Chemical Engineering Science 64 (2009) 3683 -- 3687

Contents lists available at ScienceDirect

Chemical Engineering Science

journal homepage: w w w . e l s e v i e r . c o m / l o c a t e / c e s

Adsorption of carbon dioxide, ethane, and methane on titanosilicate type

molecular sieves
Alejandro Anson a , Christopher C.H. Lin a , Steven M. Kuznicki a, ∗ , James A. Sawada b
a
Department of Chemical and Materials Engineering, University of Alberta, Edmonton, AB, Canada T6G 2V4
b
QuestAir Technologies Inc, 6961 Russell Ave, Burnaby, BC, Canada V5J 4R8

A R T I C L E I N F O A B S T R A C T

Article history: The separation of carbon dioxide from light hydrocarbons is a vital step in multiple industrial processes
Received 31 March 2009 that could be achieved by pressure swing adsorption (PSA), if appropriate adsorbents could be identified.
Received in revised form 5 May 2009 To compare candidate PSA adsorbents, carbon dioxide, methane, and ethane adsorption isotherms were
Accepted 7 May 2009
measured for cation exchanged forms of the titanosilicate molecular sieves ETS-10, ETS-4, and RPZ. Mixed
Available online 22 May 2009
cation forms, such as Ba/H-ETS-10, may offer appropriate stability, selectivity, and swing capacity to be
Keywords:
utilized as adsorbents in CO2 /CH4 PSA processes. Certain cation exchanged forms of ETS-4 were found to
Adsorption partially or completely exclude ethane by size, and equivalent RPZ materials were observed to exclude
Carbon dioxide both methane and ethane, while allowing carbon dioxide to be substantially adsorbed. Adsorbents such
Ion exchange as Ca/H-ETS-4 and Ca/H-RPZ are strong candidates for use in PSA separation processes for both CO2 /C2 H6
Molecular sieves and CO2 /CH4 , potentially replacing current amine scrubber systems.
Selectivity © 2009 Elsevier Ltd. All rights reserved.
Separations
Titanosilicates
Zeolites

1. Introduction
The separation of carbon dioxide from light hydrocarbons is an
important stage in the industrial production of methane and ethane
from natural gas (Cavenati et al., 2006). For some applications, such
as the synthesis of ethylene by decomposition of ethane, the carbon
dioxide content in ethane or methane must be reduced to trace levels.
This can be accomplished by chemical absorption with amines, but
the purification process is feasible only when the carbon dioxide
content in the inlet stream is already moderately low (< 3–5%) (Rao
and Rubin, 2002). The total or partial extraction of carbon dioxide
from ethane or methane by pressure swing adsorption (PSA) could
reduce the chemical and energy costs of the purification process,
while obviating the need for amine scrubbers, a technology which
is causing increasing environmental concern (Collie and Hlavinka,
1998; Rao and Rubin, 2002).
The separation of carbon dioxide from methane or ethane could
be performed by pressure swing adsorption (PSA) using an appro-
priate adsorbent. Carbon dioxide, ethane, and methane adsorption
isotherms for various standard adsorbents are shown in Fig. 1. The
adsorption capacities on carbon molecular sieves at gas pressures
below 1 bar (Fig. 1a) typically decrease in the following order:
∗ Corresponding author. Tel.: +1 780 492 8819; fax: +1 780 492 8958.
E-mail address: steve.kuznicki@ualberta.ca (S.M. Kuznicki).
C2 H6 > CO2 > CH4 (Laukhuf and Plank, 1969; Reich et al. 1980). Car-
bon molecular sieves are thus selective towards ethane over carbon
dioxide and cannot be utilized to reduce carbon dioxide levels in
ethane streams.
Most classical aluminosilicate zeolites selectively adsorb carbon
dioxide over ethane and methane, as is illustrated for 13X zeolite in
Fig. 1b (Choudhary et al., 1995). Similar selectivity is observed for
NaY, Na-mordenite and 4A zeolites (Choudhary et al., 1995; Breck,
1974). The shape of the CO2 isotherms for classical zeolites varies,
but the curves are generally too steep for normal PSA processes
and, therefore, classical zeolites are best suited for thermal swing
adsorption processes.
Not all molecular sieves preferentially adsorb carbon dioxide over
ethane; some follow a capacity sequence of C2 H6 > CO2 > CH4 , as is
illustrated for ALPO-5 in Fig. 1c (Choudhary and Mayadevi, 1996a).
Similar trends are observed for SAPO-5 and silicalite-1, with ethane
selectivities following the order: silicalite-1 > ALPO-5 > SAPO-5
(Choudhary and Mayadevi, 1996a,b).
Templated, high silica zeolites usually demonstrate a capacity se-
quence of CO2 > C2 H6 > CH4 , but present very small CO2 /C2 H6 se-
lectivity, as is observed for ZSM-5 in Fig. 1d (He and Seaton, 2006).
These adsorbents cannot be used for a PSA separation of carbon diox-
ide from ethane but, with their nearly linear CO2 isotherms, may be
suitable for PSA separations of carbon dioxide from methane. Prod-
uct recovery would suffer, however, due to the adsorbent's signifi-
cant methane capacity.

