Novel ionic liquid analogs formed by triethylbutylammonium carboxylate-water mixtures for CO2 absorption

Article abstract of DOI:10.1016/j.molliq.2011.12.006

This work presents ten ionic liquid analogs of triethylbutylammonium carboxylate and water as CO₂ absorbents. The mixtures are liquids with relatively low viscosities at ambient temperature. The absorption experimental results have shown that each of them has fast absorption rate and large absorption capacity of CO₂.

Full text of DOI:10.1016/j.molliq.2011.12.006

Journal of Molecular Liquids 168 (2012) 17–20
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Liquids
journal homepage: www.elsevier.com/locate/molliq
Novel ionic liquid analogs formed by triethylbutylammonium carboxylate-water
mixtures for CO2 absorption
⁎
G.N. Wang, Y. Dai, X.B. Hu, F. Xiao, Y.T. Wu , Z.B. Zhang, Z. Zhou
Key Laboratory of Mesoscopic Chemistry of MOE, School of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210093, China
a r t i c l e i n f o
a b s t r a c t
Article history:
This work presents ten ionic liquid analogs of triethylbutylammonium carboxylate and water as CO2 absor-
bents. The mixtures are liquids with relatively low viscosities at ambient temperature. The absorption exper-
imental results have shown that each of them has fast absorption rate and large absorption capacity of CO2.
© 2011 Elsevier B.V. All rights reserved.
Received 30 September 2011
Received in revised form 5 December 2011
Accepted 20 December 2011
Available online 30 December 2011
Keywords:
Ionic liquid analogs
Triethylbutylammonium
Carboxylic acid
Carbon dioxide
Absorption
1. Introduction
compound as ionic liquid analogs is that the samples stay in the liquid
state at a large temperature range just like the ILs though they are not
The removal of CO2 from industrial flue gasses is now widely con-
cerned because of the adverse effects such as global warming [1]. Var-
ious methods have been proposed, such as selective adsorption [2]/
absorption [3,4] and membrane processes [5,6]. Although these
methods have been successfully used in industries, they are still asso-
ciated with energy consumption, corrosion and pollution problems
[7]. In order to overcome such obstacles, many novel absorbents
such as ionic liquids (ILs) [8], salt hydrates [9] and metal oxide com-
plexes [10], have been developed to provide high CO2 absorption ca-
pacity and thermal stability. However, such absorbents are strongly
affected by the presence of water vapor in the gas stream so that
the absorption capacity [11] and absorption mechanism [12] are
much more different. Unfortunately, water is very common in flue
gasses from fossil fuel combustion or other industrial gas streams
[13]. Therefore, it is crucial to provide novel compounds which can
absorb CO2 in the presence of water.
composed entirely of ions. It is further proved that the [N2224][carbox-
ylate]-nH2O mixtures ([N2224][CA]-nH2O) have high CO2 absorption ca-
pacity and fast absorption rate of CO2, which are very attractive for the
capture and separation of acidic gasses in industrial applications.
2. Experimental section
2.1. Materials
Triethylamine ((C2H5)3N), n-butyl bromide (C4H9Br), silver oxide
(Ag2O), and ethanol, obtained commercially from Sinopharm Chemi-
cal Reagent Co., Ltd (Nanjing, China), were of analytical grade and
used without further purification. All carboxylate acids of analytical
grade were purchased from Nanjing Chem. Co. (Nanjing, China) and
used as received.
In 1995, Quinn reported hydrated salts based on acetic acid and cit-
rate acid, which showed good performance of CO2 capture [9]. Inspired
by his publication, ten novel ionic liquid analogs are prepared by cou-
pling asymmetric triethylbutylammonium ([N2224]) cation with car-
boxylate anion. It is found that each of them is in the liquid state at
room temperature with a certain amount of water and can be detected
only a glass transition temperature. The reason to define this type of
2.2. General procedure for [N2224][carboxylate]-nH2O synthesis
Triethylbutylammonium bromide ([N2224][Br]) was prepared via
the alkylation of (C2H5)3N (70.0 g) with C4H9Br (95.0 g) in ethanol
(200 ml) under reflux and vigorous stirring for 12 h. The solvent
and unreacted reactants were removed by rotary evaporation,
[N2224][Br] (160 g) as white solid was obtained after being dried at
60 °C under vacuum (≈0.1 kPa) for 24 h.
