pubs.acs.

org/IECR Article

Nonaqueous MEA/PEG200 Absorbent with High Efficiency and Low

Energy Consumption for CO2 Capture
Wen Tian, Kui Ma, Junyi Ji, Siyang Tang, Shan Zhong, Changjun Liu, Hairong Yue, and Bin Liang*
Cite This: Ind. Eng. Chem. Res. 2021, 60, 3871−3880 Read Online

ACCESS Metrics & More Article Recommendations *

sı Supporting Information
Downloaded via NATL TAIWAN UNIV SCIENCE & TECHLGY on October 18, 2024 at 16:08:02 (UTC).
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

ABSTRACT: CO2 capture by stripping with aqueous amine is a

well-understood and widely employed technology. However, there
still exist several drawbacks in the practical applications, such as
high energy consumption and serious equipment corrosion. In this
paper, an organic monoethanolamine/poly(ethylene glycol) 200
(MEA/PEG200) absorbent with low specific heat capacity, low
vapor pressure, and a high boiling point was developed to
substitute the aqueous amine systems. The absorption rate and
capacity, the desorption rate, and regeneration efficiency, as well as
the thermal stability of the absorbent were systematically
investigated. For instance, the 3 M MEA/PEG200 absorbent
exhibited the highest desorption rate (838 mgCO2/(Lsol·min)) and
excellent cyclic stability (deviation < 4%), which was much superior to the 3 M MEA/H2O absorbent (447 mgCO2/(Lsol·min)). In
addition, the desorption energy consumption of the MEA/PEG200 absorbent was 33% lower than that of the MEA/H2O. Moreover,
both fresh and CO2-rich MEA/PEG200 solutions exhibited only slight corrosion to carbon steels even after being immersed for 400
days. Therefore, this low-energy-consuming and anticorrosive MEA/PEG200 absorbent reveals the potential for industrial CO2
capture.

1. INTRODUCTION The current research studies on a CO2 capture solvent

Global annual energy-related CO2 emissions rose to a historic mainly focus on developing an absorbent with a high
high value of 33.1 Gt in 2018.1 The excessive amount of CO2 absorption rate, large capacity, low specific heat capacity, and
low saturated vapor pressure. For example, ionic liquids
emission has a significant impact on agriculture, water
possess large CO2 absorption capacity, but the high cost and
resources, ecosystem, and human health. It is of great
high viscosity hinder their industrial applications.11,12 Phase-
significance to promote the low-carbon technologies to reduce
change absorbents can automatically separate a CO2-rich-phase
the CO2 emissions. Particularly, the CO2 capture, utilization,
from a poor-phase solution; the energy consumption can be
and storage (CCUS) is considered as one of the most effective
reduced by cutting the amount of the absorbent loaded in a
methods to achieve CO2 emission reduction.2 Alkanolamine
regenerator.13,14 However, phase changeable solvents are still
solvent absorption from the industrial flue gas is one of the
expensive, and the CO2-rich liquid phase generally possesses
dominant CO2 capture technologies due to its high separation
high viscosity, which is not conducive to desorption.15
efficiency, good selectivity, and cycling stability. Meanwhile,
Nonaqueous solutions, especially the organic alcohol solvents
the purity of CO2 regenerated from an organic amine solvent is
[e.g., triethylene glycol, propylene glycol, ethylene glycol, and
relatively high, which can be applied as a raw material to
poly(ethylene glycol) (PEG)], are considered as one class of
further produce the downstream products, such as carbonate,
the promising solvents. The polar hydroxyl groups on the
urea, organic chemicals, etc.3,4 Aqueous amines are a major organic alcohol solvents have higher affinity for CO 2
solvent group for commercial absorption. However, they molecules, thereby improving the CO2 capture capacity.16−18
possess large specific heat capacity, high sensible heat and On the other hand, the organic alcohols usually exhibit lower
latent heat of vaporization, and high decomposition temper-
ature (120−150 °C) in regeneration.5 The heat duty for
solvent regeneration can constitute up to 70% of the total Received: October 29, 2020
operating costs in a CO2 capture plant.6−8 Meanwhile, the Revised: February 20, 2021
amine aqueous solution is also corrosive for the equipment.9,10 Accepted: February 23, 2021
Therefore, developing various new solvents is crucial to Published: March 8, 2021
improve the energy efficiency and operation safety during the
CO2 capture process.

