Towards Solar Thermochemical Carbon Dioxide Capture via Calcium Oxide Looping : A Review

This article presents a review of research in the field of solar thermochemical carbon dioxide capture based on calcium oxide looping. Topics covered include thermodynamics, kinetics, and heat and mass transfer as well as reactor design, modeling, and testing.


INTRODUCTION
Carbon dioxide (CO 2 ) makes up 77% of all greenhouse gas emissions worldwide and 80% in the United States, primarily due to the use of fossil fuels as the source of energy (U.S. Energy Information Administration, 2009).Worldwide energy use continues to grow at a rate of about 1.6% per year, and 87% of that energy comes from fossil fuel sources (BP Statistical Review of World Energy, 2012; U.S. Energy Information Administration, 2011).At the same time, CO 2 is a valuable commodity with applications in the pharmaceutical, medical, environmental, food and beverages, petrochemical, oil recovery, metallurgy, and manufacturing industries (Stalkup, 1978;Pierantozzi, 2003).In 2005, industrial CO 2 use was estimated to be 6.3 Gt-CO 2 /yr (Metz et al., 2005).An emerging use of CO 2 is for the solar-driven production of synthesis gas, which in turn can be used to obtain synthetic hydrocarbon fuels (Steinfeld et al., 1998;Chueh et al., 2005;Leckel, 2009;Hathaway et al., 2011).Employment of CO 2 capture systems may become necessary in the near future to stabilize CO 2 concentration in the atmosphere if the world continues to rely on fossil fuels for energy production.Such systems can also become a key technical solution to close the anthropogenic carbon cycle in a renewable-based energy system involving solarderived hydrocarbon fuels as the primary chemical energy carriers.
The theoretical minimum energy required to capture CO 2 is set by the reversible separation work of a binary CO 2 -containing gas mixture, given by min ln( ) (1 ) ln(1 ) 1000 where min W is the minimum work of separation in kJ per mole of gas separated, R is the universal gas constant, T is the temperature, and x is the mole fraction of the gas being separated from the mixture (Kalafati, 1991).min W is shown in Fig. 1 as a function of mole fraction of CO 2 in the binary mixture.At CO 2 molar fractions below 8 × 10 -4 the work required is over 20 kJ/mol CO 2 and decreases to 7 kJ/mol CO 2 at a CO 2 molar fraction of 0.15.
Many carbon dioxide capture technologies have been investigated to date, including pre-combustion capture, post-combustion capture, oxycombustion, and capture from industrial processes (Metz et al., 2005).Methods used by capture technologies include physical or chemical absorption, adsorption, membrane separation, and cryogenic separation (Rackley, 2010).Absorption technologies have been in commercial use for over 50 years but have only recently been applied to low concentrations of CO 2 such as those found in exhaust gases.Commonly used solvents and sorbents in chemical absorption capture are amines, carbonates, and aqueous ammonia.Monoethanolamine (MEA) is the most common solvent used in commercial applications (Wilson and Gerard, 2007).Some solvents used in physical absorption are methanol, propylene carbonate, and dimethyl ethers of polyethylene glycol.Some disadvantages to absorption capture are the high amount of energy required to regenerate the solvent, degradation of the solvent by other components of flue gas such as particulates, SO x and NO x , and decreased Fig. 1.Thermodynamic minimum separation work of a binary gas mixture (Matthews and Lipiński, 2012).Reproduced with permission from Elsevier.reactivity of the solvent with successive cycles (Chapel et al., 1999;Aaron and Tsouris, 2005).Chemical sorbents for CO 2 capture by adsorption include various metal oxides and solid amines.Physical sorbents in use include activated carbon, activated alumina, and silica gel (Rackley, 2010).The main disadvantages to adsorption are low reaction rates and reduced effectiveness at CO 2 concentrations higher than 0.04 to 1.5% (IEA Greenhouse Gas R&D Programme, 1994).In membrane separation, the membrane essentially works as a filter, separating out CO 2 from the rest of the gas flow (Aaron and Tsouris, 2005).There are many materials used as membranes, but they can generally be divided into the categories of porous and nonporous.The main disadvantages to membrane separation are higher energy requirements, lower percentage of CO 2 removed than a chemical absorption based process, and reduced effectiveness at lower CO 2 concentrations (IEA Greenhouse Gas R&D Programme, 1994).Cryogenic separation relies on the difference in boiling points of the exhaust gas constituents.Cryogenic separation of CO 2 is typically used in the production of sales-quality natural gas.The main disadvantage to cryogenic separation of CO 2 is the large energy cost associated with cooling the gases (Aaron and Tsouris, 2005).The costs of each type of capture vary greatly depending on the specific situation.For example, Tuinier et al. (2011) state that cryogenic capture is favorable when low cost liquefied natural gas (LNG) is available, absorption is favorable when low cost steam is available, and membrane separation has advantages when neither LNG nor steam are cheaply available.As noted in Metz et al. (2005), the majority of carbon capture technologies are in the development stage, so costs are estimated based on analysis of hypothetical power plants.The sensitivity of these analyses to assumptions makes it difficult to draw firm conclusions about cost when comparing different studies.
One chemical absorption approach to CO 2 capture, often referred to as CaO-CaCO 3 looping, uses calcium oxide (CaO) carbonation combined with calcium carbonate (CaCO 3 ) calcination: Carbonation step (exothermic): Calcination step (endothermic): is the molar reaction enthalpy change at 298 K.A 3 step cycle involving hydration of CaO to form calcium hydroxide, Ca(OH) 2 , has also been proposed to avoid the kinetic limitations of the CaO-carbonation reaction: Hydration step (exothermic): Carbonation step (exothermic): Calcination step (endothermic): The cycles (2a)-( 2b) and ( 3a)-(3c) operate at low pressures, can capture more CO 2 per unit mass of sorbent than other processes, and use CaO as a low cost sorbent that is obtained from natural limestone and dolomite (Gupta and Fan, 2002;Abanades et al., 2004a).A challenge for realizing the CaO carbonation-calcination cycles is the high temperature required for the calcination step, typically achieved via combustion of fossil fuels.An alternative to combustion is to employ concentrated solar energy as the high-temperature source of process heat.The latter approach allows mitigation of additional CO 2 emissions associated with the generation of the process heat.The use of concentrated solar radiation has been proposed to drive numerous hightemperature processes, which can potentially run continuously if coupled with thermal storage (Fletcher, 2001;Steinfeld and Palumbo, 2001;Kodama and Gokon, 2007).
