Removal of low concentration CO

2
at ambient temperature using several
potassium-based sorbents
Chuanwen Zhao, Yafei Guo, Changhai Li, Shouxiang Lu

State Key Laboratory of Fire Science, University of Science and Technology of China, Hefei 230026, China
h i g h l i g h t s
Several potassium-based sorbents
were applied to low concentration
CO
2
removal at ambient temperature.
K
2
CO
3
/AC shows high carbonation
conversion efciency and CO
2
sorption capacity.
K
2
CO
3
/5A, K
2
CO
3
/13X, and K
2
CO
3
/SG
show low CO
2
sorption capacities.
K
2
CO
3
/Al
2
O
3
shows low carbonation
conversion efciency and needs high
temperature of 350 C to be
regenerated.
K
2
CO
3
/AC is a more efcacious choice
for CO
2
removal in conned space.
g r a p h i c a l a b s t r a c t
a r t i c l e i n f o
Article history:
Received 30 December 2013
Received in revised form 12 February 2014
Accepted 15 February 2014
Keywords:
CO
2
removal
Conned space
Potassium-based sorbent
Porous support
a b s t r a c t
The requirement for long-duration human operation in a conned space has made removal of low con-
centration CO
2
a critical technology. The incorporation of organic amines into a porous support is thought
to be a promising approach, but the low amine utilization ratio and the loss of amine components due to
evaporation in regeneration process make it necessary to try new sorbents. As K
2
CO
3
are more difcult to
be decomposed than organic amine compound, potassium-based sorbents may be more effective for CO
2
removal in conned spaces. In this work, activated carbon (AC), Al
2
O
3
, zeolite 5A, zeolite 13X, and silica
aerogels (SG) were chosen as the supports and K
2
CO
3
was provided as the active component. An exper-
imental demonstration of the CO
2
sorption performances of these sorbents was present in detail in the
condition of ambient temperature and the CO
2
concentration of 5000 ppm. The CO
2
sorption capacities
are calculated as 0.87, 1.18, 0.34, 0.53, and 0.15 mmol CO
2
/g for K
2
CO
3
/AC, K
2
CO
3
/Al
2
O
3
, K
2
CO
3
/5A,
K
2
CO
3
/13X, and K
2
CO
3
/SG, respectively. The reacted products are completely regenerated in the temper-
ature range of 100200 C for K
2
CO
3
/AC and K
2
CO
3
/SG. Other sorbents, however, require a higher temper-
ature of 350 C in order to be regenerated. K
2
CO
3
/Al
2
O
3
shows the highest CO
2
sorption capacity, while
K
2
CO
3
/AC shows the highest bi-carbonation conversion efciency. The CO
2
sorption capacities of
K
2
CO
3
/5A, K
2
CO
3
/13X, and K
2
CO
3
/SG do not reach the expected values. Among these sorbents, K
2
CO
3
/
AC is a new, more efcacious choice for CO
2
removal in conned space at ambient temperature.
2014 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.apenergy.2014.02.054
0306-2619/ 2014 Elsevier Ltd. All rights reserved.

