Artificial intelligence is reshaping X-ray spectroscopy workflows at major user facilities. Researchers at Argonne National Laboratory have developed an AI-guided method for X-ray absorption near-edge structure (XANES) spectroscopy that reduces the number of measurements required by as much as 80 percent while maintaining accuracy. The approach automates key experimental decisions traditionally made by beamline scientists, reducing human error and shortening data acquisition time.
X-ray spectroscopy is a family of analytical techniques used to probe elemental composition and chemical state by measuring how materials absorb X-rays at varying energies. XANES spectroscopy, a widely used form of X-ray absorption spectroscopy, focuses on the region near the absorption edge—where a sharp increase in absorption occurs as tightly bound core electrons are ejected from atoms.
How AI changes XANES spectroscopy workflows
In conventional XANES spectroscopy, scientists must manually decide where to measure along the X-ray energy range and how long to dwell at each point. Regions near the absorption edge contain dense chemical information and often require fine energy steps and longer acquisition times. Other regions contribute less insight.
“It is often not easy for experimenters to set the optimal number of measurements to make in a given energy region,” said Shelly Kelly, an APS physicist and group leader. “AI is helping us take the guesswork out of XANES.”
The Argonne team’s AI-driven algorithm replaces manual selection of measurement points. The system identifies where the absorption edge is likely to occur, determines which regions contain the most chemical information, and minimizes redundant measurements.
“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.”
By reducing unnecessary scans, the method cuts total measurements by up to 80 percent, enabling faster data acquisition and improved temporal resolution.
Enabling AI-directed experiments at beamlines
Beyond efficiency gains, the AI system supports real-time decision-making during XANES spectroscopy experiments.
By comparing evolving spectra against known reference states—for example, fully charged versus fully discharged battery materials—the algorithm determines when sufficient information has been collected and signals when to move to the next stage of the experiment.
“It’s not just speeding up the measurement,” Kelly said. “It’s making decisions during the experiment — decisions a human used to make.”
This shift moves X-ray spectroscopy toward semi-autonomous operation, particularly at high-brightness synchrotron facilities such as the Advanced Photon Source (APS), a US Department of Energy Office of Science user facility. The team demonstrated the AI-guided XANES spectroscopy method at beamlines 25-ID-C, 20-BM, and 10-ID at the APS.
Implications for laboratory operations and data quality
For laboratory professionals working with synchrotron facilities or advanced X-ray spectroscopy tools, the development signals a shift in how experimental design and data acquisition may be managed.
Automating energy selection and dwell times reduces operator-dependent variability and lowers the risk of sample damage from prolonged exposure to X-ray beams. It also allows researchers to capture rapid chemical transformations in real time, particularly in systems such as batteries, catalysts, and electronic materials.
“There is a lot of hype around AI today in the media,” said Mathew Cherukara, computational scientist and group leader at the APS. “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 light sources grow brighter and datasets larger, AI-guided X-ray spectroscopy and AI-directed experiments may become standard components of beamline operations. For lab managers overseeing advanced materials characterization programs, the technology underscores the growing integration of machine learning into analytical instrumentation and experimental workflows.
This article was created with the assistance of Generative AI and has undergone editorial review before publishing.