CO 2 capture using a CaO based cycle was investigated by Shimizu et al. (1999).A pair of conceptual fluidized beds connected by solid transport pipes and attached to a 1000 MW air-fired coal power plant was studied.The carbonation temperature was 873 K and the calcination temperature was 1223 K. Heat was recovered from the exothermic carbonation reaction and the cooling stream of CO 2 using a secondary steam cycle.Use of heat recovery boosted the plant efficiency to 33.4% compared to 32% for an oxygen-fired coal plant with heat recovery.
An 18 kW pilot scale batch mode fluidized bed for the CaO carbonation-calcination cycle studied by Abanades et al. (2004b) is shown in Fig. 2. The fluidized bed was shown to be effective for CO 2 capture at a temperature near 650°C.The carbonation reaction was fast enough at atmospheric Fig. 2. Pilot scale fluidized-bed carbonator (Abanades et al., 2004b).Reproduced with permission from John Wiley and Sons.pressure to completely remove CO 2 from the input gas at a bed height of 0.25 m.The reactivity of the CaO decreased with an increased number of cycles, although the rate of decay and the residual conversion depended on the type of limestone used.Havelock limestone behaved similarly to limestone in other studies and performed better than Cadomin limestone.
A 75 kW pilot scale dual fluidized bed setup was investigated by Lu et al. (2008).It was the first to feature a continuous cycle utilizing two beds rather than a batch cycle where carbonation and calcination occurred in the same location.Cyclones collected particles from the top of the carbonator and cycled them to the calciner and vice versa.The solids flow was controlled by solenoid valves.CO 2 concentration entering the carbonator was nominally 15%.Heat was supplied to the calciner by electric heaters or by burning biomass or coal.The calcination temperature ranged from 850 to 950°C and the carbonation temperature ranged from 580 to 720°C.During the first 10 cycles, the fraction of CO 2 removed from the inlet gas, was greater than 90%, and it decreased to approximately 70% after 25 cycles.Increasing the carbonation temperature improved the efficiency in later cycles.The authors speculated that this is due to particle sintering and pore plugging blocking the ability for CO 2 to reach deep pores after cycling of the sorbent.
Several other laboratory scale dual fluidized bed setups have been investigated, and three were summarized by Rodriguez et al. (2011).The facilities were 10-75 kW dual fluidized beds operating in several regimes of fluidization.
Carbonation temperatures ranged from 600-700°C and calcination temperatures ranged from 800-900°C.Two of the facilities operated continuously while the other was semi-continuous.The semi-continuous setup allowed for the effect of cycling to be observed at the expense of achieving steady state, while the continuous setups contained a mixture of particles with different histories but could achieve steady state operation.Each facility used different fuel and limestone sources, but all achieved CO 2 capture efficiencies above 70%, demonstrating that CO 2 capture using fluidized beds is industrially viable for a wide range of conditions.
To date, researchers have demonstrated that CO 2 capture via the CaO carbonation-calcination cycle is a feasible method.It has been shown that fluidized beds are suitable for the process because they provide favorable heat and mass transfer conditions in the reactor and can be scaled up as needed.However, while there has been a large body of research conducted regarding CO 2 capture with CaO using fossil energy, using concentrated solar energy to drive the cycle is a relatively new concept.This paper presents a review of past work on CaO-based CO 2 capture, with a particular focus on the use of solar energy to drive the process.Thermodynamic analyses of the cycles are discussed, followed by particle level studies of kinetics and heat and mass transfer.The paper concludes with a summary of reactor level design, modeling, and experiments.Nikulshina et al. (2006) performed a thermodynamic analysis for CO 2 capture from atmospheric air using the three step carbonation cycle (3a)-(3c).The model system is shown in Fig. 3. Calcium hydroxide was carbonated with atmospheric air containing 500 ppm CO 2 at 500 K, the resulting CaCO 3 calcined at 1500 K to form CaO, and finally slaked at 353 K to form Ca(OH) 2 .Assumptions made included ideal heat exchange, adiabatic carbonation, pure substances, and chemical equilibrium.Pumping work was not included in the analysis.The total heat required for the cycle without heat recovery was 12.1 MJ/mol CO 2 captured.The addition of two heat exchangers to recover heat from the CO 2 -depleted air exiting the carbonator and the CO 2 exiting the calciner reduced the heat requirement to 2.5 MJ/mol CO 2 captured.Rodriguez et al. (2008) examined the heat requirements in a coal fired calciner as a function of carbonation conversion, coal composition, and sorbent makeup flow.At a residual carbonation conversion of 0.075 and no makeup flow of sorbent, the heat requirement for the calciner was 36.9% of the plant input.Assuming a 1 mol CO 2 /s capture rate and using the study data this translates to about 3.9 MJ/mol CO 2 captured.Martinez et al. (2012) examined the effects of adding solid heat recovery to the cycle (2a)-( 2b) by simulating four different solid heat exchange configurations.Indirect solid heat exchange between hot CaO and cool CaCO 3 was the most efficient option, increasing thermal efficiency by up to 2% compared to a case with no solid heat recovery.A solid heat exchange system with a mixing seal valve showed no improvement compared to the base case.A system with a heat recovery fluidized bed preheating CaCO 3 with hot CO 2 increased thermal efficiency by up to 1.4%.

