Article

pubs.acs.org/est

Enhancement of Fast CO2 Capture by a Nano-SiO2/CaO Composite at

Ca-Looping Conditions
J. M. Valverde,*,† A. Perejon,‡ and L. A. Perez-Maqueda‡
†
Faculty of Physics, University of Seville, Avenida Reina Mercedes s/n, 41012 Seville, Spain
‡
Instituto de Ciencia de Materiales de Sevilla (C.S.I.C.-Univ. Seville), Americo Vespucio 49, 41092 Seville, Spain

ABSTRACT: In this paper we show the performance of a new CO2 sorbent

consisting of a dry physical mixture of a Ca-based sorbent and a SiO2 nanostructured
powder. Thermo-gravimetric analysis (TGA) performed at conditions close to the Ca-
looping process demonstrate that the rate of CO2 capture by the mixture is enhanced
during the fast carbonation stage of practical interest in applications. Moreover, the
residual capture capacity of the mixture is increased. SEM/EDX, physisorption, and
XRD analyses indicate that there is a relevant interaction between the nanostructured
SiO2 skeleton and CaO at high temperatures, which serves to improve the efficiency of
the transfer of CO2 to small reactive pores as well as the stability of the sorbent pore
structure.

1. INTRODUCTION the capture capacity of CaO decreases with increasing the

The Ca-looping process involves the carbonation of CaO and number of cycles, which is attributed to the decrease of the
the subsequent calcination of limestone (CaCO3) to regenerate reactive surface area as a result of sintering.3,7,8 In a very recent
the sorbent.1,2 Carbonation−calcination cycles are realized in work, Chen et al.10 have presented a synthetic sorbent prepared
practice by means of two interconnected fluidized beds by mixing rice husk ash (of high SiO2 content) with calcined
operating under atmospheric pressure. The CaO powder reacts limestone in an aqueous solution. CO2/SO2 capture capacity
in a fluidized bed reactor (carbonator) with the CO2 present in was enhanced for an optimum SiO2/CaO molar ratio of 0.2
the gas to form CaCO3 at temperatures around 600 °C. The under pressurized carbonation. In a previous work, Li et al.11
spent sorbent is then regenerated by calcining it at temper- had shown that a hydrated mixture of rice husk ash/CaO also
atures around 900 °C in a second fluidized bed reactor exhibited enhanced carbonation conversion at atmospheric
(calciner). In the calciner, limestone decomposes to yield CaO pressure for certain preparation conditions (hydration temper-
and a concentrated stream of CO2 ready to be stored. This ature of 75 °C, hydration time of 8 h, and SiO2/CaO molar
process is currently implemented in large-scale pilot plants, ratio of 1.0). The authors of these works argued that, because of
showing considerable potential for reducing postcombustion the higher melting point of calcium silicates formed by the
CO2 emissions.3,4 It also has a promising application in the reaction between SiO2 and Ca(OH)2, the resistance to sintering
production of hydrogen from natural gas by means of the steam was increased.10 It is also remarkable that in their experiments,
methane reforming process, which is enhanced by in situ the carbonation time was 15 min in ref 11 and 25 min in ref 10.
capture of CO2.5 When the carbonation time was decreased to 10 min the
Carbonation of CaO takes place in two stages.3,6−8 A first fast sorbent performance was reduced.10 As demonstrated by Wang
carbonation stage consists of CO2 sorption on the surface of et al.,12 it could be also possible that CO2 is sorbed by calcium
CaO particles, which is controlled by chemical reaction (kinetic silicates over long carbonation periods.
regime). Fast sorption finishes when the carbonate layer In our study, a fine Ca(OH)2 powder (Merck) is used as
reaches a thickness of about 30−50 nm.7 Then, it continues at a original CO2 sorbent. When this cohesive powder is fluidized,
much slower rate controlled by diffusion of CO2 through the its macroscopic behavior is heterogeneous (Geldart C
carbonated layer. The stage of practical interest is fast
carbonation since the sorbent must react with CO2 over Received: January 25, 2012
short contact times.1 Ideally, the sorbents should react fast Revised: May 4, 2012
toward an optimum capture capacity and maintain it as the Accepted: May 8, 2012
number of carbonation−calcination cycles is increased.1,9 Yet Published: May 2, 2012

© 2012 American Chemical Society 6401 dx.doi.org/10.1021/es3002426 | Environ. Sci. Technol. 2012, 46, 6401−6408
Environmental Science & Technology Article