Corresponding author. Tel.: +86 551 63603141; fax: +86 551 63601669.
E-mail address: sxlu@ustc.edu.cn (S. Lu).
Applied Energy 124 (2014) 241247
Contents lists available at ScienceDirect
Applied Energy
j our nal homepage: www. el sevi er . com/ l ocat e/ apener gy
1. Introduction
The requirement for long-duration human operation in a con-
ned space such as underground warehouses, mines, submarines
and space capsules, has made removal of low concentration CO
2
a critical technology [1]. The primary method used to remove
CO
2
from enclosed inhabited environments until the early 1990s
was based on LiOH [2]. Although the CO
2
storage capacity of LiOH
is high, the material cannot be practically regenerated. The long-
term occupation of a space station or a submarine would require
a CO
2
sorbent that can be easily regenerated to reduce launch
weight and storage volume [3]. Recently, the incorporation of or-
ganic amines into a porous support is thought to be a promising
approach, and this process has been widely studied by many
researchers [431]. Various amine solutions were used as the ac-
tive components. These divide into three categories: primary, sec-
ondary, and tertiary amine groups. The widely used supports
include polymers [46], various silica material (SBA series [710],
MCM series [1115], Silica gel [1620], and KIT series [21]), carbon
materials [2227], and zeolites [2831]. Although various amine-
based sorbents show high CO
2
sorption capacities, the utilization
of amine compound is not high for all sorbents. Their amine utili-
zation ratio was reported as only 2067%. Another signicant
drawback for amine-based sorbents is loss of amine components
due to evaporation when the sorbent is regenerated at a tempera-
ture higher than 100 C. The requirement for solving these prob-
lems makes it necessary to test new sorbents.
Recently, carbon capture and storage is recognized as a prom-
ising option for CO
2
emission reduction [3236], and alkali metal
carbonates can be used to capture CO
2
in the presence of H
2
O to
form either sodium or potassium bicarbonate at temperatures
below 100 C [37]. A moderate temperature swing of 120
200 C then causes the bicarbonate to decompose and release a
mixture of CO
2
/H
2
O that can be converted into a sequestra-
tion-ready CO
2
stream by subsequently condensing the steam.
Compared with the amine-based sorbent, alkali metal carbonates
are more difcult to be decomposed; this is even the case if the
temperature is increased to 800 C. These sorbents may therefore
be more effective for CO
2
removal in conned space. If the alkali
metal-based sorbents are used for this technology, they should
feature high sorption capacities at low CO
2
partial pressures
and at ambient temperature. However, in previous studies most
researchers have only examined the CO
2
capture behaviors of al-
kali metal-based sorbents under the reaction conditions of 50
100 C, 520% CO
2
, 520% H
2
O [3746]. It is unclear whether
the carbonation reaction paths change when the temperature is
the ambient temperature and the CO
2
concentration is lower
than 1%. It is also unknown if their CO
2
sorption capacities keep
high values in this situation.
Firstly, it was reported that K
2
CO
3
/AC [3841,43,45] (acti-
vated carbon) and K
2
CO
3
/Al
2
O
3
[4246] had the potential to be
employed as suitable sorbents for CO
2
capture in ue gas from
fossil fuel-red power plants due to their high CO
2
capture
capacities and rapid carbonation reaction rates. Secondly, the
zeolites [2831,4750] possess well-dened pore structures and
their pore sizes are of the same magnitude as CO
2
molecules,
they are therefore expected to have good afnities for CO
2
cap-
ture, and these pore sizes can accommodate both the immobi-
lized alkali metal carbonates and the adsorbed CO
2
. Thirdly,
silica aerogels (SG) possessing an open pore structure, a high
porosity, a high surface area and a low density, have a great po-
tential for utilization as CO
2
adsorbents [1620]. Nonetheless in
order to reach high adsorption capacities, SG should be function-
alized. Here, AC, Al
2
O
3
, zeolite 5A, zeolite 13X, and SG were cho-
sen as the supports, and K
2
CO
3
was chosen as the active
component. An experimental demonstration of the CO
2
sorption
performances of these sorbents was presented in detail in the
condition of ambient temperature and the CO
2
concentration of
5000 ppm. Additionally, the CO
2
sorption capacities of these
potassium-based sorbents were compared with other sorbents
such as amine-tethered solid sorbents.
2. Experimental section
2.1. Samples
The supported potassium-based sorbents were prepared by
impregnating K
2
CO
3
on AC, Al
2
O
3
, zeolite 5A, and 13X, and SG,
respectively. K
2
CO
3
was provided as an analytical reagent. Al
2
O
3
and AC were supplied by Research Institute of Nanjing Chemical
Industry Group. Zeolite 5A and 13X were supplied by Shanghai
Zeolite Molecular Sieve Co., LTD. SG was supplied by Shanghai
Shangchen Industrial Co., LTD. The preparation process of these
sorbents consisted of three steps as follows: (1) 10 g of support
material were added to an aqueous solution of anhydrous K
2
CO
3
in 100 ml of deionized water. Then it was mixed with a magnetic
stirrer for 20 h at room temperature. (2) After stirring, the mixture
was dried in a rotary vacuum evaporator at 105 C for dehydration.
(3) The dried samples were calcined in a furnace under a N
2
ow
(100 ml/min) for 3 h at 300 C.
All sorbents were prepared with theoretical K
2
CO
3
loadings of
30%. However, the actual loadings are 12.49%, 28.5%, 11.2%,
16.2%, and 37.5% for K
2
CO
3
/AC, K
2
CO
3
/Al
2
O
3
, K
2
CO
3
/5A, K
2
CO
3
/
13X, and K
2
CO
3
/SG, respectively.
2.2. Apparatus and procedure
The CO
2
sorption performances of these sorbents were assessed
using a modied xed bed reactor system coupled with a gas ana-
lyzer. This experimental setup is shown in Fig. 1.
The atmosphere was simulated by a constant ow of CO
2
,
H
2
O and a balance of N
2
. CO
2
and N
2
were obtained from
high-purity cylinders and delivered to the experimental appara-
tus with mass ow controllers. H
2
O was fed using a high-preci-
sion syringe pump and then heated to ensure complete
vaporization. The xed-bed reactor with an inner diameter of
0.05 m was made of stainless steel and was set in an electrically
heated furnace. The electric heating was used to heat the reactor
to the given temperature. A water pipe was buried in the cast
copper. This water-cooling heat exchanger was used to keep
the temperature constant, when the exothermic reaction oc-
curred. A gravimetric sensor was used to detect the change of
the sorbents weight online. The Relative Humidity (RH) and
temperature of the simulated atmosphere were also monitored
online by a hygrothermograph. A gas analyzer was used to mea-
sure the CO
2
concentration of the outlet gas continuously. A total
of 10 g of the sorbent was placed into the reactor, and the height
of the material was 0.01 m. A simulated gas composition consist-
ing of 5000 ppm CO
2
+ 1.8% H
2
O at a total ow rate of 0.05 m
3
/h
was used for the carbonation reactions. The reaction temperature
is 20 C. When the carbonation reaction was nished, the gas
composition was changed to 100% N
2
, and the reaction temper-
ature was increased to 400 C. In this way, the regeneration pro-
cesses of these sorbents were carried out.
The amount of K
2
CO
3
impregnated was determined by an
Inductively Coupled Plasma Mass Spectrometer. An ASAP 2020 sys-
tem with N
2
adsorptiondesorption was used for surface area and
pore structure determination.
242 C. Zhao et al. / Applied Energy 124 (2014) 241247
3. Results and discussion
3.1. Microscopic structure of different sorbents
Because the microscopic structures of these sorbents will play
an important role in their CO
2
sorption performances. It should
be investigated that how does the microscopic structure change
after K
2
CO
3
impregnated in these supports. To achieve this objec-
tive, all samples were characterized by N
2
adsorptiondesorption
analysis. The specic surface areas of these samples were calcu-
lated by the BrunauerEmmettTeller (BET) method. The pore vol-
umes and the pore size distribution were obtained using
desorption branches of the isotherms. They are calculated by the
BarrettJoynerHalenda (BJH) method for the pores between 1.7
and 300.0 nm, and are calculated by the HorvathKawazoe method
for the pores smaller than 1.7 nm. BET surface areas, pore volumes,
and pore size distributions of these sorbents are shown in Figs. 2
and 3, respectively.
Fig. 3 shows that the specic surface area of AC is much higher
than other supports, but its pore volume is only 0.372 cm
3
/g, which
is lower than other supports. The reason is that most pores of AC
are distributed in the pore size of less than 5 nm which is shown
7
3
4
9
10
1 2
5
F
T P M T RH
6
8
11 12 13 14 15
Fig. 1. Schematic for the experimental system. (1) N
2
; (2) CO
2
; (3) pressure regulating valve; (4) mass ow controller; (5) H
2
O; (6) high-precision syringe pump; (7)
evaporator; (8) mixture tank; (9) xed-bed reactor; (10) gas analyzer; (11) temperature controller; (12) pressure monitor; (13) gravimetric sensor; (14) relative humidity
monitor; (15) temperature meter.
0
250
500
750
1000
454
186
200
564
0.99
71
215
7.89
185
957
B
E
T

