Article

Cite This: Ind. Eng. Chem. Res. 2017, 56, 15094−15101 pubs.acs.org/IECR

Energy Consumption and Exergy Analysis of MEA-Based and

Hydrate-Based CO2 Separation
Nan Xie,† Bin Chen,‡ Chenghua Tan,† and Zhiqiang Liu*,†,§
†
School of Energy Science and Engineering, Central South University, Changsha 410083, China
‡
AECC Hunan Aviation Powerplant Research Institute, Zhuzhou 412002, China
§
Collaborative Innovation Center of Building Energy Conservation & Environmental Control, Zhuzhou 412007, China
*
S Supporting Information

ABSTRACT: Carbon capture and storage (CCS) is regarded

as the most efficient approach in handling the global warming
crisis. MEA-based CO2 capture is a well-developed chemical
absorption method with a long period of industrial application.
A novel hydrate-based gas separation (HBGS) method, with a
wide range of advantages, has recently received special attention
from researchers. In this study, two different CO2 separation
processes were simulated utilizing Aspen Plus software. The fea-
sibility of both processes was validated, and the process energy
consumption and exergy loss were also compared at the same
flue gas condition. Some efforts have also been made to inves-
tigate the effects of different operation parameters on the process energy efficiency. Results show that the first law efficiency of
the MEA-based CO2 separation system is 88.19% and the second law efficiency of the system is 38.32%, while the corresponding
values of the hydrate-based separation system are 74.15% and 38.85%, respectively. Cooling of the lean amine solution and the
regeneration process occupy the largest portion of exergy loss in the MEA separation system. In the hydration separation, the flue
gas compression and cooling are the major causes for exergy loss.

1. INTRODUCTION sizes of the gas molecules. Most research is focused on the

Rising CO2 emission has led to various environmental issues, formation/dissociation conditions, enthalpy, additives, and
and global warming, most of all, remains a major concern phase equilibrium.9−19 Kumar et al.10 investigated the effects
among governments and scientists. It is common sense to see of thermodynamic and kinetic promoters, in which the pro-
that the anthropogenic emission of CO2 is mainly due to the moters used in this study were sodium dodecyl sulfate (SDS)
burning of fossil fuels, especially the combustion process in and tetrahydrofuran (THF). Their results showed the gas
large power plants. Carbon capture and storage (CCS) tech- uptake was increased by almost three-fold in the presence of
nology is now seen as the most valid solution to reduce CO2 kinetic promoter SDS. The hydrate formation pressure was
emission, essentially known as precombustion, oxy-combustion, obviously reduced with the addition of thermodynamic pro-
and postcombustion methods. Physical absorption, chemical moter THF. Three different ternary systems of CO2 + water +
absorption, pressure swing adsorption, cryogenic condensation, (TBAB/TBAC/TBAF) were reported by Li et al.11 An iso-
and membrane separation are common techniques in capturing choric pressure-search method was utilized to measure the
carbon dioxide, particularly the MEA-based absorption.1,2 As a equilibrium data of these semiclathrate hydrates. The lowest
well-developed amine-based absorption technique, it has a long formative pressure at the same temperature was found in the
history of industrialization. However, it is still hindered by its system of tetra-n-butyl ammonium fluoride (TBAF). The syn-
own weakness, such as the energy intensive process of solvent ergic effect of TBAB in the presence of both cyclopentane and
regeneration, high corrosion on equipment, etc.3,4 dodecyl trimethylammonium chloride (DTAC) was carefully
Hydrate-based gas separation (HBGS) is now considered as examined by Li et al.12,13 Gas uptakes, CO2 selectivity, CO2 sep-
a novel way to avoid the massive atmospheric buildup of CO2. aration efficiencies, induced time, and hydrate formation rate
Hydrate-based CO2 capture is realized by the selective partition were also investigated by them.20 They found that the CO2
of CO2 between the hydrate phase and the gaseous phase. Gas concentration reaches approximately 93% by separation involv-
hydrates, also known as clathrate hydrates, are nonstoichiometric, ing only one stage. Similar investigations were made by many
crystal compounds which have cagelike structures formed by
water and gas molecules.5 Structure identification and inves- Received: September 10, 2017
tigation are carried out by researchers to better understand the Revised: November 20, 2017
nature of gas hydrates.6−8 Generally, there are three different Accepted: November 29, 2017
kinds of crystalline structures (sI, sII, sH) depending on the Published: November 29, 2017