THERMODYNAMIC ANALYSES
A thermodynamic analysis of the cycle (2a)-( 2b) was performed to determine the effect of input CO 2 concentration, reaction temperatures, and gas and solid phase heat recovery on the total solar energy required (Matthews and Lipiński, 2012).The system used for the analysis is shown in Fig. 4. Input gas at T 0 containing a variable amount of CO 2 was preheated by CO 2 -depleted gas exiting the carbonator, CO 2 exiting the calciner, and the exothermic carbonation reaction before entering the carbonator at T 3 .CO 2 in the input gas reacted with CaO to form CaCO 3 at T carb .CaCO 3 exited the carbonator and was preheated by CaO and CO 2 exiting the calciner.In the calciner, concentrated solar energy provided heat for the endothermic calcination reaction at T calc .CaO and CO 2 exited the calciner and were used for preheating before the cycle began again.Chemical equilibrium was assumed for each reaction and any heat not recovered was rejected to the environment.(Nikulshina et al., 2006).Reproduced with permission from Elsevier.

Fig. 4.
Cycle for CO 2 capture with gas and solid heat recovery (Matthews and Lipiński, 2012).Reproduced with permission from Elsevier.
The required solar energy for a carbonation temperature of 673 K and a calcination temperature of 1273 K was greater than 45 MJ/mol CO 2 captured for a CO 2 concentration of 300 ppm and no heat recovery.The minimum value of solar Q for 100% solid heat recovery and 0% gas heat recovery was 303 kJ/mol CO 2 captured .solar Q for 100% gas heat recovery and 0% solid heat recovery was 283 kJ/mol CO 2 captured .solar Q for 100% gas and solid heat recovery was 207 kJ/mol CO 2 captured .Gas heat recovery was more important at low CO 2 input concentrations because of the large amount of inert gas that must be heated to the carbonation temperature.At higher CO 2 input concentrations near 15%, solid heat recovery became nearly as important as gas heat recovery.Gas heat recovery could reduce the required solar energy by 22-99% and solid heat recovery could reduce the required solar energy by 0.1-26%.
These thermodynamic analyses have demonstrated the need for heat recovery in CaO-looping based CO 2 capture systems to reduce the energy requirements and thus the overall cost of using such systems on an industrial scale.Gas heat recovery has a larger potential to reduce energy requirements, but non-negligible reduction can also be obtained by implementing solid heat recovery at higher CO 2 concentrations in the input gas stream such as that produced by a combustion-based power plant.

