regeneration conditions

Journal of Cleaner Production 125 (2016) 296e308

Contents lists available at ScienceDirect

Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro

Detailed experimental study on the performance of Monoethanolamine, Diethanolamine, and Diethylenetriamine at absorption/regeneration conditions Weidong Fan*, Yacheng Liu, Kang Wang School of Mechanical and Power Engineering, Shanghai Jiao Tong University, Shanghai 200240, People's Republic of China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 24 September 2015 Received in revised form 4 March 2016 Accepted 14 March 2016 Available online 7 April 2016

Experiments were carried out in a lab-stripper to investigate the regeneration performance of CO2loaded solutions of Monoethanolamine (MEA) under various operational conditions (energy input, rich solution flow rate, rich solution loading). To better understand the regeneration process of a rich MEA solution, appropriate evaluation of the absorption and regeneration capacities of an aqueous MEA solution were first determined. Thus, first of all, a bubbling reaction system was employed to assess the absorption and desorption capacities of MEA, and other amines, such as Diethanolamine (DEA) and Diethylenetriamine (DETA), were introduced to compare these capacities in terms of amino group. The influences of the solution temperature and the solution concentration on CO2 absorption and desorption performances were investigated. The results show that these factors significantly affect CO2 absorption and desorption. Moreover, experiments in the bubbling reaction system were conducted to compare the regeneration capacities of the three CO2-loaded solutions (MEA, DEA, DETA) with increasing concentrations under the condition of different sweep gases (100% N2, 14% CO2 and 100% CO2). The data show that under the condition of each type of sweep gas, the regeneration capacity of the three types of amino groups followed the rule: DEA > DETA > MEA. On the basis of these data obtained from the reaction system, regeneration experiments with loaded 2 mol/L MEA solutions in a stripper column were conducted to analyze the relationships between the regeneration ratio and the stripper operating parameters. © 2016 Elsevier Ltd. All rights reserved.

Keywords: Monoethanolamine (MEA) Diethanolamine (DEA) Diethylenetriamine (DETA) Sweep gas Regeneration ratio Stripper

1. Introduction Because global warming has caused worldwide concern and carbon dioxide (CO2) emissions are continually growing, reducing CO2 emission has become an important task for the whole world (Mao et al., 2014; Kainiemi et al., 2015). The main source of CO2 is from the production of electricity, the majority of which is derived from pulverized coal-fired boilers. Therefore, people have made a great effort to develop suitable technologies for capturing CO2 from large-scale CO2 emission sources in industry. Although numerous technologies are generally compatible with carbon capture and storage (CCS) activity and are applied in the chemical and petroleum industries, relatively few have gained any measure of acceptance for reducing the considerable CO2 emissions from coal-fired

* Corresponding author. Tel.: þ86 21 34208287; fax: þ86 21 34206115. E-mail address: [email protected] (W. Fan). http://dx.doi.org/10.1016/j.jclepro.2016.03.144 0959-6526/© 2016 Elsevier Ltd. All rights reserved.

plants (Glier and Rubin, 2013). In general, for large point sources, especially coal-fired power plants, three acceptable technology options are suitable for commercial deployment; these are aminebased post-combustion CO2 capture, oxy-fuel combustion, and calcium looping technologies (MacDowell et al., 2010). However, amine-scrubbing technology may be a more feasible technology for CO2 capture from coal-fired power plants because of its retrofit approach for capturing CO2 from existing power plants (Mazari et al., 2015). Moreover, it is well-suited to capturing CO2 from dilute, low-pressure streams. The amine solvents absorption method as one main mode of post-combustion CO2 capture has become a more mature CO2 capture approach among CCS technologies applied in large coal-fired power plants for capturing CO2 because of its higher reliability, greater suitability, and lower investment and running cost for decarburization retrofit (Molina and Bouallou, 2015). Monoethanolamine (MEA), as a primary amine, has been widely employed as the CO2-absorbent of the chemical absorption method

W. Fan et al. / Journal of Cleaner Production 125 (2016) 296e308

because of its high CO2 reactivity. Therefore, researchers gave more attention to MEA-based processes, such as the development of novel technology using MEA, its model setup, its applications in pilot plants and energy consumption analysis (Kuntz and Aroonwilas, 2009; Faramarzi et al., 2010; Moser et al., 2011). Energy consumption in the regeneration process is higher for MEA, which is its main drawback compared to Diethanolamine (DEA) and Diethylenetriamine (DETA) (Zhang et al., 2014). DEA, as a secondary amine, is also suitable for the removal of CO2 from flue gas but exhibits slow kinetics. DETA, as an alkyl amine, has also received great attention because it owns three amino groups in each single molecule, and it has been used as an activator to improve CO2 absorption performance because of its fast reaction kinetics and high absorption capacity (Fu et al., 2012). DETA has three function groups, two primary amino groups and one secondary amino group, while MEA has one primary amino group and DEA has one secondary amino group. Most studies compared absorption/desorption capacities among different aqueous solvents in terms of the same mole amine concentration rather than the same amino group concentration (Rivera-Tinoco and Bouallou, 2010). The absorption capacity of DETA is higher than that of MEA and DEA because it has two primary amino groups and one secondary amino group per mole amine concentration (Liu et al., 2014). However, studies comparing CO2 absorption and desorption capacity for various solvents in terms of amino group concentration may be €ffer et al. (2012) did research on the commore reasonable. Scha parison of MEA and Triethylenetetramine (TETA) under the same mole amino groups, but they analyzed the results from the aspect of amine molecule instead of amino group. In comparison with the major attention paid to improving the efficiency of the CO2 absorption process, experimental research on the regeneration performance of aqueous MEA solution in a bubbling reaction system and a stripping column is less abundant. A similar experimental apparatus was employed in the literature reported by Galindo et al. (2012). They compared the regeneration characteristics between 20 wt. % MEA and 34 wt. % DEA for the same number of amino groups with the temperature ranging from 80  C to 100  C, and the results showed a higher cyclic capacity for DEA and lower dependence on CO2 partial pressure for desorption compared to MEA. However, in their lab-scale stripper, the stripping steam was mainly produced in the reboiler. The objective of this paper was to examine and compare the performances of three aqueous solutions of MEA, DEA, and DETA in terms of amino group under various operational conditions in the bubbling reaction system. Then, the regeneration experiments of CO2-loaded solutions of MEA were conducted under various operational conditions in a lab-stripper column. The main aim was to analyze the relationships between the regeneration ratio and the stripper operational parameters (energy input, rich solution flow rate, rich solution loading). First, the absorption capacity of the three CO2-loaded solutions at different amino group concentrations and temperatures were determined. The effects of the temperature and concentration with various sweep gases (100 vol. % N2, 14 vol. % CO2 and 100 vol. % CO2) on the regeneration capacity were investigated. In light of these data obtained from the reaction system, the regeneration experiments on loaded 2 mol/L MEA solutions in a stripper column were conducted. 2. Theoretical background 2.1. Absorption reaction mechanism The reaction mechanism between CO2 and MEA (DEA) can be described by the zwitterion mechanism, and both MEA as the primary amine and DEA as the secondary amine can react with CO2 to

