How Can (or Why Should) Process Engineering Aid the Screening and Discovery of Solid Sorbents for CO₂ Capture?

Conspectus Adsorption using solid sorbentsis emerging as a serious contenderto amine-based liquid absorption for postcombustion CO₂ capture. In the last 20+ years, significant efforts have been investedin developing adsorption processes for CO₂ capture. Inparticular, significant efforts have been invested in developing newadsorbents for this application. These efforts have led to the generationof hundreds of thousands of (hypothetical and real) adsorbents, e.g.,zeolites and metal-organic frameworks (MOFs). Identifying theright adsorbent for CO₂ capture remains a challenging task.Most studies are focused on identifying adsorbents based on certainadsorption metrics. Recent studies have demonstrated that the performanceof an adsorbent is intimately linked to the process in which it isdeployed. Any meaningful screening should thus consider the complexityof the process. However, simulation and optimization of adsorptionprocesses are computationally intensive, as they constitute the simultaneouspropagation of heat and mass transfer fronts; the process is cyclic,and there are no straightforward design tools, thereby making large-scaleprocess-informed screening of sorbents prohibitive. This Accountdiscusses four papers that develop computational methodsto incorporate process-based evaluation for both bottom-up (chemistryto engineering) screening problems and top-down (engineering to chemistry)inverse problems. We discuss the development of the machine-assistedadsorption process learning and emulation (MAPLE) framework, a surrogatemodel based on deep artificial neural networks (ANNs) that can predictprocess-level performance by considering both process and materialinputs. The framework, which has been experimentally validated, allowsfor reliable, process-informed screening of large adsorbent databases.We then discuss how process engineering tools can be used beyond adsorbentscreening, i.e., to estimate the practically achievable performanceand cost limits of pressure vacuum swing adsorption (PVSA) processesshould the ideal bespoke adsorbent be made. These studies show whatconditions stand-alone PVSA processes are attractive and when theyshould not be considered. Finally, recent developments in physics-informedneural networks (PINNS) enable the rapid solution of complex partialdifferential equations, providing tools to potentially identify optimalcycle configurations. Ultimately, we provide areas where further developmentsare required and emphasize the need for strong collaborations betweenchemists and chemical engineers to move rapidly from discovery tofield trials, as we do not have much time to fulfill commitments tonet-zero targets.