PEESE researchers have published several impactful articles in Science Advances, contributing valuable insights into AI for science, energy systems, sustainability, and materials discovery.
Agentic AI for Electron Microscopy and Materials Discovery
Our 2026 paper, “Bridging Electron Microscopy and Materials Analysis with an Autonomous Agentic Platform,” introduces EMSeek, a modular, provenance-tracked multi-agent AI platform that turns raw electron microscopy (EM) data into actionable materials insight in minutes rather than weeks. Conventional EM workflows remain fragmented across segmentation, crystallographic reconstruction, property prediction, and literature review, often requiring thousands of manual actions. EMSeek unifies these steps through five specialized agents, covering reference-guided one-for-all atom/defect segmentation, mask-aware crystal-structure reconstruction, a gated mixture-of-experts property predictor with uncertainty calibration, literature retrieval with citation anchoring, and physical-consistency checks with audit-ready reporting, all coordinated by a central planner. Evaluated on 20 material systems and five canonical tasks, EMSeek completes an end-to-end analysis in just 2–5 minutes per image, roughly 50× faster than expert workflows, while maintaining transparency, reproducibility, and scientific rigor. Case studies on two-dimensional lattices and nanoparticles demonstrate how the platform can accelerate the observe–model–screen loop across catalysis, energy storage, and semiconductor research.
Cornell Chronicle: AI turns electron microscopy into materials insights in minutes
- Also featured in Phys.org, etc.
AI-Enabled Liquid Crystal Interfaces for Microplastics Characterization
In collaboration with Professor Nicholas L. Abbott’s team, our paper “Liquid Crystal–Driven Interfacial Ordering of Colloidal Microplastics: Advancing Microplastic Characterization Below the Macroscale” presents a new AI-enabled approach for characterizing environmentally relevant microplastics at the colloidal scale. The study shows that microplastics self-organize at liquid crystal-aqueous interfaces into distinct patterns that reflect material composition, concentration, environmental aging, and the presence of natural organic matter. By combining liquid crystal-templated self-assembly with computer vision and deep learning, the work advances AI for materials approaches to accessible and scalable microplastics analysis.
AI-Driven Design for Solid-State Batteries
Our recent article, “Toward AI ecosystems for electrolyte and interface engineering in solid-state batteries,” provides a field-shaping review on how artificial intelligence can accelerate the discovery of solid electrolytes and the design of stable interfaces in next-generation solid-state batteries. The paper highlights advances in screening, simulation, and generative design, offering a vision for an integrated AI ecosystem to accelerate future battery breakthroughs.
Molecular Design and Microplastic Mitigation
Our 2024 paper, “Designing Microplastic-Binding Peptides with a Variational Quantum Circuit-Based Hybrid Quantum-Classical Approach,” highlights a collaboration with Professor Carol Hall. This study introduces an advanced AI method for identifying peptides that effectively bind to microplastics using a hybrid quantum-classical algorithm, paving the way for innovative solutions to mitigate microplastic pollution.
- Cornell Chronicle: AI, quantum drive discovery of peptides for microplastic cleanup
Reshoring EV Battery Manufacturing and Climate Goals
The 2023 paper, “Will reshoring manufacturing of advanced electric vehicle battery support renewable energy transition and climate targets?” examines the potential policy impacts of bringing EV battery production back to domestic soil. This research has garnered attention in prominent media outlets:
- Cornell Chronicle: Study unveils policy insights for reshoring EV battery production
- Also featured in TechXplore, News Azi, etc.
Sustainability of Retired EV Batteries
The 2021 article, “Second life and recycling: Energy and environmental sustainability perspectives for high-performance lithium-ion batteries,” presents collaborative work with Professors Lynden Archer and Chris Rahn. The study explores innovative strategies to extend the life and enhance the sustainability of retired EV batteries. Media coverage includes:
- The Hill: Cornell researchers propose new solutions to make retired EV batteries more sustainable
- National Science Foundation (NSF) Research News: Retired electric vehicle batteries could be used to store renewable energy
- Popular Science: The race to close the EV battery recycling loop
- ScienceDaily: That new EV battery will be a headache to recycle: These solutions can help
- TechXplore: Study provides keys to managing influx of EV batteries
- Mirage: Retired electric vehicle batteries could be used to store renewable energy
- TechnologyNetwork: Addressing the Influx of Spent EV Batteries
- AZoMaterials: Reusing and Recycling Li-Ion Batteries
Metal-Organic Framework Discovery for CO2 Capture
PEESE researchers co-authored the 2016 paper, “In silico discovery of metal-organic frameworks for precombustion CO2 capture using a genetic algorithm.” This collaboration included Professor J. Fraser Stoddart, who was awarded the 2016 Nobel Prize in Chemistry just a week before the paper’s online publication.
