Você está na página 1de 10

Chemical Engineering Journal 232 (2013) 139148

Contents lists available at ScienceDirect

Chemical Engineering Journal


journal homepage: www.elsevier.com/locate/cej

CO2 capture with a novel solid uidizable sorbent: Thermodynamics


and Temperature Programmed CarbonationDecarbonation
Muhammad B.I. Chowdhury, Mohammad R. Quddus, Hugo I. deLasa
a
Department of Chemical and Biochemical Engineering, Western University, London, ON N6G 5B9, Canada

h i g h l i g h t s g r a p h i c a l a b s t r a c t

 We prepared a new and stable CO2 Consecutive CO2 absorptiondesorption with the new uidizable Li4SiO4. Temperature ramp: 5 C/min;
sorbent based on lithium Input Gas composition: 10 vol% CO2 in helium; ow rate: 50 mL/min.
orthosilicate.
 We develop a CO2 capture 5.30
thermodynamic analysis for various
CO2 concentration (molx106/cm3)

1
5.10
uidizable sorbents.
Inversion Point 10
 We develop TPC and TPDC using a
4.90
temperature programmed xed bed
unit. 4.70
 We develop TPCTPDC runs using CO2 Absorption
calcium carbonate and lithium 4.50
orthosilicate. 300 400 500 600 700
4.30
 We are able to predict sorbent 10
CO2 Desorption
regeneration and temperature 1
4.10
inversion points. Temperature (oC)

a r t i c l e i n f o a b s t r a c t

Article history: This study investigates several sorbents for CO2 capture with emphasis on the development of a novel
Received 24 April 2013 lithium orthosilicate based sorbent. Thermodynamic analysis is considered to predict sorbent regenera-
Received in revised form 15 July 2013 tion conditions and thermal levels where sorbent kinetics change from CO2 absorption to CO2 desorption.
Accepted 17 July 2013
Temperature Programmed Carbonation (TPC) and Temperature Programmed Decarbonation (TPDC) are
Available online 27 July 2013
developed using a temperature programmed xed bed unit. Sorbents are kept in contact with a gas
stream containing a 10% CO2 mole fraction, and are subjected to a 5 C/min temperature ramp. Calcium
Keywords:
carbonate, lithium orthosilicate and a novel lithium orthosilicate modied sorbent are considered for
CO2 capture
Fluidizable sorbents
these runs. TPCTPDC runs conrm thermodynamic predictions for thermal inversion points. Further-
Thermodynamics more, TPDTPDC runs show that the novel lithium orthosilicate based sorbent provides a very stable
Temperature Programmed Carbonation and increased CO2 sorption capacity over 10 absorptionregeneration cycles, while calcium carbonate
Temperature Programmed Decarbonation displays a reduced CO2 sorption capacity with cyclic operation. This uidizable novel sorbent can signif-
icantly contribute towards CO2 removal from ue gases emitted by power plants.
2013 Elsevier B.V. All rights reserved.

1. Introduction tions of carbon dioxide have increased from 280 ppm in the pre-
industrial age to more than 380 ppm today. In fact, they increase
Global warming by increasing anthropogenic emissions of by more than 2 ppm per year [1]. Thus, novel technologies are re-
greenhouse gases is important in the world today. These emissions, quired to capture carbon dioxide and to store it.
which are mainly carbon dioxide, must be mitigated to minimize In this respect, CO2 removal from large stationary emission
the on-going irreversible environmental damages. The concentra- sources such as power plants should be, especially targeted.
Stationary sources release approximately half of the global CO2
Corresponding author. Tel.: +1 (519) 661 2144; fax: +1 (519) 661 3498. emissions. One of the most viable techniques for CO2 capture is
E-mail address: hdelasa@eng.uwo.ca (H.I. deLasa). chemical absorption using a solid sorbent. Calcite, dolomite,

1385-8947/$ - see front matter 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cej.2013.07.044
140 M.B.I. Chowdhury et al. / Chemical Engineering Journal 232 (2013) 139148

Nomenclature

ai activity coefcient for the i chemical species i Greek Symbols


C CO2 concentration in the gas exiting stream (mol/cm3) DG standard Gibbs Free Energy change (kJ/mol)
C0 CO2 concentration in the incoming stream (mol/cm3) DGrxn change of Gibbs Free Energy of reaction (kJ/mol)
Cp heat capacity (J/mol-K) DH standard enthalpy change (kJ/mol)
fi fugacity of chemical species i (bar) Ui fugacity co-efcient of chemical species i
K equilibrium reaction constant mi stoichiometric co-efcient of species i
Prxn total reaction pressure (bar)
pi partial pressure of species i (bar) Sorbent designation
pCO2 partial pressure of carbon dioxide (bar) Calcite CaCO3 (99%)
R universal gas constant (8.314 J/mol-K) Limestone (Cadomin 1) CaO (54%)
T temperature (K) Lithium orthosilicate Li4SiO4 (99.9%)
yi mole fraction of species i Lithium zirconate Li2ZrO3 (94%)
Sodium silicate Na2SiO3