© 2017 American Chemical Society 15094 DOI: 10.1021/acs.iecr.7b03729

Ind. Eng. Chem. Res. 2017, 56, 15094−15101
Industrial & Engineering Chemistry Research Article

other researchers,13,21−23 who discovered that the captured CO2 absorbent on equipment is neglected; (2) compression, drying,
concentration in all these studies can be enriched to a relatively and transportation processes of products are also ignored;
high level in only one or two separation stages. (3) there is no heat dissipation in the heat transfer process;
A comprehensive review was given to the properties of (4) the main reactions of the present separation process are given
CO2 hydrates and the hydrate-based CCS by Ma et al.24 This in Table 2.2,30
technology is still at its early stage because of the compli- Gaseous CO2 forms a semiclathrate hydrate with the pro-
cated and stochastic process of hydrate formation and diverse moter tetra-n-butyl ammonium fluoride (TBAF). In this crystal
experimental conditions. Hence, they suggested that large-scale structure, a part of the cage structure is broken to encage the
system design and economic and environmental impact ana- large ion TBA+. And the fluoride anion can form a polyhedral
lysis should be further conducted. Xu and Li made original hydrogen-bonded cage structure with water molecules and
comments and suggestions on further development of the change the structure of this crystal inclusion compound, mod-
hydrate-based CCS technology, including equilibrium condi- erating the formation condition as a result. Basic assumptions
tions, additives, molecular-level measurement methods, process of the hydrate-based CO2 capture process are given as follows:
optimization, new apparatuses and cost, etc.25 Energy consump- (1) The pressure of flue gas after the first compression is
tion estimation of the HBGS process was carried out by Tajima assumed to be 0.5 MPa, and 2.5 MPa after the second com-
et al.26 Hydration separation processes of three greenhouse pression, and the compression is considered as an adiabatic
gases were studied and compared with liquefaction separation process with the efficiency of 0.8; (2) the minimum tempera-
processes in terms of the energy consumed. They found that pure ture difference in exchangers is specified at 10 °C and the COP
CO2 hydrate separation consumed a huge amount of energy due (coefficient of performance) of the refrigeration unit is equal
to the extremely high formation pressure. This high pressure can to 3; (3) the hydrate dissociation temperature is set to be
be dramatically alleviated by adding appropriate additives. 25 °C; (4) compression work imposed on the feed gas is
After reviewing the literature, we found no research on the recovered by the expansion of the off-gas stream in a turbine of
comparison between the MEA-based and the hydrate-based which the efficiency is 0.8; (5) the molar fraction of the additive
separation processes. Thus, it is essential to make a compre- TBAF is assumed to be 0.29%; thus, the dissociation enthalpy
hensive comparison in terms of energy efficiency. As a com- of the CO2 + TBAF hydrate is estimated as 144 kJ/mol using
monly used tool for simulating various CO2 capture processes the Clausius−Clapeyron equation based on the available equilib-
in power plants,27−29 Aspen Plus is utilized in this study as well. rium data.
Based on the simulated results, the energy consumption esti-
mation and the exergy analysis of a two-stage hydrated-based 3. PROCESS SIMULATION
CO2 separation process are carried out. The results are then 3.1. System description. Simulation diagrams of two sep-
compared in detail with those of a MEA-based separation aration processes are shown in Figure 1 and Figure 2, respec-
process. Furthermore, sensitivity studies of various operational tively. In Figure 1, the feed gas is cooled to the absorption
factors which influence the process energy efficiencies are per- temperature at first and then introduced into the ABSORBER,
formed as well. Our research presents a thorough comparison in which the CO2 is absorbed by a lean solution to form a CO2-
between a well-developed and a novel CO2 separation tech- rich solution. The clean flue gas leaves the vessel from the
nology, providing results with a deeper understanding of CCS upper outlet. The carbon-rich solvent is then pumped into the
technology. This work will also be useful in the energy con- exchanger (EXCHANGE) and heated up by the lean solution
servation and the process optimization of CCS technology in issuing from the STRIPPER to an appropriate temperature for
large power plants. regeneration. After leaving the exchanger, the rich CO2 solvent
is heated by the superheated steam from the reboiler to release
2. BASIS PARAMETERS OF FLUE GAS AND PROCESS its burden of CO2. The CO2-rich gas exits the stripper after
ASSUMPTIONS condensation as the product of the current system. The regen-
2.1. Flue gas parameters. Based on the flue gas emitted erated solution (CO2-lean solution) is cooled (in COOLER2)
from a typical 1000 MW thermal power plant after desul- to the absorbing temperature and enters the absorber again to
furization and denitration, the temperature and the pressure of conduct further work. The MEA solution loss is supplemented
the flue gas are fixed at 140 °C and 0.1 MPa, respectively. The by using a MIXER.
concentration of CO2 in the flue gas is assumed to be 16%, In Figure 2, a two-stage separation process is described to
which is within the reasonable range of 14% to 18%. The produce gaseous CO2 with a concentration higher than 90% by
composition and the simplified parameters of the flue gas are volume. The high temperature flue gas is precooled by an
shown in Table 1. ambient fluid before it enters the compressor. After the first
2.2. Process assumptions. Some reasonable assumptions compression in an adiabatic compressor (SCOMP1), the pres-
are currently made in the simulation of two different CO2 surized gas mixture is cooled again to a given temperature by
separation processes. The basic assumptions of the MEA-based the lean CO2 gas emitted from the hydrate formation reactor
CO2 capture process are given as (1) corrosion behavior of the (FORMAT1). And then, the feed gas is adiabatically pres-
surized for the second time to reach the formation pressure.
Table 1. Basis Parameters of the Flue Gas in Present Study The feed gas is cooled to the formation temperature by using a
refrigeration unit (REFRIG1) and is then introduced into the
Mole Molar Feed rate Feed rate hydrate reactor to form a CO2 hydrate in the TBAF solution.
Component fraction mass (kmol/h) (t/h) In the dissociation reactor (DISSOC1), the hydrate is heated
CO2 0.16 44 7143 314 up by an ambient fluid until it breaks down to the CO2-rich gas
N2 0.71 28 31697 888 and the CO2-lean solvent. The latter is piped back to the
O2 0.05 32 2232 71 formation unit to maintain the cycle operation. The former is
H2O 0.08 18 3571 64 pressurized once again to undergo another similar process to a
15095 DOI: 10.1021/acs.iecr.7b03729
Ind. Eng. Chem. Res. 2017, 56, 15094−15101
Industrial & Engineering Chemistry Research Article