In the anion exchange reaction, Ag2O was added in portions into
[N2224][Br] solution, the mole ratio of Ag2O to [N2224][Br] is 1:2, and
stirred without air and light at room temperature for 6 h. After filtra-
⁎
Corresponding author at: Key laboratory of Mesoscopic Chemistry of MOE, School
of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210093, China.
Tel./fax: +86 25 83593772.
tion, the [OH−
] concentration of the resulting solution was
E-mail address: ytwu@nju.edu.cn (Y.T. Wu).
0167-7322/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.molliq.2011.12.006

18
G.N. Wang et al. / Journal of Molecular Liquids 168 (2012) 17–20
determined from the titration with standard HCl solution. Then, the
triethylbutylammonium hydroxide ([N2224][OH]) solution reacted
with a slight excess of carboxylate acid via neutralization at room
temperature for 2 h. Water was evaporated to generate a residual so-
lution that contained the [N2224]-based samples. After being further
dried at 60 °C under vacuum (≈0.1 kPa), the residue was extracted
with diethyl ether for several times to remove the unreacted acid.
The crude product was dissolved in the mixed solvent of dichloro-
methane with 2 vol.% methanol, and the solution was passed through
a column filled with neutral silica. After the removal of the solvent by
evaporation, the product was dried in a vacuum (≈0.1 kPa) oven at
80 °C for at least 5 days to remove water before the characterization
experiments. If the samples were dried at other conditions (at 50 to
recovered by extruding CO2 at 55 °C and under vacuum (0.1 kPa)
for 4 h and repeated CO2 absorptions into two typical samples were
presented in Fig. 2.
3. Results and discussion
3.1. The measurement of water content
As is known to all, carboxylate anions are highly hydrophilic and it
is almost impossible to dry the samples completely as can be seen in
Table 1. However, the evidences of FTIR and 1H NMR analysis indicate
that small amounts of water interact with the anions instead of being
isolated molecules and form complexes with the [N2224] salts. The
DSC spectrum results suggest that the samples remain in the liquid
state at a large range of temperature. Based on the phenomenon men-
tioned above, it is reasonable to regard [N2224][CA]-nH2O as ionic liq-
uid analogs.
70 °C and for 1 to 3 d), triethylbutylammonium carboxylate ([N2224
]
[CA]) with different amount of water would be obtained [9]. The
structures of the compounds are determined by the 1H NMR spectros-
copy (BRUKER DPX 300) and confirmed by the elemental analysis
(Elementar Vario MICRO). Photos of the ten obtained samples can
be found in Fig. 1 and details of characterization experiments were
shown in Supplementary Data.
3.2. Physical properties
The important physical properties of all samples in this work are
summarized in Table 1. It is obvious that only glass transition temper-
atures (Tg) can be detected ranging from −75 °C to −50 °C, and Tg in-
creases generally with the number of carboxyl groups in the anion.
For example, [N2224]2[phthalate]-2H2O has the highest Tg value
whereas the [N2224][acetate]-1H2O, [N2224][propionate]-0.5H2O and
[N2224][butyrate]-0.5H2O have about 20 °C lower values. Those later
three salts have almost identical Tg values, indicating that the length
of alkyl chain in the mono-carboxylate anions has less effect on the
Tg values. In contrast to [N2222][acetate]-4H2O that has a melting tem-
perature at 45 °C [9], no melting temperatures were detected from
−100 to 50 °C for all samples in this work, reflecting that the
[N2224] salts and water complexes were also eutectic liquid mixtures
at the experimental temperature range.
The temperature of water loss (Tloss) is the temperature when the
samples start to lose water and it is found that the values of Tloss are
all above 100 °C and increase with the number of the functional
groups in the anions. All [N2224][CA]-nH2O have the decomposition
temperatures from 160 to 190 °C, which are a little lower than normal
functional ionic liquids [14]. The fully ionized carboxylate anions are
basic and triethylbutylammonium hydroxide was formed in the pres-
ence of water. So the mechanism of decomposition may relate to a
Hoffman Elimination Reaction explained by MacFarlane [15]. From
this respect, the reason for low thermal stability of the ILs may be
the interaction of basic anions and cation in the presence of water.