0009-2509/$ - see front matter © 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ces.2009.05.024
3684 A. Anson et al. / Chemical Engineering Science 64 (2009) 3683 -- 3687
3 6
C 2H 6
2 4
1 CO2 2 or ethane.
CH4
n [mmol/g]
0 0
0 50 100 0 50
0.8 0.8
C2H6
0.4 0.4
CO2
CH4
0 0
0 50 100 0 50
p [kPa]
Fig. 1. Carbon dioxide, ethane, and methane adsorption isotherms for: (a) BPL
activated carbon at 301 K (Reich et al., 1980), (b) 13X zeolite at 305 K (Choudhary
et al., 1995), (c) ALPO-5 at 305 K (Choudhary and Mayadevi, 1996a), and (d) ZSM-5
at 298 K (He and Seaton, 2006).
ETS-10 is a large-pored titanosilicate molecular sieve with an av-
erage pore size of about 0.8 nm (Kuznicki, 1991). Carbon dioxide and
light hydrocarbons are all small enough to enter the pores of ETS-10
without constrictions, and no size-selectivity effects can be expected
for the separation of carbon dioxide from methane or ethane. How-
ever, the shape of the adsorption isotherms and the strength of the
adsorption for different gases on ETS-10 can be widely modified by
ion-exchange (Anson et al., 2008), and this characteristic could be
exploited to design appropriate adsorbents for PSA-based CO2 re-
moval processes.
ETS-4 is a synthetic titanosilicate molecular sieve with a crys-
talline structure related to the mineral zorite and a pore size of
0.3–0.5 nm (Kuznicki, 1990). Upon appropriate ion exchange (usu-
ally with strontium or barium) followed by temperature treatment,
ETS-4 has been reported to be a size-selective adsorbent for small
molecules with similar size, such as O2 /N2 , O2 /Ar, N2 /CH4 , CH4 /C2 H6 ,
and C2 H4 /C2 H6 (Kuznicki et al., 1999, 2000, 2001; Kuznicki and Bell,
2003). This phenomenon forms the basis of a pressure swing nitro-
gen rejection (from methane) process finding expanding use in the
United States (Butwell et al., 2001). Because the kinetic diameter of
CO2 is smaller than those of N2 , O2 , Ar, CH4 , and C2 H6 , adsorptive
separation of CO2 from light hydrocarbons by size-selectivity may
be possible on ETS-4.
Reduced pore zorite (RPZ), a synthetic titanium silicate, is a struc-
tural analog of ETS-4 and the mineral zorite (Kuznicki et al., 2008).
The size-selectivity and adsorption characteristics of RPZ can be read-
ily tailored. The pore size of RPZ in the regime of 3–5 Å is controlled
by the level of structural chloride ions (Lin et al., 2009). As the con-
centration of chloride ions increases in the framework, the pores of
RPZ constrict progressively and systematically, which allows stable,
size-selective materials to be prepared. In addition to size-selectivity,
ion-exchange of the extra-framework cations can also modify the
isotherm profiles and adsorption strength for selected gases (Lin
et al., 2009). It may, therefore, be possible to combine these features
to prepare a highly selective adsorbent for PSA-based separations of
carbon dioxide from methane and/or ethane.
In this work, we study the adsorption of CO2 , C2 H6 , and CH4 on
different ion exchanged forms of ETS-10, ETS-4, and RPZ. The shape
CO2
of the adsorption isotherms and adsorptive selectivity is modified
by controlling the strength of the adsorption sites and/or controlling
pore size to induce size-selectivity, in order to prepare enhanced