lithium ceramics are gaining attention for large scale CO2 capture the gassolid CaOCO2 reaction. The rst step involves a rapid het-
via this type of process. In this process, CO2 reacts with solid sor- erogeneous chemical reaction, while the second step consists of a
bents producing metal carbonates and/or metal oxides or silicates. slow CO2 and CaO reaction resulting in the formation of a CaCO3
layer [2,11]. This CaCO3 layer considerably limits the access to
CaO CO2 () CaCO3 1
unreacted CaO for further carbonation. Therefore, the reactivity
of calcite decreases over the carbonationdecarbonation cycles.
Li2 ZrO3 CO2 () Li2 CO3 ZrO2 2
In contrast, lithium zirconate and lithium orthosilicate are
claimed to be stable even after 100 cycles [11]. Recent studies
Li4 SiO4 CO2 () Li2 CO3 Li2 SiO3 3
showed the potential of CO2 capture using lithium zirconate
[68] and lithium orthosilicate [9]. Kato et al. observed that Li4SiO4
Na2 SiO3 CO2 () Na2 CO3 SiO2 4
absorb CO2 at a faster rate than Li2ZrO3 at the same sorption con-
These sorbents can be thermally regenerated by heating the ditions [12]. From the calculated relationship of the chemical po-
metal carbonate beyond the carbonation temperature. Under these tential of the CO2 capture reaction as a function of temperatures
conditions, the regeneration or reverse of reactions (1)(4) are and CO2 partial pressures, Li4SiO4 appears to be a good candidate
allowed. for high temperature CO2 capture[13]. In addition, Li4SiO4 shows
Thus, solid sorbents can be used in principle for CO2 capture if lower costs of the raw materials involved. This is the case when
carbonation may be achieved at ue gas conditions. Furthermore, comparing the expensive ZrO2 with the less costly SiO2 [12].
metal carbonates decomposition temperatures are equally impor- Seggiani et al. doped Li4SiO4 with potassium and found that
tant. For instance, metal carbonates requiring more than 1000 C the sorption capacity increased as function of temperature be-
may not be suitable for this process, as they require extra supply tween 500 C and 580 C [14]. However, above 580 C, the sorp-
of energy and costly equipment [2]. tion capacity decreased. Wang et al. developed a Li4SiO4-based
It is envisioned that the chemical looping cycle using uidized sorbent using waste materials like rice husk ash (RHA) [15]. These
bed units could involve two interconnected uidized beds treating authors found that metals in RHA doped with Li4SiO4 inhibit par-
the exhaust of power stations. This conguration facilitates the ticle growth and increase pore volume and surface area. In addi-
reversible CO2 adsorption and desorption process. Therefore, at tion, RHA doped with Li4SiO4 maintains its high CO2 capacity
the outlet of the system an essentially pure stream of CO2 is pro- during multiple cycles compared to pure Li4SiO4 [15]. Olivares-
duced, ready for geological sequestration. Major advantages of this Marn et al. synthesized lithium based sorbent from y ash in
technology are that it uses high-temperatures and can be imple- the presence of Li2CO3 at 950 C. These authors found that CO2
mented in large-scale uidized bed reactors. Given the existing capture was a function of added K2CO3 [16].
knowledge with uidized bed reactor engineering, it is expected CO2 chemisorption using lithium based ceramics is controlled
the migration of this process from a lab-scale prototype to a pilot by different factors. Zhang et al. studied CO2 sorption/desorption
plant-scale or commercial scale unit can be accomplished rela- kinetics with Li4SiO4. These author found that in the sorption pro-
tively easily [3,4]. cess, the CO2 molecules come into contact with the sorbents and
As a result, an actual industrial installation would consist of a rapidly react to form a double-shell structure consisting of Li2CO3
cycling looping process involving a carbonation reactor and a and Li2SiO3 [17]. Initially, CO2 chemisorption is controlled by the
decarbonation reactor. However, the choice of specic sorbent reactivity of the lithium ceramics [17,18] and the ow of CO2
would depend on the range of carbonation and decarbonation tem- [19]. However, once the external lithium shell is produced at the
peratures, the reactivity of the particular sorbent and its stability lithium ceramic particle outer surface, the whole process may be-
under cyclic operation [5]. come diffusionally controlled[1719].
One can notice from Eqs. (2)(4) that some sorbents produce It is thus concluded that an ideal sorbent being used in a cyclical
metal oxides and/or silicates, in addition to carbonates. Therefore, system capturing CO2 from hot ue gases, requires properties of
both various chemical and physical properties of the formed spe- both good CO2 absorptiondesorption capacity and high tempera-
cies have to be considered in the case of cyclical sorbent applica- ture stability. To address this issue, this research focuses on the
tions. For example, the melting point of Li2CO3 is 720 C. development of a novel and stable lithium orthosilicate based
Therefore, the operation of Li2CO3 has to be limited to 720 C. sorbent, with enhanced structural properties to minimize CO2
CO2 capture using calcium oxide has been widely studied absorption/desorption diffusional limitations. The aim of this study
[2,10,11]. There are two possible rate-controlling steps inuencing is also to determine the best conditions for CO2 capture; with these
M.B.I. Chowdhury et al. / Chemical Engineering Journal 232 (2013) 139148 141

conditions being identied on the basis of both carbonationdecar- where aCaO ; aCO3 CaO are activity coefcients of the calcination reac-
bonation thermodynamics and kinetics. tion and pCO2 ;e is the partial pressure of CO2.
Since, the activity coefcients for solids and for the fugacity of
the gases at the expected operating conditions are close to one,
2. Thermodynamics at reaction equilibrium the equilibrium constant for the calcination unit becomes:

Thermodynamics allow establishing the dependence of equilib- K p;decarbonation1 pCO2 ;e 15


rium constants on temperature and pressure [20]. One can thus, Similarly, during carbonation, the equilibrium constant can be
assess how operating conditions may inuence CO2 capture at expressed as:
chemical equilibrium.
1
K p;carbonation1 16
pCO2 ;e
2.1. Effect of temperature on chemical equilibrium constants
A comparable analysis can be developed for CO2 removal at
At reaction equilibrium, the CO2 molar function is a variable equilibrium conditions using Li4SiO4. In this case, Kp,decarbonation2
function of both temperature and pressure. At a constant pressure, can be expressed as:
the dependence of DG on T is given by the GibbsHelmholtz equa-
aLi4 SiO4 pCO2 ;e
tion as follows [20], K p;decarbonation2 17
aLi2 SiO3 :aLi2 CO3
  
dDG =RT  DH 
7
dT p RT 2 Therefore; K p;decarbonation2 pCO2 ;e 18

where R is a universal gas constant and T is the temperature. Both Furthermore, in the case of the Li4SiO4 carbonation, the equilib-
DG and DH are affected by pressure. At 1 atm or 1 bar standard rium constant for carbonation can be calculated as:
condition Eq. (7) can be represented as: 1
K p;carbonation2 19

dDG =RT DH  pCO2 ;e
8
dT RT 2 Similar expressions can be established for calculating decarbon-
ation equilibrium constants for both lithium zirconate and sodium
2.2. Effect of pressure on chemical equilibrium constants silicates such as: K p;decarbonation3 pCO2 ;e and K p;carbonation3 p 1 .
CO2 ;e

According to the denition of the standard Gibbs Energy change


The effect of pressure on chemical equilibrium can be formu- of reaction, it can be postulated that:
lated by applying Le Chateliers principle. As a result, the equilib-
DG
rium constant, K, of a heterogeneous reaction can be expressed  ln K 20
RT
as follows [20]
Therefore, for CO2 release i.e. decarbonation, at the equilibrium
Y v aqp  arq    Eq. (20) can be described as:
KT ai i 9
aax aby   
i1 DG RT ln pCO2 0 21
where K is the equilibrium constant, ai is the activity coefcient and
As a result, Eq. (21) is satised at equilibrium for each one of the
mi is the stoichiometric coefcient of the species i. While activity
CO2 sorbents considered in Eqs. (1)(4).
coefcients for solids are set to 1. For gases, activity coefcients
One should notice that at initial conditions, in the decarbon-
are dened as the ratio of the species fugacity in the mixture to
ation unit, if the partial pressure pCO2 is too low, Eq. (21) becomes
the fugacity at its standard state,
f negative:
i
ai o 10 DG RT ln pCO2 < 0 22
fi
with f oi P o = 1 bar or 1 atm, dening the gas species at the stan- This negative value in Eq. (22) occurs when all absorbed CO2
dard state: decomposed in the decarbonation unit but could not reach equilib-
Furthermore f i can be expressed as: rium condition.
On the other hand, if pCO2 is high enough in the decarbonation
f y P / 11
i i rxn i unit, Eq. (21) may become positive:
where yi is the mole fraction of species i, Prxn is the total reaction DG RT ln pCO2 > 0 23
pressure, and /i is the fugacity coefcient. Moreover, the fugacity
coefcient [21] for 15 bars total pressure and temperatures be- This shows that as long as Eq. (23) is positive, no decomposition
tween 5001000 C can be approximated to /i 1. Thus, of sorbent will occur. Rather carbonation will proceed until CO2
f y Prxn p 12 partial pressure decreases enough so that Eq. (21) is fully satised.
i i i

