How can we use science to build better gingerbread houses?
That’s something Miranda Schwacke spent a lot of time thinking about. Graduate students in MIT’s Department of Materials Science and Engineering (DMSE) are part of Kitchen Matters, a group of graduate students who use food and kitchen tools to explain scientific concepts through short videos and support events. Past topics have included why chocolate “sets”, why it becomes unwieldy when it melts (spoiler: it lets in water), and how to make isomalt, the glass of sugar that stunt performers jump over in action movies.
Two years ago, when the group was making a video on how to build a structurally sound gingerbread house, Schwacke scoured cookbooks for the variables that would make the most dramatic differences in the cookies.
“I read about what determines the texture of a cookie and tried several recipes in my kitchen until I found two gingerbread recipes that I was happy with,” says Schwacke.
She focused on the butter. Butter contains water, which turns into steam when baked at high temperatures, creating air pockets within the cookies. Schwacke predicted that reducing the amount of butter would produce a denser gingerbread, strong enough to hold up well as a house.
“This hypothesis is an example of how changes in structure can affect the properties and performance of materials,” Schwacke said in the eight-minute video.
The same curiosity about materials properties and performance drives her research into the high energy costs of computing, especially artificial intelligence. Schwacke is developing new materials and devices for neuromorphic computing that mimic the brain by processing and storing information in the same place. She studies electrochemical ionic synapses. This is a small device that can be “tuned” to adjust its conductivity, similar to neurons that strengthen or weaken connections in the brain.
“If you look at AI in particular, training very large models, it consumes a lot of energy. And if you compare that to the amount of energy humans use to learn things, the brain uses much less energy,” Schwacke says. “That led to this idea of finding a more brain-inspired, energy-efficient way of AI.”
Her advisor Bilge Yildiz emphasizes this point. One reason the brain is so efficient is that it doesn’t need to move data back and forth.
“In the brain, the connections between neurons called synapses are where information is processed. That’s where the signal transmission is. The signals are processed in the same place, they’re programmed, and they’re also stored in the same place,” says Yildiz, the Breen M. Carr Professor (’51) in the Department of Nuclear Science and Engineering and DMSE. Schwacke’s device aims to replicate that efficiency.
scientific roots
Born to a marine biologist mother and an electrical engineer father, Schwacke was immersed in science from an early age. Science “has always been part of how I understand the world.”
“I was obsessed with dinosaurs and wanted to be a paleontologist when I grew up,” she says. But her interests expanded further. In middle school in Charleston, South Carolina, she participated in the First Lego League Robotics Competition, building robots that completed tasks like pushing and pulling objects. “My parents, especially my father, were very involved with the school team and helped us design and build small robots for competitions.”
Meanwhile, her mother was researching how dolphin populations are affected by pollution for the National Oceanic and Atmospheric Administration. It had a lasting impact.
“This was an example of how science can be used to understand the world and how it can be improved,” Schwacke said. “That’s what I always wanted to do in science.”
It was later in her high school’s magnet program that she became interested in materials science. There, she was introduced to an interdisciplinary subject that combines physics, chemistry, and engineering, studying the structure and properties of materials and using that knowledge to design new materials.
“I’ve always loved studying everything from this very basic science of studying how atoms are ordered, to these solid materials that we interact with in everyday life, and how that gives them those properties that we can see and play with,” Schwacke says.
As a fourth-year student, she participated in a research program for a thesis project on dye-sensitized solar cells. Dye-sensitized solar cells are a low-cost, lightweight solar cell technology that uses dye molecules to absorb light and generate electricity.
“What drove me was to really understand how we transition from light to usable energy, and also how this could help us get more renewable energy sources,” Schwacke says.
After high school, she headed across the country to Caltech. “I wanted to try a completely new field,” she says. There, I learned about materials science, including nanostructured materials that are thousands of times thinner than a human hair. Her focus was on material properties and microstructure (the tiny internal structures that determine a material’s behavior), which led her to electrochemical systems such as batteries and fuel cells.
AI Energy Challenge
At MIT, he continued his research in energy technology. She met Yildiz during a Zoom meeting during her first year of graduate school in the fall of 2020, when campus was still operating under strict COVID-19 protocols. Yildiz’s lab studies how charged atoms, or ions, move through materials in technologies such as fuel cells, batteries, and electrolyzers.