Table 2. Reactions of the MEA-Based CO2 Capture Process2,30

Coefficients for calculating the equilibrium constants (Keq) in the following equations:
No. Type Reaction equations ln(Keq) = A + B/T + C* ln(T) + D*T
1 Equilibrium MEA+ + H2O ↔ C2H7NO + H3O+ A = −3.038, B = −7008.36, C = 0, D = −0.0031
2 Equilibrium CO2 + 2H2O ↔ H3O+ + HCO3− A = 231.47, B = −12092.1, C = −36.78, D = 0
3 Equilibrium HCO3− + H2O ↔ H3O+ + CO3−2 A = 216.05, B = −12431.7, C = −35.48, D = 0
4 Equilibrium 2H2O ↔ H3O+ + OH− A = 132.899, B = −13445.9, C = −22.48, D = 0
5 Equilibrium MEACOO− + H2O ↔ C2H7NO + HCO3− A = −0.52, B = −2545.53, C = 0, D = 0

Figure 1. Process flowsheet of the MEA-based CO2 separation.

Figure 2. Process flowsheet of the hydrate-based CO2 separation.

higher concentration. In the current system, part of the com- Table 3. Detailed Configurations of the Absorber and the
pression work imposed on the feed gas is recovered from the Stripper
off-gas by using a TURBINE.
absorber stripper
3.2. Selection of model. Thermodynamics and kinetics of
Model type RadFrac RadFrac
the MEA-based CO2 capture system are modeled using the
Calculation type Equilibrium Equilibrium
“MEA Property Insert” of Aspen Plus software. Thus, the
Number of stages 10 10
electrolyte−NRTL model is currently utilized to describe the Condenser type None Partial-Vapor
MEA−water−CO2 system. Besides, Elec Wizard in Aspen Plus Reboiler type None Kettle
is employed to describe the complex chemistry and ther- Reflux 1.2
modynamics of the MEA property. GMELCC-1, GMELCD-1, Entry position On-Stage On-Stage
GMELCE-1, and GMELCN-1 (electrolyte−molecules/electro- Stage 1/Condenser pressure (kPa) 220 69
lyte−electrolyte binary energy parameters C, D, E, and non- Column pressure drop (kPa) 10
randomness parameter alpha in the electrolyte NRTL activity
coefficient model) are modified for the present research. The Table 4. Block Types and Parameters of Individual Device in
ELECNTRL model is utilized to handle various problems the MEA-Based Process
involving aqueous electrolyte solution for different concentra- Name Type Specification
tions. Detailed configurations of the absorber and the stripper
PUMP Pump Discharge pressure, 0.26 MPa
are listed in Table 3. The block types and the parameters of COOLER1 HeatX Hot stream outlet temperature, 40 °C
other individual equipment are shown in Table 4. EXCHANGE HeatX Cold stream outlet temperature, 92 °C
The properties of the streams in the hydrate-based separation COOLER2 HeatX Hot stream outlet temperature, 40 °C
system of interest are simulated by using the PRMHV2 model.
Based on the Peng−Robinson equation of state and the mod- degree of predictive accuracy for high pressure cases (>150 bar),
ified Huron−Vidal mixing rule, this model has an excellent like the streams in the current process. Table 5 shows the
15096 DOI: 10.1021/acs.iecr.7b03729
Ind. Eng. Chem. Res. 2017, 56, 15094−15101
Industrial & Engineering Chemistry Research Article

