Neuromorphic computing, emulating structures and principles based on the human brain, provides an alternative and promising approach for efficient and low-consumption information processing.

Bian, Cao and Zhou examine the recent progress made in neuromorphic computing, highlighting significant improvements in devices, hardware and systems based on 2D materials, such as graphene and black phosphorus. These emerging 2D materials have layered structures and rich electronic properties, making them ideal for creating resistive-switching devices, ferroelectric-related devices, and novel structure transistors.

“The rapid development of artificial intelligence put demands on computing systems to handle large amounts of data,” said co-author Peng Zhou. “Conventional computing hardware based on the von Neumann architecture cannot cope with artificial intelligence tasks, because the processing and memory units are usually physically separated, inducing computing speed limitation and power cost.”

In contrast, neuromorphic computer architecture utilizes 2D materials to mimic biological neural networks through transistors and capacitors, improving computing efficiency and power consumption. While conventional computing performs tasks in a linear sequence, 2D materials in neuromorphic computing can be stacked vertically to form functional blocks that can be further interconnected to create a grid, providing system-level integration.

“Human brains consist of a network of neurons and synapses and can handle multiple complex tasks in efficient parallelism with lower power consumption,” said Zhou. “Thus, neuromorphic computers simulating the working principle and functions of the human brain attracts broad interest.”

While the use of neuromorphic computers based on 2D materials is still in its infancy, developments will greatly aid and complement current techniques, and contribute to satisfying the demands of artificial-intelligence applications.

Source: “Neuromorphic computing: Devices, hardware and system application facilitated by two-dimensional materials,” by Jihong Bian, Zhenyuan Cao, and Peng Zhou, Applied Physics Reviews (2021). The article can be accessed at