Applied Energy 185 (2017) 362–375

Contents lists available at ScienceDirect

Applied Energy
journal homepage: www.elsevier.com/locate/apenergy

Enhancing the energetic efficiency of MDEA/PZ-based CO2 capture

technology for a 650 MW power plant: Process improvement
Bin Zhao a, Fangzheng Liu a, Zheng Cui a, Changjun Liu a,⇑, Hairong Yue a, Siyang Tang a, Yingying Liu b,
Houfang Lu a,b, Bin Liang a,b
a
Multi-phases Mass Transfer and Reaction Engineering Laboratory, College of Chemical Engineering, Sichuan University, Chengdu 610065, China
b
Center of CCUS and CO2 Mineralization and Utilization, Sichuan University, Chengdu 610065, China

h i g h l i g h t s

MDEA/PZ solutions are more energetically efficient for post-combustion CO2 capture.

Various process improvements effectively enhance the process energy efficiency.

A reboiler heat duty (Qreb) of 2.24 G J/t CO2 was achieved.

The net power efficiency penalty was reduced from 9.13% to 7.66%.

a r t i c l e i n f o a b s t r a c t

Article history: Post-combustion CO2 capture (PCC) facilities are set up at the power plants to reduce their carbon foot-
Received 2 September 2016 print. However, the high energy demand of the amine-based PCC technologies significantly decreases the
Received in revised form 31 October 2016 net energy efficiency of a power plant. This work focused on developing an optimal amine-based PCC
Accepted 3 November 2016
technology by carefully analyzing and integrating the energy flow. The methyldiethanolamine (MDEA)
Available online 11 November 2016
solution of lower reaction heat with CO2 is found to be a superior solvent than monoethanolamine
(MEA). Process analysis using the equilibrium stage model in Aspen Plus was performed to investigate
Keywords:
the effects of piperazine (PZ) promoter and various operation parameters such as absorption pressure,
CO2 capture
MDEA/PZ solutions
stripping pressure, CO2 loading in lean solution, and CO2 removal ratio on the net efficiency penalty of
Process improvements MDEA/PZ-based PCC process. Several enhancement measures, including the absorber intercooling, simple
Net power efficiency penalty rich-split, advanced rich-split, and stripper interheating, were found significantly improving the process
energetic efficiency. A reboiler heat duty (Qreb) as low as 2.24 G J/t CO2 was achieved which is 27.7% lower
than that of its counterpart (MEA). The enhanced MDEA/PZ-based technology leads to a low net power
efficiency penalty of 7.66% which is 16.1% less than that of the benchmark MEA-based technology.
This improvement corresponds to an absolute increase of 1.47% in net efficiency of a 650 MW power
plant with PCC units.
Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction CO2 emission sites such as power plants are setting up CO2 capture
facilities. However, the power generation cost was increased by
Anthropogenic CO2 emissions are believed to be the primary 40–85% after installation of a CO2 capture unit [8,9]. Thus develop-
contributing factor toward the increasing global warming [1,2], ing low-cost and energy-saving CO2 capture technologies is very
and they are mainly attributed to the combustion of fossil fuels attractive [10,11].
[3]. In the near future, fossil energy will continue to dominate Alkanolamine solution is the most commonly used solvent for
our energy supply because the application of renewable energies PCC process and is widely applied in industry [10,12,13]. Unlike
is still far from matching the total energy consumption [4]. Carbon CO2 separation in natural gas production and ammonia production,
capture and storage (CCS) is a post treatment strategy to the partial pressure of CO2 (13 kPa) from a power plant is signifi-
reduce CO2 emissions [5–7]. Thus, increasingly more stationary cantly lower [14,15]; monoethanolamine (MEA) is usually adopted
as the absorbent because of its high reactivity with CO2 [13,16,17].
⇑ Corresponding author. Fig. 1 shows the pattern of energy consumption for an MEA-
E-mail address: liuchangjun@scu.edu.cn (C. Liu). based PCC process with 4.2 Mt CO2/a capture capacity, in which

http://dx.doi.org/10.1016/j.apenergy.2016.11.009
0306-2619/Ó 2016 Elsevier Ltd. All rights reserved.
 B. Zhao et al. / Applied Energy 185 (2017) 362–375 363

