Calcination Reaction Kinetics of a Limestone Sorbent in low CO2 Partial Pressures Using TGA Experiments

Available online at www.sciencedirect.com

ScienceDirect Energy Procedia 114 (2017) 259 – 270

13th International Conference on Greenhouse Gas Control Technologies, GHGT-13, 14-18 November 2016, Lausanne, Switzerland

Determination of carbonation/calcination reaction kinetics of a limestone sorbent in low CO2 partial pressures using TGA experiments Mohammad Ramezani, Priscilla Tremain, Elham Doroodchi and Behdad Moghtaderi* The Priority Research Centre for Frontier Energy Technologies & Utilisation, Chemical Engineering, School of Engineering, Faculty of Engineering and Built Environment, The University of Newcastle, Callaghan, New South Wales 2308, Australia

Abstract The carbonation/calcination reaction cycle of a limestone sorbent was used in a novel greenhouse calcium looping process to provide the heat and CO2 requirement of greenhouses. The unique conditions of this process (low temperature and low CO2 partial pressure) led to the determination of the carbonation/calcination reaction kinetic parameters in a temperature range of 400 to 800 oC and CO2 partial pressures of 0.04 to 0.16% (400 to 1600 ppm) which were carried out experimentally via a thermogravimetric analyzer (TGA). Various gas-solid reaction mechanisms were considered to determine the best reaction mechanism for the carbonation reaction. The diffusion function or G(x) = x2 had the best least-square linear fit, which resulted in a first order reaction for the carbonation reaction in the greenhouse calcium looping process. Moreover, the activation energy and preexponential factor of the carbonation reaction were established to be 19.7 kJ mol-1 and 295.8 min-1 kPa-1. The same methodology for reaction mechanism determination was carried out for the calcination reaction and G(x)=1-(1-x)1/3 was found to have the best linear fit. The activation energy and pre-exponential factor determined for the calcination reaction were 103.6 kJ mol-1 and 6.9 × 106 min-1, respectively. © 2017 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license © 2017 The Authors. Published by Elsevier Ltd. (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of GHGT-13. Peer-review under responsibility of the organizing committee of GHGT-13. Keywords: Type your keywords here, separated by semicolons ;

* Corresponding author. Tel.: +61-2 -4033 9062; fax: +61-2-4033 9095 E-mail address: [email protected]

1876-6102 © 2017 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of GHGT-13. doi:10.1016/j.egypro.2017.03.1168

260

Mohammad Ramezani et al. / Energy Procedia 114 (2017) 259 – 270

1. Introduction Greenhouses generally need CO2 during the day when photosynthesis reactions occur. At the same time due to warmer conditions because of sunlight, no heat is usually required. However, in contrast to the day time, during the night, vegetable plants in greenhouses release CO2. Heat is also required to keep the greenhouses warmer during the night. Generally, the CO2 requirements during the day and the heat requirements during the night are sufficed by combusting natural gas or biomass or alternatively by using solar systems. However, such arrangements are expected to increase both operating cost as well as CO2 emissions. A solution to this, is to implement the novel greenhouse calcium looping (GCL) process as proposed in our previous publication [1] which is based on cyclic carbonation/calcination reactions of a limestone sorbent. Table 1. Summary of kinetic studies of carbonation reaction reviewed. Investigator

System

Bhatia and Perlmutter [2]

Commercial limestone (98% CaCO3) in TGA Synthesised limestones in TGA

Gupta and Fan [3]

CO2 partial pressure (%)

Temperature (oC)

Order of reaction

Activation energy (kJ mol-1)

Pre-exponential factor

2-42

400-725

First order

88.9±3.7

3.576 ± 0.108 × 10-5 (m4 kgmol-1 min-1)

100

550-650

First order

72.7

1.16 × 104

550-850

First order (CO2 < 10%) Zero order (CO2 > 10%)

Sun et al. [4]

Natural limestone and dolomite in TGA and PTGA

6-100

Grasa et al. [5]

Natural limestones in TGA

10-100

Yu et al. [6]

Synthesised and natural limestones in magnetic suspension balance (MSB)

10- 30

Dolomite in TGA

Sedghkerdar et al. [8]