As a result, the activity coefcient in Eq. (10) for the gas species
becomes: 3. Thermodynamic model predictions for different sorbents

ai pi 13
3.1. Gibbs Energy changes
For instance, in the case of calcite in the decarbonation unit, the
equilibrium constant, Kp,decarbonation1, for the CO2 removal can be ex- The Gibbs Free Energy changes can be calculated on the basis of
pressed as: thermodynamic data [22]. Fig. 1 reports the Gibbs Free Energy
changes for carbonation and decarbonation for the four sorbents
aCaO pCO2 ;e
K p;decarbonation1 14 considered in the present study as a function of the temperature
aCO3 CaO and at 1 bar CO2 partial pressure.
142 M.B.I. Chowdhury et al. / Chemical Engineering Journal 232 (2013) 139148

Negative values of Gibbs Energies in Fig. 1 shows conditions 100

Equilibrium partial pressure of


Li 4 SiO4 CO2 Absorption
where reactions proceed towards the right side of Eqs. (1)(4) i.e.
10 Li 2 ZrO3
CO2 is absorbed. These results suggest that CO2 can be absorbed CaO
even at room temperatures. However, due to kinetic limitations, 1 Na2 SiO3
CO2 absorption at room temperatures is not viable. Thus, high tem-

CO2 (bar)
0.1
peratures are required for adequate CO2 capture kinetics, with the
exothermicity of carbonation reactions also beneting the CO2 cap- 0.01
ture [23]. This is true for lithium orthosilicate, lithium zirconate [6] 0.001 CO2 Desorption
and calcium oxide [24].
0.0001
One can notice in Fig. 1, that there are specic temperatures to
achieve zero Gibbs Energies for carbonation. These temperatures 0.00001
are 325 C, 720 C, 750 C and 900 C for sodium silicate, lithium
0.000001
orthosilicate, lithium zirconate and calcite respectively. These tem- 50 150 250 350 450 550 650 750 850 950
peratures yielding zero Gibbs Energies are designated as Inversion Temperature ( oC)
Point Temperatures. One should also note that lower inversion
temperatures not only allow CO2 capture processes with lower en- Fig. 2. Equilibrium partial pressure dependence with temperature. Note: reported
ergy demands but also help to avoid sorbent sintering. Sintering is curves separate the CO2 absorption conditions (carbonation) and CO2 desorption
(decarbonation) conditions. Broken horizontal line provides the inversion temper-
one of the main reasons for sorbent deactivation. Preventing sin-
ature for 0.1 bar equilibrium pressure.
tering is a main factor for sorbent choice.

4.1. Preparation of as received solid sorbent


3.2. Equilibrium partial pressure change
Solid sorbents were thermally treated to increase thermal sta-
Fig. 2 reports equilibrium partial pressures for CO2 absorption bility and remove absorbed CO2, water vapor and other adsorbed
and CO2 desorption for several sorbents and their changes with gaseous impurities. To achieve this, CaCO3 was calcined at 930 C
temperature. It shows that Li4SiO4 can desorb CO2 at 560 C if while Li2ZrO3 and Li4SiO4 were calcined at 675 C under atmo-
CO2 partial pressure is below 0.1 bar. However, if the partial pres- spheric condition. Conditions for pre-treatment for sorbent calci-
sure of CO2 is higher than 0.1 bar, absorption takes place at this nations were set to exceed considerably the equilibrium
temperature. temperatures. As reported in Fig. 2 for a 0.00039 bar CO2 partial
Fig. 2 also reports that CO2 capture from hot ue gases (gener- pressure in air these equilibrium temperatures were: 527 C for
ally 400800 C), using lithium orthosilicate, lithium zirconate and CaO, 427 C for Li2ZrO3 and 374 C for Li4SiO4.
calcite is thermodynamically viable. These materials belong to the
so-called high temperature CO2 capture sorbents. From Fig. 2, one
4.2. Preparation of the novel uidizable solid sorbent
can also see that in principle, sorbents such as Li4SiO4, Li2ZrO3, and
CaCO3 are capable of absorbing CO2 from hot ue gases.
The preparation of the new uidizable solid sorbent involves
the following steps: (a) Li4SiO4 is mixed with glacial HAc excess
(molar ratio is 1:10) followed by continuous stirring. (b) After mix-
4. Experimental ing, a packed bed of precursor sorbent particles is calcined under
nitrogen ow from room temperature to 620 C in 2 h. (c) Follow-
All the chemicals used in this work were of reagent grade. The ing calcination, the sorbent particles are exposed to a helium ow
99.9 wt% lithium orthosilicate (Li4SiO4), the 99.0 wt% calcium car- at 675 C for 1 h. This allows the removal of any carbonaceous spe-
bonate (CaCO3), and the 94 wt% lithium zirconate (Li2ZrO3) were cies or gaseous products remaining adsorbed on the sorbents.
obtained from Alfa Aesar, Massachusetts, USA. The 99.7 v/v% gla-
cial acetic acid (HAc) was from Aldrich Chemical Co., Wisconsin,
4.3. Sorbent characterization
USA; and the 10 v/v% carbon dioxide with helium being the bal-
ance was from Praxair, Canada. All chemicals were used as received
Sorbent characterization was developed using the Brunauer
without further purication.
EmmettTeller (BET) surface area, pore size and distribution, and
pore volume from nitrogen absorption and desorption isotherms
40 at 77 K. A constant-volume absorption apparatus (Micromeritics
Li4 SiO4 ASAP 2010) using N2 gas (99.995% pure; obtained from Praxair,
Li2 ZrO3 Canada) was utilized. The prepared samples were degassed at
20 CaO 150 C for 6 h before measurements were taken. Micromeritics
Na2 SiO3 DFT Plus software for Windows was used to calculate pore vol-
G (kJ/mol)

CO2 Desorption ume and pore size distribution by applying Density Functional
0 Theory (DFT) based on slit-shape pore geometry.
CO2 Absorption X-ray diffraction (XRD) patterns of catalysts were obtained to
establish crystallinity and crystallinity changes, given that its
-20 importance on sorbent performance. XRD was performed utilizing
a Rigaku Difractometer (Ultima IV) unit with a CuKa1 + Ka2 equal
to 1.54184 radiation. The XRD instrument was operated at 45 kV
-40
50 150 250 350 450 550 650 750 850 950 and 160 mA, using the normal scan rate of 10 per minute (equiv-
alent to 0.5 two-theta on conventional diffractometers) in the 2h
Temperature ( oC)
range spanning from 10 to 90. X-rays were collimated using 1
Fig. 1. Temperature dependence of Gibbs Energy of reaction for carbonation and divergent and scatter slits, and a 0.2 mm receiving slit. XRD peaks
decarbonation reactions (CO2 partial pressure 1 bar). were identied in the present study, using a Rigaku Difractometer
M.B.I. Chowdhury et al. / Chemical Engineering Journal 232 (2013) 139148 143

