promising CO 2 /CH 4 separation properties, with a selectivity >1,500 at 293 K and
0.5 bar [25]. Using synchrotron powder X-ray diffraction (XRD) and Rietveld
analyses, they showed that both K
+ and Cs
+ ions are located within the 8-ring
windows in the LTA structure. Thus, these cations appear to obstruct the passage
of CH 4 through the zeolite, whereas the CO 2 molecules can pass with a temporary
displacement of the cations. The Long group have reported that Ca-A can adsorb
3.72 mmol CO 2 g
À1 at 0.15 bar and 313 K from the dry flue gas from a coal-fired
power plant, together with a CO 2 /N 2 selectivity of 250 [26]. Even compared to
Mg-MOF-74, one of the best performing CO 2 adsorbents among those reported thus
far, therefore, Ca-A has a greater volumetric CO 2 uptake, CO 2 /N 2 selectivity, and a
longer breakthrough time.
3.2 CHA-Type Zeolites
Webley and co-workers [27–29] have studied the N 2 and CO 2 adsorption properties
of various cation-exchanged forms of chabazite (CHA) zeolites with different Si/Al
ratios (2.4–2.5). They have shown that the adsorption properties of the fully
exchanged Li
+
, Na
+
, and K
+ forms of chabazite with Si/Al ¼ 2.5 at 273 K and
1.0 bar depend strongly on the type of extraframework cations, with N 2 loadings of
1.53, 1.25, and 0.87 molecules per unit cell and with CO 2 loadings of 6.37, 5.82, and
4.77 molecules per unit cell, respectively [27]. The relatively low adsorption capacity of N 2 on K-chabazite was initially suggested to be due to kinetic screening, i.e.,
pore blockage by larger K
+ cations. However, the “molecular trapdoor” mechanism
has been subsequently forwarded to explain the enhanced selectivity of Cs
+
-
exchanged chabazite [15] and has also been observed for other cage-based smallpore zeolites like rho (RHO) [30]. According to this mechanism, the very low uptake
(theoretically zero) of bigger, less polarizable N 2 and CH 4 molecules is not due to the
molecular sieving effect. Instead, more polarizable CO 2 interacts with the “trapdoor”
cation strongly enough that the cation deviates from the center of the 8-ring windows
(temporarily and reversibly), allowing sufficient space for CO 2 to enter the zeolite
pores. Since the molecules that have weaker interactions with the extraframework
cations do not induce extraframework cation movement, they cannot be adsorbed.
As shown in Fig. 3, therefore, the molecular trapdoor effect can be improved (viz.,
complete occupation of all pore aperture “doorways” by the trapdoor cation) by
tuning the framework Si/Al ratio.
The existence of such an optimized Si/Al ratio was demonstrated by comparing
the N 2 adsorption isotherms obtained at 77 K on the K
+ form of chabazite zeolites
with the same Si/Al ratios described above [31]. The samples with Si/Al < 3.0 show
negligible N 2 adsorption, indicating a potentially optimized trapdoor effect. However, those with Si/Al > 3.0 have an obvious adsorption of N 2, suggesting the
absence of complete occupation of pore windows by trapdoor cations. It is worth
noting that the optimized Si/Al ¼ 3.0 is only a theoretical critical threshold, as
K-chabazite with a Si/Al close to 3.0 (e.g., 3.2) still shows a very small yet
Small Gas Adsorption and Separation in Small-Pore Zeolites
7
0.5 bar [25]. Using synchrotron powder X-ray diffraction (XRD) and Rietveld
analyses, they showed that both K
+ and Cs
+ ions are located within the 8-ring
windows in the LTA structure. Thus, these cations appear to obstruct the passage
of CH 4 through the zeolite, whereas the CO 2 molecules can pass with a temporary
displacement of the cations. The Long group have reported that Ca-A can adsorb
3.72 mmol CO 2 g
À1 at 0.15 bar and 313 K from the dry flue gas from a coal-fired
power plant, together with a CO 2 /N 2 selectivity of 250 [26]. Even compared to
Mg-MOF-74, one of the best performing CO 2 adsorbents among those reported thus
far, therefore, Ca-A has a greater volumetric CO 2 uptake, CO 2 /N 2 selectivity, and a
longer breakthrough time.
3.2 CHA-Type Zeolites
Webley and co-workers [27–29] have studied the N 2 and CO 2 adsorption properties
of various cation-exchanged forms of chabazite (CHA) zeolites with different Si/Al
ratios (2.4–2.5). They have shown that the adsorption properties of the fully
exchanged Li
+
, Na
+
, and K
+ forms of chabazite with Si/Al ¼ 2.5 at 273 K and
1.0 bar depend strongly on the type of extraframework cations, with N 2 loadings of
1.53, 1.25, and 0.87 molecules per unit cell and with CO 2 loadings of 6.37, 5.82, and
4.77 molecules per unit cell, respectively [27]. The relatively low adsorption capacity of N 2 on K-chabazite was initially suggested to be due to kinetic screening, i.e.,
pore blockage by larger K
+ cations. However, the “molecular trapdoor” mechanism
has been subsequently forwarded to explain the enhanced selectivity of Cs
+
-
exchanged chabazite [15] and has also been observed for other cage-based smallpore zeolites like rho (RHO) [30]. According to this mechanism, the very low uptake
(theoretically zero) of bigger, less polarizable N 2 and CH 4 molecules is not due to the
molecular sieving effect. Instead, more polarizable CO 2 interacts with the “trapdoor”
cation strongly enough that the cation deviates from the center of the 8-ring windows
(temporarily and reversibly), allowing sufficient space for CO 2 to enter the zeolite
pores. Since the molecules that have weaker interactions with the extraframework
cations do not induce extraframework cation movement, they cannot be adsorbed.
As shown in Fig. 3, therefore, the molecular trapdoor effect can be improved (viz.,
complete occupation of all pore aperture “doorways” by the trapdoor cation) by
tuning the framework Si/Al ratio.
The existence of such an optimized Si/Al ratio was demonstrated by comparing
the N 2 adsorption isotherms obtained at 77 K on the K
+ form of chabazite zeolites
with the same Si/Al ratios described above [31]. The samples with Si/Al < 3.0 show
negligible N 2 adsorption, indicating a potentially optimized trapdoor effect. However, those with Si/Al > 3.0 have an obvious adsorption of N 2, suggesting the
absence of complete occupation of pore windows by trapdoor cations. It is worth
noting that the optimized Si/Al ¼ 3.0 is only a theoretical critical threshold, as
K-chabazite with a Si/Al close to 3.0 (e.g., 3.2) still shows a very small yet
Small Gas Adsorption and Separation in Small-Pore Zeolites
7