Natural and synthesised limestones in TGA

Yin et al. [9]

Pure CaO and synthesised limestone in TGA

50

Natural limestone in pressurized thermogravimetric analyser (PTGA)

100

Grasa et al. [11]

First order

700

First order (CO2 partial pressure < 100 kPa) Zero order (CO2 partial pressure > 100 kPa)

29±4

20-50

550-725

0.544 ± 0.015 × 10-5 (m4 kmol-1 s-1)

Nil

Nil

Nil

Nil

Zero order -25.71 (T>650)

Natural limestone in TGA

28.3 (Natural) 50

5-10

500-650

550-800

975-1025

700-800

Zero order

1.67 × 10-4 (mol m-2 s-1 kPa-1) 1.67 × 10-3 (mol m-2 s-1)

20.3±1

51.77 (T<650)

Mostafavi et al. [7]

Butler et al. [10]

550-750

29±4

7.51 × 10-3 (T<650) (mol m-2 s-1) 2.45 × 10-7 (T>650) (mol m-2 s-1) 2.22 × 10-4 (Natural) (mol m-2 s-1)

21.6 (Synthesised)

2.06 × 10-4 (Synthesised) (mol m-2 s-1)

43.68 (CaO)

0.33 (CaO) (mol m-2 s-1)

28.28-43.49 (Synthesised)

0.76-8.6 × 10-3 (mol m-2 s-1)

Zero order

Nil

Nil

First order

230 (according to the equation related to the concentration of CO2 (kmol m-3)) 200 (according to the equation related to the volume fraction of CO2)

Zero order

10.5 × 103 (m4 kmol-1 s-1)

28 × 106 (s-1)

It was concluded in our previous publication [1] that the GCL process can reduce energy usage by up to 70% in

261

Mohammad Ramezani et al. / Energy Procedia 114 (2017) 259 – 270

comparison to a conventional burner system in addition to producing near zero CO2 emissions. These results were derived through Aspen Plus simulations considering the carbonator and calciner as RGibbs reactors. The RGibbs unit operation simulated the reaction on the basis of Gibbs free energy minimization and predicts the thermodynamic equilibrium of reactants and products at given process conditions, without needing the reactions’ stoichiometry kinetics. To establish the required size of the carbonator/calciner in the GCL process, an RPlug reactor can be used instead of an RGibbs reactor. The RPlug unit works on the basis of an ideal plug flow reactor which assumes perfect radial mixing and no mixing along length of the reactor [12]. To simulate the process with an RPlug block, the stoichiometry and kinetics of the reaction are required. Table 2. Summary of kinetics studies of calcination reaction reviewed. Investigator System CO2 partial Temperature pressure (%) (oC) 0% 866 Barker [13] Commercial limestone (99.5 % CaCO3) in a TGA Guler et al. Natural limestones in a 0% 650-770 [14] TGA Borgwardt Natural limestones in a 0% 515-1000 [15] differential reactor Natural limestones in a Dennis and 20 % 800-975 fluidised bed reactor Hayhurst [16] Calvo et al. Commercial limestones 0% 860-920 [17] (99.5 % CaCO3) Rao [18] Ar and Dogu [19] Garcia et al [20]

Gonzalez et al. [21]

Commercial limestone in a TGA Natural limestone in a TGA Natural limestones and dolomite particles in a TGA and a thermobalance for pressurized experiments

Martinez et al. [22]

A commercial limestone (99.5 % CaCO3) in a TGA Natural limestones in a TGA

Yin et al. [9]

Pure CaO and synthesised limestones in a TGA

Order of reaction Nil

Activation energy (kJ mol-1) 147

Pre-exponential factor

First order Zero order First order

203.8

1.1×108 (s-1) 1.6×10-5 (kmol m-2 s-1) 1.0×10-5 (kmol m-2 s-1)

201-205 169

Zero order

193.8 110.5

Nil

1.9×109 1.3×106 (min-1) 2.2×102 - 3.3×104 (min-1) 3.65×107 - 14.3×107 (kmol m-2 s-1) Blanca limestone: 6.7×106 Mequinenza limestone: 2.5×102 dolomite: 29.5 (mol m-2 s-1) 5.3×104 (min-1)

