Designing 'Zeolite' with Artificial Intelligence

[Asia Economy Reporter Junho Hwang] A technology to develop porous materials in desired forms using artificial intelligence has been developed by domestic researchers.

It is expected that the efficiency of energy storage materials such as gas adsorption will be greatly improved by designing porous materials like zeolites into desired structures.

KAIST announced on the 7th that Professor Jihan Kim's research team from the Department of Bio and Chemical Engineering succeeded in inverse designing porous materials with desired properties using artificial intelligence.

Porous materials refer to materials with tiny holes. These holes can store gases or drugs, making them useful in various industries. A representative material is zeolite, which is silicon oxide combined with metals.

Zeolite Generation Using Artificial Intelligence

The research team developed a method to design the structure of zeolites, a type of porous material, by utilizing artificial intelligence technology and molecular simulation technology. They created a structure that allows artificial intelligence to learn and generate the complex characteristics of porous materials by using a generative adversarial network, an AI generative model, and three-dimensional grid data used in molecular simulations.

In particular, the research team excluded some of the previously known zeolite structures during the AI training process. Through this, they succeeded in generating zeolites with new structures that the AI had not learned before.

Although research has been conducted to develop materials with desired properties using artificial intelligence, it has been limited to simple small molecules.

Applicable to Various Properties Such as Gas Adsorption

Professor Jihan Kim said, "This is the first case of designing porous materials using artificial intelligence," and added, "It is not limited to gas adsorption purposes but can be easily applied to other properties, so it is expected to be used in developing materials for other fields such as catalysts, separation, and sensors."

This research was conducted with support from BK21, the Korea Research Foundation Mid-Career Researcher Support Project, and the Energy Cloud Project Group.

This content was produced with the assistance of AI translation services.