PNNL has developed a next-generation electrical resistivity tomography system for DOE that uses E4D software and AI-enhanced modeling to produce real-time subsurface images that help guide environmental remediation decisions.

Replacing commercial acid with acidic waste enables researchers to improve nickel extraction efficiency, lower projected costs, and improve process economics.

Localized gradients in magnetic fields have long-range effects on the concentration of rare earth ions in solution, facilitating field-driven extraction of critical minerals.

Studying a series of rare earth element compounds revealed differences in their magnetic behavior that will be leveraged for separations.

The Scientific Discovery through Advanced Computing institute will focus on making scientific machine learning accessible to domain scientists.

This achievement earned DIMPLES a Best Paper award at the ACM International Conference on Supercomputing 2025.

Nanoscale domains of magnetically susceptible critical materials encounter enhanced magnetic interactions under external magnetic fields, providing a promising new avenue for separations.

The size of graphene oxide sheets affects the ability of assembled membranes and adsorbents to separate rare earth elements.

A PNNL team has developed an energy- and chemical-efficient method of separating valuable critical minerals from dissolved solutions of rare earth element magnets.

Surface manganese mining from nodules in South Dakota is now possible with a new scalable selective precipitation process.

Cobalt substitution for magnesium in a model mafic mineral dramatically slows reaction with carbon dioxide to form a cobalt-doped carbonate.

Precipitation with sodium hydroxide in a flow-gel device enables the rapid extraction of magnesium from seawater in a new scalable process.

Klickitat Valley Health's fuel cell and microgrid project boosts rural healthcare, enhancing energy resilience and sustainability for social equity.

In a recent publication in Nature Communications, a team of researchers presents a mathematical theory to address the challenge of barren plateaus in quantum machine learning.

The atomic-level distribution of nickel and cobalt in forsterite depends on the concentration of the different elements.

Data from SOSAT web tool informs site selection and management for the subsurface injection of captured carbon dioxide.

To improve our ability to “see” into the subsurface, scientists need to understand how different mineral surfaces respond to electrical signals at the molecular scale.

The surface oxygen functionality of graphene oxide may be tuned using ultraviolet light, affecting how differently charged ions move through the material.

A new approach validates quantum algorithms for low-energy nuclear physics applications

PNNL played host in mid-May to the Artificial Intelligence for Robust Engineering & Science workshop, an annual event that explores advances in artificial intelligence