C2H6 adsorbents for the PSA separation of carbon dioxide from methane
CH4
2. Experimental
100
ETS-4 and ETS-10 were synthesized hydrothermally as previously
reported (Kuznicki, 1990, 1991). RPZ titanosilicate was synthesized
by mixing 25.1 g sodium silicate (28.8% SiO2 , 9.14% Na2 O, Fisher),
CO2
4.6 g sodium hydroxide (97+% NaOH, Fisher), 3.0 g KCl (anhydrous,
Fisher), and 16.3 g TiCl3 (Fisher) solution, as previously reported (Lin
C 2H6 et al., 2009; Kuznicki et al., 2008). The mixture was stirred in a
blender for 1 h, and then reacted in a 125 ml sealed autoclave (PARR
Instruments) at 200 ◦ C for 48 h. The resultant material was washed
CH4
with de-ionized water and dried in an oven at 100 ◦ C. All molecular
sieves were reduced to fine powders (< 150 m).
Complete ion-exchange was performed by exposing the as-
100
synthesized molecular sieves to an excess of aqueous ionic solution
(i.e. BaCl2 , ZnCl2 , CuSO4 ) at 100 ◦ C with stirring for 24 h. The fully
exchanged materials were washed with de-ionized water and dried
at 100 ◦ C.
The mixed cationic forms of ETS-4 (Ba/H, Ca/H), ETS-10 (Ba/H),
and RPZ (Ca/H) were prepared by first exposing the as-synthesized
molecular sieves to concentration-controlled aqueous solutions of
BaCl2 or CaCl2 at 100 ◦ C with stirring for 24 h. These solutions were
prepared by mixing 1, 4, or 6 meq/g of BaCl2 or CaCl2 with 10 g of
deionized water per gram of molecular sieve. The partially exchanged
materials were then exposed to an HCl solution at pH 2 for 16 h at
20 ◦ C. The final products were washed with de-ionized water, and
dried at 100 ◦ C.
Carbon dioxide, ethane, and methane adsorption isotherms at
25 ◦ C and up to 1 bar were measured in an Autosorb-1-MP volumet-
ric system from Quantachrome Instruments, Boynton Beach, FL. This
system is equipped with low pressure range transducers (0–1 Torr)
that allow the direct measurements of very low pressure points of
the adsorption isotherms, including those corresponding to the ini-
tial linear region (Henry's law regime) of carbon dioxide isotherms.
Samples were dried at 200 ◦ C for 6 h under vacuum before the ad-
sorption experiments, unless otherwise indicated.
Henry's law constants (K) were determined directly from the lin-
ear region of the isotherms, and were also calculated from the Lang-
muir equation applied to the low pressure adsorption data. Both
methods for the determination of the Henry's law constants gave
equivalent results. The limiting selectivity () of the adsorbent for
the gas A over the gas B is calculated as the ratio of their respective
Henry's law constants:
KA
(A/B) = .
KB
In the present report, the swing capacity (or working capacity)
of a given adsorbent is defined as the adsorption capacity between
pressures of 1–100 kPa. A partial pressure of 1 kPa was selected to
allow a common basis of comparison and because highly rectangular
isotherms tend to hold a significant portion of their capacity below
1 kPa. Working capacity percentage is normalized with respect to
the total adsorption at 1 bar.
3. Results and discussion
Fig. 2 depicts carbon dioxide, ethane, and methane adsorption
isotherms for three different cationic forms of ETS-10: Na-, Ba-,
 A. Anson et al. / Chemical Engineering Science 64 (2009) 3683 -- 3687 3685