s
u
r
f
a
c
e

a
r
e
a
,

m
2
/
g
the supports
the supported sorbents
(a)
1 2 3 4 5
1 2 3 4 5
0
250
500
750
1000
305
930
402
214
5.9
275
212
21
445
372
P
o
r
e

v
o
l
u
m
e
,

1
0
-
3
c
m
3
/
g
the supports
the supported sorbents
(b)
Fig. 2. Microscopic structures of samples: (a) BET surface area, (b) pore volume.
(1) K
2
CO
3
/AC, (2) K
2
CO
3
/Al
2
O
3
, (3) K
2
CO
3
/5A, (4) K
2
CO
3
/13X, (5) K
2
CO
3
/SG.
1 10 100
0.0
0.5
1.0
1.5
d
V
/
d
l
o
g
(
R
P
)
,

c
m
3
/
(
g

n
m
)
Pore radius, nm
AC
Al
2
O
3
5A
13X
SG
(a)
1 10 100
0.0
0.5
1.0
1.5
d
V
/
d
l
o
g
(
R
P
)
,

c
m
3
/
(
g

n
m
)
Pore radius, nm
K
2
CO
3
/AC
K
2
CO
3
/Al
2
O
3
K
2
CO
3
/5A
K
2
CO
3
/13X
K
2
CO
3
/SG
(b)
Fig. 3. Pore size distribution curves for different sorbents (a) the supports, (b) the
supported sorbents.
C. Zhao et al. / Applied Energy 124 (2014) 241247 243
in Fig. 3. After K
2
CO
3
has been loaded on AC, the surface area de-
creases from 957 to 564 m
2
/g, and its pore volume decreases to
0.214 cm
3
/g. Conrmed from the change of pore distribution be-
fore and after loading K
2
CO
3
which is shown in Fig. 3, most active
components are distributed in the pore size of 25 nm. The pore
structure with the pore size of less than 2 nm is not lled.
Most pores are distributed in the pore size of 2150 nm for
Al
2
O
3
and zeolite 5A, and most active components are willing to
distribute in this pore size range. After K
2
CO
3
has been loaded
on, the specic surface area of Al
2
O
3
decreases from 185 to
71 m
2
/g, and its pore volume decreases from 0.445 to 0.27 cm
3
/g.
For zeolite 5A, the specic surface area and pore volume greatly
decrease from 200 m
2
/g and 0.402 cm
3
/g to 7.89 m
2
/g and
0.021 cm
3
/g, respectively.
Most pores of zeolite 13X are distributed in two pore size
ranges of 12 nm and 5100 nm. After K
2
CO
3
has been loaded
on, its specic surface area decreases from 454 to 215 m
2
/g, and
its pore volume decreases from 0.305 to 0.212 cm
3
/g. It is interest-
ing that most K
2
CO
3
are distributed in the pore size of 12 nm,
while the pore structure with the pore size of 5100 nm is not
lled.
The specic surface area of SG is only 186 m
2
/g, but its pore vol-
ume of 0.930 cm
3
/g is much higher than other supports. Most
pores of SG are distributed in the pore size of 5150 nm. After
K
2
CO
3
has been loaded on, its pore volume greatly decreases to
0.006 cm
3
/g, and its specic surface area greatly decreases from
186 to 0.99 m
2
/g. Most pores of SG are blocked.
3.2. CO
2
sorption performances of different sorbents
3.2.1. CO
2
sorption curves of different sorbents
The CO
2
sorption processes were performed in the composition
of 5000 ppm CO
2
, 1.8% H
2
O, and N
2
balanced at 20 C. The H
2
O
concentration (corresponding to the RH of 78%) was chosen to
simulate a typical condition in a submarine. The CO
2
sorption ef-
ciency, which is used for expressing the CO
2
sorption performance
of the sorbent, is calculated by the change of CO
2
concentration be-
tween the outlet gas and the inlet gas. The results are shown in
Fig. 4.
As shown in Fig. 4(a), CO
2
sorption efciencies of all supports
decrease to zero in less than 20 min. The CO
2
sorption capacity
can be easily obtained by integrating the data of their CO
2
removal
curves. The CO
2
sorption capacities are 1.37 10
1
, 1.38 10
1
,
9.48 10
2
, 1.09 10
1
, and 1.20 10
1
mmol CO
2
/g for AC,
Al
2
O
3
, zeolite 5A, zeolite 13X, and SG, respectively. Because CO
2
sorption performances of all supports are mainly depended on
the physical adsorption process in this condition, the microscopic
structure of the sample plays an important role in this process.
Several results were obtained from the adsorption isotherms of
pure CO
2
on adsorbents at 2530 C. It was reported that the CO
2
sorption capacity of AC increased from 0.2 to 8 mmol/g when the
pressure increased from 0.05 to 60 atm [47,5154]. The similar
conclusion was obtained for some zeolite materials. The CO
2
sorp-
tion capacities of zeolite b, 5A, 4A, and 13X increased from 0.3, 4.8,
2.8, and 3.2 to 1.7, 6.3, 4.9, and 5.1 mmol/g respectively, when the
pressure increased from 0.1 to 60 atm at 30 C [30,47,51]. The CO
2
sorption capacity of AC decreased as the temperature increased at
the same pressure of pure CO
2
[47,5154]. Some other results were
obtained by passing the mixture 20% CO
2
+ 80% N
2
at 1 atm total
pressure. The CO
2
sorption capacity decreased from 0.734 to
0.072 mmol/g for AC when the temperature increased from 28 to
150 C [27,55]. The CO
2
sorption capacity of zeolite 13X decreased
from 1.25 to 0.18 mmol/g as the temperature increased from 30 to
120 C [28,29,48]. It can be concluded that the CO
2
sorption capac-
ities of these materials will be greatly affected by temperature and
CO
2
partial pressure. In the reaction conditions of this paper, low
temperature is an advantage factor, but the low CO
2
partial pres-
sure is an adverse factor for CO
2
sorption capacities of these sor-
bents. A chemical active component is needed to improve the
CO
2
sorption capacities of these materials.
Except for K
2
CO
3
/SG, CO
2
sorption capacities are greatly im-
proved for these supported sorbents. The time for nearly 100%
CO
2
removal is longer for K
2
CO
3
/AC as 27 min than that of other
sorbents, and its total reaction time is 132.8 min. The total reaction
time is longest for K
2
CO
3
/Al
2
O
3
as 215 min. Their CO
2
sorption
capacities are calculated as 8.73 10
1
, 1.18, 3.45 10
1
,
5.27 10
1
, and 1.54 10
1
mmol CO
2
/g for K
2
CO
3
/AC, K
2
CO
3
/
Al
2
O
3
, K
2
CO
3
/5A, K
2
CO
3
/13X, and K
2
CO
3
/SG, respectively. It is de-
duced that these potassium-based sorbents can be used to remove
CO
2
at ambient temperature by chemical reactions. Based on that
one mole of K
2
CO
3
can absorb one mole of CO
2
, the theoretical
bi-carbonation conversion efciencies are calculated as 96.4%,
57.3%, 42.5%, 44.9%, and 5.7%, respectively. K
2
CO
3
/Al
2
O
3
shows
the highest CO
2
sorption capacity, while K
2
CO
3
/AC shows the high-
est bi-carbonation conversion efciency. The CO
2
sorption capaci-
ties of K
2
CO
3
/5A, K
2
CO
3
/13X, and K
2
CO
3
/SG do not reach the
expected values.
3.2.2. Mass changes of different sorbents
To achieve the objective of nding out the carbonation reaction
paths of different sorbents at ambient temperature, mass changes
in the reaction process of all sorbents were measured with the
gravimetric sensor. The results are shown in Fig. 5.
Fig. 5(a) shows that the reaction rate and the nal mass change
of AC are the lowest among these support materials. The curves of
mass change are similar for zeolite 5A, zeolite 13X, and SG. Calcu-
lated from the curves of CO
2
removal, the amounts of CO
2
removal
are 60.1, 60.7, 41.7, 47.9, and 52.8 mg for AC, Al
2
O
3
, zeolite 5A, zeo-
lite 13X, and SG, respectively. The nal mass change of each sample
is higher than this value. The extra increment is attributed to the
0
25
50
75
100
C
O
2