© 2021 American Chemical Society https://dx.doi.org/10.1021/acs.iecr.0c05294

3871 Ind. Eng. Chem. Res. 2021, 60, 3871−3880
Industrial & Engineering Chemistry Research pubs.acs.org/IECR Article

isobaric specific heat capacity compared with that of the water,

which can effectively decrease the desorption energy
consumption.19−21 However, solvents such as methanol,
ethylene glycol (vapor pressure of 133 Pa, 20 °C), and
propanediol (viscosity of 330 mPa·s, 40 °C) have short-
comings of high vapor pressure, toxicity, and high viscosity,
which are not suitable for industrial operations. Therefore, the
physical properties of solvents are crucial factors affecting their
incorporation into industrial applications.
Poly(ethylene glycols) (PEGs), a kind of polar solvents, are
frequently used as cosolvents in chemical and pharmaceutical
processes. The end hydroxyl group forms hydrogen bonds Figure 1. Schematic diagram of the CO2 absorption and desorption
between molecules, resulting in low saturated vapor pressure device.
and high thermal stability. The highly polar hydroxyl group of
PEGs can also interact with CO2 molecules to promote the
absorption.18,22 Li et al. previously utilized the PEGs as 2.2.2. Desorption. A CO2-rich solution with a CO2/MEA
cosolvents to prepare the low concentrated amine/PEG200 molar ratio of 0.45−0.50 was desorbed in a three-necked flask,
and ionic liquid/PEGs absorbents, which exhibited excellent in which the CO2-rich solution was heated to 80 °C under
absorption and desorption performance.19,23,24 Among various magnetic agitation at 300 rpm. N2 was used as a purge gas at a
PEG substances, the diethylene glycol and triethylene glycol flow rate of 200 mL/min. The off gas was detected until the
are generally toxic, while PEG200 possesses excellent proper- CO2 concentration stabilized at less than 1.0% for 10 min. The
ties such as nontoxicity and no stimulation. In addition, the regenerated solution was also analyzed via titration analysis.
viscosity of PEGs increases with the molecular weights (e.g., 2.2.3. Corrosion Evaluation. C20 steel was selected as the
25.05 mPa·s for PEG200, 34.94 mPa·s for PEG300, and 44.4 probe material to evaluate the corrosion behavior of the
mPa·s for PEG400 at 313 K). The increase of viscosity leads to absorbent system. The weight of each C20 sample was 1.40−
the enhancement of mass-transfer resistance between a gas and 1.50 g and the size was about 8.0 × 30.0 × 1.0 mm. The
liquid interface. Therefore, PEG200 is chosen as an auxiliary samples were immersed in the solutions (both the fresh and
solvent when considering the nonaqueous amine solution. CO2-rich MEA/PEG200 or MEA/H2O) for 30, 120, and 400
Since these absorbents are developed for the industrial days, respectively.
applications, in this paper, monoethanolamine (MEA) with 2.3. Calculation and Analysis Methods. 2.3.1. Absorp-
relatively low viscosity (10.02 mPa·s at 313 K),25 a fast CO2 tion and Desorption Calculation. The absorption and
absorption rate, and low cost is selected to prepare the MEA/ desorption rates of CO2 were calculated by eqs 1 and 2,
PEG200 absorbents. The CO2 absorption rate and capacity, respectively. The CO2 content of absorbed and desorbed
desorption rate and efficiency, cyclic stability, and the viscosity solutions can be obtained by eqs 3 and 4, respectively. The
of the absorbent that varies with the absorption process are volumetric CO2 content in a liquid phase was also determined
systematically investigated, and the performance of MEA/ by titration and calculated by eq 5. The desorption ratio was
PEG200 is compared with that of MEA/H2O. Finally, the defined as the ratio of CO2 contents in the solutions before
isobaric specific heat capacity, reaction heat, and thermal and after desorption (eq 6).
stability of the 5 M MEA/PEG200 absorbent are also in out
measured to evaluate the desorption energy consumption, as VSFG(yCO − yCO ) PMCO2
2 2
well as the corrosion behavior of absorbents on carbon steel. rabs = out ×
1− yCO RTVsol (1)
2

2. EXPERIMENTAL SECTION out

VN2yCO PMCO2
2
2.1. Materials. Poly(ethylene glycol) 200 (PEG200, AR) rdes = out ×
1 − yCO RTVsol (2)
and monoethanolamine (MEA, 99%) were purchased from 2

Chengdu Kelong Reagent co. Ltd. Nitrogen (99.99%) and t

carbon dioxide (99.99%) were purchased from Chengdu ∫0 rabsdt
Qiaoyuan Gas Plant. αabs =
MCO2 × Vsol × CMEA (3)
2.2. Apparatus and Procedures. 2.2.1. Absorption.
MEA/PEG200 and MEA/H2O absorbents were prepared t
with the MEA molar concentrations of 0.5, 1, 2, 3, and 5 mol/ ∫0 rdesdt
L, respectively. As shown in Figure 1, 100 mL of the as- αdes =
MCO2 × Vsol × CMEA (4)
prepared solution was transferred into a three-necked flask
(250 mL) and heated to 40 °C in a water bath with magnetic ′
VL − CO2 = V2 − V1 − V HCl (5)
stirring at 300 rpm. A feeding stream of mixed gas (CO2
volume fraction of 15%) was bubbled through the solution at a αCO2 ‐ rich − αCO2 ‐ poor
constant flow rate of 300 mL/min. The off gas of absorption η=
αCO2 ‐ rich (6)
was measured using an infrared CO2 analyzer (FN316B,
Shanxi Feiente Instrument Technology Co., Ltd.). The 2.3.2. Estimation of CO2 Desorption Energy Consump-
absorption experiment ceased after the CO2 concentration of tion. The CO2 desorption is the major energy-consuming
the off gas stabilized for 10 min. The CO2 content absorbed in process in CO2 capture.27 The energy consumption of
the solution was determined by titration analysis.26 desorption can be calculated by eq 7, including sensible heat
3872 https://dx.doi.org/10.1021/acs.iecr.0c05294
Ind. Eng. Chem. Res. 2021, 60, 3871−3880
Industrial & Engineering Chemistry Research pubs.acs.org/IECR Article

Figure 2. CO2 absorption rates and capacity of the MEA/PEG200 and MEA/H2O absorbents, respectively. (a) 0.5 M, (b) 1 M, (c) 2 M, (d) 3 M,
(e) and 5 M. (f) CO2 capacity of absorbents at different MEA concentrations determined by titration analysis.

of solution, reaction heat of the desorption process, and latent (Y2 − Y1)
heat of solvent vaporization. q=
44 (12)
Q = Hsensible + Hreaction + H vapor (7) ΔHadsorbent
ΔHabs,CO2 = × 10−3
For 1 kg of CO2, the sensible heat of solution was calculated by q (13)
eqs 8−10 The latent heat of evaporation, as shown in eq 14, the partial
Hsensible = msolv C PΔT (8)
pressure ratio of the volatile solvent, and CO2 can be calculated
by the method proposed by Wang et al.29
1000 Psol 1
msolv = × η × αrich × C MEA × ρsolv H vapor = ΔHν ,sol × ×
MCO2 (9) PCO2 MCO2 (14)
2.3.3. Analysis Method. 2.3.3.1. Determination of
c P = c MEA × ϕ + csolvent × (1 − ϕ) (10) Reaction Heat and Thermal Capacity. Differential scanning
from 20 to 92 °C was performed on a calorimeter (NETZSCH
According to the literature,28 the reaction heat was calculated DSC214) to measure the isobaric specific heat capacity of the
by eqs 11−13, solvent at a heating rate of 2 °C/min under N2 gas protection
mCO2 (20 mL/min).30 As the gas stream containing CO2 bubbled
Hreaction = ΔHabs,CO2 through the absorbents at 40 °C with a flow rate of 20 mL/
MCO2 (11) min, a synchronous thermal analyzer (NETZSCH STA 449
3873 https://dx.doi.org/10.1021/acs.iecr.0c05294
Ind. Eng. Chem. Res. 2021, 60, 3871−3880
Industrial & Engineering Chemistry Research pubs.acs.org/IECR Article

Figure 3. Viscosity changes of the MEA/PEG200 and MEA/H2O solutions under different CO2 and MEA concentrations at 313 K.