Figure 1. (a) Thermogram obtained during pretreatment of Ca(OH)2 and nano-SiO2/Ca(OH)2 mixtures by heating them to 850 °C in a N2 dry
atmosphere (nominal wt% of SiO2 in the freshly prepared mixtures is indicated). The inset shows the wt % (on the total weight of the freshly
prepared samples) of SiO2 and CaO remaining after pretreatment, which was calculated from this thermogram. (b) CO2 capture capacity of the
nano-SiO2/CaO mixtures as the number N of carbonation−calcination cycles is increased (nominal SiO2 wt % is indicated). Results from
independent tests on different samples (not shown) demonstrated data reproducibility within a 10% relative variation. (c) and (d) Relative mass of
CO2 sorbed and desorbed as a function of time during carbonation−calcination cycles number 1 (c) and 100 (d) for CaO and a nano-SiO2/CaO
mixture (15 wt % SiO2). Variation of temperature as a function of time is shown in (d) (right axis).

behavior13). Stable gas channels are developed through which Ca(OH)2 particles due to electrostatic charge transfer.14,18 A
the gas bypasses the bed, hindering the efficiency of the gas− test performed at ambient pressure and temperature (reported
solids contact. As reported in ref 14, the fluidizability of this in 14) showed that the time for CO2 breakthrough in the
powder is enhanced by dry mixing it with a nanostructured effluent gas from a fluidized bed was increased for the nano-
SiO2 powder (Aerosil R974 supplied by Evonik Industries), SiO2/Ca(OH)2 mixture because of fluidization homogeniza-
which helps the destabilization of gas channels. Aerosil R974 is tion. Besides of the preparation method, a main difference with
an amorphous silica manufactured by flame hydrolysis with a SiO2/CaO sorbents presented elsewhere10,11 is that the SiO2
reported primary particle size of 12 nm and BET surface area of powder used in our work is nanostructured. As a result, it will
200 m2/g. Due to long pathways of primary nanoparticles in be shown in the present manuscript that the rate of fast
the flame reactor at high temperatures, these nanoparticles have carbonation (taking just tens of seconds) and residual capture
extremely short life as individual particles. They form fractal capacity are enhanced at Ca-looping conditions and atmos-
aggregates, of size on the order of micrometers, wherein pheric pressure for small values of the SiO2/CaO molar ratio.
nanoparticles are permanently held together by strong chemical
bonds due to sintering.15 Subsequently, these aggregates join 2. EXPERIMENTAL RESULTS
together due to attractive van der Waals forces, forming light A Q5000IR TG analyzer (TA Instruments) with a fast infrared
and fluidizable agglomerates of size between tens and hundreds furnace and a high sensitivity balance (<0.1 μg) characterized
of micrometers and density of the order of tens of kg/m3.16,17 by a minimum baseline dynamic drift (<10 μg) was used. The
When Ca(OH)2 is physically dry-mixed with the SiO2 powder, samples tested were the Ca(OH)2 powder and dry physical
SiO2 nanostructured agglomerates become coated with the mixtures of Ca(OH) 2 with nano-SiO 2 (nominal SiO 2
6402 dx.doi.org/10.1021/es3002426 | Environ. Sci. Technol. 2012, 46, 6401−6408
Environmental Science & Technology Article

Figure 2. (a) Sorption rates in the fast carbonation stage as the number of carbonation−calcination cycles is increased. (b) Time of duration of the
fast carbonation stage. (c) Capture capacities in the fast and slow carbonation stages, XKN and XDN, respectively. (d) Decarbonation rates during
calcination. Results are shown for CaO and for a nano-SiO2/CaO mixture (15 wt % SiO2). Solid thin lines in (c) are best power law fits to the data:
XKN = 0.365N−0.253 for the nano-SiO2/CaO mixture and XKN = 0.386N−0.432 for CaO.