0%

680-875

0.67

109.8-175.5

0%

900

155.6-212.5

0-80 % with increment of 5%

775-900

Zero order Nil

10 %

950-1150

Nil

205

0-100 % with increment of 25 % 0%

825-920

First order

91.7 112.4

2.5×105 2.0×106 (m3 kmol-1 s-1)

550-800

Zero order

173.7 (CaO) 148.5-159.5 (Synthesised)

2.1×103 (CaO) 2.0-6.5 (mol m-2 s-1)

Blanca limestone: 166 Mequinenza limestone: 131 dolomite: 114

A detailed literature review has been performed covering the carbonation/calcination reaction kinetics of of calcium rich compounds at different CO2 partial pressures (Please refer to Table 1 and 2). Although a broad number of studies have focused on deriving carbonation/calcination kinetics, there is a knowledge gap when considering the carbonation/calcination reactions in the unique operational conditions pertinent to the GCL process (low temperatures and low CO2 partial pressures) and taking into account different kinetic models for the carbonation/calcination reactions. Therefore, the main objectives of this study are as follows: x Establish the kinetic parameters of the carbonation/calcination reactions at the unique operating conditions of the GCL process (i.e. low temperature range of 400-500 oC (carbonation) and 600-800 oC (calcination), and low CO2 partial pressures of 400-1600 ppm). x Derive suitable kinetic models for the carbonation/calcination reaction process.

262

Mohammad Ramezani et al. / Energy Procedia 114 (2017) 259 – 270

2. Experimental methods A series of experiments utilizing a thermogravimetric analyzer (TGA) were conducted in order to establish kinetic data for the GCL process. 5 mg of raw Omya limestone sorbent (75-150 μm), provided by a local supplier, was inserted into a platinum crucible of a Q50 TGA. Elemental analysis, carried out in X-ray fluorescence (XRF) apparatus (Spectro X-lab 2000), of the limestone is presented in Table 3. To determine kinetic parameters of the carbonation reaction, the sample was initially heated to 700 oC at a ramp rate of 20 oC min-1, in a N2 environment, and held isothermally for 10 minutes to ensure the limestone sorbent was completely converted to CaO. To start the carbonation reaction, industrial grade CO2 diluted in air at a concentration of 2000 ppm, supplied by Coregas Co., was introduced to the TGA by a mass flow controller (MFC). The total flow rate of gases was set at 200 ml min-1 at atmospheric pressure inside the TGA. Experiments were conducted at a high flow rate of 200 ml min-1, to minimize external mass transfer resistance due to the formation of gas films around the particles. The carbonation reaction experiments were performed in a temperature range of 400 to 500 oC and CO2 partial pressures of 0.05 to 0.1% (500 to 1000 ppm). When the carbonation reaction was completed (i.e. there was insignificant changes to sample’s weight due to the carbonation reaction), the gases inside the TGA (air and CO2) were switched to 100 % CO2 and the temperature was increased to a temperature appropriate for calcination reaction (e.g. 600-800 oC). The reason for introducing pure CO2 to the TGA is to stop the incidence of calcination reaction before the TGA’s temperature reached the desired calcination temperature. Once the desired temperature was reached, the appropriate CO2 partial pressure was provided to the TGA (e.g. 400-1600 ppm). All experiments were repeated 3 times to obtain reliable data. The conversion of CaO particles via the carbonation reaction was calculated on a molar basis using Eq. (1):

x

mCaO t  mCaO 0 44 mCaO 0 56

Eq (1)

where x is the CaO conversion, mCaO (t) is the weight of the sample at time t and mCaO (0) is the initial sample weight. 44 and 56 are the molecular weights of CO2 and CaO, respectively. For the determination of reaction kinetics, firstly, the general equation for a gas-solid reaction is a function of equilibrium gas pressure ( PCO ,eq ), temperature and the reaction mechanism as follows [23]: 2

dx dt



kf  x PCO  PCO 2



n

2

, eq

Eq (2)

where t is time, n is the reaction order, f(x) represents a function for the reaction mechanism and k is the reaction rate constant following the Arrhenius-type equation: k