s
o
r
p
t
i
o
n

e
f
f
i
c
i
e
n
c
y
,

%
Time, min
AC
Al
2
O
3
5A
13X
SG
(a)
0 6 12 18 24 30
0 60 120 180 240
0
25
50
75
100
C
O
2

s
o
r
p
t
i
o
n

e
f
f
i
c
i
e
n
c
y
,

%
Time, min
K
2
CO
3
/AC
K
2
CO
3
/Al
2
O
3
K
2
CO
3
/5A
K
2
CO
3
/13X
K
2
CO
3
/SG
(b)
Fig. 4. CO
2
sorption performances of different sorbents (a) the supports, (b) the
supported sorbents.
244 C. Zhao et al. / Applied Energy 124 (2014) 241247
adsorption of H
2
O. The amounts of H
2
O adsorption are calculated
as 19.6, 108.2, 265.1, 302.7, and 295.7 mg for AC, Al
2
O
3
, zeolite
5A, zeolite 13X, and SG, respectively. Compared the curve of mass
change with the curves of CO
2
removal, the H
2
O adsorption perfor-
mances can be obtained for these materials. The average H
2
O sorp-
tion efciencies are 10%, 50%, 94%, 94%, and 90% through the whole
CO
2
sorption process for AC, Al
2
O
3
, zeolite 5A, zeolite 13X, and SG,
respectively. This result is in consistent with the results shown in
pure water vapor adsorption isotherms on various sorbents at
25 C [51].
The nal mass changes through the whole CO
2
sorption process
are 541.8, 1175.6, 357.6, 1082.8, and 97.1 mg for K
2
CO
3
/AC, K
2
CO
3
/
Al
2
O
3
, K
2
CO
3
/5A, K
2
CO
3
/13X, and K
2
CO
3
/SG, respectively. Based on
that one mole of K
2
CO
3
can adsorb one mole CO
2
and one mole H
2
O
to produce two mole of KHCO
3
, the theoretical nal mass changes
of these sorbents are 541.2, 733.6, 213.6, 326.7, and 95.5 mg,
respectively, which are calculated from their CO
2
removal curves.
The values obtained from these two curves are similar for K
2
CO
3
/
AC and K
2
CO
3
/SG. On the contrary, the mass changes are much
higher than these theoretical values for K
2
CO
3
/Al
2
O
3
, K
2
CO
3
/5A,
and K
2
CO
3
/13X. It is deduced that, except for the chemical reaction
to remove CO
2
, the adsorption processes of H
2
O take place in the
same time for these samples. It is reported that zeolites and zeo-
lite-based sorbents show preferable water adsorption capacities,
hence their CO
2
sorption capacities are greatly reduced by
moisture [28,29]. The reason is deduced that the increase of the li-
quid lm or water layer on the solid surface leads to a decrease in
the concentration of CO
2
in the liquid lm, or an increase in the
HCO
3

