Researchers at The University of Texas at Austin and Peking University claim to have developed the world's thinnest memory storage device capable of offering dense capacity. The device is designed to efficiently and permanently store information while switching in nanoseconds, making the technology a strong candidate for use with several advanced next-generation memory and computing applications.
Formed with 2D nanomaterials, the atomristor is essentially a next-generation memristor, a new type of storage technology featuring lower memory scalability.
Until now, memory storage and transistors have existed as separate components. "However, atomristor affords 3D integration of memory with transistors on the same chip for advanced computing systems, beneficial for mobile device and big data applications," said Deji Akinwande, an associate professor in The University of Texas at Austin's Cockrell School of Engineering's Department of Electrical and Computer Engineering.
The entire memory cell is a 1.5 nanometer (nm)-thick sandwich. Semiconducting molybdenum sulfide atomic sheets serve as the active layer, while metallic atomic sheets composed of graphene function as electrodes. The design of the memory storage device makes it possible to densely pack atomristors layer by layer in a plane. The technique offers a significant space advantage over bulkier conventional flash memory. The trimmed-down device also facilitates faster and more efficient electric current flow.
According to Akinwande, memristor technology's nonvolatile resistive switching phenomenon has been observed to overcome a fundamental limitation in ultrascaled bulk materials. "These devices can be collectively labeled atomristor, in essence, [a] memristor effect in a single atomic sheet," Akinwande said. By replacing metal electrodes with graphene, the entire memory cell can be scaled below 2 nm, making it the current thinnest memory storage device. "This characteristic enables atomristors to densely pack layer by layer in a plane and thus occupies much less space than the conventional memory," he added.
Funding for The University of Texas at Austin team's work was provided by the National Science Foundation and the Presidential Early Career Award for Scientists and Engineers, which Akinwande was awarded in 2015.
There could be a bright future for atomristor technology, particularly in fields such as analytics and AI. Given their size, capacity and integration flexibility, atomristors can be packed together to make advanced 3D chips with a memory architecture featuring 3D connections similar to those found in the human brain.
Atomristors are also able to integrate with conventional silicon chips "in a facile way via back-end-of-line integration," Akinwande noted. "Considering size, capacity and integration flexibility, atomristor can advance emerging applications, such as 3D crossbar networks for neuromorphic memory computing," he added.
Another unique application for atomristor technology is as a nonvolatile, low-power radio frequency switch. "This is a rapidly emerging field because of the massive growth in wireless technologies and the need for very low-power switches," Akinwande explained.
In smartphones, tablets and other mobile business and consumer technologies, radio frequency switches direct incoming signals from the antenna to one of several wireless communication bands, an inefficient process that wastes precious battery life.
"Atomristors offer an unprecedented advancement for high-frequency systems owing to its low-voltage operation, small form factor, fast switching speed and low-temperature integration compatible with [silicon] or flexible substrates," Akinwande said. Better yet, the memory storage device consumes no static power, which can lead to longer battery life for mobile devices.
While Akinwande is upbeat about atomristors' long-term potential, he acknowledged that the ultrathin memory storage device is still several years away from becoming a mainstream technology. Practical matters, such as improving storage retention, endurance, uniformity and device yield, must be addressed. "As these practical parameters are optimized, a feasible device technology can be expected in a 10-year timeframe," Akinwande said.
From a scientific view, Akinwande and his co-researchers would like to gain more insight into the underlying mechanisms of nonvolatile resistive switching in single-layer atomic sheets. "It inspires a new field of research in defects, ion transport and energetics at the sharp interfaces between atomically thin sheets and conducting electrodes," Akinwande said." "Now, we are working with theorists and scientists and trying to get the detailed understanding of [its] physics."