(Ultima IV) and the International Center for Diffraction Data li- Concerning lithium zirconate and untreated lithium orthosili-
brary. Crystallinity was dened for each one of the analyzed sam- cate, calcination led to signicant reductions of surface area. More
ples using the following equation: specically, when using untreated lithium orthosilicate, there was

Scattered intensity from crystalline phase


Crystallinity
Scattered intensity from crystalline phase Scattered intensity from amorphous phase

The CO2 absorptiondesorption sorbent capacity and the both a reduction of pore volume and surface area, with this being
inversion temperature were established using Temperature Pro- attributed to the collapse of the pore sorbent network. Finally, cal-
grammed CarbonationDecarbonation (TPC-TPDC). The TPC-TPDC cinations of glacial acetic acid treated Li4SiO4, retained both a sig-
experiments were carried out using a Micromeritics Autochem nicant fraction of surface area and pore volume. This result is
2920. Prior to TPC measurements, 100150 mg of the fresh sorbent consistent with the observed lower apparent density of the acid
was calcined at 675 C under a helium ow to remove moisture, treated Li4SiO4 (0.98 g/cm3). This is also the case with respect to
entrap CO2 and other loosely bonded impurities. The TPC runs the higher apparent density of untreated Li4SiO4 (1.24 g/cm3).
were performed by circulating a stream of gas containing 10% Fig. 3a and b compare the nitrogen adsorptiondesorption iso-
CO2 with He making the balance at a rate of 50 mL/min. The tem- therms and the corresponding pore size distributions of sorbents
perature was raised from ambient to the required temperature at a after calcination. Isotherm shapes and hysteresis loops are those
rate of 5 C/min. of type IV isotherms with H3 and H4 loops [25]. CaO and un-
A Thermal Conductivity Detector (TCD) was used to record the treated Li4SiO4 show small H3 loops, conrming that these materi-
change in carbon dioxide concentration in the gas stream exiting als are mostly nonporous with some slit-like pores [25]. On the
the bed of sorbent. CO2 amounts were recorded at the sorbent other hand, large H4 loops are observed for the acid treated Li4SiO4,
bed exit and used to calculate the carbon dioxide consumed or re- suggesting the presence of large irregular shaped mesopores and a
leased during the carbonation and decarbonation (regeneration). broad pore size distribution with slit like pores [25].
Fig. 4a and b compare isotherms and pore size distributions for
5. Results and discussions both carbonated and decarbonated sorbents. Reported pore size
distributions were calculated using the Density Functional Theory
5.1. Surface area and pore size distribution model (DFT model) based on slit-shape pore geometry. The shape
of the isotherms and their hysteresis loops remained close. How-
Surface area, average pore diameter and pore volume of the sor- ever, as shown in Fig. 4b, there was a reduction in pore volume
bents of the present study are summarized in Table 1. in carbonated sorbents attributed to CO2 absorption. Fig. 4b also
It can be observed in Table 1 that the specic surface area of reports a comparatively larger pore size distribution and pore vol-
CaCO3 increased with calcination. This gain in specic surface area ume, with this being consistent with the ndings for acid treated
can be attributed to the CO2 release from CaCO3. Regarding lime- Li4SiO4. Thus, pore size distributions show that the acid treated Li4-
stone (Cadomin 1), the specic surface area also augmented signif- SiO4 provides a stable sorbent, with a pore network providing good
icantly with calcination. However, pore volume remained low for CO2 accessibility even under carbonated conditions.
both CaCO3 and limestone, even after calcination.

Table 1
Physiochemical properties of the sorbents.

Catalysts BET surface area (m2/gm) Average pore diameter () Pore volume (cm3/gm)
Fresh Calcined After 10 cycle Fresh Calcined After 10 cycle Fresh Calcined After 10 cycle
CaCO3 3.08 5.05 22.0 49.7 0.002 0.006
Limestone 3.49 9.83 3.93 24.6 93.9 115.4 0.002 0.023 0.011
Li2ZrO3 6.37 2.24 43.2 16.4 0.007 0.001
Li4SiO4 36.74 4.73 2.82 167.3 39.5 107.3 0.154 0.005 0.004
Li4SiO4/HAc 9.35 7.21 85.9 104.9 0.020 0.015

30.0 0.0012
CaO a CaO b
Incremental Volume

25.0 0.001
Vol (cm3 /g STP)

Li4 SiO4 Li4 SiO4


20.0 Li 4 SiO4 /HAc
Li4 SiO4 /HAc 0.0008
(cm 3 /g)

15.0 0.0006
10.0 0.0004
5.0 0.0002
0.0 0
0 0.2 0.4 0.6 0.8 1 10 100 1000
Relative pressure (P/P0 ) Pore width (A)

Fig. 3. (a) N2 adsorption/desorption isotherms and (b) pore-size distributions of calcined sorbents. (D) Fresh calcium oxide calcined at 930 C, () lithium orthosilicates
calcined at 650 C, (j) HAc treated Lithium orthosilicates calcined at 650 C.
144 M.B.I. Chowdhury et al. / Chemical Engineering Journal 232 (2013) 139148

30 0.0012

Incremental Volume (cm3/g)


Li4 SiO4 -Carbonated
Li4 SiO4 -Decarbonated
a Li 4 SiO4 b
25 0.001 Li4 SiO4 -Carbonated
Li4 SiO4 /HAc-Carbonated
Vol (cm3/g STP) Li 4 SiO4 /HAc-Decarbonated Li 4 SiO4 /HAc
20 0.0008 Li4 SiO4 /HAc-Carbonated

15 0.0006

10 0.0004

5 0.0002

0 0
0 0.2 0.4 0.6 0.8 1 10 100 1000
Relative pressure ( P/Po) Pore width (A)

Fig. 4. (a) N2 adsorption/desorption isotherms and (b) pore-size distributions for carbonated and decarbonated sorbents.

1000 1800 1
4 1 4 4 3 d
1-Li 4 SiO4
900 1600 1
2-SiO 2 5 1-Li4SiO4
800 3-Li 2 CO 3 1400 1 2-SiO2
4-Li 2 SiO 3 3 11 1 3 3 3 3
1
c
700 3-Li2CO3
c 4
3 4 3 3 3 3 c 1200 5 4-Li2SiO3
4 5
600 5-CH3COOLi
1000

a. u.
a.u.

