Performance evaluation on complex absorbents for CO2 capture

Separation and Purification Technology 82 (2011) 87–92

Contents lists available at SciVerse ScienceDirect

Separation and Purification Technology journal homepage: www.elsevier.com/locate/seppur

Performance evaluation on complex absorbents for CO2 capture Jian-Gang Lu ⇑, Ai-Chun Hua, Lan-Lan Bao, Shi-Xin Liu, Hui Zhang, Zheng-Wen Xu Jiangsu Key Laboratory of Atmospheric Environment Monitoring and Pollution Control, Nanjing University of Information Science and Technology, Nanjing 210044, PR China

a r t i c l e

i n f o

Article history: Received 3 May 2011 Received in revised form 29 July 2011 Accepted 24 August 2011 Available online 7 September 2011 Keywords: CO2 capture Methylmonoethanolamine Phosphate Piperazine Bubble column

a b s t r a c t Piperazine (PZ) and phosphates as additives were added into an aqueous N-methylmonoethanolamine (MMEA) to form complex absorbents, respectively. Performances of CO2 capture by the complex absorbents were evaluated in a bubble column reactor. Reaction mechanisms and activations of the additives were presented theoretically. Effects of type and concentration of additives, and gas flowrates on volumetric mass transfer coefficient were investigated, and effects of orifice size of the gas sparger and stirring rates on average absorption velocity were also discussed. Results show that CO2 loadings of MMEA–PZ and MMEA–K3PO4 complex absorbents were larger than that of single MMEA and MEA absorbents. The MMEA–PZ complex absorbent gave a highest CO2 loading in all complex absorbents. The overall mass transfer coefficient increased, subsequently reached a maximum and then decreased with the increase of K3PO4 concentration in the complex absorbent. The overall mass transfer coefficient increased with the increase of the gas flow rates. Average absorption velocities increased with the decrease of the orifice size and with the increase of the orifice numbers. The average absorption velocities in moderate intensity of stirring rates were higher than that in the high intensity of stirring rates. Ó 2011 Elsevier B.V. All rights reserved.

1. Introduction Greenhouse gases, including carbon dioxide, methane, nitrous oxide, ozone, water vapor, hydrofluorocarbons (HFCs), perfluorocarbons (PFCs) and sulfur hexafluoride (SF6), can significantly contribute to global warming. It has been known that global warming resulted in serious global environmental problems [1]. Various climate models estimated that average global temperature may rise by about 1.4–5.8 °C by the year 2100 because of the enhanced greenhouse effect [2]. Carbon dioxide (CO2), accounting for over 50% in the amount of greenhouse gases and currently responsible for over 60% of the enhanced greenhouse effect, is the primary species of greenhouse gases. It has turned to be a worldwide issue to reduce CO2 emission and decrease CO2 concentration in the atmosphere. The bulk of the CO2 emission comes mostly from the fossil fuel-based energy industries [3], such as coal-combusted power generating and petroleum and metallurgy processing that are the foundations of sustaining economic growth. The capture of CO2 from the industrial sources seems to be an important measure for this issue. Therefore, low energy-consumption, available, efficient technologies have attracted significant attention for the capture and removal of CO2 from gas mixtures produced by industrial sources. Current separation and capture technologies based on a variety of physical and chemical processes include absorption, adsorption, conversion, cryogenic separation, and membrane techniques. Such ⇑ Corresponding author. Tel.: +86 25 5873 1090; fax: +86 25 5873 1089. E-mail address: [email protected] (J.-G. Lu). 1383-5866/$ - see front matter Ó 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.seppur.2011.08.029