2.5 2.5 2.5

2 2 CO2 2
CO2
CO2
n [mmol/g]

1.5 1.5 1.5

C2H6
C2H6 C 2H6
1 1 1

0.5 CH4 0.5 CH4 0.5

CH4
0 0 0
0 50 100 0 50 100 0 50 100
p [kPa]
0.5 0.5 0.5
CO2 CO2
0.4 0.4 0.4
C2H6
n [mmol/g]

0.3 0.3 0.3

C2H 6
0.2 CO2
0.2 0.2

0.1 0.1 0.1

CH4 CH4 C2H6
CH4
0 0 0
0 0.5 1 0 0.5 1 0 0.5 1
p [kPa]

Fig. 2. Carbon dioxide, ethane, and methane adsorption isotherms at 25 ◦ C for: (a) Na-ETS-10, (b) Ba-ETS-10, (c) Ba/H-ETS-10 (1 meq Ba/g) up to 100 kPa, (d) Na-ETS-10, (e)
Ba-ETS-10, and (f) Ba/H-ETS-10 (1 meq Ba/g) up to 1 kPa.

and Ba/H-ETS-10 (1 meq Ba/g). Isotherms for Na-ETS-10 (Fig. 2a) are Table 1
Henry's law constants (K) and limiting selectivity () for the adsorption of carbon
similar in shape to those for zeolite 13X (Fig. 1b), and follow the
dioxide, ethane, and methane on various titanosilicate molecular sieves.
capacity sequence: CO2 > C2 H6 > CH4 . Carbon dioxide and ethane
isotherms for Ba-ETS-10 (Fig. 2b) are less rectangular than for Na- K (mmol g−1 kPa−1 )  (CO2 /C2 H6 )  (CO2 /CH4 )
ETS-10. This effect is opposite to what is generally observed in clas- CO2 C2 H6 CH4
sical zeolites exchanged with divalent cations, such as Ca2+ , Sr2+ ,
Na-ETS-10 5.2 9.5×10−1 1.6×10−2 5.5 3.2×102
or Ba2+ (Breck, 1974). For Ba/H-ETS-10 (1 meq Ba/g), carbon dioxide Ba-ETS-10 2.1 6.2×10−1 1.2×10−2 3.4 1.7×102
and ethane isotherms are even less rectangular than for fully ex- Ba/H-ETS-10a 1.2 1.4×10−1 7.8×10−3 8.8 1.5×102
changed Ba-ETS-10. This result is in agreement with observations for Ba-ETS-4 34 1.1×10−1 9.8×10−3 3.1×102 3.5×103
classical zeolites, where H+ is the weakest of the possible cationic Ca/H-ETS-4b 1.3 7.8×10−2 4.4×10−3 16 2.8×102
Zn-ETS-4 1.6 3.6×10−1 9.7×10−3 4.6 1.7×102
adsorption sites.
Cu-ETS-4 1.3 3.2×10−1 8.6×10−3 4.2 1.5×102
The Henry's law constants for the adsorption of carbon dioxide, Ba-RPZ 22 8.0×10−3 0.7×10−3 2.8×103 3.2×104
ethane, and methane on ETS-10 materials decrease with cation ex- Ca/H-RPZb 7.7 7.2×10−2 1.2×10−3 1.1×102 6.4×103
change in the order of Na > Ba > Ba/H, but the limiting selectivity of Zn-RPZ 2.0 2.1×10−1 6.6×10−3 9.2 3.0×102
CO2 over ethane and methane does not undergo appreciable changes Cu-RPZ 9.5×10−1 2.0×10−1 5.1×10−3 4.8 1.8×102