§E· k 0 exp¨ ¸ © RT ¹

Eq (3)

where k0 is the pre-exponential factor, E is the activation energy, R is the gas constant and T is the temperature. Table 3. XRF analysis of Omya limestone. Ca

Fe

Mg

Al

Si

Mn

K

97.56

0.23

0.43

0.15

1.21

0.38

0.04

The CO2 equilibrium partial pressure is dependent on the carbonation/calcination reaction temperature according to the correlation suggested by Baker [24]:

263

Mohammad Ramezani et al. / Energy Procedia 114 (2017) 259 – 270

log PCO2 ,eq >kPa @ 9.079 

8307 .83 T >k @

Eq (4)

Integrating Equation (2) gives G(x) of which various forms were postulated to show different reaction mechanisms: x

Gx

t

1

³ f x dx ³ k P 0



n

CO2

 PCO2 ,eq dt

Eq (5)

0





n

k PCO2  PCO2 ,eq t

In the case of the calcination reaction, the conversion of CaCO3 particles via the calcination reaction was calculated using Equation (6): x

mCaCO  mCaCO t 3

3

Eq (6)

mCaCO  mCaO 3

where x is the CaCO3 conversion, mCaCO3 (t) is the weight of the sample at time t, mCaCO3 is the initial weight of CaCO3 just before the start of calcination reaction and mCaO is the sample weight at the end of calcination reaction. Table 4. G(x) equation for different reaction mechanisms a. Gx Reaction mechanism D1

x2

D2

1 x ln1 x  x

D3

ª1  1  x 13 º «¬ »¼

D4

2 2x  1  x 3 3

C1

 ln1  x

C2

1  x 1  1

A2

> ln1  x @2 1 > ln1  x @3 1 1  1  x 2

A3 R2

a

1

2

1

1

R3

1  1  x 3

P1

x

P2

x2

P3

x3

P4

x4

1

1

1

D1 to D4: diffusional function, C1 and C2: first and second order chemical reaction, A2 and A3: Avrami-Erofe’ev random nucleation and

subsequent growth function, R2 and R3: phase boundary reaction, P1 to P2: Mampel power law function.

Similarly to the carbonation reaction, for the determination of reaction kinetics, firstly we start with the general equation for a gas-solid reactions which is a function of temperature, reaction mechanism and equilibrium gas pressure ( PCO,eq) [23]: 2

dx dt



kf  x PCO

2

, eq

 PCO



n

2

Eq (7)

264

Mohammad Ramezani et al. / Energy Procedia 114 (2017) 259 – 270

where t is time, f(x) is a function for the reaction mechanism, n is the reaction order and k is the reaction rate constant based on the Arrhenius-type equations. The CO2 equilibrium partial pressure which is dependent on the carbonation/calcination reaction temperature can be calculated using an empirical equation postulated by Baker [24] (Equation (4)). G(x) function represents different reaction mechanisms resulted by integrating Equation (7): G x



x

t

1

³ f x dx ³ k PCO ,eq  PCO 0 0 2

k PCO ,eq  PCO 2

2

t

2

dt n

Eq (8)

n

Table 4 summarises G(x) functions for various reaction mechanisms which were used to determine the carbonation and calcination reaction mechanisms in this study.

3. Results and discussion Firstly, the results for the determination of the reaction mechanism of the carbonation reaction are described followed by the results of the calcination reaction under conditions pertinent to the GCL process.

3.1. Carbonation reaction Figure 1 shows the CaO conversion versus time, calculated by Eq. (1), over a temperature range of 400-500 oC and a CO2 partial pressure of 1000 ppm (0.1%). It can be seen that the reaction rate increased with increasing temperature from 400 oC to 450 oC, while after this threshold the reaction rate started to decrease as the temperature increased to 500 oC. This was attributed to the temperatures 470 oC and 500 oC occurring close to the equilibrium temperature at which the carbonation reaction occurs for CO2 partial pressures of 1000 ppm (0.1 %). Figure 1 also shows that the reaction progress became almost zero at around 35% CaO conversion for 470 and 500 oC, 32% for 430 and 450 oC and 26% for 400 oC. This is due to the formation of a carbonate layer on the surface of the CaO particles, hence hindering the access of CO2 to the active CaO sites of the particles [25].