processes have been used in the chemical industry and others. The chemical absorption is an efficient method for CO2 capture, especially for capture of low concentration CO2 in gas source [4–6]. Amine is one of efficient absorbents, which has caught more attention from researchers. Aqueous alkanolamines (mono-, di-, or tri-ethanolamine), such as monoethanolamine (MEA), 2-amino2-methyl-1-propanol (AMP), diethanolamine (DEA), di-isopropanolamine (DIPA) and methyldiethanoamine (MDEA) are widely used in petrolic and chemical industries for acid gases removal. Recently, the complex absorbents (or blended absorbents) composed of two single absorbents (e.g. alkanolamines) with varying compositions have been used to improve the CO2 adsorbing performances [7–10]. The principle of formulation is to combine the favorable characteristics of different absorbents, for instance, higher absorbing capacity, higher absorbing velocity, lower energy consumption, reduced corrosion, and lower oxidative degradation; and simultaneously their unfavorable characteristics are suppressed [10]. Therefore, the complex absorbents combine the advantages of single absorbents. As a rule, one component (i.e., a solute) in the complex absorbent is as an additive (or so-called an activating agent) to be added into another component as a main solvent. The main solvent is usually large in varying compositions of concentration compared with the additives. Product costs of the additives are generally higher than the main solvent. For example, piperazine (PZ) or 2-amino-2-methyl-1-propanol (AMP), as an additive, is added into methyldiethanolamine (MDEA) as a main solvent to form the MDEA-based blended absorbent [9]. Bubble columns have been widely used in industry for carrying out a variety of chemical reactions such as absorption and purification [11].

88

J.-G. Lu et al. / Separation and Purification Technology 82 (2011) 87–92

Nomenclature a C d D E G H Ha K k kLa k0L

specific interfacial area (m1) concentration (kmol m3 or mol L1) bubble diameter (m or mm) diffusion coefficient (m2 s1) enhancement factor for chemical reaction gas flow rate (m3 s1 or L min1), Henry constant Hatta number overall mass-transfer coefficient (m s1) individual mass-transfer coefficient (m s1), or reaction rate constant (m3 mol1 s1 or s1) volumetric mass transfer coefficient (s1) physical mass transfer coefficient (s1)

Because they can offer favorable performances such as a simple structure without moving parts, high gas/liquid contact areas, good mass/heat transfer rates, and large liquid hold-ups [12]. Mass transfer in bubble columns has been extensively investigated [13–14]. MMEA is a kind of the secondary amine, which molecule structure is shown in Fig. 1. It was selected as a main solvent for the complex absorbents because it is commercially available, low product cost and lower corrosiveness due to weaker alkalinity compared with MEA. The reaction schemes of CO2 captured by various alkanolamines such as MEA, DEA, TEA, MDEA, diglycolamine (DGA), DIPA, and AMP, have been proposed in literature [15–16]. In this work, MMEA blended with other alkanolamines such as bis-amino amine PZ (its molecule structure is shown in Fig. 1) or phosphates was used as CO2 absorbents to form novel composite solutions, which was expected to have a higher CO2 reaction rate and loading capacity. Therefore, the alkanolamine composite solution would result in substantial lower solution circulation rates while it was compared with a unique amine solution. Both lower circulation rate and resulted lower pumping energy cost would lead to a reduction of regeneration energy requirement. In this work, Piperazine (PZ) and phosphates as additives were added into the aqueous MMEA to form the MMEA-based complex absorbents, respectively. CO2 capture performances of the complex absorbents were evaluated in a bubble column reactor. Reaction mechanism and activation of the additives were schemed theoretically. Effects of type and concentration of additives, and gas flowrate on absorption performance were investigated, and effects of orifice size and stirring rates on average absorption velocity were also discussed. 2. Theory section 2.1. Reaction scheme and mechanism Zwitterion mechanism is the recently most widely accepted mechanisms for primary and secondary amines reacting with CO2 [10]. Following reactions occur in the solution:

N

L  R t

CO2 loading of liquid phase (mol L1) average absorption velocity (mol L1 s1) time (s)

Subscripts A gas component a additive g gas phase L liquid phase ov overall Greek letter e gas hold-up

R1 R2 NH þ CO2 R1 R2 NHþ COO R1 R2 NHþ COO þ B R1 R2 NCOO þ BHþ

ð1Þ ð2Þ

where B is a base present in the solution. The zwitterionic mechanism indicates that, an amine reacts with CO2 to form a zwitterion, successively a base (B) makes the zwitterion deprotonate to form carbamate. B could be R1 R2 NH;OH ; or H2 O. In addition, CO2 also reacts with the hydroxide ions present in solution:

CO2 þ OH  HCO3

ð3Þ

The contribution of reaction of Eq. (3) to the overall reaction rate can be considered negligible. On the one hand, alkanolamines are weak bases [6]. They hydrolyze to create a little OH ion. On the other hand, a great deal of molecule is R1R2NH in the solution. CO2 collides with R1R2NH rather than OH. Hereby, reaction of Eq. (3) was negligible in the contribution to the overall reaction rate. Amines protonation, bicarbonate and carbonate formation, and water dissociation are also negligible as a result of they are not dominant in the aforementioned system. 2.2. Activation scheme and mechanism When phosphates as additives are added into the aqueous MMEA, following multilevel hydrolytic reactions would occur in the solution. 2  PO3 4 þ H2 O HPO4 þ OH

ð4Þ

  HPO2 4 þ H2 O H2 PO4 þ OH

ð5Þ

H2 PO4 þ H2 O H3 PO4 þ OH

ð6Þ

Phosphates and their hydrolytes in the solution could serve as the B. Accordingly, the additives would impact the reaction 2 between MMEA and CO2. B could be PO3 or OH . 4 ; HPO4 ; As PZ as an additive is added into the aqueous MMEA, the following reactions could occur in the solution:

ð7Þ ð8Þ

CH3 NH CH 2 CH 2 OH ð9Þ

N N-methylmonoethanolamine (MMEA)

piperazine (PZ)

Fig. 1. Molecule structures of MMEA and PZ

ð10Þ

89

J.-G. Lu et al. / Separation and Purification Technology 82 (2011) 87–92

PZ can act as both reactant with CO2 and the B, and its reaction products could also serve as the B. Therefore, phosphates and PZ in aqueous MMEA would homogeneous catalyze and accelerate the reaction between MMEA and CO2.

was continuous with respect to the gas phase and batch as regards the absorbent liquid. Operation conditions for general evaluation experiments were gas flowrate of 1 L/min, orifice diameter of 1 mm and static liquid (no stirring).

3. Experimental section

3.3. Analytical method and data processing

3.1. Materials

The total amount of alkalinities was determined by titration with a standard acid solution using methyl orange as an indicator. The CO2 content in the solution was determined by the method of chemical analysis. A known volume of the liquid sample was acidulated with a diluted H2SO4 aqueous solution (volume ratio of H2SO4:H2O was 1:4), and the volume of the evolved gas was measured with a gas burette. After temperature and pressure corrections, the CO2 content of the liquid sample was calculated. The gas burette precision was ±0.02 mL. CO2 loading of liquid phase, LCO2 and volumetric mass transfer coefficient, kLa are adopted to evaluate the adsorption performances of the complex absorbents. The LCO2 is the amount of CO2 dissolved in an absorbent (including physical and chemical absorption) and defined as CO2 moles per liter absorbent. kLa is one of the most important parameters that express absorption performances of bubble reactors. In the gas–liquid mass transfer process that is enhanced by chemical reactions, the total resistance (1/K) to mass transfer is modeled as the sum of gas film (1/kg) and liquid film resistance (1/kL):

In this work, CO2 was commercial cylinder gas with purity more than 99.99% (Nanjing Real Special gas Co., China). MMEA and MEA were analytical reagents (Weixing Biochemical Reagent Co. Ltd., China). K3PO4, K2HPO4, KH2PO4 and H2SO4 were analytical reagents from China Medicament Group Shanghai Chemical Reagent Co. PZ with purity more than 99% was purchased from China Medicament Group Shanghai Chemical Reagent Co. Concentrations of MMEA and various complex absorbents were 0.5 and 0.4 mol/L MMEA + 0.1 mol/L additives (PZ or phosphates), respectively. 3.2. Experimental unit and procedure Fig. 2 shows the schematic flow chart of experimental unit. The evaluation experiments were performed in a cylindrical bubble column absorber (5) with dimension of U60  200 mm (internal diameter  height) which was made of polymethacrylate. In the cylindrical column a gas sparger with a porous cross-type tube (7) was set up in order to disperse uniform bubbles into the liquid phase at room temperature (19–22 °C). There were three crosstype tubes with orifice diameters of 1, 2, and 3 mm, respectively. Pure CO2 gas was used in evaluation experiments. The flowmeter (2) was calibrated beforehand. The foam-film flowmeter (6) at the gas outlet of the bubble column (5) was used as mensuration of outlet gas flowrate and calculation of mass balance. The buffer (3) and water saturator (4) were used to stabilize and humidify the CO2 stream, respectively. A magnetic stirrer (8) (Shanghai Precision Instrument Co. Ltd., China) under the bubble column could be used to stir the solution. The solution (i.e. absorbent) was prepared with deionized water to a given concentration in the column. CO2 stream passing through a gas flowmeter (2) (LZB-II, Yutao Automation Instrument Co. Ltd., Zhejiang, China) under a desired flow rate, was fed into the buffer (3), and then into water saturator (4) at a pressure slightly higher than atmospheric pressure. Subsequently, CO2 stream was bumbled into the bubble column reactor (5) by the gas sparger (7), where CO2 was absorbed by 250 mL liquid absorbent. The unabsorbed CO2 was then released from gas outlet on the top of the column reactor. The operational regime