Nomenclature

nCO2 amount of CO2 desorbed from the rich solution, kmol/h nH2 O; cond amount of condensed water refluxed from the con-
L circulation rate of the absorbent, kg/h denser (E4), kg/h
cp heat capacity of the rich solution, kJ/(kgK) alean CO2 loading in the lean solution, mol CO2/mol total
Trich temperature of the rich stream (S1) at the stripper in- amine (MEA or MDEA/PZ blends)
let, °C arich CO2 loading in the rich solution, mol CO2/mol total
Tlean temperature of the lean stream (S2) at the absorber in- amine (MEA or MDEA/PZ blends)
let, °C Da CO2 loading difference between the rich and lean solu-
Treb temperature of the reboiler (E2), °C tions, mol CO2/mol total amine (MEA or MDEA/PZ
TE4 temperature of the stripper condenser (E4), °C blends)
TE1 logarithmic mean temperature difference of lean/rich DPloss equivalent electric power loss of power plant, kW
heat exchanger (E1), °C M CO2 total CO2 capture rate for a 650 MW power plant, t
Pstrip stripping pressure, MPa CO2/h
Pabs absorption pressure, MPa hCarnot carnot efficiency of the water/steam cycle, %
Qreb reboiler (E2) heat duty, GJ/t CO2 heff effective efficiency of power loss in the water/steam
DHabs; CO2 reaction enthalpy of CO2 with amine (MEA or MDEA/ cycle due to steam extraction, %
PZ blends), kJ/kmol Tcond temperature of the main condenser in the water/steam
DHvap; H2 O vaporization heat of H2O, kJ/kg cycle of power plant, °C
cp; H2 O heat capacity of H2O, kJ/(kgK) Tsat, steam temperature of the extracted steam at the extraction
nH2 O; vap Amount of the hot steam entering the condenser (E4), pressure, °C
kg/h

Blower work Reaction Heat Reaction Heat

8.1% 2.12 GJ/t CO2 2.12 GJ/t CO2
59% 68%

parameter flowsheet
optimization modifications
Sensible Heat Sensible Heat
Reboiler duty (57.5%)
3.6 GJ/t CO2 0.68 GJ/t CO2 3.11 GJ/t CO2 0.62 GJ/t CO2
4.0 GJ/t CO2
19% 20%
Total energy
consumption

Useless Heat Useless Heat

Compression work
0.8 GJ/t CO2 0.37 GJ/t CO2
29.7%
22% 12%

Pump work
4.7%

Fig. 1. Distribution of total energy consumption and three components of Qreb in an MEA-based PCC process [13,18].

Qreb accounts for 57.5% of the total energy consumption [10,13,15] gations have been done on the CO2-MDEA-H2O system such as
and the operation cost accounts for 79.0% of the life cycle invest- the chemical [26–29] and physical absorption data [30–34] as well
ment [8,19]. The minimum Qreb of the MEA-based process is as as the equilibrium model [35]. However, MDEA itself shows low
high as 3.1 GJ/t CO2, which leads to that the net power efficiency CO2 absorbing rate which limits its application in CO2 separation
decreases from 38.9% to 29.8% [13,18,20]. The high energy con- processes [36]. Piperazine (PZ) was found to be an absorption rate
sumption for sorbent regeneration is mainly ascribed to the high promoter for MDEA [37]. The MDEA/PZ blends also have a low reac-
reaction enthalpy of MEA and CO2. The decomposition heat of car- tion heat of 70 kJ/mol CO2 [14,25], corresponding to only 1.6 GJ/
bamate is 2.12 GJ/t CO2 which accounts for about 68% of Qreb [18]. t CO2. And the resultant aqueous MDEA/PZ solution shows both a
The absorption reaction heat is 95 kJ/mol CO2 for MEA [21,22], and high CO2 absorption capacity and adequate CO2 absorption rate
it cannot be changed by process improvements. Hence, further [23]. The equilibrium constants of the CO2-PZ system was reported
reduction of Qreb for MEA-based process becomes increasingly dif- [38–40]. The CO2 loading of various aqueous MDEA/PZ solutions
ficult and alkanolamine of low reaction heat with CO2 would be was found increasing with the increase of PZ content [41–43]. Bish-
expected. noi et al. developed an absorption model for the CO2-MDEA-PZ-H2O
Methyldiethanolamine (MDEA) is one of such absorbents and is system which is capable of predicting the CO2 absorption rate,
promising for a low-energy-consuming PCC process. The decompo- enhancement factor, the height and volume of the transfer unit well
sition enthalpy of the resultant bicarbonate of MDEA and CO2 [44]. However these works were limited to the low PZ content.
(61 kJ/mol CO2) is much smaller than that of the carbamate formed Generally an MDEA/PZ solution of 45 wt.% MDEA and 5 wt.% PZ
by MEA and CO2 (95 kJ/mol CO2) [14,21,23–25]. Extensive investi- was used to remove the CO2 from the feed stream in natural gas
364 B. Zhao et al. / Applied Energy 185 (2017) 362–375