transport resistance and thus a decrease in the interaction

between and the active sites on support [18]. This conclusion is
in consistent with it for zeolite 5A, zeolite 13X, and K
2
CO
3
/13X,
because they show high H
2
O adsorption capacities, and low CO
2
sorption capacities. The H
2
O adsorption capacity of K
2
CO
3
/5A is
much lower than it of zeolite 5A, because its pore volume is greatly
decreased from 402 to 21 cm
3
/g. It was reported that the CO
2
sorption capacity increased in presence of moisture for amine-
modied silica gels [18,20]. However, both the CO
2
sorption capac-
ity and the H
2
O adsorption capacity are much lower than expected
for K
2
CO
3
/SG. The reason is that their microscopic structure is com-
pletely destroyed by the impregnating of K
2
CO
3
. Both the CO
2
sorp-
tion capacity and the H
2
O adsorption capacity are excellent for
K
2
CO
3
/Al
2
O
3
. Its carbonation reaction path is deduced that H
2
O is
adsorbed rst, and then CO
2
reacts with the adsorbed H
2
O and
K
2
CO
3
to produce KHCO
3
. The H
2
O adsorption capacity of AC is
poor. However, it was reported that moisture played a promoting
role in the removal of CO
2
for K
2
CO
3
/AC [40]. The reason is deduced
that the hydration reaction occurs rst and then KHCO
3
is pro-
duced for the CO
2
sorption process of K
2
CO
3
/AC at ambient
temperature.
3.3. Regeneration behaviors of different sorbents
In practical application of CO
2
removal in a closed loop environ-
mental control system, the sorbent after CO
2
sorption process
should be regenerated by increasing temperature, or entering
high-temperature steam. In this section, the effect of temperature
on the regeneration behaviors of these reacted sorbents is investi-
gated, and the results are shown in Fig. 6.
It is shown that the regeneration processes for all sorbents had
nished before the reaction temperature increased to the nal va-
lue of 375 C. The nal mass changes for all sorbents are equal to
the values increased through their whole CO
2
sorption processes.
It means that all sorbents are completely regenerated in this pro-
cess. They are completely regenerated in the temperature range
of 100200 C for K
2
CO
3
/AC and K
2
CO
3
/SG. It was reported that
the decomposition of KHCO
3
started at 117.2 C and ended at
219.7 C [56]. The regeneration processes of K
2
CO
3
/AC and
K
2
CO
3
/SG could be explained by the decomposition of KHCO
3
.
The regeneration processes of K
2
CO
3
/Al
2
O
3
, K
2
CO
3
/5A, and K
2
CO
3
/
13X consists of two steps as the temperature increases: the rst
step is carried out in the temperature range of 60150 C, and
the next step is carried out in the temperature range of 150
350 C. Because a large amount of H
2
O is adsorbed on these sor-
bents, the rst step of regeneration is mainly attributed to the
desorption of H
2
O for K
2
CO
3
/Al
2
O
3
, K
2
CO
3
/5A, and K
2
CO
3
/13X.
KAl(CO
3
)
2
(OH)
2
was found in the carbonation product of K
2
CO
3
/
Al
2
O
3
in a xed-bed reactor [41] and a uidized-bed reactor [46].
The decomposition of KAl(CO
3
)
2
(OH)
2
was reported in the temper-
ature range of 260350 C [41,46]. As a result, it is deduced that
KAl(CO
3
)
2
(OH)
2
was produced in the carbonation process of
K
2
CO
3
/Al
2
O
3
at ambient temperature, its second step of regenera-
tion is attributed to the decomposition of KHCO
3
and KAl(CO
3
)
2
(-
OH)
2
. It was reported that zeolite 13X after CO
2
removal had to
be regenerated at 350 C [20]. It is deduced that the regeneration
0
90
180
270
360

m
,

m
g
Time, min
AC
Al
2
O
3
5A
13X
SG
(a)
0 6 12 18 24 30
0 60 120 180 240
0
400
800
1200

m
,

m
g

Time, min
K
2
CO
3
/AC
K
2
CO
3
/Al
2
O
3
K
2
CO
3
/5A
K
2
CO
3
/13X
K
2
CO
3
/SG
(b)
Fig. 5. Mass changes of different sorbents (a) the supports, (b) the supported
sorbents.
0 25 50 75
-1200
-800
-400
0
T
e
m
p
e
r
a
t
u
r
e
,

