International Journal of Greenhouse Gas Control 44 (2016) 115–123
Contents lists available at ScienceDirect
International Journal of Greenhouse Gas Control journal homepage: www.elsevier.com/locate/ijggc
Performance and reaction kinetics of CO2 absorption into AMP solution with [Hmim][Gly] activator Zuoming Zhou, Bingsong Guo, Bihong Lv, Hongxiu Guo, Guohua Jing ∗ College of Chemical Engineering, Huaqiao University, Xiamen, Fujian, 361021, China
a r t i c l e
i n f o
Article history: Received 27 April 2015 Received in revised form 13 October 2015 Accepted 10 November 2015 Available online 7 December 2015 Keywords: Carbon dioxide Mechanism Kinetics 1-hexyl-3-methylimidazolium glycinate 2-amino-2-methyl-1-propanol
a b s t r a c t Absorption and regeneration of a new aqueous blend of 1-hexyl-3-methylimidazolium glycinate ([Hmim][Gly]) and 2-amino-2-methyl-1-propanol (AMP) were investigated. The physicochemical properties of [Hmim][Gly]-AMP mixed solution with varying molar ratios were obtained at different temperature, and the results demonstrated that the mixed solution had a low viscosity with a high solubility of CO2 . It was found that the reactivity of AMP was significantly enhanced with the addition of [Hmim][Gly], and the absorption of CO2 into the mixed solution was independent of temperature within a range of 298–323 K. The optimal regeneration temperature and time were 333 K and 60 min, respectively. The mixed solution could be regenerated efficiently after three cycles. Furthermore, the study of 13 C NMR spectrum indicated that the reaction between CO2 and the [Hmim][Gly]-AMP solution was zwitterionic mechanism. Based on zwitterionic mechanism and two-film model, the kinetics of CO2 absorption into the mixed solution was investigated in the temperature range of 298–323 K by employing a double stirred-cell absorber. The kinetics region was considered to be a fast pseudo-first order, and the second order reaction rate constant was calculated as follows: k2,[Hmim][Gly]−AMP = 1.57 × 107 exp(− 3088.92/T). © 2015 Elsevier Ltd. All rights reserved.
1. Introduction Large emissions of CO2 have caused negative effect on atmospheric environment. Many countries have done jointing effort to reduce CO2 emissions (Miller et al., 2010). Chemical absorption using alkanolamine aqueous solution is generally considered as a reliable technology for CO2 capture. Monoethanolamine (MEA), a primary amine, is widely used for post-combustion capture from power plant for its rapid reaction rate with CO2 and low molecular weight (Tan et al., 2011). However, the regeneration of MEA requires high energy due to the high absorption heat. Some studies also showed that the degradation rate of MEA increased as temperature rose (Zheng et al., 2012). Therefore, there is a highly requirement to develop new chemical absorbents. Ionic liquids (ILs), generally consist of large organic bulky asymmetric cation (e.g. quaternary ammonium, imidazolium, pyridiniumand phosphonium ions) and small symmetrical shaped anion. They are regarded as promising green absorbents for CO2
∗ Corresponding author at: Jimei Road 668, Xiamen, Fujian 361021, China. Tel.: +86 592 6166216; fax: +86 592 6162300. E-mail addresses:
[email protected] (Z. Zhou),
[email protected] (B. Guo),
[email protected] (B. Lv),
[email protected] (H. Guo),
[email protected] (G. Jing). http://dx.doi.org/10.1016/j.ijggc.2015.11.010 1750-5836/© 2015 Elsevier Ltd. All rights reserved.
absorption, due to many fascinating properties, such as negligible volatility, non-flammability, high chemical and thermal stability, excellent solubility for CO2 , easy to be recycled and fine structure and property tunability (Wu et al., 2011; Zhang et al., 2006, 2013). Compared with other ILs, amino acid ionic liquids (AAILs) are more biodegradable and biocompatible, even when they contain imidazolium cation. Meanwhile, amino acids (AAs) are easy to be obtained in large quantities at low cost (Sharma et al., 2012). In 2005, Ohno’s group (2005) prepared AAILs ([emim][AA]) from 20 natural AAs. Lu et al. (2011) synthesized three imidazolium AAILs absorbents ([bmim][Gly], [bmim][Ala], [bmim][Lys]) for CO2 absorption. Their results revealed that the absorption loading and the absorption rate of these AAILs were much larger than that of the original AAs solution. In appropriate conditions, the regeneration efficiency of these imidazolium AAILs almost reached 100% (Zhao et al., 2011). Since pure AAILs have a higher viscosity than the ordinary ILs and are not conducive to the actual absorption, the industrial application of them as CO2 absorbent is still unfeasible. Fortunately, most of the AAILs are soluble in water, and the viscosity reduces obviously in aqueous solution. Zhang et al. (2010) found that CO2 absorption into a new aqueous solution of tetramethylammonium glycinate [N1111 ][Gly] exhibited a high absorption capacity. The CO2 loading of the saturated absorption solution was increased from 0.169 to 0.601 mol CO2 /mol IL as the mass fraction of [N1111 ][Gly]
116
Z. Zhou et al. / International Journal of Greenhouse Gas Control 44 (2016) 115–123
Nomenclature N A P R T V L H* H H E Ta Tb VL C M M D E K kG kL k2 x
absorption rate of CO2 per unit interfacial area (mol m−2 s−1 ) gas–liquid interfacial area (m2 ) pressure (Pa) ideal gas law constant (kmol m−3 s−1 ) temperature (K) CO2 volumetric flow rate (m3 s−1 ) CO2 loading (mol CO2 /mol absorbent) Henry’s constant (atm) Henry’s coefficient (mol m−3 Pa−1 ) vaporization enthalpy cohesion energy density solution temperature boiling point molar volume (cm3 mol−1 ) concentration (mol L−3 or mol m−3 ) molecular weight (g mol−1 ) average molecular weight (g mol−1 ) diffusion coefficient (m2 s−1 ) enhancement factor equilibrium constant gas-phase physical mass transfer coefficient (mol m−2 s−1 Pa−1 ) liquid-phase physical mass transfer coefficient (m s−1 ) the second-order rate constant (m3 mol−1 s−1 ) mole fraction
