Pennsylvania State Univ University Park

NSF Awards · FY 2025 · 2025-10

Data centers (DC) are the backbone of the modern digital economy, critical to the U.S. economical growth, national security, public health, and enhanced data security and management. DCs are an energy intensive infrastructure, accounting for over 4% of total electricity use worldwide. As demand for AI and cloud computing grows, efficient cooling systems are critical to ensuring reliable and resilient DC operations. A critical failure in DC cooling systems can have catastrophic consequences, including total system shutdown, loss of data, and IT equipment. To prevent such catastrophic events, novel Fault Detection and Diagnostics (FDD) and mitigation techniques are essential. Currently, most FDD methods rely on conventional statistical techniques, machine learning models, or ad-hoc estimations. However, these methods are often limited in scope and may fail to detect rare or complex failure scenarios – particularly those arising from complex cascading events or malicious cyber-attacks. To tackle this challenge, this project develops a new FDD method based on failure and cyber-attack detection in supervisory control theory of discrete event systems. The intellectual merits of this project are: (1) new FDD methods for detecting and mitigating cascading faults and cyber-attacks resulting in resilient DC cooling system operation, (2) an open-source virtual testbed for evaluating performance of the proposed algorithms, and (3) a hardware-in-the-loop testbed to understand the challenges of FDD-enabled controls in real-world DC cooling equipment. The broader impacts of this project include new FDD methods to transform conventional DC cooling system design and management into future resilient DC cooling infrastructure and a field-validated computational framework for advanced FDD analysis of resilient cyber-physical infrastructure. By integrating the event-driven supervisory control and physics-based modeling, the goal of this project is to develop a field-validated, FDD-enabled, model-based control and computation framework for the robust design and reliable operation of next-generation resilient DC cooling systems. The proposed FDD method: (1) identifies, analyzes, and captures complex dynamics of benign and malicious faults, with rigorous detection guarantees, (2) characterizes critical attack vectors that pose a severe threat to cooling system management and DC operation, and (3) generates robust, real-time control responses that enable adaptive system adjustments to sustain normal cooling operation during major disruptions. The findings of this research are expected to have a broad range of real-world applications, particularly in design and development of attack-resilient DC cooling infrastructure across the United States and can be used by DC developers, technology companies, utilities, and HVAC manufacturers. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-10

This project aims to serve the national interest by advancing engineering design education through the use of artistic practice to support student learning of additive manufacturing. The field of additive manufacturing (AM) and colloquially known as 3D printing is expanding to a worldwide revenue of more than $21 billion, making it an important facet of the United States' future global manufacturing competitiveness. However, due to the organic, freeform geometries possible with AM, rapidly educating a capable and innovative future AM workforce requires an interdisciplinary approach that is not always leveraged in engineering education. To address this limitation, this project aims to establish new, relevant paradigms in engineering education for AM within the domain of Science, Technology, Engineering, Art, and Math, better known as STEAM education. Through this lens, it will be possible to investigate and understand the potential impact that the integration of arts-based educational practice can have on AM educational outcomes with engineering undergraduates. Without this understanding, the ultimate potential of STEAM as an inherent part of undergraduate AM design education remains limited. The objective of this project is to establish foundational knowledge of how (1) engineering students' intrinsic STEAM agency and (2) the introduction of arts-based epistemic practice influence the outcomes of the early-stage design process within the context of additive manufacturing. This foundational knowledge will result in the creation of an evidence-based, educational STEAM framework capable of implementation across Design for Additive Manufacturing (DfAM) coursework. Knowledge will be generated through empirical study of arts-based interventions as applied to undergraduate DfAM education with engineering students across academic levels. The project will result in (1) quantitative evidence demonstrating the impact of student experiences in AM, their interest in art, and their design self-efficacy, (2) identification of statistically significant differences in student DfAM outcomes from problem exploration, concept generation, and design evaluation when presented with arts-based epistemic practice, and (3) a robust framework for synergistically integrating arts-based educational practices into DfAM education for engineering undergraduates. These deliverables have transformative potential to expand the use and perceived validity of STEAM education across the field of AM in undergraduate institutions across the United States. The NSF IUSE: EDU Program supports research and development projects to improve the effectiveness of STEM education for all students. Through the Engaged Student Learning track, the program supports the creation, exploration, and implementation of promising practices and tools. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-10

Graphs are powerful tools for representing relationships in complex systems, from social networks to weather monitoring stations. Graph Neural Networks (GNNs) have emerged as effective methods for analyzing these interconnected systems, but their "black box" nature poses significant challenges in critical applications such as environmental monitoring, healthcare, and finance. This project develops a comprehensive framework for making GNN predictions explainable and trustworthy. The research addresses the urgent need for artificial intelligence systems that can not only make accurate predictions but also explain their reasoning in ways that domain experts can understand and verify. For instance, in South Florida's water management network where monitoring stations form a graph connected by hydrological pathways, emergency managers need to understand which stations and their interconnections are most influential in flood predictions. This capability is essential for building trust in these systems and ensuring their responsible deployment in applications that affect public safety and welfare. The project will train students in interdisciplinary research combining machine learning, information theory, and practical applications, while developing educational materials that bridge theoretical foundations with real-world implementations. This project establishes a unified framework using information theory concepts for explainable GNNs through two complementary research thrusts. The first thrust develops rigorous mathematical foundations for quantifying explainability in graph learning, including necessary and sufficient conditions for classifier explainability, methods to address out-of-distribution challenges, and ways to demonstrate how accurate the finding are. The second thrust translates these theoretical insights into practical architectures and algorithms, including computationally efficient explainers, generative models for robust explainers, and co-design frameworks that balance prediction accuracy with explainability. The research introduces novel concepts such as nonverbal signatures for characterizing explanation patterns and explanation-assisted learning mechanisms that leverage extracted explanations to improve model performance. Extensive evaluations will be conducted on benchmark datasets and specially curated weather forecasting datasets from South Florida's water management systems. The project advances the state-of-the-art by providing both theoretical rigor in quantifying explainability and practical solutions for deploying trustworthy GNN systems in critical applications. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-10