concentrations between 10% and 30% by SiO2 weight on As discussed in ref 14 this SiO2 percentage gives the optimum
Ca(OH)2 weight). The mass of the samples ranged between 10 amount needed to achieve a maximum contact of the surface
and 20 mg. nano-SiO2 with Ca(OH)2 in the initial mixture.
Once the sample was placed in the TG balance, it was heated Figure 1c and 1d show the relative mass of CO2 sorbed and
to 850 °C in a N2 atmosphere (20 °C/min). As the desorbed as a function of time during the first and last cycles
temperature was raised, the sample went across dehydration (15 wt % SiO2). As well documented in the literature,3,6−8
and desorption of CO2 as seen in the thermogram shown in carbonation takes place in two stages: an initial fast carbonation
Figure 1a. After preheating, the sorbent consisted thus of a stage controlled by surface reaction and a subsequent slower
mixture of CaO and nano-SiO2 at weight percentages shown in stage controlled by diffusion. The presence of nano-SiO2 has a
the inset of Figure 1a. Once preheating was finished, the relevant influence on the rate of sorption during fast
temperature was decreased to 650 °C and the sample was carbonation (dXKN/dt) and the rate of decarbonation during
subjected to a flow (100 cm3 min−1) of a gas mixture of CO2 calcination (−dX/dt), which take place at a faster rate. This is
(15% volume concentration) and air for 5 min. Then, the clearly seen in Figure 2a and 2d, where dXKN/dt and −dX/dt
temperature was raised to 850 °C (heating rate 300 °C/min) are plotted versus N. Figure 2b illustrates the time lag of the fast
and the sample was calcined in an air flow (100 cm3 min−1) carbonation stage (Δt). In the first cycle, fast carbonation for
during 5 min for decarbonation. After calcination, the CaO takes about ΔtCaO ≃ 50 s, which is similar to the value
temperature was decreased (cooling rate 300 °C/min) to reported for calcined samples of fine natural limestones in
proceed again with carbonation. similar conditions.7 ΔtCaO decreases with N and tends to ΔtCaO
Data of the mass of CO2 sorbed relative to the mass of CaO ≃ 20 s for large N (Figure 2b). On the other hand, Δt is initially
(capture capacity X) are plotted in Figure 1b as a function of smaller for the nano-SiO2/CaO mixture (ΔtCaO ≃ 30 s) but it
the number of carbonation−calcination cycles N. X(N) decreases at a slower rate as N is increased and tends to
decreases with increasing N and tends to a residual value Xr ΔtSiO2/CaO ≃ 20 s for N > 20, which is similar to ΔtCaO that
around 0.1 for CaO, which is similar to the value typically keeps slowly decreasing even at large N. The data plotted in
reported for natural limestones.19 The values of X(N) for the Figure 2c show the capture capacities during the fast stage
nano-SiO2/CaO mixtures become relatively higher as N is (XKN) and the slow stage (XDN). There is a clear effect of nano-
increased. The highest residual capacity is obtained for a 15 wt SiO2 on the rate of decrease of XKN with N as compared to the
% of nano-SiO2 (SiO2/CaO molar ratio of 0.25), Xr ≃ 0.18, rate of decrease of XDN, which is similar for both sorbents. XKN
which is almost twice the residual capacity obtained for CaO. decreases at a relatively higher rate in the case of the original
6403 dx.doi.org/10.1021/es3002426 | Environ. Sci. Technol. 2012, 46, 6401−6408
Environmental Science & Technology Article

Figure 3. BJH desorption (dV/dD) pore volume distributions for the Ca(OH)2 (a) and nano-SiO2 powders (b) used in our study as received and
after being heated to 850 °C. The inset of (a) shows SEM pictures of the surface of Ca(OH)2 particles before (a and b) and (c) after thermal
pretreatment (CaO). The inset of (b) shows a TEM picture of nano-SiO2 agglomerates before thermal pretreatment (a) and a SEM picture after
thermal pretreatment (b).