Vent

6

2

9 7 P

3

4 5

1 8

1 H 1 ¼ þ K kg kL

where K, kg and kL are overall mass transfer coefficient, gas and liquid phase mass transfer coefficients, respectively. H is Henry constant. Gas film resistance is negligible as a result of the pure CO2 gas used in gas phase. Therefore, 1/K = 1/kL, namely K = kL, kLa could be calculated by the following expressions:

Ka ¼ kL a

ð12Þ

0

kL a ¼ EkL a

ð13Þ 0 kL

where E is the enhancement factor and is physical mass transfer coefficient. Taking into account the operation regime employed in this 0 work, mathematical expression of the kL a for physical absorption is given in Eq. (14).

dC 0 ¼ kL aðC   CÞ dt

ð14Þ

where C⁄ and C are the physical solubility and carbon dioxide concentration in the liquid phase, respectively. t is operation time. The reaction of CO2 with aqueous alkalinities or activated alkalinities can be treated as a fast pseudo-first-order reaction [17]. An approximate enhancement factor would be [18]:

E¼

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1 þ Ha2

ð15Þ

where the Hatta number, Ha, is shown in Eq. (16).

Ha ¼

pffiffiffiffiffiffiffiffiffiffiffiffi kov DA 0

kL

ð16Þ

where the kov is the overall reaction kinetic constant. DA is the gas diffusivity. kov has the following expression:

kov ¼ k2 C MMEA þ ka C a Fig. 2. Schematic diagram of experimental unit for CO2 capture in a bubble column reactor (1) gas cylinder, (2) flow meter, (3) buffer, (4) water saturator, (5) bubble reactor, (6) foam film flowmeter, (7) gas sparger, (8) magnetic stirrer, (9) liquid sampler.

ð11Þ

ð17Þ

k2 and ka are reaction rate constants for MMEA and additives, respectively. CMMEA and Ca are the bulk MMEA and additive concentrations, respectively. The dC/dt can be obtained experimentally

J.-G. Lu et al. / Separation and Purification Technology 82 (2011) 87–92

4. Results and discussion

18 16 14 -1 4

between two consecutive measurements dt = t2  t1 (dt = 15 min in this work). The value of k2 was obtained from the literature [18] and the parameter ka was obtained from the literature [19–20]. The 0 value of kL was obtained from the literature [21]. The values of C⁄ and DA were estimated from the literature [22]. Hence, the values of kLa for chemical absorption can be obtained combining Eqs. (11)–(17).

kLa 10 (s )