297

form carbamate and protonated amine molecules, limiting the theoretical loading to 0.5 mol CO2/mol amine by stoichiometry. However, the formation of bicarbonate or carbonate results in CO2 loading over the theoretical value (Choi et al., 2014). Thus, overall reactions of CO2 with MEA (RNH2) and DEA (RR0 NH) can be represented as follows:

CO2 þ 2RNH2 ⇔RNHCOO þ RNHþ 3

(1)

RNHCOO þ H2 O⇔RNH2 þ HCO 3

(2)

2 þ HCO 3 þ H2 O⇔CO3 þ H3 O

(3)

MEA as a primary amine can react with CO2 faster and form more stable carbamates than DEA. The reaction of CO2 with MEA consists of two steps: first, the formation of zwitterions, and then, deprotonation of the zwitterions (Choi et al., 2009).  CO2 þ RNH2 ⇔RNHþ 2 COO

(4)

  þ RNHþ 2 COO þ RNH2 ⇔RNHCOO þ RNH3

(5)

As an alkyl amine, DETA has two primary amino groups and one secondary amino group. Hartono et al. (2007) performed the qualitative determination of the species present in the DETAeH2OeCO2 system by using the nuclear magnetic resonance (NMR) technique; the results suggested that carbamate was the main species in the system for loading below 1.0 mol CO2/mol amine, and, while exceeding that value, dicarbamate was dominating and bicarbonate or carbonate were also formed. 2.2. Regeneration reaction mechanism During the regeneration process of aqueous solutions of amine, the regenerable carbamate and bicarbonate or carbonate are thermally decomposed to liberate CO2 from the rich solutions. Because of the unstable ions' rank (Choi et al., 2014): bicarbonate or carbonate > secondary carbamate > dicarbamate > primary carbamate, the regeneration of a rich MEA solution is more difficult than that of DEA or DETA. Thus, the overall regeneration process consists of the following reactions (6 and 7) and the reverse reactions (1, 2, and 3).  HCO 3 ⇔OH þ CO2

(6)

 CO2 3 þ H2 O⇔CO2 þ 2OH

(7)

3. Experimental considerations 3.1. Experimental apparatus and procedure of the bubbling reaction system The absorption and regeneration capacities of three aqueous solutions (MEA, DEA, DETA) were measured in the bubbling reaction apparatus illustrated in Fig. 1; the main experimental operational parameters are summarized in Table 1. The solution temperature was controlled by an electrical heating jacket with a deviation of ±1 K. A condenser circulating cooling water was used at the outlet of the three-neck flask to reduce the loss of water and solvent because of evaporation. The inlet gas was fed into the amine solution through the transfer switches and the mass flowmeter, and the outlet gas flow rate was monitored by another mass flowmeter. Prior to the absorption test, aqueous solutions with concentrations of from 1 to 4 mol amino group/L were prepared.

298

W. Fan et al. / Journal of Cleaner Production 125 (2016) 296e308

Fig. 1. Schematic diagram of the bubbling reaction system.

Table 1 Main experimental operating parameters in the bubbling reaction system. Parameters

Value

Amino group concentration (mol/L) Absorption/regeneration capacities test duration Inlet gas flow rate for absorption (L/min) Sweep gas flow rate (L/min) Solution volume for absorption test (mL) Solution volume for regeneration test (mL) Regeneration temperature (K) Condenser temperature ( C)

1e4 30 0.5 0.2 70 60 353e373 ~20

Then, the solutions would be used for the absorption test until they were saturated with CO2 before the regeneration tests. Absorption experiments were conducted to assess and compare the absorption capacity among the three aqueous solutions, and pure CO2 was used to react with the amine solvents. As described in the research of Galindo et al. (2012), cyclic capacity was calculated as the difference between the loadings at 313 K and 363 K. The reaction temperature for absorption was set at 313, 333, 348, and 363 K, respectively, and 70 mL of aqueous

solution was taken for each absorption test. Pure CO2 gas was bubbled through the aqueous solutions at 0.5 L/min after the solution's temperature reached the desired value, and the outlet gas flow rate was recorded every minute after that. The absorption test lasted 30 min for each condition, and the total CO2 loading in the solvent was determined by the desorption/titration procedure described in the research conducted by Aroonwilas and Tontiwachwuthikul (1997). In the regeneration capacity tests, the aqueous solutions were saturated with CO2 to a certain concentration, and 60 mL of CO2loaded solution was added to the flask for each regeneration test. The regeneration temperature was set at 353, 363 and 373 K, and sweep gas (100% N2 or 14% CO2 or 100% CO2) was bubbled through the CO2-loaded aqueous solutions at 0.2 L/min after the solutions reached the desired temperature. The sweep gas was employed to simulate the effect of the stripping steam on the regeneration process in the stripper column. Each regeneration capacity test lasted for 30 min under each condition. The CO2 loading was determined after regeneration by the same procedure applied after the absorption test.

W. Fan et al. / Journal of Cleaner Production 125 (2016) 296e308

3.2. Experimental apparatus and procedure of the lab-scale stripper The experimental apparatus for regeneration of CO2-loaded solutions of MEA in a lab-stripper is shown in Fig. 2. The regeneration apparatus mainly consists of a stripper column, a thermostatic oil bath for supplying the desorption energy input, a condenser for cooling the water and solvent to reflux, and a rich solution storage tank for continuously supplying the CO2-loaded solutions of MEA. The stripper column consists of three sections: the upper section for introducing the CO2-loaded MEA solutions, the middle section for releasing CO2 from the rich solution, and the bottom section for reboiling the rich solution to achieve a higher degree of regeneration. The middle section of the lab-stripper is well-packed with glass spring rings with a size of Ф7  14 mm; the packing height is 500 mm. The packing of the glass spring rings can supply enough area to achieve a higher desorption of the rich MEA solution. The outer wall of the packing section is actually a doublelayer structure. The interlayer space between the two layers' walls can be filled with hot oil. The hot oil enters from the bottom of the interlayer space, and drains away from its top. The flow of hot oil is continuous; therefore, the wall temperature of the packing section can be kept at a certain constant temperature by this thermostatic oil bath to maintain a constant optimum regeneration temperature condition inside the packing section. Moreover, a set of thermocouples and a temperature display device are arranged at the entrance of the middle section for monitoring the actual inlet rich solution temperature. The bottom section of the lab-stripper is full of spiral pipe, and the hot oil is introduced into the pipe to reboil the MEA solution. Additionally, a set of thermocouples and a temperature display device are arranged at the entrance of the bottom