Calcination Kinetics
The kinetics of the calcination reaction have been extensively studied.Borgwardt (1985) studied the kinetics of the calcination of two different limestone types with particle diameters ranging from 1 to 90 µm using a differential reactor.At 850°C, the reaction was first order with respect to CaCO 3 surface area with a rate constant of 1.6 × 10 -6 mol/cm 2 /s.The activation energy was 205 kJ/mol.The calcination reaction conversion as a function of time at a reaction temperature of 720°C is shown in Fig. 5. Dennis and Hayhurst (1987) found the rate constant to be 1.0 × 10 -6 mol/cm 2 /s and the activation energy to be 169 kJ/mol at the same temperature, but noted that the rate was a function of both CO 2 partial pressure and total pressure.(Borgwardt, 1985).Reproduced with permission from John Wiley and Sons.
They also found that the reaction was chemically controlled and that increasing the total pressure increased the reaction time regardless of CO 2 partial pressure.Escardino et al. (2008) summarized the results of several studies from 1930-1974, noting that the reported reaction order varied from 0 to 1 and the reported activation energy varied from 147 to 397 kJ/mol.Their own study resulted in an activation energy of 175 kJ/mol and a reaction order of 1/3.Garcia-Labiano et al. (2002) studied the calcination reaction using several sources of limestone.They found that the preexponential factor of the Arrhenius equation ranged from 29.5 to 6.7 × 10 6 mol/cm 2 /s and that the activation energy ranged from 114 to 166 kJ/mol depending on the sorbent source.Acharya et al. (2012) studied the calcination reaction in atmospheres of N 2 , steam, and CO 2 .They found that the pre-exponential factor ranged from 2.12-4.82× 10 10 s -1 and that activation energy ranged from 180 to 257 kJ/mol.
The N 2 atmosphere had the highest pre-exponential factor and activation energy and the CO 2 atmosphere had the lowest.A summary of rate equations for the calcination reaction is shown in Table 1.

Carbonation Kinetics
The kinetics of the carbonation reaction have also been widely studied.The reaction is generally considered to be chemically controlled initially and switches to a diffusion controlled regime due to formation of a CaCO 3 product layer.Bhatia and Perlmutter (1983) found the reaction rate constant to be an average of 0.0595 cm 4 /mol/s for temperatures ranging from 550 to 725°C and gas atmospheres containing 2 to 10% CO 2 .Shimizu et al. (1999) found the reaction rate to be first order with respect to CO 2 concentration, but the maximum conversion decreased as the CaO was cycled.The reaction rate constant was unaffected by cycling and had an average value of 25 m 3 /kmol/s for reaction atmospheres containing 5 to 15% CO 2 .Fang et al. (2009) found the rate constant to be 2.1 × 10 -3 m 3 /mol/s for the chemically controlled initial reaction and 2.5 × 10 -3 m 3 /mol/s for the diffusion controlled regime for a reaction atmosphere of 20% CO 2 .A summary of rate equations for the carbonation reaction is shown in Table 2. Nikulshina et al. (2007) conducted a thermogravimetric (TG) analysis of the carbonation of CaO and Ca(OH) 2 using atmospheric levels of CO 2 .The CaO carbonation reaction rate was chemically controlled initially and switched to a diffusion controlled regime after 20 minutes due to formation of a CaCO 3 product layer.The CaO reaction n CaO,0 , is shown in Fig. 6(a).The final reaction extent at 100 minutes ranged from 0.008 at 300°C to 0.05 at 450°C.Above 450°C, the calcination reaction began to become favorable, so the final reaction extent decreased above this temperature.The reaction rate of Ca(OH) 2 carbonation did not switch regimes with time, instead demonstrating diffusion controlled kinetics below 350°C and chemically controlled kinetics above 350°C.The Ca(OH) 2 reaction extent as a function of time is shown in Fig. 6(b).The Ca(OH) 2 had a higher final reaction extent of 0.25 after 100 minutes at 425°C.The experimental data for CaO was fitted using an unreacted core model and the data for Ca(OH) 2 was fitted using a kinetic model accounting for the formation of an interface of water or OH -ions.Numerical modeling results are shown in Figs.6(a) and 6(b) with solid lines.

