Effect of sulfur dioxide on adsorption capacity of zeolite sorbents for carbon dioxide

Abstract

The article discusses a rather serious problem limit-ing the use of adsorption for the CO2 capture from flue gas, in the presence of sulfur dioxide. An apparatus with a vertical batch adsorber was constructed to study adsorption under elevated pressure in a wide range of temperatures and evaluation of CO2 breakthrough curves with an infrared analyzer. The article summarizes the results of experiments conducted with zeolite clinoptilolite, which represented natural materials, and molecular sieve 13X as a representative of synthetic sorbents. Adsorption capacities achieved during cyclically repeated tests with a model gaseous mixture free of SO2 and a mixture of the same composition but enriched with a low volume fraction of SO2 (0.3 %) were compared. Adsorption took place at a temperature of 20 °C and at two overpressures (200 and 500 kPa) of the gas with a 13 % volume fraction of CO2. Each sub-experiment consisted of five adsorption and desorption cycles, where desorption was based on depressurization followed by temperature increase to 120 °C under nitrogen atmosphere. There were no changes in capacities when tested with the gaseous mixture without SO2. Relative to the weight of the sample, the 13X sample at an overpressure of 500 kPa had a capacity of 11.3 % and clinoptilolite 3.8 %. Tests in the presence of SO2 led to a permanent reduction of the equilibrium capacities for both samples and at both pressures. At the overpressure of 500 kPa, the capacity decreased to 7.4 % for the 13X and to 2.5 % for the clinoptilolite. A more intensive desorption involving a thermal and vacuum step did not lead to any improvement for the 13X sample. In contrast, the effect for clinoptilolite was very positive. Its capacity in the fifth cycle reached 3.4 % close to the state without SO2 exposition. In the case when SO2 in the gas was accompanied with 40 % relative humidity, vacuum desorption did not lead to positive results in any case. After five cycles, the capacity of 13X dropped to 3.2 % and clinoptilolite to 1.4 %. When moisture, SO2 and the presence of O2 (volume fraction of 6 %) in the model mixture were further combined, the capacity of 13X decreased to 1.4 % and clinoptilolite to 0.4 % after five cycles. Tests with SO2 (dry gas) caused a decrease in the specific surface area from 512 to 211 m2.g‒1 for the 13X sample. On the other hand, for clinoptilolite it decreased from only 29 to 28 m2.g‒1 under the same conditions. Ac-cording to XRF, it was not possible to remove sorbed SO2 from the 13X sample even by evacuation followed by heating up to 200 °C. Using the XRD method, it was found that SO2 remains in the matrix, although it does not undergo transition to the crystalline phase. The study verified that synthetic molecular sieve 13X, unlike natural clinoptilolite, is not applicable for CO2 adsorption from SO2 containing flue gas.