299

section for monitoring the solution temperature streaming down from the middle section. After the temperature in the stripper became stable, a magnetic pump was started to transport CO2-loaded amine solution from the storage tank; the solution flow rate could be controlled by a liquid flowmeter. Then it was pumped into the heat exchanger, where it acquired heat from high-temperature water (366 K) provided from the thermostatic water bath. After this process, the solution was injected into the column by a sprayer to liberate CO2 from the rich solution. 3.3. Analytical methods The CO2 absorption/regeneration capacities of three aqueous solvents were investigated by measurements of solution loading, absorption rate, and cyclic capacity under per mole amino groups. Moreover, the influence of operational parameters (such as amino group concentration, solution temperature, and the inlet CO2 concentration in the sweep gas) on solution loading were investigated and compared for MEA, DEA, and DETA. Solution loading is a vital parameter that it refers to the total amount of CO2 dissolved in an aqueous amine solution by chemical and physical absorption, which is expressed by the following equation:

b¼

MCO2  103 V0  ½A

(8)

where b is the CO2 loading in the aqueous solution in mol CO2/mol amino group; MCO2 is the CO2 amount absorbed by aqueous amine

Fig. 2. Schematic diagram of the lab-stripper experiment system.

300

W. Fan et al. / Journal of Cleaner Production 125 (2016) 296e308

solution in mol; V0 is the volume of aqueous solvent in mL; and [A] represents the amino group concentration of the solvent in mol/L. The CO2 absorption rate is a dominant parameter in determining the degree of the CO2 absorption process; it is expressed as follows:

N¼

Q in  Q out V0  60  22:4

(9)

where N is the absorption rate in mol CO2/(L$s); Qin is the inlet gas flow rate in mL/min; and Qout is the outlet gas flow rate in mL/min. The cyclic capacity is the gap of the CO2 loading amount between the gaseliquid equilibrium of absorption and regeneration under the per mole amino group, which is an important parameter for estimating the absorption characteristics of the solvent. In this study, the CO2 absorption and regeneration temperatures of the three aqueous solvents are approximately 313 K and 363 K, respectively. In the regeneration experiments on CO2-loaded solutions of MEA in a lab-stripper, the regeneration ratio ε (%) was employed to reflect the amount of active free amine molecules regenerated from the aqueous solution and is defined as the difference between the initial rich loading and the achieved lean loading divided by the rich loading. Thus, it can be expressed by the following equation:

ε¼

b0  b∞  100% b0

(10)

where b0 is the initial amount of CO2 loading in the rich solution in mol CO2/mol MEA; and b∞ is the amount of CO2 loading in the lean solution taken out from the lean solution outlet in mol CO2/mol MEA. 3.4. Chemicals and sample preparation Reagent grade Monoethanolamine (MEA), Diethanolamine (DEA), and Diethylenetriamine (DETA) were all used as purchased without purification. Three aqueous solutions were formed by weighing the required amount of amine and then mixing with deionized water. In the study, a series of samples covering a range of amino group concentrations from 1 mol/L to 4 mol/L were prepared; that is to say, for the aqueous solution of DETA, the amine concentration ranged from 1/3 mol/L to 4/3 mol/L because of the three amino groups in the chemical structure of DETA. 4. Results and discussion 4.1. Absorption capacity 4.1.1. Equilibrium loading To assess the theoretical cyclic capacity of the three aqueous solutions (MEA, DEA, and DETA), the equilibrium curves of the amount of CO2 loading of the three aqueous solutions over time at 313 K and 363 K are shown in Fig. 3 at amino group concentrations of 1, 2 and 3 mol/L, respectively. These temperatures of 313 K and 363 K roughly represent the absorption and regeneration condition, respectively. From Fig. 3 it can be seen that the comparison for equilibrium loadings follows the rule: MEA > DEA > DETA for the absorption condition and MEA > DETA > DEA for the regeneration condition in terms of amino group concentration. When the results for the absorption capacities of MEA are compared with the data on CO2 solubility given by Choi et al. (2014), the results in this study are slightly higher than those in the literature. This is because pure CO2 gas was used in this study and less reaction time was needed to reach the final equilibrium. Furthermore, it can be inferred that

DEA has the highest cyclic capacity, and it decreases as the amino group concentration increases. The equilibrium loadings of MEA and DEA reach approximately 0.8 mol CO2/mol amino group at a concentration of 1 amino group/L, which obviously exceed those of DETA. Loading capacities range from 0.6 to 0.8 mol CO2/mol amino group for MEA and DEA for the absorption condition with the amino group concentration increasing, while that of DETA ranges from 0.58 to 0.67 mol CO2/mol amino group. However, all of the equilibrium loadings exceed theoretical values (0.5 mol CO2/mol amino group), which implies the formation of bicarbonate or carbonate in Eqs. (2) and (3). For the CO2 absorption condition, DETA has the lowest loading amount because the three amino groups have not been consumed totally in every amine molecule, which is validated by the fact that no tricarbamate species been found in the DETAeH2OeCO2 system (Hartono et al., 2007). It can also be seen that it takes longer for DEA to reach equilibrium because of its slower kinetics. 4.1.2. Comparison of absorption rate Fig. 4 illustrates the absorption rate curves of MEA, DEA, and DETA in terms of amino group at 313 K under various concentrations. It can be shown that the change rules for the absorption rate for DETA are quite similar to those for MEA. The initial absorption rates for MEA and DETA are higher than those of DEA with an equivalent number of amino groups, but the absorption rate of DEA exceeds the others after a period of reaction. The reason may be as follows: the kinetics of a secondary amino group are slower than those of a primary amino group, and MEA has the maximum ratio of primary amino groups, while DEA has the largest component of secondary amino groups. Therefore, the initial absorption rate of MEA is the fastest and that of DEA is the lowest when the reaction starts with same number of amino groups. However, the number of amino groups in MEA and DETA becomes less for carbamate formation than that of DEA after a period of reaction. That is to say, the bicarbonate formation reaction takes the lead in aqueous MEA and DETA solutions when the reaction is going on, but the kinetics for bicarbonate formation are slower than that for carbamate formation; therefore, the absorption rate of DEA exceeds that of MEA and DETA. It can also be seen that the initial absorption rate of the three solutions increases with increasing concentration of amino groups because more free active amino groups can react with CO2 in a unit volume. However, the difference of the initial absorption rate for MEA among the different concentrations varies less than that of DEA, which can be explained by the volumetric overall masstransfer coefficient (KGaV). Because the inlet gas flow rate for absorption is maintained at 0.5 L/min, KGaV tends to be stable despite increases in the amino groups' concentration of aqueous MEA solution (Fu et al., 2014). 4.1.3. Effect of solution temperature Solution temperature is a vital parameter impacting CO2 capture performance and the reverse absorption reaction would gradually dominate the reaction process as the temperature increases because the CO2 absorption reaction is exothermic. In Fig. 5, the equilibrium loadings from 313 K to 363 K are shown in terms of amino group as the concentration of the amino groups increase from 1 to 3 mol/L. It can be seen that DEA spans the greatest loading range, from 0.34 to 0.82 mol CO2/mol amino group among the various concentrations. The highest loadings are achieved at the concentration of 1 mol amino group/L, and the loadings decrease with increasing concentration for the same temperature. Moreover, for a concentration of 1 mol amino group/L, the trend of the curve for MEA is linearly decreasing. It can also be seen that the equilibrium loadings for MEA show little temperature dependence

loading(mol CO2/mol amino group)

