Available online at www.sciencedirect.

com

Journal of Environmental Sciences 2012, 24(3) 494498

JOURNAL OF
ENVIRONMENTAL
SCIENCES
ISSN 1001-0742
CN 11-2629/X

www.jesc.ac.cn

Vaporization reduction characteristics of aqueous ammonia solutions by the

addition of ethylene glycol, glycerol and glycine
to the CO2 absorption process
Jong-Beom Seo1 , Soo-Bin Jeon1 , Je-Young Kim2 , Gang-Woo Lee3 ,
Jong-Hyeon Jung4 , Kwang-Joong Oh1,
1. Department of Environmental Engineering, Pusan National University, Busan 609735, Korea. E-mail: jbseo@pusan.ac.kr
2. Research Institute of Industrial Science & Technology, Pohang 790330, Korea
3. Research & Development Institute of Yoosung Co. Ltd., Ulsan 680070, Korea
4. Faculty of Health Science, Daegu Haany University, Daegu 712715, Korea
Received 17 February 2011; revised 18 April 2011; accepted 05 May 2011

Abstract
Aqueous ammonia (NH3 ) solution can be used as an alternative absorption for the control of CO2 emitted from ue gases due to its
high absorption capacity, fast absorption rate and low corrosion problem. The emission of CO2 from iron and steel plants requires
much attention, as they are higher than those emitted from power plants at a single point source. In the present work, low concentration
ammonia liquor, 9 wt.%, was used with various additives to obtain the kinetic properties using the blast furnace gas model. Although
a solution with a high ammonia concentration enables high CO2 absorption eciency, ammonium ions are lost as ammonia vapor,
resulting in reduced CO2 absorption due to the lower concentration of the ammonia absorbent. To decrease the vaporization of ammonia,
ethylene glycol, glycerol and glycine, which contain more than one hydroxyl radical, were chosen. The experiments were conducted at
313 K similar to the CO2 absorption conditions for the blast furnace gas model.
Key words: ammonia; vaporization; blast furnace gas; ethylene glycol; glycerol; glycine
DOI: 10.1016/S1001-0742(11)60797-3

Introduction
Various technologies are used to separate CO2 from
the ue gas of conventional fossil fuel-red power
plants and petroleum chemical plants, including chemical absorption, physical adsorption, cryogenic methods,
membrane separation and biological xation (Choi et al.,
2009). Industrially chemical absorbents mainly include
monoethanolamine (MEA), diethanolamine (DEA) and
N-methyldiethanolamine (MDEA). A dierent class of
chemical absorbents; amine solutions, have commonly
been used for the commercial production of CO2 and tested
for the capture of CO2 on a pilot scale. However, such
technologies require a large amount of energy, especially
during the desorption stage of the process. In addition, the
use of amines also entails problems in relation to solvent
degradation and corrosion. Aqueous ammonia (NH3 ) solution can be used as an alternative for the control of the CO2
emitted from ue gas due to its high absorption capacity,
fast absorption rate and low corrosion problem (Seo et al.,
2011).
Aqueous ammonia is a good example to serve these
characteristics and has been used for many years at several
* Corresponding author. E-mail: kjoh@pusan.ac.kr

steel works for the removal of CO2 and sulfur compounds

from blast furnace gas and coke oven gas. Since the studies
of Yeh and Bai (1999) and Yeh et al. (2004) on aqueous
ammonia for the absorption of greenhouse gases, further
investigations have been undertaken. Moreover, Resnik et
al. (2004) and Diao et al. (2004) subsequently reported
an aqua ammonia process and ammonia scrubbing for the
capture of CO2 . Therefore, aqueous ammonia, which has
a low regeneration temperature, has been recommended
for the process of CO2 absorption at iron and steel works,
which produce many heat sensitive wastes at low and
medium temperatures (Kim et al., 2009).
However, although a solution with a high ammonia
concentration enables high CO2 absorption eciency,
ammonium ions are lost as ammonia vapor, resulting in reduced CO2 absorption due to the lower concentration of the
ammonia absorbent. Ethylene glycol, glycerol and glycine
are materials that could potentially be used to control the
vapor pressure of NH3 and improve the CO2 absorption
characteristics. Their performance is most likely due to the
interactions between the hydroxyl groups of the additives
and ammonia via hydrogen bonding (You et al., 2008).
To analyze the eects of additives, vaporization experiments were conducted in three-stages, including before