F3) was employed to measure the weight change and heat When H2O exists in the solution, RHNCOO− is hydrolyzed to
effect of absorbents during absorption. Therefore, the reaction form RNH2. Therefore, from the reaction formulas eqs 16-1
heat and volatility could be estimated. Thermogravimetric and 16-2, the reaction stoichiometry in the aqueous solution is
(TG) analysis of the fresh and CO2-rich absorbents was 1:1, that is, the maximum absorption amounts of 1 mol
evaluated in 5 mL/min N2 at 80 °C for 60 min. alkanolamine is 1 mol CO2. Therefore, in principle, MEA/H2O
2.3.3.2. Determination of Vapor Pressure. The vapor could absorb more CO2 than MEA/PEG200.
pressure of the absorbents was tested using an automatic vapor
RHNCOO− + H 2O ⇔ RNH 2 + HCO−3 (16-1)
pressure tester (HY8017 Jilin Hong Yuan Scientific Instrument
Co., Ltd.).
2.3.3.3. Viscosity Measurement. The CO2-absorbed rich- CO2 + RNH 2 + H 2O ⇔ RHNH+2 + HCO−3 (16-2)
phase solution and the original MEA/PEG200 solution were
Figure 2a−e shows the CO2 absorption rate and capacity of
mixed to obtain the solutions with CO2 content of 0.10−0.45
MEA/PEG200 and MEA/H2O absorbents with the MEA
mol/mol MEA. The viscosity was measured using a rotating
concentration in the range of 0.5−5 M. When the MEA
viscometer (Shanghai Changji Geological Instrument Co.,
concentration was 0.5 and 1 M, the initial absorption rates
Ltd.) under 40 °C.
2.3.3.4. Inductively Coupled Plasma Optical Emission (492 and 614 mgCO2/(Lsol·min)) in the MEA/PEG200
Spectrometer (ICP-OES) Analysis. The iron contents of these absorbents were lower than that of MEA/H2O absorbents
solutions were detected by an inductively coupled plasma (667 and 722 mgCO2/(Lsol·min)), as shown in Figure 2a,b.
emission spectrometer (ICP-OES) using an Arco’s detector Notably, when the MEA concentration exceeded 2 M (Figure
(Spectro). 2c−e), the initial absorption rate of MEA/PEG200 absorbents
2.3.3.5. Determination of the Polarization Curve. The was comparable to that of the MEA/H2O absorbents (∼720
corrosion current density of C20 steel in different solutions mgCO2/(Lsol·min)). The increase in the MEA concentration
was tested by the electrochemical method. The polarization may lead to the high MEA molecule density in the solution,
curve (Tafel curve) was determined using the electrochemical which can shorten the CO2 diffusion path and lower the
workstation (CHI 760E Shanghai Chenhua Instrument Co., diffusion resistance. Figure 2f reveals the CO2 absorption
Ltd.) capacity of absorbents with varying MEA concentrations. The
2.3.3.6. 13C NMR Analysis. 13C NMR spectra were obtained CO2 absorption capacity in the MEA/PEG200 absorbents with
using a JNM ECZ400S/L1 spectrometer. To provide enough 0.5, 1, 2, 3, and 5 M are 0.48, 0.46, 0.44, 0.46, and 0.45
signals for the deuterium lock, deuterated dimethyl sulfoxide molCO2/molMEA, respectively, which is close to the theoretical
(DMSO-d6) or D2O was introduced into the NMR tube
containing the fresh and CO2-rich absorbents. saturated capacity (0.50 molCO2/molMEA). On the contrary, the
saturated CO 2 absorption capacity in the MEA/H2 O
3. RESULTS AND DISCUSSION absorbents decreases with the increase of the MEA
3.1. CO2 Absorption Rate and Capacity. According to concentration, and the actual capacity is about 50% of their
the zwitterion mechanism, primary amine can react with CO2 theoretical capacity (1.0 molCO2/molMEA) when the MEA
to form the zwitterions.31−33 Under nonaqueous conditions, concentration exceeds 3 M. The low capacity may be caused
the hydrogen ion transfer occurs between the zwitterion and by the insufficient hydrolyzation of the RHNCOO− in aqueous
the amine (MEA) to produce RHNH2+ and RHNCOO−, as solutions. With the increase of the MEA concentration, the
indicated in eqs 15-1−15-3. When the nonaqueous solvent is ratio of hydrolyzed RHNCOO− becomes smaller, thus the
employed, the stoichiometric coefficient of amine and CO2 is corresponding saturated capacity of MEA/H2O decreases.
2:1 in the reaction. In addition, the CO2 absorption rate of MEA/PEG200
absorbents decreases rapidly compared with that of the MEA/
RNH 2 + CO2 ⇔RNH+2 COO− (zwitterion) (15-1) H2O absorbents. As shown in Figure 3, the viscosity of MEA/
PEG200 absorbents exhibits an order of magnitude higher than
RNH 2 + RNH+2 COO− ⇔ RHNH+2 + RHNCOO− that of the MEA/H2O absorbents, which further increases
(15-2) significantly with the elevated CO2 content. The viscosity of
the 0.5 M MEA/PEG200 solution increased from 28.4 to 52.4
2RNH 2 + CO2 ⇔ RHNH+2 + RHNCOO− (15-3) mPa·s, as the content of CO2 in the solution increased from 0
3874 https://dx.doi.org/10.1021/acs.iecr.0c05294
Ind. Eng. Chem. Res. 2021, 60, 3871−3880
Industrial & Engineering Chemistry Research pubs.acs.org/IECR Article