This EAGER project is developing hybrid (human-AI) technologies for forecasting complex sociotechnical outcomes. The project considers community experiences of social media as a proving ground for design and development. In doing so, it is laying theoretical and empirical groundwork for future technologies supporting user governance in shared spaces online. The project’s research has the main objectives of advancing machine learning and artificial intelligence to support the integration of human inputs through hybrid prediction markets, wherein humans participate alongside bot of AI-enabled traders to buy and sell assets representing future outcomes and deploying hybrid prediction markets to forecast user experiences, perspectives, and narratives online. This project is the first comprehensive effort to build and test hybrid prediction markets for integrated human-AI forecasting, and thus is a high risk-high reward EAGER effort. The approach will fold in human input as other machine learning algorithms cannot, i.e., directly within the algorithm's deployment, showcasing an opportunity for the integration of first-person perspectives and machine learning for complex tasks where participatory design is desirable. Markets will be studied theoretically and in practice, and thoroughly compared against a suite of computational, crowdsourced and hybrid benchmarks, including supervised and semi-supervised algorithms, survey elicitation, and ensembling. The market approach could enable explanations not generally afforded by deep learning-driven approaches yet critical to platforms and policy makers. Design of hybrid technologies will build on the first-hand accounts of impacted individuals and development will engage with community members throughout. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-10

Topography influences stress in the Earth’s crust and the pattern of deformation that arises during mountain building. Because the material properties of rocks near the surface of the Earth depend on their burial and exhumation history, in collisional mountain belts feedbacks may exist between the burial history of rocks in the subsurface, their strength at the surface, and topography. However, few field data exist to evaluate the strength of such feedbacks between surface and deep-Earth processes. This project will address this knowledge gap by quantifying the tectonic origin and geomorphic implications of variations in rock material properties across the Taiwan Central Range. In addition, this project will facilitate international collaboration between US and Taiwanese scientists through paired field expeditions and a student exchange program, leveraging new and existing research collaborations. To explore relationships among burial and exhumation history, rock strength, and topography, this project will collect and integrate a paired dataset of structural and geomorphic observations. Existing and new measurements of erosion rates and thermal and deformation history will constrain patterns in burial and exhumation. Field surveys of river corridors assisted by small unmanned aerial vehicles will facilitate characterization of how rock properties influence river incision and hillslope-channel coupling at high resolution (centimeter-scale) and large spatial extent (kilometer-scale). Repeat surveys will enable high-resolution analysis of river response to individual storms. Data from this project will serve as a baseline for quantifying landscape response to future extreme events and help to address hazards associated with landslides, floods, and sedimentation. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-10

NONTECHNICAL SUMMARY This CAREER award supports theoretical research and education in the broad field of nonequilibrium quantum dynamics with a view towards emerging quantum technologies. Quantum computing is a new mode of computation that harnesses the principles of quantum mechanics to store and process information. Rapid experimental progress is ushering in an era of useful near-term quantum computing platforms, with potential applications ranging from cryptography to drug design. Quantum computation relies on controlling the nonequilibrium dynamics of systems of many interacting quantum particles. Comprehending such dynamics thus plays a critical role in enabling future advances. Many-particle quantum systems tend to lose information about the state in which they were prepared. This tendency, known as quantum ergodicity, is detrimental to quantum computation, which hinges on the ability to preserve delicate quantum states and perform operations on them. The research component of this project will develop a theoretical understanding of a variety of mechanisms through which quantum ergodicity can be avoided. It will also consider how to realize these mechanisms on present-day quantum hardware with the long-term goal of expanding the toolkit for the study and manipulation of complex quantum systems. Quantum science and technology has been identified as a key national research priority. Sustaining leadership in this field requires nurturing a robust quantum workforce. To this end, the education component of this project includes developing an interdisciplinary quantum computing curriculum, whose success will grow the quantum talent pipeline. The principal investigator will also partner with institutional initiatives to engage in outreach around quantum physics topics. These activities will reach hundreds of precollege and college students and tap into popular excitement about quantum physics using hands-on activities and active learning approaches. TECHNICAL SUMMARY This CAREER award supports theoretical research and education in the broad field of nonequilibrium quantum dynamics with a view towards emerging quantum technologies. The research component addresses the foundational question of how quantum many-body systems can fail to relax to thermal equilibrium and maintain quantum coherence in the presence of strong interactions. Research activities are organized into three interrelated thrusts whose goals are: (1) Elucidate the role of emergent dynamical constraints in the far-from-equilibrium behavior of gauge theories and related quantum spin models. This question will be investigated using a combination of perturbation theory and numerical exact diagonalization to extract the timescales of relaxation processes in such systems. (2) Explore a general construction of quantum many-body scars, a dynamical regime where thermalization can be avoided by preparing the system in a special class of initial states. The construction hinges on the use of infinite-temperature thermofield-double states, which are of interest in both the high-energy physics and quantum information science communities. (3) Discover novel roadblocks to thermalization and decoherence intrinsic to quantum circuits, the natural setting for quantum computation. This project focuses on quantum circuits related to classical cellular automata, as well as circuits based on deterministic aperioidic sequences. All three topics are relevant to present-day experiments on platforms including cold atomic gases of bosons and fermions, arrays of Rydberg atoms in optical tweezers, and present day noisy intermediate-scale quantum hardware. Quantum science and technology has been identified as a key national research priority. Sustaining leadership in this field requires nurturing a robust quantum workforce. To this end, the education component of this project includes developing an interdisciplinary quantum computing curriculum, whose success will grow the quantum talent pipeline. The principal investigator will also partner with institutional initiatives to engage in outreach around quantum physics topics. These activities will reach hundreds of precollege and college students and tap into popular excitement about quantum physics using hands-on activities and active learning approaches. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-10

Large Language Models (LLMs) are increasingly deployed as the backbone of real-world applications such as Google Search with AI Overviews and Microsoft Bing Copilot. When data and code are not properly separated within an application, the latter (including AI applications) is vulnerable to cyber-attacks. This project's novelties are twofold: (1) conducting a systematic study to deepen the understanding of such threats, and (2) developing new defenses to mitigate such attacks. Its broader significance and importance lie in establishing foundational security principles for the rapidly growing ecosystem of AI applications, which are now widely deployed across diverse societal domains. Moreover, the released code and materials produced by this project will not only help secure real-world LLM-integrated applications but also serve as valuable educational resources for undergraduate and graduate courses, fostering the next generation of researchers and practitioners in this emerging security area. Security history shows that when data and instructions are not properly separated within a system, injection attacks can emerge—for example, SQL injection attacks in traditional software. Similarly, due to the lack of a clear boundary between instructions and data in prompts, LLM-integrated applications are inherently vulnerable to prompt injection attacks. To understand and mitigate such threats this project adopts a holistic approach comprising three interconnected research thrusts to systematically investigate the security vulnerabilities of LLM-integrated applications to prompt injection attacks and to develop new methods to prevent, detect, and attribute such attacks. The project will also open-source a platform that integrates our developed algorithms along with a comprehensive tutorial on prompt injection attacks and defenses. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-10