Sorbent Degradation
Degradation of the CaO sorbent over the course of multiple cycles has been identified as a limiting factor in the use of the cycle for CO 2 capture.Abanades (2002) proposed a model for carbonation conversion over a number of cycles: where X c,N is the carbonation conversion at the end of the chemically controlled portion of the carbonation reaction, N is the cycle number, and f = 0.782 and b = 0.174 are constants obtained from a curve fit of several experimental studies by other researchers.The results of this curve fit compared to experimental data are shown in Fig. 7.A wide variety of synthetic sorbents designed to reduce degradation over multiple cycles have been studied recently, including calcium aluminate supported CaO (Manovic and Anthony, .0595 cm 4 /mol/s Bhatia and Perlmutter (1983) 2 2 , max (1 ) ( )   4) (Abanades, 2002).Reproduced with permission from Elsevier.
2009), CaO with a mayenite (Ca 12 Al 14 O 33 ) binder (Li et al., 2006), CuO/CaO composite (Manovic et al., 2011), CaO calcined from mesoporous CaCO 3 (Iyer et al., 2004), thermal pretreatment of CaO (Can Ozcan et al., 2011), and CaO prepared with various precursors (Lu et al., 2006).This summary of the kinetic analyses demonstrates that the kinetics of the calcination and carbonation reactions vary widely depending on the sorbent source and gas atmosphere composition.Overcoming the degradation of the sorbent is an important challenge addressed by on-going research in the field.A careful choice of sorbent is necessary to ensure fast reactions and an economical process without a need for frequent sorbent replacement.

HEAT AND MASS TRANSFER ANALYSES
In addition to chemical kinetics, interparticle physical processes of interest include mass transfer of gaseous reactants through the pore network, heat transfer to the endothermic reactions or from the exothermic reactions, and changing particle morphology.Global reaction rates and total particle conversion are influenced by local chemical kinetics, local reactant concentration, and local total pressure and temperatures.Numerical models have been developed to investigate mass and heat transfer limitations on reaction rates and total reaction extent at the particle level.

Calcination Modeling
Mass transfer in particles undergoing calcination has been investigated by Silcox et al. (1989) and García-Labiano et al. (2002).Diffusion through particle pores is the predominant mode of mass transfer in particles undergoing calcination.The local, transient concentration of carbon dioxide was predicted using the following one-dimensional (1D) conservation equation: Eq. ( 5) has been utilized in models with ψ representing partial pressure, concentration, or density.It has been employed in several analyses of both the calcination and carbonation reactions (Silcox et al., 1989;García-Labiano et al., 2002;Stendardo and Foscolo, 2009;Khoshandam et al., 2010;Ebner andLipiński, 2011, 2012;Yu et al., 2012).The source term S   was defined in terms of the chemical reaction rates.Effective diffusivity of carbon dioxide in a porous media has been treated in the above models as a combination of bulk diffusivity and Knudson diffusivity and scaled by particle porosity.Boundary conditions were defined by spherical symmetry at the particle center (zero mass flux) and external mass transfer at the surface.García-Labiano et al. (2002) investigated different rate expressions and determined kinetic parameters by fitting the changing grain size and shrinking core chemical reaction models to experimental data for the calcination of three different calcium based sorbents.They also investigated the effect of total pressure and postulated that total pressure influences the reaction rate because of its influence on effective gas diffusivity.
Models for investigating heat transfer by conduction and radiation within reacting particles have also been developed.Heat transfer properties related to conduction and radiation at the particle surface and within the particle volume vary greatly with temperature and the morphological changes in porosity and pore structure that accompany chemical reaction.A numerical model for investigating semi-transparent particles with spectral radiative properties in high-temperature applications was developed by Dombrovsky and Lipiński (2007).The model was applied to the transient heating of calcium carbonate and zinc oxide particles by concentrated solar radiation and predicts considerable temperature gradients within the particles as they are heated.Lipiński and Steinfeld (2004a) first investigated a fixed bed of reacting particles.They numerically simulated the coupled radiative heat transfer and chemical reaction in a non-gray, non-isothermal, fixed bed of CaCO 3 particles undergoing calcination and found good agreement between model results and experimental data.The unsteady energy equation was solved inside the fixed bed by finite volume techniques.The incident solar radiation distribution was established with the Monte Carlo ray-tracing method.The radiative heat transfer through the bed was modeled with the Rosseland diffusion approximation.A contracting geometry rate law with an Arrhenius temperature dependence of the rate constant was employed.The heat transfer model was validated via comparison of computed and experimentally measured variations of temperatures and reaction extents.Lipiński and Steinfeld (2004b) investigated a slab of suspended, reacting particles.They used a similar formulation for chemical kinetics as Lipiński and Steinfeld (2004a) to numerically simulate radiative heat transfer and chemical reaction in an absorbing, emitting, and Mie-scattering suspension of CaCO 3 particles undergoing calcination.They considered both shrinking and non-shrinking particles and showed radiation absorption was less efficient for shrinking particles.As a result of the less efficient absorption, the shrinking particle reaction rate was slower than the nonshrinking particle reaction rate.In the shrinking particle case, radiative properties varied as particle size changed and the absorption coefficient, scattering coefficient, and scattering phase function had to be calculated at each time step.
A transient model incorporating heat transfer, mass transfer, and chemical kinetics has been analyzed by Ebner and Lipiński (2011) (Fig. 8).Eq. ( 5) was solved along with the energy conservation equation formulated as Two methods for modeling radiative heat transfer were considered: Rosseland diffusion approximation and Monte Carlo ray tracing method.Ebner and Lipiński (2011) found that including radiation significantly increased the rate of particle heating and subsequently decreased the time to complete particle calcination.Temperatures predicted by the Rosseland diffusion approximation were in good agreement with those predicted by the Monte Carlo ray tracing method.The effect of the grain size of calcium oxide strongly influence the optical behavior of the particle, but was investigated and found to only weakly influence temperature, reaction extent, and total reaction time.
Ebner and Lipiński (2012) used the same formulation for heat and mass transfer to investigate two methods for modeling the calcination rate and resulting particle: changing grain size model and shrinking core model.The two approaches were compared and conditions under which the reaction was heat transfer limited or mass transfer limited were identified.

