Highly efficient and selective formation of hydrogencarbonate in CO 2 absorption process using piperidine and piperazine derivatives

Article abstract of DOI:10.1246/cl.2012.142

This investigation demonstrated that bicarbonate ions were selectively formed over carbamate in a CO₂ absorption process using piperidine and piperazine derivatives based on ¹³CNMR. Piperidines with methyl or hydroxymethyl substituent at 2 position (PiP-Me and PiP-MeOH) and 2,5-dimethylpiperazine (DM-PiZ) generated the bicarbonate ions as main adducts in reaction with CO₂. The absorptions of CO₂ by those aqueous amines (PiP-Me and DM-PiZ) were faster than those of MEA (2-aminoethanol).

Full text of DOI:10.1246/cl.2012.142

doi:10.1246/cl.2012.142
142
Published on the web January 28, 2012
Highly Efficient and Selective Formation of Hydrogencarbonate in CO2 Absorption Process
Using Piperidine and Piperazine Derivatives
Man Sub Shin,1 Yoon Kook Park,1 Sung Chan Nam,2 and Kwang-Jin Hwang*1
1Department of Bio and Chemical Engineering, Hongik University,
300 Shinanri, Jochiwon, Chungnam, Korea 339-701
2Greenhouse Gas Research Center, Korea Institute of Energy Research,
102 Gajeong-ro, Yuseong-gu, Daejeon, Korea 305-343
(Received October 4, 2011; CL-110816; E-mail: kjhwang@hongik.ac.kr)
This investigation demonstrated that bicarbonate ions were
selectively formed over carbamate in a CO2 absorption process
using piperidine and piperazine derivatives based on 13C NMR.
Piperidines with methyl or hydroxymethyl substituent at 2
position (PiP-Me and PiP-MeOH) and 2,5-dimethylpiperazine
(DM-PiZ) generated the bicarbonate ions as main adducts in
reaction with CO2. The absorptions of CO2 by those aqueous
amines (PiP-Me and DM-PiZ) were faster than those of MEA
(2-aminoethanol).
H+ ions, these amines do not generate carbamate; CO2 is
converted to bicarbonate, but slowly in the same molar ratio
(eq 3).16-19
(1) Carbamate formation (1° and 2° amines)
þ
2 RR0NH þ CO2 ꢀ RR0NCOOꢀ þ RR0NH2
(2) Conversion of carbamate to bicarbonate
RR0NCOOꢀ þ H2O ꢀ HCO3ꢀ þ RR0NH
(3) Bicarbonate formation (3° amine)
ð1Þ
ð2Þ
ð3Þ
R3N þ CO2 þ H2O ꢀ HCO3ꢀ þ R3NHþ
Recently, there has been increasing demand for green
energy and environmentally related areas: specifically, post
combustion CO2 capture (PCC). In order to address this
problem, many different technologies have been tried, including
water-soluble amines,1,2 using solid amines,3 ionic liquid,4 or
porous supports,5 cryogenic and membrane separation,6 and
biological conversion.7 Thanks to its high capacity for CO2
absorption at a low CO2 partial pressure, amine-based absorption
techniques have been considered as the most promising
technology in PCC.8
In the progress of amine-based absorption technology, it has
been reported that a certain type of amine,9,10 a mixture of
different types of amine,11,12 finding the optimum absorption
process conditions and controlling the stability and reactivity
enhance the CO2 absorption efficiency. However, many re-
searchers have focused on which alternative amine makes it
possible to overcome amine disadvantages, including solvent
loss and high desorption energy.13,14 Recently, we proposed that
enhancing the hydrogencarbonate (bicarbonate, BC) selectivity
process may solve the cumbersome problem of high desorption
energy in CO2 absorption.15
In principle, the conversion of CB to BC as in eq 2 would
depend on the structures of amines resulting in different steric
or electronic effects. With this fact in mind for the selective
formation of BC in the amine-based CO2 absorption process, we
carried out the CO2 absorption process in the presence of
secondary amine derivatives, as shown in Figure 1. In terms of
selectivity, the primary amines are highly inclined to form
carbamates. Meanwhile, tertiary amines are known to react with
CO2 too slowly to form BC ions in practical PCC applications.20
The experimental devices for amine-based CO2 absorption
are composed of a supplier, a reactor, and a data collector, as
shown in Figure 2. In the experiment, 99.99% purity CO2 gas
PiP
PiP-Me
PiP-MeOH
PiZ
DM-PiZ
MEA
Figure 1. The structures, names, and abbreviations of amines
used for CO2 absorption.