Extended reality (XR) technologies, including virtual reality (VR) and augmented reality (AR), are transforming how people interact with the world by merging virtual content with the physical world and creating immersive, interactive experiences. To reach their full potential, next-generation XR systems demand a high degree of context awareness, that is, a detailed understanding of both user behaviors and surrounding environmental conditions. With enhanced context awareness, XR systems can deliver virtual content that is personalized, timely, and highly relevant, adapting to user interactions and responding to changes in the surrounding environments. To achieve this goal, this project builds a new class of retrieval-augmented generation (RAG)-empowered XR systems that bring together the power of large language models (LLMs) and localized, context-rich knowledge databases to make XR systems more intelligent and adaptive. The project builds and maintains an accurate, up-to-date, and diverse knowledge database, which integrates diverse sources of contextual information, such as 3D object data, egocentric images, text inputs, and user-specific data like head pose, eye gaze, and user preferences. The project also makes context-aware XR more resource-efficient and low-latency, by strategically leveraging the collaboration of XR devices with nearby edge servers to process complex and dynamic contextual inputs. This project will lay the foundation for context-aware XR applications across multiple domains, such as commerce, entertainment, manufacturing, and social interactions. It will support a wide range of use cases, including smart homes, intelligent manufacturing, and collaborative social XR platforms, where systems can intelligently adapt to both individual and group user behaviors, as well as dynamic environmental conditions. The project will train several cohorts of undergraduate and graduate students. The findings of this project will be presented at K-12-oriented events, and at multiple venues in the field. This project designs, implements, and evaluates context-aware XR systems that can understand and respond to user states and environmental conditions. The research focuses on three thrusts. The first thrust builds a RAG system with a comprehensive knowledge database that integrates multimodal context data of XR users and environments. The second thrust designs resource-efficient update mechanisms that minimize latency for executing context-aware XR algorithms while maintaining high accuracy of context awareness under the resource constraints of edge servers. The third thrust tests the context-aware XR systems through building XR emulators and implementing real-world prototypes, with user studies evaluating performance based on real-time interactions. The project makes intellectual contributions to the studies of XR, edge computing, and machine learning. It augments LLMs with a personalized, localized knowledge database collected in real time from users’ This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-10

The National STEM Teacher Corps Pilot Program Regional Alliance project "Elevating Elementary Teachers as the Frontier of STEM Education" aims to recognize outstanding STEM educators in high-need schools that advance educational excellence in the Mid-Atlantic Region. This project intends to support 27 STEM Teacher Corps members by recognizing and rewarding their accomplishments and efforts to enhance STEM teaching and learning in K-5 classrooms. To elevate these teachers and their work, Teacher Corps members will receive annual stipends, specialized professional development in STEM curriculum and educational leadership, and classroom materials and travel support to regional workshops and science teacher conferences. The project will also provide significant professional development, mentorship, and support to hundreds of other STEM teachers within the region through activities led by the Teacher Corps members. The proposed project components will enable exemplary and innovative STEM teachers to participate in state and national science education conferences, lead workshops for K-5 teachers who are not Corps members, and build an important network of support for K-5 STEM education in the region. This project at Penn State University includes partnerships with West Virginia University, The College of New Jersey, K-12 schools and school districts, award winning teachers in the region, state-level STEM education networks, science teachers' associations, and key local education agencies from each state. Project goals include empowering K-5 teachers to become agents of change in STEM education through engaging students in scientific and engineering practices. By providing targeted professional development and fostering peer leadership, the project intends to improve student outcomes and create a network of educators equipped to lead a regional transformation. This project will recruit and train a group of dedicated K-5 teachers who will participate in summer workshops to enhance their knowledge and leadership abilities. These teachers will then support their peers to create engaging STEM learning environments for young students. Evaluation of the project is iterative and guided by the questions: Did the project's efforts elevate the STEM teacher profession and Teacher Corps members in the region by promoting STEM teaching and learning opportunities in elementary schools? Did sustained STEM-focused professional development and the efforts of the Communities of Practice enhance elementary teachers' STEM knowledge, confidence, and leadership skills? Evaluation will include formative assessments to guide program improvements and measure the project's effectiveness. The project team will also engage regional and national organizations to disseminate project outcomes and share best practices. This Regional Alliance project is supported through the National STEM Teacher Corps Pilot Program. The NSF National STEM Teacher Corps Pilot Program supports outstanding STEM educators in high-needs schools that advance educational excellence in our Nation’s preK-12 classrooms. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-10

This award supports research that focuses on the development of Human-centric AI for Reconfigurable Manufacturing Operation and desigN sYnthesis (HARMONY) to harmonize human workers with increasingly autonomous resources. Emerging technologies, such as automated guided vehicles, mobile manipulators, collaborative and reconfigurable robots, offer new opportunities to address persistent challenges like machine breakdowns, material shortages, and demand fluctuations through dynamic reconfiguration of hardware, software, and logistics. However, fully realizing these benefits requires advanced analytical and programming skills to orchestrate multiple autonomous resources while meeting multi-facet performance targets and operational constraints. While generative artificial intelligence (GenAI) offers an intuitive interface for human-autonomy collaboration, its current application in manufacturing is limited by a lack of domain-specific knowledge. This project seeks to create preference-aligned decision options that humans can explore, select, and refine through low-barrier, multimodal interfaces. The project looks to also include an educational program featuring an innovative curriculum in AI and digital manufacturing, hands-on K–12 engagement and professional development opportunities delivered via workshops, webinars, industry partnerships, and Smart Learning Factories. Successful implementation of HARMONY has the potential to transform manufacturing systems to incorporate autonomous resources in rebuilding national manufacturing capacity and prepare future workforce with advanced technologies. The goal of this research is to innovate GenAI solutions along with digital twins, and progressive data cultivation strategies for translating the autonomy of individual resources into system-level performance gain, all while under human supervision. Existing GenAI models face significant limitations due to a lack of domain specificity, limited manufacturing decision data, and the risk of hallucination. To overcome these challenges, this project pursues three key objectives: (i) the development of a progressive data cultivation strategy leveraging digital twins and statistical surrogate models; (ii) the design of tailored GenAI architectures that embed manufacturing-specific constraints, objectives, and interdependencies; and (iii) the systematic evaluation through Smart Learning Factories and industry collaborations. This project aims to advance the next generation of GenAI technologies for human-centric decision-making in dynamic, reconfigurable manufacturing environments. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-10