90

12 10 8 6

4.1. Comparison of CO2 loadings of complex absorbents

4 In order to evaluate the performances of complex absorbents for CO2 capture and better understand gas/liquid mass transfer characteristics of bubble column reactor, experiments for determination of CO2 loadings of liquid phase were carried out using single MMEA and MEA, and various complex absorbents. Determination experiment was terminated as a constant CO2 loading of the liquid sample appeared and the outlet gas flowrate equal to the inlet was indicated. Experiment data are given in Fig. 3. Results show that, LCO2 of various absorbents rapidly increased and then reached to an equilibrium with absorption time. Absorbent MMEA + PZ exhibited a highest LCO2 up to 0.49 mol/L among all absorbents. The LCO2 were in the order of (MMEA + PZ) > (MMEA + K3PO4) > (MMEA + K2HPO4) > MMEA > MEA > (MMEA + KH2PO4) under the same total concentration of 0.5 mol/L. The LCO2 of MMEA–PZ and K3PO4 complex absorbents were larger than that of single MMEA, indicating that the performance of a single absorbent could be improved by additives. This can be explained by reaction mechanism. PZ cannot only serve as the B to accelerate the reaction of MMEA with CO2 (Eqs. (1) and (2)), but also directly react with CO2 as an amine [19]. Among phosphates, The MMEA–K3PO4 complex absorbent gave the highest LCO2 . LCO2 of MMEA–K2HPO4 was slightly larger than that of the single MMEA. The LCO2 of single MMEA was slightly larger than that of the single MEA. However, LCO2 of MMEA–KH2PO4 was lower than that of the single MMEA and MEA. The fact is that the hydrolyzation of PO3 4 was much eas 3 ier than that of HPO2 and H PO , because k of PO 2 b 4 4 4 is far higher 2  than that of HPO4 and H2 PO4 . The ion concentration of OH pro2  duced from PO3 4 is far higher than that from HPO4 and H2 PO4 .

2 0.0

0.1

0.2

0.3

0.4

0.5

-1

CK PO (mol L ) 3

4

Fig. 4. Effect of additive concentration on overall mass transfer coefficient.

Under the constant overall concentration of 0.5 mol/L, the concentration of K3PO4 in the complex absorbent can be obtained by altering the proportion of K3PO4 and MMEA, i.e. the concentrations of (K3PO4 + MMEA) in mol/L were 0 + 0.5, 0.1 + 0.4, 0.2 + 0.3, and 0.3 + 0.2, and so on. As Fig. 4 shows that, the overall mass transfer coefficient increased with increasing the concentration of K3PO4 until a maximum value was reached at a K3PO4 concentration of about 0.26 mol/L, subsequently decreased with further increasing of K3PO4 concentration. As the concentration of K3PO4 in the complex absorbent increased or the concentration of MMEA decreased, the concentration of hydroxyl ion increased from the phosphate hydrolyzation. Therefore, the reaction could be expressed as follows:

CO2 þ OH HCO3

ð18Þ

However, its velocity is far less than that of the reaction given by Eq. (1) [23]. As a result, the overall mass transfer coefficient decreased when K3PO4 concentration was more than 0.26 mol/L. This shows that only a certain amount of additive could effectively promote the complex absorbent for CO2 capture. The effect of the additive is essentially identical to the homogeneous catalysis.

4.2. Effect of additive concentration on overall mass transfer coefficient 4.3. Effect of gas flow rate on overall mass transfer coefficient The effect of the additive K3PO4 concentration in the complex absorbent on the overall mass transfer coefficient was studied.

0.6

0.4

-1

LCO (mol L )

0.5

0.3 0.5M MMEA 0.4M MMEA+0.1MPZ 0.4M MMEA+0.1MK 3PO4

0.1

0.4M MMEA+0.1MK 2HPO4

2

0.2

0.4M MMEA+0.1MKH 2 PO4

0.0

Gas flow rate is one of the most important operating conditions of bubble column reactor. It has a very significant effect on gas– liquid mass transfers. The effect of the gas flow rate on the overall mass transfer coefficient in the bubble column reactor was experimentally investigated using concentration of 0.25 mol/L MMEA + 0.25 mol/L K3PO4 As shown in Fig. 5, the overall mass transfer coefficient increased with the increase of the gas flow rate. The fluid state in the reactor varied with the increase of gas flow rates. Reynolds number increased with the increase of the gas flow rates. The turbulence degree trended to be enlarged and the boundary layer thicknesses of gas and liquid phase in the gas– liquid interface trended to diminish. The mass transfer resistance then decreased and the overall mass transfer coefficient increased accordingly. This observation was in accord with that of Kawase and Hashiguchi [24], whom indicated the dependence was kLa  G1.2.

0.5M MEA 0

1000

2000

3000

4000

5000

t (s) Fig. 3. CO2 loading as a function of absorption time for complex absorbents.