Greek letters ˛ interaction coefficient regeneration efficiency (%) ı solubility parameter [(J cm−3 )1/2 ] viscosity of solution (mPa s) density of the pure [Hmim][Gly] (g mL−1 ) average molecular weight Subscripts solution a b boiling point n the nth absorption cycle i gas–liquid interface gas phase G L liquid phase m the mixture of aqueous [Hmim][Gly] solution
in the solution decreased from 100% to 30%. Our previous work (Jing et al., 2012) found that the absorption rate of CO2 into 30% [N1111 ][Gly] solution was much higher than that of 30% TEA and 30% DEA, and only a little bit smaller than that of 30% MEA. It should be noted that high concentrations of ILs aqueous solution is not suitable for industrial application due to their relatively high viscosity and cost. In the light of the methods and theories of mixed amines (Benamor and Aroua, 2005), using ILs as active components of alkanolamine solution becomes a possible and potential way to re-formulate the CO2 absorbents (Chinn et al., 2005; Wang et al., 2009). These mixed solvents decreased the regeneration energy and improve the physical properties of the solvents (e.g. densities, viscosities, corrosion and volatility), which increased CO2 absorption capacity accordingly. Zhang et al. (2010) used the mixed solution of [N1111 ][Gly]-MDEA to capture CO2 . The results indicated that [N1111 ][Gly] could obviously increase the absorption rate of CO2 in MDEA aqueous solution. Compared with
MDEA, the sterically hindered amine AMP has the same high equilibrium capacity for CO2 , but a higher reaction rate constant (Saha et al., 1995) and a lower regeneration energy cost (Mandal et al., 2001, 2003). Therefore, in our previous work (Zhou et al., 2012), [N1111 ][Gly] was mixed with AMP aqueous solution to form a new solvent for CO2 absorption. The results indicated that [N1111 ][Gly] promoted the absorption of CO2 in 0.95 M (mol L−1 ) AMP aqueous solution. In our previous work, the second-order reaction rate of CO2 absorption into [Hmim][Gly] was found to be k2 = 3.04 × 107 exp(−3050/T) (Guo et al., 2013), while that of [N1111 ][Gly] was found to be k2 = 4.6930 × 105 exp(−1.856/T) (Jing et al., 2012). Thus, the absorption rate of CO2 into [Hmim][Gly] solution was higher than that of the [N1111 ][Gly] solution. Hence the mixed solution of [Hmim][Gly]-AMP seems to be worth exploring for CO2 capture. In this work, a mixed aqueous solution consisted of [Hmim][Gly] and AMP was proposed for CO2 absorption. The performance of CO2 absorption and regeneration with the new absorbent were investigated at different mole ratio of [Hmim][Gly]-AMP, reaction temperature, regeneration temperature and regeneration time. The mechanism of CO2 capture into this new absorbent was proposed based on 13 C NMR spectra analysis, and the kinetic rate parameters were determined by a double stirred-cell absorber based on two-film model. 2. Experimental and methods 2.1. Chemicals [Hmim][Gly] (97%) was obtained from Shanghai Cheng Jie Chemical Co. LTD. China. AMP was purchased from J&K Scientific Ltd. Aqueous solution was prepared with distilled water. The gases were pure CO2 (>99.999% purity) and N2 (>99.999% purity) obtained commercially and used as received. 2.2. Absorption and regeneration methods A double stirred-cell absorber with a smooth gas–liquid interface was used as the absorption instrument, which was reported in detail in our previous work (Jing et al., 2012). The experiments were conducted with simulated flue gas, which mixed with 85% N2 and 15% CO2 of volume ratio. The gas flow rate was adjusted to 120 mL min−1 by mass flow controller. The stirring speed for gas phase was 250 rpm and for liquid phase was 130 rpm. The total concentration of the absorbent was 1.0 M. The absorption temperature was from 298 K to 323 K. The kinetics region was determined according to the judge method suggested by Levenspiel and Godfrey (1974). The absorption rate of CO2 can be calculated as follows: N=
P × (Vout − Vin ) ART
(1)
where A, P, R, T are gas–liquid interfacial area (m2 ), pressure (Pa), ideal gas law constant (kmol m−3 s−1 ), temperature (K), respectively. Vin , Vout are volumetric flow rate of inlet, outlet CO2 (m3 s−1 ), respectively. The CO2 saturated loading in the [Hmim][Gly]-AMP mixed solution was measured by a bubbling absorber, which had used in our previous work (Guo et al., 2013). The absorption took place in a flask with 100 mL mixed solution. The simulated flue gas flowed through a soap bubble meter to measure the flow rate after heated to the desired temperature of 303 K by a heating oven. The volumetric flow rate of the exit gas was also determined by another soap bubble meter. The CO2 loading was obtained by calculating the integral characteristic value of absorption rate to time, using Origin 8.0. Regeneration experiments were investigated under 0.5 kPa vacuum for a certain time by using a rotary evaporator to obtain
Z. Zhou et al. / International Journal of Greenhouse Gas Control 44 (2016) 115–123
the optimal regeneration temperature and regeneration time. After regeneration, the repeated absorption was performed under the same conditions with the barren solution. The regeneration efficiency can be calculated as: =
Ln L1
2.3.