Figure 1. The calcium oxide (CaO) conversion, on molar basis, versus time for different carbonation temperatures with the following conditions: CO2 partial pressure: 1000 ppm (0.1 %), total pressure: 1 atm, total flow rate: 100 ml min-1, sample size: 5 mg, particle size: 75-150 μm.

Mohammad Ramezani et al. / Energy Procedia 114 (2017) 259 – 270

k P  P



n

eq Figure 2. G (x) = x2 reaction mechanism versus time for determination of of the carbonation reaction at 450 oC and CO2 partial pressures of 500-1000 ppm. Symbols represent experimental data calculated by the D1 model and the continuous lines are the linear fits of the data which are time based on the linear least-square regression method.

The right hand term in Eq. (5) is temperature dependent and the most appropriate reaction mechanism can be determined by plotting G(x) functions, tabulated in Table 4, versus time for an isothermal experiment. For CO2 partial pressures of 100-1000 ppm and a temperature range of 400-500 oC, G(x) functions were plotted and the R2 value of the least-square linear fits for all reaction mechanisms were determined of which the highest value was considered to be the most appropriate reaction mechanism for the carbonation reaction. It is concluded that the most appropriate reaction mechanism is D1 or G(x) = x2. Figure 2 illustrates the variation of the G(x) = x2 reaction mechanism versus time, at a temperature of 450 oC, in which the slope of each curve determines the k P  Peq n term of the carbonation reaction (see Eq. (5)). As can be clearly seen, the carbonation reaction increases as the CO2 partial pressure increases from 500 to 1000 ppm, due to Le Châtelier's principle in which increasing the CO2 partial pressure shifts the equilibrium and promotes the carbonation reaction forward. A sigmoidal shape also can be seen in Figure 2, especially at 500 ppm. This is because of nucleation growth which occurs for particular gas-solid reactions at specific temperatures and gas phase compositions [25]. Nucleation is a dynamic process which initiates the carbonation reaction and the progress of this reaction depends on the rate of nuclei growth. Following this step, the chemical reaction plays a key role in making a linear slope which was considered the intrinsic reaction rate in this study [25]. As mentioned before, the slope of the least-square linear regression presented in Figure 2 determines the value of o o n k P  P eq at 450 C. These values were also determined, in the same manner, for 400, 430, 470 and 500 C, as presented in Figure 3 in the format of ln>k P  Peq n @ versus lnP  Peq to establish the order of reaction (n) and the reaction constant (k). Symbols represent the experimental data while the lines are the least-square linear fits in which the slope determines the order of reaction and the intercept of the linear fit indicates ln (k). It is worth mentioning that k and n are both temperature dependent in which the order of the reaction (n) increased from 1 to 1.6 for an increase in temperature from 400 oC to 500 oC. These results are in agreement with those of Sun et al. [4] and Grasa et al. [5] who reported a first order reaction for CO2 partial pressures up to 10 kPa and 100 kPa, respectively. Utilizing the ln (k) results obtained from Figure 3, Arrhenius plots were created as shown in Figure 4 for the temperature range of 400 to 500 oC and CO2 partial pressures of 500 to 1000 ppm. A least-square linear fit was then performed, where the slope of the fit corresponded to the activation energy, while the pre-exponential factor was the intercept of the linear fit. The Arrhenius’s factors are summarized in Table 5.

265

266

Mohammad Ramezani et al. / Energy Procedia 114 (2017) 259 – 270

>

Figure 3. ln k P  Peq

@ versus ln P  P in the range of 400-500 n

eq

o

C temperatures and 500-1000 ppm CO2 partial pressure.

Figure 4. Arrhenius plot for the carbonation reaction at 400-500 oC and CO2 partial pressure of 500-1000 ppm.