sorbent (rK  − Δ ln XKN/Δ ln N ≃ 0.432), approaching XDN produced by thermal decomposition of Ca(OH)2, it is K ≃
for large N. In contrast, rK is notably lessened for the nano- 10−4 min−1 at 650 °C, K ≃ 10−2 min−1 at 850 °C. Accordingly,
SiO2/CaO mixture, being rK ≃ 0.253, which is similar to the no appreciable surface area reduction is expected in our
decay rate of the capture capacity in the diffusion regime. Thus, experiments during the transitional periods (t = 40 s). On the
XKN is kept at a relatively high value as compared with XDN at other hand, heating rates typically employed in conventional
large N for the modified sorbent. These results illustrate that TGA instruments (of about 50 °C min−1 or even smaller21) are
fast capture capacity and sintering resistance are enhanced at likely to yield nonnegligible sintering in the transitional periods,
conditions close to the Ca-looping process. which would affect the CO2 capture performance.
A relevant aspect of TGA experiments is the heating and The effect of heating the sorbents during pretreatment and
cooling rates between cycles. The use of an infrared furnace in subsequent calcination cycles on their BET surface area and
our TGA runs allowed us for heating and cooling the sample pore size distribution (BJH method) was investigated by means
very quickly (300 °C min−1), thus the carbonation−calcination of a TriStar II 3020 V1.03 physisorption analyzer operated by
transitions took place in just 40 s. To assess whether sintering N2 sorption at 77 K. This analysis was complemented by SEM
in this period might influence the performance of the sorbent, observations using a HITACHI Ultra High-Resolution S-5200
the results reported by Borgwardt20 can be used. The reduction equipped with an EDX spectroscopy unit (Bruker AXS
of the specific surface area ΔS is expressed as a function of time Microanalysis GmbH) for chemical characterization of the
t as (ΔS/S0)γ = Kt where S0 is the initial specific surface area sample. Figure 3 shows pore size distributions obtained for
and K is the sintering rate constant, which increases with both the Ca(OH)2 and nano-SiO2 powders as received and
temperature T according to the law Δ log K ∝ −1/T.20 The after being preheated individually. The pore size distribution of
exponent γ measured for CaO is γ ≃ 2.7.20 In the case of CaO the untreated Ca(OH)2 powder (Figure 3a) has a pronounced
6404 dx.doi.org/10.1021/es3002426 | Environ. Sci. Technol. 2012, 46, 6401−6408
Environmental Science & Technology Article

peak at a pore size close to 3−4 nm in accordance with SEM Physisorption measurements were also carried out after the
images showing tiny crystals with a typical size of less than tens powders were subjected in a furnace to 10 cycles changing the
of nm (see inset of Figure 3a). The morphology of the temperature between 650 and 850 °C as in TGA tests. The
Ca(OH)2 surface is severely changed after heating due to pore size distribution of the nano-SiO2 powder remained
sintering as seen in the inset of Figure 3a. Accordingly, the almost unaffected (Figure 5a). After these 10 calcinations the
region in the pore size distribution comprising pore sizes
smaller than 10 nm is flattened and the distribution weight is
shifted toward larger pore sizes (Figure 3a). As regards nano-
SiO2, SEM images (inset of Figure 3b) show coalescence and
growth of the nanoparticles in the agglomerates but changes in
the pore size distribution are not relevant (Figure 3b). Pore size
distributions of the mixtures are plotted in Figure 4. As

Figure 5. BJH desorption (dV/dD) pore volume distributions of

nano-SiO2 (a) and CaO (b) preheated at 850 °C and after 10
calcinations. The insets show SEM pictures of the samples after 10
calcination cycles.

BET surface area was 180 m2/g, which is similar to the value
obtained for the powder after first calcination. In contrast, the
Figure 4. BJH desorption (dV/dD) pore volume distributions for
Ca(OH)2 powder suffered a considerable decrease of the BET
Ca(OH)2 and for the mixtures before (a) and after (b) being heated to surface area: from 16.3 m2/g for the powder as received down
850 °C (wt % of nano-SiO2 is indicated). The inset of (a) shows the to 13 m2/g after first calcination and 9 m2/g after 10
relative decrease of BET surface area after heating the mixtures as a calcinations (Figure 5b). The pore distributions of the mixtures
function of the SiO2 wt %. The inset of (b) shows the pore volume were less distorted as the percentage of nano-SiO2 was
distributions of a 20 wt % SiO2 sample preheated at 850 °C and after increased (inset of Figure 4b), which indicates that they have
10 calcinations. a higher thermal stability. From SEM images of the calcined
mixtures (Figure 6a and 6b), regular cells (not visible in the
calcined original sorbent) and reaction spots were observed.
expected from the pore size distributions obtained for both Figure 6c and 6d illustrate another feature observed in the
powders individually, the mixtures containing a high percentage modified samples after the TGA cycles, which is the
of nano-SiO2 preserve a relatively larger amount of pores development of surface cracks. An EDX spectrum of the
smaller than 10 nm after being heated. Accordingly, the relative surface of a calcined mixture is shown in Figure 6e. The X-ray
decrease of the BET surface area caused by sintering (inset of mapping image shows Si and Ca local spots homogeneously
Figure 4a) is lessened as the percentage of nano-SiO2 is distributed, indicating that SiO2 and CaO have reacted during
increased. calcination. A well-defined peak in the EDX spectrum reveals
6405 dx.doi.org/10.1021/es3002426 | Environ. Sci. Technol. 2012, 46, 6401−6408
Environmental Science & Technology Article