No. 3

Vaporization reduction characteristics of aqueous ammonia solutions by the addition of ethylene glycol, glycerol

CO2 absorption, initial CO2 absorption and the stabilization phase; whereby, the vaporization generation can be
predicted during the CO2 treatment process.

495

be the optimum aqueous ammonia solution concentration

for CO2 absorption process and the blended additives were
1 wt.% in aqueous solution.
1.3 Total vaporization measurement of NH3 -water

1 Materials and methods

1.1 Materials
Ammonia solution (28 wt.%), ethylene glycol (99.5 wt.%),
glycerol (99.0 wt.%) and glycine (99.9 wt.%) were supplied by Junsei Chemical Co. Ltd. (Japan). All chemicals
were used without further purication. Aqueous solutions
were prepared with distilled water. Double-distilled water,
degassed by boiling, was used to prepare the ammonia
solutions. The CO2 gas was of commercial grade, with a
purity of 19.99%.
1.2 Selection of additives
The operation of the aqueous ammonia process requires
special consideration with respect to the vaporization of
NH3 due to its highly volatile nature. Although a solution
with a high ammonia concentration enables high CO2
absorption eciency, ammonium ions are lost as ammonia
vapor, resulting reduced CO2 absorption due to the lower
concentration of the ammonia absorbent.
In general, amine radicals have been shown to improve
the absorption capacity of acidic gases by increasing the
basicity. Hydroxyl radicals are also known to prevent
solvate by hydrogen bond and to reduce vaporization.
In this work, three additives, including amine/hydroxyl
or hydroxyl radicals, were selected as aqueous ammonia
absorbents. To decrease the vaporization of ammonia,
ethylene glycol, glycerol and glycine, which contain more
than one hydroxyl radical, were chosen.
In the case of the aqueous ammonia absorbents, the
concentration of ammonia was 9 wt.%, which is known to

The experimental apparatus for measuring the vaporization

of the NH3 -water is shown in Fig. 1. The reactor, with
a height of 160 mm and inside diameter of 95 mm,
was located inside a temperature-controlled vessel, with
four, 5-mm-wide glass plates adhered to the inner wall
of the reactor as baes. The total volume of the reactor
was about 1134 cm3 , with an active interface area of
70.88 cm2 . The reactor temperature was measured using
a K-type thermocouple, with an accuracy of 0.1 K. All
experiments were conducted with 1000 mL of solution.
The vaporization of the NH3 -water was measured using
a pressure data logger (PR2000, Madgetech, USA), with
an accuracy of 0.25% at 313 K. The measured data
from the pressure transducer were stored automatically to a
computer. The NH3 gas concentration was measured using
an NH3 analyzer, with an accuracy of 0.2 ppm.
The vaporization experiments proceeded via threestages. The rst stage was the measurement of the
vaporization of the aqueous ammonia solutions and blended alcohol solutions, with no CO2 reaction. These results
were observed through a pressure indicator and saved to a
computer. The second stage was the measurement of the
initial gaseous ammonia concentration when the blended
ammonia solutions reacted with CO2 , which was measured
for 30 min. Lastly, the vaporization of the aqueous ammonia solutions and blended alcohol solutions loaded with 1
mol/L CO2 were measured for 120 min.
The three-stages showed the before CO2 absorption, the
initial CO2 absorption and the stabilization phase in the
whole process, from which it was possible to predict the
vapor generated in the CO2 treatment process.