Table 5. Block Types and Parameters of Individual Device in Ex in − (Exout + Ir) = W (5)
the Hydrate-Based Process
where Exin represents the exergy that enters the system, kJ; and
Name Type Specification Exout means the exergy that leaves the system, kJ; Ir refers to
SCOMP1 Compr Isentropic; Discharge pressure, 0.5 MPa; the exergy loss and also represents the degradation of energy
Mechanical efficiency 0.8 caused by process dissipation, kJ; W is the net work output of a
SCOMP2 Compr Isentropic; Discharge pressure, 2.5 MPa; system or device, kJ. Exergy efficiency, also known as the
Mechanical efficiency 0.8
SCOMP3 Compr Isentropic; Discharge pressure, 0.9 MPa;
second law efficiency, can be calculated by
Mechanical efficiency 0.8
ηex = W /Extotal = 1 − ηex,loss = 1 − Ir /Extotal (6)
COOLER HeatX Hot stream outlet temperature, 50 °C
EXCHANGE1 HeatX Hot stream outlet temperature, 90 °C where Extotal represents the total exergy input into the system, kJ;
EXCHANGE2 HeatX Hot stream outlet temperature, 60 °C
and ηex,loss means the exergy loss efficiency, which shows the
TURBINE Compr Isentropic; Mechanical efficiency, 0.8
distribution of exergy loss in a certain process. Therefore, the
REFRIG1 HeatX Hot stream outlet temperature, 4.5 °C
REFRIG2 HeatX Hot stream outlet temperature, 4.5 °C
exergy analysis for both CO2 separation processes is conducted
based on the results obtained in the simulation and then
block types and the simulation parameters of the hydrate-based compared and discussed in detail. Simulation results of both
separation. processes for specific calculations of exergy loss are illustrated
in Figure S1 and Figure S2 in the Supporting Information.
4. ENERGY CONSUMPTION ESTIMATION AND
EXERGY ANALYSIS METHOD 5. RESULTS AND DISCUSSION
4.1. Energy balance algorithm. It is a fundamental 5.1. MEA-based CO2 separation process. Energy balance
principle that energy cannot be created or destroyed, but it can and distribution of the MEA-based CO2 separation system
be transferred or transformed into different forms with its total (MEA system) are carefully calculated (details are shown in
amount unchanged. Based on the first law of thermodynamics, Table S1 in the Supporting Information), at the reference state
the energy balance can be described as T0 = 298.15 K, P0 = 1 atm. The major energy input is the super-
heated steam (Stream 12) from the reboiler, which occupies
Q in − Q out = ΔQ (1) 78.29% of the total energy input. Only 0.18% of the total
where Qin represents the energy entering the system, kJ; and energy is taken away by the regenerating gas (Stream 16), while
Qout means the energy that leaves the system, kJ; ΔQ is the 36.59% of the total energy output is caused by the cooling
energy change in the system, kJ. The thermal efficiency can be water (Stream 20) leaving the cooler. And the overall first-law
defined as efficiency of the MEA system is 88.19%. Exergy balance and
distribution are also calculated based on the simulation results
ηt = W /Q total (2) (details in Table S2). It can be found that 78.93% of the total
exergy input is also achieved by the superheated steam. The
where W means the useful work, kJ; and Qtotal is the energy
maximum exergy loss exists in the processes of the MEA solvent
consumed in the system, kJ. Thus, the thermal efficiency is
regeneration and the regenerating gas cooling in the stripper
currently adopted as a major evaluation index of the present
(37.14% of the total). The exergy loss of all devices is 61.68% of
study. Simulation results of both processes for energy consump-
the total; hence, the exergy efficiency of the current system is
tion calculations are illustrated in Figure S1 and Figure S2 in
only 38.32%.
the Supporting Information.
According to our energy consumption estimation and exergy
4.2. Exergy analysis method. On the basis of the second
analysis, the main consumption in this process is the heating of
law of thermodynamics, the exergy efficiency is also taken as a
the CO2-rich solution (Stream 6) to release its burden of target
major criterion to evaluate the quality of the energy, defined as
gas. So, energy saving in the stripper is the key problem. High
the maximum potential of the energy that can be converted into
cost of the heating process is determined by the essential char-
useful work. In the processes of our interest, the total exergy
acteristics of the absorbents. Thus, searching for new absor-
consists of physical exergy and chemical exergy. Physical exergy
bents and new additives seems to be the only way, which is the
can be given as follows:
main constraint of this technique.
Exphys = − (H0 − H1) + T0(S0 − S1) (3) 5.2. Hydrate-based CO2 separation process. Composi-
tions of several typical streams in this process are shown in
where the H0 means the enthalpy of the stream at reference Figure 3. The concentration of gaseous CO2 is increased from
state (T0, P0), kJ; S0 means the entropy at reference state, kJ/K; 16% (mole fraction) in the flue gas (a) to 60% in the first-stage
H1 is the enthalpy of the stream, kJ; S1 is the entropy of the separated gas (c), and eventually to 90% in the second-stage
stream, kJ/K. And chemical exergy can be given as separated gas (e). The CO2 concentrations in the off-gas (b, d)
are very low, which proves the validity and feasibility of the
Exchem = ∑ xiEx0,i + ∫ RT0∑ xi ln xi CO2 separation by using hydrate. Energy consumption as well
i i (4)
as exergy analysis are carried out (as shown in Table S3 and
where xi represents the mole fraction of component i; Ex0,i is Table S4 of the Supporting Information). The electric power
the standard chemical exergy of component i, kJ; R represents for pressurizing and cooling the streams is the major energy
the universal gas constant, 8.314J/mol·K; T0 is the temperature input of the hydrate system. The chemical reaction heat needed
of the reference state, K. in the dissociation units can be provided by an environmental
All processes in nature are irreversible and, hence, lead to fluid because the endothermal CO2 hydrate dissociation can
exergy loss, which represents the irreversibility of the system or happen at ambient temperature. But the heat removal in the for-
process; thus, the exergy balance equation can be described as mation units can only be realized by consuming the refrigeration
15097 DOI: 10.1021/acs.iecr.7b03729
Ind. Eng. Chem. Res. 2017, 56, 15094−15101
Industrial & Engineering Chemistry Research Article