Carbonation Modeling
Mass transfer in particles undergoing carbonation has been approached in a similar manner as in particles undergoing calcination.Stendardo and Foscolo (2009) and Khoshandam et al. (2010) developed models coupling mass transfer and carbonation chemical kinetics.In carbonating particles, as the product layer of calcium carbonate develops, pores shrink and a large fraction of them eventually close, a product layer shell is formed, and mass transfer is restricted.As pore diffusion is impeded, diffusion through the product layer of calcium carbonate becomes a significant Fig. 8. Semitransparent particle undergoing heterogeneous decomposition under high flux irradiation (Ebner and Lipinski, 2011).Reproduced with permission from Elsevier.
factor.Both phenomena were carefully considered in the model developed by Stendardo and Foscolo (2009).They were able to predict total reaction extent for cases in which complete conversion could not be experimentally reached due to mass transfer limitations through the developed product layer.
A complete model solving equations for conservation of mass, Eq. ( 5), conservation of energy, Eq. ( 6) with   q limited to conduction, and chemical kinetics for carbonation was numerically analyzed for a particular type of sorbent particle by Yu et al. (2012).They did not consider pure or relatively pure calcium oxide sorbent samples but analyzed a special sorbent particle made of very small calcium oxide particles suspended in a porous spherical supporting material.Good correlation was observed between the model and experimental data.