11

10

4
3
8
2

6
7

Fig. 1 Schematic diagram of the experimental apparatus used to measure the vaporization. (1) reactor; (2) temperature controller; (3) thermo couple;
(4) cooling water; (5) pressure indicator; (6) electric bend; (7) cooling coil; (8) gas supply valve; (9) NH3 analyzer; (10) pressure data logger; (11)
computer.

Journal of Environmental Sciences 2012, 24(3) 494498 / Jong-Beom Seo et al.

496

2 Results and discussion

9 wt.% NH3
9 wt.% NH3 + 1 wt.% Glycine
9 wt.% NH3 + 1 wt.% Ethylene glycol
9 wt.% NH3 + 1 wt.% Glycerol

2.1 NH3 vaporization non-reacted with CO2

9 wt.% NH3
9 wt.% NH3 + 1 wt.% Ethylene glycol
9 wt.% NH3 + 1 wt.% Glycerol
9 wt.% NH3 + 1 wt.% Glycine
NH3 and water vaporization (kPa)

6
a

NH3 concentration (ppm)

a
40000

30000

20000

10000

10
15
20
Reaction time (min)

25

30

45000
b
40000
35000
30000
25000
20000
15000

10000
NH3

Glycine Ethylene glycol Glycerol

Fig. 3 NH3 gas concentration vs. time (a) and plots of median and
percentiles of NH3 gas concentration (b) for 9 wt.% NH3 and 1 wt.%
additives with CO2 absorption at 313 K.

20

40

60
80
Reaction time (min)

100

120

6
b
NH3 and water vaporization (kPa)

50000

NH3 concentration (ppm)

Figure 2a shows the vapor emitted from the aqueous

ammonia solutions and blended alcohol solutions with no
CO2 reaction. Glycerol, ethylene glycol and glycine played
the role of reducing vaporization at 313 K.
The 9 wt.% aqueous ammonia solutions vaporized within the range 0 to 5.28 kPa. On the addition of 1 wt.%
ethylene glycol to the aqueous ammonia solution, the
vapor was generated within the range 0 to 4.36 kPa. While,
with 1 wt.% glycerol and 1 wt.% glycine, the vaporization
occurred within the ranges 0 to 1.52 kPa and 0 to 4.58 kPa,
respectively. Figure 2b presents the stable vaporization
after 1 hr in the form of a box plot, with the plotted data
represented as the median and percentiles. As a result, the
median of the 9 wt.% aqueous ammonia was 5.18 kPa,
and those of the blended solutions with the additions of
ethylene glycol, glycerol and glycine were 4.00, 1.34 and
4.21 kPa, respectively. The reduction in the vaporization
due to the addition of glycerol exhibited the most eective

Vol. 24

result compared to the 9 wt.% aqueous ammonia with the

additions of 1 wt.% ethylene glycol or glycine, which were
decreased by 22.7% and 18.6%, respectively. The rate of
decrease induced by 1 wt.% glycerol was 74.1%, which
was higher than those with the other additives.
Also, during the initial thirty minutes, the rate of vaporization increase of the 9 wt.% aqueous ammonia was 0.18
kPa/min; whereas, those of the blended solutions with 1
wt.% ethylene glycol or glycine were similar, at 0.11 and
0.12 kPa/min. In the case of 1 wt.% glycerol, this value
was lowest, at 0.11 kPa/min.
2.2 Loss rate of ammonia gas during absorption of CO2

NH3 Ethylene glycol Glycerol

Glycine

Fig. 2 NH3 and water vaporization vs. time (a) and plots of the median
and percentiles of NH3 and water vaporization (b) for 9 wt.% NH3 + 1
wt.% additives that had not reacted with CO2 at 313 K.

