NSF PCL-Test Bed: ATHENA: Advanced Testbed for High-throughput Experimentation in Nano- and Atomic Science

Award ID: 2607469

NSF PCL-Test Bed: ATHENA: Advanced Testbed for High-throughput Experimentation in Nano- and Atomic Science Imagine a future where novel materials for abundant energy, resilient infrastructure, and quantum computing are discovered and deployed as quickly as software updates. Today, development of new materials is a slow, expensive, and manual process, which significantly hinders innovation and advancement of nearly every industry. ATHENA will disrupt this status quo by creating the experimental backbone for the next industrial revolution. ATHENAs acceleration of advanced materials development will unlock unprecedented innovations in energy production, manufacturing, transportation, and national security. This research hub will leverage AI to transition to online use of advanced scientific instruments so that the capacity to create, build, and innovate material research is no longer bound by geography or institution. As a national platform, ATHENA will work with other NSF PCL nodes to develop standards, train users, and scale access to next-generation tools for experimental science. Thevision of ATHENA is to go small and fast in order to go big, by overcoming the materials characterization bottleneck at atomic and nanoscales. This approach will transform the way materials are discovered and utilized. Electron and scanning probe microscopes will operate with greater autonomy, boosting performance to the physical limits of current engineered systems and will be paired with a matching sample preparation pipeline. Operationally, ATHENA will build bottom-up autonomy that enables these instruments to plan, execute, and refine experiments from local single-task optimization toward long-horizon scientific goals. The approach begins with automated, fixed-policy, and reward-based experimental primitives that enable robust, high-throughput measurements at the nanoscale. This autonomy is scaled using agentic workflows, i.e., decision-making systems that leverage probabilistic planning, reward functions that are predictive of material functionality, and digital twins that anticipate instrument and sample responses. These are orchestrated by theory-in-the-loop architecture, allowing simulations to shape experimental decisions and enable multi-instrument rollouts across platforms. The ATHENA scientific focus will be threefold. For solution-processable materials, ATHENA will learn generalizable, physically consistent heuristicsfor energy-relevant materials prediction spanning atomic bonding through mesoscale defect structures, grounded in first principles, and accessible through efficient small-volume random libraries. For refractory materials, ATHENA will unravel complex phase transformation pathways and defect networksvia in-situ observation of multicomponent material libraries, connecting microstructure evolution to emergent mechanical properties. For atomically precise fabrication, the approach will pioneer autonomous approaches to build and characterize quantum and low-dimensional systems atom-by-atom. These goals, unified by a machine learningenabled infrastructure, position ATHENA as a national accelerator node for predictive, interpretable, and scalable experimental science. 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 Program Director: Waleed Nasser