Stability of amine-functionalized CO2 adsorbents: a multifaceted puzzle.

This review focuses on important stability issues facing amine-functionalized CO2 adsorbents, including amine-grafted and amine-impregnated silicas, zeolites, metal-organic frameworks and carbons. During the past couple of decades, major advances were achieved in understanding and improving the performance of such materials, particularly in terms of CO2 adsorptive properties such as adsorption capacity, selectivity and kinetics. Nonetheless, to pave the way toward commercialization of adsorption-based CO2 capture technologies, in addition to other attributes, adsorbent materials should be stable over many thousands of adsorption-desorption cycles. Adsorbent stability, which is of utmost importance as it determines adsorbent lifetime and operational costs of CO2 capture, is a multifaceted issue involving thermal, hydrothermal, and chemical stability. Here we discuss the impact of the adsorbent physical and chemical properties, the feed gas composition and characteristics, and the adsorption-desorption operational parameters on the long-term stability of amine-functionalized CO2 adsorbents. We also review important insights associated with the underlying deactivation pathways of the adsorbents upon exposure to high temperature, oxygen, dry CO2, sulfur-containing compounds, nitrogen oxides, oxygen and steam. Finally, specific recommendations are provided to address outstanding stability issues.

Insights into the Hydrothermal Stability of Triamine-Functionalized SBA-15 Silica for CO2 Adsorption.

The hydrothermal stability of triamine-grafted, large-pore SBA-15 CO2 adsorbents was studied by using steam stripping. Following two 3 h cycles of steam regeneration, lower CO2 uptakes, lower CO2 /N ratios, and slower adsorption kinetics were observed relative to fresh samples, particularly at the lowest adsorption temperature (25 °C). CO2 adsorption measurements for a selected sample exposed to 48 h of steam stripping depicted that after the initial loss during the first exposure to steam (3-6 h), the adsorptive properties stabilized. For higher adsorption temperatures (i.e., 50 and 75 °C), however, all adsorptive properties remained almost unchanged after steaming, indicating the significance of diffusional limitations. Thermogravimetric analysis and FTIR spectroscopy on grafted samples before and after steam stripping showed no amine leaching and no change in the chemical nature of the amine groups, respectively. Also, a six-cycle CO2 adsorption/desorption experiment under dry conditions showed no thermal degradation. However, N2 adsorption measurement at 77 K showed significant reductions in the BET surface area of the grafted samples following steaming. Based on the pore size distribution of calcined, grafted samples before and after steaming, it is proposed that exposure to steam restructured the grafted materials, causing mass transfer resistance. It is inferred that triamine-grafted, large-pore SBA-15 adsorbents are potential candidates for CO2 capture at relatively high temperatures (50-75 °C; for example, flue gas) combined with steam regeneration.

The role of beaded activated carbon's surface oxygen groups on irreversible adsorption of organic vapors.

The objective of this study is to determine the contribution of surface oxygen groups to irreversible adsorption (aka heel formation) during cyclic adsorption/regeneration of organic vapors commonly found in industrial systems, including vehicle-painting operations. For this purpose, three chemically modified activated carbon samples, including two oxygen-deficient (hydrogen-treated and heat-treated) and one oxygen-rich sample (nitric acid-treated) were prepared. The samples were tested for 5 adsorption/regeneration cycles using a mixture of nine organic compounds. For the different samples, mass balance cumulative heel was 14 and 20% higher for oxygen functionalized and hydrogen-treated samples, respectively, relative to heat-treated sample. Thermal analysis results showed heel formation due to physisorption for the oxygen-deficient samples, and weakened physisorption combined with chemisorption for the oxygen-rich sample. Chemisorption was attributed to consumption of surface oxygen groups by adsorbed species, resulting in formation of high boiling point oxidation byproducts or bonding between the adsorbates and the surface groups. Pore size distributions indicated that different pore sizes contributed to heel formation - narrow micropores (<7Å) in the oxygen-deficient samples and midsize micropores (7-12Å) in the oxygen-rich sample. The results from this study help explain the heel formation mechanism and how it relates to chemically tailored adsorbent materials.

The role of beaded activated carbon's pore size distribution on heel formation during cyclic adsorption/desorption of organic vapors.