As population growth and urbanization accelerate worldwide, the demand for infrastructure that is both resilient and resource-efficient is becoming increasingly urgent. Concrete, the most widely used human-made material on Earth, plays a central role in meeting this demand but faces escalating challenges related to aging when exposed to the elements, material scarcity, and the limitations of current construction methods. This United States/Saudi Arabia workshop brings together leading researchers from both nations to identify transformative solutions for next-generation concrete materials and systems. Both the United States (US) and the Kingdom of Saudi Arabia (KSA) are leaders in research and technology transfer related to advanced cementitious materials. However, both countries face important challenges linked to the availability of quality materials, aging infrastructure, and the growing frequency and severity of extreme natural events. By fostering international collaboration and integrating expertise across materials science, construction engineering, and structural engineering, the workshop advances the national interest by promoting scientific progress and laying the foundation for innovative infrastructure that enhances public welfare, defense readiness, and economic resilience. A particular focus on mentoring early-career researchers and building global research networks supports the development of a skilled, internationally engaged STEM workforce. Technically, the workshop will bring approximately 20 US researchers to Riyadh to collaborate with counterparts from the KSA in four key research areas: (1) resource-efficient construction, (2) durability and resilience, (3) innovative materials and construction methods, and (4) next-generation cementitious binders and mixture design. The role of artificial intelligence, automation, and machine learning in accelerating innovation will be integrated into each focal area. The workshop will identify critical knowledge gaps and catalyze new directions in cementitious materials research, advancing a platform for interdisciplinary collaboration. Outcomes will include a joint research agenda, a publicly available workshop report, and a digital platform to support ongoing collaboration and proposal development aligned with the priorities of NSF and KSA’s Research, Development, and Innovation Authority. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-10

This NSF project aims to design and evaluate honeybee-inspired virtual electric peer-to-peer networks to enable efficient and resilient control of distributed energy resources. These resources include electric vehicles, heat pumps, electric water heaters, and battery energy storage systems at the distribution level. Existing electrical infrastructures are undersized to handle increasing loads, and a lack of effective coordination among these resources further exacerbates this challenge. Inspired by the decentralized coordination mechanisms observed in honeybee colonies—where energy (food) is exchanged among members in a process called trophallaxis—this project will develop a bio-inspired cyber-physical system where distributed resources (“bees”) and storage systems (“hive”) dynamically allocate energy. By applying principles from collective insect behavior, this research seeks to transform energy coordination, benefiting grid operators and consumers alike. The intellectual merits of the project include novel mathematical models based on trophallaxis, development of bio-inspired control strategies, and validation through virtual testbeds and real-world demonstrations. The broader impacts include advancing non-wire alternatives that enhance grid resilience, improving access to electricity services, and fostering interdisciplinary knowledge exchange between biology, computing, and engineering. Additionally, the project will provide publicly available open-source software, engage students through an undergraduate design competition, and disseminate findings through workshops and outreach initiatives. This project will address existing technical challenges in energy coordination by translating honeybee trophallaxis into mathematical models and integrating them into an innovative cyber-physical framework. It will develop predictive models for uncertain building loads and energy behaviors using stochastic transfer learning. Additionally, it will create bidirectional biology-technology knowledge transfer frameworks to inform control-oriented models across multiple system layers. New bio-inspired control methods will be designed to optimize peer-to-peer energy sharing and grid operations. The project will leverage virtual testbeds using Python and GridLAB-D for rigorous evaluation, with experimental demonstrations conducted at the University of Colorado Boulder’s microgrid. By combining expertise from biology, computer science, and engineering, this research will generate novel strategies for resilient, adaptive, and efficient grid operations. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-10

Non-technical Description: A central goal of Materials Genome Initiative is to predict superior intrinsic material properties on the atomic scale and translate them to superior technologies we can touch and hold. However, most such material discoveries are lost in translation due to the “mesoscale cliff.” Materials promising on the nanoscale may fail in devices where microstructures up to hundreds of micrometer in size dominate device performance. This team addresses this materials challenge to develop a fundamental knowledge base to deploy the next generation of cryogenic electro-optic materials integrated on silicon for chip-scale quantum integrated circuits. The electro-optic effect describes a material’s optical refractive index change upon the application of an electric field. Electro-modulators power our internet today by converting electrical to optical signals. They are also fundamental building blocks for the emerging scalable optical quantum computing hardware, on-chip trapped ion quantum computing schemes and developments in low temperature science. With these, new materials’ challenges arise, in that, electro-optic modulators must now respond in the gigahertz frequency range, be operated at cryogenic temperatures with low energy budget and must be integrated directly on silicon. This requires materials with cryogenic electro-optic coefficients that are many orders of magnitude higher than the current industry standard. In addition, they require large index and low optical loss at the telecom wavelength with low microwave dielectric constant and loss for low power, low loss operation. A lack of fundamental understanding on the mesoscale is undermining this effort with severely degraded performances in going from atoms to devices. This research team aims to make an impact here with a new theory approach informing experimental breakthroughs. Technical Description: The research team plans an Atoms-to-Devices design approach that is firmly rooted in the materials genome framework. It has three interlocking thrusts: (1) Density Functional Theory informed Thermodynamic Theory of Electro-Optics builds the foundational bridge between the atomic scale and the mesoscale using a modern thermodynamic theory of electro-optics to predict and validate new materials with superior intrinsic electro-optic properties. (2) Thermodynamics integrated Phase-Field Simulation implements the thermodynamic electro-optic theory using phase-field modeling to predict and experimentally validate complex mesoscale microstructures and their effective electro-optic properties. (3) An Open-source Phase-Field-integrated Electrodynamics simulation software package integrates phase-field modeling and electrodynamics simulations to design a digital twin of the physical modulator devices and their performances. A robust experimental testing and validation is built into each Thrust. Graduate and undergraduate students along with postdoctoral researchers and principal investigators will train together in an iterative closed-loop materials design crucible. Undergraduate, graduate and post-doctoral mentoring of personnel will ensure a robust pipeline for the next generation of workforce in quantum science. A website will be developed that will serve as a medium for disseminating the team’s work. Research breakthroughs with potential for technology translation potential will be communicated through Industry outreach that would also benefit student career development. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-10

