CO₂ recovery from ultra-dilute streams: Bridging a novel temperature vacuum swing adsorption model with materials screening

Separation is the single most important Chemical Engineering unit operation that is responsible for 15% of our totalenergy consumption, and with the increase of the global population this number is expected to rise to as much as 45% (Sholl and Lively, 2016) . Increasing the efficiency of separation processes will therefore have an enormous impact onsociety (National Academies of Sciences, Engineering , and Medicine, 2019) . As an example, separation of a dilutecomponent from a gas mixture (e.g., the removal of carbon dioxide from a confined space or mercury vapour from air)implies processing of extremely large volumes of gas to capture adequate amounts of the dilute component. This facthas significant implications on the energy consumption of the process and the plant size (Santori et al., 2018) . Studies attested that adsorption processes are best suited for the efficient removal of trace gas impurities aiming toachieve extremely high purities (Sircar et al., 1996) . The two main reasons for this are: (a) the availability of a largespectrum of microporous adsorbents with varying pore structures and surface properties, and (b) the possibility ofdesigning many different process schemes by tailoring generic adsorption separation methods. Materials like metal-organic frameworks (MOFs), whose chemical composition and pore shape can be optimally tuned for particularapplications are good candidates, but we are currently lacking a meaningful systematic way, from a process viewpoint,to rank them for particular applications. In our previous work, we developed a molecular simulation tool which allows us to screen thousands of materials foradsorption using their physicochemical and adsorptive properties (Boyd et al., 2019) . In this work, we extend thatstudy by coupling the molecular simulation tool with process modelling to rank materials for dilute gas separations andfor a given set of process metrics, including the overall energy efficiency and process productivity. A novel 5-steptemperature-vacuum swing adsorption (TVSA) process was developed, and its performance was evaluated by usingan equilibrium-based shortcut approach similar to that described by Joss et al., 2015 . Our TVSA process model is thencoupled with molecular simulations, which allows the evaluation of both, physicochemical and adsorptive properties ofthousands of microporous structures, including ~300,000 MOFs (Boyd et al., 2019) . We applied the model to a particular ultra-dilute binary mixture of N and CO (CO concentration of 2000 ppm) forapplications such as CO capture from confined spaces like spacecraft, aircraft, and submarines, and by using thecommercial adsorbent zeolite 13X. The CO purity achieved after each step of the process, together with the totalamount of CO and N species, is presented in Fig. 1. This novel process yields higher CO purities than the onesobtained by standard previously used 5-step TVSA processes. The performance of this process is expected to be further improved by coupling molecular modelling with processoptimization. By doing so, we aim to screen the large pool of available MOF materials to find the best performing onesfor this particular application. This analysis provides new insights into which relevant metrics need to be included whenranking materials for ultra-dilute gas separations, and identifies the top performing ones given the relevant processmetrics.

Challenge “There is currently no ability to quickly identify what processes and process conditions are optimal for a particular adsorbent to achieve the required specifications for a capture application” (Mission Innovation report).¹,.