Pressure swing adsorption (PSA) is regarded as a promising technology for pre-combustion CO2 capture, due to the boundary conditions of the separation process (high feed pressure of 35 to 40 bar and high CO2 feed concentration of about 40%). At the same time, metal organic frameworks have attracted research interest because of their exceptional adsorption properties and their flexibility in tailoring. This work aims at assessing the potential of a metal organic framework (USO-2-Ni) with respect to the performance in separating CO2 from H2 in a pressure swing adsorption pre-combustion process. This is done by comparing it to a base case which was designed using commercial activated carbon as adsorbent. Both experimental and theoretical aspects related to the PSA process are investigated in this study. First a sound characterization of the two adsorbents namely activated carbon (Chemviron, Germany) and MOF[1] is done by measuring equilibrium adsorption isotherms of CO2, H2 and N2 using a Magnetic Suspension Balance (Rubotherm, Germany). Additionally physical material properties such as material and bed densities as well as heat capacities were defined. Secondly kinetic parameters, such as mass and heat transfer parameters, are determined by conducting breakthrough experiments. The interpretation of the dynamic experiments is done by describing the process with a detailed one-dimensional model consisting of mass and heat balances and several constitutive equations such as adsorption isotherms, equation of state and pressure drop correlation. The model was further validated by predicting breakthrough curves in a broad temperature and pressure range and compared to the experiments. The comparison of the results of the aforementioned experiments with simulations allowed to confirm the static adsorption measurements and to determine the rapid adsorption kinetics of CO2 on USO-2-Ni.
For the design of the PSA process several aspects are important: the energy penalty has to be minimized whereas the capture rate is aimed to be higher than 90% and the CO2 purity has to meet the specifications for the subsequent transportation and storage. Therefore after the validation of the process parameters, the simulation tool is used to describe various complete PSA cycles with different combinations of the individual steps. To assess the performance of the newly developed PSA cycles a multi-objective optimization including the CO2 capture rate and the CO2 purity is conducted.
The results of the PSA simulations considering USO-2-Ni as adsorbent have been compared to those considering an activated carbon adsorbent. The former show a slightly better performance, as expected based on the favorable thermodynamics of the metal organic framework material compared to the thermodynamics of the activated carbon. It was shown that high bed and particle porosities can have a negative effect on the process performance, due to the reduced CO2 uptake per unit volume and the increased fluid ratio in the column. It is therefore essential to consider the formulation of the adsorbent as much as the adsorption properties in further process development.
This work is part of the European Union’s Framework Program 7 project DECARBit (“Decarbonise it”, 2008-2011).
[1] B. Arstad et al., Adsorption 14, 755 (2008)
See more of this Group/Topical: Separations Division