W. Fan et al. / Journal of Cleaner Production 125 (2016) 296e308

0.9

1 m ol am ino group/L 0.9

2 m ol am ino group/L 0.9

0.8

0.8

0.8

0.7

0.7

0.7

0.6

0.6

0.6

0.5

0.5

0.5

0.4

0.4

0.4

0.3

0.3

0.3

0.2

0.2

0.2

0.1

0.1

0.1

0.0

0.0 0

5 10 15 20 25 30

301

3 m ol am ino group/L

MEA 313 K DEA 313 K DETA 313 K MEA 363 K DEA 363 K DETA 363 K

0.0 0

5 10 15 20 25 30

0

5 10 15 20 25 30

tim e (m in) Fig. 3. Equilibrium loading curves at 313 K and 363 K for three aqueous solutions.

Fig. 4. Comparison of absorption rate among three aqueous solutions.

between 313 K and 348 K compared to the others and have a prominent decline at 363 K. Meanwhile, the equilibrium loading of DEA decreases more rapidly with increasing temperature, but there is a moderate decline for DETA. Therefore, the equilibrium loading of DEA is the most susceptible to increasing temperature, which means that the secondary amino group is more easily influenced by temperature. It can also be seen from Fig. 5 that the loading amounts of DEA and DETA are lower than the theoretical values (0.5 mol CO2/mol amino group) at the regeneration temperature (363 K) at the concentration of 1 mol amino group/L, but the loading amount of MEA is nearly 0.6 mol CO2/mol amino group. That is to say, the primary carbamate has not been decomposed, while the secondary carbamate and dicarbamate are partially decomposed. When the results

for the loading amounts of MEA are compared with the literature data given in Table 2, it can be concluded that a higher concentration of amino groups and a higher temperature would decrease CO2 loading, which is in accordance with the results in this work. 4.1.4. Effect of solution concentration The amino group concentration of the solution is a vital parameter worthy of serious discussion because it determines the amount of free active amine molecules per unit volume. Equilibrium loadings at 313 K and 363 K with increasing concentration for the three solvents are illustrated in terms of amino groups in Fig. 6. It can be seen that the equilibrium loadings decrease with increasing number of amino groups and achieve maximum value at the concentration of 1 amino group/L for both absorption and

302

W. Fan et al. / Journal of Cleaner Production 125 (2016) 296e308

2 m ol am ino group/L

loading(mol CO2/mol amino group)

1 m ol am ino group/L 0.85

0.85

0.85

0.80

0.80

0.80

0.75

0.75

0.75

0.70

0.70

0.70

0.65

0.65

0.65

0.60

0.60

0.60

0.55

0.55

0.55

0.50

0.50

0.50

0.45

0.45

0.45

0.40

0.40

0.40

0.35

0.35

0.35

313 333 348 363

313 333 348 363

3 m ol am ino group/L

MEA DEA DETA

313 333 348 363

tem perature(K) Fig. 5. Temperature dependence of three aqueous solutions.

Table 2 Summary of literature review for CO2 loading of MEA solution. Amino group concentration

Temperature (K)

Loading (mol CO2/mol amino group)

Ref.

4 mol/L

313 363 313 333 353

0.57 0.42 0.459 0.448 0.382

Galindo et al., 2012 Galindo et al., 2012 Choi et al., 2014 Choi et al., 2014 Choi et al., 2014

5 mol/L

loading(mol CO2/mol amino group)

0.90 0.85

MEA 313 K DEA 313 K DETA 313 K MEA 363 K DEA 363 K DETA 363 K

0.80 0.75 0.70 0.65 0.60 0.55 0.50 0.45 0.40 0.35 0.30

1

2

3

4

concentration(mol amino group/L) Fig. 6. Effect of concentration on loadings of three aqueous solutions.

regeneration conditions. It can be inferred that a lower concentration is beneficial for an amino group to reach higher equilibrium loading because an increase of solution concentration would result in the rise of solution viscosity. This means that the liquid film thickness at the border and the mass transfer resistance between the liquid phase and the gas phase would increase, which militates against loading as the solution concentration increases

(Fu et al., 2012). Equilibrium loading of each solvent between 3 mol amino group/L and 4 mol amino group/L shows little change. Increasing concentration can increase the absorption amount by supplying more free active amino groups per unit volume, but it would not improve the absorption capacity for each amino group. It can also be seen that the curves of the three aqueous solutions at the absorption condition (313 K) is steeper than that at the regeneration condition (363 K) as the concentration of amino group ranges from 1 mol/L to 3 mol/L. That is because rising solution temperature would make the solution viscosity decrease and promote a second-order reaction rate constant, which are conducive to CO2 capture. Thus, for higher temperature the loading decreases slowly for the three solvents. 4.1.5. Comparisons of cyclic capacity Comparisons of cyclic capacity are illustrated among the three aqueous solutions in terms of amino groups in Fig. 7. As described above, the cyclic capacities are calculated from the equilibrium loadings shown in Fig. 6. It can be seen that the cyclic capacities decrease with increasing concentration by per mole amino group and by per unit mass amine. However, cyclic capacities increase with increasing amino group by unit mass solvent. The cyclic capacity of DEA ranges between 0.25 and 0.42 mol CO2/mol amino group, while those of MEA and DETA range from 0.15 to 0.23 mol CO2/mol amino group and from 0.13 to 0.19 mol CO2/mol amino group, respectively. Comparing the cyclic capacities on an amine mass basis, the cyclic capacity of DETA ranges from 0.16 to 0.25 g CO2/g amine, which is obviously higher than that of MEA and DEA. This is a predictable result because the aqueous solution of DETA has more amino groups than the others, based on the same mass amine. DEA presents the highest cyclic capacity in terms of mole amino group basis and unit mass solvent basis while DETA performs the best in the aspect of cyclic capacity on unit mass amine basis. It can also be seen that the cyclic capacity of DETA is similar to that of MEA on amole amino group basis and unit mass solvent basis, and the cyclic capacity of DETA and MEA vary less in terms of absorbed CO2 by each unit mass of solvent.

cyclic capacity (mol CO2/mol amino group)

W. Fan et al. / Journal of Cleaner Production 125 (2016) 296e308

0.44

0.25

cyclic capacity (g CO2/kg solvent)

cyclic capacity (g CO2/g amine)

46 44 42 40 38 36 34 32 30 28 26 24 22 20 18 16 14 12 10 8

0.42

0.24

0.40

0.23

0.38

0.22

0.36

0.21

0.34

0.20

0.32

0.19

0.30

0.18

0.28

0.17

0.26

0.16

0.24

0.15

0.22

0.14

0.20

0.13

0.18

0.12

0.16

0.11

0.14 0.12

0.10

1

2

3

4

303

1

2

3

4

MEA DEA DETA

1

2

3

4

concentration(mol amino group/L) Fig. 7. Cyclic absorption capacity of three aqueous solutions.