13 C
the pure solvents and of the nonideality of the solute-free solvent mixture as described by Eq. (8).
˛jk ≈
2.4. Physicochemical data measurement and calculation 2.4.1. Density, viscosity and pH The density, viscosity and pH of the [Hmim][Gly]-AMP solution were measured by acidity meter (pHS-25), Anton Paar Density meter (DMA 4500) and digit display viscometer (NDJ-5S), respectively. 2.4.2. Solubility of CO2 in the mixed solution of [Hmim][Gly]-AMP The solubility of CO2 in the pure [Hmim][Gly] was calculated by regular solution theory (Scovazzo et al., 2004), which was reported in our previous works (Guo et al., 2013). Moreover, the Henry’s constant (H* ) and the solubility parameter (ı1 ) of CO2 in [Hmim][Gly] could be estimated. The Henry’s coefficient (H, mol Pa−1 m−3 ) of CO2 in the pure AMP was determined by “N2 O analogy” method (Versteeg and Van Swaaij, 1988), the expression as follows: H2,4 = HN2 O,4 ×
H2,3 HN2 O,3
(3)
Subscript 1, 2, 3, 4 stand for IL, CO2 , water and AMP, respectively. HN2 O,4 , HN2 O,3 , H2,3 are determined by the expressions:
1205.2
HN2 O,4 = 8.648 × 104 exp −
T
2284
HN2 O,3 = 8.55 × 106 exp −
T
2044
H2,3 = 2.82 × 106 exp −
T
(4)
(5)
(6)
The data should be transformed between Henry’s coefficient and Henry’s constant (Pa). At the beginning of CO2 absorption, the following expression can be used: H≈
106 · M · H∗
(7)
(g cm−3 ), M (g mol−1 ) denote the average density and average molecular weight of the absorbent. O’connell and Prausnitz (1964) deduced the thermodynamic of H* in mixed solvents based on regular solution theory. They considered H* in the mixed solvent as a function of the nonideality H* in
n−1 n
∗ xj ln H2,j −
j=1
j=1 k>j
/ 2 j=
/ 2k= / 2 j=
˛jk xj xk
(8)
x stands for the mole fraction, subscript m for the mixed solution of [Hmim][Gly]-AMP, and j, k for the different components in the mixed solution. ˛ describes the interaction coefficient between the two different solvents, and ˛jk can be estimated by a simplified form of Hildebrand’s equation (O’connell and Prausnitz, 1964):
NMR analysis
To gain a further insight into the absorption/desorption mechanism, 0.5 mL of the absorption samples of solvents before and after CO2 absorption were characterized by 13 C NMR (Bruker AVIII500 MHz), using an internal standard of 0.1 mL D2 O for the deuterium lock.
n
∗ ln H2,m =
(2)
is the regeneration efficiency (%); L1 and Ln are the saturated absorption loading (mol CO2 /mol absorbent) for the first cycle of the fresh solution and n cycle of the regenerated barren solution, respectively.
117
ıj + ık
2
VJL + VkL
(9)
2RT
By definition, the solubility parameter of AMP can be calculated by Eq. (10):
ı4 =
E
1/2
=
V4L
H − RT
1/2
V4L
(10)
H stands for the vaporization enthalpy of AMP, and the value is 47.05 kJ mol−1 at the boiling point of 400 K obtained from SciFinder database. E stands for cohesion energy density. VL represents the molar volume (cm3 mol−1 ). The density of AMP was obtained according to the following equation (Xu et al., 1991). 4 = 0.9507 × 103 − 0.7102 × (T − 273) − 7.1246 × 10−4 × (T − 273)2
(11)
The solubility decreased with increasing solution temperature (Ta ) in the condition of |Ta − Tb | < 150 K, while Ta is below boiling point (Tb ), and the expression can be shown as follows: ıa = ıb
1.13 b
(12)
a
Therefore, the solubility of CO2 in the mixed solution can be obtained from Eq. (3) to Eq. (12). 2.4.3. Diffusivity of CO2 in the mixed solution of [Hmim][Gly]-AMP The diffusivity of CO2 in the [Hmim][Gly]-AMP solution can be calculated by the model of Danckwerts (1970), shown in Eq. (13).