1 6-H2Si2O5
500 6
1
11 1 800
1 1 1 5 2 5 5
400 2 1 1 1 b 6
b 600 3 13
6
2 1 1
1 6 6 b
300 2 2
2 3 400 2 1
2
200 1 2 2 1 3
3 200 1 1
11 4 3 1 4 3 1 1 3
3
a
100 4
3
3 14 a 1 4 3 3 3 1 14
a
0
0
10 20 30 40 50 60 10 20 30 40 50 60

2 2

Fig. 5. XRD diffractogram: (a) as received Li4SiO4, (b) calcined Li4SiO4 675 C, and Fig. 6. XRD diffractogram for: (a) As received Li4SiO4, (b) acid treated Li4SiO4 (dried
(c) Li4SiO4 after 10 absorptiondesorption cycles. at room temperature). (c) Acid treated calcined Li4SiO4 and (d) after 10 cycles of CO2
absorptiondesorption with acid treated Li4SiO4.
5.2. XRD analysis

XRD allows the identication of various crystalline phases on ance all peaks of type 5 in the c diffractogram of Fig. 6. One
Li4SiO4 and on Li4SiO4 pre-treated with acetic acid as shown in can also observe that lithium silicates are less intense using an acid
Fig. 5. It was observed that the Li4SiO4 sample as received con- treated sorbent. However, peaks of LiCO3 appear to also be promi-
tained crystalline Li2CO3, Li2SiO3 and SiO2 phases. Upon calcination nent in the c diffractogram while Li2SiO3 is insignicant. Thus,
at 675 C however, these phases disappear, forming additional lith- from these observations, the following reactions appear to occur
ium orthosilicate. As a result, the following reactions can be con- during calcination of acid treated Li4SiO4 sorbents:
sidered to occur: 2CH3 COOLi ! Li2 CO3 CH3 COCH3 27
2Li2 CO3 SiO2 ! Li4 SiO4 2CO2 24
H2 Si2 O5 3Li2 CO3 ! Li4 SiO4 Li2 SiO3 H2 O CO2 28
Li2 CO3 Li2 SiO3 () Li4 SiO4 CO2 25
Li2 SiO3 Li2 CO3 ! Li4 SiO4 2CO2 29
However, after repeated CO2 absorptiondesorption cycles, the
characteristic Li2CO3 and Li2SiO3 peaks re-appeared (reverse reac- One can thus see that when comparing untreated Li4SiO4 with
tion of Eq. (25)) conrming that the regeneration process of the Li4- acetic acid treated Li4SiO4, the Li4SiO4 microstructure is signi-
SiO4 sorbent remained incomplete. This was likely due to the cantly altered by the acetic acid addition. There is also, signicant
diffusion resistance consequence of pore blockage and the forma- re-crystallization of Li4SiO4, as shown in Fig. 6b after acetic acid
tion of core-shell structures of Li2CO3 and Li2SiO3 on the sorbent. addition. One can also notice that after acetic acid treatment fol-
Acetic acid treatment of Li4SiO4, however, initially created crys- lowed by calcination, the dominant Li4SiO4 peaks appear again in
talline lithium acetate salts and silicic acids, as shown in Fig. 6. A the XRD difractograms with these peaks being observed at lower
possible reaction leading to the formation of acetate salts can be intensities. Moreover, after repeated cycles of operation, the dom-
considered as follows: inant peaks remain with similar approximate intensities conrm-
ing the stability of the acid treated Li4SiO4 sorbent.
2Li4 SiO4 8CH3 COOH ! 8CH3 COOLi H2 Si2 O5 3H2 O 26
Fig. 5b and c report XRD analysis for calcined Li4SiO4 and Li4SiO4
Silicic acid and in situ water undergo polycondensation follow- after 10 absorptiondesorption cycles. One can see comparing
ing the sol gel process forming polymeric three dimensional net- these two gures that there is an increase in the Li2CO3 content.
works [26]. This network may remain even after calcinations, On this basis one can claim that changes in sorption activity can
helping to increase the sorbent porosity. be expected due to incomplete sorbent regeneration. Fig. 6c and
After calcination of the acid treated sorbent however, the d shows that on the other hand, in the case of acid treated Li4SiO4,
formed acetate salts are converted mainly into porous lithium the Li2CO3 content (Eq. (27)) remains steady for both calcined Li4-
orthosilicates. This can be supported by observing the disappear- SiO4 (Fig. 6c) and Li4SiO4 after 10 repeated cycles (Fig. 6d).
M.B.I. Chowdhury et al. / Chemical Engineering Journal 232 (2013) 139148 145

XRD was also valuable to establish the changes in crystallinity ture could not be raised above 720 C. This was due to the melting
during the pre-treatment process. For instance, Fig. 5b shows that point of Li2CO3, which is one of the products formed by absorbing
Li4SiO4 was 82% crystalline following calcination at 675 C. On the CO2 on Li2ZrO3. Thus, Li2ZrO3 could not be regenerated. Therefore,
other hand, once the lithium orthosilicate sorbent is treated with Li2ZrO3 is considered unsuitable for CO2 capture from hot ue
glacial acid acid, there is an increased crystallinity up to 96% due gases.
to the formation of acetate salt as shown in Fig. 6b. Formed acetate On the other hand, when using Li4SiO4, TPCTPDC curves dis-
salts are converted however into porous lithium orthosilicates play promising results in terms of CO2 absorption and desorption
upon calcination with reduced crystallinity to 64% (Fig. 6c). at lower temperatures than the other sorbents considered in the
As a result, the acetic acid addition leads to Li4SiO4 sorbents present study. CO2 is absorbed below the 595 C inversion temper-
with enhanced pore structure with a higher pore volume, a wider ature condition with a maximum TPC peak at 567 C. This thermal
pore size distribution and reduced particle density. level (inversion point temperature) established with a heating
ramp of 5 C/min is in agreement with the thermodynamics which
5.3. Performance evaluation with Temperature Programmed predicts 575 C at 0.1 bar CO2 partial pressure. Moreover, close
CarbonationTemperature Programmed Decarbonation (TPCTPDC) inversion point temperatures were also obtained in separate
runs performed using 2.5 C/min, 7.5 C/min and 10 C/min tem-
TPCTPDC is a valuable experimental tool for establishing the perature ramps. Deviations remained in the 2 C range.
following: (a) the CO2 absorption and desorption capacity of a sor- Regarding the Li4SiO4 carbonation process, one can observe a
bent and its evolution with carbonationdecarbonation sequential double peak. This double peak is assigned to the formation of outer
cycles, (b) the inversion temperature or the temperature where the shell structures consisting of Li2CO3 and Li2SiO3 [17]. In fact, while
sorbent interaction with CO2 is neither in the absorption mode nor CO2 chemisorption controls the reactivity of Li4SiO4 during CO2
in desorption mode, (c) the kinetics of CO2 absorption/desorption. absorption at the initial carbonation stages (rst peak), once an
As stated above, the temperature ramp selected for the present external lithium shell is formed the absorption process likely be-
study was 5 C/min. This temperature ramp allows increasing the comes diffusionally controlled (second peak). Under these condi-
temperature to 100 C in about 20 min. This is the typical solid res- tions, higher TPC temperatures are required for full Li4SiO4
idence time that one can expect in large scale dense uidized beds carbonation.
using this type of sorbent material. Thus, these CO2 absorption CO2 absorptiondesorption processes using Li4SiO4 however,
desorption capacity results were obtained under dynamic have to be limited to thermal levels below 720 C. This is due to
conditions with a sorbent time-on-stream of 1520 min. These the melting point of Li2CO3. To compensate for this and to com-
times-on-stream are close to the times-on-stream expected in a plete the TPCTPDC, the sorbent temperature was held at 700 C
twin reactor conguration involving CO2 absorption and desorp- for an hour.
tion in large scale uidized bed units. Fig. 7 also shows that Li4SiO4 and CaO display good agreement
Fig. 7 reports three interesting absorptiondesorption (carbon- between thermodynamics and experimental ndings for the
ationdecarbonation) cases. One can see that when using a com- inversion point temperature. However, Li2ZrO3 with 94 wt% pur-
mercial calcite, with a CO2 partial pressure of 0.1 bar, the CO2 is ity display a signicant difference between the equilibrium inver-
continuously absorbed until the temperature reaches 722 C. This sion point temperature and the practical observed value. Even at
thermal level (thermal inversion point) is in agreement with ther- the Li2CO3 melting temperature (720 C), the inversion point tem-
modynamics predicting 750 C for this same condition. perature for Li2ZrO3 is still not reached.
In the case of calcite, a maximum Temperature Programmed Furthermore, for cyclic use in CO2 capture, sorbents need to be
Carbonation (TPC) peak is obtained at 665 C while the desorption regenerated. In this respect, Li4SiO4 requires a lower regeneration
of CO2 starts only after 722 C. On the other hand, the desorption temperature when compared to CaO. The difference of equilibrium
peak displays a maximum Temperature Programmed Decarbon- inversion point temperatures between CaO and Li4SiO4 is
ation (TPDC) at 768 C reaching a plateau above 810 C. Thus, this approximately 180 C. Regarding Li4SiO4 and Li2ZrO3, while Li4SiO4
shows that the CO2 desorption processes require very high temper- display a 30 C lower inversion temperature than Li2ZrO3 this dif-
atures to be completed. ference is in practice much larger (>110 C).
For Li2ZrO3, only a partial absorption peak ending at 700 C and Fig. 8 shows the CO2 absorption using commercial calcite
no desorption peak were observed. As described earlier, tempera- (99.98%) and limestone (Cadomin 1). Although both are CaO based