SOLAR REACTOR ENGINEERING
Early experimental work has focused on the calcination of CaCO 3 for the solar production of lime as a commodity or as an energy storage material.Several reactor types have been used for this process, including fluidized bed (Badie et al., 1980;Flamant et al., 1980), rotary kiln (Badie et al., 1980;Flamant et al., 1980;Meier et al., 2003;Meier et al., 2004;Meier and Cella, 2004;Meier et al., 2005;Meier et al., 2006), and cyclone or vortex flow reactors (Imhof et al., 1991;Nikulshina et al., 2009a;Steinfeld et al., 1992).CaCO 3 was fed to the reactors in the form of particles with sizes ranging from a few µm to a few mm.The reactors were subjected to 1-5 kW of concentrated solar radiation input, provided by a solar furnace, consisting of a sun-tracking heliostat and a stationary parabolic concentrator, or by a high-flux solar simulator, consisting of highpressure Ar arc lamps close-coupled to elliptical reflectors.Experimentally measured quantities include temperatures, measured with thermocouples or with a pyrometer, reaction extents, determined via X-ray diffraction or by complete calcination of product samples in a TG, and radiative power input to the reactor, determined from calibrated gray-scale images of a Lambertian (diffusely reflecting) target at the focus of the concentrating system.Flamant et al. (1980) and Badie et al. (1980) conducted the calcination of CaCO 3 particles in fluidized bed and rotary kiln reactors (Fig. 9).The fluidized bed reactor was comprised of a vertical silica tube (5) with a gas distributor (3) consisting of inert (metallic or ceramic; 0.6-0.8mmdia.)beads lying on a metallic grid.The lower part of the tube was surrounded by a cylindrical reflector (9) to reduce reflection and reradiation losses.The solar absorption and consequently the solar energy conversion efficiency of the reactor was enhanced by doping the CaCO 3 particles with 1 weight-% of graphite.Complete decomposition of a batch of 10 g of particles was accomplished after 6-8 min of operation at around 850°C bulk temperature, resulting in an energy conversion efficiency (ratio of sensible energy plus reaction enthalpy to incident solar energy) of 10-15%.The rotary kiln consisted of a rotating water-cooled cylindrical metallic shell (7), with its inner wall lined with a refractive (a) (b) Fig. 9. Fluidized bed and rotary kiln reactors tested by (Badie et al., 1980).Reproduced with permission from Elsevier.
tube (1) (alumina or lanthanum chromite).Particles were fed through a vibrating tube in the back of the reactor (2) and were transported to the front due to the rotation of the 5° inclined cavity.For the rotary kiln reactor in continuous operation, optimum efficiency of 7% was achieved at 30% conversion with a feed rate of 0.077 g/s and a rotation speed of 8 rpm; the highest reported conversion was 50%.While nearly uniform temperature distribution was measured over a large portion of the fluidized bed, temperature variations of more than 1000°C were observed in the rotary kiln.Salman and Khraishi (1988) partially calcined a CaCO 3 pellet contained in a graphite tube, surrounded by a stainless steel box with a windowed opening.Conversions ranging from 17% at 1165°C to 65% at 1446°C were measured.Nonuniform heating of the pellet was identified to be the reason for the lower reaction rates as compared with experiments conducted in electric furnaces, and the use of a fluidized bed reactor was suggested to improve the heat and mass transfer.Imhof (1991), Imhof et al. (1991) and Steinfeld et al. (1992) developed and tested cyclone reactors.The cyclone reactor tested by Steinfeld et al. (1992) consisted of a 30 cm-long truncated steel cone (10° opening angle), insulated Reich et al., Aerosol and Air Quality Research, 14: 500-514, 2014 509 with 12 cm-thick alumina-silica insulation.Two concentric conical plates with a 6-cm open aperture covered the larger end section of the cone.A tangential air flow with suspended CaCO 3 particles entered the reactor beneath the lower conical cover plate.Due to the swirl of the flow, the particles separated from the flow at the wall and exited through the smaller end section of the cone.The reactor was operated at 1200-1450 K wall temperature, and degrees of particle conversion in the range 53 to 100% were measured.Meier et al. (2003Meier et al. ( , 2004)), Meier and Cella (2004) and Meier et al. (2005aMeier et al. ( , b, 2006) ) developed and tested 10 kW rotary kiln reactors for the calcination of 1-5 mm large CaCO 3 particles.Meier et al. (2004) developed a horizontal cylindrical cavity reactor, consisting of a conical reaction chamber, made from refractory concrete.The reactor was insulated with fibrous ceramic material encapsulated in a stainless steel shell.The entire reactor sat on four rubber wheels which were driven by an electric motor to rotate the cavity.Solar radiation entered through a 8 cm-dia.circular aperture in the reactor's water cooled aluminum front.The CaCO 3 particles (98% purity) were fed from a hopper to the back of the reactor.Due to the rotation of the conical reaction chamber, the particles were transported through the reactor, where they were directly exposed to solar radiation.The calcined particles were collected in a bin below the reactor front.The reactor was tested on sun for a total of 100 h.Tests were run for 30 min at steady-state with a constant particle flow rate.Degrees of calcination > 98% and efficiencies up to 20% were determined experimentally.The efficiency was defined as the ratio of the enthalpy of the produced CaO to the solar power input to the reactor: For a typical test run with η = 13%, energy losses were due to natural convection from the open aperture and losses via unrecovered products (50% of the solar power input), reradiation losses (17%), sensible heat of products (10%), and conduction through insulation (10%).Meier et al. (2003Meier et al. ( , 2005bMeier et al. ( , 2006) ) developed an indirectly heated rotary kiln reactor, consisting of 16 SiC absorber tubes, which were contained in a tilted rotating cylindrical solar cavity with a circular aperture (Fig. 10).1-5 mm large limestone particles were fed from the back of the reactor through a preheating chamber into the tubes, and the products were collected with a funnel at the front of the cavity.Experimentally determined efficiencies, calculated according to Eq. ( 7), reached 32%.The energy balance of the reactor accounted for the solar power input to the reactor, the heat rate to heat the reactants, the energy input to the dissociation reaction of CaCO 3 , reradiation and conduction heat losses, a lumped heat loss term that accounted for unknown heat losses such as convection heat losses through the aperture and heat losses via unrecovered products, and the heat stored in the reactor during transients: The relative magnitude of the energy balance terms are shown in Fig. 11 for experiments with various parameter settings.The design and principal dimensions of scale-up reactors with solar power input of 0.5 to 20 MW were described.
A 10-kW th solar-heated rotary kiln for multiple applications in solar thermochemical processing was developed, tested and modeled at the German Aerospace Center (DLR) (Neises et al., 2012;Tescari et al., 2013).While it was experimentally demonstrated for thermal reduction of , for on-sun experiments with various parameter settings: rotational speed (8-18 rpm), particle feed rate (36-136 g/min), limestone grain size (2-3 mm), and temperature (1200-1400 K) (Meier et al., 2006).Reproduced with permission from Elsevier.metal oxides for thermochemical energy storage, this reactor concept, due to the possibility of implementing a moving particle bed in a rotary kiln, can be considered for solardriven calcination of limestone particles.Nikulshina et al. (2009b) conducted 5 consecutive carbonation-calcination cycles in a fluidized-bed solar reactor (Fig. 12).The reactor consisted of a silica tube (22 mm i.d., 260 mm length) containing 7 g of a mixture of 97 wt% inert silica particles (229 µm mean particle size) and 3 wt% CaO particles (9µm mean particle size), directly exposed to concentrated solar radiation.Carbonation was conducted during 600 s at 365-400°C in a flow of 2 L(STP)/min of synthetic air containing 500 ppm of CO 2 and 17% water vapor.Calcination was conducted during approx.600 s at 800-875°C in an Ar-flow of 1.75 L(STP)/min.The CO 2 evolution during the 5 cycles shown in Fig. 13 demonstrated that completely CO 2 -depleted air exited the reactor during the carbonation steps and that the calcination reaction reached completion after around 500 s.Sorbent deactivation was not observed after 5 cycles, which was attributed to the addition of steam during carbonation, and to the relatively low calcination temperature.A reduction of the particle size from 9 to 1.5 µm and consequently an increase of the BET surface area were observed, which was explained by the cracking of the particles due to the Ca(OH) 2 formation inside the particles.Single long-term (4000 s) carbonation experiments yielded higher reaction extent with CaO than with Ca(OH) 2 particles (71 vs. 61%).
The coproduction of syngas and lime by combined calcination of CaCO 3 particles and dry-reforming of methane in a 5 kW vortex flow reactor has been experimentally demonstrated by Nikulshina et al. (2009a).A flow of CH 4 , diluted in Ar and laden with CaCO 3 particles was injected behind the aperture of a cylindrical cavity reactor and exited through an opening in the back wall.Methane conversion (X CH 4 = 1 -n CH 4 ,out /n CH 4 ,in ) was determined via measured inlet and outlet flow rates and gas compositions; CaCO 3 conversions (X CaCO 3 = n CaO /(n CaO 3 + n CaO ) was determined from TG measurements during the complete calcination of solid product samples.Highest CH 4 and CaCO 3 conversions were X CH 4 = 0.38 and X CaCO 3 = 0.83, respectively, at a reactor temperature of 1223 K, and the energy conversion efficiency was 7-10%.

