Scientists have developed a new artificial intelligence system that dramatically speeds up X-ray spectroscopy, allowing researchers to analyze materials up to five times faster while reducing human error, according to new research from Argonne National Laboratory.
Researchers developed an AI-driven approach that automates parts of X-ray absorption near-edge structure (XANES) spectroscopy, a powerful method for studying the chemical properties of materials. The technology can reveal how atoms behave inside materials such as batteries, catalysts, and advanced electronic components.
Traditionally, scientists must manually decide where to take measurements across different X-ray energy levels. These decisions can require dozens or even hundreds of adjustments, making experiments time-consuming and sometimes prone to human error.
The new AI system replaces much of this manual process by automatically selecting the most useful measurement points. By focusing only on the regions of the spectrum that contain valuable chemical information, the system reduces the number of required measurements by up to 80 percent without sacrificing accuracy.
This optimization dramatically shortens data-collection time, enabling scientists to observe chemical reactions and material changes much more quickly. It also reduces the risk of damaging sensitive samples during lengthy X-ray exposure.
The technology also introduces a new concept known as AI-directed experiments. During an experiment, the algorithm continuously analyzes incoming data and can determine when enough information has been collected or when conditions should change.
Researchers say this capability allows scientists to track chemical transformations in real time, improving the study of complex systems such as battery charging cycles or catalytic reactions in industrial processes.
Experts believe the approach could accelerate discoveries in energy technology, materials science, and electronics, fields that depend heavily on precise measurements of atomic-scale chemical behavior. By reducing experimental time and human guesswork, the AI system may also help large research facilities process the growing volumes of data generated by advanced X-ray instruments.
The breakthrough highlights the expanding role of artificial intelligence in scientific research, where machine learning tools are increasingly used to automate complex experiments and uncover insights that would be difficult to obtain through traditional methods.
