Five researchers in the University of Illinois Urbana-Champaign’s Grainger Engineering are participating in four 2024 Multi-University Research Initiatives, or MURIs, including one project led by civil & environmental engineering professor Tugce Baser. Sponsored by the U.S. Department of Defense, these initiatives fund teams of investigators spanning multiple institutions of higher education as they conduct basic, interdisciplinary research on critical topics.
“The science and engineering challenges we face today are highly complex and cross-disciplinary,” said Bindu Nair, the director of the Basic Research Office in the Office of the Under Secretary of Defense for Research and Engineering. “The MURI program acknowledges these complexities by supporting teams whose members have diverse sets of expertise as well as creative scientific approaches to tackling problems.”
The program is highly competitive. Of the 276 research teams that submitted white papers this fiscal year, 102 were selected for a full proposal review, and 30 received awards. Each team will receive $1.5 million each year for three years with the possibility of a two-year extension subject to satisfactory research progress.
Baser will serve as the principal investigator on the project titled “I’M-SHARP: Interdisciplinary Material Science for the Hyperspectral Remote Sensing of Permafrost.” By drawing on new tools developed in remote sensing and electromagnetic theory, her team will study the properties of the permafrost – the ground that remains frozen year-round – under current and changing environmental conditions. “Frozen ground is a crucial element affecting the rate and magnitude of Earth’s global warming because it plays a critical role in our planet’s water, heat and carbon cycle,” Baser said. The advances made in this study will enable the design of high-resolution sensors for tracking permafrost landscapes and addressing the challenges in the Defense Department’s climate action plan.
Materials science & engineering professor Paul Braun is a co-principal investigator on the project titled “Multiscale Self‐Assembly of Non‐Hermitian Devices: Meta‐Atoms to Nanomaterials” led by New York University. Inspired by how ordinary atoms combine to form complex molecules and crystalline structures, the team will consider how artificial nanoparticles called “meta-atoms” can arrange themselves to form large-scale devices with non-Hermitian elements. This will enable the devices to have properties such as unidirectional invisibility, enhanced photonic crystal cavity sensing and high-power single-mode lasing with high spatial and spectral purity.
The Illinois team led by Braun will synthesize the meta-atoms and investigate how they self-assemble into larger structures. The team will measure the optical properties of the structures and provide data to the rest of the team to compare to theory and simulation.
Mechanical science & engineering professors Bill King and Nenad Miljkovic are co-principal investigators on the project titled “Fundamentals of Machine Learning for Phase Change Heat Transfer” led by the University of California, Irvine. Heat flows that result from phase changes such as boiling to vapor condensation are crucial to modern building systems, transportation, refrigeration and power generation. However, the physics of these processes remains incredibly difficult to model and simulate because these processes are fundamentally chaotic and involve multiple physical processes. Researchers will use state-of-the-art computer vision and physics-based machine learning techniques to interpret and predict phase changes in system designs.
The Illinois team led by King and Miljkovic will collect high-resolution vision information of phase changes and use it to train machine learning models. The team will then use these models to design better-performing phase change heat transfer systems.
Computer science professor Edgar Solomonik is a co-principal investigator on the project titled “Tensor Approaches for Simulating Kinetic Systems” led by the University of Delaware. Problems such as the kinetics of and particle distributions in nonequilibrium plasmas require solving partial differential equations in very high dimensions, and standard simulation techniques have proven inadequate. The research team will study these problems using the tensor network approximation that promises to greatly reduce the computational cost of studying high-dimensional objects. Although nonequilibrium plasmas are the inspiration for this work, the tools developed will extend into multiple problems of interest to the Defense Department.
The Illinois team led by Solomonik will develop methods for decomposition and optimization tools with a focus on tensor network topology optimization, low-rank tensor network representation of high-dimensional functions and operators and data-driven learning of tensor network representations.