Non-Technical Summary. Semiconductors are critical components in a wide range of technologies including smart phones, solar cells, and radiation detectors, among many others. Mixed-chalcogen solids are an important class of semiconductors that have at least two negatively charged atoms, or anions, which can arrange themselves in ordered or disordered ways within a crystal. The arrangement of anions strongly impacts the electrical and thermal properties of semiconductors, but scientists do not have reliable tools to predict when or how ordering occurs. To overcome this challenge, researchers at Pennsylvania State University and Portland State University, with support from the Solid State and Materials Chemistry program in NSF’s Division of Materials Research, combine computer modeling, machine learning, and laboratory experiments to develop a framework that can predict atomic ordering. The project outcomes advance scientific knowledge and can be used to design materials with improved properties and atomic-scale precision that is needed for next-generation applications. Additionally, this collaborative project includes hands-on training for graduate students, generates new content for college courses and promotes participation in science through mentoring and community interactions, thereby helping to develop a highly skilled workforce in the chemical and materials sciences. Technical Summary. Metal chalcogenides are a highly tunable class of materials due to the high miscibility of chalcogen anions in the solid state, which allows their properties to be fine-tuned using anion alloying strategies. Although chalcogens tend to form solid solutions in simple crystals, ordering does occur in structures with more than one anion site. Order-disorder phenomena affects their properties in numerous ways, but often goes unnoticed and is poorly understood. With this project, supported by the Solid State and Materials Chemistry program in NSF’s Division of Materials Research, the research team establish a deeper understanding of the fundamental mechanisms that define atomic-scale ordering in metal chalcogenides and other extended solids. This interdisciplinary approach, unifying experiment and theory, establishes chemical models for predicting local and long-range ordering phenomena. The integration of machine learning with density functional theory enables efficient screening of thousands of potential materials, creating a powerful platform for materials discovery. Beyond chalcogenides, the methodology provides a template for studying ordering phenomena in other multicomponent systems. Investigations of how ordering influences transport properties reveal hidden mechanisms governing macroscopic behavior, establishing new design principles that bridge atomic structure and functional performance in multianion solids. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

This project uses advances in artificial intelligence, computer science, and statistics to develop the GeoRDyn toolkit to allow researchers to identify the spatial and relational complexities in more fine-grained data. This will permit new approaches to investigating longstanding questions in the field, such as if and how networks emerge and influence collective action and the conditions under which events diffuse across time and space. Events are often observed at different temporal and spatial scales, involving complex interactions among actors at the micro, macro, and meso levels. Methodological tools commonly used in the study of collective-action events are often less dynamic and/or do not account for complex dependencies. This project develops the GeoRDyn toolkit to help researchers dissect the spatial and relational complexities in events and draw rigorous causal inferences. The toolkit incorporates recent advances in artificial intelligence (AI), computer science, and statistics. The project develops new methods and software for studying complex and dynamic processes, such as diffusion, that can be used by researchers as well as those outside of the academic community who need to understand the causes and potential implications of events. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

Nontechnical Summary The SPARK project aims to address a critical need for new light sources to advance quantum information science by developing materials and tools for the generation of single photons from atomic-scale defects in ultra-thin semiconductors. These single photons serve as the basic units of information in next-generation quantum computing and secure communications. The activity is a collaboration between Penn State University and the Swiss Federal Laboratories for Materials Science and Technology. It combines U.S. and international expertise in materials synthesis, microscopy, and quantum optics to examine how atoms embedded in two-dimensional semiconductors emit light. Understanding and controlling this process will help build new types of devices for communication and computing that are faster, more secure, and more energy-efficient than those used today. The project includes hands-on education and training for undergraduate and graduate students. A central education component is the creation of a new initiative called the Semiconductor Training and Research Initiatives for Veterans in Engineering - STRIVE. This program aims to provides U.S. military veterans with training in semiconductor research and manufacturing aligned with national workforce and security needs. Additional educational activities include specialized coursework, scientific writing workshops, and international exchange opportunities. These efforts ensure that students not only gain technical expertise but also develop communication and collaboration skills required in today’s global research environment. By integrating research and education, this project promotes innovation, international partnership, and workforce development in the rapidly growing field of quantum materials and technologies. Technical Summary The research investigates electrically controlled sources of single photons in atomic-scale materials known as two-dimensional semiconductors. The principal investigators aim to understand how defect complexes—such as missing atoms or foreign atoms embedded in the material—can act as stable and tunable quantum light emitters. The team studies monolayer transition metal dichalcogenides, a class of materials only a few atoms thick, combined with atomically thin metallic layers that act as optical cavities to enhance photon emission. These hybrid semiconductor/metal structures are engineered with embedded atomic defects that can be addressed electrically and optically. The project uses a combination of advanced synthesis techniques, atomic-resolution scanning probe microscopy, ultrafast tunneling spectroscopy, and in situ optical characterization. The research team maps the relationship between the atomic structure of defects and their electronic and optical behavior at the single-photon level. A key focus is on understanding how energy is transferred between the semiconductor and metal interfaces and how this interaction affects the brightness, timing, and direction of emitted photons. The activity directly supports the goals of the Electronic and Photonic Materials program by advancing the fundamental understanding of light-matter interactions at the atomic scale and enabling the development of scalable platforms for quantum light generation. These insights lay the foundation for future applications in quantum communication, computing, and sensing. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