Figure 3a shows the ammonia gas concentration discharged from the batch reactor with CO2 absorption,
which was measured using an ammonia analyzer. The
concentration of CO2 gas that reacted with the aqueous
ammonia solutions and blended alcohol solutions was
20.0 kPa, which was the same as that for coke oven
gas. The experiments generally lasted for 30 min. The
outlet ammonia concentration is also shown for the 9 wt.%
aqueous ammonia, with the additions of 1 wt.% glycine,

Vaporization reduction characteristics of aqueous ammonia solutions by the addition of ethylene glycol, glycerol

3
a

0
0

20

40
60
80
Reaction time (min)

100

120

3.0
b
2.5

2.0

1.5

1.0

0.5

2.3 NH3 vaporization loaded with 1 mol/L CO2

Figure 4a shows the vapor emitted from the aqueous
ammonia solutions and blended alcohol solutions loaded
1 mol/L CO2 . The 9 wt.% aqueous ammonia solutions
vaporized within the range 0 to 2.72 kPa, decreases of
between 0 to 48.5% compared with no CO2 reaction. With
the addition of 1 wt.% ethylene glycol, glycerol or glycine,
the vaporizations were within the ranges of 0 to 1.14, 0 to
0.15, and 0 to 0.27 kPa, respectively.
Figure 4b presents the stable vaporization after 1 hour
in the form of box plots using the plotted data as the
median and percentiles. The median of the 9 wt.% aqueous
ammonia solution was 2.62 kPa, and those of the solutions
blended with 1wt.% ethylene glycol, glycerol and glycine
were 1.12, 0.12 and 2.44 kPa, respectively. Compared with
the no CO2 reaction, the median ammonia gas concentrations on the addition of 1 wt.% ethylene glycol and 1
wt.% glycine decreased by 72.0% and 42.2%, respectively.
The rate of decrease induced by 1 wt.% glycerol was
91.3%, which exhibited the most eective reduction of
vaporization.

Table 1

497

9 wt.% NH3
9 wt.% NH3 + 1 wt.% Glycine
9 wt.% NH3 + 1 wt.% Ethylene glycol
9 wt.% NH3 + 1 wt.% Glycerol
NH3 and water vaporization (kPa)

ethylene glycol and glycerol, respectively.

The initial ammonia gas concentration at the outlet for
the reaction of 20 kPa CO2 and aqueous ammonia solution
was 41800 ppm, which stabilized at 34300 ppm after
30 min. On the addition of ethylene glycol, the initial
ammonia gas and the stabilized ammonia gas concentrations were 22700 and 21100 ppm, respectively. On the
addition of glycine, the initial and stabilized ammonia gas
concentrations were 40300 and 32550 ppm, respectively.
On the addition of glycerol, the initial and stabilized
ammonia gas concentrations were the lowest, at 14500 and
13750 ppm, respectively.
When the data were plotted as the median and percentiles around the box plot (Fig. 3b), the median ammonia
gas concentrations for the 9 wt.% aqueous ammonia solution, and those with the additions of either 1 wt.% ethylene
glycol, glycine or glycerol were 37300, 36000, 21400 and
13450 ppm, respectively. Compared with the 9 wt.% aqueous ammonia, the median ammonia gas concentrations
on the additions of 1 wt.% ethylene glycol and 1 wt.%
glycine decreased by 3.4% and 42.6%, respectively. The
rate of decrease induced by 1 wt.% glycerol was 63.9%,
which was higher than those for the other additives. The
absorption on the addition of 1 wt.% glycerol was most
eective at reducing the ammonia concentration, which
was similarly to that of the solution that had not reacted
with CO2 .

NH3 and water vaporization (kPa)

No. 3

0.0

NH3

Glcine Ethylene glycol Glycerol

Fig. 4 NH3 and water vaporization vs. time (a) and plots for the median
and percentiles of NH3 and water vaporization (b) for 9 wt.% NH3 + 1
wt.% additives loaded with 1 mol/L CO2 at 313 K.

The absorption of CO2 into an aqueous ammonia solution cause the crystallized CO2 -containing ammonium
salts to be produced, including ammonium carbonate
((NH4 )2 CO3 ), ammonium bicarbonate (NH4 HCO3 ) and
ammonium carbamate (NH2 COONH4 ). As a result, the
ammonium ions, which cause the vaporization of NH3 ,
were reduced and subsequently reduced the vapor pressure
of the solution.
To compare the result for blended conditions, the data
for median of vaporization in continuous CO2 absorption
system are listed in Table 1. The aqueous ammonia solution adding glycerol was an most eective material to
reduce vaporization in continuous CO2 absorption system.

