Dynamic and equilibrium-based investigations of CO2-removal from CH4-rich gas mixtures on microporous adsorbents

Abstract

The removal of CO2 from CH4-rich gas mixtures is one of the key technologies for CH4-production and purification (Silva et al., Microporous Mesoporous Mat 158:219, 2012; 187:100, 2014). For this purpose, different techniques like adsorption on porous solids, membrane technologies or absorptive methods are employed (Scholes et al., Fuel 96:15, 2012; Sridhar et al., Sep Purific Rev, 2007). In any case, the appropriate separation technique as well as the optimal separation active material must be found. However, the choice of the optimal ensemble depends on many parameters, particularly CO2-concentration, the presence of other components i.e., water, the content of higher hydrocarbons, the pressure of the raw gas and the gas throughput (Andriani et al., Appl Biochem Biotechnol 172:4, 2014). In this work the focus is put on adsorption technologies. Therefore, three different commercially available adsorbents were investigated in the context of their applicability in separation processes by adsorption. One zeolite, a commercial activated carbon and a carbon molecular sieve were chosen as adsorbents. The classification of the materials is based on the characterization with N2 at 77 K, a series of adsorption isotherms and breakthrough curves (CO2 in the presence of CH4). Isotherms were measured by a volumetric method at temperatures of 293–333 K and pressures up to 2 MPa. Due to very long equilibration times in case of CH4 on the carbon molecular sieve, isotherm data for 313–353 K up to 1 MPa were taken from reference (Moller et al., Chem. Ing. Tech 86:1–2, 2014). Dynamic experiments were carried out with a ternary mixture of He/CH4/CO2 (molar fractions: 0.80/0.15/0.05) at 0.5 MPa and 293 K. A simplified mathematical model, based on mass- and energy balances, was applied to simulate breakthrough curves on packed adsorbent beds. The suitability of the investigated adsorbents for CO2-removal by adsorption was classified with the help of the obtained experimental data. It can be shown, that an evaluation of the separation performance of such materials, based only on textural parameters like the BET surface area or N2-isotherms at 77 K is limited in its confidence and can cause a substantial misinterpretation of the whole separation process.