Table 5. The carbonation reaction kinetic parameters for CO2 partial pressures of 500-1000 ppm and the temperatures of 400-500 oC. Temperature (oC) Order of reaction (n) Activation energy (E) (kJ mol-1) Pre-exponential factor (k0) (min-1 kPa-1) 400-500 1 19.7 295.8

The activation energy determined was less than that of Sun et al. [4] who reported a value around 30 kJ mol-1, but this was higher than zero which was reported by Bhatia and Perlmutter [2]. The discrepancy in results between the current study and previous studies, is likely related to the differences in the limestone samples structure and impurities. For the CaO-CO2 reaction, the rate of reaction is not only dependent on the chemical energy, but also on

267

Mohammad Ramezani et al. / Energy Procedia 114 (2017) 259 – 270

the mechanical energy such as strain energy between different grains. During nucleation and solid product formation, a sorbents structure changes and the extent of changes may differ from one sorbent to another. Moreover, previous investigations carried out at higher CO2 partial pressures lead to a higher amount of sintering issues, increasing the required amount of activation energy. Sintering refers to a change in pore structure, pore shrinkage and grain growth that particles of CaO undergo during heating [26]. At high temperatures and high partial pressure of CO2 and steam, CaO sintering increases [15].

3.2. Calcination reaction Figure 5 illustrates the CaCO3 conversion versus time, determined by Eq. (6), for a CO2 partial pressure of 400 ppm (0.04 %) and temperature range of 600 to 800 oC. As expected, the calcination reaction rate increased with increasing temperature from 600 oC to 800 oC. According to Eq. (4), proposed by Baker [24], raising temperature causes an increase in the CO2 equilibrium partial pressure and similarly in the driving force of PCO ,eq  PCO



2

2



o

which resulted in a higher calcination reaction rate. At temperatures more than 600 C, the calcination reaction was completed in less than 12 minutes, whereas at 600 oC the reaction took 40 minutes to complete. This is due to the fact that 600 oC is close to the equilibrium temperature at which the calcination reaction occurs for a CO2 partial pressure of 400 ppm. It is worth mentioning that the longevity of the calcination reaction is favourable in the GCL process, as the conversion of CaCO3 inventory may provide a sufficient level of CO2 to the greenhouse. It is true that working at low temperatures, in which the reaction rate is comparatively slow, the design of larger reactors is required. However, the operational costs and the related issues such as sintering which occur at high temperatures would be decreased.

Figure 5. The calcium carbonate (CaCO3) conversion, versus time for different calcination temperatures with the following conditions: CO2 partial pressure: 400 ppm (0.04 %), total pressure: 1 atm, total flow rate: 200 ml min-1, sample size: 5 mg, particle size: 75-150 μm.

The k parameter in the right hand term of Eq. (7) depends on temperature and, therefore, plotting the G(x) functions, summarized in Table 4, versus time for an isothermal temperature is an approach to establish the most appropriate reaction mechanism. The highest R2 value represents the most appropriate G (x) function of which for the calcination reaction was R3 or G(x)=1-(1-x)1/3. According to Eq. (8), to establish the k P  Peq n term of calcination reaction, it is required to plot the G (x) function versus time and determine the slope of the curves. These functions are demonstrated in Figure 6 for a CO2 partial pressure of 400 ppm and covering a temperature range of 600-800 oC.

268

Mohammad Ramezani et al. / Energy Procedia 114 (2017) 259 – 270

Figure 6. G(x)=1-(1-x)1/3 reaction mechanism versus time for determination of k of the calcination reaction in a temperature range of 600 to 800 C and CO2 partial pressure of 400 ppm. Symbols represent experimental data calculated by the R3 model and the continuous lines are the linear fits of the data which are time based on the linear least-square regression method.

o

The values of k P  Peq n were also determined using the same method, for the other CO2 partial pressures ranging from 400 to 1600 ppm and are presented in Figure 7 and Figure 8, for temperatures of 700 oC and 800 oC, respectively. The slope of the curves in Figure 7 and Figure 8, showing ln>k Peq  P n @ versus ln Peq  P , can be used to establish the order of calcination reaction and the intercepts are the reaction constant (k). It can be clearly noticed that the calcination reaction follows a zero order reaction based on the CO2 partial pressure in the unique operational conditions of the GCL process (i.e. low temperature range of 600-800 oC and low CO2 partial pressure of less than 0.2 %). The independence of the calcination reaction rate to the CO2 partial pressure was also reported by other authors [15, 19].

Figure 7.

>

ln k Peq  P

n

@ versus lnP

eq

 P

for a temperature of 700 oC and a CO2 partial pressure range of 400-1600 ppm.

