PNNL

Scientists Are Speeding Up the Development of “Rad-Hard” Materials for Satellites, Nuclear Reactors

Scientists are using agentic agents to speed up the process of toughening materials for their future in space or in nuclear reactors.

Projects

This research initiative aims to transform the nuclear forensics analysis process through microanalysis, real-time process optimization, and machine learning

Leveraging Viscous Hamilton-Jacobi PDEs for Uncertainty Quantification in Scientific Machine Learning

Toward Predictive Speciation in Complex Electrolyte Solutions

Scientists uncovered the molecular basis of nitric acid speciation across dilute and concentrated solutions and identified competing hydration-shell rearrangements around lanthanide ions at the dilute limit.

Evaluation of Artificial Intelligence (AI) and Machine Learning (ML) Approaches for Bill-of-Material (BOM) Analysis

Hybrid Decision and Control Methods for Multi-Scale Grid Resilience and Oscillation Damping

A machine learning approach to generate residual stress distributions using sparse characterization data in friction-stir processed parts

An AI Agent for Physics Could Speed Up Quantum Computing and Dark Matter Searches

An AI agent that can auto-tune an amplifier could speed up quantum computing development and dark matter discovery. Researchers at PNNL are working to find out with a Genesis Mission–funded project.

Publications

From the Periodic Table to New Nuclear Energy Materials

Nuclear reactors need tough alloys that can withstand extreme conditions. PNNL researchers developed a computational approach called robust alloy design (RAD) to search for materials that meet multiple nuclear requirements.