the flue gas to the formation pressure reaches more than one-
half of the total input.
The energy consumption per unit weight of CO2 is currently
estimated as well. The energy consumption per unit in the hydrate
system is 1.117 kWh/kg, being much lower than 2.848 kWh/kg
of the MEA system. This indicates that the separation process
by hydration has a significant advantage. These results are also
compared with previous literature, in which the energy con-
sumption per unit is 0.853 kWh/kg in the research of Tajima
et al.26 Our results are higher because the energy consumed
for cooling the flue gas before its compression is considered in
our research, and the separated CO2 flow rate in our study is
284 t/h higher than the value (186 t/h) in the literature. More-
Figure 3. Compositions of different streams: (a) flue gas; (b) off-gas over, the enthalpy in their study was 73 kJ/mol while the value
after the first-stage separation; (c) first-stage separated gas; (d) off-gas in our work is 144 kJ/mol because of the addition of TBAF.
after the second-stage separation; (e) second-stage separated gas. Figure 5 shows the exergy loss in each device. It can be
observed that the exergy loss has a concentrative distribution
for both processes. In the MEA system, a considerable amount
power 3 and 4 (in Table S3). Only 0.09% of the energy input is of exergy loss which can barely be recovered was taken away by
taken away by the gaseous mixtures (Stream 20) after the final the cooling process in the condenser. Hence the regeneration
separation process. A turbine is utilized to recover the energy in and the cooling in the stripper led to the major part of exergy
the first-stage off-gas (Stream 8), but still, around 3% leaves the loss. For this reason, further energy saving in the MEA system
system with the off-gas (Stream 9). The overall first-law might be hard to achieve. In the hydrate-based system, nearly
efficiency of this system is only 74.15%. In the exergy esti- one-third of the total exergy loss was found in the refrigeration
mation, the electric power input for the compression occupies units (23.62%) as well as the adiabatic compressors (8.62%).
more than 55% of the total input, but the maximum exergy loss Thus, to improve the energy efficiency of both systems, sen-
is in the refrigerators. A part of exergy loss is caused by the sitivity studies focusing on the stripper, compressors, and refrig-
emission of the off-gas (Stream 9 and 17). The exergy efficiency eration units are quite necessary.
of the hydrate separation system is 38.85%.
5.4. Sensitivity analysis. In the hydrate-based separation
Except for the heat-exchange network optimization aiming at
system, a large amount of exergy loss was caused by the cooling
a certain power plant, or the improvements on compressors
process in refrigeration units. There is a value to study the effect
and refrigerators, effective energy saving can also be realized by
moderating the hydrate formation condition using suitable of refrigeration performance on the process energy efficiency.
additives, which declares a remarkable potential of energy Thus, the effect of COP on the energy consumption per unit
conservation. weight of CO2 was investigated as shown in Figure 6. With the
5.3. Comparison of both separation processes. By increase of COP of refrigeration units, the energy consumption
contrast, the energy efficiency of the hydrated-based separation per unit goes down apparently, but this trend slows down with
system is lower than that of the MEA absorption system, but its continuous increase. Therefore, improving COP value is
with a slightly higher exergy efficiency. Higher exergy efficiency favorable in the hydrate-based system, but its effect is less
means a slower energy degradation. Figure 4 shows the dis- marked for COP larger than 2.5. This means pursuing an
tributions of the energy input in these two systems. The energy excessive COP value will be both ineffective and impractical.
distribution in the MEA system is quite uneven, and the Figure 7 shows the variation of the energy consumption per unit
thermal energy of the reboiler steam plays the dominant role in by improving the efficiency of compressors. The variation curve
the whole process, determined by the solution regeneration also presents a descending tendency with the rising efficiency of
of the absorbent. In the hydrate system, the distribution also compressors. Thus, energy conservation in this system can also
seems to be an uneven split. The electric power for pressurizing be achieved by improving the efficiency of compressors.

Figure 4. Input energy distribution of both CO2 separation processes: the MEA-based (left) and the hydrate-based (right).

15098 DOI: 10.1021/acs.iecr.7b03729

Ind. Eng. Chem. Res. 2017, 56, 15094−15101
Industrial & Engineering Chemistry Research Article

Figure 5. Exergy loss in different devices.

Figure 6. Effect of COP on the process energy consumption per unit.

Figure 8. Effect of solvent flow rate on the heat duty of the reboiler
and the COOLER2.

heat duty in the reboiler and the lean solvent cooler.