2.3. Water content measurement
The water contents in the samples were determined by Karl-
Fisher Analysis (Metrohm 787 KF Titrino) and 1H NMR spectroscopy
(BRUKER DPX 300).
2.4. Physical properties measurement
The glass transition temperature (Tg) was recorded at the temper-
ature range from −100 to 0 °C under N2 atmosphere at the scan rate
of 10 °C/min by different scanning calorimetry (Perkin-Elmer Dia-
mond DSC), while melting temperature (Tm) was recorded at the
temperature range from −20 °C to 50 °C by thermogravimetric anal-
ysis (Perkin-Elmer Pyris 1 TGA).
The temperatures of water loss (Tloss) and decomposition (Tdec
)
were determined at the scan rate of 10 °C/min under N2 at the tem-
perature range from room temperature to 500 °C by thermal analysis
(TG) instrument (NETZSCH STA 449C). The details of DSC and TG
spectrum can be seen in Supplementary Data.
The viscosity and density were measured by a cone-plate viscom-
eter (HAAKE Rheostress 600, an uncertainty of 1% in relation to the
full scale) and a densitometer (DMA 5000 DENSITY METER, an uncer-
tainty of 0.03 kg·m−3).
2.5. CO2 absorption experiment
The water content in the samples was adjusted to integral num-
bers by adding corresponding amount of fresh deionized water before
the absorption experiments. The absorption of CO2 (99.99%, from
Nanjing Gas Supply Inc., China) into [N2224][CA]-nH2O was carried
out at 25 °C and at the same initial CO2 pressure of 0.1 MPa, according
to the standard procedure [12]. The saturated samples were
Fig. 1. [N2224][CA]-nH2O prepared in this work. Left to right: [N2224][propionate]-
0.5H2O, [N2224]2[fumarate]-3.6H2O, [N2224]2[malonate]-3.5H2O, [N2224][acetate]-1H2O,
Fig. 2. Cycles of CO2 absorption into [N2224][CA]-nH2O at 25 °C and at the same initial
CO2 pressure of 0.1 MPa. The saturated samples were recovered by extruding CO2 at
55 °C and under vacuum (0.1 kPa) for 4 h. ● [N2224][butyrate]-1H2O and ▲ [N2224][pro-
pionate]-1H2O.
[N2224][butyrate]-0.5H2O, [N2224]2[malate]-3H2O, [N2224]2[succinate]-3H2O, [N2224
[phthalate]-2H2O, [N2224]3[citrate]-2.5H2O, [N2224]2[maleate]-2H2O.
]
2

G.N. Wang et al. / Journal of Molecular Liquids 168 (2012) 17–20
19
Table 1
Physical properties of [N2224] [carboxylate]-nH2O.
Scheme 1. Proposed reaction mechanism between [N2224][carboxylate]-nH2O and CO2.
Anion
Tg/°C
Tloss/°C
Tdec/°C
n
ρ/g·cm−3
η/mPa·s
Acetate
−73
−74
−74
−60
−57
−49
−60
−54
−55
−52
103
107
101
105
111
113
110
119
117
125
164
162
161
161
162
161
156
167
187
190
1
0.9981
0.9807
0.9694
1.0393
1.0415
1.0654
1.0486
1.0477
1.0578
1.0762
79
also faster and more efficient CO2 absorbents than methyldiethanola-
mine aqueous solutions (51.28 w/w%, K=0.066 min−1) [19] widely
used in industrial processes [20].
The CO2 absorption capacities of all [N2224][CA]-nH2O at a certain
equilibrium partial pressure of CO2 are also summarized in Table 2.
Generally speaking, all samples exhibit large CO2 absorption capaci-
ties at relatively low partial pressures, i.e., one mole of [N2224][buty-
rate]-1H2O (cheap) absorbs 0.45 mol of CO2 at 25 °C and 44 kPa,
which is comparable to the CO2 absorption capacity of [N2224][L-Ala]
Propionate
Butyrate
Malonate
Succinate
Phthalate
Maleate
Fumarate
Malate
0.5
0.5
3.5
3
2
2
3.6
3
2.5
120
157
509
NA1
NA
NA
NA
NA
NA
Citrate
1
NA: higher than 3800 mPa·s at 25 °C.