D2,m m T = D2,3 3 T = const
(13)
Here, D denotes the diffusion coefficient (m2 s−1 ). denotes viscosity (mPa s). The value of 3 was reported in the literature (Korson et al., 1969). D2,3 can be calculated by the following equation (Korson et al., 1969). log D2,3 = −8.1764 +
712.5 2.591 × 105 − T T2
(14)
3. Results and discussion 3.1. The physicochemical properties of [Hmim][Gly]-AMP solution Table 1 shows the physicochemical value of [Hmim][Gly]-AMP solution with various components (0:1, 1:9, 2:8, 3:7) at different temperature (298–303 K). The pH, density and viscosity of the mixed solution were all increased with increasing [Hmim][Gly] concentration, but decreased as temperature increased. For each composition, there was a good linear relationship between ln(m ) and temperature (T), ln(m ) and (T − 273.15) in a temperature range of 298–323 K. A higher IL concentration means a higher density and
118
Z. Zhou et al. / International Journal of Greenhouse Gas Control 44 (2016) 115–123
Table 1 The physicochemical properties of [Hmim][Gly]-AMP solution. n1 :n4 *
T (K)
pH
m (g cm−3 )
m (mPa s)
ln m = a(T − 273.15) + b
ln m = aT + b
0:1
298 303 313 323
12.11 ± 0.01 11.95 ± 0.01 11.76 ± 0.02 11.57 ± 0.04
0.99645 ± 1.73E−5 0.99521 ± 5.77E−6 0.99137 ± 3.06E−5 0.98685 ± 8.89E−5
1.25 ± 0.00 1.10 ± 0.00 0.87 ± 0.00 0.70 ± 0.00
a = −0.0004b = 0.0165r2 = 0.9972
a = −0.0236 b = 7.35 r2 = 0.9989
1:9
298 303 313 323
11.42 ± 0.03 11.28 ± 0.01 11.08 ± 0.03 10.88 ± 0.02
0.99968 ± 6.10E−4 0.99833 ± 2.65E−5 0.9944 ± 5.77E−6 0.98992 ± 2.08E−5
1.34 ± 0.00 1.15 ± 0.00 0.90 ± 0.00 0.74 ± 0.00
a = −0.0004b = 0.0317r2 = 0.9965
a = −0.0236b = 7.3398r2 = 0.9929
2:8
298 303 313 323
11.45 ± 0.03 11.3 ± 0.01 11.12 ± 0.03 10.97 ± 0.04
1.00283 ± 1.00E−5 1.00109 ± 5.77E−6 0.99704 ± 1.00E−5 0.99228 ± 1.15E−5
1.36 ± 0.00 1.18 ± 0.00 0.93 ± 0.00 0.75 ± 0.00
a = −0.0004b = 0.0099r2 = 0.9940
a = −0.0236 b = 7.318 r2 = 0.9929
3:7
298 303 313 323
11.45 ± 0.02 11.27 ± 0.03 11.11 ± 0.03 10.91 ± 0.02
1.00545 ± 1.15E−5 1.00358 ± 5.77E−6 0.99942 ± 1.15E−5 0.99457 ± 1.15E−5
1.39 ± 0.00 1.24 ± 0.00 0.96 ± 0.00 0.77 ± 0.00
a = −0.0004b = 0.0066r2 = 0.9917
a = −0.0235b = 7.2163r2 = 0.9988
n1 , n4 stand for [Hmim][Gly] and AMP, respectively.
viscosity, which will decrease the solubility and diffusivity of CO2 in the mixed solution. The solubility of CO2 decreased and the diffusivity of CO2 increased with increasing temperature. The results were consistent with the predictions of Arrhenius equation and the thermodynamic diffusivity mechanism. 3.2. Absorption of CO2 into [Hmim][Gly]-AMP aqueous solution 3.2.1. Effect of [Hmim][Gly]-AMP mole ratio on CO2 absorption rate [Hmim][Gly] was added into AMP aqueous solution as an activator to form mixed absorbent. CO2 absorption into the mixed solution was investigated at three different molar ratio of [Hmim][Gly]-AMP (1:9, 2:8, 3:7) with a total concentration of 1.0 M at 303 K and atmospheric pressure. As shown in Fig. 1, the absorption rate and absorption amount of CO2 into the mixed solution were increased with increasing [Hmim][Gly] concentration at the same absorption time. Although the viscosity of [Hmim][Gly]-AMP aqueous solution also increased as [Hmim][Gly] concentration rose, the increasing rate was rather small (seen in Table 1). The results indicated that the promotive effect of [Hmim][Gly] concentration on CO2 absorption was more significant than that on the viscosity. With all the concentrations investigated, the absorption rates were fast at the first 15 min and decreased quickly due to the chemical absorption. Further increasing the reaction time resulted in a decrease of the absorption rate due to the saturation of the 2.1
3.2.3. Interaction effect of [Hmim][Gly] and AMP in the mixed solution As shown in Fig. 3, CO2 absorption into [Hmim][Gly]-AMP aqueous solution was compared with that into single AMP solution and [Hmim][Gly] solution, respectively. The absorption rate of 0.3 M [Hmim][Gly] was higher than that of 0.7 M AMP at the first 30 min. As increasing the reaction time, the absorption rate of [Hmim][Gly] was finally lower than that of 0.7 M AMP. However, the absorption 2.0
0.040
-3
0.032 1.2 0.024 0.9
0.016
0.6
0.008
0
30
60
90
120
150
180
210
0.000
Absorption time (min) Fig. 1. CO2 absorption into different molar ratio of [Hmim][Gly]-AMP solution at 303 K and atmospheric pressure.
1.6
-2
-1
0.048
Absorption amount (mol)
-2
1.5
298 K 303 K 313 K 323 K
1.8
1.4
-3
0.7 M AMP+0.3 M IL 0.8 M AMP+0.2 M IL 0.9 M AMP+0.1 M IL 1.0 M AMP
-1
Absorption rate (10 mol⋅m ⋅s )
3.2.2. Effect of temperature on the absorption In order to study the effect of temperature on CO2 absorption, [Hmim][Gly]-AMP solution with a molar ratio of 3:7 was selected for CO2 capture in a temperature range of 298–323 K. As shown in Fig. 2, temperature had little influence on the absorption rate in the investigated range. This result was similar to that of the system consisted of 15% [N1111 ][Gly] + 15% MDEA for CO2 absorption (Zhang et al., 2010). Our previous research found that temperature (298–323 K) could slightly affect the absorption of CO2 in 1.0 M [Hmim][Gly] solution (Guo et al., 2013). Weng et al. (2011) had revealed that temperature (298–323 K) had little influence on the absorption of CO2 in aqueous solutions of AMP or AMP + SG (sodium glycinate).
0.056
1.8
0.3
[Hmim][Gly]-AMP absorbent. The detailed discussion of the reaction would be shown in Section 4.
Absorption rate (10 mol.m .s )
*
1.2 1.0 0.8 0.6 0.4
0
30
60
90
120
Absorption time (min)
150
180
210
Fig. 2. The effect of temperature on absorption rate into [Hmim][Gly]-AMP aqueous of 3:7 mole ratio.
