AI for Accelerated Discovery of Materials for Carbon Capture

Abstract

In 2018, the IPCC reported that limiting atmospheric warming to less than 2 °C will require both capture and storage of anthropomorphic carbon dioxide (CO2) emissions. Despite estimates that we will need to capture upwards of 10 GtCO2 per year by 2050, today we capture less than 1% of all emitted CO2. Dramatic reductions in the cost of carbon capture processes are needed to drive adoption at scale. IBM Research is building AI-driven tools to accelerate the discovery of materials with improved performance for carbon capture, utilization, and storage including solvents, solid sorbents, and membranes. We have also developed automated synthesis platforms for production of carbon capture materials and novel catalysts for carbon dioxide conversion. Finally, we have created a digital rock platform coupled with high-precision microfluidics to accurately simulate CO2 fluid flow at pore scales relevant for geological storage.