Figure 4. CO2 desorption rate and the desorption ratio of the CO2-rich MEA/PEG200 and CO2-rich MEA/H2O, respectively. (a) 0.5 M; (b) 1 M;
(c) 2 M; (d) 3 M; (e) 5 M; and the (f) desorption ratio of the CO2-rich absorbents at different MEA concentrations.

to 0.45 mol/molMEA. Moreover, under the same CO2 content, PEG200. When the concentration further increases to 5 M, the
the viscosity of the 5 M MEA/PEG200 solution increased from maximal desorption rate exhibits a certain decrease (782
40.2 to 481 mPa·s. In comparison, the viscosity of MEA/H2O mgCO2/(Lsol·min)), which may be due to the sluggish diffusion
solutions remains below 5.0 mPa·s in any condition. The high efficiency of the desorption generated CO2 from the high-
viscosity values result in high gas diffusion resistance, thereby viscosity solution. However, the desorption rate is still higher
reducing the CO2 absorption rate. than that of the CO2-rich MEA/H2O absorbents, which may
3.2. CO2 Desorption Rate and Efficiency. The CO2 be due to the much lower decomposition energy barrier of a
desorption rate and efficiency of the CO2-rich MEA/PEG200 nonaqueous solution than that of an aqueous solution.34
and MEA/H2O absorbents are shown in Figure 4. The initial The desorption efficiencies of the MEA/PEG200 and MEA/
desorption reaction temperature (60 °C) of the MEA/PEG200
H2O absorbents are calculated by eq 6 and shown in Figure 4f.
absorbents is detected to be lower than that of the MEA/H2O
With the increase of the MEA concentration, the desorption
(70 °C). The maximum CO2 desorption rate of MEA/PEG200
efficiency of MEA/PEG200 decreases from 88.0% (0.5 M, 50
(0.5 M) is 323 mgCO2/(Lsol·min), which is greater than that of min) to 44.4% (5.0 M, 120 min), which is still higher than that
the MEA/H 2 O (129 mg CO 2 /(L sol ·min)). Notably, the of the corresponding MEA/H2O absorbents (decreased from
desorption rates of absorbents with varying MEA concen- 43.0% (0.5 M, 50 min) to 26.0% (5.0 M, 120 min)). The
trations reveal a volcanic-like trend along with the reaction absorbents with a high MEA concentration generally possess
time (Figure 4a−e). With the increase of the MEA high viscosity, which restricts the mass transfer between the
concentration in a CO2-rich solution, the maximal CO2 CO2 gas and liquid phases, thus resulting in the decreased
desorption rate increases; the highest maximal desorption desorption efficiency of the desorbed CO2 gas in the viscous
rate of 838 mgCO2/(Lsol·min) can be obtained for 3 M MEA/ solution.
3875 https://dx.doi.org/10.1021/acs.iecr.0c05294
Ind. Eng. Chem. Res. 2021, 60, 3871−3880
Industrial & Engineering Chemistry Research pubs.acs.org/IECR Article

Figure 5. Cyclic absorption and desorption rates of the MEA/PEG200 and MEA/H2O at different concentrations. (a) 0.5 M; (b) 1 M; (c) 2 M;
(d) 3 M; and (e) 5 M.

3.3. Cyclic Absorption and Desorption Ability. To temperature is about 80 °C and the evaporation latent heat
further evaluate the cyclic performance of the absorbents, the greatly affects the energy consumption. Figure 6 exhibits the
cyclic absorption and desorption efficiency of all absorbents are TG curves of the fresh and CO2-rich absorbents in a N2
investigated. As shown in Figure 5a−d, the absorption atmosphere at 80 °C for 60 min. The results indicate a weight
capacities of the 0.5−3 M MEA/PEG200 absorbents are
loss of 10.19% for the fresh 5 M MEA/PEG200 absorbent and
stable within three cycles, while the desorption efficiency
presents a high level (about 85.0% for 0.5 to 2 M and 75.0%
for 3 M), especially under the MEA concentration lower than
0.1 molCO2/molMEA. Figure 5e reveals that the desorption
efficiency of the 5 M MEA/PEG200 remains above 55.0% after
passing through two cycles. By contrast, a high value of the
CO2 residue (0.25−0.30 molCO2/molMEA) is detected in all of
the MEA/H2O absorbents after desorption, leading to a low
cyclic desorption efficiency. Therefore, the MEA/PEG200
absorbents exhibit high cyclic stability and high CO 2
absorption/desorption efficiency, and the results are also
shown in Table S1.
3.4. Absorbent Thermal Stability. The thermal stability
of the absorbent is also crucial for a continuous CO2 Figure 6. Thermogravimetric analysis curves of the fresh and CO2-
absorption and desorption process since the desorption rich absorbents (5 M MEA/PEG200 and 5 M MEA/H2O).

3876 https://dx.doi.org/10.1021/acs.iecr.0c05294
Ind. Eng. Chem. Res. 2021, 60, 3871−3880
Industrial & Engineering Chemistry Research pubs.acs.org/IECR Article

