(292f) Performance-based screening of porous materials for carbon capture

The discovery of new classes of porous materials such as metal-organic frameworks (MOFs) has opened access to a very large number of structures with a wide range of functionalities, which can be potentially exploited in different separation applications. Experimental evaluation of all these materials for specific applications is not feasible, and as a result, this prompted the development of high throughput computational screening methods. So far, these screening methods have been predominantly based on certain adsorbent metrics, such as pore volume and surface area, and equilibrium and dynamic properties, such as adsorption isotherms, selectivity and diffusivity, obtained from molecular simulations. It is now becoming apparent that a more complete picture of the performance of porous materials in a PSA or VSA process should be obtained from the actual process simulation. In my presentation, I will reflect on the development of the multiscale strategies that combine structural characterization methods, machine learning approaches, molecular simulations and processes modelling and optimization to predict performance of the materials on the process scale. Several studies that employ these strategies have already emerged, ranking MOFs and other materials for post-combustion carbon capture. Here we specifically focus on the challenges associated with the interface between molecular and process levels of description, on the systematic analysis of the propagation of errors across scales, and demonstrate that the emerging picture is quite complex.