NON-TECHNICAL SUMMARY This award supports theoretical and computational research, and education to advance fundamental understanding of particles in materials that lie between light and matter. One of the most striking insights from the quantum mechanics is that all fundamental particles in nature fall into two categories: fermions and bosons. This distinction underpins much of modern technology. For example, the fact that electrons behave as fermions is what enables the operation of electronic devices, from smartphones to supercomputers. In contrast, photons, namely the particles of light, are bosons, which is what makes lasers possible. In recent decades, a surprising new chapter has emerged, namely anyons, which are particles that are neither fermions nor bosons, but something in between. First proposed as a theoretical curiosity, anyons have recently been confirmed in several ingenious experiments in a system known as the fractional quantum Hall effect, obtained when two-dimensional electrons are placed in a magnetic field. Moreover, these exotic particles may hold the key to building fault-tolerant quantum computers. This project aims to deepen the understanding of anyons and their unique properties. The Principal Investigator and his students have already developed accurate and reliable theoretical tools to study the fractional quantum Hall effect. With support from this grant, they will explore how different types of anyons behave—examining their spatial profiles, how they tunnel through barriers, how they cluster into molecules, how they reveal themselves in photoluminescence experiments, and how they may appear without a magnetic field in certain twisted bilayer systems. TECHNICAL SUMMARY This award supports theoretical and computational research, and education to advance fundamental understanding of anyons. Two-dimensional systems—such as quantum wells, graphene monolayers and multilayers, topological insulators, and twisted bilayers—form a vibrant subfield of condensed matter physics. Many of the central ideas in this field trace back to the phenomena of the integer and fractional quantum Hall effects, which revealed new states of matter governed by emergent particles. A striking feature of the fractional quantum Hall effect is the emergence of anyons—quasiparticles that obey fractional statistics. These are of deep theoretical interest and are also central to certain ideas for fault-tolerant topological quantum computation. Recent discoveries of fractional quantum Hall-like states at zero magnetic field in twisted semiconductor and graphene multilayers mark a major advance, potentially enabling more accessible platforms for realizing anyons. This project aims to undertake a quantitative theoretical study of the properties of anyons in these systems. The goals include: (i) Molecular anyons: The PI and his team will explore the possibility of whether the abelian and non-Abelian anyons tend to cluster together to form molecules, and if so, how the charge of the molecule depends on the interaction between the electrons. Such molecular anyons may explain the anomalous charge measured in shot noise experiments and will have many implications for future experiments. (ii) Photoluminescence Theory: A framework will be developed to assess whether photoluminescence studies of fractional quantum Hall effect can serve as a probe of anyonic excitations and their bound states. (iii) Composite Fermion Pairing: Non-Abelian anyons are excitations of paired states of composite fermions. PI and his collaborators have shown that pairing of composite fermions can be induced by going to a higher Landau level, by increasing the width, or by enhancing Landau level mixing. They will investigate whether pairing of composite fermions can also be induced by proximity coupling to a superconductor or by screening the interaction by nearby metallic layer. Implementation of a Bardeen-Cooper-Schrieffer type superconductivity of composite fermions in the spherical geometry will be sought. (iv) Wave Functions for Twisted Bilayers: Variational wave functions for composite fermions will be constructed to describe fractional quantum Hall states in twisted bilayer systems. The project will use a combination of exact diagonalization, variational methods, quantum Monte Carlo (including fixed-phase diffusion Monte Carlo), and field theoretical and mean-field approaches. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

This project concerns the mathematical analysis and also the computation of certain phenomena in fluid mechanics and elasticity that are modeled using so-called partial differential equations. The aim of the project is both to advance a scientific understanding of important physical phenomena, as well as to make quantitative predictions whenever possible. The problems under study are motivated by real-life applications with potential societal benefits. In the first part of the project, the Principal Investigator studies the interaction between incompressible fluid flows and walls, focusing on two problems. The first problem pertains to the motion of inviscid fluids in containers with permeable walls that allow for injection and suction, and how the rate and direction of injection and suction affects the flow. This problem has many applications from the study of fluids in sections of pipelines, to modeling of underground wells. The second problem focuses on a simplified model of the Earth, consisting in a fluid-filled solid shell, representing the Earth’s crust and mantle, containing a solid core, and on the long-time combined motion of the fluid-solid system. In the second part of the project, the Principal Investigator examines the effects of material transport and diffusion in certain fluid models. Two problems are considered. The first problem concerns flame front propagation in combustion and phase separation in fluid mixtures, for instance binary alloys. The second problem concerns modeling of fluid planets, such as certain exoplanets, combining the effects of rotation, gravitation, convection and magnetism. The last part of the project focuses on elastic materials, more specifically problems stemming from applications in seismology, where the Earth’s crust is modeled as an elastic solid. Remote monitoring of buried faults and magma chambers from satellite data during quiescent periods with no detectable seismic activity is investigated. The project includes also training activities for both graduate and undergraduate students. The focus of this proposal is to study various problems characterized by singularities and ill-posedness of partial differential equations modeling the behavior of incompressible fluids and elastic solids. These problems are tackled using primarily analytic techniques, but also computational tools. The goal of the project is to make progress in our understanding of important processes, such as phase separation in mixtures, and make quantitative predictions, such as the reconstruction of unknown parameters in inverse problems. The project consists in three main parts: I. Wall interactions in hydrodynamics models; II. Transport, dissipation, and their interplay in non-linear systems; III. Boundary inverse problems for geophysical models. The project addresses questions motivated by fundamental physical phenomena. Hence, the project is both timely and relevant. Part I concerns the behavior of incompressible flows near rigid walls. Injection and suction can stabilize boundary layers. Fluid-structure interaction is mediated by boundary forces. Part II concerns the combined effect of advection and dissipation in non-linear systems. Long-wave instability can be mitigated by dissipation. Strong advection can accelerate phase separation. Convection, dissipation, and geodynamo all contribute to planetary motion. Part III concerns remote probing of geophysical systems. Fault monitoring is key to predicting earthquake nucleation. In volcanoes, the magma chamber is often inaccessible. The three parts of the project are distinct, but connected in a cohesive research program by the underlying themes of singularity and ill-posedness.This research brings about challenging mathematical questions, which requires novel ideas and techniques: singularities in various forms, from non-smooth domains, to non-standard interface and boundary conditions, to ill-posedness and instability, for instance, permeate the entire project. Several of the problems under study, such as shape optimization in Part III, have a computational component that is also addressed. Progress on these problems is likely to have impact in other fields. The project offers training opportunities for both graduate and undergraduate students, as well as dissemination to the broader scientific community and outreach to society. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

