Schematic illustration of an edge computing system based on monolithic 3D-integrated, 2D material-based electronics. The system stacks different functional layers, including AI computing layers, signal-processing layers and a sensory layer, and integrates them into an AI processor. Image: Sang-Hoon Bae, McKelvey School of Engineering, Washington University in St. Louis.Multifunctional computer chips have evolved to do more with integrated sensors, processors, memory and other specialized components. However, as chips have expanded, the time required to move information between functional components has also grown.
“Think of it like building a house,” said Sang-Hoon Bae, an assistant professor of mechanical engineering and materials science in the McKelvey School of Engineering at Washington University in St. Louis. “You build out laterally and up vertically to get more function, more room to do more specialized activities, but then you have to spend more time moving or communicating between rooms.”
To address this challenge, Bae and an international team of collaborators from the US, Korea and France have demonstrated monolithic three-dimensional (3D) integration of layered two-dimensional (2D) materials into novel processing hardware for artificial intelligence (AI) computing. They envision that their new approach will not only provide a material-level solution for fully integrating many functions into a single, small electronic chip, but also pave the way for advanced AI computing. They report their work in a paper in Nature Materials.
The team’s monolithic 3D-integrated chip offers several advantages over existing laterally integrated computer chips. It contains six atomically thin 2D layers, made from 2D materials such as tungsten diselenide and hexagonal boron nitride.
Each layer has its own function and achieves significantly reduced processing time, power consumption, latency and footprint. This is accomplished by tightly packing the processing layers to ensure dense interlayer connectivity. As a result, the hardware offers unprecedented efficiency and performance in AI computing tasks.
This discovery offers a novel solution to integrating electronics and also opens the door to a new era of multifunctional computing hardware. With ultimate parallelism at its core, this technology could dramatically expand the capabilities of AI systems, allowing them to handle complex tasks with lightning speed and exceptional accuracy.
“Monolithic 3D integration has the potential to reshape the entire electronics and computing industry by enabling the development of more compact, powerful and energy-efficient devices,” Bae said. “Atomically thin 2D materials are ideal for this, and my collaborators and I will continue improving this material until we can ultimately integrate all functional layers on a single chip.”
Bae said these devices also are more flexible and functional, making them suitable for more applications.
“From autonomous vehicles to medical diagnostics and data centers, the applications of this monolithic 3D integration technology are potentially boundless. For example, in-sensor computing combines sensor and computer functions in one device, instead of a sensor obtaining information then transferring the data to a computer. That lets us obtain a signal and directly compute data, resulting in faster processing, less energy consumption and enhanced security because data isn’t being transferred.”
This story is adapted from material from Washington University in St. Louis, with editorial changes made by Materials Today. The views expressed in this article do not necessarily represent those of Elsevier. Link to original source.