19.98% for the CO2-rich one, attributing to the partial CO2 1795 kJ/kgCO2 for MEA/H2O. This slight difference lies in the
desorption. However, a significant weight loss is observed with different reaction intermediates (Section 3.6 in detail).
a value of 73.98 and 81.87% for the fresh and CO2-rich 5 M 3.5.3. Latent Heat during the Desorption Process. The
MEA/H2O absorbents, respectively. As shown in Table 1, the vaporization latent heat of the solution during desorption is the
phase-change heat. As the desorption temperature (80 °C) is
Table 1. Saturated Vapor Pressure of Different Solutions lower than the vaporization temperature of each component in
T (°C) PS (Pa) the solution, the sensible heat is the main energy input in the
process. However, some absorbent solutions can also be
MEA 80 223735
removed during the continuous escape of the CO2 gas flow. As
H2O 80 47370
shown in eq 14, the vaporization latent heat of the solution is
PEG200 85 9.936
mainly determined by partial pressures of the outlet CO2 and
5 M MEA/PEG200 80 1800
the solution. Since the 5 M MEA/PEG200 absorbent with a
5 M MEA/H2O 80 26 700
high boiling point and low volatility exhibits excellent thermal
stability (Table 1 and Figure 6), the latent heat of the solution
saturated vapor pressure of the MEA is 2237 Pa (80 °C).35 vaporization can be ignored. On the contrary, according to the
Additionally, the saturated vapor pressure of PEG200 is only calculation method proposed by Wang et al.,29 the latent heat
9.9 Pa (95 °C),36 which is much lower than that of water of vaporization of 5 M MEA/H2O is 65.59 kJ/kgCO2 at the
(47 370 Pa, 80 °C). Thus, when PEG200 is used as the
cosolvent, the saturated vapor pressure of the 5 M MEA/ temperature of 80 °C. The previous work has reported that the
PEG200 solution (1800 Pa, 80 °C) is much lower than that of capture of 1 kg of CO2 would produce 0.03 kg of steam,21
the 5 M MEA/H2O solution (26 700 Pa, 80 °C). The low which is in line with the results in this work.
volatility of the MEA/PEG200 absorbent can significantly As summarized in Table 2, the estimated energy
decrease the mass loss ratio of the absorbent and effectively consumption for CO2 regeneration in the 5 M MEA/
reduce the latent heat during the carbon capture process. PEG200 solution is 2.547 MJ/kgCO2, which is 33% lower
3.5. Estimation of the Desorption Energy Consump- than that of the MEA/H2O (3.821 MJ/kgCO2). Moreover, the
tion. To obtain the energy consumption in the regeneration loss of organic amines during the desorption process can also
process, the energy consumption of the CO2-rich MEA/ be reduced when utilizing PEG200 as a cosolvent.
PEG200 and MEA/H2O absorbents is evaluated during the 3.6. Corrosion Evaluation of Carbon Steel. The
desorption process using 1 kg of CO2 as the basis (Figure S1). corrosion of the absorbents on the C20 steel is an important
The energy consumption during the desorption is mainly parameter to determine the equipment safety and operation
composed of the sensible heat of solution, reaction heat of the cost of the process. The corrosion behaviors of the fresh and
desorption process, and latent heat of vaporization.27 CO2-rich absorbents are tested by immersing the carbon steel
3.5.1. Sensible Heat of the Solution. The cyclic stability plates in the solutions for 30, 120, and 400 days, respectively.
test results (Figure 5e) are used to evaluate the sensible heat As shown in Figure S4, the MEA/PEG200 absorbents exhibit a
during the desorption process. The average isobaric specific slightly corrosive effect on the steel, and the solution remains
heat capacity (CP) of the PEG200 solution detected by clear after immersion for 400 days (Figure 7a). On the
differential scanning calorimetry (DSC) at 30−90 °C is 2.37 contrary, the steel samples in the MEA/H2O absorbents are
kJ/(kg·K) (Figure S2), which is consistent with the reported seriously corroded and the solution is obviously dyed. The
value (CP = 2.13 kJ/(kg·K) at 60 °C).21 In addition, the CP, polarization curves of C20 steel in different solutions are also
MEA value of 3.36 kJ/(kg·K) can also be observed. According shown in Figure 7b. The relationship between the current
to the eqs 8−10, the calculated sensible heat values (Table 2) density of the solutions is MEA/PEG200 < CO2-rich MEA/
of MEA/PEG200 and MEA/H2O absorbents are 1052 and PEG200 < MEA/H 2O < CO2-rich MEA/H 2O, which
1961 kJ/kgCO2, respectively. corresponds to the conductivity value of the solutions (Figure
7a). According to Faraday’s law, there is a strict correspond-
Table 2. Desorption Energy Consumption of the Two ence between the metal corrosion rate and self-corrosion
Absorbents (kJ/kgCO2) current. As shown in Figure 7a, ICP results also confirmed the
slight corrosion on carbon steel in the MEA/PEG200
MEA/H2O MEA/PEG200 solutions. The concentrations of the Fe ions are only 25.2
MEA concentration 5M 5M and 12.6 mg/L in the MEA/PEG200 and CO2-rich MEA/
sensible heat 1961.00 1052.00 PEG200 solutions after 30 days, respectively, while that in
reaction enthalpy 1795.0038 1495.00 MEA/H2O and CO2-rich MEA/H2O are 31.8 and 483.0 mg/
latent heat of vaporization 65.59 0.00 L, respectively. By prolonging the immersion period, only a
total energy consumption for regeneration 3821.59 2547.00 tiny increment of Fe ions is detected in the fresh (32.0 mg/L)
and CO2-rich (45.0 mg/L) MEA/PEG200 solutions, respec-
tively, indicating the excellent anticorrosion ability of the
3.5.2. Reaction Heat during the Desorption Process. The MEA/PEG200 solution. However, in MEA containing aqueous
desorption reaction heat in the CO2-rich 5 M MEA/PEG200 solutions, the concentrations of Fe ions of fresh and CO2-rich
absorbent is measured by TG and DSC. The calculated MEA/H2O are 714.0 and 2114.0 mg/L, respectively. These
reaction heat is 65.8 kJ/mol for the CO2 desorption reaction in high values also suggest the serious corrosion of the steel.
the CO2-loaded (0.4 molCO2/molMEA) MEA/PEG200 absorb- Moreover, the weight values of the C20 steel before and after
ent (Figure S3), while that of the MEA/H2O absorbent is 79− immersing for different times are shown in Table S2, and the
84 kJ/mol.22,23 The reaction heat values are also listed in Table results are in line with the ICP results. The corrosion
2, that is, 1495 kJ/kgCO2 for MEA/PEG200 absorbents and experiments further reveal that the MEA/PEG200 absorbent
3877 https://dx.doi.org/10.1021/acs.iecr.0c05294
Ind. Eng. Chem. Res. 2021, 60, 3871−3880
Industrial & Engineering Chemistry Research pubs.acs.org/IECR Article

Figure 7. (a) ICP results of the Fe concentration after various C20 steel immersion times in various solutions. (a) Electrical conductivity and (b)
polarization curves of the various solutions.