4.2. Regeneration capacity 4.2.1. Regeneration tests with N2 as the sweep gas The measurements of regeneration capacities with N2 as the sweep gas for the three solutions were carried out to analyze the changing trend of lean loading by each amino group with increasing concentration at different regeneration temperatures. The loading curves are shown in Fig. 8. It can be seen that lean loadings increase with increasing concentration in terms of amino group concentration for MEA and DETA, while that of DEA varies little among the various concentrations. DEA achieves the best regeneration result among the three solvents and DETA the second. Average loadings of DEA are 0.27 mol CO2/mol amino group at 353 K, 0.19 mol CO2/mol amino group at 363 K, and 0.12 mol CO2/ mol amino group at 373 K, which are obviously lower than those of MEA and DETA. The reason for this phenomenon may be described as follows: the sweep gas N2 can decrease the CO2 partial pressure in the regeneration atmosphere, which is beneficial for the release of CO2 to promote the regeneration process. However, increasing concentration is unfavorable to the regeneration process because it can raise energy demand for desorption and water evaporation. Therefore, lean loadings of MEA and DETA become higher with increasing concentration. Because DEA is prone to regenerate at a lower temperature, increasing concentration would not obviously influence the regeneration results. 4.2.2. Regeneration tests with 14% CO2 as the sweep gas CO2 at 14 vol. % was also used as a sweep gas to determine the regeneration results of the three saturated solutions. The 14% CO2 mixture as the sweep gas was prepared by mixing pure CO2 and pure N2 in the gas mixer. Loading curves of the three solvents are shown in terms of amino group in Fig. 9. It can be seen that the regeneration results for the three solutions at the same temperature and concentration follow the rule: DEA > DETA > MEA. The changing trends with 14% CO2 for the three solvents were similar to each other. The minimum loadings for DETA are achieved at 2 mol amino group/L at each regeneration temperature, which are 0.39, 0.32, and 0.255 mol CO2/mol amino group individually at 353 K,

363 K, and 373 K, respectively. The minimum loadings for DEA are achieved at 3 mol amino group/L, which are 0.30, 0.21, and 0.135 mol CO2/mol amino group at the three regeneration temperatures. MEA also achieves minimum loadings at 3 mol amino group/L. However, the data are still higher than those of the corresponding case in Fig. 8; that is, the lower the CO2 partial pressure in the sweep gas is, the lower the loading after regeneration is. The concentrations for reaching minimum loading exhibit differences among the three aqueous solutions. The following reasoning may explain this phenomenon. With 14% CO2 as the sweep gas, the N2 in the mixture gas can promote regeneration, but the CO2 in the mixture gas can react with unsaturated amino groups (Aroonwilas et al., 1999). As discussed in the absorption results, higher loading can be accomplished at lower concentration, and high concentration is unfavorable to regeneration. Therefore, each solvent achieves the highest loading at 1 mol amino group/L, and as the concentration increases, loading decreases with declining water-amine ratio (Hartono et al., 2009). Because unfavorable factors for regeneration become obvious with increasing concentration, loadings will increase after a certain concentration. The minimum loading for each solvent is achieved under the combined effects of both sweep gas and concentration. 4.2.3. Regeneration tests with CO2 as the sweep gas Pure CO2 was adopted as another sweep gas. The loading curves in terms of amino group are shown in Fig. 10. It can be seen that the loadings decrease with increasing amino group concentration. As mentioned above, low concentration is favorable to the reaction of CO2 with the amino group, and there is enough CO2 for the reaction in the regeneration atmosphere (100% CO2). Therefore, loadings decrease with increasing concentration. It can also be seen that the loading of DEA declines rapidly with increasing regeneration temperature. By comparing the loading curves of the three solvents for each sweep gas in terms of amino group in Figs. 8e10, it can be clearly seen that loadings of DEA are much lower than those of MEA and DETA for each sweep gas. This means that DEA performs best in the regeneration results and DETA is second. Lean loadings become

304

W. Fan et al. / Journal of Cleaner Production 125 (2016) 296e308

sweep gas:100% N2

loading(mol CO2/mol amino group)

0.60

0.60

353 K

0.60

363 K

0.56

0.56

0.56

0.52

0.52

0.52

0.48

0.48

0.48

0.44

0.44

0.44

0.40

0.40

0.40

0.36

0.36

0.36

0.32

0.32

0.32

0.28

0.28

0.28

0.24

0.24

0.24

0.20

0.20

0.20

0.16

0.16

0.16

0.12

0.12

0.12

1

2

3

4

1

2

3

4

373 K MEA DEA DETA

1

2

3

4

concentration(mol amino group/L) Fig. 8. Comparison of regeneration among three aqueous solutions with pure N2 as the sweep gas.

Fig. 9. Comparison of regeneration among three aqueous solutions with 14% CO2 mixture as the sweep gas.

higher with increasing CO2 ratio in the sweep gas, so regeneration tests with N2 as the sweep gas are the best. For example, for DETA at the concentration of 2 mol amino group/L, the value of 0.4 mol CO2/ mol amino group is achieved with CO2 as the sweep gas, while 0.25 mol CO2/mol amino group is achieved with N2 as the sweep gas. Moreover, it can also be seen that the loadings of the three solvents are higher than 0.5 mol CO2/mol amino group with pure

CO2 as the sweep gas, an amino group concentration of 1 mol/L, and a solution temperature of 353 K, but the measurements for the corresponding cases of various sweep gases are less than the theoretical values. So the decomposition of carbamate has not proceeded with pure CO2 as the sweep gas, and it was confirmed that the regeneration capacity increases with lower CO2 concentration in the sweep gas.