5.50 8.50
CO2 concentration (molx10 6 /cm3 )

CO2 concentration (molx10 /cm 3 )

Calcite
Li4 SiO4 Calcite
7.50
Cadomin1 (Limestone)
6

Li 2 ZrO3

5.00 CO2 Desorption 6.50


CO2 Desorption

5.50

4.50 4.50
0 200 400 600 800 1000 0 200 400 600 800 1000
3.50 CO2 Absorption
CO2 Absorption
4.00 2.50
o
Temperature (oC) Temperature ( C)

Fig. 7. The CO2 absorptiondesorption cycle with respect to temperature (Tem- Fig. 8. CO2 absorptiondesorption with temperature by calcite and Cadomin 1
perature ramp: 5 C/min; Input Gas composition: 10 vol% CO2 in helium; ow rate: (limestone). (Temperature ramp: 5 C/min; Input Gas composition: 10 vol% CO2 in
50 mL/min). helium; ow rate: 50 mL/min).
146 M.B.I. Chowdhury et al. / Chemical Engineering Journal 232 (2013) 139148

sorbents, Cadomin1 contains 53 wt% CaO only. It can be noticed, in Furthermore, one can also notice in both Figs. 9 and 10, that the
this respect, that CaO doping with other materials such as MgO sorbent absorption capacity of both limestone and lithium ortho-
may help to increase CO2 capacity. For instance, it can be observed silicate decreases over repeated cycles. As shown with XRD, the
that Cadomin1 has a comparatively higher surface area and higher appearance of new phases after CO2 absorptiondesorption cycles
porosity than calcite. These two morphological parameters may af- suggest partial decomposition of Li4SiO4. This decay of absorption
fect CO2 absorption and desorption. capacity can be assigned to sorbent sintering and losses of sorbent
It can also be noticed that the inversion temperature, was unaf- surface area, with a signicant reduction of the mesopore-small
fected by the CaO specic composition. This is the case since the pore volume necessary for carbonate product storage.
inversion temperature is a thermodynamic property which re- As described earlier, to improve absorption stability, a novel u-
mains unchanged for calcite and Cadomin 1, at 700 C and 710 C idizable Li4SiO4 based sorbent was developed. Fig. 11 reports the
respectively. CO2 absorptiondesorption capacity of the novel Li4SiO4 for over
repeated cycles. One can see that the absorption capacity of Li4SiO4
did not change signicantly over repeated cycles, conrming the
6. Sorbent stability studies stability of the acetic acid treated Li4SiO4. It is postulated that ace-
tic acid modies the microstructure of Li4SiO4, resulting in a more
Figs. 9 and 10 display the CO2 sorbent capacity for CO2 capture stable sorbent. In this respect, Li et al. assumed that an acetic acid
over repeated cycles at approximately 0.1 bar partial pressure of solution leads to a better sorbent sintering resistance given that
CO2. It has already been discussed in the present study that at a modied limestone protects CaO grain growth [27].
constant partial pressure of CO2, the absorbed CO2 is desorbed if Fig. 12 provides a comparison of sorbent efciencies using the
the temperature is increased beyond an equilibrium thermal level. following efciency denition:
This so-called temperature inversion point, as reported in Figs. 9
and 10 remains almost constant over repeated absorptiondesorp- Experimental CO2 uptake

tion cycles. This predictability of temperature inversion points Maxium CO2 uptake as Li4 SiO4 expected by stoichiometry
condently supports the soundness of the thermodynamic analysis  100% 30
described earlier.