An obvious linear relation between the heat duty and the mass
flow rate was observed for both devices when increasing the mass
flow rate from 6100 t/h to 7000 t/h. The thermal energy from
the reboiler was partly absorbed by the rich CO2 solution as the
reaction heat was needed for maintaining the CO2 desorption
process. The rest was taken out of the system by the lean
solvent and the regenerating gas. In the lean solvent cooler, the
heat received from the reboiler was eventually taken away by an
environmental fluid. In the regenerating gas, a large amount of
water vapor and some volatile gases (in practical operation)
Figure 7. Effect of compressor efficiency on the process energy must be condensed and removed before leaving the system.
consumption per unit.
This also helps acquire a higher concentration of the target gas.
Figure 9 shows the relation between the heat duty of the
In the MEA-based separation process, nearly a half of condenser and the mass flow rate of solvent. This positive func-
the total exergy was expended on the regeneration of the MEA tion also shows a linear variation but with a small difference.
solvent and the cooling of the lean solvent and the regenerat- An increasing slope was spotted on the curve in the range of
ing gas. This involves three major devices: the lean solvent 6100 t/h to 6300 t/h. It can be explained as when the flow rate
cooler (COOLER2), the reboiler, and the condenser (both in of flue gas remains unchanged, increasing the solvent flow rate
STRIPPER). Hence the effect of the mass flow rate of MEA will certainly augment the CO2 absorption, which leads to a
solvent on the thermal load of these three devices was inves- nonlinear increase in the heat duty of the condenser because of
tigated carefully. Figure 8 presents the variation curves of the the rising amount of the regenerating gas, while continuously
15099 DOI: 10.1021/acs.iecr.7b03729
Ind. Eng. Chem. Res. 2017, 56, 15094−15101
Industrial & Engineering Chemistry Research Article

between the literature and the present work. Furthermore, the

energy efficiency of the removal process can also be enhanced by
implementing various modifications, such as solvent properties
variation, device structural adjustment, and operation optimiza-
tion.

6. CONCLUSIONS
High energy consumption is seen as the major weakness of
CO2 capture and separation technology. Much research was
conducted to improve the energy performance. In this work,
energy consumption and exergy loss of two different CO2 sep-
aration systems were estimated from the simulation results, and
then compared in detail. The main conclusions are as follows:
(1) Carbon dioxide can be efficiently separated from the flue
gas emitted from a thermal plant in both systems. At the
Figure 9. Effect of solvent flow rate on the heat duty of the condenser.
same flue gas condition, the first law efficiency of the
MEA-based CO2 capture system is 88.19%, higher than
increasing the MEA solvent to an excessive degree will not the value 74.15% of the hydrate-based system. The
augment the absorbed amount anymore because the CO2 in the second law efficiency is 38.32%, being slightly lower than
flue gas is limited. Thus, the heat duty of the condenser the value 38.85% of the latter.
transforms into a single value linear function of the solvent flow (2) In the process of MEA absorption, most energy is con-
rate. sumed by the solvent regeneration, during which the
In addition, it can also be found in Figure 8 and Figure 9 that energy conservation should be considered carefully. The
the thermal load in the cooler was one order higher than that in heat loads of the reboiler, the condenser, and the cooler
the condenser. This indicates that the thermal energy in the are significantly influenced by the solvent flow rate. The
regeneration process was mostly taken away by the lean solvent, separation rate of CO2 can obviously affect the energy
not the regenerating gas. Thus, the lean solvent cooler is also
performance as well.
considered as a key point of energy saving, where a consid-
(3) In the hydrate-based process, the refrigeration process
erable thermal energy might be recovered somehow.
causes the largest portion of the total consumption.
Most energy was carried into the MEA system with the high
temperature steam, providing the regenerating energy for CO2 Raising the COP of refrigerators helps achieve energy-
desorption. As presented in Figure 10, the regenerating energy saving, but this effect disappears with its continuous
increase. Besides, improving the compressor efficiency
can also dramatically enhance the energy efficiency of the
removal process.
(4) Searching for new absorbents and additives might be the
only way to reduce the regeneration energy consumption
in the MEA-based CO2 separation. Therefore, the hydrate-
based separation can be an alternative method with a
promising prospect, although it has a lot of requirements
of devices and limits for now.
In summary, with some improvements to the hydrate-based
CO2 separation process, this technique will show a distinct
advantage of energy savings. Future research focusing on the
energy efficiency analysis, technology optimization, as well as
structural design of devices is greatly encouraged before its
industrial application.

