Scott Institute announces 2026 seed grant winners

For the fourteenth year, the Scott Institute for Energy Innovation at Carnegie Mellon University has proudly supported an interdisciplinary selection of energy and climate research projects through its seed grant program. 

This year, a total of ten projects received over $500K through seed funding and the hardware and software tool upgrade program, an additional resource for supporting the vitality of energy and climate research among Scott Institute faculty affiliates. 

“The Scott Institute seed grant program helps our researchers take bold risks and create what’s next in energy,” said Costa Samaras, director of the Scott Institute. “We are thrilled to support faculty-led energy research projects this year representing nearly every part of campus.”

This year's award winners are based in the College of Engineering, School of Computer Science, Heinz College of Information Systems and Public Policy, Mellon College of Science, College of Fine Arts, and Tepper School of Business.  

Seed grant recipients

Katherine Flanigan

Katherine Flanigan, an assistant professor of civil and environmental engineering, will leverage AI to fill the civic infrastructure gap, a term that describes the missing link between community knowledge of, and institutional responses to, events like blackouts that are caused by extreme weather and threaten fragile energy systems. 

Flanigan will add an AI layer to a civic technology platform called re+connect so that community-generated data can inform the tool’s recommendations for how to respond to energy disruptions, resulting in improved preparedness, response, and recovery for affected communities. Pilot deployments of the updated re+connect platform will occur in areas (e.g., northwest Florida, Puerto Rico) that experience frequent blackouts and other recurring energy challenges. Flanigan’s work is in partnership with re+collective, Inc., and community organization Surcando La Historia. 

As co-leader of the Center for Engineering Resilience and Climate Adaptation (CERCA) and with established research in community-engaged infrastructure system adaptation, Flanigan is poised to continue building on a strong foundation thanks to additional support from the Scott Institute, including partial support from the Rankin Grand Challenge Partnership.

Shawn Litster 

A research team led by Shawn Litster, a professor of mechanical engineering, will explore an alternative fuel cell to power untethered robotics and unmanned aerial vehicles (UAVs), both of which are capable of consuming more energy resources than lithium ion batteries and internal combustion engines can provide. 

With support from the Scott Institute, the collaborators will purchase experimental components to produce proof-of-concept data, allowing them to answer fundamental and applied questions about the fuel cell’s properties. 

Rounding out the research team are Jay Whitacre, a professor of materials science and engineering and engineering and public policy; and Chris Pistorius, a professor of materials science and engineering.

Amir Barati Farimani 

Lead researcher Amir Barati Farimani, a professor of mechanical engineering; and collaborator, Paulina Jaramillo, a professor of engineering and public policy, will use support from the Scott Institute to address the current “data-rich yet insight poor” approach to monitoring and forecasting U.S. power grid conditions. 

Improved forecasting of grid conditions, detection of anomalies, and recommendation of real-time resilience actions is increasingly important as the grid contends with extreme weather events and accommodates new energy demands from the electrification of transportation, industry, and buildings. 

Using seed funding partially supported by the Scott Institute’s Anthropic Grand Challenge Partnership, Farimani and Jaramillo will construct an initial multimodal dataset, develop a prototype foundation model for short-term forecasting and anomaly detection, and create preliminary AI agents meant to communicate findings, interpret emerging risks, and assist grid operators during rapidly evolving events. 

Alan McGaughey

Multi-departmental collaboration between Alan McGaughey, a professor of mechanical engineering; and Gerald Wang, an associate professor of civil and environmental engineering, enables a computational science- and engineering-informed approach to discover new fluids that can serve as refrigerants. This project is partially supported by the Scott Institute’s Trane Technologies Grand Challenge Partnership.

Refrigerants support cooling, a tool for climate adaptation that only becomes more critical, including for health and safety, as global temperatures continue to rise. McGaughey and Wang aim to develop a set of models that can predict thermophysical properties relevant to refrigeration, resulting in a refrigerant recommendation engine that can propose new candidates capable of achieving specified cooling goals.

Woody Zhu 

Wildfires, destructive and hazardous to human life and made more intense by extreme weather and aging infrastructure among other factors, have inspired both short- and long-term solutions that each come with tradeoffs. 

A seed project to be led by Shixiang (Woody) Zhu, an assistant professor of data analytics at the Heinz College; and Ramteen Sioshansi, a professor of engineering and public policy, will develop an integrated causal-inference and optimization framework to determine the effectiveness of different wildfire mitigation strategies. 

Looking specifically at two methods, Public Safety Power Shutoffs and fast-trip programs, Zhu and Sioshansi will combine statistical machine learning, power-system modeling, and economic assessments of outage impacts to study where and when these interventions are most effective, enabling long-term resilience planning for wildfires. 

Riccardo Paccagnella 

Riccardo Paccagnella, assistant professor in the Software and Societal Systems Department in the School of Computer Science, will address a problem for sustainability and security: processor overheating. 

Processor overheating underscores the fact that 40 percent of the electricity used by data centers is used for cooling, according to the IEA. Moreover, overheated processors can be exploited to leak sensitive information, a form of attack known as Hertzbleed. 

Paccagnella will design and implement a solution called Goldilocks to help maintain a frequency level that’s “just right” for each machine and workload, providing a tool for mitigating security risks and reducing overheating in data centers. 

Alison Barth 

Alison Barth, a professor in the Department of Biological Sciences in the Mellon College of Science, aims to draw inspiration for engineered systems from the energy-efficiency of biological networks like the mammalian brain. 

The cerebral cortex is particularly important for learning and decision making, though there’s much to be discovered about its functionalities. With support from the Scott Institute, Barth hopes to identify learning principles in cortical networks that can be applied to the design of AI systems, like deep learning networks and large-language models, for more energy-efficient learning.

Hardware & software tool upgrade program 

Marc De Graef

The Materials Characterization Facility (MCF) at Carnegie Mellon University supports the discovery of the structure and chemistry of advanced materials. 

Marc De Graef, MCF’s faculty director, will apply support from the hardware and software tool upgrade program towards acquiring an integrated backscatter electron (BSE) imaging and energy dispersive x-ray spectroscopy (EDS) detector for a scanning electron microscope in the MCF, also requiring a software upgrade. 

By upgrading the detector and the software, the Scott Institute will by extension support the work of numerous affiliated faculty, research groups, and startups who make use of the MCF’s resources.

Nick Muller

Do power outages at oil refineries impact retail gasoline prices in the U.S.? 

To research a most accurate answer to this question, Nick Muller, professor of economics, engineering, and public policy; and Ph.D. student, John Mantus, will use support from the Scott Institute to purchase comprehensive county or state daily gasoline price data, increasing the scope and resolution of their analysis.

Azadeh Sawyer

Formerly, a rooftop weather station at the Intelligent Workplace in the School of Architecture collected valuable local environmental data. Azadeh Sawyer, an assistant professor in the School of Architecture, and co-investigator Fred Betz, will use resources from the Scott Institute to get the weather station up and running again, acquiring new environmental sensors, a data logger, mounting hardware, and other accessories. A refurbished weather station will provide data relevant to researchers across the university, particularly for urban microclimate analysis and building performance analysis modeling.