Most fossil-fuel based power plants use post-combustion technology. Addressing the CO₂ capture problem in these power plants is critical if a dent in CO₂ emissions is to be expected. Post-combustion CO₂ capture using solid sorbents has been identified as a powerful alternative for solvent based processes. Some accessible novel solid sorbents have the potential to further reduce parasitic energy.

However, the design, optimization, and integration of processes involving solid sorbents are poorly understood. Current models are computationally expensive making their use in process and systems optimization challenging. Further, the solution to the problem of finding optimal design and integration of these processes into existing plants remains elusive.

We propose to attack this problem by developing meta-models and surrogate models using machine-learning techniques, that significantly reduce the computational efforts; validating them against detailed models and experiments, and finally integrating them into systems-level models that provide the optimal way to integrate CO₂ capture processes into power plants. This approach will enable further reduction in parasitic energy and allow for the scale-up and implementation of solid-based capture processes into power plants.