Z. Zhou et al. / International Journal of Greenhouse Gas Control 44 (2016) 115–123 0.064
2.4
0.7 M AMP+0.3 M IL 0.7 M AMP 0.3 M IL
0.048 0.040
1.2
0.032
0.9
0.024
0.6
0.016
0.3
0.008
-3
1.5
0
30
60
90
120
150
180
210
Absorption amount (mol)
1.8
0.056
-2
-1
Absorption rate (10 mol⋅m ⋅s )
2.1
0.0
119
0.000
Absorption time (min) Fig. 3. CO2 absorption into [Hmim][Gly]-AMP solutions at 303 K.
rate of 0.3 M [Hmim][Gly] or 0.7 M AMP was lower than that of their mixture. On the other hand, the absorption amount of 0.3 M [Hmim][Gly] + 0.7 M AMP, 0.3 M [Hmim][Gly], 0.7 M AMP solution were 0.04886, 0.0355, 0.0247 mol, respectively. The absorption amount of the mixed solution was less than that of the sum of each single solution. It could be seen that there was a weak negative interaction of [Hmim][Gly] and AMP on CO2 absorption in the mixed system. 3.3. Regeneration of [Hmim][Gly]-AMP saturated aqueous solution Besides absorption characteristics, regeneration performance is also an important criterion for absorbent selection, especially in regards to the energy penalty in the regenerator. Hence, 0.3 M [Hmim][Gly] + 0.7 M AMP was chosen to study the effect of regeneration cycle, regeneration temperature (298–323 K) and regeneration time on CO2 regeneration. 3.3.1. Effect of regeneration temperature on regeneration efficiency The mixed absorbent of 0.3 M [Hmim][Gly] + 0.7 M AMP was regenerated at temperature of 323–338 K. CO2 absorption loading of the mixed solution after regenerating at different temperature was compared with that of the fresh solution, and the results were shown in Fig. 4. The absorption rates of CO2 in all the solutions were rather fast in the first 90 min, and then decreased with increasing
Fig. 4. CO2 loading of 0.3 M [Hmim][Gly] + 0.7 M AMP under different regeneration temperature.
Fig. 5. CO2 loading of 0.3 M [Hmim][Gly] + 0.7 M AMP under different regeneration time.
reaction time until the aqueous solution was saturated. The results showed that the mixed solution could be well regenerated by thermal method. On the other hand, CO2 loading increased with the increase of the regeneration temperature. However, it was of interest to note that there was only 7.20% increase in the regeneration efficiency from 333 to 338 K. For energy-saving, the appropriate regeneration temperature was 333 K. 3.3.2. Effect of regeneration time on regeneration efficiency To explore the optimal regeneration time, the regeneration experiments were performed at four different regeneration time. As shown in Fig. 5, the CO2 saturated loading of the second absorption were 0.390, 0.463, 0.473, and 0.481 mol CO2 /mol absorbent, while the regeneration time were 30, 60, 90 and 120 min, respectively. The CO2 saturated loading rose very slightly over 60 min. Thus, the optimal regeneration time was 60 min, and the corresponding regeneration efficiency was 94.68%. 3.3.3. Effect of absorption/regeneration cycles on regeneration efficiency The mixed solution of [Hmim][Gly]-AMP was regenerated for three cycles at 333 K and 60 min. As illustrated in Fig. 6, the regeneration efficiency in turn was 96.73%, 95.30% and 94.07% for the first, the second and the third cycle regeneration, respectively. The
Fig. 6. Consecutive absorption and desorption of CO2 in recycled [Hmim][Gly]-AMP solution against time.
120
Z. Zhou et al. / International Journal of Greenhouse Gas Control 44 (2016) 115–123
regeneration efficiency of the solution after different regeneration cycles changed slightly, which revealed that the aqueous blend of [Hmim][Gly]-AMP was well recycled with a rapid absorption rate by thermal regeneration. There were two possible reasons for the slightly decrease in regeneration efficiency: the reaction of [Hmim][Gly] with CO2 produced a number of heat-stable salts, which were unregenerate under the regeneration conditions (Winyu et al., 2006); the volatilization and degradation of AMP in the thermal regeneration process (Zheng et al., 2012).
could be seen in Fig. 8. In the 13 C NMR spectra (Fig. 7(b)), after capturing CO2 , a new peak observed at 160 ppm was attributed to HCO3 − or CO3 2− . Therefore, the reaction mechanism between CO2 and AMP (RNH2 ) (where R = C(CH3 )2 CH2 OH) is different from that of unhindered amines as shown in the following equations: 2R NH2 + CO2 → R NHCOO− + RNH3 +
(15)
R NHCOO− + H2 O → R NH2 + HCO3 −
(16)
Overall reaction: 4. Absorption mechanism To gain a further insight into the absorption mechanism, the 13 C NMR spectrum of CO -treated solution was compared with the 2
initial ones in Fig. 7. The molecular structure and the types of carbon nuclei in the mixed solution visualized with the 13 C NMR technique
R NH2 + CO2 + H2 O → R NH3 + + HCO3 −
(17)
The carbamate of AMP is easily hydrolyzed into AMP and bicarbonate (Sartori and Savage, 1983). As shown in Fig. 7(b), the 13 C NMR spectrum of CO -saturated AMP solution did not con2 tain a new resonance that corresponding to the formation of a
Fig. 7. 13 C NMR (100 MHz, D2 O) spectra of aqueous solution. (a) AMP + H2 O; (b) AMP + H2 O + CO2 ; (c) [Hmim][Gly] + H2 O; (d) [Hmim][Gly] + H2 O + CO2 ; (e) AMP + [Hmim][Gly] + H2 O; (f) AMP + [Hmim][Gly] + H2 O + CO2 .