The effect of activated carbon's pore size distribution (PSD) on heel formation during adsorption of organic vapors was investigated. Five commercially available beaded activated carbons (BAC) with varying PSDs (30-88% microporous) were investigated. Virgin samples had similar elemental compositions but different PSDs, which allowed for isolating the contribution of carbon's microporosity to heel formation. Heel formation was linearly correlated (R(2)=0.91) with BAC micropore volume; heel for the BAC with the lowest micropore volume was 20% lower than the BAC with the highest micropore volume. Meanwhile, first cycle adsorption capacities and breakthrough times correlated linearly (R(2)=0.87 and 0.93, respectively) with BAC total pore volume. Micropore volume reduction for all BACs confirmed that heel accumulation takes place in the highest energy pores. Overall, these results show that a greater portion of adsorbed species are converted into heel on highly microporous adsorbents due to higher share of high energy adsorption sites in their structure. This differs from mesoporous adsorbents (low microporosity) in which large pores contribute to adsorption but not to heel formation, resulting in longer adsorbent lifetime. Thus, activated carbon with high adsorption capacity and high mesopore fraction is particularly desirable for organic vapor application involving extended adsorption/regeneration cycling.

Two-dimensional modeling of volatile organic compounds adsorption onto beaded activated carbon.

A two-dimensional heterogeneous computational fluid dynamics model was developed and validated to study the mass, heat, and momentum transport in a fixed-bed cylindrical adsorber during the adsorption of volatile organic compounds (VOCs) from a gas stream onto a fixed bed of beaded activated carbon (BAC). Experimental validation tests revealed that the model predicted the breakthrough curves for the studied VOCs (acetone, benzene, toluene, and 1,2,4-trimethylbenzene) as well as the pressure drop and temperature during benzene adsorption with a mean relative absolute error of 2.6, 11.8, and 0.8%, respectively. Effects of varying adsorption process variables such as carrier gas temperature, superficial velocity, VOC loading, particle size, and channelling were investigated. The results obtained from this study are encouraging because they show that the model was able to accurately simulate the transport processes in an adsorber and can potentially be used for enhancing absorber design and operation.

Effect of the adsorbate kinetic diameter on the accuracy of the Dubinin-Radushkevich equation for modeling adsorption of organic vapors on activated carbon.

This paper investigates the effect of the kinetic diameter (KD) of the reference adsorbate on the accuracy of the Dubinin-Radushkevich (D-R) equation for predicting the adsorption isotherms of organic vapors on microporous activated carbon. Adsorption isotherms for 13 organic compounds on microporous beaded activated carbon were experimentally measured, and predicted using the D-R model and affinity coefficients. The affinity coefficients calculated based on molar volumes, molecular polarizabilities, and molecular parachors were used to predict the isotherms based on four reference compounds (4.3≤KD≤6.8 Å). The results show that the affinity coefficients are independent of the calculation method if the reference and test adsorbates are from the same organic group. Choosing a reference adsorbate with a KD similar to that of the test adsorbate results in better prediction of the adsorption isotherm. The relative error between the predicted and the measured adsorption isotherms increases as the absolute difference in the kinetic diameters of the reference and test adsorbates increases. Finally, the proposed hypothesis was used to explain reports of inconsistent findings among published articles. The results from this study are important because they allow a more accurate prediction of adsorption capacities of adsorbents which allow for better design of adsorption systems.

Adsorption and desorption of mixtures of organic vapors on beaded activated carbon.

In this study, adsorption and desorption of mixtures of organic compounds commonly emitted from automotive painting operations were experimentally studied. A mixture of two alkanes and a mixture of eight organic compounds were adsorbed onto beaded activated carbon (BAC) and then thermally desorbed under nitrogen. Following both adsorption and regeneration, samples of the BAC were chemically extracted. Gas chromatography-mass spectrometry (GC-MS) was used to quantify the compounds in the adsorption and desorption gas streams and in the BAC extracts. In general, for both adsorbate mixtures, competitive adsorption resulted in displacing low boiling point compounds by high boiling point compounds during adsorption. In addition to boiling point, adsorbate structure and functionality affected adsorption dynamics. High boiling point compounds such as n-decane and 2,2-dimethylpropylbenzene were not completely desorbed after three hours regeneration at 288 °C indicating that these two compounds contributed to heel accumulation on the BAC. Additional compounds not present in the mixtures were detected in the extract of regenerated BAC possibly due to decomposition or other reactions during regeneration. Closure analysis based on breakthrough curves, solvent extraction of BAC and mass balance on the reactor provided consistent results of the amount of adsorbates on the BAC after adsorption and/or regeneration.