Figure 10. Effect of the CO2 removal rate on the regenerating energy. ■
*
ASSOCIATED CONTENT
S Supporting Information

in the MEA-based separation system obviously augments with The Supporting Information is available free of charge on the
the rising removal rate. At removal rate around 90%, the regen- ACS Publications website at DOI: 10.1021/acs.iecr.7b03729.
erating energy in our study was 5.3424 MJ/kg, being slightly Total energy and exergy input/output of both CO2
higher than 5.13 MJ/kg in previous research.31 This proves the separation systems (Table S1, Table S2, Table S3, and
reliability of the results in the present work. The suitable range Table S4) and simulation results of the streams in both
for industrial application was recommended by the authors as CO2 separation processes (Figure S1 and Figure S2)
3.2 MJ/kg to 4.2 MJ/kg. Hence, a lot of efforts were made by (PDF)

■
them in order to optimize the operation, and preferable results
were obtained. The average value of the regenerating energy
in the literature32 was 4.12 MJ/kg, within the reasonable range AUTHOR INFORMATION
mentioned above. This difference might be due to the dis- Corresponding Author
crepancy of the amount and the composition of flue gas *Tel: +86-731-88879863. E-mail: liuzq@csu.edu.cn.
15100 DOI: 10.1021/acs.iecr.7b03729
Ind. Eng. Chem. Res. 2017, 56, 15094−15101
Industrial & Engineering Chemistry Research Article

ORCID (16) Xia, Z. M.; Chen, Z. Y.; Li, X. S.; et al. Thermodynamic
Nan Xie: 0000-0002-9502-4501 equilibrium conditions for simulated landfill gas hydrate formation in
aqueous solutions of additives. J. Chem. Eng. Data 2012, 57 (11),
Zhiqiang Liu: 0000-0002-6504-0543 3290−3295.
Notes (17) Lin, W.; Delahaye, A.; Fournaison, L. Phase equilibrium and
The authors declare no competing financial interest. dissociation enthalpy for semi-clathrate hydrate of CO2+TBAB. Fluid

■
Phase Equilib. 2008, 264, 220−227.
ACKNOWLEDGMENTS (18) Sabil, K. M.; Witkamp, G. J.; Peters, C. J. Estimations of
enthalpies of dissociation of simple and mixed carbon dioxide hydrates
This research was financially supported by the National Natu- from phase equilibrium data. Fluid Phase Equilib. 2010, 290, 109−114.
ral Science Foundation of China (Grant No. 51376198 and (19) Pivezhani, F.; Roosta, H.; Dashti, A.; Mazloumi, S. H.
51676209), and Fundamental Research Funds for the Central Investigation of CO2 hydrate formation conditions for determining
Universities of Central South University (No. 2017zzts129). the optimum CO2 storage rate and energy: Modeling and experimental