(Table 1). The swing capacity of CO2 for ETS-10 materials increases a
1 meq Ba2+ /g.
in the order of Na < Ba < Ba/H (Table 2). More than 90% of the total 4 meq Ca2+ /g.
b

CO2 capacity for Ba/H-ETS-10 (1 meq Ba/g) is observed inside the gas
pressure region of 1–100 kPa, yet this material still demonstrates CO2
selectivity over ethane (nearly 9) and high selectivity over methane
(around 150). From this, Ba/H-ETS-10 (1 meq Ba/g) could be an
effective adsorbent for the PSA separation of CO2 from light hydro-
carbons, especially methane. Fully exchanged Ba-ETS-10 also demon-
strates a large CO2 swing capacity, and could also be appropriate for
use in the PSA separation of CO2 from CH4 , under certain conditions.
The shape of the adsorption isotherms for ETS-4 and RPZ can be
modified, much like ETS-10, by cation exchange. However, some ex-
changed forms of ETS-4 and RPZ are not stable to thermal activa-
tion, and unlike ETS-10, cannot be utilized as effective adsorbents.
Na-ETS-4 losses its crystallinity upon mild temperature treatments
(Kuznicki et al., 2001), as do other ETS-4 materials exchanged with
monovalent cations. In order to induce thermal stability for activa-
tion (and adsorption), ETS-4 and RPZ require exchanges with large
size di- or tri-valent cations such as Ba2+ or La3+ (Kuznicki, 1990).
Fig. 3 depicts the carbon dioxide isotherms for Ba/H-ETS-4 materials
exchanged to different levels of Ba2+ and H+ and activated at dif-
ferent temperatures before adsorption measurements. The CO2 ca-
pacity for Ba/H-ETS-4 (1 meq Ba/g) depends on the temperature of
activation (Fig. 3), because the interatomic distances and the pore
sizes of ETS-4 progressively shrink with increased activation tem-
perature (Kuznicki et al., 2001). Ba/H-ETS-4 (4 meq Ba/g), Ba/H-ETS-4
(6 meq Ba/g), and fully exchanged Ba-ETS-4 were found to be stable
upon heating up to 150–200 ◦ C, at which point they are sufficiently
activated for adsorption. Differences in the CO2 adsorption capaci-
ties (mmol/g) observed for thermally stable materials are influenced
by the charge and mass density of the cationic sites.
As with Ba2+ , exchange in the presence of at least 4 meq of Sr2+ ,
Ca , Zn2+ , or La3+ per gram of Na-ETS-4 stabilizes the ETS-4 crystal
2+
3686 A. Anson et al. / Chemical Engineering Science 64 (2009) 3683 -- 3687

