(687d) Systems Design and Analysis of Direct Air Capture (DAC) of CO2 Via Temperature Vacuum Swing Adsorption

CO₂ concentrations in the atmosphere have risen from a pre-industrial level of 270 ppm to around 400 ppm today. This rise has led to increased trapping of heat by the atmosphere and a potential for catastrophic climate change. To reduce the rate of CO₂ concentration increase, post combustion technologies have been proposed which are focused on removing CO₂ from point sources. However, this constraints the location of the CO₂ capture plant, and cannot mitigate other CO₂ emission sources, such as transportation, and cannot reduce the atmospheric concentration, only slow its increase.  On the other hand Direct Air Capture (DAC) aims at removing CO₂ directly from air increasing the flexibility of deployment and enabling CO₂ concentration reduction in the future.  

Our work proposes capturing CO₂ from air via temperature vacuum swing adsorption (TVSA) process using solid adsorbents. The solid sorbent is an amine impregnated MOF material (MIL-PEI) grown as films.  Monolith contactors are used to support the films due to the low gas pressure drop that can be achieved during the adsorption phase of the cycle.  Steam is used as a stripping agent during the desorption step. The amines in MIL-PEI are prone to oxidation/leaching during the desorption step if operated at high temperature. Vacuum swing is integrated with temperature swing process to minimize this risk. This will require less heating of the MOF surface since vacuum facilitates lowering the partial pressure of CO₂ over the MOF surface. The captured CO₂ will be utilized as part of the carbon source to develop a novel route for synthesis of 2,5 furandicarboxylic acid (FDCA) based on enzymatic carboxylation of furfural at near ambient conditions.

This talk will address the cyclic simulation of the DAC process and its overall economics. A partial differential algebraic model has been developed to solve coupled heat and mass transfer profiles of the three component system, viz, CO₂, the inert gas and the water. The experimental data from isotherm measurements of MIL-PEI-50 were used to estimate the parameters for CO₂ adsorption using the temperature dependent Langmuir isotherm model. Furthermore, the rate of adsorption of CO₂ is approximated using the linear driving force model (LDF). During the desorption step the steam is assumed to condense on the surface of the MOF but not to penetrate the MOF pores.  The simulations for the TVSA model are performed using gPROMS, which is a commercial dynamic process modeling and optimization software. The governing equations are solved using method of lines with a finite difference based discretization for the axial and radial domains. The energy required for CO₂ removal is compared with the theoretical minimum energy required for ideal gas unmixing. Parasitic losses during the desorption step have been identified. They are mainly the heat of enthalpy to raise the sorbent and the monolith wall temperature during the desorption step. Energy analysis has been performed suggesting areas of improvement for reduction in these parasitic losses. It is shown that the energy requirements of the TVSA process will decrease as the sorbent capacity increases. It will also bring down the parasitic loss percentage. Optimization studies will further improve the process.