4.4. Effect of orifice parameter on absorption performance The process of bubble formation is governed by many factors such as operating parameters (e.g., gas flow rate through the

91

J.-G. Lu et al. / Separation and Purification Technology 82 (2011) 87–92

30

0

1.0

1.0

25

3mm

4

-1

G (L min )

-1

kLa 10 (s )

500 1000 1500 2000 2500 3000 3500 4000 4500

20

2mm 0.5

0.5 1mm

15

0

1

2

3

4

5

6

7

8

t1

0

-1

G (L min )

Fig. 7. Changes of outlet gas flowrates with operational time and determination of ultimate t.

orifice, mode of operation, and static condition of the liquid), system properties (e.g., orifice size), and the physicochemical properties (e.g., liquid viscosity, liquid density). Orifice parameters (e.g., orifice size and orifice number) are one of the most important key factors. Experiments were conducted to study effects of orifice parameters of the gas sparger on absorption performance using MMEA–PZ complex absorbent of 0.4 mol/L MMEA + 0.1 mol/L PZ. Three orifice sizes in 1, 2 and 3 mm of the gas spargers were used in these experiments, respectively. Parameters of orifice are listed in Table 1. Fig. 6 shows that LCO2 as a function of operational time. Results showed that absorption performance was significantly affected by orifice parameters. In order to characterize effects of orifice parameters on absorption performance, average absorption  is introduced and defined as (Eq. (19)): velocity, R

Lt;CO2  L0;CO2 DLCO2 ¼ t  t0 Dt

ð19Þ

Orifice number

1 mm 2 mm 3 mm

24 20 16

where Lt;CO2 , L0;CO2 and DLCO2 represent CO2 loadings at operational ultimate time t, initial time t0 and interval Dt, respectively. As a fresh absorbent is used, t0 = 0, L0;CO2 is equal to 0, and then Dt = t  t0 = t. t was determined by means of the foam film flowmeter. As the foam film flowmeter indicated the gas outlet flowrate G equal to the inlet, for instance, the value of ultimate t was noted.  calculations are given in Fig. 8. The Data of t are shown in Fig. 7. R results showed that average absorption velocities increased with the decrease of the orifice sizes and with the increase of the orifice numbers. Absorption velocity of the bubble column reactor primarily depends on the interfacial area of bubbles [25]. The specific interfacial area depends on the bubble size and the gas hold-up on the basis of correlation of Eq. (20) [26]

a¼

6e dð1  eÞ

ð20Þ

where a, d and e are the specific interfacial area, the bubble diameter and the gas hold-up, respectively. Therefore, a smaller orifice diameter could form a smaller bubble diameter. A larger orifice number could give a larger gas hold-up. Consequently, a larger interfacial area could be obtained.

Table 1 Parameters of orifice over the gas spargers. Orifice diameter

t3

t (s)

Fig. 5. Effect of gas flow rate on overall mass transfer coefficient.

4.5. Effect of stirring rate on average absorption velocity With an attempt to improve the contact between gas/liquid phases and to enhance the mass transfer, magnetic stirring was

2.2

0.5

2.0

Orifice number: 24

-1

R 10 (mol L s )

0.6

-1

0.4 0.3

1.8 Orifice number: 20

1.6

4

2

-1

LCO (mol L )

R¼

t2

0.0 500 1000 1500 2000 2500 3000 3500 4000 4500

0.0

9

0.2

1mm 2mm 3mm

0.1 0.0 0

500 1000 1500 2000 2500 3000 3500 4000 4500

t (s) Fig. 6. Effect of orifice parameters on CO2 loading.

1.4 Orifice number: 16

1.2 1.0 0

1

2

3

Orifice diameter (mm) Fig. 8. Effect of orifice parameters on average absorption velocity.

4

92

J.-G. Lu et al. / Separation and Purification Technology 82 (2011) 87–92

the orifice diameters and with the increase of the orifice numbers. The average absorption velocities increased in moderate intensity of the stirring rates, and decrease in the high intensity of the stirring rates.

2.6

2.2 Acknowledgments

-1

-1

R 10 (mol L s )

2.4

2.0

This work was supported by the University Natural Science Project of Jiangsu Province (09KJB610003), University Project of Industrial Promotion of Jiangsu Province (JH10-17) and the Undergraduate Training Program for Practical Innovation of NUIST (2011).