structure and allows activation at 150–200 ◦ C. The levels of di- or Fig. 4 depicts carbon dioxide, ethane, and methane adsorption
tri-valent cations that are needed to stabilize the crystal structure of isotherms for Ba-, Ca/H-, and Zn-ETS-4. Sieving properties of ETS-4
RPZ are similar to those for ETS-4. are clearly observed from the ethane isotherms. Ethane adsorption
capacity is low in all cases and decreases with cation exchange in
the order of Zn > Ca/H > Ba, indicating a progressive reduction in
Table 2
the pore size of the adsorbent with exchange. The carbon dioxide
Swing capacity (1–100 kPa) for the adsorption of carbon dioxide, ethane, and
methane on various titanosilicate molecular sieves. isotherm for Ba-ETS-4 is nearly rectangular, with swing capacity of
only 37% under test conditions. CO2 isotherms for Ca/H- and Zn-ETS-
CO2 (mmol/g) % C2 H6 (mmol/g) % CH4 (mmol/g) %
4 are less rectangular and show swing capacities in the order of 80%
Na-ETS-10 1.14 57 0.82 61 0.76 98 (Table 2).
Ba-ETS-10 1.78 77 1.26 72 0.64 98 The CO2 /C2 H6 Henry's law selectivity for Ba-ETS-4 is approxi-
Ba/H-ETS-10a 1.88 91 1.14 90 0.51 99
mately 310, which is nearly two orders of magnitude higher than for
Ba-ETS-4 0.78 37 0.04 66 0.53 98
Ca/H-ETS-4b 1.94 82 0.24 85 0.27 98 ETS-10 materials (Table 1), indicating significant enhancement from
Zn-ETS-4 1.59 76 0.53 73 0.49 98 the size exclusion of ethane. The limiting selectivity of CO2 /C2 H6 for
Cu-ETS-4 1.36 76 0.54 75 0.44 98 ETS-4 materials decreases with the extra-framework cation type in
Ba-RPZ 0.44 34 0.01 80 0.04 98 the order of Ba > Ca/H > Zn ∼ Cu (Table 1). Ca/H-ETS-4 appears to
Ca/H-RPZb 1.13 60 0.12 79 0.11 99
Zn-RPZ 1.36 73 0.41 76 0.33 98
partially exclude ethane and demonstrates a CO2 /C2 H6 limiting se-
Cu-RPZ 1.30 81 0.49 79 0.28 98 lectivity of 16. The selectivities of Zn- or Cu-ETS-4 are similar to the
values obtained for ETS-10, indicating little if any enhancement by
a
1 meq Ba2+ /g.
size-selectivity. From this group, Ca/H-ETS-4 exchanged with at least
b
4 meq Ca2+ /g.
4 meq of Ca2+ per gram of sample seems to be the most appropriate
ETS-4 material for use in the PSA separation of carbon dioxide and
ethane.
Values of CO2 /CH4 selectivity for ETS-4 materials also reflect size
2.5
effects, although these effects are much smaller than in the case
of CO2 /C2 H6 . CH4 molecules are smaller than C2 H6 , and the pores
(a)
of ETS-4 are too large to constrict the passage of CH4 in stabilized
2 (b) materials, including fully exchanged Ba-ETS-4. The limiting CO2 /CH4
selectivity for ETS-4 materials decreases with cation exchange in the
following order: Ba > Ca/H > Zn ∼ Cu (Table 1). Because of its nearly
rectangular isotherms for CO2 , the selectivity for Ba-ETS-4 is quite
n [mmol/g]

1.5
(c) high, reaching approximately 3500, but values for the other cation
exchanged ETS-4 are similar to those obtained for ETS-10 materials.
Taking into account CO2 swing capacities (Table 2) and CO2 /CH4
1
(d) selectivity values (Table 1), Ca/H- and Zn-ETS-4 would appear to be
the best adsorbents for the PSA separation of carbon dioxide and
(e) methane.
0.5 Size effects for the adsorption of CO2 /C2 H6 and CO2 /CH4 on
RPZ titanosilicates are greater than on ETS-4 containing equivalent
(f)
extra-framework cations. Fig. 5 represents carbon dioxide, ethane,
0 and methane adsorption isotherms for Ba-, Ca/H-, and Zn-RPZ. The
0 20 40 60 80 100 methane adsorption capacity for Ba-RPZ is very low due to the size
exclusion of CH4 molecules, while CO2 is still adsorbed. CO2 /C2 H6
p [kPa]
and CO2 /CH4 limiting selectivity values for RPZ materials are uni-
Fig. 3. Carbon dioxide adsorption isotherms for: (a) Ba/H-ETS-4 (6 meq Ba/g) dried formly higher than for ETS-4 exchanged with the same cations
at 200 ◦ C, (b) Fully exchanged Ba-ETS-4, (c) Ba/H-ETS-4 (4 meq Ba/g) dried at 150 ◦ C, (Table 1). From this, it appears that RPZ channels are more con-
(d) Ba/H-ETS-4 (1 meq Ba/g) dried at 100 ◦ C, (e) Ba/H-ETS-4 (1 meq Ba/g) dried at stricted than the ETS-4 channels. For Ba- and Ca/H-RPZ (4 meq Ca/g),
120 ◦ C, and (f) Ba/H-ETS-4 (1 meq Ba/g) dried at 200 ◦ C. the selectivity values for both CO2 /CH4 and CO2 /C2 H6 are