Thus far researchers know that CO2 is converted into
mixtures of carbamate (CB) and BC in the amine-based CO2
absorption process. In general, CB is formed quickly, but it
requires high desorption energy. Whereas carbonate is formed
more slowly and it requires only low desorption energy. In other
words, CB can be stably stored and BC can easily reproduce
CO2.8 Thus, if CO2 is converted and separated only in one form
of the intermediates CB or BC, then the storage and desorption
processes for CO2 could be implemented in PCC more
efficiently than the CB-BC mixture could.
The reaction mechanisms for the amine-based CO2 capture
vary depending on the structure of the amines (eqs 1-3). In
primary or secondary amines, which can provide protic cations,
two moles of amines absorb one mole of CO2 to form carbamate
ions (eq 1) then undergo hydrolysis to produce bicarbonate or
carbonate. In tertiary amines, which are not capable of releasing
Figure 2. Experimental apparatus for CO2 absorption: 1) CO2
gas cylinder, 2) reservoir, 3) reactor, 4) pressure transducer,
5) thermometer, 6) vacuum pump, 7) data acquisition and
analysis, 8) water bath, 9) magnetic stirrer, (V1-V5) valves.
Chem. Lett. 2012, 41, 142-144
© 2012 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/

143
Table 1. 13C NMR data of amines before and after CO2 absorption for 2 h
Chemical shift values/ppm
Structure
Sample
sp3-Carbon
sp2-Carbon
a
b
c
d
e
f
CB
BC
PiPa
45.7
44.5
25.7
22.6
24.2
21.8
®
®
®
®
®
O
O
C
N
H
N
d
e
a
b
a
b
d
e
+
PiP + CO2
45.3
25.5
24.1
163 0.3
161.1
c
f
H
f
PiP-Mea
45.7
44.3
24.9
21.6
24.2
21.5
33.5
30.0
51.4
52.7
21.7
18.5
®
®
N
e
a
nob
160 0.5
d
PiP-Me + CO2
b
c
H
f
PiP-MeOHa
45.7
44.6
28.0
24.7
25.4
21.4
23.9
22.1
57.4
58.1
66.1
62.0
®
®
N
e
a
OH
d
PiP-MeOH + CO2
no
160 0.9
b
c
O
O
O
O
O
PiZa
44.7
42.6
®
®
®
®
®
®
®
®
®
C
N
C
N
H
N
d
d
d
d
a
a
a
a
b
c
b
c
+
+
N
H
N
C
N
H
PiZ + CO2
43.5
41.5
44.2
162 0.3
160 0.5
O
H
N
DM-PiZa
49.6
48.9
52.0
49.2
18.5
16.6
®
®
®
®
®
®
®
®
c
a
b
a
DM-PiZ + CO2
no
160 0.8
N
H
c
b
b
aIn neutral condition before addition of CO2. Not observed.
was used. The reactor was made of stainless steel and its volume
was 49.5 cm3.
All the experiments were conducted at room temperature.
Initially, 10 g of 30 wt % (distilled water 7 g + absorbents 3 g)
amine absorbents were injected into the reactor. The exception
was PiZ, which was used at 8 wt % as its solubility in water was
too low. Any gases remaining in the reactor were removed
by opening valve V5 and using a vacuum pump. When the
pretreatment had been completed, valve V5 was closed and
valve V4 was opened to inject CO2 in the supplier into the
reactor. The pure CO2 pressure was controlled to be around 1.0-
1.2 bar. This process was repeated until amine absorbents did not
absorb CO2 any further. The numbers of CO2 injections were
between six and fourteen depending on the amine absorbents.
