Left: A porous structure from Michael Deem’s database of predicted zeolites, which consist of silicon (tan) and oxygen (red) atoms. Right: Voronoi Hologram describes the cavities of the structure shown on the left. The size and shape of these cavities determines which molecule will pass through the structure, and which will be absorbed. The colored points show how many times a certain shape appears in the structure.
Approximately 75 percent of electricity used in the U.S. is produced by coal-burning power plants that expel carbon dioxide into the atmosphere. Berkeley Lab researchers are searching for porous materials to filter out the CO2 before it reaches the atmosphere, but identifying these materials is easier said than done.

“There are a number of porous substances — including crystalline porous materials, such as zeolites, and metal-organic frameworks — that could be used to capture carbon dioxide from power plant emissions,” says Maciej Haranczyk, a scientist in Berkeley Lab's Computational Research Division.

In the category of zeolites alone, Haranczyk notes that there are around 200 known materials and 2.5 million structures predicted by computational methods. That’s why Haranczyk and colleagues have developed a computational tool that can help researchers sort through vast databases of porous materials to identify promising carbon capture candidates—and at record speeds. They call it Zeo++.

Using Zeo++, researchers have already sifted through one such database of millions of materials and have identified a few that could outperform current technologies. The tool works not by simulating each atom of a material, but by mapping what isn’t there - the voids in the materials.

Drawing from a database of the coordinates of all the atoms in each porous structure, Zeo++ generates a 3D map of the voids in each material. This 3D network allows researchers to see where the channels between atoms intersect to create cavities. The size and shape of these cavities determine whether a molecule will pass through the system or be absorbed. Using a tool called Voronoi Holograms, also developed by the Berkeley Lab team, researchers can automatically compare these 3D maps to identify materials with similar pore sizes and structures.

(Berkeley Lab)