W. Fan et al. / Journal of Cleaner Production 125 (2016) 296e308

305

sweep gas:100% CO2

loading(mol CO2/mol amino group)

0.60

353 K

0.60

363 K

0.60

0.56

0.56

0.56

0.52

0.52

0.52

0.48

0.48

0.48

0.44

0.44

0.44

0.40

0.40

0.40

0.36

0.36

0.36

0.32

0.32

0.32

0.28

0.28

0.28

0.24

0.24

0.24

0.20

0.20

0.20

0.16

0.16

0.16

0.12

0.12

0.12

1

2

3

4

1

2

3

4

373 K MEA DEA DETA

1

2

3

4

concentration (mol amino group/L) Fig. 10. Comparison of regeneration among three aqueous solutions with pure CO2 as the sweep gas.

4.2.4. Comparison of regeneration ratio To clearly exhibit and compare the regeneration capacities of the three CO2-loaded aqueous solutions (MEA, DEA, DETA), regeneration ratios were calculated to study the influence of various operational parameters (sweep gas, amino group concentration, solution temperature) on the three solvents. As shown in Fig. 11, the data obtained are presented as a function of the amine type, concentration, temperature, and sweep gas. As expected, the regeneration ratio increased with lower CO2 partial pressure in the sweep gas, lower amino group concentration, and higher solution temperature. The regeneration capacity of DEA can be seen to be highest, which confirmed the pattern showed in the loading experiments. The regeneration ratio of the aqueous MEA solution under the conditions of 373 K, 1 mol amino group concentration, and pure N2 as the sweep gas showed a superior result compared to the others in the MEA solution. Over 70% of the CO2 dissolved in the solution can be released, reaching the lean loading of 0.22 mol CO2/mol amino group shown in Fig. 8. And the regeneration ratio of the aqueous MEA solution under the condition of 363 K, 1 mol amino group concentration, and pure CO2 as the sweep gas showed the worst result. Only approximately 35% of the CO2 was liberated from the rich MEA solution (lean loading of 0.53 mol CO2/mol amino group), which means that only bicarbonate ions and carbonate ions were decomposed under this condition. Moreover, it can be inferred that solution temperature is the most effective parameter for the regeneration ratio of the three solvents because, at higher temperature levels, four results can be achieved: 1. Supplying enough energy to decompose carbamate; 2. Decreasing the solution viscosity and increasing the mass transfer coefficient; 3. Declining CO2solubility in the water (Vrachnos et al., 2006); 4. Intensifying the evaporation of water to reduce the CO2 partial pressure in the sample flask.

4.3. Stripper performance In the lab-scale stripper, a series of experiments were carried out to better understand the desorption process under various operational parameters. By reference to other scholars' experiments and previous experiments, the reasonable operational condition (MEA concentration of 2 mol/L; rich solution flow rate of 5 L/ h; rich loading of 0.52 mol CO2/mol MEA; and temperature of the thermostatic oil bath of 410 K) was selected as the baseline case for comparison. 4.3.1. Effect of energy input The energy input to the stripper was supplied by the thermostatic oil bath. By changing the temperature of the heating oil in the thermostat, the input energy can be varied. The heat duty can be estimated as the sum of the energy used for creating the necessary driving force for desorption, warming up the rich amine liquid feed to reboiler temperature, and the heat of condensation of water from the saturated CO2 from the stripper. Heat duty is introduced to compare energy input for the same CO2 release amount for different oil-bath temperatures. The loading of the CO2-loaded MEA solution was measured at oil-bath temperatures ranging from 394 K to 411 K, and the changed loading in the packing section and the spiral pipe section of the stripper is shown in Fig. 12. It can be seen that the changed loading in both sections increases with increasing oil-bath temperature, supplying higher energy into the rich solution. However, the changed loading amount in the packing section is more than that in the spiral pipe section. More specifically, the changed loading in the packing section is approximately 0.16 mol CO2/mol MEA at 411 K, which is nearly two times higher than that value (0.08 mol CO2/mol MEA) at 394 K. In the spiral pipe section, the changed loading is 0.11 mol CO2/mol MEA at 394 K, which is close to that value (0.14 mol CO2/mol MEA) at 411 K. Moreover, the changed loading amount in both sections is

306

W. Fan et al. / Journal of Cleaner Production 125 (2016) 296e308

Regeneration ratio (%)

363 K, 1mol amino group/L

363 K, 100% N2

90

90

80

MEA DEA 80 DETA

80

70

70

70

60

60

60

50

50

50

40

40

40

30

30

30

20

20

20

10

10

10

3 37

3 36

K

K

K

2 3 4 mol amino group/L

100% N2, 1mol amino group/L

3 35

N2

2 CO 0% 10 2 CO

% 14

0% 10

1

90

0.35

packing section spiral pipe section

lean loading (mol CO2 /mol MEA)

changed loading (mol CO2 /mol MEA)

0.40

0.30 0.25 0.20 0.15 0.10 0.05 0.00

0.50

70

0.45

65

0.40

60

0.35

55

0.30

50

0.25

45

0.20

40

0.15

35

0.10

30

0.05

25

0.00

20

2 41 0 41 8 40 6 40 4 40 2 40 0 40 8 39 6 39 4 39 2 39

2 41 0 41 8 40 6 40 4 40 2 40 0 40 8 39 6 39 4 39 2 39

temperature (K)

temperature (K)

Rengeneration ratio (%)

Fig. 11. Comparison of regeneration ratio under various operational parameters.

Fig. 12. Regeneration ratio and lean loading with different oil-bath temperature.

approximately equal despite the different temperatures. The reason may be that the solution temperature in the spiral pipe section is higher than that in the packing section. Therefore the changed loading in the spiral pipe section is higher at lower temperature (394 K). However, the evaporation of water from the rich solution in the spiral pipe section would be more intensive with increasing oilbath temperature, and would result in reducing the CO2 partial pressure in the packing section, which is more conducive to regeneration in the packing section. In conclusion, the lean loading decreases with increasing temperature, and the regeneration ratio increases with increasing temperature. The loading amount of MEA decreases from 0.52 mol CO2/mol MEA to 0.23 mol CO2/mol MEA at 411 K. The main disadvantage of MEA is its high energy requirement for regeneration. The desorption energy consumption occupies a large part in the decarbonization system (Vatopoulos and Tzimas, 2012). Thus, it is meaningful to introducing heat duty to compare the energy input for the same CO2 release amount.

Fig. 13 shows the characteristics of heat duty and CO2 release at different oil-bath temperatures. The curve in the figure indicates that the CO2 release amount (20 mol CO2/kWh) is the highest value for a unit energy consumption at 397 K, and corresponds to the lowest heat duty (4.1 GJ/t CO2). The lowest CO2 release amount, at 407 K, is 13 mol CO2/kWh and corresponds to the highest heat duty (6.36 GJ/t CO2). It can be inferred that a 397 K oil-bath temperature is reasonable to guarantee the necessary amount of energy to the stripper with the minimum cost compared to others. Values in the literature (Stec et al., 2015; Lin and Wong, 2014) have been reported at approximately 4 GJ/t CO2 for the standard MEA-based process, which agrees with the minimum heat duty. The overall trend of heat duty increases with increasing oil-bath temperature, but the value at 411 K is lower than that at 407 K because the viscosity of the oil would decrease with increasing temperature, which is conducive to promoting the heat transfer coefficient from the high temperature oil to the rich solution.