(expensive). It can be found that the solubility of CO2 into [N2224
]
[mono-carboxylate]-nH2O samples grows with the increasing length
of the alkyl chain in the anion. The free volume between ions will in-
crease if the alkyl chain in the anion is enlarged. However, for multi-
carboxylate species, such a trend does not exist probably because the
interactions between the carbonyl groups are strong enough to pre-
vent the affinity with CO2. In addition, the conformation of the anions
affects not only the physical parameters but also the absorption ca-
pacity of CO2. [N2224]2[fumarate]-6H2O traps about 58% less CO2
than [N2224]2[maleate]-6H2O, which indicates that the trans-
conformation favors strong intermolecular hydrogen bonds between
anions rather than the combination with CO2 [21]. The small differ-
ence in the absorption capacity between [N2224]3[citrate]-10H2O and
[N2222]3[citrate]-10.1H2O [9] confirms the conclusion that the cation
plays a less important role in determining the solubility of gasses
into ILs [8].
The viscosities of all samples range from 79 mPa·s up to more
than 3800 mPa·s (NA in Table 1) at 25 °C, depending on the anion
types. [N2224][mono-carboxylate] water complexes are found to
have viscosities lower than 160 mPa·s at 25 °C, primarily due to the
lighter molecular weight of these anions. The reason for the high vis-
cosities may be attributed to the multi-carbonyls in the anions which
would enlarge the interactions between the ions. Relatively low vis-
cosities make the [N2224][CA]-nH2O attractive as industrial absorbents
for CO2 capture.
In addition, [N2224][CA]-nH2O are of lower densities than tradi-
tional ILs. It is distinct that the densities of [N2224][CA]-nH2O increase
with a rise in the volume of the carboxylate anions, like the trend of
other kinds of ILs, such as imidazolium-based ILs [16].
3.2.1. CO2 absorption properties
The large CO2 absorption capacities of the mixtures are resulted
from the carboxylate anions which are conjugated bases of corre-
sponding weak acids, being the main driving force for the chemical
reaction. Scheme 1 was claimed by Quinn [9] and it was found to be
suitable for our case from the results of 13C NMR and 1H NMR analysis
for both CO2-free and CO2-treated samples (see Supplementary Data).
The 13C NMR spectrum detects a new resonance at 159.604 ppm, to
be the bicarbonate carbonyl carbon [9] while other features of the
spectrum stay the same. Correspondingly, there is an extra resonance
signal at 2.084 ppm in the CO2-saturated IL detected by 1H NMR
spectra.
In order to test the accuracy of the absorption apparatus, the solu-
bility of CO2 into 1 mol/L NaOH solution and [bmim][PF6] was deter-
mined and compared with those in the references [17,18]. The
experimental results have shown strong constancy that demonstrates
the liability of the absorption apparatus in this work.
The absorption capacity is shown to reach 95% of the equilibrium
value within 30 min which is even shorter than the amino acid ionic liq-
uids [14]. [N2224][butyrate]-1H2O had been recycled with no obvious
loss of absorption capacity. However, the capacity of [N2224][propio-
nate]-1H2O in the later three cycles is 8% smaller than the first time. It
may be reasoned that an absolute vacuum was not really achieved in
the subsequent three cycles. Based on the absorption data in the first
30 s, the initial absorption rate constant (K/min−1) was calculated
and shown in Table 2. It is found that [N2224][CA]-nH2O with the high
CO2 capacities usually has the large K values. Small K values may result
from low CO2 loadings or high viscosities. Nevertheless, all [N2224][CA]-
nH2O, which are comparable to the aqueous monoethanolamine (36 w/
w%, K=0.24 min−1, results from our own experimental device), are
4. Conclusion
From all aspects mentioned above, the ionic liquid analogs of
[N2224][CA]-nH2O are thought to be of significance for the industrial
CO2 capture due to the high absorption capacity, fast absorption
rate and relatively low cost.
Acknowledgments
Table 2
The authors are grateful for the financial support from National
Natural Science Foundation (No. 21076101), Technological Support
Project of Jiangsu Province (No. BE2010142) and Fundamental Re-
search Funds for the Central Universities (No. 1114020504).