Mohammad Ramezani et al. / Energy Procedia 114 (2017) 259 – 270

Figure 8.

>

ln k Peq  P

n

@ versus ln P

eq

 P

269

for a temperature of 800 oC and a CO2 partial pressure range of 400-1600 ppm.

On the other hand, a first order reaction was reported by other researchers [22]. Abanades’s team [22] concluded that the calcination reaction follows a first order reaction under the operational conditions of a temperature range of 825-920 oC and CO2 partial pressure of 0 to 100 %. It is worth mentioning that the level of CO2 examined in their experiments was considerably more than in this study. With a CO2 concentration of 25 %, this could cause a significant film resistant layer which hinders CO2 mass transfer from the interface reaction to the bulk phase. In contrast, in our experiments the maximum CO2content was 0.2 % which causes a very thin layer on the product CaO and consequently an insignificant effect on the mass transfer resistance. Using the reaction constants (k) obtained from Figure 7 and Figure 8, the kinetic parameters of the calcination reaction over the unique operational conditions of the GCL process were calculated and the results are summarized in Table 6. The calculated activation energy of 103.6 kJ mol-1, was less than those reported in previous studies which concluded that the activation energy for the calcination reaction would be in a range of 147-212 kJ mol-1 [13, 15, 19, 21]. The discrepancy between those results and the activation energy determined in the current study might be related to differences in the porous structure of the samples and/or operational conditions of which the calcination reaction occurred (i.e. temperature of 850-1150 oC and CO2 partial pressure of 10-100 %). Table 6. The calcination’s kinetic parameters for CO2 partial pressures of 400-1600 ppm and the temperatures of 600-800 oC. Pre-exponential factor (k0) (min-1) Order of reaction (n) Activation energy (E) (kJ mol-1) 0.02 103.6 6.9 × 106

4. Conclusion Over the CO2 partial pressures of 0.04 to 0.16% (400 to 1600 ppm) and temperature range of 400 to 800 oC, the carbonation/calcination reaction kinetics of a limestone sorbent were determined using a TGA. A range of gas-solid reaction mechanisms were considered to predict the carbonation/calcination kinetics at the unique conditions of the GCL process. The best models were found to be a diffusion function or G(x) = x2 for the carbonation reaction and G(x) = 1-(1-x)1/3 for the calcination reaction, determined via least-square linear regression of experimental data. Interestingly, up to 450 oC the carbonation reaction rate increased with increasing temperature, while after this threshold the carbonation reaction rate decreased until 500 oC. Conversely, the carbonation reaction rate increased with an increase in CO2 partial pressure. In the case of the calcination reaction, with an increase in the temperature

270

Mohammad Ramezani et al. / Energy Procedia 114 (2017) 259 – 270

the calcination reaction increased while the calcination reaction followed a zero order reaction rate based on the CO2 partial pressure. The activation energy and pre-exponential factor for both carbonation and calcination reactions were determined as follows: x The carbonation reaction follows a first order reaction with the pre-exponential factor of 295.8 min-1 kPa-n and activation energy of 19.7 kJ mol-1. x The calcination reaction followed a zero order reaction rate with a pre-exponential factor of 6.9 × 106 min-1 and an activation energy of 103.6 kJ mol-1.

Acknowledgements The authors appreciate the financial support provided by the University of Newcastle, Australia.