8.50
CO2 concentration (mol x10 6/cm3 )

7.50 5.30
CO2 concentration (molx10 /cm )
3

Inversion Point 1 1
6.50 5.10
6

Inversion Point 10

5.50 4.90
CO2 Absorption 10

4.50 4.70
10
300 400 500 600 700 800 900 CO2 Absorption
3.50 1 4.50
CO2 Desorption 300 400 500 600 700
2.50 4.30 10
CO2 Desorption
Temperature ( o C)
1
4.10
Fig. 9. The CO2 absorptiondesorption (10 cycles) prole of limestone with respect Temperature (oC)
to temperature. (Temperature ramp: 5 C/min; Input Gas composition: 10 vol% CO2
in helium; ow rate: 50 mL/min). Note: direction of the arrows describes successive
Fig. 11. The CO2 absorptiondesorption (10 cycles) prole of modied Li4SiO4 with
runs: from run 110.
respect to temperature (Temperature ramp: 5 C/min; Input Gas composition:
10 vol% CO2 in helium; ow rate: 50 mL/min). Note: direction of the arrows
describes successive runs: from run 110.
4.80
CO2 concentration (mol x10 6/cm )
3

80
Cadomin1 (adsorption)
Inversion Point 70 Cadomin1 (desorption)
4.65 Li4 SiO4 (without treatment)
60 Li4 SiO4 /HAc (adsorption)
% Efficiency

1 Li 4 SiO4 /HAc (desorption)


50
CO2 Absorption Li4 SiO4 /HAc (adsorption-2nd run)
10 40 Li 4 SiO4 /HAc (desorption-2nd run)
4.50 10
30
300 400 500 600 700
1
20
CO2 Desorption
10
4.35
Temperature (oC) 0
1 2 3 4 5 6 7 8 9 10
Fig. 10. The CO2 absorptiondesorption (10 cycles) prole of Li4SiO4 with respect to Number of cycles
temperature. (Temperature ramp: 5 C/min; Input Gas composition: 10 vol% CO2 in
helium; ow rate: 50 mL/min). Note: direction of the arrows describes successive Fig. 12. The CO2 absorptiondesorption efciencies over repeated cycles. (Closed
runs: from run 110. symbols are for absorption and open symbols are for desorption).
M.B.I. Chowdhury et al. / Chemical Engineering Journal 232 (2013) 139148 147

It can be noticed that although Cadomin1 shows a higher sor- 7. Conclusions


bent efciency than Li4SiO4 during the initial cycles of absorp-
tiondesorption, the Cadomin1 sorbent efciency as dened in (1) It is shown that gassolid sorption thermodynamics helps to
Eq. (30), drops drastically over repeated cycles. predict sorbent regeneration conditions, and the tempera-
Table 2 shows the quantitative information of CO2 uptake and tures where sorbent kinetics change from CO2 absorption
release during cyclic process. Li4SiO4 without acid treatment to CO2 desorption.
showed a moderate CO2 capacity reduction over absorption (2) It is observed that Temperature Programmed Carbonation
desorption repeated cycles: from 48.3 mL/gm in cycle one to (TPC) and Temperature Programmed Decarbonation (TPDC)
23.1 mL/gm in cycle ten. This CO2 reduced capacity was assigned help to establish absorptiondesorption performance for
to diffusional resistance effects, which are the result of the forma- several CO2 sorbents.
tion of a core shell structure. On the other hand, acid treated Li4- (3) It is noticed that limestone absorption capacity is drastically
SiO4 increases the fraction of mesopores preventing the reduced over repeated cycles while modied Li4SiO4 is stable
formation of the core shell structure. This yields as a result a more and reproducible.
stable sorbent, exceeding even Cadomin1 efciency in the 9th (4) It is found that an acid treated uidizable Li4SiO4, provides a
cycle. promising sorbent for CO2 capture with increased specic
Fig. 13 also compares the theoretical and experimental tem- surface area, higher porosity and stable sorbent capacity.
perature inversion conditions and thermal levels where CO2
absorption and desorption rates are in balance. One can notice a Acknowledgments
consistency between the theoretical values as expected by thermo-
dynamics and the experimental data. These predictions which are We would like to thank the nancial support of the Carbon
approximately similar to each other are valuable, given that they Management Canada who provided a Post-Doctoral fellowship to
allow setting adequate conditions for a twin conguration involv- Dr. Muhammad B.I. Chowdhury. We would also like to acknowl-
ing absorption and desorption uidized bed units. edge the Natural Sciences and Engineering Research Council of
Canada for the nancial support of the research. Authors would
also like to acknowledge Ms F. de Lasa for her assistance during
the preparation of this manuscript.

Appendix A. Thermodynamic properties of the CO2 sorbents


Table 2 studied
Quantitative capacity of the sorbents.
A.1. Heat capacity of various chemical species
Cycle Limestone Li4SiO4 Li4SiO4/HAc
CO2 CO2 CO2 CO2 CO2 CO2 Heat capacities of the various chemical species were dened in
captured released captured released captured released
(mL/gm) (mL/gm) (mL/gm)
J/mol-K [2831] with the temperature, T involved in the following
correlations given in degrees Kelvin:
1 151.6 147.1 48.3 46.9 52.3 51.9
2 109.6 105.7 43.1 41.9 51.3 50.9 Cp CO2 22:243 5:977  102 T  3:499  105 7:464  109
3 92.5 89.1 40.9 39.1 51.8 51.1
4 82.8 79.8 38.8 37.5 50.9 49.7
5 72.7 69.7 34.7 33.5 49.9 48.4 Cp CaCO3 82:34 0:0497T  1287000=T 2
6 65.7 62.3 31.8 30.2 48.4 46.6
7 60.6 56.7 27.8 26.4 46.5 45.0
8 55.6 50.1 25.9 23.7 45.2 44.1 Cp CaO 41:84 0:02025T  451870=T 2
9 49.8 46.9 24.4 22.9 44.8 43.5
10 43.1 40.9 23.1 21.1 44.4 42.9 Cp LiCO3 68:33 146:6103 T  162:5106 T 2 248:0109 T 3
 0:702106 T 2 298  683 K

900
Cp LiCO3 6:394 261:4103 T  102:0106 T 2 42:47109 T 3

800  0:168106 T 2 683 K  993 K


700
Temperature ( C)

Cp ZrO2 69:20 8:54103 T  0:862106 T 2 0:246109 T 3


o

600
500  1:382106 =T 2
400 Theoretical (CaO)
Experimental (Cadomin 1) Cp Li2 ZrO3 125:81 47:71103 T  25:78  105 =T 2
300 Theoretical (Li 4SiO4 )
Experimental (Li4 SiO4 )
200
Experimental (Li4 SiO4 /HAc) Cp Li2 SiO3 171:3 18:82103 T  17:66105 =T 2  971:4=T 1=2
100
0 Cp Li4 SiO4 1160:1 0:401T  3:263  106 =T 2
0 2 4 6 8 10
Cycles
A.2. Standard Gibbs Free Energies and standard enthalpies for various
Fig. 13. Theoretical and experimental inversion point temperature for CO2 chemical reactions
absorptiondesorption at the partial pressure of 0.1 bar. Full line: CaO theoretical
inversion point temperature, broken line: theoretical inversion point tempera-
ture for Li4SiO4, (j) experimental inversion point temperature for Cadomin 1,
Table A reports the standard Gibbs Free Energies and the stan-
(s) experimental inversion point temperature for Li4SiO4, (D) experimental dard heat of reaction for the various chemical species involved in
inversion point temperature for Li4SiO4 pretreated with acetic acid.). the calculations of the present study [22].
148 M.B.I. Chowdhury et al. / Chemical Engineering Journal 232 (2013) 139148

Table A [13] Y. Duan, K. Parlinski, Density functional theory study of the structural,
Standard Gibbs free energies and standard heat of enthalpies for various reactions. electronic, lattice dynamical, and thermodynamic properties of Li4SiO4 and its
capability for CO2 capture, Phys. Rev. B 84 (2011) 104113.
 