Figure 6. SEM pictures of a nano-SiO2/CaO mixture: (a) after thermal pretreatment (the arrow indicates a reaction spot); (b) after 10 calcinations;
(c) and (d) after 100 carbonation-calcination cycles. (e) EDX spectrum averaged on the region shown in the inset (500 × 500 nm2) of a 10 times
calcined mixture. The inset is an X-ray mapping image of the region shown, where green and red spots indicate the presence of Ca and Si,
respectively. A quantitative mapping analysis yields a wt % of 9.7% for C, 2.5% for Si, 60.5% for Ca, and 27.3% for O.

also the presence of C, which suggests the occurrence of surface A general problem to apply the Ca-Looping process for
carbonation. postcombustion CO2 capture is elutriation of fine particles at
XRD analysis was performed on modified sorbent samples the typically high postcombustion gas velocities (above 1 m/
using a Bruker D8 Advance powder diffractometer equipped s).3 However, the effective particles in our preheated sorbent
with a high temperature chamber, which allows for analyzing consist of large agglomerates wherein primary particles are
the samples under controlled atmosphere and temperature. strongly linked due to sintering at contacts. Particle size was
Thermal pretreatment and a carbonation−calcination cycle was measured by means of a Mastersizer 2000 (Malvern Instru-
thus performed in situ. Temperature conditions were the same ments) based on laser diffraction of the sample dispersed in a
as in TGA tests. Preheating and calcination were performed in fast-moving compressed air stream (10 kPa). A high-velocity air
N2 whereas carbonation was performed in CO2. XRD patterns jet subjects the agglomerates to very high velocities (above 10
are shown in Figure 7 showing calcium silicate formation. m/s), which might yield mechanical attrition.22 The average
During carbonation, most of the CaO was carbonated to particle size of the dispersed sample measured was 91 μm
CaCO3, which subsequently decarbonated during calcination. (volume weighted mean D[4,3]), which is similar to the particle
On the other hand, the proportion of calcium silicates formed size used in some Ca-looping pilot scale tests.23 On the other
during preheating changed only slightly after carbonation and hand, it must be remembered that the Ca-looping process also
calcination as revealed by the quantitative Rietveld analyses has an application in SE-SMR (sorption-enhanced steam
performed on the XRD patterns (Figure 7c inset). Most of CO2 methane reforming), where fluidized beds operate at lower
sorption occurred thus by carbonation of CaO. gas velocities in the bubbling regime (∼0.01−0.1 m/s).5 With
6406 dx.doi.org/10.1021/es3002426 | Environ. Sci. Technol. 2012, 46, 6401−6408
Environmental Science & Technology Article

Figure 7. XRD patterns of a nano-SiO2/CaO mixture preheated (a) and subjected to carbonation (b) and calcination (c) in situ in the DRX
chamber at same temperature conditions and time periods as in TGA tests. The inset in (c) shows the variation of the composition percentages
(measured by means of a Rietveld analysis) after carbonation (b) and calcination (c).

an average particle size close to 100 μm, elutriation of the yield a larger number of reactive small pores directly exposed to
preheated sorbent in a lab-scale bubbling bed was not CO2, which otherwise would be placed in the interior of CaO
significant as it was checked by performing a fluidization test. agglomerates wherein the diffusive transfer of CO2 is slow. The
At a gas velocity of 0.12 m/s, the relative mass of powder increase of the effective contact area to which CO2 can be
elutriated from the bed (Δm/m0) as a function of time t could rapidly transferred would make fast carbonation proceed at a
be well fitted to the equation Δm/m0 = a(1−e−bt), where a = higher rate as seen in our results (Figure 2a). Even though the
0.0746, b = 0.1207, and t is in minutes (Rsqr = 0.998). After 1 higher proportion of exposed small pores would enhance
min just about 1% of the initial mass (m0 = 3.787 g) was surface carbonation, these small pores become quickly
elutriated. saturated, which would explain the initially shorter duration
of fast carbonation for the nano-SiO2/CaO mixtures (Figure
3. DISCUSSION 2b). As the number of cycles is increased, the duration of fast
Fast carbonation in sorbents with small pore size is limited by carbonation is decreased at a larger rate for the CaO powder
the lack of void space available for the growth of the carbonate probably due to more intense sintering. The active small pores
layer in the kinetic regime.24 Moreover, fast carbonation on the in the nano-SiO2/CaO mixture would still allow for a relatively
external surface of the large CaO agglomerates may give rise to high contact area for fast carbonation as N increases due to
a carbonate layer that blocks the transfer of CO2 to the pores in thermal stability enhancement. This would contribute to lessen
the interior of these agglomerates. A main effect of nano-SiO2 is the progressive shortening of the fast carbonation period
to improve the dispersibility of CaO agglomerates. This would (Figure 2b) while the carbonation rate is kept still high as
6407 dx.doi.org/10.1021/es3002426 | Environ. Sci. Technol. 2012, 46, 6401−6408
Environmental Science & Technology Article