o
C

m
,

m
g
Time, min
K
2
CO
3
/AC
K
2
CO
3
/Al
2
O
3
K
2
CO
3
/5A
K
2
CO
3
/13X
K
2
CO
3
/SG
0
100
200
300
400
Fig. 6. Regeneration behaviors of different sorbents.
C. Zhao et al. / Applied Energy 124 (2014) 241247 245
behaviors of K
2
CO
3
/5A and K
2
CO
3
/13X depend mainly on their
supports.
3.4. Comparison of CO
2
sorption capacities for these sorbents with
other sorbents
The CO
2
sorption capacities of these sorbents are compared
with that of potassium-based sorbents with same supports
[3746] which were obtained under the reaction conditions of
50100 C, 520% CO
2
, 520% H
2
O, the support materials
[4755], and amine-modied sorbents with same supports
[1620,2531]. Their CO
2
sorption capacities are listed in Fig. 7.
The CO
2
sorption capacity of AC is in the range of 0.05
0.734 mmol/g. As mentioned above, it is greatly affected by tem-
perature and CO
2
partial pressure. The CO
2
sorption capacity of
K
2
CO
3
/AC is 0.772.32 mmol CO
2
/g under the ue gas from fossil
fuel-red power plants conditions. It of amine-modied AC sorbent
is 0.571.17 mmol/g, which is lower than that of other amine-
based sorbents. The reason is that, most pores of AC are distributed
in the pore size of less than 2 nm, but no active components can ll
in such small pores which lead to a limited loading amount of ac-
tive component. In this paper, AC was lled with 30% aqueous
K
2
CO
3
, but its actual loadings is only 12.49%. As a result, its CO
2
sorption capacity is only 0.87 mmol/g. With an increase in the
ingredient ratio of K
2
CO
3
/AC in preparation, the amount of impreg-
nated K
2
CO
3
could increase to 23.1% [38]. The CO
2
sorption capac-
ity will increase for K
2
CO
3
/AC with higher K
2
CO
3
loadings.
The CO
2
capture capacity of K
2
CO
3
/Al
2
O
3
is 1.822.43 mmol
CO
2
/g under the ue gas from fossil fuel-red power plants condi-
tions. In this paper, K
2
CO
3
/Al
2
O
3
shows the highest CO
2
sorption
capacity of 1.18 mmol/g. However, KAl(CO
3
)
2
(OH)
2
was produced
in the carbonation process of K
2
CO
3
/Al
2
O
3
at ambient temperature,
K
2
CO
3
/Al
2
O
3
after CO
2
removal had to be regenerated at 350 C. To
fulll the specic task for the application in a closed loop environ-
mental control system such as submarines and space craft, regen-
eration of the sorbent should be possible at low temperature to
reduce the energy demand.
Zeolites have shown promising results for separating CO
2
from
gas mixtures and can potentially be used in the pressure swing
adsorption (PSA) process [4750]. The CO
2
sorption capacities are
0.360.5 mmol/g for zeolite 5A, and 0.71.25 mmol/g for zeolite
13X, respectively. Several previous works were to synthesize novel
adsorbents by immobilization of different primary and secondary
amines on zeolite matrix [2831]. This immobilization was ex-
pected to impart high sorption capacity for CO
2
as compared to
the bare zeolite matrix. The CO
2
sorption capacities increased to
0.521.11 and 2.062.56 mmol/g for amine modied zeolite 5A
and 13X, respectively. The CO
2
capture capacities of K
2
CO
3
/zeolites
are only 0.350.52 mmol CO
2
/g under the ue gas from fossil fuel-
red power plants conditions. In this paper, they are only 0.34 and
0.53 mmol CO
2
/g for K
2
CO
3
/5A and K
2
CO
3
/13X, respectively. The
reason is that their CO
2
sorption capacities are greatly reduced
by moisture. For application of potassium-based sorbent, the CO
2
sorption performance is deeply depended on the existing of H
2
O
and its concentration. Moreover, the sorbent can be regenerated
in water vapor in real operations to improve its CO
2
sorption
capacity. As a result, zeolites are not good choices as potassium-
based sorbent for CO
2
removal.
The CO
2
sorption capacity of SG is in the range of 0.75
1.31 mmol/g. One promising way for the improvement of the CO
2
sorption capacity of silica gels is the incorporation of amino into
the silica structure [1620]. Functionalisation of aerogels with
amine groups was also used for improving their strength and elas-
ticity. It was reported that the CO
2
sorption capacity of amine func-
tionalized SG achieved such a high value of 6.97 mmol/g [18].
However, the CO
2
sorption capacity of sorbents prepared by
impregnating K
2
CO
3
on SG is only 0.120.31 mmol/g. The reason
is that their microscopic structure is completely destroyed by the
impregnating process. It is deduced that the CO
2
sorption capacity
of K
2
CO
3
/SG can be improved by optimizing the prepared method
for this sorbent.
Among these sorbents, K
2
CO
3
/AC may be a new choice for CO
2
removal in conned space at ambient temperature as it shows
the highest bi-carbonation conversion efciency. Because K
2
CO
3
/
AC can be easily regenerated at low temperatures of 100200 C,
this sorbent can be used in a continuous CO
2
sorptiondesorption
operation system. The CO
2
sorption performance is mainly deter-
mined by the gas residence time and the reaction rate, as the pro-
cess is carried out in the continuous solid circulation mode.
4. Conclusions
The CO
2
sorption performances of K
2
CO
3
/AC, K
2
CO
3
/Al
2
O
3
,
K
2
CO
3
/5A, K
2
CO
3
/13X, and K
2
CO
3
/SG were detailed investigated
in the condition of ambient temperature and the CO
2
concentration
of 5000 ppm using a modied xed bed reactor system coupled
with a gas analyzer. It is found that the CO
2
sorption capacities
are 1.37 10
1
, 1.38 10
1
, 9.48 10
2
, 1.09 10
1
, and
1.20 10
1
mmol CO
2
/g for AC, Al
2
O
3
, zeolite 5A, zeolite 13X,
and SG, respectively. After K
2
CO
3
has been loaded on, their CO
2
sorption capacities are greatly improved to 0.87, 1.18, 0.34, 0.53,
and 0.15 mmol CO
2
/g, respectively. K
2
CO
3
/AC shows the highest
bi-carbonation conversion efciency, it can be completely regener-
ated in 100200 C. The CO
2
sorption capacity will increase for
K
2
CO
3
/AC with higher K
2
CO
3
loadings. K
2
CO
3
/Al
2
O
3
shows the
highest CO
2
sorption capacity, but it has to be regenerated at
350 C as the product of KAl(CO
3
)
2
(OH)
2
in the CO
2
sorption pro-
cess. Zeolites and amine modied zeolites show promising results
for separating CO
2
from gas mixtures. However, the CO
2
sorption
capacities of K
2
CO
3
/zeolites are only 0.350.52 mmol CO
2
/g, be-
cause their CO
2
sorption capacities are greatly reduced by mois-
ture. The CO
2
sorption capacities of SG and amine functionalized
SG are high. However, the CO
2
sorption capacity of sorbents pre-
pared by impregnating K
2
CO
3
on SG is only 0.120.31 mmol/g, be-
cause their microscopic structure is completely destroyed by the
impregnating process. Among these sorbents, K
2
CO
3
/AC may be a
new choice for CO
2
removal in conned space at ambient temper-
ature. Besides a high CO
2
sorption capacity, a high reaction rate is
needed for the sorbent in such a low temperature and a low CO
2
partial pressure. Future research will investigate the CO
2
sorption
reaction rate of these sorbent at ambient temperature.
Acknowledgments
Financial support from the National Natural Science Foundation
(No. 51206155) and the China Postdoctoral Science Foundation
0
1
2
3
4
V IV III II I
In this work
potassium-based sorbents in Ref. 34-43
support materials in Ref. 44-52
amine-modified sorbents in Ref. 18-22, 27-33
C
O
2