Figure 8. Summary of the reaction mechanism, desorption ratio after 120 min, and total regeneration energy consumption of CO2 captured by 5 M
MEA/PEG200 and 5 M MEA/H2O.

may lower the cost of anticorrosion materials required from MJ/kg), which is mainly due to the different sensible heat
the equipment for industrial CO2 capture. values of the aqueous and nonaqueous absorbents.
3.7. Reaction Mechanism in MEA/PEG200 and MEA/
H2O Absorbents. To further confirm the difference in the 4. CONCLUSIONS
reaction mechanism between the CO2 trapping process in the
aqueous and nonaqueous absorbents, the qualitative analysis of A nonaqueous MEA/PEG200 absorbent is successfully
the NMR is performed. As shown in Figures S5 and S6, only developed for CO2 capture using an organic solvent. The
RHNCOO− is generated during the CO2 capture process using absorption rate and capacity, desorption rate and efficiency,
the MEA/PEG200 absorbent. However, the products of the and cyclic stability of the absorbents are systemically
CO2-rich MEA/H2O absorbent are more complicated, investigated in the MEA/PEG200 sorbent and the results are
including RHNCOO−and HCO3−/CO32−, which is consistent compared with that of the MEA/H2O absorbent. The MEA/
with the refs 20 and 37 and eqs 15-1−15-3 and 16-1−16-2. To PEG200 absorbent with the MEA concentration above 3 mol/
summarize the typical difference between the MEA/PEG200
L exhibits the highest desorption rate (838 mgCO2/(Lsol·min))
and MEA/H2O absorbents, the reaction routes, the desorption
ratio after 120 min, and total energy consumption are and excellent cyclic stability (deviation < 4%), which is much
proposed in Figure 8. First, the aqueous and nonaqueous superior to the MEA/H2O absorbent (the desorption rate of
absorbents exhibit different reaction routes, and more 447 mgCO2/(Lsol·min)). In addition, the energy consumption
complicated products are produced in the aqueous absorbent for CO2 regeneration in the 5 M MEA/PEG200 solution is
in the presence of H2O. Second, the desorption ratio after 120
33% lower than that in the MEA/H2O solution. Moreover, low
min of the MEA/PEG200 absorbent is obviously higher than
that of the MEA/H2O system (44.4 and 26.0%, respectively), corrosion on carbon steel is found during 400-day immersion
which may be due to the much lower decomposition energy with a weight loss of 0.65 and 0.68% for fresh and CO2-rich
barrier of a nonaqueous solution.34 Finally, the total desorption MEA/PEG200 solutions, respectively. Therefore, this low-
energy consumption of the MEA/PEG200 absorbent (2.547 energy-consuming and anticorrosion MEA/PEG200 absorbent
MJ/kg) is lower than that of the MEA/H2O system (3.821 possesses enormous potential for CO2 capture in the industry.
3878 https://dx.doi.org/10.1021/acs.iecr.0c05294
Ind. Eng. Chem. Res. 2021, 60, 3871−3880
Industrial & Engineering Chemistry Research

■ ■
pubs.acs.org/IECR Article

ASSOCIATED CONTENT NOMENCLATURE

*
sı Supporting Information
rabs absorption rate of CO2 (mgCO2/(Lsol·min))
The Supporting Information is available free of charge at rdes desorption rate of CO2 (mgCO2/(Lsol·min))
https://pubs.acs.org/doi/10.1021/acs.iecr.0c05294.
VSFG inlet gas flow (mL/min)
The flow chart of the CO2 capture process, the DSC yinCO2 molar concentration of CO2 imported
curve, the TG-DSC diagram, the optical images, 13C yout
CO2 molar concentration of CO2 exported
NRM spectra, cyclic performance, and the corrosion rate V N2 N2 flow (mL/min)
(PDF) αabs absorption of CO2 (mol/molMEA)

■
αdes desorption volume of CO2 (mol/molMEA)
AUTHOR INFORMATION VL‑CO2 content of CO2 in the liquid phase (mL)
η desorption efficiency of CO2
Corresponding Author Hsensible sensible heat of solution (kJ/kg)
Bin Liang − Low-Carbon Technology and Chemical Reaction msolv solution quality (kg)
Engineering Laboratory, School of Chemical Engineering and αrich capacity of CO2 in the rich liquid (molCO2/molMEA)
Institute of New Energy and Low-Carbon Technology,
Sichuan University, Chengdu 610065, China; orcid.org/ Csolv molar content of MEA in the solution (mol/L)
0000-0003-2942-4686; Phone: (+86) 28-85990278; ρsolv density of the mixed solution (kg/L)
Email: [email protected] MCO2 molecular weight of CO2
cP specific heat of the solution (kJ/kg·K)
Authors φ the mass fraction of MEA
Wen Tian − Low-Carbon Technology and Chemical Reaction Hreaction heat of the reaction (kJ/kg)
Engineering Laboratory, School of Chemical Engineering, q the molar amount of CO2 per gram of the absorbent
Sichuan University, Chengdu 610065, China (molCO2/g)
Kui Ma − Low-Carbon Technology and Chemical Reaction ΔHabs,CO2 heat of the reaction per mole of CO2 (kJ/mol)
Engineering Laboratory, School of Chemical Engineering, PCO2 CO2 partial pressure (Pa)
Sichuan University, Chengdu 610065, China; orcid.org/ T temperature (K)