compared to the carbonation rate in CaO. In addition, surface (6) Bathia, S. K.; Perlmutter, D. D. Effect of the product layer on the
structures and cracks observed in SEM images of the nano- kinetics of the CO2-lime reaction. AIChE J. 1983, 19 (1), 79−86.
SiO2/CaO mixtures (Figure 6), which could be originated by (7) Grasa, G.; Murillo, R.; Alonso, M.; Abanades, J. C. Application of
the increase of molar volume associated with the formation of the random pore model to the carbonation cyclic reaction. AIChE J.
2009, 55 (5), 1246−1255.
calcium silicates in the heated mixture, might contribute to (8) Arias, B.; Abanades, J. C.; Grasa, G. S. An analysis of the effect of
further increase the effective surface area for carbonation. carbonation conditions on CaO deactivation curves. Chem. Eng. J.
Let us finally remark that the conditions used in the TGA 2011, 167 (1), 255−261.
runs were chosen to assess the sorbents performance as being (9) Grasa, G.; Gonzalez, B.; Alonso, M.; Abanades, J. C. Comparison
close to conditions in the Ca-Looping process.9 Manovic et al. of CaO-based synthetic CO2 sorbents under realistic calcination
have pointed out that it would be more realistic to test conditions. Energy Fuels 2007, 21 (6), 3560−3562.
calcination at high CO2 volume concentrations (80−90%) and (10) Chen, H.; Zhao, C.; Ren, Q. Feasibility of CO2/SO2 uptake
higher temperature.21,25 Under these harsher calcination enhancement of calcined limestone modified with rice husk ash during
conditions, the capture capacity of CaO is decreased due to pressurized carbonation. J. Environ. Manage. 2012, 93 (1), 235−244.
enhanced sintering.25 The results obtained here suggest that the (11) Li, Y.; Zhao, C.; Ren, Q.; Duan, L.; Chen, H.; Chen, X. Effect of
rice husk ash addition on CO2 capture behavior of calcium-based
effect of nano-SiO2 will yield also a better performance when
sorbent during calcium looping cycle. Fuel Process. Technol. 2009, 90
calcination is carried out at higher temperatures due to (6), 825−834.
enhanced contact area and improved thermal stability. A (12) Wang, M.; Lee, C.-G.; Ryu, C.-K. CO2 sorption and desorption
detailed study of the performance of the mixtures as calcination efficiency of Ca2SiO4. Int. J. Hydrogen Energy 2008, 33 (21), 6368−
conditions become harsher will be the subject of a separate 6372.
work. (13) Geldart, D. Types of gas fluidization. Powder Technol. 1973, 7

■
(5), 285−292.
AUTHOR INFORMATION (14) Valverde, J. M.; Pontiga, F.; Soria-Hoyo, C.; Quintanilla, M. A.
S.; Moreno, H.; Duran, F. J.; Espin, M. J. Improving the gas solids
Corresponding Author contact efficiency in a fluidized bed of CO2 adsorbent fine particles.
*Phone: +34 954550960; fax: +34 954239434; e-mail: Phys. Chem. Chem. Phys. 2011, 13, 14906−14909.
jmillan@us.es; mail: Department of Electronics and Electro- (15) Hyeon-Lee, J.; Beaucage, G.; Pratsinis, S. E.; Vemury, S. Fractal
magnetism, Faculty of Physics, Avda. Reina Mercedes s/n, analysis of flame synthesized nanostructured silica and titania powders
41012 Seville, Spain. using small-angle x-ray scattering. Langmuir 1998, 14, 5751−5756.
(16) Valverde, J.; Castellanos, A. Fluidization of nanoparticles: A
Notes simple equation for estimating the size of agglomerates. Chem. Eng. J.
The authors declare no competing financial interest. 2008, 140 (1−3), 296−304.