s
o
r
p
t
i
o
n

c
a
p
a
c
i
t
y
,

m
m
o
l

C
O
2
/
g

s
o
r
b
e
n
t
Fig. 7. Comparison of CO
2
sorption capacities for these sorbents with different
support. I: AC, II: Al
2
O
3
, III: zeolite 5A, IV: zeolite 13X, V: SG.
246 C. Zhao et al. / Applied Energy 124 (2014) 241247
(Nos. 2011M501061 and 2012T50579) are sincerely acknowl-
edged. The authors wish to thank Mr. Christopher Love for his help
in English revision of this article.
References
[1] Ge J, Cowan RM, Tu C, McGregor ML, Trachtenberg MC. Enzyme-based CO
2
capture for advanced life support. Life Support Biosphere Sci: Int J Earth Space
2002;8(34):1819.
[2] Mao JF, Han X, Zhang H, Zhou SL. Studies on the performance of CO
2
adsorption
using LiOH combined with activated carbon ber. In: Proceedings 3rd Asian
conference refrigeration air-conditioning, vols. I and II; 2006. p. 297301.
[3] Satyapal S, Filburn T, Trela J, Strange J. Performance and properties of a solid
amine sorbent for carbon dioxide removal in space life support applications.
Energy Fuels 2001;15(2):2505.
[4] Filburn T, Helble JJ, Weiss RA. Development of supported ethanolamines and
modied ethanolamines for CO
2
capture. Ind Eng Chem Res
2005;44(5):15426.
[5] Gray ML, Champagne KJ, Fauth D. Performance of immobilized tertiary amine
solid sorbents for the capture of carbon dioxide. Int J Greenhouse Gas Control
2008;2(1):38.
[6] Schladt MJ, Filburn TP, Helble JJ. Supported amine sorbents under temperature
swing absorption for CO
2
and moisture capture. Ind Eng Chem Res
2007;46(5):15907.
[7] Chang ACC, Chuang SSC, Gray M, Soong Y. In-situ infrared study of CO
2
adsorption on SBA-15 grafted with c-(aminopropyl) triethoxysilane. Energy
Fuels 2003;17(2):46873.
[8] Liu XW, Zhou L, Fu X, Sun Y, Su W, Zhou YP. Adsorption and regeneration study
of the mesoporous adsorbent SBA-15 adapted to the capture/separation of CO
2
and CH
4
. Chem Eng Sci 2007;62(4):110110.
[9] Stuckert NR, Yang RT. CO
2
capture from the atmosphere and simultaneous
concentration using zeolites and amine-grafted SBA-15. Environ Sci Technol
2011;45(23):1025764.
[10] Choi S, Drese JH, Eisenberger PM, Jones CW. Application of amine-tethered
solid sorbents for direct CO
2
capture from the ambient air. Environ Sci Technol
2011;45(6):24207.
[11] Xu X, Song C, Andresen JM, Miller BG, Scaroni AW. Novel polyethylenimine-
modied mesoporous molecular sieve of MCM-41 type as high-capacity
adsorbent for CO
2
capture. Energy Fuels 2002;16(6):14639.
[12] Franchi RS, Harlick PJE, Sayari A. Applications of pore-expanded mesoporous
silica. 2. Development of a high-capacity, water-tolerant adsorbent for CO
2
. Ind
Eng Chem Res 2005;44(21):800713.
[13] Yue MB, Sun LB, Cao Y, Wang Y, Wang ZJ, Zhu JH. Efcient CO
2
capturer derived
from as-synthesized MCM-41 modied with amine. Chem A Eur J
2008;14(11):344251.
[14] Serna-Guerrero R, Belmabkhout Y, Sayari A. Further investigations of CO
2
capture using triamine-grafted pore-expanded mesoporous silica. Chem Eng J
2010;158:5139.
[15] Heydari-Gorgi A, Belmabkhout Y, Sayari A. Polyethylenimine-impregnated
mesoporous silica: effect of amine loading and surface alkyl chains on CO
2
adsorption. Langmuir 2011;27(20):124116.
[16] Leal O, Bolvar C, Ovalles C, Garca JJ, Espidel Y. Reversible adsorption of carbon
dioxide on amine surface-bonded silica gel. Inorg Chim Acta 1995;240(1
2):1839.
[17] Tang Y, Landskron K. CO
2
-sorption properties of organosilicas with bridging
amine functionalities inside the framework. J Phys Chem C
2010;114(6):24948.
[18] Cui S, Cheng W, Shen X, Fan M, Russell A, Wu Z, et al. Mesoporous amine-
modied SiO
2
aerogel: a potential CO
2
sorbent. Energy Environ Sci
2011;4:20704.
[19] Wrmeyer K, Alnaief M, Smirnova I. Amino functionalised silica-aerogels for
CO
2
-adsorption at low partial pressure. Adsorption 2012;18:16371.
[20] Wrmeyer K, Smirnova I. Adsorption of CO
2
, moisture and ethanol at low
partial pressure using aminofunctionalised silica aerogels. Chem Eng J
2013;225:3507.
[21] Son W, Choi J, Ahn W. Adsorptive removal of carbon dioxide using
polyethyleneimine-loaded mesoporous silica materials. Microporous
Mesoporous Mater 2008;113(13):3140.
[22] Gray ML, Soong Y, Champagne KJ, Baltrus J, Stevens Jr RW, Toochinda P, et al.
CO
2
capture by amine-enriched y ash carbon sorbents. Sep Purif Technol
2004;35(1):316.
[23] Gebald C, Wurzbacher JA, Tingaut P, Zimmermann T, Steinfeld A. Amine-based
nanobrillated cellulose as adsorbent for CO
2
capture from air. Environ Sci
Technol 2011;45(20):91018.
[24] Ye Q, Jiang J, Wang C, Liu Y, Pan H, Shi Y. Adsorption of low-concentration
carbon dioxide on amine-modied carbon nanotubes at ambient temperature.
Energy Fuels 2012;26:2497504.
[25] Plaza MG, Pevida C, Arenillas A, Rubiera F, Pis JJ. CO
2
capture by adsorption
with nitrogen enriched carbons. Fuel 2007;86(14):220412.
[26] Aroua MK, Daud WMAW, Yin CY, Adinata D. Adsorption capacities of carbon
dioxide, oxygen, nitrogen and methane on carbon molecular basket derived
from polyethyleneimine impregnation on microporous palm shell activated
carbon. Sep Purif Technol 2008;62:60913.
[27] Wang J, Chen H, Zhou H, Liu X, Qiao W, Long D, et al. Carbon dioxide capture