This REU targets critical advanced technologies including aviation, marine propulsion, and power generation, while simultaneously developing a robust US-based workforce with skills to advance their careers. Addressing challenges in these technologies requires both deep fundamental research and strong connections with industry, which are hallmarks of Penn State’s approach to research. This REU site renewal builds off the successes of a prior site (grant EEC-2149667) that has a proven track record of engaging students in high-impact research, meaningful interactions with industry professionals, and developing career skills, all while doing rigorous human-subjects research on the impact that REU experiences can have on students’ attitudes towards research. This project will provide 10 REU students with a unique and enriching experience that encompasses research, professional development, and networking. A student cohort from a range of schools will spend 10 weeks at Penn State. A number of research-related activities, including lab and facility tours, industry workshops, and a research symposium at the Site's industry partner, Pratt and Whitney, will provide students with a broader view of research beyond their own individual project. Students will also participate in workshops on technical communication, applying for jobs in industry or graduate school, and fellowship applications to provide students with the skills to transfer their research experiences to career outcomes. Mentoring from faculty, near-peer graduate students, and industry professionals will provide REU students with a strong support network both during and after the summer experience. The site will provide each student with compelling research on topics that range from fundamental to applied in areas such as fluid mechanics, material science, heat transfer, manufacturing, combustion, sensor development, and high-performance computations. Proposed projects include high-impact topics like design for additive manufacturing, infrared thermal imaging for efficiency improvements, advanced design of grid-level battery storage, and use of alternative fuels for grid resilience and fuel security. Because the project involves faculty with a range of expertise and laboratories, students will use the latest research tools such as metal additive manufacturing, machine learning, high-speed diagnostics, and more. The success of the REU site will be assessed by experts in engineering education using qualitative and quantitative formative, summative, and longitudinal research methods, which correspond with a rigorous educational research plan to meet critical gaps in engineering education literature. In addition to ensuring the quality of the site, the team will also seek to translate what they learn to the scholarly community more broadly. This evaluation method, based in rigorous engineering education theory, will build on the team's prior research to provide data to both the REU site and in the open literature, improving REU outcomes for all NSF REU sites. Overall, this REU site will provide a cohort of students with a unique and impactful summer experience while informing them about the power and propulsion industries. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

This award provides funding for participants in “Semi-annual Workshop in Dynamical Systems and Related Topics at Penn State,” to be held on the Penn State University Park campus November 13-16, 2025. This represents the Penn State half of the Semi-annual Workshop in Dynamical Systems and Related Topics, cosponsored for the last 35 years by the dynamics groups of Penn State and University of Maryland; the Maryland half is held each spring. The workshop is devoted to recent developments in dynamical systems and many applications to related fields. The goals of this conference are to promote the communication of mathematical results, to facilitate interaction and progress in dynamical systems and related fields, to nurture the sense of community and common mission in these fields, and to contribute to the training of graduate students and recent Ph.D. recipients and to their integration into the dynamics community. This conference series began as a meeting place for mathematicians based primarily in the northeastern United States working in dynamical systems and related topics, and during the past twenty years has grown to attract outstanding mathematicians from around the world. In recent years the conference has featured several special sessions focused on the latest achievements and current research trends in dynamics and its applications. The themes for the 2025 edition of the workshop include classification of measures for homogeneous and non-homogeneous systems, ergodic and mixing properties of parabolic flows and sharp bounds on their deviation of ergodic averages, new developments for random systems, recent trends in the theory of Hamiltonian dynamics, statistical properties of non-uniformly hyperbolic systems and their thermodynamic formalism. More information can be found on the conference website at https://science.psu.edu/math/research/dynsys/workshop. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

The BRIDGE-TECH (Building Relationships in Innovation and Development for Growth in Emerging Technologies) project brings together a cohort of five primarily undergraduate institutions (PUIs) committed to strengthening regional innovation ecosystems by removing barriers to research, workforce development, and technology commercialization. These institutions each serve regions with untapped economic and technological potential but face common challenges such as limited research infrastructure, fragmented partnerships, and underdeveloped support for technology transfer. Through a coordinated, cohort-based approach, the project will empower faculty and communities by developing sustainable industry partnerships, supporting entrepreneurial programming, and expanding access to emerging technologies. Ultimately, BRIDGE-TECH seeks to build ecosystems where innovation thrives and contributes to long-term economic benefits. Technically, BRIDGE-TECH proposes a scalable model to increase innovation capacity at PUIs through strategic capacity-building and shared institutional development. The project focuses on five interrelated goals: (1) increase industry-sponsored research, (2) enhance innovation culture and technology transfer, (3) develop design thinking and innovation accelerator, (4) drive and expand innovation in rapid manufacturing, and (5) create synergistic opportunities for academic research. Each institution will implement customized, locally grounded activities, such as creating innovation centers, forming regional councils, conducting ecosystem assessments, and launching faculty fellows programs, while leveraging cohort-wide training workshops and shared resources. The anticipated outcomes include increased research funding, expanded partnerships with industry entities, improved IP and commercialization processes, and the training of over 40 faculty members. The collaboration ultimately aims to elevate the role of PUIs in national innovation networks while supporting sustainable economic development. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

Bacteria commonly swim through complex biological fluids like mucus, playing a crucial role in health and disease, from infections in the lungs to microbial imbalances in the gut. Understanding how bacteria move through biological fluids is the first step toward developing new ways to cure and prevent such infections. Many mathematical tools describing how microorganisms swim through fluids like water were developed in the 1950s-1970s. These foundational theories continue to be used today. However, mucus is a far more complex and challenging environment than water. It is composed of macromolecular proteins (mucins) that confer it viscoelastic properties, simultaneously flowing like a fluid, yet capable of recoiling like elastic solids. Mathematical tools for studying bacterial locomotion through such complex biological fluids are lacking. This research will combine mathematics, computer simulations, and laboratory experiments to create a more comprehensive picture of this process. It will first investigate the fluid mechanics of propulsion through complex fluids using a single bacterial flagellum. This will be followed by a study of how multiple flagella bundle together, a standard feature of many bacteria like E. coli. Finally, the collective behavior of large groups of bacteria in fluids like mucus will be investigated. Knowledge so gained will be instructive in the design of new medicines, the prevention of dangerous infections of mucosal surfaces, and in the management of stubborn biofilms. The research focuses on bacterial flagellar propulsion in mucus, and in a better-controlled anisotropic, viscoelastic fluid: a lyotropic liquid crystal (LC). Using mathematical modeling and analysis, numerical simulations, and experiments, this project will address three interconnected problems. First, a novel slender body theory will be derived from first principles, alongside controlled experiments, to quantify the forces, flow fields, and resulting dynamics of individual bacterial flagella within a nematic LC environment. Theories will be tested against full numerical simulations of Ericksen-Leslie and Beris-Edwards model LC fluids. The first aim will be extended to encompass the coordinated behavior of multiple flagella forming helical bundles, a key aspect of bacterial locomotion. Finally, the emergent behavior and dynamics of many bacteria interacting within LCs will be modeled and analyzed, bridging the gap between individual flagellar mechanics and population-level phenomena. The expected outcomes include significant advances in our understanding of general fluid-structure interactions in complex biological media. The mathematical machinery developed will be applicable to a wide range of nearby problems in biology and engineering and will illuminate new mechanical aspects of evolutionary biology. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