The reactions were completed within about two hours.
Figure 3. 13C NMR data of PiP before (a) and 1 h after (b) CO2
absorption.
For NMR spectroscopic analysis, a certain amount (ca.
200 ¯L) of reaction mixture was diluted in D2O or CDCl3 and
analyzed using a 200 or 500 MHz NMR spectrometer (Bruker
Ltd.) and the result is presented in Table 1. The formation of CB
or BC ions were identified by the C=O peak near 160-168
ppm.21,22 As examples, 13C NMR spectra of PiP and PiP-Me are
shown in Figures 3 and 4. The C=O peaks of the CB formed
from reactions with CO2 with PiP and PiZ derivatives were
identified at near 163 and 162 ppm, respectively. The chemical
shift (¤) values of the C=O peaks of CB were not affected by
pH changes caused by different CO2 loading, as represented by
other researchers.8 However, the chemical shift (¤) value of BC
showed different values depending on the pHs of surrounding
solutions.22 As reaction time went on, the position of C=O
peaks of BC in PiP and PiZ derivatives shifted toward the
upfield regions. In the case of PiP-Me, it moved to 160.5 ppm
from 165.5 ppm after 2 h as shown in Figure 4. The carbonate
2¹
(CO3 ) was not distinguished from the bicarbonate because
only one peak was formed22 as protons were quickly exchanged
between bicarbonate and carbonate.
Figure 4. 13C NMR data of PiP-Me before (a) and after (b, c)
CO2 absorption for 1 and 2 h, respectively.
Chem. Lett. 2012, 41, 142-144
© 2012 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/

144
when their position numbers 2 or 5 had been substituted by alkyl
or hydroxy alkyl groups, only bicarbonate was formed. Initial
absorption rates of piperidine (PiP and PiP-Me) and piperazine
(DM-PiZ) derivatives were faster than those of MEA, which is
currently used in the industrial market for PCC.
Based on these results, we conclude that it will be possible
to selectively collect bicarbonate only in the process of amine-
based CO2 absorption and therefore, this process can be used
to improve the PCC process to be more efficient in terms of
desorption energy.
This work was supported by Energy & Resource R&D
program under the Ministry of Knowledge Economy, Republic
of Korea (2009-2011).
References
1
R. Shao, A. Stangeland, Bellona Report, The Bellona
Foundation, Oslo, Norway, 2009.
J. D. Davis, M.S. Thesis or Ph.D. Dissertation, The
University of Texas at Austion, 2009.
Figure 5. The absorption of 1st injected CO2 by aqueous
amine solution (30 wt % amine solutions in water were used
except 8 wt % for PiZ).
2
In general, the secondary amines PiP and PiZ compounds in
the neutral state were not distinguished from the protonated
forms (PiPH+ and PiZH+) because protons were quickly
exchanged between the amines in equilibrium. Specifically,
these results were confirmed by the NMR spectroscopies of the
PiP derivatives as shown in Figures 3 and 4.
3
4
M. Kim, J.-W. Park, Chem. Commun. 2010, 46, 2507.
D. Wappel, G. Gronald, R. Kalb, J. Draxler, Int. J.
Greenhouse Gas Control 2010, 4, 486.
X. Xu, C. Song, J. M. Andrésen, B. G. Miller, A. W. Scaroni,
Microporous Mesoporous Mater. 2003, 62, 29.
M. Mavroudi, S. P. Kaldis, G. P. Sakellaropoulos, Fuel 2003,
82, 2153.
C. Stewart, M.-A. Hessami, Energy Convers. Manage. 2005,
46, 403.
Q. Yang, M. Bown, A. Ali, D. Winkler, G. Puxty, M. Attalla,
Energy Procedia 2009, 1, 955.