SUMMARY
Carbon dioxide is an important commodity in multiple state-of-the-art industries.The demand for pure CO 2 may substantially increase in mid-to long-term as it is considered, beside water, as an input for solar-driven hydrocarbon fuel production.Carbon dioxide capture technologies are one option to provide carbon dioxide for these applications which simultaneously address the challenge of climate change mitigation.Carbon dioxide capture methods can be divided into the categories of absorption, adsorption, membrane separation, and cryogenic separation.One absorption method that has been investigated is the CaO-based carbonationcalcination cycle.Thermodynamic analyses have shown that the cycle is well suited to separation applications with high CO 2 concentrations, such as power plant flue gases.Heat recovery will play an important role in reducing the energy required for the process.Kinetic analyses have shown that the calcination reaction is relatively fast, but the carbonation reaction speed is limited by the switch to a Fig. 12. Schematic of the fluidized bed solar reactor setup tested by (Nikulshina et al., 2009) and (Nikulshina and Steinfeld, 2009).Reproduced with permission from Elsevier.Fig. 13.CO 2 evolution during carbonation-calcination cycling (Nikulshina et al., 2009).Reproduced with permission from Elsevier.
diffusion controlled regime due to the formation of a CaCO 3 product layer on the outside of the particles.Sorbent degradation over multiple cycles is also a known problem, but many potential solutions have been proposed for reducing it.Heat and mass transfer modeling at the particle level has shown good agreement with experimental data.Several solar reactors with input power of up to 10 kW to accomplish the calcination step of the process have been investigated, reaching efficiencies of up to 32%.Advancements towards more efficient and cost effective solar-driven CO 2 capture via thermochemical looping include development of new sorbents that resist degradation and development of solar reactors that accomplish the full cycle at high efficiencies and CO 2 capture rates.

Fig. 3 .
Fig. 3. Cycle for CO 2 capture from air including a slaker.Solid lines indicate a CO 2 capture only cycle and dotted lines indicate a cycle that co-produces hydrogen(Nikulshina et al., 2006).Reproduced with permission from Elsevier.

Fig. 5 .
Fig. 5. Calcination conversion as a function of time at 720°C(Borgwardt, 1985).Reproduced with permission from John Wiley and Sons.

Fig. 6 .
Fig. 6.Carbonation reaction extent as a function of time for CaO (a) and Ca(OH) 2 (b).Data points indicate experimental measurements and solid lines indicate numerical modeling results(Nikulshina et al., 2007).Reproduced with permission from Elsevier.

Table 1 .
Summary of rate equations for calcination.

Table 2 .
Summary of rate equations for carbonation.