New AI Tools Propel Materials Discovery Forward

"Create a materially better future." That's the slogan of the Accelerate Conference presented by the Acceleration Consortium, a global network of researchers working to fast-track materials discovery. This year's conference featured talks by four scientists from Pacific Northwest National Laboratory, including a keynote speech from Sergei Kalinin, who is a PNNL joint appointee and Weston Fulton Professor at the University of Tennessee Knoxville. 

In his speech, titled "Integrating autonomous systems for advanced material discovery: bridging experiments and theory through optimized rewards," Kalinin explained how he believes scientists can achieve autonomous experimentation.

"Scientists can have the best instrumentation and the best machine learning algorithms, but if they cannot talk to each other, nothing will get done," said Kalinin. "If the system needs a human intermediary between the two, scientists lose many of the potential advantages of including machine learning in the research, including speed, working with large volumes of data, or precision. But it is up to human to define what the reward is that machine learning will try to maximize." 

PNNL researchers are working hard to develop and train machine learning algorithms to aid in materials discovery. AutoEM, an artificial intelligence-guided microscopy platform, recently won an R&D100 Award. The Adaptive Tunability for Synthesis and Control via Autonomous Learning on Edge (AT SCALE) Initiative integrates theory and simulation with experiments to go beyond automation into autonomy.

Mitra Taheri, PNNL joint appointee and professor at Johns Hopkins University, also gave a talk, titled "Autonomous, High Throughput Characterization Enabling Discovery of Cobalt-Lean Heusler Alloys," at the conference. At PNNL, she serves as the chief materials scientist on the leadership team of AT SCALE.

"By integrating machine learning and artificial intelligence into synthesis and characterization methods, including microscopy, we can achieve both high throughput and precision in materials screening," said Taheri. "This opens up new possibilities for materials discovery and design, such as reducing the amount of rare or critical elements in functional alloys for cheaper materials." 

In addition to Kalinin and Taheri, PNNL scientists Henry Sprueill and Yangang Liang spoke on "ChemReasoner: bridging generative AI and computational chemistry" and "Automation-accelerated redox flow battery research," respectively. Chief Data Scientist Neeraj Kumar also participated in the Accelerate Conference by moderating the session "Accelerating computational predictions of materials, molecules, and reactions".

Tell Us What You Think

Do you have a review, update or anything you would like to add to this news story?

Leave your feedback
Your comment type
Submit

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.