■
study. Energy 2016, 113, 215−226.
REFERENCES (20) Li, X. S.; Xu, C. G.; Chen, Z. Y.; Wu, H. J. Hydrate-based pre-
combustion carbon dioxide capture process in the system with tetra-n-
(1) Song, C. F.; Liu, Q. L.; Ji, N.; Deng, S.; Zhao, J.; Kitamura, Y. butyl ammonium bromide solution in the presence of cyclopentane.
Advanced cryogenic CO2 capture process based on Stirling coolers by Energy 2011, 36 (3), 1394−1403.
heat integration. Appl. Therm. Eng. 2017, 114, 887−895. (21) Park, S.; Lee, S.; Lee, Y.; Seo, Y. CO2 capture from simulated
(2) Li, B. H.; Zhang, N.; Smith, R. Simulation and analysis of CO2 fuel gas mixtures using semiclathrate hydrates formed by quaternary
capture process with aqueous monoethanolamine solution. Appl. ammonium salts. Environ. Sci. Technol. 2013, 47 (13), 7571−7577.
Energy 2016, 161, 707−717. (22) Gholinezhad, J.; Chapoy, A.; Tohidi, B. Separation and capture
(3) Liu, L. X.; Zhao, J.; Deng, S.; An, Q. S. A technical and economic of carbon dioxide from CO2/H2 syngas mixture using semi-clathrate
study on solar-assisted ammonia-based post-combustion CO2 capture hydrates. Chem. Eng. Res. Des. 2011, 89, 1747−1751.
of power plant. Appl. Therm. Eng. 2016, 102, 412−422. (23) Babu, P.; Yao, M.; Datta, S.; Kumar, R.; Linga, P.
(4) Li, K. K.; Leigh, W.; Feron, P.; Yu, H.; Tade, M. Systematic study Thermodynamic and kinetic verification of tetra-n-butyl ammonium
of aqueous monoethanolamine (MEA)-based CO2 capture process: nitrate (TBANO3) as a promoter for the clathrate process applicable
Techno-economic assessment of the MEA process and its improve- to precombustion carbon dioxide capture. Environ. Sci. Technol. 2014,
ments. Appl. Energy 2016, 165, 648−659. 48 (6), 3550−3558.
(5) Sun, Q. B.; Kang, Y. T. Review on CO2 hydrate formation/ (24) Ma, Z. W.; Zhang, P.; Bao, H. S.; Deng, S. Review of
dissociation and its cold energy application. Renewable Sustainable fundamental properties of CO2 hydrates and CO2 capture and
Energy Rev. 2016, 62, 478−494. separation using hydration method. Renewable Sustainable Energy Rev.
(6) Ye, N.; Zhang, P. Equilibrium data and morphology of tetra-n- 2016, 53, 1273−1302.
butyl ammonium bromide semiclathrate hydrate with carbon dioxide. (25) Xu, C. G.; Li, X. S. Research progress of hydrate-based CO2
J. Chem. Eng. Data 2012, 57 (5), 1557−1562. separation and capture from gas mixtures. RSC Adv. 2014, 4, 18301−
(7) Lee, Y. H.; Lee, S. M.; Lee, J. Y.; Seo, Y. W. Structure 18316.
identification and dissociation enthalpy measurements of the CO2+N2 (26) Tajima, H.; Yamasaki, A.; Kiyono, F. Energy consumption
hydrates for their application to CO2 capture and storage. Chem. Eng. J. estimation for greenhouse gas separation processes by clathrate
2014, 246, 20−26. hydrate formation. Energy 2004, 29, 1713−1729.
(8) Lee, Y. H.; Lee, D. Y.; Lee, J. W.; Seo, Y. W. Enclathration of CO2 (27) Park, S. H.; Chung, S. W.; Lee, S. K.; Choi, H. K.; Lee, S. H.
as a co-guest of structure H hydrates and its implications for CO2 Thermo-economic evaluation of 300 MW class integrated gasification
capture and sequestration. Appl. Energy 2016, 163, 51−59. combined cycle with ash free coal (AFC) process. Appl. Therm. Eng.
(9) Kastanidis, P.; Romanos, G. E.; Michalis, V. K.; Economou, I. G.; 2015, 89, 843−852.
Stubos, A. K.; Tsimpanogiannis, I. N. Development of a novel (28) Canepa, R.; Wang, M. H. Techno-economic analysis of a CO2
experimental apparatus for hydrate equilibrium measurements. Fluid capture plant integrated with a commercial scale combined cycle gas
Phase Equilib. 2016, 424, 152−161. turbine (CCGT) power plant. Appl. Therm. Eng. 2015, 74, 10−19.
(10) Kumar, A.; Bhattacharjee, G.; Barmecha, V.; Diwan, S.; (29) Escudero, A. I.; Espatolero, S.; Romeo, L. M.; Lara, Y.; Paufique,
Kushwaha, O. S. Influence of kinetic and thermodynamic promoters C.; Lesort, A. L.; Liszka, M. Minimization of CO2 capture energy
on post-combustion carbon dioxide capture through gas hydrate penalty in second generation oxy-fuel power plants. Appl. Therm. Eng.
crystallization. J. Environ. Chem. Eng. 2016, 4 (2), 1955−1961. 2016, 103, 274−281.
(11) Li, S. F.; Fan, S. S.; Wang, J. Q.; Lang, X. M.; Wang, Y. H. (30) Duan, L. Q.; Zhao, M. D.; Yang, Y. P. Integration and
Semiclathrate hydrate phase equilibria for CO2 in the presence of optimization study on the coal-fired power plant with CO2 capture
tetra-n-butyl ammonium halide (bromide, chloride, or fluoride). J. using MEA. Energy 2012, 45, 107−116.
Chem. Eng. Data 2010, 55 (9), 3212−3215. (31) Rezazadeh, F.; Gale, W. F.; Akram, M.; Hughes, K. J.;
(12) Li, X. S.; Xu, C. G.; Chen, Z. Y.; Cai, J. Synergic effect of Pourkashanian, M. Performance evaluation and optimization of post
cyclopentane and tetra-n-butyl ammonium bromide on hydrate-based combustion CO2 capture processes for natural gas applications at pilot
carbon dioxide separation from fuel gas mixture by measurements of scale via a verified rate-based model. Int. J. Greenhouse Gas Control
gas uptake and X-ray diffraction patterns. Int. J. Hydrogen Energy 2012, 2016, 53, 243−253.
37, 720−727. (32) Oh, S. Y.; Binns, M.; Cho, H.; Kim, J. K. Energy minimization of
(13) Li, X. S.; Xu, C. G.; Chen, Z. Y.; Wu, H. J. Tetra-n-butyl MEA-based CO2 capture process. Appl. Energy 2016, 169, 353−362.
ammonium bromide semi-clathrate hydrate process for post-
combustion capture of carbon dioxide in the presence of dodecyl
trimethyl ammonium chloride. Energy 2010, 35, 3902−3908.
(14) Xu, C. G.; Yu, Y. S.; Ding, Y. L.; Cai, J.; Li, X. S. The effect of
hydrate promoters on gas uptake. Phys. Chem. Chem. Phys. 2017, 19,
21769−21776.
(15) Yang, M. J.; Jing, W.; Zhao, J. F.; Ling, Z.; Song, Y. C.
Promotion of hydrate-based CO2 capture from flue gas by additive
mixtures (THF (tetrahydrofuran)+TBAB (tetra-n-butyl ammonium
bromide)). Energy 2016, 106, 546−553.

15101 DOI: 10.1021/acs.iecr.7b03729

Ind. Eng. Chem. Res. 2017, 56, 15094−15101