device to system-level implementation.

Main Methods:

• Comprehensive literature review of recent advancements in 2D material-based memristive devices and arrays.
• Analysis of device performance metrics, including switching characteristics, endurance, and retention.
• Examination of various array structures and their suitability for in-memory computing and neuromorphic applications.

Main Results:

• 2D materials exhibit unique properties like layered structures and heterojunction formation, enabling high-performance memristive devices.
• 2D material-based memristors and memtransistors show promise for dense, energy-efficient in-memory computing arrays.
• Significant progress has been made in device fabrication and characterization, with ongoing efforts in array-level integration.

Conclusions:

• 2D material-based memristive arrays represent a critical technology for realizing next-generation in-memory computing and neuromorphic systems.
• Bridging the gap between single-device performance and array/system-level implementation is essential for practical applications.
• Continued research in materials, device design, and integration strategies will accelerate the development of these advanced computing paradigms.