Argonne Applies AI to Speed Chemical Analysis at DOE’s Advanced Photon Source
AI-guided approach takes over manual steps in powerful X-ray technique
March 2, 2026 — Artificial intelligence (AI) is transforming nearly every branch of science. And researchers at the U.S. Department of Energy’s (DOE) Argonne National Laboratory are helping lead the way.
Artistic rendering shows new AI-guided approach capturing absorption edge from atomic structure of material analyzed by XANES at a light source. Credit: Argonne National Laboratory.
“There is a lot of hype around AI today in the media,” said Mathew Cherukara, a computational scientist and group leader at Argonne’s Advanced Photon Source (APS), a DOE Office of Science user facility. “Yet there is no question that AI can help researchers at APS and other light sources make breakthroughs in advanced chemical processes critical to American industry.”
As proof, the Argonne team has developed an AI-guided method that dramatically speeds up a widely used X-ray technique known as X-ray absorption near-edge structure (XANES) spectroscopy. It does so with far less risk of human error or damage to the sample from the X-ray beams.
This powerful analytical tool reveals the hidden chemistry inside materials important to modern life, such as batteries, catalysts and materials through which electricity flows without resistance. The team’s AI approach cuts the number of measurements previously needed by as much as 80%, with no loss of accuracy. The result is a dramatic shortening of data acquisition duration, allowing researchers to capture fast chemical changes in real time.
Here’s how XANES works: Scientists shine X-ray beams with increasing energy onto a material. Each X-ray beam is a tiny packet of energy. When the energy is high enough to knock a tightly bound electron out of an atom, the material suddenly absorbs more X-rays. This sharp jump in absorption is called the absorption edge.
By tracking how X-ray absorption changes before, during and after this edge, researchers can watch the chemistry of a specific element unfold within a material, from how a metallic catalyst reacts with other chemicals to how the charge state of a battery element changes during cycling.
“XANES is incredibly powerful, but until now, scientists had to make dozens or even hundreds of choices about where to measure and how long to measure at each X-ray energy level,” said Shelly Kelly, an APS physicist and group leader.
Some regions of X-ray energy are rich with chemical information, calling for numerous measurements. Other parts are not, meaning far fewer measurements are needed. “It is often not easy for experimenters to set the optimal number of measurements to make in a given energy region,” Kelly said. “AI is helping us take the guesswork out of XANES.”
The team’s new approach replaces the manual measurement process with an AI algorithm that automatically selects the most useful measurement points. The algorithm identifies where the absorption edge is likely to occur, which regions hold the most chemical detail and which regions offer little added information.
“Our AI method measures only where needed,” said Ming Du, a computational scientist and lead author on the paper. “It’s smarter, faster and more efficient, and it lets researchers focus on the big picture.”
The system also enables something new: AI-directed experiments. By comparing a sample’s evolving spectrum against known starting and ending states (for example, a fully charged electrode versus fully discharged), the AI can tell researchers in real time the state of the chemical progress, when enough information has been collected, and when it’s time to move on.
“It’s not just speeding up the measurement,” Kelly said. “It’s making decisions during the experiment — decisions a human used to make.”
The work points toward a future in which X-ray beamlines, such as those at the APS, are more autonomous and better able to track complex chemical reactions as they happen.
“This brings us closer to intelligent X-ray stations that make the most of every photon,” Cherukara said. “Argonne plans to continue developing AI-driven tools for next-generation X-ray science, especially as the upgraded APS delivers beams up to 500 times brighter than before.”
The team demonstrated the method using beamlines 25-ID-C, 20-BM and 10-ID at the APS. The project was supported by the DOE Office of Science, Office of Basic Energy Sciences.
The research first appeared in npj Computational Materials. In addition to Du, Kelly and Cherukara, contributors include Mark Wolfman and Chengjun Sun.
Source: Joseph E. Harmon, Argonne National Laboratory
Related
Argonne Applies AI to Speed Chemical Analysis at DOE’s Advanced Photon Source
AI-guided approach takes over manual steps in powerful X-ray technique March 2, 2026 — Artificial…
OpenAI and Amazon Announce Strategic Partnership
SAN FRANCISCO and SEATTLE, Feb. 27, 2026 — OpenAI and Amazon today announced a multi-year strategic partnership to accelerate…
SDSC Powers AI Model to Improve Prostate Cancer Care
Feb. 27, 2026 — Researchers and clinicians from six leading medical centers and academic institutions…
Lilly Launches LillyPod NVIDIA DGX SuperPOD for Genomics and Drug Discovery AI
Built with over 1,000 NVIDIA Blackwell Ultra GPUs, LillyPod is now online to power scientific…
Domino Introduces Fastest, Safest Path to Scale Enterprise Agentic AI Systems
SAN FRANCISCO, Feb. 27, 2026 — Domino Data Lab has announced a major new platform…
CCC Report Examines How AI and Citizen Science Can Advance Research Together
Feb. 26, 2026 — In a new workshop report published by the Computing Community Consortium…
