Making the rules

A member of the National Academy of Sciences and American Academy of Arts & Sciences, as well as a fellow of the American Association for the Advancement of Science and four discipline-specific societies, Sharon Glotzer is internationally recognized for her foundational research into the self-assembly of nanoparticles. She is also a first generation scientist and former Valley Girl who never worried she wouldn’t be taken seriously in science and engineering.

Her remarkable insights into nanoparticle assembly have made her a top theorist – and a favorite collaborator of leading nano-engineering experimentalists seeking to understand their own creations. Her ultimate goal is even more ambitious: she wants to understand the design rules for building new materials from the nanoscale up, enabling properties that were previously impossible.

Although Glotzer found her intellectual home in chemical engineering, her path to the field was hardly conventional. While finishing her physics degree at the University of California, Los Angeles, her career launched almost as soon as she set foot in a research laboratory at the aerospace company TRW.

Yet the start was inauspicious. After two weeks of searching through filing cabinets of data taken by the Voyager I satellite for the signature of a plasma phenomenon, she went to her supervisor.

“I said, ‘This isn’t working for me. You’re not using my talents. I need something more challenging,’” Glotzer recounted.

He saw her point and reassigned her to the superconductivity group, where she re-derived the physics behind a phenomenon used for precise control of light. That first paper, recasting the theory into a simpler and more accessible formulation, lent a new clarity to the field and went on to be widely cited – including by the renowned string theorist Ed Witten.

When her TRW colleagues recommended her for a doctoral fellowship, she jumped at the chance, joining the group of their collaborator at Boston University. But even from grade school, Glotzer says she was never great at executing experiments, always more interested in planning them and analyzing the data.

“A vacuum pump blew up all over me one day, and I was taking a class in statistical mechanics,” said Glotzer. “My to-be-future-advisor saw me and said, ‘You look like a theorist. Come talk to me later.’”

By the end of her first year, she had transferred into the group of the world-renowned statistical physicist H. E. Stanley where she was introduced to computational science. Then, when the university acquired a new-fangled supercomputer, she was among the first there to learn its programming language.

“I wanted to program the biggest computers on the planet,” she said. “And then it turned out that I was actually really good at it – that was my thing.”

After her PhD, she won a fellowship with the National Institute of Standards and Technology (NIST).

“NIST is a wonderful place. It’s a great environment for doing science, especially as a junior researcher,” said Glotzer. “Nobody’s teaching, nobody’s writing proposals – just doing research.”

As a post-doctoral research fellow, she had all the equipment and expertise of the lab at her fingertips while still having the freedom to pursue her own ideas. She and three other young scientists trained in computational science started a group specializing in simulation, with the aim of supporting projects throughout the Materials Science and Engineering Laboratory. That center is still running today.

We went from studying a single nanoparticle shape to studying 145 different shapes in a single paper, to tens of thousands, to now hundreds of millions.

Sharon Glotzer

Once she became research staff, the problems were more often handed down from above. She built a “baby group” of graduate students from nearby universities and post-doctoral fellows who could work outside the official agenda, but it wasn’t enough.

Observing her knack for growing and supervising her group, her own supervisor began preparing the way for Glotzer to move up in management at NIST. While the influence of the director’s office, with its Congressional briefs and testimonies, held some appeal, Glotzer knew she had more to add to the research field.

“I just had so much fun working with really smart people, coming up with these great questions, answering them, and then changing the way people thought about stuff,” she said.

That’s when Northwestern University first approached her about becoming a chemical engineering professor. She was mulling over their offer at a conference when she ended up next to U-M’s Ron Larson, an internationally recognized chemical engineer, in the cafeteria line.

She spilled the beans about the offer from Northwestern over lunch. Larson, about to begin his first year as chair of chemical engineering at U-M, saw an opportunity.

“He said, ‘Don’t do anything! Give me three days,’” said Glotzer.

U-M matched the offer, bringing her in as an associate professor with tenure. U-M’s Department of Materials Science and Engineering hired her husband-to-be, John Kieffer, away from the University of Illinois Urbana-Champaign, and the couple moved to U-M together in 2001.

Glotzer’s expertise in the growing field of computational materials science led her to establish two new graduate courses at U-M: Computational Nanoscience of Soft Matter (ChE/MSE 557) and Assembly Engineering (ChE 696).

She turned her leadership and management skills toward growing an exceptionally large research group: currently three full-time staff, six postdoctoral researchers, 26 graduate students and six undergraduate researchers.

Glotzer helped lead the College of Engineering and campus in building U-M’s computational research infrastructure, now Advanced Research Computing, and she also engaged in outreach among other universities in the Great Lakes region, helping them start local computer simulation cohorts.

“The idea was, how can we leverage critical mass in high performance computing at universities that have it to develop and teach curricula to students at schools everywhere?” said Glotzer.

The program was so successful at building up regional strength in scientific computing that it’s now obsolete.

But as computing resources became faster, and simulation results began sending back terabytes of data at a time, Glotzer soon had to reckon with the fact that scientific computing skills were no longer enough: she needed to harness data science. Her students were on top of the trend, studying data science online and forming a club to apply the new techniques to their simulation data.

“In our simulations, we went from studying a single nanoparticle shape to studying 145 different shapes in a single paper, to tens of thousands, to now hundreds of millions,” said Glotzer.

This tactic has enabled her group to produce some of the first design rules for self-assembling nanoparticles. The entropy in the system, governed in part by particle shape, can be an unexpectedly significant driver. Building on this finding, Glotzer and her small army are trying to develop a theory that will guide the design of new materials.

“These complex building blocks can be made with attributes that you just don’t get when you work on the molecular scale. It enables you to make materials that you can’t just get naturally out of atoms and molecules,” said Glotzer.

Sharon Glotzer is the Anthony C. Lembke Department Chair of Chemical Engineering, the John Werner Cahn Distinguished University Professor of Engineering, the Stuart W. Churchill Collegiate Professor of Chemical Engineering, a professor of materials science and engineering, a professor of macromolecular science and engineering, and a professor of physics.