A team of IBM researchers are creating a cloud-based knowledge base of existing methods and materials to capture CO2. It employs IBM technology for annotation and natural language processing to mine information contained in patents and papers and applies AI to digest information and present findings to the researcher, like a ranking of the best-known materials for CO2 separation.
Based on this knowledge, scientists are able to define desired properties of molecules to be considered for CO2 capture and separation processes. Teams can then employ AI algorithms to predict the optimal molecules to be used as building blocks for more effective polymer membranes for CO2 separation.
Once captured, CO2 can be put to use. IBM researchers are also working on a sustainable materials development platform for harnessing CO2 as a feedstock or raw material for monomers and polymers such as plastic. The new CO2-based materials are designed with a focus towards recyclability that allows for recovery and reuse.
Progressing carbon capture and sequestration before it’s too late requires an acceleration of the discovery process through the close integration of high-performance computing infrastructure, sophisticated AI systems, and AI-guided automatic lab experiments to test large numbers of chemical reactions. The reactions should illustrate the design rules for molecules and chemical processes that enable the efficient synthesis of materials optimized for CO2 capture, separation and conversion.
The goal over the next five years is to make CO2 capture and reuse efficient enough to scale globally so we can significantly reduce the amount of CO2 released into the atmosphere and, ultimately, slow climate change.