5
6
7
8
9
In terms of selectivity of BC over CB, as summarized in
Table 1, this result illustrates that piperidine and piperazine
derivatives containing methyl or hydroxymethyl groups at the
2 or 5 position, PiP-Me, PiP-MeOH, and DM-PiZ, yielded BC
ions only in the course of the CO2 absorption process. On
reviewing the 13C NMR of PiP-Me after absorption of CO2
(Figure 4), CB was not formed and only a BC peak existed near
160 ppm. Similarly, in the case of PiP-MeOH, only a carbonyl
peak from BC ions was observed at 160.9 ppm. The NMR of PiZ
after CO2 absorption represents the CB/BC peaks of PiZ at
162.3/160.5 ppm, respectively. Meanwhile, only a BC peak of
DM-PiZ appeared at 160.8 ppm. It is quite an intriguing result
that piperidine and piperazine derivatives with alkyl substituent
at the ¡ carbon produced BC only, while it was also expected to
form CB over BC in the early stage of CO2 absorption process.
One of the possible explanations would be that the formation of
carbamate from PiP-Me and PiP-MeOH was suppressed because
they were readily converted to BC due to steric effects.
In addition to the selectivity of BC, the CO2 absorption was
also investigated based on the CO2 pressure drop in the presence
of different types of secondary amines along with MEA as a
reference, as shown in Figure 5. The amines, PiP, PiP-Me, and
DM-PiZ, showed faster pressure drop than the standard sample
MEA. This may be attributed the fact that secondary amines
have higher basicity than MEA due to the inductive effect of
alkyl groups. PiP-MeOH, which has hydrophilic hydroxy (OH)
groups showed much slower absorption than the structurally
similar amines PiP-Me or MEA. This result agrees with the
previous report that hydroxyamines usually represent a lower
CO2 loading capacity than that of alkylamines.23
P. Singh, G. F. Versteeg, Process Saf. Environ. Prot. 2008,
86, 347.
10 P. Singh, J. P. M. Niederer, G. F. Versteeg, Int. J. Greenhouse
Gas Control 2007, 1, 5.
11 J. T. Cullinane, G. T. Rochelle, Fluid Phase Equilib. 2005,
227, 197.
12 J.-G. Shim, J.-H. Kim, J. H. Lee, K.-R. Jang, Energy
Procedia 2009, 1, 779.
13 H. Lepaumier, D. Picq, P.-L. Carrette, Ind. Eng. Chem. Res.
2009, 48, 9068.
14 H. Lepaumier, D. Picq, P.-L. Carrette, Ind. Eng. Chem. Res.
2009, 48, 9061.
15 S.-C. Nam, K.-J. Hwang, Y. Yoon, Y. Park, M. S. Shin,
Korean Patent Applied 10-2010-0121664, 2010.
16 M. Caplow, J. Am. Chem. Soc. 1968, 90, 6795.
17 P. V. Danckwerts, Chem. Eng. Sci. 1979, 34, 443.
18 K. H. Han, J. S. Lee, B. M. Min, Korean Chem. Eng. Res.
2007, 45, 197.
19 G. Sartori, D. W. Savage, Ind. Eng. Chem. Fundam. 1983,
22, 239.
20 F. Barzagli, F. Mani, M. Peruzzini, Int. J. Greenhouse Gas
Control 2011, 5, 448.
21 W. Böttinger, M. Maiwald, H. Hasse, Fluid Phase Equilib.
2008, 263, 131.
22 F. Mani, M. Peruzzini, P. Stoppioni, Green Chem. 2006, 8,
995.
23 P. Singh, J. P. M. Niederer, G. F. Versteeg, Chem. Eng. Res.
Des. 2009, 87, 135.
In conclusion, cyclic secondary amines, piperidine and
piperazine, generally formed both carbamate and bicarbonate in
the CO2 absorption process. However, it was discovered that
Chem. Lett. 2012, 41, 142-144
© 2012 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/