IISc researchers develop new language system to unlock potential of nanoporous materials

In order to better comprehend and build nanoporous materials, researchers at the Indian Institute of Science (IISc), Bengaluru developed a new language system that could have a big impact on gas separation and carbon capture technologies.

The system, called STRONG (STring Representation Of Nanopore Geometry), converts complex nanopore geometries into simple character sequences, much like genetic code does for DNA.

The random production of nanopores, which are tiny holes in materials like graphene, has long hampered scientific understanding in the field of 2D materials research. This system will solve that problem.

“The problem is that these 2D materials have a wide distribution of nanopores, both in terms of shape and size. You don’t know what is going to form in the material, so it is very difficult to understand what the property of the resulting membrane will be,” said Ananth Govind Rajan, assistant professor at IISc’s Department of Chemical Engineering.

According to the STRONG system, distinct letters are assigned to various atomic configurations. For example, "F" stands for a fully bound atom with three bonds, and "C" for a corner atom with two bonds.

Similar to how ChatGPT handles text, this method enables machine learning models to "read" and analyze nanopore structures.

The research team demonstrated the usefulness of STRONG in environmental technology in a paper that was published in the Journal of the American Chemical Society.

The method discovered viable nanoporous structures for removing CO2 from flue gas, which could help reduce industrial carbon emissions.

Piyush Sharma, the study's lead author, used STRONG together with neural networks to assess materials for carbon capture potential.

Without requiring a lot of physical testing, the discovery also made it possible to create "digital twins" of 2D materials, which could aid in predicting material performance and finding new applications.

This discovery could speed up the creation of sophisticated materials for a number of applications, from DNA sequencing to water desalination, by simplifying the understanding and predicting the nanoporous material features.