DGR kJ=mole DHR kJ=mole [14] M. Seggiani, M. Puccini, S. Vitolo, High-temperature and low concentration
CO2 sorption on Li4SiO4 based sorbents: study of the used silica and doping
CO3 Ca () CaO CO2 130.9 178.3 method effects, Int. J. Greenhouse Gas Control 5 (2011) 741748.
SiO3 Li2 Li2 CO3 () LiSiO4 CO2 92.83 141.97 [15] K. Wang, X. Guo, P. Zhao, F. Wang, C. Zheng, High temperature capture of CO2
CO3 Li2 ZrO2 () Li2 ZrO3 CO2 110.6 159.8 on lithium-based sorbents from rice husk ash, J. Hazard. Mater. 189 (2011)
Na2 CO3 SiO2 ! Na2 SiO3 CO2 42.66 86.614 301307.
[16] M. Olivares-Marin, T.C. Drage, M.M. Maroto-Valer, Novel lithium-based
sorbents from y ashes for CO2 capture at high temperatures, Int. J.
Greenhouse Gas Control 4 (2010) 623629.
References [17] Z. Qi, H. Daying, L. Yang, Y. Qian, Z. Zibin, Analysis of CO2 sorption/desorption
kinetic behaviors and reaction mechanisms on Li4SiO4, AlChE J. (2012). doi:
10.1002/aic.13861.
[1] J.G. Canadell, C. Le Qur, M.R. Raupach, C.B. Field, E.T. Buitenhuis, P. Ciais, T.J.
[18] J. Ortiz-Landeros, T.L. valos-Rendn, C. Gmez-Yez, H. Pfeiffer, Analysis and
Conway, N.P. Gillett, R.A. Houghton, G. Marland, Contributions to accelerating
perspectives concerning CO2 chemisorption on lithium ceramics using thermal
atmospheric CO2 growth from economic activity, carbon intensity, and
analysis, J. Therm. Anal. Calorim. 108 (2012) 647655.
efciency of natural sinks, Proc. Natl. Acad. Sci. 104 (2007) 1886618870.
[19] R. Rodrguez-Mosqueda, H. Pfeiffer, Thermokinetic analysis of the CO2
[2] H. Gupta, L.S. Fan, Carbonationcalcination cycle using high reactivity calcium
chemisorption on Li4SiO4 by using different gas ow rates and particle
oxide for carbon dioxide separation from ue gas, Ind. Eng. Chem. Res. 41
sizes, J. Phys. Chem. A 114 (2010) 45354541.
(2002) 40354042.
[20] J.W Tester, M. Modell, Thermodynamics and its Application, 3rd ed., Prentice
[3] J.R. Grace, A.A. Avidan, T.M. Knowlton, Circulating Fluidized Beds, Blackie
Hall PTR, New Jersey, USA, 1997.
Academic and Professional, London, UK, 1997.
[21] K. Annamalai, I.K. Puri, Advanced Thermodynamics Engineering, CRC Press LLC,
[4] M.A. Cuenca, E.J. Anthony, Pressurized Fluidized Beds, Blackie Academic &
New York, 2002.
Professional, Glasgow, U.K., 1995.
[22] I. Barin, Thermochemical Data of Pure Substances, VCH, Newyork, 2008.
[5] L.I. Eide, M. Anheden, A. Lyngfelt, C. Abanades, M. Younes, D. Clodic, A.A. Bill,
[23] V.L. Mejia-Trejo, E. Fregoso-Israel, H. Pfeiffer, Textural, structural, and CO2
P.H.M. Feron, A. Rojey, F. Giroudiere, Novel capture processes, Oil Gas J. Sci.
chemisorption effects produced on the lithium orthosilicate by its doping with
Technol. 60 (2005) 497508.
sodium (Li4xNaxSiO4), Chem. Mater. 20 (2008) 71717176.
[6] J. Ida, R. Xiong, Y.S. Lin, Synthesis and CO2 sorption properties of pure and
[24] D.K. Lee, An apparent kinetic model for the carbonation of calcium oxide by
modied lithium zirconate, Sep. Purif. Technol. 36 (2004) 4151.
carbon dioxide, Chem. Eng. J. 100 (2004) 7177.
[7] K. Yi, D. Eriksen, Low temperature liquid state synthesis of lithium zirconate
[25] M. Kruk, M. Jaroneic, Gas adsorption characterization of ordered
and its characteristics as a CO2 sorbent, Sep. Sci. Technol. 41 (2006) 283296.
organicinorganic nanocomposite materials, Chem. Mater. 13 (2001) 3169.
[8] Y. Duan, Electronic structural and electrochemical properties of lithium
[26] S.A. Greenberg, D. Sinclair, The polymerization of silicic acid, J. Phys. Chem. 59
zirconates and their capabilities of CO2 capture: a rst-principles density-
(1955) 435440.
functional theory and phonon dynamics approach, J. Renewable Sustainable
[27] Y. Li, C. Zhao, H. Chen, C. Liang, L. Duan, W. Zhou, Modied CaO-based sorbent
Energy 3 (2011) 013102.
looping cycle for CO2 mitigation, Fuel 88 (2009) 697704.
[9] K. Essaki, M. Kato, H. Uemoto, Inuence of temperature and CO2 concentration
[28] M.W. Chase Jr., NIST-JANAF themochemical tables, J. Phys. Chem. Ref. Data
on the CO2 absorption properties of lithium silicate pellets, J. Mater. Sci. 40
(1998) 11951.
(2005) 50175019.
[29] G.K. Moiseev, N.A. Vatolin, Interaction of lithium zirconates with lithium
[10] D.Y. Lu, R.W. Hughes, E.J. Anthony, Ca-based sorbent looping combustion for
under equilibrium conditions, Dokl. Phys. Chem. 388 (2003) 3337.
CO2 capture in pilot-scale dual uidized beds, Fuel Process. Technol. 89 (2008)
[30] M. Asou, T. Terai, Y. Takahashi, High-temperature enthalpy-increment
13861395.
measurements and derived heat capacity of lithium orthosilicate (Li4SiO4) at
[11] J. Blamey, E.J. Anthony, J. Wang, P.S. Fennell, The calcium looping cycle for
temperatures from 300 K to 1000 K, J. Chem. Thermodyn. 24 (1992) 273280.
large-scale CO2 capture, Prog. Energy Combust. Sci. 36 (2010) 260279.
[31] C. Tequi, P. Grinspan, P. Richet, Thermodynamic properties of alkali silicates:
[12] M. Kato, S. Yoshikawa, K. Nakagawa, Carbon dioxide absorption by lithium
heat capacity of Li2SiO3 and lithium bearing melts, J. Am. Ceram. Soc. 75 (1992)
orthosilicate in a wide range of temperature and carbon dioxide
26012604.
concentrations, J. Mater. Sci. Lett. 21 (2002) 485487.