Z. Zhou et al. / International Journal of Greenhouse Gas Control 44 (2016) 115–123
121
Table 2 Effect of liquid-phase volume and liquid-phase transfer mass coefficient on CO2 absorption rate in the 0.3 M [Hmim][Gly] + 0.7 M AMP solution. N (10−3 mol m−2 s−1 )
t (min)
Fig. 8. Molecular structure and type of carbon nuclei in aqueous solution system.
carbamate carbonyl carbon. This result indicated that AMPcarbamate had been fully hydrolyzed. It was expected that CO2 reacted with a solution of AAILs through a zwitterionic mechanism similar to amines (Caplow, 1968; Danckwerts, 1979; Kumar et al., 2003), because of the similar functional groups. According to the zwitterionic mechanism and the 13 C NMR spectrum (Fig. 7(c) and (d)), CO2 reacted with [Hmim][Gly] via the formation of a zwitterion, subsequently deprotonated by a base present in the solution ([Hmim]+ was represented as R+ in the following): K,k2
H2 N-CH2 -COO− R+ + CO2 − OOC+ H2 N-CH2 -COO− R+ − −
+
− +
(18)
− +
−
OOC H2 N-CH2 -COO R + Bi OOCHN-CH2 -COO R + Bi H − +
− +
OOCHN-CH2 -COO R + H2 OH2 N-CH2 -COO R + HCO3
−
+
(19) (20)
where K and k2 are the equilibrium constant and the rate constant of the second order reaction. Bi is a base, which could be amine, OH− , or H2 O. Different from the AMP-carbamate, as shown in Fig. 7(d), the 13 C NMR spectrum contained a new resonance that corresponding to the formation of a carbamate carbonyl carbon. It showed that [Hmim][Gly]-carbamate did not thoroughly hydrolyzed. In the [Hmim][Gly]-AMP solution, [Hmim][Gly] could react quickly with CO2 to form zwitterions, which would transfer protons to AMP, therefore, the absorption rate of CO2 increased greatly compared with the AMP aqueous solution. During CO2 absorption, + H N-CH -COO− R+ was expected to be eliminated by plenty of 3 2 AMP. The interaction between the protonated and the unprotonated amines, as shown in Eq. (24), involved only a proton transfer and was considered to be instantaneous and at equilibrium. According to the 13 C NMR analysis (Fig. 7(f)), the forms of CO2 in an aqueous amine solution consisted of carbamate, bicarbonate and carbonic acid, thus the equilibrium reactions in the liquid phase were suggested as follows: −
OOCHN-CH2 -COO− R+ + H2 OH2 N-CH2 -COO− R+ + HCO3 −
(21)
0 3 5 10 15 20 25 30 40 50 60
VL = 100 mL
VL = 200 mL
VL = 200 mL
nL = 130 rpm
nL = 130 rpm
nL = 180 rpm
1.77 ± 0.03 1.67 ± 0.03 1.53 ± 0.01 1.31 ± 0.03 1.16 ± 0.03 1.01 ± 0.02 0.92 ± 0.03 0.88 ± 0.01 0.80 ± 0.02 0.74 ± 0.03 1.84 ± 0.01
1.76 ± 0.00 1.67 ± 0.02 1.54 ± 0.02 1.31 ± 0.03 1.15 ± 0.01 1.02 ± 0.02 0.93 ± 0.04 0.87 ± 0.03 0.80 ± 0.03 0.74 ± 0.03 1.85 ± 0.02
1.76 ± 0.00 1.66 ± 0.02 1.53 ± 0.02 1.31 ± 0.02 1.15 ± 0.01 1.02 ± 0.04 0.95 ± 0.02 0.87 ± 0.03 0.81 ± 0.01 0.74 ± 0.02 1.86 ± 0.03
CO2 absorption rate had little effect and could be neglected. As was mentioned in Sections 3.2.1 and 3.2.3, the CO2 absorption rate increased with the increase of the absorbent concentration within our investigation conditions. According to the judgment suggested by Levenspiel and Godfrey, the kinetics region of CO2 absorption into the mixed solution was considered to be a fast pseudo-first order (Levenspiel and Godfrey, 1974). The reaction kinetics could be determined by the two-film model which was detailed in our previous works (Jing et al., 2012). The gas-phase and liquid-phase mass transfers in the gas–liquid interface can be considered as follows:
N = kG P2 − P2,i
N = EkL C2,i -C2
(28)
(29)
Here, kG and kL denote the gas-phase and liquid-phase mass transfer coefficients, obtained from the literature (Mandal et al., 2001), respectively. At the beginning, the concentration of CO2 in the bulk solution can be ignored (C2 = 0) relative to the ILs concentration. The Ci can be calculated by Eq. (30), on the basis of the phase balance in the gas–liquid interface: C2,i = H2 P2,i
(30)
Throughout the experiments, the kinetic parameters were calculated based on the initial CO2 absorption rates. As a result, the enhancement factor (E) for CO2 absorption can be obtained from Eq. (31) by combining with Eqs. (28)–(30). E=
N H2 kL (P2 − N/kG )
(31)
H2 N-CH2 -COO− R+ + H++ H3 N-CH2 -COO− R+
(22)
RNH2 + H+ RNH3 +
(23)
For the fast pseudo-first order reaction regime, the absorption rate was also given by Eq. (32). Details of the theory were introduced by Danckwerts (1970).
