UI researchers use AI to find ideal carbon capture material

Simmy Wood
·2 min read
1

URBANA, Ill. (WCIA) — Researchers from the University of Illinois are using artificial intelligence to find new materials for capturing carbon dioxide emissions before they spread into the atmosphere.

The process, called carbon capture, is critical in reducing greenhouse gases from industrial sources like power plants. However, it is not cheap to design metal-organic frameworks (MOFs), which are one of the most effective materials for carbon capture, to selectively absorb large amounts of carbon dioxide.

Virtual virtuosos: UI pianists use new tech to learn remotely

UI researchers Emad Tajkhorshid and Hyun Park are working with scientists from Argonne National Laboratory, the University of Illinois at Chicago, and the University of Chicago to rearrange the building blocks of MOF molecules and develop the most ideal configurations for carbon capture.

The team is using several innovative methods to achieve this. They are running chemical screenings of materials, and using molecular dynamics to create theory-based simulations. They are also using AI for machine learning and to help think outside the box in discovering more arrangements.

These generative AI methods have allowed researchers to significantly speed up the usually-pain-staking MOF assembly process. The team was able to create more than 120,000 new MOF designs within 30 minutes.

Illini grads pitch on ‘Shark Tank’

Tajkhorshid said the study demonstrates the great potential of applying AI to molecular sciences.

After running calculations on the Argonne Leadership Computing Facility’s Polaris supercomputer, researchers use the Delta supercomputer at the U of I to quickly narrow down the field to only the most ideal carbon capture candidates. They are then tested for stability, chemical properties, and capacity.

“I believe the key to making successful molecule targets, whether they are MOFs or biomolecules, [is a] tight marriage of [a] diverse set of old and new techniques,” Park said. “In addition to contributing my expertise in biophysics and AI, I am glad I had hands-on experience working with different scientists on MOF project, merging many fields of science.”

For the latest news, weather, sports, and streaming video, head to WCIA.com.