■
0000-0002-5667-4755
Junyi Ji − Low-Carbon Technology and Chemical Reaction
Engineering Laboratory, School of Chemical Engineering, REFERENCES
Sichuan University, Chengdu 610065, China; orcid.org/ (1) Energy Industry Review. Global Energy & CO2 Status Report
0000-0002-5545-1357 2019. In The Latest Trends in Energy and Emissions in 2019; IEA, 2019.
Siyang Tang − Low-Carbon Technology and Chemical (2) Matter, J. M.; Stute, M.; Snaebjornsdottir, S. O.; Oelkers, E. H.;
Reaction Engineering Laboratory, School of Chemical Gislason, S. R.; Aradottir, E. S.; Sigfusson, B.; Gunnarsson, I.;
Engineering, Sichuan University, Chengdu 610065, China; Sigurdardottir, H.; Gunnlaugsson, E.; Axelsson, G.; Alfredsson, H. A.;
orcid.org/0000-0003-4757-3992 Wolff-Boenisch, D.; Mesfin, K.; Taya, D. F. D. l. R.; Hall, J.;
Dideriksen, K.; Broecker, W. S. Rapid carbon mineralization for
Shan Zhong − Low-Carbon Technology and Chemical permanent disposal of anthropogenic carbon dioxide emissions.
Reaction Engineering Laboratory, School of Chemical Science 2016, 352, 1312−1314.
Engineering, Sichuan University, Chengdu 610065, China; (3) Wang, W.; Liu, X.; Wang, P.; Zheng, Y.; Wang, M. Enhancement
orcid.org/0000-0002-0247-6788 of CO2 Mineralization in Ca2+-/Mg2+-Rich Aqueous Solutions Using
Changjun Liu − Low-Carbon Technology and Chemical Insoluble Amine. Ind. Eng. Chem. Res. 2013, 52, 8028−8033.
Reaction Engineering Laboratory, School of Chemical (4) Shangguan, W.; Song, J.; Yue, H.; Tang, S.; Liu, C.; Li, C.; Liang,
Engineering and Institute of New Energy and Low-Carbon B.; Xie, H. An efficient milling-assisted technology for K-feldspar
Technology, Sichuan University, Chengdu 610065, China; processing, industrial waste treatment and CO2 mineralization. Chem.
orcid.org/0000-0003-3735-4112 Eng. J. 2016, 292, 255−263.
Hairong Yue − Low-Carbon Technology and Chemical (5) Feng, B.; Du, M.; Dennis, T. J.; Anthony, K.; Perumal, M. J.
Reduction of Energy Requirement of CO2 Desorption by Adding
Reaction Engineering Laboratory, School of Chemical Acid into CO2-Loaded Solvent. Energy Fuels 2010, 24, 213−219.
Engineering and Institute of New Energy and Low-Carbon (6) Idem, R.; Wilson, M.; Tontiwachwuthikul, P.; Chakma, A.;
Technology, Sichuan University, Chengdu 610065, China; Veawab, A.; Aroonwilas, A.; Gelowitz, D. Pilot Plant Studies of the
orcid.org/0000-0002-9558-0516 CO2 Capture Performance of Aqueous MEA and Mixed MEA/
Complete contact information is available at: MDEA Solvents at the University of Regina CO2 Capture Technology
https://pubs.acs.org/10.1021/acs.iecr.0c05294 Development Plant and the Boundary Dam CO 2 Capture
Demonstration Plant. Ind. Eng. Chem. Res. 2006, 45, 2414−2420.
(7) Chen, S. J.; Zhu, M.; Fu, Y.; Huang, Y. X.; Tao, Z. C.; Li, W. L.
Notes Using 13X, LiX, and LiPdAgX zeolites for CO2 capture from post-
The authors declare no competing financial interest. combustion flue gas. Appl. Energy 2017, 191, 87−98.

■ ACKNOWLEDGMENTS
The authors are grateful for the support from the National Key
(8) Yue, H.; Zhao, Y.; Zhao, L.; Lv, J.; Wang, S.; Gong, J.; Ma, X.
Hydrogenation of dimethyl oxalate to ethylene glycol on a Cu/SiO2/
cordierite monolithic catalyst: Enhanced internal mass transfer and
stability. AIChE J. 2012, 58, 2798−2809.
R&D Program of China (Grant No. 2018YFB0605700) and (9) Kittel, J.; Gonzalez, S. Corrosion in CO2 Post-Combustion
the National Natural Science Foundation of China Capture with Alkanolamines - A Review. Oil Gas Sci. Technol. 2014,
(22008157). We also appreciate Dr. Yanping Huang from 69, 915−929.
Center of Engineering Experimental Teaching for the NMR (10) Kittel, J.; Fleury, E.; Vuillemin, B.; Gonzalez, S.; Ropital, F.;
measurement and analysis. Oltra, R. Corrosion in alkanolamine used for acid gas removal: From

3879 https://dx.doi.org/10.1021/acs.iecr.0c05294
Ind. Eng. Chem. Res. 2021, 60, 3871−3880
Industrial & Engineering Chemistry Research pubs.acs.org/IECR Article