H2 N-CH2 -COO− R+ + RNH3 ++ H3 N-CH2 -COO− R+ + RNH2
(24)
N = C2,i
(25)
According to Eqs. (28)–(32), the enhancement factor (E) and the second-order forward rate constant (k2 ) can be determined. The kinetic parameters for CO2 absorption are listed in Tables 3 and 4. The data summarized that E increased as the concentration of [Hmim][Gly] increased. Linear regression analysis of the values of the second-order reaction rate constant at various temperatures is shown in Fig. 9. The Arrhenius equation led the following expression:
+
CO2 + H2 OH + HCO3 −
+
HCO3 H + CO3
−
−
H2 OH+ + OH−
(26) (27)
5. Absorption kinetics The kinetics region of CO2 absorption into the mixed absorbent of 0.3 M [Hmim][Gly] + 0.7 M AMP was carried out in different liquid-phase volumes and mass transfer coefficient at 303 K. As shown in Table 2, the change of the volume of liquid-phase on
k2 D2 C1
(32)
3088.92
k2,[Hmim][Gly]−AMP = 1.57 × 107 exp −
T
(33)
122
Z. Zhou et al. / International Journal of Greenhouse Gas Control 44 (2016) 115–123
Table 3 The kinetic data for CO2 absorption in [Hmim][Gly]-AMP solution of 1 M total amine concentration with varying mole ratio at 303 K. n1 :n4 *
N0 × 103 (mol m−2 s−1 )
E
k2 (L mol−1 s−1 )
0:1 1:9 2:8 3:7
1.74 ± 0.04 1.76 ± 0.00 1.90 ± 0.03 1.96 ± 0.02
40.90 ± 2.93 43.65 ± 0.28 58.46 ± 3.59 66.77 ± 2.95
496.12 ± 69.50 587.76 ± 8.29 1091.69 ± 131.66 1491.95 ± 132.57
*
n1 , n4 stand for [Hmim][Gly] and AMP, respectively.
Table 4 The kinetic data for CO2 absorption in 0.3 M [Hmim][Gly] + 0.7 M AMP solution at different temperature. T (K)
N × 103 (mol m−2 s−1 )
E
k2 (L mol−1 s−1 )
298 303 313 323
1.72 ± 0.03 1.76 ± 0.00 1.79 ± 0.03 1.84 ± 0.01
39.67 ± 1.95 43.65 ± 0.28 49.25 ± 2.25 57.43 ± 1.02
488.77 ± 48.19 587.75 ± 8.29 778.77 ± 71.91 1107.49 ± 39.20
Fig. 9. Arrhenius plot of k2 for CO2 -[Hmim][Gly]-AMP reaction as a function of temperature.
A second-order reaction rate constant for the reaction between CO2 and AMP was obtained from the paper (Gabrielsen et al., 2006). k2,AMP = 1.943 × 107 exp
−5176.49 T
(34)
CO2 absorption into the mixed solution had a higher absorption rate compared with that of AMP. This might attribute to the increase of the reaction rate constant. The activation energy of k2,[Hmim][Gly]-AMP , k2,AMP were 28.0, 43.0 kJ mol−1 , respectively. This result showed that the reactivity of AMP was significantly enhanced with an addition of [Hmim][Gly]. 6. Conclusions A mixed solution of [Hmim][Gly]-AMP with a low viscosity, high solubility and high absorption capacity of CO2 was proposed for CO2 capture in this work. The results showed that the reactivity of AMP was significantly enhanced with the addition of [Hmim][Gly]. Moreover, CO2 absorption into the mixed solution was stable in a temperature range of 298–323 K. The optimum regeneration temperature and regeneration time of [Hmim][Gly]-AMP solution were 333 K and 60 min, respectively. The mixed system could be satisfactorily regenerated after three cycles. The study of 13 C NMR spectrum indicated that CO2 reacted with [Hmim][Gly]-AMP conformed to zwitterionic mechanism. The reaction region was found to be fast pseudo-first order. Some important kinetic parameters
(e.g. k2 and E) of CO2 absorption into [Hmim][Gly]-AMP solution were obtained. Acknowledgments This work was sponsored by the National Natural Science Foundation of China (No. 21277053), the Program for New Century Excellent Talents in the University of China (No. NCET-11-0851), and the Promotion Program for Young and Middle-aged Teacher in Science and Technology Research of Huaqiao University (ZQNYX104). References Benamor, A., Aroua, M.K., 2005. Modeling of CO2 solubility and carbamate concentration in DEA, MDEA and their mixtures using the Deshmukh-Mather model. Fluid Phase Equilib. 231, 150–162. Caplow, M., 1968. Kinetics of carbamate formation and breakdown. J. Am. Chem. Soc. 90, 6795–6803. chinn, D., Vu, D., Driver, M.S., Boudreau, L.C., 2005. CO2 removal from gas using ionic liquid absorbents. Patent 20,060,251,558, United States. Danckwerts, P.V., 1970. Gas-Liquid Reactions. McGraw-Hill, New York. Danckwerts, P.V., 1979. The reaction of CO2 with ethanolamines. Chem. Eng. Sci. 34, 443–446. Fukumoto, K., Yoshizawa, M., Ohno, H., 2005. Room temperature ionic liquids from 20 natural amino acids. J. Am. Chem. Soc. 127, 2398–2399. Gabrielsen, J., Michelsen, M.L., Stenby, E.H., Kontogeorgis, G.M., 2006. Modeling of CO2 absorber using an AMP solution. AIChE J. 52, 3443–3451. Guo, H.X., Zhou, Z.M., Jing, G.H., 2013. Kinetics of carbon dioxide absorption into aqueous [Hmim][Gly] solution. Int. J. Greenhouse Gas Control 16, 