Carbon dioxide capture by pressure swing adsorption

Abstract

The high concentration of carbon dioxide (CO2) in the atmosphere originated from combustion processes has been frequently pointed as the main responsible for global warming and climatic changes. To mitigate the adverse effects of global warming and to reduce CO2 concentration, many technologies have been developed in the last decades to capture CO2 in different scenarios. Among the available technologies for post-combustion Carbon Capture and Storage (CCS), one of the most studied processes is the adsorption in porous media, such as activated carbon. Pressure Swing Adsorption (PSA) is a cyclic adsorption process, which allows continuous separation of gas streams. The performance of a PSA process is usually evaluated by the purity, recovery, and productivity when the process reaches the cyclic steady state. This study presents experimental and simulated data obtained from a bench-scale PSA. It aims to improve the simulation of PSA process using more adequate parameter values and detailed models thermodynamically consistent to describe accurate results according to experimental data. Also, it aims to evaluate the model validation using three different activated carbons: Norit RB4, Filtron N and Charbon 500. The first one, for instance, presents pellet shape and relative high microporosity in comparison to the other ones. The unit was tested with a mixture simulating dry flue gases containing 85% of N2 and 15% of CO2 (on a molar basis). A mathematical-phenomenological model combining momentum, mass and heat balances, using the Linear Driving Force approach (LDF) for mass transport were applied. This parameter is commonly determined by experimental data of breakthrough curves, conversely in this work it has been estimated by experimental results of the respective gas uptake. Sips model was used to describe the adsorption equilibrium of single component, and for binary mixture Ideal Adsorption Solution Theory (IAST) was applied in this study to simulate the dynamic behavior of the process. In this work, the IAST equations were directly applied in the simulation which is not very common in literature for simulation of gas separation on PSA. The model predicted reasonably well the breakthrough curves and temperature history. PSA process simulation was validated according to experimental data and has shown to be in agreement with them. Estimation procedure of Linear Driving Force (LDF) parameter has shown to be reliable. Model parameters were adequately determined and IAST was more appropriate to simulate separation process of a binary mixture.