■
(17) Quintanilla, M. A. S.; Valverde, J. M.; Espin, M. J.
ACKNOWLEDGMENTS Electrofluidization of silica nanoparticle agglomerates. Ind. Eng.
Chem. Res. 2012, 51 (1), 531−538.
This work was supported by the Andalusian Regional (18) Gallo, C. F.; Lama, W. L. Some charge exchange phenomena
Government (Junta de Andalucia, contract FQM-5735) and explained by a classical model of the work function. J. Electrostat. 1976,
Spanish Government Agency Ministerio de Ciencia e 2 (2), 145−150.
Innovacion (contracts FIS2011-25161 and CTQ2011-27626). (19) Grasa, G. S.; Abanades, J. C. CO2 capture capacity of CaO in
We are grateful to Drs. MAS Quintanilla, M. J. Espin, and C. long series of carbonation/ calcination cycles. Ind. Eng. Chem. Res.
Soria-Hoyo for their help with sample preparation and 2006, 45 (26), 8846−8851.
characterization. We also acknowledge technical assistance (20) Borgwardt, R. H. Sintering of nascent calcium oxide. Chem. Eng.
Sci. 1989, 44 (1), 53−60.
from the Microscopy, X-ray, and Functional Characterization
(21) Lu, D. Y.; Hughes, R. W.; Anthony, E. J.; Manovic, V. Sintering
services of the Innovation, Technology and Research Center and reactivity of CaCO3-based sorbents for in situ CO2 capture in
(CITIUS, University of Seville), and the Physisorption service fluidized beds under realistic calcination conditions. J. Environ. Eng.
of the Instituto de Ciencia de Materiales de Sevilla (CISC-Univ. 2009, 135 (6), 404−410.
Seville). (22) Mueller, N.; Ecker, F. Determining granule breaking strength by

■
laser diffraction method. Pharm. Ind. 2011, 73 (11), 2061−2070.
REFERENCES (23) Gonzalez, B.; Alonso, M.; Abanades, J. C. Sorbent attrition in a
carbonation/calcination pilot plant for capturing CO2 from flue gases.
(1) Abanades, J. C.; Anthony, E. J.; Lu, D. Y.; Salvador, C.; Alvarez, Fuel 2010, 89, 2918−2924.
D. Capture of CO2 from combustion gases in a fluidized bed of CaO. (24) Alvarez, D.; Abanades, J. C. Pore-size and shape effects on the
AIChE J. 2004, 50 (7), 1614−1622. recarbonation performance of calcium oxide submitted to repeated
(2) Alonso, M.; Rodriguez, N.; Gonzalez, B.; Grasa, G.; Murillo, R.; calcination/recarbonation cycles. Energy Fuels 2005, 19, 270−278.
Abanades, J. C. Carbon dioxide capture from combustion flue gases (25) Manovic, V.; Charland, J.-P.; Blamey, J.; Fennell, P. S.; Lu, D. Y.;
with a calcium oxide chemical loop. Experimental results and process Anthony, E. J. Influence of calcination conditions on carrying capacity
development. Int. J. Greenhouse Gas Control 2010, 4 (2), 167−173. of CaO-based sorbent in CO2 looping cycles. Fuel 2009, 88, 1893−
(3) Blamey, J.; Anthony, E. J.; Wang, J.; Fennell, P. S. The calcium 1900.
looping cycle for largescale CO2 capture. Prog. Energy Combust. Sci.
2010, 36 (2), 260−279.
(4) Wang, W.; Ramkumar, S.; Li, S.; Wong, D.; Iyer, M.; Sakadjian, B.
B.; Statnick, R. M.; Fan, L. S. Subpilot demonstration of the
carbonation-calcination reaction (CCR) process. High temperature
CO2 and sulfur capture from coal-fired power plants. Ind. Eng. Chem.
Res. 2010, 49 (11), 5094−5101.
(5) Johnsen, K.; Ryu, H. J.; Grace, J. R.; Lim, C. J. Sorption-enhanced
steam reforming of methane in a fluidized bed reactor with dolomite as
CO2-acceptor. Chem. Eng. Sci. 2006, 61 (4), 1195−1202.

6408 dx.doi.org/10.1021/es3002426 | Environ. Sci. Technol. 2012, 46, 6401−6408