197–205. Jing, G.H., Zhou, L.J., Zhou, Z.M., 2012. Characterization and kinetics of carbon dioxide absorption into aqueous tetramethylammonium glycinate solution. Chem. Eng. J. 181, 85–92. Korson, L., Drost-Hansen, W., Millero, F.J., 1969. Viscosity of water at various temperatures. J. Phys. Chem. 73, 34–39. Kumar, P.S., Hogendoorn, J.A., Versteeg, G.F., Feron, P.H.M., 2003. Kinetics of the reaction of CO2 with aqueous potassium salt of taurine and glycine. AIChE J. 49, 203–213. Levenspiel, O., Godfrey, J.M., 1974. A gradientless contactor for experimental study of interface mass transfer with/without reaction. Chem. Eng. Sci. 29, 1723–1730. Lu, J.G., Ji, Y., Zhang, H., Suo, F.X., Fan, F., Liu, C., 2011. A type of ionic liquids for CO2 capture and the synthesis method. Patent 102,008,870, China. Mandal, B.P., Biswas, A.K., Bandyopadhyay, S.S., 2003. Absorption of carbon dioxide into aqueous blends of 2-amino-2-methyl-1-propanol and diethanolamine. Chem. Eng. Sci. 58, 4137–4144. Mandal, B.P., Guha, M., Biswas, A.K., Bandyopadhyay, S.S., 2001. Removal of carbon dioxide by absorption in mixed amines: modeling of absorption in aqueous MDEA-MEA and AMP-MEA solutions. Chem. Eng. Sci. 56, 6217–6224. Miller, M.B., Luebke, D.R., Enick, R.M., 2010. CO2 - philic oligomers as novel solvents for CO2 absorption. Energy Fuels 24, 6214–6219. O’connell, P.J., Prausnitz, J.M., 1964. Thermodynamics of gas solubility in mixed solvents. Ind. Eng. Chem. Fund. 3, 347–364. Saha, A.K., Bandyopadhyay, S.S., Biswas, A.K., 1995. Kinetics of absorption of CO2 into aqueous solutions of 2-amino-2-methyl-1-propanol. Chem. Eng. Sci. 50, 3587–3598. Sartori, G., Savage, D.W., 1983. Sterically hindered amines for CO2 removal from gases. Ind. Eng. Chem. Res. 22, 239–249. Scovazzo, P., Camper, D., Kieft, J., Poshusta, J., 2004. Regular solution theory and CO2 gas solubility in room-temperature ionic liquids. Ind. Eng. Chem. Res. 43, 6855–6860. Sharma, P., Park, S.D., Baek, I.H., Park, K.T., Yoon, Y.I., Jeong, S.K., 2012. Effects of anions on absorption capacity of carbon dioxide in acid functionalized ionic liquids. Fuel Process. Technol. 100, 55–62. Tan, J., Shao, H., Xu, J., Du, L., 2011. Mixture absorption system of monoethanolamine- triethylene glycol for CO2 capture. Ind. Eng. Chem. Res. 50, 3966–3976. Versteeg, G.F., Van Swaaij, W.P.M., 1988. Solubility and diffusivity of acid gases (carbon dioxide, nitrous oxide) in aqueous alkanolamine solutions. J. Chem. Eng. Data 33, 29–34. Wang, Y.T., Fang, C.G., Zhang, F., Geng, J., Zhang, Z.B., 2009. The performances of CO2 absorption in mixed aqueous solution of MDEA and amino acid ionic liquids. CIESC J. 60, 2781–2786. Weng, Y., Zhao, W., Ye, Q., Zhang, Y., Ye, Z.P., Shi, Y., 2011. Absorption of carbon dioxide with aqueous blends of 2-amino-2-methyl-1-propanol and sodium glycinate. China Environ. Sci. 31, 729–733. Winyu, T., Amornvadee, V., Bryce, M., 2006. Electrochemical investigation on the effect of heat-stable salts on corrosion in CO2 capture plants using aqueous solution of MEA. Ind. Eng. Chem. Res. 45, 2586–2593. Wu, C., Wang, J., Wang, H., Pei, Y., 2011. Effect of anionic structure on the phase formation and hydrophobicity of amino acid ionic liquids aqueous two-phase systems. J. Chromatogr. A 1218, 8587–8593.
Z. Zhou et al. / International Journal of Greenhouse Gas Control 44 (2016) 115–123 Xu, S., Frederick, D.O., Alan, E.M., 1991. Physical properties of aqueous AMP solutions. J. Chem. Eng. Data 36, 71–75. Zhang, F., Fang, C.G., Wu, Y.T., Wang, Y.T., Li, A.M., Zhang, Z.B., 2010. Absorption of CO2 in the aqueous solutions of functionalized ionic liquids and MDEA. Chem. Eng. J. 160, 691–697. Zhang, J., Jia, C., Dong, H., Wang, J., 2013. A novel dual amino-functionalized cation-tethered ionic liquid for CO2 capture. Ind. Eng. Chem. Res. 52, 5835–5841. Zhang, J.M., Zhang, S.J., Dong, K., Zhang, Y.Q., Shen, Y.Q., Lv, X.G., 2006. Supported absorption of CO2 by tetrabutylphosphonium amino acid ionic liquids. Chemistry 12, 4021–4026.
123
Zhao, Y., Zhang, X., Zhen, Y., Dong, H., Zhao, G., Zeng, S., Tian, X., Zhang, S., 2011. Novel alcamines ionic liquids based solvents: Preparation, characterization and applications in carbon dioxide capture. Int. J. Greenhouse Gas Control 5, 367–373. Zheng, C., Tan, J., Wang, Y.J., Luo, G.S., 2012. CO2 solubility in a mixture absorption system of 2-Amino-2-methyl-1-propanol with ethylene glycol. Ind. Eng. Chem. Res. 51, 11236–11244. Zhou, Z.M., Jing, G.H., Zhou, L.J., 2012. Characterization and absorption of carbon dioxide into aqueous solution of amino acid ionic liquid [N1111 ][Gly] and 2-amino-2-methyl-1-propanol. Chem. Eng. J. 204, 235–243.