4

1.8 1.6 1.4

References

1.2 -500

0

500

1000

1500

2000

stirring rate (rpm) Fig. 9. Effect of stirring rate on average absorption velocity.

used in this study. Stirring as a mechanical mixing is widely used in the chemical and other industrial processes. Experiments were carried out for investigating effect of stirring rate on absorption performance, using MMEA–PZ complex absorbent of 0.4 mol/L MMEA + 0.1 mol/L PZ. Results are shown in Fig. 9. The average absorption velocities increased and then decreased with the increase of stirring rates. Stirring can change the hydrodynamics of the flow field such as the degree of turbulence, the bubble size and distribution as well as the rising velocity of the bubbles. The turbulence degree and Reynolds number increased with the increase of stirring rates. The mass transfer was enhanced and thus absorption velocity increased. Stirring induced the liquid revolving and prolonged the rising route of the bubble in this work. Therefore, gas duration time was increased. The mass transfer was enhanced, and hence absorption velocity increased. However, the average absorption velocities decreased in higher intensity of the stirring rates in comparison with that of moderate intensity of the stirring rates and still state, e.g., more than 1000 rpm in this work. Some bubbles collided each other and then got together into a big bubble in the case of high intensity of the stirring rates. Bubble coalescence led to a decrease of gas–liquid contact area. Some bubbles were elongated, deformed and broken up. These phenomena could be observed in the experiments. The phenomena essentially resulted in a decrease of interfacial area, and the absorption velocities decreased ultimately in the high intensity of stirring rates.

5. Conclusions In this work, performances of an amine complex absorbent of MMEA–PZ and an inorganic salt complex absorbent of MMEA– phosphates for CO2 capture were evaluated in a bubble column reactor. Experiments have been conducted to investigate effects of various factors such as CO2 loadings, additive concentrations, gas flow rates, orifice parameters and stirring rates on absorption performances. For a series of complex absorbents of MMEA–PZ and MMEA–phosphates, the MMEA–PZ complex absorbent gave the highest CO2 loading. CO2 loadings of the MMEA–PZ and MMEA– K3PO4 complex absorbents were larger than that of single MMEA and MEA. The overall mass transfer coefficient increased, subsequently reached a maximum and then decreased with the increase of K3PO4 concentration in the complex absorbent. It indicates that an additive added in small amounts rather than in large amounts into an amine absorbent could improve mass transfer. The overall mass transfer coefficient increased with the increase of the gas flow rates. Average absorption velocities increased with the decrease of