References [1] M. Ramezani, K. Shah, E. Doroodchi, B. Moghtaderi. Application of a novel calcium looping process for production of heat and carbon dioxide enrichment of greenhouses. Energy Convers. Manage. 2015;103:129-138. [2] S. Bhatia, D. Perlmutter. Effect of the product layer on the kinetics of the CO2ϋlime reaction. AIChE J. 1983;29:79-86. [3] H. Gupta, L.-S. Fan. Carbonation-calcination cycle using high reactivity calcium oxide for carbon dioxide separation from flue gas. Ind. Eng. Chem. Res. 2002;41:4035-4042. [4] P. Sun, J.R. Grace, C.J. Lim, E.J. Anthony. Determination of intrinsic rate constants of the CaO–CO2 reaction. Chem. Eng. Sci. 2008;63:4756. [5] G. Grasa, R. Murillo, M. Alonso, J.C. Abanades. Application of the random pore model to the carbonation cyclic reaction. AIChE J. 2009;55:1246-1255. [6] F.-C. Yu, L.-S. Fan. Kinetic study of high-pressure carbonation reaction of calcium-based sorbents in the calcium looping process (CLP). Ind. Eng. Chem. Res. 2011;50:11528-11536. [7] E. Mostafavi, M.H. Sedghkerdar, N. Mahinpey. Thermodynamic and Kinetic Study of CO2 Capture with Calcium Based Sorbents: Experiments and Modeling. Ind. Eng. Chem. Res. 2013;52:4725-4733. [8] M.H. Sedghkerdar, N. Mahinpey, N. Ellis. The effect of sawdust on the calcination and the intrinsic rate of the carbonation reaction using a thermogravimetric analyzer (TGA). Fuel Process. Technol. 2013;106:533-538. [9] J. Yin, C. Qin, B. Feng, L. Ge, C. Luo, W. Liu, H. An. Calcium looping for CO2 capture at a constant high temperature. Energy Fuels 2013;28:307-318. [10] J.W. Butler, C. Jim Lim, J.R. Grace. Kinetics of CO2 absorption by CaO through pressure swing cycling. Fuel 2014;127:78-87. [11] G.S. Grasa, I. Martínez, M.E. Diego, J.C. Abanades. Determination of CaO carbonation kinetics under recarbonation conditions. Energy Fuels 2014. [12] R. Schefflan, Teach yourself the basics of Aspen plus, John Wiley & Sons; 2011. [13] R. Barker. The reversibility of the reaction CaCO3֎ CaO+ CO2. Journal of applied Chemistry and biotechnology 1973;23:733-742. [14] C. Guler, D. Dollimore, G.R. Heal. The investigation of the decomposition kinetics of calcium carbonate alone and in the presence of some clays using the rising temperature technique. Thermochim. Acta 1982;54:187-199. [15] R. Borgwardt. Calcination kinetics and surface area of dispersed limestone particles. AIChE J. 1985;31:103-111. [16] J.S. Dennis, A.N. Hayhurst. the effect of CO2 on the kinetics and extent of calcination of limestone and dolomite particles in fluidised beds. Chem. Eng. Sci. 1987;42:2361-2372. [17] E.G. Calvo, M. Arranz, P. Leton. Effects of impurities in the kinetics of calcite decomposition. Thermochim. Acta 1990;170:7-11. [18] T.R. Rao. Kinetics of calcium carbonate decomposition. Chem. Eng. Technol. 1996;19:373-377. [19] I. Ar, G. Do÷u. Calcination kinetics of high purity limestones. Chem. Eng. J. (Lausanne) 2001;83:131-137. [20] F. GarcÕa-Labiano, A. Abad, L. De Diego, P. Gayan, J. Adanez. Calcination of calcium-based sorbents at pressure in a broad range of CO< sub> 2 concentrations. Chem. Eng. Sci. 2002;57:2381-2393. [21] B. González, G.S. Grasa, M. Alonso, J.C. Abanades. Modeling of the Deactivation of CaO in a Carbonate Loop at High Temperatures of Calcination. Ind. Eng. Chem. Res. 2008;47:9256-9262. [22] I. Martínez, G. Grasa, R. Murillo, B. Arias, J.C. Abanades. Kinetics of Calcination of Partially Carbonated Particles in a Ca-Looping System for CO2 Capture. Energy Fuels 2012;26:1432-1440. [23] J. Szekely, J.W. Evans, Gas-solid reactions, Academic Press, UK; 1976. [24] E. Baker. The calcium oxide-carbon dioxide system in the pressure range 1-300 atmospheres. J. Chem. Soc. (Resumed) 1962;464-470. [25] M.M. Hossain, H.I. de Lasa. Chemical-looping combustion (CLC) for inherent separations—a review. Chem. Eng. Sci. 2008;63:4433-4451. [26] J. Blamey, E.J. Anthony, J. Wang, P.S. Fennell. The calcium looping cycle for large-scale CO2 capture. Progr. Energy Combust. Sci. 2010;36:260-279.