2.5 2.5 2.5

CO2 CO2
2 2 CO2 2
n [mmol/g]

1.5 1.5 1.5

1 1 1 C2H6
CH4
0.5 0.5 C2H6 0.5
C2 H 6 CH4
CH4
0 0 0
0 50 100 0 50 100 0 50 100
p [kPa]

Fig. 4. Carbon dioxide, ethane, and methane adsorption isotherms at 25 ◦ C for: (a) Ba-ETS-4, (b) Ca/H-ETS-4 (4 meq Ca/g), and (c) Zn-ETS-4.
 A. Anson et al. / Chemical Engineering Science 64 (2009) 3683 -- 3687
2 2
n [mmol/g] 1.5 CO2 1.5
1 1
0.5 0.5
CH4
C 2H 6
0 0
0 50 100
Fig. 5. Carbon dioxide, ethane, and methane adsorption isotherms at 25 ◦ C for: (a) Ba-RPZ, (b) Ca/H-RPZ (4 meq Ca/g), and (c) Zn-RPZ.
approximately an order of magnitude higher than for Ba- and Ca/H-
ETS-4 (4 meq Ca/g). The carbon dioxide adsorption isotherm for
Ba-RPZ is essentially rectangular, but the CO2 swing capacity im-
proves for Ca/H-RPZ (4 meq Ca/g) (Table 2). Ca/H-RPZ (4 meq Ca/g)
exchanged with at least 4 meq of Ca2+ per gram of the Na-RPZ
precursor could be a promising adsorbent for the PSA separation of
carbon dioxide from ethane, methane, or both.
4. Conclusions
Titanosilicate molecular sieves (ETS-10, ETS-4, and RPZ) can be
modified by cation exchange to prepare adsorbents with the appro-
priate characteristics of selectivity and swing capacity for the PSA
separation of carbon dioxide from C2 H6 or CH4 . Large-pored ad-
sorbents such as Ba/H-ETS-10 (1 meq Ba/g) with low Ba2+ content
could be utilized for the separation of CO2 from CH4 by site modifi-
cation. This effect can be further enhanced by size-selectivity in the
small-pored titanosilicates ETS-4 and RPZ. While adsorbing substan-
tial amounts of CO2 , stabilized ETS-4 can exclude C2 H6 by size and
similarly modified RPZ materials can exclude both C2 H6 and CH4 .
Ca/H-ETS-4 and Ca/H-RPZ containing high levels of Ca2+ appear to
be excellent candidates for the PSA separation of CO2 from C2 H6 and
CH4 .
Acknowledgements
The authors thank Dr. Amy Dambrowitz for assistance with
manuscript development. Support from the Natural Sciences and
Engineering Research Council Industrial Research Chair in Molecu-
lar Sieve Separations Technology (S.M.K.) and the Alberta Ingenuity
Fund Chair in Separation Technology for Oilsands Extraction and
Upgrading (S.M.K.) is gratefully acknowledged.
References
Anson, A., Wang, Y., Lin, C.C.H., Kuznicki, T.M., Kuznicki, S.M., 2008. Adsorption
of ethane and ethylene on modified ETS-10. Chemical Engineering Science 63,
4171–4175.
3687
2
CO2 CO2
1.5
1
C2H6
C2 H6
0.5
CH4
CH4
0
0 50 100 0 50 100
p [kPa]
Breck, D.W., 1974. Zeolite Molecular Sieves: Structure, Chemistry and Use. Wiley-
Interscience, Wiley, London.
Butwell, K.F., Dolan, B.B., Kuznicki, S.M., 2001. Selective removal of nitrogen from
natural gas by pressure swing adsorption. US Patent no. 6315817.
Cavenati, S., Grande, C.A., Rodrigues, A.E., 2006. Removal of carbon dioxide from
natural gas by vacuum pressure swing adsorption. Energy and Fuels 20,
2648–2659.
Choudhary, V.R., Mayadevi, S., Singh, A.P., 1995. Sorption isotherms of methane,