394

396

398

400

402

404

406

408

410

412

heat duty (GJ/t CO2 )

6.5 6.0 5.5 5.0

70

0.45

65

0.40

60

0.35

55

0.30

50

0.25

45

0.20

40

0.15

35

0.10

30

0.05

25

0.55

0.60

20 0.65

rich loading (mol CO 2/mol MEA)

4.0 3.5 392

Fig. 15. Regeneration ratio and lean loading for different rich loading values.

394

396

398

400

402

404

406

408

410

412

temperature (K) Fig. 13. Heat duty and CO2 release with different oil-bath temperatures.

0.50

70

0.45

65

0.40

60

0.35

55

0.30

50

0.25

45

0.20

40

0.15

35

0.10

30

0.05

25

Rengeneration ratio (%)

4.3.2. Effect of rich solution flow rate Solution flow rate has a significant impact on the regeneration performance of the CO2-loaded MEA solution. As shown in Fig. 14, the regeneration ratio decreases and lean loading increases as the solution flow rate increases from 3 L/h to 8 L/h. This observation can be explained by the following analysis: increasing solution flow rate means that more loaded amine molecules are provided to release CO2, but unfortunately, some rich solutions drain away without being fully desorbed. Additionally, a higher solution flow rate shortens the residence time of the solution in the stripper and results in a lower temperature of the rich solution in the packing section and the spiral pipe section as the solution flow rate increases from 3 L/h to 8 L/h. More specifically, the rich solution temperature in the stripper decreases from 370 K to 367 K as the solution flow rate ranges from 3 L/h to 8 L/h. Thus, the lean loading achieved is at its maximum value (0.33 mol CO2/mol MEA) at a flow rate of 8 L/h and a minimum value (36%) of the regeneration ratio. The curves in Fig. 14 show that the lean loading still remains at a lower value (0.16e0.23 mol CO2/mol MEA) and the regeneration ratio remains at a higher value (36%e44%) as the solution flow rate ranges from 3 L/h to 5 L/h. That is, the vaporization of water for CO2

lean loading (mol CO2/mol MEA)

0.50

0.00 0.50

4.5

307

Regeneration ratio (%)

21 20 19 18 17 16 15 14 13 12 392

lean loading (mol CO 2/mol MEA)

CO2 release( mol CO2 /kWh)

W. Fan et al. / Journal of Cleaner Production 125 (2016) 296e308

20

0.00 2

3

4

5

6

7

8

9

rich solution flow rate (L/h) Fig. 14. Regeneration ratio and lean loading with different solution flow rates.

stripping from the reboiler is still enough to achieve a lower lean loading and lower CO2 partial pressure in the stripping steam, which contributes to desorption of the rich solution. And as the solution flow rate increases, a larger amount of CO2 would be released by per energy consumption for regeneration. However, these values would obviously change as the flow rate increases from 5 L/h to 8 L/h. This means that less stripping steam than actually necessary is generated in the reboiler and a higher rate of heat supply is required. These results are in good agreement with those measured by different authors (Galindo et al., 2012; Perevertaylenko et al., 2015). Therefore, perhaps 5 L/h is an optimal value to achieve more CO2 at the same energy input. We chose 5 L/h as the baseline condition. 4.3.3. Effect of rich solution loading The loading of the rich solution is an important factor for calculation of the regeneration ratio. According to the data in Fig. 3, for a concentration of 2 mol/L, the absorption capacity of MEA can reach 0.68 mol CO2/mol MEA. In this section, the loading and regeneration ratio are compared at different concentrations from 0.52 to 0.64 mol CO2/mol MEA, and the results obtained are shown in Fig. 15. As we know, the theoretical loading of MEA is limited to 0.5 mol CO2/mol amine by stoichiometry; thus, part of the CO2 contributes to the formation of bicarbonate or carbonate in the solution. Moreover, bicarbonate or carbonate ions are rather unstable and desorption occurs easily without increasing the energy requirement. This is the reason why the data in Fig. 15 show that the lean loading is approximately 0.24 mol CO2/mol MEA whatever the initial rich loading is. As a result, the regeneration ratio of CO2loaded MEA solution rises gradually with increasing rich loading, and the maximum value of approximately 61% can be achieved at an initial rich loading of 0.64 mol CO2/mol MEA. Moreover, the results of the research by Galindo et al. (2012) showed that a greater energy requirement was needed to desorb CO2 from lower rich amine loading. Thus, to achieve a higher regeneration ratio in CO2-loaded MEA solution, it is recommended that the absorption achieve higher CO2 loading. In actual industrial applications, MEA can reach rich loadings as high as 0.47 mol CO2/mol MEA depending on the operational conditions and the absorption column performance (Tobiesen et al., 2008). 5. Conclusion First, experiments were carried out in a bubbling reaction system to compare the absorption/regeneration capacities of three

308

W. Fan et al. / Journal of Cleaner Production 125 (2016) 296e308

amine solvents (MEA, DEA, DETA) in terms of amino group under various operational parameters. After appropriate evaluation of these capacities of the aqueous MEA solution, experiments were carried out in a lab-stripper to investigate the regeneration performance of CO2-loaded solutions of MEA under various combinations of energy input, rich solution flow rate, and rich solution loading. The main focus was the relationships between the regeneration ratio and the stripper operational conditions.  The order of loadings is: MEA > DEA > DETA at absorption conditions and MEA > DETA > DEA at regeneration conditions in terms of each amino group with the same concentration of amino groups. In the comparison of cyclic capacity, DEA presents the best in terms of cyclic capacity on a mole basis, while DETA performs the best in terms of cyclic absorbed CO2 by unit mass of amine.  Regeneration tests show that higher temperature and lower CO2 partial pressure in the sweep gas could result in a higher regeneration ratio. Low amino group concentration is beneficial to regenerating with N2 as the sweep gas. Temperature is the most effective parameter affecting the regeneration of the three solvents.  A higher temperature for the oil-bath could contribute to a higher regeneration ratio, but 397 K is a reasonable temperature because it guarantees the necessary amount of energy to the stripper at the minimum cost.  The lower the flow rate of rich solution is, the higher regeneration ratio is. Meanwhile, 5 L/h is an optimal value to achieve a greater amount of CO2 release per energy consumption for regeneration.  The lean loading is approximately 0.24 mol CO2/mol MEA whatever the initial rich loading is. The higher the rich solution loading is, the higher the regeneration ratio is. References Aroonwilas, A., Tontiwachwuthikul, P., 1997. High-efficiency structured packing for CO2 separation using 2-amino-2-methyl-1-propanol (AMP). Sep. Purif. Technol. 12, 67e79. Aroonwilas, A., Veawab, A., Tontiwachwuthikul, P., 1999. Behavior of the masstransfer coefficient of structured packings in CO2 absorbers with chemical reactions. Ind. Eng. Chem. Res. 38, 2044e2050. Choi, S.Y., Nam, S.C., Yoon, Y.I., Park, K.T., Park, S.-J., 2014. Carbon dioxide absorption into aqueous blends of methyldiethanolamine (MDEA) and alkyl amines containing multiple amino groups. Ind. Eng. Chem. Res. 53 (37), 14451e14461. Choi, W.-J., Seo, J.-B., Jang, S.-Y., Jung, J.-H., Oh, K.-J., 2009. Removal characteristics of CO2 using aqueous MEA/AMP solutions in the absorption and regeneration process. J. Environ. Sci. 21 (7), 907e913. Faramarzi, L., Kontogeorgis, G.M., Michelsen, M.L., Thomsen, K., Stenby, E.H., 2010. Absorber model for CO2 capture by monoethanolamine. Ind. Eng. Chem. Res. 49, 3751e3759. Fu, K., Chen, G., Liang, Z., Sema, T., Idem, R., Tontiwachwuthikul, P., 2014. Analysis of mass transfer performance of monoethanolamine-based CO2 absorption in a packed column using artificial neural networks. Ind. Eng. Chem. Res. 53 (11), 4413e4423.

