Researchers have employed X-ray microscopy to identify a ferroelectric material that adjusts its reactions to controlled ultrafast external stimuli, like light pulses. This discovery holds potential for use in energy-efficient microelectronics.
“Modern supercomputers and data centers consume several megawatts of power,” stated Haidan Wen, a physicist at the U.S. Department of Energy (DOE) Argonne National Laboratory. “Our goal is to find materials that contribute to more energy-efficient microelectronics. A viable option is a ferroelectric material suitable for artificial neural networks, which can be integrated into energy-efficient electronics.”
Ferroelectric materials are utilized in various information processing devices, including computer memory, transistors, sensors, and actuators. Researchers at Argonne have observed unexpected adaptive behaviors in a ferroelectric material that can progressively change to achieve a specific outcome, based on the number of photons from light pulses that hit the material. They collaborated with scientists from Rice University, Pennsylvania State University, and DOE’s Lawrence Berkeley National Laboratory.
The ferroelectric material studied contains interconnected regions or domains, which are as distinct as oil droplets in water. These nanometer-sized domains can rearrange in response to light pulses, offering potential for efficient information transmission in microelectronics.
The ferroelectric sample is designed as a layered structure, alternating between lead and strontium titanate. With assistance from Rice University collaborators, this seven-layer structure is an astonishing 1,000 times thinner than a standard sheet of paper. Previously, the team had illuminated a single, powerful light pulse onto a sample, resulting in uniformly ordered nanostructures.
“This time, we applied numerous weaker light pulses, each lasting a quadrillionth of a second,” explained Wen. “As a result, we produced a variety of domain structures instead of just one, depending on the light dosage.”
The researchers utilized the Nanoprobe (beamline 26-ID) at the Center for Nanoscale Materials and the Advanced Photon Source (APS) to visualize the nanoscale reactions. Both belong to the DOE Office of Science user facilities at Argonne. The Nanoprobe employed a focused X-ray beam measuring tens of nanometers to scan the sample as it received a series of ultrafast light pulses.
The captured images demonstrated how nanodomains were formed, erased, and reorganized in response to the light pulses. The dimensions and boundaries of these domains varied from 10 nanometers — approximately 10,000 times smaller than a human hair — to 10 micrometers, similar to the size of a cloud droplet. The resulting arrangement depended on the number of light pulses that stimulated the sample.
“By integrating an ultrafast laser with the Nanoprobe beamline, we can initiate and manipulate changes in the networked nanodomains using light pulses while using minimal energy,” noted Martin Holt, an expert in X-ray and electron microscopy and a group leader.
The sample starts with a spiderweb-like configuration of nanodomains. The impact of the light pulses disrupts this web, leading to the formation of entirely new arrangements that adapt toward a specific goal, much like an adaptive network.
“We’ve uncovered completely new arrangements of these nanodomains,” stated Stephan Hruszkewycz, an Argonne physicist and group leader. “This opens the door to numerous new discoveries. In the future, we plan to explore various light stimulation methods and examine even more unknown nanodomains and networks.” The newly upgraded APS significantly enhances the ability to visualize nanoscale changes over time, offering X-ray beams that are up to 500 times brighter.
Through this significant discovery of time-dependent changes in the networked nanodomains, researchers are progressing towards the creation of adaptive networks for information storage and processing, paving the way for more energy-efficient computing systems.
This research is documented in a paper published in Advanced Materials. Alongside Wen, Holt, and Hruszkewycz, contributors include Marc Zajac, Tao Zhou, Tiannan Yang, Sujit Das, Yue Cao, Burak Guzelturk, Vladimir Stoica, Mathew Cherukara, John Freeland, Venkatraman Gopalan, Ramamoorthy Ramesh, Lane Martin, and Long-Qing Chen.
Funding for this research was provided by the DOE Office of Basic Energy Sciences.