ethane, ethane and carbon dioxide on NaX, NaY and Na-mordenite zeolites.
Journal of the Chemical Society, Faraday Transactions 91, 2935–2944.
Choudhary, V.R., Mayadevi, S., 1996a. Sorption isotherms of methane, ethane,
ethylene, and carbon dioxide on ALPO-5 and SAPO-5. Langmuir 12, 980–986.
Choudhary, V.R., Mayadevi, S., 1996b. Adsorption of methane, ethane, ethylene, and
carbon dioxide on silicalite-1. Zeolites 17, 501–507.
Collie, J., Hlavinka, M., 1998. An analysis of BTEX emissions from amine sweetening
and glycol dehydration facilities. In: 1998 Laurance Reid Gas Conditioning
Conference Proceedings, Norman, Oklahoma, University of Oklahoma.
He, Y., Seaton, N.A., 2006. Heats of adsorption and adsorption heterogeneity for
methane, ethane, and carbon dioxide in MCM-41. Langmuir 22, 1150–1155.
Kuznicki, S.M., 1990. Preparation of small-pored crystalline titanium molecular sieve
zeolites. US Patent no. 4,938,939.
Kuznicki, S.M., 1991. Large-pored crystalline titanium molecular sieve zeolites. US
Patent no. 5,011,591.
Kuznicki, S.M., Bell, V.A., Petrovic, I., Blosser, P.W., 1999. Separation of nitrogen from
mixtures there of with methane utilizing barium exchanged ETS-4. US Patent
no. 5,989,316.
Kuznicki, S.M., Bell, V.A., Petrovic, I., Desai, B.T., 2000. Small-pored crystalline
titanium molecular sieve zeolites and their use in gas separation processes. US
Patent no. 6,068,682.
Kuznicki, S.M., Bell, V.A., Nair, S., Hillhouse, H.W., Jacubinas, R.M., Braunbarth, C.M.,
Toby, B.H., Tsapatsis, M., 2001. A titanosilicate molecular sieve with adjustable
pores for size-selective adsorption of molecules. Nature 412, 720–723.
Kuznicki, S.M., Bell, V.A., 2003. Olefin separations employing CTS molecular sieves.
US Patent no. 6,517,611 B1.
Kuznicki, S.M., J. A. Sawada, J.A., Rode, E.J., Lin, C.C.H., 2008. Silicate materials,
method for their manufacture, and the method for using such silicate materials
for adsorptive fluid separations. International Patent no. WO 2008/002463, 2008.
Laukhuf, W.L.S., Plank, C.A., 1969. Adsorption of carbon dioxide, acetylene, ethane,
and propylene on charcoal at near room temperatures. Journal of Chemical and
Engineering Data 14, 48–51.
Lin, C.C.H., Sawada, J.A., Wu, L., Haastrup, T., Kuznicki, S.M., 2009. Anion-controlled
pore size of titanium silicate molecular sieves. Journal of the American Chemical
Society 131, 609–614.
Rao, A.B., Rubin, E.S., 2002. Details of a technical, economic and environmental
assessment of amine-based CO2 capture technology for power plant greenhouse
gas control. Environmental Science and Technology 36, 4467–4475.
Reich, R., Ziegler, W.T., Rogers, K.A., 1980. Adsorption of methane, ethane, and
ethylene gases and their binary and ternary mixtures and carbon dioxide on
activated carbon at 212–301 K and pressures to 35 atmospheres. Industrial and
Engineering Process Design and Development 19, 336–344.