[1] H. Herzog, B. Eliasson, O. Kaarstad, Capturing greenhouse gases, Sci. Am. 182 (2000) 72–79. [2] M. Williams, Climate change: information kit. Geneva: the United Nations Environment Programme (UNEP) and the United Nations Framework Convention on Climate Change (UNFCCC), 2002. [3] D.J. Wuebbles, A.K. Jain, Concerns about climate change and the role of fossil fuel use, Fuel Process Technol. 71 (2001) 99–119. [4] L.-L. Lia, B.-H. Chen, Review of CO2 absorption using chemical solvents in hollow fiber membrane contactors, Sep. Purif. Technol 41 (2005) 109–122. [5] S.B.P. Mandal, A.K. Biswas, S.S. Bandyopadhyay, Absorption of carbon dioxide into aqueous blends of 2-amino-2-methyl-1-propanol and diethanolamine, Chem. Eng. Sci. 58 (1996) 4137–4144. [6] P. Singh, J.P.M. Niederer, G.F. Versteeg, Structure and activity relationships for amine based CO2 absorbents-I, Int J Greenhouse Gas Control 1 (2007) 5–10. [7] L.Q. Hua, Y. Shuo, T.J. Lin, A new complex absorbent used for improving propylene carbonate absorbent for carbon dioxide removal, Sep. Purif. Technol. 16 (1999) 133–138. [8] C.H. Liao, M.H. Li, Kinetics of absorption of carbon dioxide into aqueous solutions of monoethanolamine + N-methyldiethanolamine, Chem. Eng. Sci. 57 (2002) 4569–4582. [9] J.-G. Lu, L.-J. Wang, X.-Y. Sun, J.-S. Li, X.-D. Liu, Absorption of CO2 into aqueous solutions of methyldiethanolamine and activated methyldiethanolamine from a gas mixture in a hollow fiber contactor, Ind. Eng. Chem. Res. 44 (2005) 9230– 9238. [10] J.-G. Lu, Y.-F. Zheng, M.-D. Cheng, L. -J Wang, Effects of activators on masstransfer enhancement in a hollow fiber contactor using activated alkanolamine solutions, J. Membr. Sci. 289 (2007) 138–149. [11] W.D. Deckwer, Bubble Column Reactors, Wiley, New York, 1992. [12] H.A. Jakobsen, H. Lindborg, C.A. Dorao, Modeling of bubble column reactors: progress and limitations, Ind. Eng. Chem. Res 44 (2005) 5107–5151. [13] H. Chaumat, A.M. Billet-Duquenne, F. Augier, C. Mathieu, H. Delmas, Mass transfer in bubble column for industrial condition-effects of organic medium gas and liquid flow rates and column design, Chem. Eng. Sci. 60 (2005) 5930– 5936. [14] A. Kundu, E. Dumont, A.-M. Duquenne, H. Delmas, Mass transfer characteristics in gas–liquid–liquid system, Can. J. Chem. Eng. 81 (2003) 640–646. [15] B.P. Mandal, M. Guha, A.K. Biswas, S.S. Bandyopadhyay, Removal of carbon dioxide by absorption in mixed amines: modeling of absorption in aqueous MDEA/MEA and AMP/MEA solutions, Chem. Eng. Sci. 56 (2001) 6217–6224. [16] Y. Maham, T.T. Teng, L.G. Hepler, A.E. Mather, Volumetric properties of aqueous solutions of monoethanolamine mono- and dimethylethanolamines at temperatures from 5 to 80 °C., Thermochim. Acta 386 (2002) 111–118. [17] G.-W. Xu, C.-F. Zhang, S.-J. Xin, Y.-W. Wang, Kinetics study on absorption of carbon dioxide into solutions of activated methyldiethanolamine, Ind. Eng. Chem. Res. 31 (1992) 921–927. [18] R.J. Littel, G.F. Versteeg, W.P.M. Van Swaaij, Kinetics of CO2 with primary and secondary amines in aqueous solutions, Chem. Eng. Sci. 47 (1992). 2073–2045. [19] J.-G. Lu, Y. Ji, H. Zhang, M.-D. Chen, CO2 Capture using activated amino acid salt solutions in a membrane contactor, Sep. Sci. Technol. 45 (2010) 1240–1251. [20] J.-G. Lu, H. Zhang, Y. Ji, F. Fan, C. Liu, Amino acid-salt-based complex absorbents for CO2 capture in a membrane contactor, Energy Fuels 24 (2010) 4617–4626. [21] P.S. Kumar, J. Hogendoorn, G.F. Versteeg, P.H.M. Feron, Kinetics of the reaction of carbon dioxide with aqueous potassium salt solutions of taurine and glycine, AIChE J. 49 (2003) 203–213. [22] G.F. Versteeg, W.P.M. Van Swaaij, Solubility diffusivity of acid gases (CO2 N2O) in aqueous alkanolamine solutions, J. Chem. Eng. Data 33 (1988) 29–34. [23] B.P.M. Lauwhoff, G.F. Versteeg, W.P.M. van Swaaij, A study on the reaction between CO2 and alkanolamines in aqueous solutions, Chem. Eng. Sci. 39 (1984) 207–225. [24] Y. Kawase, N. Hashiguchi, Gas-liquid mass transfer in external-loop airlift columns with Newtonian and non-Newtonian fluids, Chem. Eng. J. Biochem. Eng. J 62 (1996) 35–42. [25] A.A. Kulkarni, J.B. Joshi, Bubble formation and bubble rise velocity in gas– liquid systems: a review, Ind. Eng. Chem. Res. 44 (2005) 5873–5931. [26] K.V. Van’t Riet, J. Tramper, Basic Bioreactor Design, Marcel Dekker, New York, 1991.