Skip to main content
Jason Kawasaki in lab with PhD student Sebastian Manzo
July 18, 2022

The whole story: Promising thin film growth technique is the result of microscopic holes

Written By:

Materials researchers at the University of Wisconsin-Madison have revealed some of the atomic-scale mechanisms behind remote epitaxy, an emerging method for synthesizing extremely thin films and membranes—and the process is very different from what many researchers believed.

These thin films could play an important role in flexible or wearable electronics and in the development of unique multilayer devices for next-generation computing.

Led by PhD student Sebastian Manzo and Jason Kawasaki, an assistant professor of materials science and engineering, the work appears in the July 2022 issue of the journal Nature Communications.

Thin films are grown or deposited on a crystalline substrate during a process called epitaxy. The structure of the substrate serves as a template, guiding the orientation of the film’s growth via atomic bonding and other atomic-scale interactions. But there are some problems with directly depositing crystals on top of a substrate; notably, removing the thin films at the end of the growth process is often difficult and can damage or destroy the film.

One workaround is called remote epitaxy, in which researchers place a layer of the 2D material graphene on top of the substrate before they synthesize the film. As the thin film grows, it still follows the orientation of the substrate, but the intermediate graphene layer makes it much easier to lift off the new film after growth since it does not bond directly to the substrate.

While that method seems to work in many situations, researchers aren’t sure exactly why, since theoretically the graphene should block or reduce the interactions between the developing crystal and the substrate. “The fundamental question is how does this process work?” says Kawasaki. “How is it that you can grow that single crystal on top of the graphene? This remote mechanism has been pretty controversial because it assumes that the interface between the graphene and the substrate is perfect.”

To investigate, Kawasaki and Manzo decided to search for defects in the graphene. Even small defects or impurities—just fractions of a percent—can have huge impacts on the properties of materials. “You learn as an undergraduate in materials science that defects are everything,” says Kawasaki. “So coming from that mindset, we might expect that defects in graphene could matter a lot.”

Manzo led a series of experiments, using high-powered microscopes to take snapshots of activity at the atomic level during remote epitaxy. First, he analyzed what happened to the graphene layer during annealing, a preliminary step in which the substrate and graphene are heated up to remove oxides from the substrate surface. “We saw how the morphology of the graphene was changing at those particular temperatures,” says Manzo. “And that’s when we noticed that 10-nanometer pinholes were forming because the oxide was bursting out of the substrate.”

During epitaxy, they saw that the thin film actually formed at these pinholes in the graphene. “We were basically seeing this kind of mushroom effect where film growth starts at the pinholes, and then it just kind of grows laterally outward,” says Manzo.

In other words, the substrate was still orchestrating the growth of the thin film directly at these pinholes, not via remote interactions.

Ultimately, the finding doesn’t change the utility of the process. However, it does provide researchers a better understanding of how the pinholes could help better control and tune the epitaxy of thin films.

Manzo says the experiments don’t close the book on remote epitaxy, either; in fact, true remote epitaxy would allow the growth of even thinner films. If researchers could produce defect-free graphene without pinholes and place that on a substrate, they could test the concept. “We are starting to do other sets of experiments to try to see if we can truly find some proof that remote interactions play a role,” says Manzo.

Other UW-Madison authors include Patrick J. Strohbeen, Zheng Hui Lim, Vivek Saraswat, Dongxue Du, Shining Xu, Nikhail Pokharel, Professor Luke Mawst, and Professor Michael Arnold.

The authors acknowledge support from the National Science Foundation via award DMR-1752797; the Defense Advanced Research Projects Administration via grant number D19AP00088; U.S. Department of Energy, Office of Science, Basic Energy Sciences, under award no. DE-SC0016007; NSF Division of Materials Research through the University of Wisconsin Materials Research Science and Engineering Center, Grant No. DMR- 1720415; National Science Foundation ECCS-1806285; Raman, electron microscopy, and PPMS facilities supported by the NSF through the University of Wisconsin Materials Research Science and Engineering Center under Grant No. DMR-1720415; and the Wisconsin Alumni Research Foundation (WARF).