The lab’s brain-based computing research sparked Schwacke’s imagination, but she was equally drawn to Yildiz’s way of talking about science.
“It wasn’t based on jargon and emphasized a very basic understanding of what’s going on: ions go here, electrons go here, to fundamentally understand what’s going on in the system,” Schwacke says.
That mindset shaped her approach to research. Her early projects focused on the properties these devices need to work properly (fast operation, low energy usage, compatibility with semiconductor technology) and the use of magnesium ions to replace hydrogen, which can leak into the environment and destabilize the devices.
Her current project, the focus of her doctoral thesis, focuses on understanding how the insertion of magnesium ions into tungsten oxide, a metal oxide whose electrical properties can be precisely tuned, changes its electrical resistance. In these devices, tungsten oxide acts as a channel layer, and the resistance controls signal strength, much like synapses regulate signals in the brain.
“I’m trying to understand exactly how these devices change channel conductance,” Schwacke says.
Schwacke’s research was awarded MathWorks Fellowships from the College of Engineering in 2023 and 2024. This fellowship supports graduate students who utilize tools such as MATLAB and Simulink in their research. Schwacke applied MATLAB to analyze and visualize critical data.
Yildiz describes Schwacke’s research as another step toward solving one of AI’s biggest challenges.
“This is electrochemistry for brain-inspired computing,” Yildiz says. “This is a new context for electrochemistry, but it also has energetic implications, as the energy consumption of computing is increasing unsustainably. We need to find new ways to do computing at much lower energies, and this is one way to help us move in that direction.”
As with any pioneering work, it comes with challenges, especially in bridging concepts between electrochemistry and semiconductor physics.
“Our group comes from solid-state chemistry, and when we started working on magnesium, no one had used it in this type of device before,” Schwacke says. “So we were looking at the magnesium battery literature for inspiration and different materials and strategies that could be used. When we started this, I wasn’t just learning the language and standards of one field. I was learning it for two fields and then trying to translate between the two fields.”
She also tackles a challenge familiar to all scientists: how to make sense of messy data.
“The main challenge is getting the data and knowing that it’s being interpreted in the right way so that you can understand what it actually means,” Schwacke says.
She overcomes hurdles by working closely with colleagues in a variety of fields, including neuroscience and electrical engineering, sometimes making small changes to experiments and seeing what happens next.
community issues
Schwacke isn’t just active in the lab. At Kitchen Matters, she and other DMSE graduate students set up booths at local events such as the Cambridge Science Fair and Steam It Up, an after-school program with hands-on activities for children.
“We did ‘pHun with Food’ and spelled ‘fun’ with pH, so we used cabbage juice as a pH indicator,” says Schwacke. “We had the kids test the pH of lemon juice, vinegar, and dish soap. They had a lot of fun mixing different liquids and seeing the different colors.”
She also served as social chair and treasurer of the Graduate Materials Council, a graduate student group at DMSE. As an undergraduate at Caltech, she led science and technology workshops for Robogals, a student-led group that encourages young women to pursue careers in the sciences, and helped students apply to the school’s summer undergraduate research fellowships.
For Schwacke, these experiences have honed her ability to explain science to a variety of audiences, a skill she believes is essential whether presenting at children’s exhibitions or research conferences.
“I’m always thinking, where should the viewer start and what do I need to explain before they understand what I’m doing?” she says.
Schwacke believes that communication skills are central to community building and believes this is an important part of conducting research. “It helps spread ideas. It always helps to get a new perspective on what you’re working on,” she says. “I also think it keeps me sane during my PhD.”
Yildiz believes Schwacke’s community involvement is an important part of her resume. “She does all of these activities to motivate the broader community to conduct research, be interested in science, and pursue science and technology, but her abilities will also help advance her own research and academic endeavors.”
After earning her Ph.D., Schwacke hopes to use her communication skills in academia to inspire the next generation of scientists and engineers. Yildiz has no doubts that she will thrive.
“I think she’s perfectly qualified,” Yildiz says. “She’s good, but that brilliance isn’t enough. She’s tenacious and resilient, which is really what we need on top of that.”