Median of vaporization in continuous CO2 absorption system

Median

Non-reacted with CO2 (kPa)

Initial absorption with CO2 (ppm)
Loaded with 1 mol/L CO2 (kPa)
Temperature 313 K.

9 wt.%
NH3

9 wt.% NH3 +
1 wt.% ethylene glycol

9 wt.% NH3 +
1 wt.% glycine

9 wt.% NH3 +
1 wt.% glycerol

5.1860
37300
2.6210

4.0065
21400
1.1206

4.2170
36000
2.4389

1.3400
13400
0.1165

498

Journal of Environmental Sciences 2012, 24(3) 494498 / Jong-Beom Seo et al.

Vol. 24

3 Conclusions

References

For the aqueous ammonia solution that had not reacted

with CO2 , the vaporization on the addition of 1 wt.%
glycerol decreased by 74.1% compared to that of the 9
wt.% aqueous ammonia solution. During the absorption
of CO2 , the rate of decrease induced by 1 wt.% glycerol
was 63.9%. For the ammonia solution loaded with 1 mol/L
CO2 , rate of decrease induced by 1 wt.% glycerol was
91.3%. During all stages, the reduction in the vaporization
induced by glycerol was shown to be the most eective
compared to those with other additives. It is possible to
show the before CO2 absorption, the initial CO2 absorption
and the stabilization phase during the entire process, as
well as predict the vapor generated in the CO2 treatment
process.
In conclusion, an ammonia liquor solution for the absorption of CO2 is highly recommended for many work
place, since glycerol could be an eective material for the
economical operation of an absorption system.

Choi W J, Seo J B, Jang S Y, Jung J H, Oh K J, 2009. Removal

characteristics of CO2 using aqueous MEA/AMP solutions
in the absorption and regeneration process. Journal of
Environmental Sciences, 21(7): 907913.
Diao Y F, Zheng X Y, He B S, Chen C H, Xu X C, 2004.
Experimental study on capturing CO2 greenhouse gas by
ammonia scrubbing. Energy Conversion & Management,
45(13-14): 22832296.
Kim J Y, Han K W, Chun H W, 2009. CO2 absorption with low
concentration ammonia liquor. Energy Procedia, 1(1): 757
762.
Resnik K P, Yeh J T, Pennline H W, 2004. Aqua ammonia process
for simultaneous removal of CO2 , SO2 , NOx. International
Journal of Environmental Technology and Management,
4(1-2): 89104.
Seo J B, Jeon S B, Choi W J, Kim J W, Lee G H, Oh
K J, 2011. The absorption rate of CO2 /SO2 /NO2 into a
blended aqueous AMP/ammonia solution. Korean Journal
of Chemical Engineering, 28(1): 170177.
Yeh A C, Bai H L, 1999. Comparison of ammonia and monoethanolamine solvents to reduce CO2 greenhouse gas
emissions. Science of the Total Environment, 228(2-3):
121133.
Yeh J T, Resnik K P, Pennline H W, 2004. Regenerable aqua ammonia process for CO2 sequestration. American Chemical
Society, Division of Fuel Chemistry, 49(1): 247248.
You J K, Park H S, Hong W H, Park J K, Kim J N, 2008. Inuence of additives including amine and hydroxyl groups on
aqueous ammonia absorbent for CO2 capture. The Journal
of Physical Chemistry B, 112(14): 43234328.

Acknowledgments
This work was supported by a grant from the Energy
Eciency & Resources of the Korea Institute of Energy
Technology Evaluation and Planning (KETEP), funded by
the Korean Governments Ministry of Knowledge Economy (No. 20092010200011-12-1-000) and by the Korean
Ministry of the Environment (MOE) as the Human Resource Development Project for Waste to Energy.