natural gas processing to CO2 capture. Mater. Corros. 2012, 63, 223− thermodynamic cycle could be used as a tool for energy efficiency
230. analysis. Energy 2019, 168, 380−392.
(11) Hasib-ur-Rahman, M.; Siaj, M.; Larachi, F. Ionic liquids for (30) Piekarski, H.; Tkaczyk, M.; Tyczyńska, M. Heat capacity and
CO2 capture-Development and progress. Chem. Eng. Process. 2010, 49, phase behavior of aqueous diethylene glycol n-pentyl ether by DSC.
313−322. Thermochim. Acta 2012, 550, 19−26.
(12) Zhang, X.; Zhang, X.; Dong, H.; Zhao, Z.; Zhang, S.; Huang, Y. (31) Jing, G.; Zhou, L.; Zhou, Z. Characterization and kinetics of
Carbon capture with ionic liquids: overview and progress. Energy carbon dioxide absorption into aqueous tetramethylammonium
Environ. Sci. 2012, 5, 6668−6681. glycinate solution. Chem. Eng. J. 2012, 181−182, 85−92.
(13) Zhu, K.; Lu, H.; Liu, C.; Wu, K.; Jiang, W.; Cheng, J.; Tang, S.; (32) Caplow, M. Kinetics of Carbamate Formation and Breakdown.
Yue, H.; Liu, Y.; Liang, B. Investigation on the Phase-Change J. Am. Chem. Soc. 1968, 6795−6803.
Absorbent System MEA plus Solvent A (SA) + H2O Used for the (33) Blauwhoff, P. M. M.; Versteeg, G. F.; Swaaij, W. P. M. V. A
CO2 Capture from Flue Gas. Ind. Eng. Chem. Res. 2019, 58, 3811− Study on the Reaction Between CO2 and Alkanolamines in Aqueous
3821. Solutions. Chem. Eng. Sci. 1984, 39, 207−225.
(14) Zhang, S.; Shen, Y.; Wang, L.; Chen, J.; Lu, Y. Phase change (34) Zhang, T. T.; Yu, Y. S.; Zhang, Z. X. Effects of non-aqueous
solvents for post-combustion CO2 capture: Principle, advances, and solvents on CO2 absorption in monoethanolamine: Ab initio
challenges. Appl. Energy 2019, 239, 876−897. calculations. Mol. Simul. 2018, 44, 815−825.
(15) Machida, H.; Ando, R.; Esaki, T.; Yamaguchi, T.; Horizoe, H.; (35) Avlund, A. S.; Kontogeorgis, G. M.; Michelsen, M. L. Modeling
Kishimoto, A.; Akiyama, K.; Nishimura, M. Low temperature swing systems containing alkanolamines with the CPA equation of state. Ind.
process for CO2 absorption-desorption using phase separation CO2 Eng. Chem. Res. 2008, 47, 7441−7446.
(36) Aschenbrenner, O.; Supasitmongkol, S.; Taylor, M.; Styring, P.
capture solvent. Int. J. Greenhouse Gas Control 2018, 75, 1−7.
Measurement of vapour pressures of ionic liquids and other low
(16) Shamiri, A.; Shafeeyan, M. S.; Tee, H. C.; Leo, C. Y.; Aroua, M.
vapour pressure solvents. Green Chem. 2009, 11, 1217−1221.
K.; Aghamohammadi, N. Absorption of CO2 into aqueous mixtures
(37) Ciftja, A. F.; Hartono, A.; Svendsen, H. F. C-13 NMR as a
of glycerol and monoethanolamine. J. Nat. Gas Sci. Eng. 2016, 35,
method species determination in CO2 absorbent systems. Int. J.
605−613. Greenhouse Gas Control 2013, 16, 224−232.
(17) Kang, M.-K.; Cho, J.-H.; Lee, J.-H.; Lee, S.-S.; Oh, K.-J. Kinetic (38) Song, H.-J.; Lee, S.; Park, K.; Lee, J.; Chand Spah, D.; Park, J.-
Reaction Characteristics of Quasi-Aqueous and Nonaqueous Sorbents W.; Filburn, T. P. Simplified Estimation of Regeneration Energy of 30
for CO2 Absorption Using MEA/H2O/Ethylene Glycol. Energy Fuels wt % Sodium Glycinate Solution for Carbon Dioxide Absorption. Ind.
2017, 31, 8383−8391. Eng. Chem. Res. 2008, 47, 9925−9930.
(18) Aschenbrenner, O.; Styring, P. Comparative study of solvent
properties for carbon dioxide absorption. Energy Environ. Sci. 2010, 3,
1106−1113.
(19) Li, J.; You, C.; Chen, L.; Ye, Y.; Qi, Z.; Sundmacher, K.
Dynamics of CO2 Absorption and Desorption Processes in Alkanol-
amine with Cosolvent Poly(ethylene glycol). Ind. Eng. Chem. Res.
2012, 51, 12081−12088.
(20) Guo, H.; Li, C.; Shi, X.; Li, H.; Shen, S. Nonaqueous amine-
based absorbents for energy efficient CO2 capture. Appl. Energy 2019,
239, 725−734.
(21) Guo, H.; Li, H.; Shen, S. CO2 Capture by Water-Lean Amino
Acid Salts: Absorption Performance and Mechanism. Energy Fuels
2018, 32, 6943−6954.
(22) Wiesmet, V.; Weidner, E.; Behme, S.; Sadowski, G.; Arlt, W.
Measurement and modelling of high-pressure phase equilibria in the
systems polyethyleneglycol (PEG)−propane, PEG−nitrogen and
PEG−carbon dioxide. J. Supercrit. Fluids 2000, 17, 1−12.
(23) Li, J.; Dai, Z.; Usman, M.; Qi, Z.; Deng, L. CO2/H2 separation
by amino-acid ionic liquids with poly(ethylene glycol) as co-solvent.
Int. J. Greenhouse Gas Control 2016, 45, 207−215.
(24) Li, J.; Chen, L.; Ye, Y.; Qi, Z. Solubility of CO2 in the Mixed
Solvent System of Alkanolamines and Poly(ethylene glycol) 200. J.
Chem. Eng. Data 2014, 59, 1781−1787.
(25) Song, J. H.; Park, S. B.; Yoon, J. H.; Lee, H. Densities and
viscosities of monoethanolamine plus ethylene glycol plus water. J.
Chem. Eng. Data 1996, 41, 1152−1154.
(26) Liang, Z.; Idem, R.; Tontiwachwuthikul, P.; Yu, F.; Liu, H.;
Rongwong, W. Experimental study on the solvent regeneration of a
CO2-loaded MEA solution using single and hybrid solid acid catalysts.
AIChE J. 2016, 62, 753−765.
(27) Oexmann, J.; Kather, A. Minimising the regeneration heat duty
of post-combustion CO2 capture by wet chemical absorption: The
misguided focus on low heat of absorption solvents. Int. J. Greenhouse
Gas Control 2010, 4, 36−43.
(28) Liu, Z. S.; Peng, Y. H.; Huang, C. Y.; Hung, M. J. Application of
thermogravimetry and differential scanning calorimetry for the
evaluation of CO2 adsorption on chemically modified adsorbents.
Thermochim. Acta 2015, 602, 8−14.
(29) Wang, J.; Sun, T.; Zhao, J.; Deng, S.; Li, K.; Xu, Y.; Fu, J.
Thermodynamic considerations on MEA absorption: Whether

3880 https://dx.doi.org/10.1021/acs.iecr.0c05294
Ind. Eng. Chem. Res. 2021, 60, 3871−3880