Thin Films “Dance” with Substrates That Are No Longer Inert
New research shows that, under voltage, substrates are in movement with thin film
Published Date
Article Content
Key Takeaways
- For decades it was believed that the substrates in thin-film devices were an inert material that did not respond to electrical activity on the thin film.
- Using a new tool called dark-field x-ray microscopy, researchers discovered that substrates respond to thin film stimulation in a back-and-forth movement of pushing and pulling.
- This discovery marks a paradigm shift in how materials scientists think about and develop quantum materials.
Many of today’s electronic devices — from the semiconductors in your cell phone to the photovoltaic cells in your solar panels — are built on thin-film substrates. The thin film is an electronically conductive material while the substrate is an inert material. Or is it?
Physicists and materials scientists have long assumed substrates do not react to electrical stimulus, but new research from the University of California San Diego and a team of collaborators has shown that substrates are not inert after all. The discovery has the potential to help engineers build the dense, three-dimensional, brain-inspired computer chips needed for more energy-efficient computing. This work appears in Science.
The research began four years ago in UC San Diego Associate Professor of Physics Alex Frañó’s lab. Frañó is a principal investigator and assistant director at the Quantum Materials for Energy-Efficient Neuromorphic Computing (Q-MEEN-C), one of the U.S. Department of Energy’s Energy Frontier Research Centers. One of the goals of Q-MEEN-C is to develop quantum materials that can be used in neuromorphic, or “brain-like” computing.
In a human brain, a vast network of neurons works across different regions, enabling every thought and action we take. Yet thoughts that are quick and simple for humans — recognizing faces, differentiating a cat from a dog or solving a captcha — can be devilishly hard and energy-intensive for computers. Quantum materials may provide a solution.
It was this quest to develop new materials that led Frañó’s lab to study vanadium dioxide thin-film devices. When a voltage is applied to the thin film, an electric filament forms — something similar happens when the Earth and atmosphere create a voltage strong enough to result in lightning. The filament in the thin film drives the device's electrical spiking, much like the signals in our neurons.
This reaction has been studied countless times in labs around the world, but Frañó’s lab was implementing a new technique, developed by graduate student Elliot Kisiel, called dark-field X-ray microscopy. Kisiel, who spent most of his graduate studies at Argonne National Laboratory and is now a Mayer Postdoctoral Fellow there, developed the tool to combine the best of two worlds: electron microscopy and X-ray diffraction.
“Dark-field X-ray microscopy gives us the ability to look at an entire device in one image, which not only greatly improves how fast we can perform these measurements, but also allows us to study the areas surrounding the devices with high fidelity,” he stated.
Since the lenses for X-rays are often thick and, therefore, absorb most of the X-rays that pass through them, the team needed to look at something that provided more signal, which is why they decided to look at the substrate during some of their preliminary testing with this new technique.
What they saw was completely unexpected: not only was there a change in the thin film, but there was also a change in the substrate, showing for the first time that the two parts were sharing energy or “coupled.” This discovery went against everything scientists had assumed about substrates for decades.
New-Age Materials for a New Age of Computing
Thin films are so called because they are vanishingly thin — around 100 nanometers — while the substrates they sit on can be 10,000 times thicker. Despite their lopsided sizes, in this experiment, the thin film was able to push and pull on the substrate. Frañó likened the discovery to a tree on a mountaintop being able to move the entire mountain. Furthermore, the substrate also acted on the thin film, each pushing and pulling on the other.
They spent the next four years making sure what they saw wasn’t a fluke. They not only reproduced the results doing the research exactly as they had done it the first time, but they also reproduced the results when they changed certain aspects of the sample, such as the thickness and material of the substrate. They also tested their results using different instrumentation. Much of that work was done at Argonne, which has a synchrotron facility that produces high brilliance X-rays, and at Brookhaven National Laboratory, home to the world’s only all-electric ultrafast electron microscope, which is able to capture exactly how a working device behaves under realistic operating conditions.
It wasn’t a fluke.
“The assumption that substrates are inert needs to be rethought,” Frañó stated. “Going forward, we have to assume that the substrate is undergoing changes when the film is. This is a transformational notion that counters decades of previous supposition.”
This paper is a companion to a paper that was published last year in ACS Nano and was recently named an ACS Nano Impact Award finalist. In that paper, they used dark-field X-ray microscopy to study the thin film, while this paper focuses on the substrate. Together they illustrate how one research result leads to the next research inquiry.
In Frañó’s case, the question he now asks himself is, “What can we do with this new substrate discovery? If a tree is moving the whole mountain, let's find a way to capitalize on that.”
One idea is that the substrate can act as a medium that can couple materials on either end of the substrate. Currently, thin film devices are connected two-dimensionally, but if researchers could build on both sides of a substrate, it would allow for three-dimensionality, opening up the possibility of denser, more interconnected computer chips, and bringing us one step closer to the next generation of computing.
UC San Diego authors include Elliot Kisiel (now at Argonne National Laboratory), Erbin Qiu, Wei He, Rourav Basak, Junjie Li, Ivan K. Schuller and Alex Frañó.
Funding was provided by the U.S. Department of Energy (DESC0019273, DE-AC02-06CH11357, DE-SC0012704 and DE-AC02-76SF00515).
Read more news about: Physical Sciences
Share This:
Stay in the Know
Keep up with all the latest from UC San Diego. Subscribe to the newsletter today.