using polyethylenimine-loaded mesoporous carbons. J Environ Sci
2013;25(1):12432.
[28] Jadhav P, Chatti R, Biniwale R, Labhsetwar N, Devotta S, Rayalu S. Monoethanol
amine modied zeolite 13X for CO
2
adsorption at different temperatures.
Energy Fuels 2007;21(6):35559.
[29] Chatti R, Bansiwal AK, Thote JA, Kumar V, Jadhav P, Lokhande SK, et al. Amine
loaded zeolites for carbon dioxide capture: Amine loading and adsorption
studies. Microporous Mesoporous Mater 2009;121(13):849.
[30] Xu X, Zhao X, Sun L, Liu X. Adsorption separation of carbon dioxide, methane
and nitrogen on monoethanol amine modied b-zeolite. J Nat Gas Chem
2009;18:16772.
[31] Su F, Lu C, Kuo SC, Zeng W. Adsorption of CO
2
on amine-functionalized Y-type
zeolites. Energy Fuels 2010;24(2):14418.
[32] Li B, Duan Y, Luebke D, Morreale B. Advances in CO
2
capture technology: a
patent review. Appl Energy 2013;102:143947.
[33] Hedin N, Andersson L, Bergstrm L, Yan J. Adsorbents for the post-combustion
capture of CO
2
using rapid temperature swing or vacuum swing adsorption.
Appl Energy 2013;104:41833.
[34] Zhang Y, Zhu Q, Lin X, Xu Z, Liu J, Wang Z, et al. A novel thermochemical cycle
for the dissociation of CO
2
and H
2
O using sustainable energy sources. Appl
Energy 2013;108:17.
[35] Chen WH, Tsai MH, Hung CI. Numerical prediction of CO
2
capture process by a
single droplet in alkaline spray. Appl Energy 2013;109:12534.
[36] Wang W, Xiao J, Wei X, Ding J, Wang X, Song C. Development of a new clay
supported polyethylenimine composite for CO
2
capture. Appl Energy
2014;113:33441.
[37] Zhao C, Chen X, Anthony EJ, Jiang X, Duan L, Wu Y, et al. Capturing CO
2
in ue
gas from fossil fuel-red power plants using dry regenerable alkali metal-
based sorbent. Prog Energy Combust Sci 2013;39(6):51534.
[38] Hayashi H, Taniuchi J, Furuyashiki N, Sugiyama S, Hirano S, Shigemoto N, et al.
Efcient recovery of carbon dioxide from ue gases of coal-red power plants
by cyclic xed-bed operations over K
2
CO
3
-on-Carbon. Ind Eng Chem Res
1998;37(1):18591.
[39] Shigemoto N, Yanagihara T, Sugiyama S, Hayashi H. Material balance and
energy consumption for CO
2
recovery from moist ue gas employing K
2
CO
3
-
on-Activated Carbon and its evaluation for practical adaptation. Energy Fuels
2006;20(2):7216.
[40] Lee SC, Choi BY, Ryu CK, Ahn YS, Lee TJ, Kim JC. The effect of water on the
activation and the CO
2
capture capacities of alkali metal-based sorbents.
Korean J Chem Eng 2006;23(3):3749.
[41] Lee SC, Choi BY, Lee TJ, Ryu CK, Ahn YS, Kim JC. CO
2
absorption and
regeneration of alkali metal-based solid sorbents. Catal Today 2006;111(3
4):38590.
[42] Lee SC, Kwon YM, Ryu CY, Chae HJ, Ragupathy D, Jung SY, et al. Development of
new alumina-modied sorbents for CO2 sorption and regeneration at
temperatures below 200 C. Fuel 2011;90(4):146570.
[43] Zhao CW, Chen XP, Zhao CS. CO
2
absorption using dry potassium-based
sorbents with different supports. Energy Fuels 2009;23(9):46837.
[44] Zhao CW, Chen XP, Zhao CS. Multiple-cycles behavior of K
2
CO
3
/Al
2
O
3
for CO
2
capture in a uidized-bed reactor. Energy Fuels 2010;24(2):100912.
[45] Zhao CW, Chen XP, Zhao CS. Carbonation and active-component-distribution
behaviors of several potassium-based sorbents. Ind Eng Chem Res
2011;50(8):446470.
[46] Zhao CW, Chen XP, Zhao CS, Wu Y, Dong W. K
2
CO
3
/Al
2
O
3
for capturing CO
2
in
ue gas from power plant Part 3: CO
2
capture behaviors of K
2
CO
3
/Al
2
O
3
in a
bubbling uidized bed reactor. Energy Fuels 2012;26(5):30628.
[47] Siriwardane RV, Shen MS, Fisher EP, Poston JA. Adsorption of CO
2
on molecular
sieves and activated carbon. Energy Fuels 2001;15:27984.
[48] Siriwardane RV, Shen MS, Fisher EP. Adsorption of CO
2
on zeolites at moderate
temperatures. Energy Fuels 2005;19:11539.
[49] Zhang J, Singh R, Webley PA. Alkali and alkaline-earth cation exchanged
chabazite zeolites for adsorption based CO
2
capture. Microporous Mesoporous
Mater 2008;111:47887.
[50] Lu C, Bai H, Wu B, Su F, Hwang JF. Comparative study of CO
2
capture by carbon
nanotubes, activated carbons, and zeolites. Energy Fuels 2008;22:30506.
[51] Sircar S, Golden TC, Rao MB. Activated carbon for gas separation and storage.
Carbon 1996;34:112.
[52] Himeno S, Komatsu T, Fujita S. High-pressure adsorption equilibria of methane
and carbon dioxide on several activated carbons. J Chem Eng Data
2005;50:36976.
[53] Pevida C, Plaza MG, Arias B, Fermoso J, Rubiera F, Pis JJ. Surface modication of
activated carbons for CO
2
capture. Appl Surf Sci 2008;254:716572.
[54] Ning P, Li F, Yi H, Tang X, Peng J, Li Y, et al. Adsorption equilibrium of methane
and carbon dioxide on microwave-activated carbon. Sep Purif Technol
2012;98:3216.
[55] Dantas TLP, Luna FMT, Silva Jr IJ, de Azevedo DCS, Grande CA, Rodrigues AE,
et al. Carbon dioxidenitrogen separation through adsorption on activated
carbon in a xed bed. Chem Eng J 2011;169(13):119.
[56] Zhao CW, Chen XP, Zhao CS. K
2
CO
3
/Al
2
O
3
for capturing CO
2
in ue gas from
power plant Part 2: Regeneration behaviors of K
2
CO
3
/Al
2
O
3
. Energy Fuels
2012;26(2):140611.
C. Zhao et al. / Applied Energy 124 (2014) 241247 247