Lightning is the most well-known component of the electrical nature of Earth’s atmosphere, but other electrical processes are critical to the composition of the atmosphere. Corona discharges are significant contributors to the development of the hydroxyl radical, which is a compound that reacts with many pollutants and has an important role in eliminating atmospheric methane and ozone. This study will investigate the role of turbulence in the development of corona discharges inside of clouds by performing controlled laboratory experiments. Beyond the implications for lightning and pollution, corona discharges are impactful in industrial manufacturing and relevant for future planetary science missions. The research team plans to test the key hypothesis that turbulence brings inertial particles carrying opposite charges together and causes significant local field enhancement, which exceeds the breakdown limit and initiates both subvisible corona and visible discharges. The primary mechanism to address this hypothesis is through controlled laboratory experiments using a Homogeneous and Isotropic Turbulence (HIT) chamber. The chamber is a cubic box with a side length of 1 meter and made of transparent acrylic panels allowing for the use of optical instruments. The research team will be able to control the turbulence, external electric field, and the droplet concentration, charges, and diameter. Measurements will include particle motion by particle image velocimetry, particle electric fields by field mill sensors, and discharge location and intensity by ultraviolet (UV) cameras. The data will be used to answer questions about how the multiscale electric fields develop and intensify in particle-laden turbulence and what the intrinsic connection is between the fluctuating electric field and the occurrence of visible and subvisible discharges. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

This project develops a new framework for digital twin modeling of Alzheimer’s disease (AD), combining clinical data, biomedical research, and advanced computational methods to support personalized medicine. A digital twin is a computational replica of an individual’s health state, enabling the prediction of disease progression and the evaluation of treatment options tailored to the patient. The project contributes to national efforts in healthcare innovation by addressing the urgent need for a better understanding, prediction, and treatment of Alzheimer’s disease, which affects millions of Americans. This work also advances the broader field of personalized medicine by demonstrating how digital twin tools, powered by large language models, machine learning, and causal inference, can accelerate discovery and improve health outcomes. In addition, the project supports interdisciplinary collaboration across artificial intelligence, mathematics, and medicine, while offering new training opportunities for students in data science, modeling, and biomedical research. This project builds a unified modeling framework for population-based and personalized digital twins of AD. The approach uses large language models (LLMs) to extract causal networks of AD biomarkers from scientific literature and combines this with clinical data to generate personalized predictions. Conformal prediction techniques are applied to quantify uncertainty in model outputs, and optimization under limited data is achieved by integrating gradient-based learning with LLM-guided parameter search. The digital twin models simulate disease trajectories and support digital clinical trials. Treatment planning is formulated as a Markov Decision Process and solved using deep reinforcement learning to identify optimal, individualized therapeutic strategies. The framework integrates causal modeling, machine learning, generative AI, and decision theory, advancing both the science of Alzheimer’s disease and the computational tools for biomedical digital twin development. While centered on AD, the methods are generalizable and contribute broadly to AI-enabled modeling under data constraints in biomedical research. This award by the Division of Mathematical Sciences in the Mathematical and Physical Sciences Directorate is jointly supported by the Office of Advanced Cyberinfrastructure in the Computer and Information Science and Engineering Directorate. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

Training methods in manufacturing have not kept pace with advances in how manufacturing processes are designed. Instead of teaching from a static handbook of unchanging material properties, novice scientists and engineers can now use AI-infused data science to understand and predict the complex range of properties and performance of materials as they are processed. In addition, traditional training does not include environmental, economic, and social sustainability considerations – from sourcing through processing to recycling, reuse, or disposal as part of materials selection or design although these aspects are increasingly demanded by end users. This skills gap necessitates a new educational paradigm in which data science is natively integrated into sustainable materials and process design, enabling consideration of the full life cycle of materials while accelerating their conceptualization and discovery. This National Science Foundation Research Traineeship (NRT) award to the Pennsylvania State University will equip the next generation of engineers, physical scientists, and social scientists with the tools required to effect transformative change in sustainable materials processing. The project (Sus-Mat for short) anticipates training 50 Ph.D. students, including 23 funded trainees, from Materials Science and Engineering, Chemical Engineering, Civil and Environmental Engineering, Computer Science and Engineering, Architecture, and Public Policy. This NRT will merge essential but commonly siloed fields of sustainability, data science, advanced materials processing, and public policy to create a holistic, data-driven materials and process design ecosystem. Trainees will learn to harness flexible data science tools including artificial intelligence (AI) integration, enabling them to understand how emergent processing approaches impact material properties and sustainability metrics and then employ those relationships to design sustainable materials and processes. The project will integrate the Sus-Mat themes of sustainability, data science, advanced materials processing, and public policy in pursuit of three core interdisciplinary research themes: (1) active learning for advanced materials processing optimization; (2) generative AI-based models for materials design; and (3) materials sustainability assessment framework. The interdisciplinary research will be enabled by the traineeship ecosystem consisting of: new core courses for foundational training, micro-credentials bolstered by experiential training, internships to facilitate knowledge translation, cohort-building activities to aid retention and community building, convergent research facilitated by co-advising, capstone experiences for broader outreach, and professional development that trains policy-savvy leaders in sustainable materials and process design. Sus-Mat’s combination of research projects using a range of materials, data availability, and processing technologies and skills training informed by public policy will modernize STEM workforce training in emergent materials processing technologies and accelerate the adoption of innovative, sustainable methodologies in high-tech domestic manufacturing with locally sourced materials. The NSF Research Traineeship (NRT) Program is designed to encourage the development and implementation of bold, new, and potentially transformative models for STEM graduate education training. The program is dedicated to effective training of STEM graduate students in high-priority interdisciplinary or convergent research areas through comprehensive traineeship models that are innovative, evidence-based, and aligned with changing workforce and research needs. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.