CO2 absorption into [N2224][CA]-nH2O at 25 °C and the same initial pressure of CO2
(0.1 MPa).
K/min−1
⁎
Anion
n
P /kPa
Absorption capacity
mol CO2/mol compound
Acetate
1
1
1
6
6
6
6
6
6
45
46
44
50
61
50
56
66
68
0.23
0.39
0.45
0.65
0.57
0.66
0.58
0.24
0.20
0.38
0.50
0.48
0.14
0.33
0.22
0.20
0.09
0.17
0.18
0.08
0.09
0.08
0.36
NA1
Appendix A. Supplementary data
Propionate
Butyrate
Malonate
Succinate
Phthalate
Maleate
Fumarate
Malate
Supplementary data to this article can be found online at doi:10.
1016/j.molliq.2011.12.006.
References
[1] W.S. Broecker, Chinese Science Bulletin 51 (2006) 1018–1029.
[2] Z. Yong, V. Mata, A.E. Rodrigues, Separation and Purification Technology 26
(2002) 195–205.
[3] F. Karadas, M. Atilhan, S. Aparicio, Energy & Fuels 24 (2010) 5817–5828.
[4] J. Huang, T. Ruther, Australian Journal of Chemistry 62 (2009) 298–308.
Citrate
10
0
10.1
66
100
100
L-Ala13
⁎7
Citrate
1
NA: not available. p* is the partial pressure of CO2 after absorption equilibrium.

20
G.N. Wang et al. / Journal of Molecular Liquids 168 (2012) 17–20
[5] M.B. Kagg, A. Lindbrathen, Industrial and Engineering Chemistry Research 44
(2005) 7668–7675.
[14] Y.Y. Jiang, G.N. Wang, Z. Zhou, Y.T. Wu, J. Geng, Z.B. Zhang, Chemical Communica-
tions (2008) 505–507.
[6] N.Y. Du, H.B. Park, G.P. Robertson, M.M. Dal-Cin, T. Visser, L. Scoles, M.D. Guiver,
Nature Materials 10 (2011) 372–375.
[15] D.R. Macfarlane, J.M. Pringle, K.M. Johansson, S.A. Forsyth, M. Forsyth, Chemical
Communications (2006) 1905–1917.
[7] E.D. Bates, R.D. Mayton, I. Ntai, J.H. Davis, Journal of the American Chemical Soci-
ety 124 (2002) 926–927.
[16] J.S. Wilkes, J.A. Levisky, R.A. Wilson, C.L. Hussey, Inorganic Chemistry 21 (1982)
1263–1264.
[8] M.J. Muldoon, S.N.V.K. Aki, J.L. Anderson, J.K. Dixon, J.F. Brennecke, The Journal of
Physical Chemistry. B 111 (2007) 9001–9009.
[17] B. Rumpf, J. Xia, G. Maure, Industrial and Engineering Chemistry Research 37
(1998) 2012–2019.
[9] R. Quinn, J.B. Appleby, G.P. Pez, Journal of the American Chemical Society 117
(1995) 329–335.
[18] J. Kumelan, A.P. Kamps, D. Tuma, J. Xia, G. Maurer, Journal of Chemical and Engi-
neering Data 51 (2006) 1802–1807.
[10] Y. Matsukura, T. Okumura, R. Kobayashi, K. Oh-Ishiet, Chemistry Letters 39
(2010) 966–967.
[19] H. Yu, Y.T. Wu, Y.Y. Jiang, Z. Zhou, Z.B. Zhang, New Journal of Chemistry 33 (2009)
2385–2390.
[11] L.A. Blanchard, Z.Y. Gu, J.F. Brennecke, The Journal of Physical Chemistry. B 105
(2001) 2437–2444.
[20] R.J. Littel, G.F. Versteeg, W.P.M. Vanswaaij, Chemical Engineering Science 47
(1992) 2027–2035.
[12] J.M. Zhang, S.J. Zhang, K. Dong, Chemistry
A
European Journal 12 (2006)
[21] Y. Fukaya, Y. Iizuka, K. Sekikawa, H. Ohno, Green Chemistry 9 (2007) 1155–1157.
4021–4026.
[13] N. Hiyoshi, K. Yogo, T. Yashima, Chemistry Letters 37 (2008) 1266–1267.