Fu, K., Sema, T., Liang, Z., Liu, H., Na, Y., Shi, H., Idem, R., Tontiwachwuthikul, P., 2012. Investigation of mass-transfer performance for CO2 absorption into diethylenetriamine (DETA) in a randomly packed column. Ind. Eng. Chem. Res. 51 (37), 12058e12064. €ffer, A., Brechtel, K., Unterberger, S., Scheffknecht, G., 2012. ExperGalindo, P., Scha imental research on the performance of CO2-loaded solutions of MEA and DEA at regeneration conditions. Fuel 101, 2e8. Glier, J.C., Rubin, E.S., 2013. Assessment of solid sorbents as a competitive postcombustion CO2 capture technology. Energy Procedia 37, 65e72. Hartono, A., da Silva, E.F., Svendsen, H.F., 2009. Kinetics of carbon dioxide absorption in aqueous solution of diethylenetriamine (DETA). Chem. Eng. Sci. 64 (14), 3205e3213. Hartono, A., Silva, E. F. d., Grasdalen, H., Svendsen, H.F., 2007. Quality determination of species in DETA-H2O-CO2 system using NMR spectra. Ind. Eng. Chem. Res. 46, 249e254. €nen, J., Ja €rvinen, M., 2015. Opportunities and Kainiemi, L., Eloneva, S., Toikka, A., Leva obstacles for CO2 mineralization: CO2 mineralization specific frames in the interviews of Finnish carbon capture and storage (CCS) experts. J. Clean. Prod. 94, 352e358. Kuntz, J., Aroonwilas, A., 2009. Mass-transfer efficiency of a spray column for CO2capture by MEA. Energy Procedia 1 (1), 205e209. Lin, P.H., Wong, D.S.H., 2014. Carbon dioxide capture and regeneration with amine/ alcohol/water blends. Int. J. Greenh. Gas Control 26, 69e75. Liu, Q., Shi, J., Zheng, S., Tao, M., He, Y., Shi, Y., 2014. Kinetics studies of CO2 adsorption/desorption on amine-functionalized multiwalled carbon nanotubes. Ind. Eng. Chem. Res. 53 (29), 11677e11683. MacDowell, N., Florin, N., Buchard, A., Hallett, J., Galindo, A., Jackson, G., Adjiman, C.S., Williams, C.K., Shah, N., Fennell, P., 2010. An overview of CO2 capture technologies. Energy Environ. Sci. 3 (11), 1645. Mao, X.Q., Zeng, A., Hu, T., Xing, Y.K., Zhou, J., Liu, Z.Y., 2014. Co-control of local air pollutants and CO2 from the Chinese coal-fired power industry. J. Clean. Prod. 67, 220e227. Mazari, S.A., Si Ali, B., Jan, B.M., Saeed, I.M., Nizamuddin, S., 2015. An overview of solvent management and emissions of amine-based CO2 capture technology. Int. J. Greenh. Gas Control 34, 129e140. Molina, C.T., Bouallou, C., 2015. Assessment of different methods of CO2 capture in post-combustion using ammonia as solvent. J. Clean. Prod. 103, 463e468. Moser, P., Schmidt, S., Sieder, G., Garcia, H., Stoffregen, T., 2011. Performance of MEA in a long-term test at the post-combustion capture pilot plant in Niederaussem. Int. J. Greenh. Gas Control 5 (4), 620e627. Perevertaylenko, O.Y., Gariev, A.O., Damartzis, T., Tovazhnyanskyy, L.L., Kapustenko, P.O., Arsenyeva, O.P., 2015. Searches of cost effective ways for amine absorption unit design in CO2 post-combustion capture process. Energy 90, 105e112. Rivera-Tinoco, R., Bouallou, C., 2010. Comparison of absorption rates and absorption capacity of ammonia solvents with MEA and MDEA aqueous blends for CO2 capture. J. Clean. Prod. 18 (9), 875e880. Sch€ affer, A., Brechtel, K., Scheffknecht, G., 2012. Comparative study on differently concentrated aqueous solutions of MEA and TETA for CO2 capture from flue gases. Fuel 101, 148e153.  ˛ zko,  tki, A., Scia _ Stec, M., Tatarczuk, A., Wie˛ cław-Solny, L., Kro M., Tokarski, S., 2015. Pilot plant results for advanced CO2 capture process using amine scrubbing at the Jaworzno II power plant in Poland. Fuel 151, 50e56. Tobiesen, F.A., Juliussen, O., Svendsen, H.F., 2008. Experimental validation of a rigorous desorber model for CO2 post-combustion capture. Chem. Eng. Sci. 63, 2641e2656. Vrachnos, A., Kontogeorgis, G., Voutsas, E., 2006. Thermodynamic modeling of acidic gas solubility in aqueous solutions of MEA, MDEA and MEAMDEA blends. Ind. Eng. Chem. Res. 45, 5148e5154. Vatopoulos, K., Tzimas, E., 2012. Assessment of CO2 capture technologies in cement manufacturing process. J. Clean. Prod. 32, 251e261. Zhang, X., Fu, K., Liang, Z., Rongwong, W., Yang, Z., Idem, R., Tontiwachwuthikul, P., 2014. Experimental studies of regeneration heat duty for CO2 desorption from diethylenetriamine (DETA) solution in a stripper column packed with Dixon ring random packing. Fuel 136, 261e267.