plants and ammonia plants [23]. Mudhasakul et al. also developed to the bottom of the absorber and contact with the downward sor-
a rate-based CO2 absorption model for this system [24]. However, bent stream, namely the lean solution, in a counter-current man-
the CO2 absorption rate of the aqueous MDEA/PZ solution with ner. The rich solution leaves the column from the bottom and
5 wt.% PZ is not good enough to efficiently remove CO2 from the enters the stripper after exchanging heat with the lean stream
flue gas of a coal-fired power plant. Svendsen et al. found that coming from the stripper. The stripper operates at a higher temper-
increasing the PZ content can improve the CO2 capture perfor- ature than that of the absorber and at a slightly higher pressure
mance of MDEA/PZ-based solvent system in CO2 removal from than the atmospheric pressure.
the blast furnace gas (100 kPa CO2) [15]. Oexmann developed a The flow rate of flue gases was assumed to be 3100 t/h, accord-
simplified semi-empirical model for MDEA/PZ solutions (8 wt.% ing to that of a 650 MW power plant (38.9% net efficiency) [57].
PZ/42 wt.% MDEA) [45] and Frailie developed a rate-based model Generally, both SO2 and NOx were removed from flue gases by flue
for MDEA/PZ solutions (8 wt.% PZ/42 wt.% MDEA and 21 wt.% gas desulfurization (FGD) and selective catalytic reduction (SCR)
PZ/29 wt.% MDEA) [14]. But the effect of PZ concentration in sol- units before entering the CO2 capture unit. Flue gases entering
vent solutions on process energetic efficiency was not investigated. the absorber were assumed to be at 0.11 MPa and 40 °C [54]. The
Process improvements such as, absorber intercooling [46–49], sim- flue gas was assumed to be composed of 12% (V/V) CO2, 78% (V/
ple rich-split [50,51], advanced rich-split [51], stripper interheat- V) N2, and 10% (V/V) H2O [56].
ing [52], has proven to be effective ways to improve the energy Thermodynamic properties of the CO2-MDEA-PZ-H2O system
efficiency of a PCC process in process simulations as well as in were determined with an Electrolyte-NRTL model in the amines
the pilot plant trials [13,18,20,53–55]. Energy efficiency is the property package of Aspen Plus. Reactions 1–5 were assumed to
major concern for a PCC process. Understanding the effect of PZ be in equilibrium. The equilibrium constants (Keq) for reactions
concentration on the process energy efficiency and optimizing (1–5) were calculated by Eq. (1) from Aspen Plus. Table 1 summa-
the process energy efficiency is of realistic interest. rized the chemical reactions and relevant calculation constants for
This work dealt with a wide composition range of MDEA/PZ- Eq. (1). Reactions 6–13 were considered kinetically controlled
based solvent. It focused on understanding the effect of PZ content because of their finite reaction rates. The reaction rate constants
on the overall energy consumption of an MDEA/PZ-based process were calculated by Eq. (2). The activation energy and the pre-
and on developing an energetically optimal amine-based PCC pro- exponential factors for Eq. (2) were summarized in Table 2.
cess by carefully analyzing and integrating the energy flow. The
energy efficiency of an MDEA/PZ-based PCC process appended to ln Keq ¼ A þ B=T þ C lnðTÞ þ DT; T in ðKÞ ð1Þ
a 650 MW power plant was evaluated by process simulation using
the equilibrium stage model in Aspen Plus. The effects of the pro-
E=RT
cess parameters such as absorption pressure, stripping pressure, Kinetic factor ¼ ke ; T in ðKÞ ð2Þ
CO2 loading in lean solution, and CO2 removal ratio on process
energy efficiency were carefully studied. The process improve-
ments including the absorber intercooling, simple rich-split, 2.2. Overall energy evaluation
advanced rich-split, stripper interheating, and their combinations
were appreciated in terms of the net efficiency of the power plant. Overall energy evaluation was conducted to assess this
advanced MDEA/PZ-based PCC system integrated with a power
2. Methodology plant.

2.1. Primary process flow chart and thermodynamics

2.2.1. Energy consumption in the reboiler
A general regenerative absorption-desorption PCC process was The energy consumed by the reboiler is consist of three compo-
adopted for process simulation (Fig. 2) [12,56]. Flue gases are fed nents (Eq. (3)) [58]:

650 MW Pretreatment CO2 capture-90% Compression

Power plant 15 MPa
Vent gas
Cooler
Wash

E4

Pump 2
Cooler
Cooler

S2
S1
Cooler E3 Pump 5
Flue gas
Cooler
Absorber Lean/rich
Direct Stripper
heat exchanger
contact
cooler
E1
Pump 1 CO2
Blower
E2 product

Pump 3 Pump 4 Reboiler

Fig. 2. Flow diagram of an amine-based PCC process [56].

 B. Zhao et al. / Applied Energy 185 (2017) 362–375 365

Table 1
Reactions and related parameters for the electrolyte property package.

No. Reaction (Equilibrium) A B C D

þ  13,446 22.48
1 2H2 O $ H3 O þ OH 132.9 0

2 HCO 2 þ 216 12,432 35.48 0

3 þ H2 O $ CO3 þ H3 O
þ þ 4.0762 7773.2
3 PZH þ H2 O $ PZ þ H3 O 0 0

4 H2 O þ Hþ PZCOO $ H3 Oþ þ PZCOO 14.042 3443.1 0 0

5 MDEAHþ þ H2 O $ MDEA þ H3 Oþ 9.4165 4234.98 0 0

Table 2
Reactions and kinetic parameters for the electrolyte property package.

No. Reaction (Kinetic) Pre-exponential factor (kmolm3s1) Activation energy (J/mol)

6 CO2 þ OH ! HCO
3 4.32  1013 55407.318
7 HCO
3 ! CO2 þ OH

2.38  1017 123164.082
 þ 4.14  10 10
8 PZ þ CO2 þ H2 O ! PZCOO þ H3 O 33616.1706

9 PZCOO þ H3 Oþ ! PZ þ CO2 þ H2 O 7.94  1021 65899.956

10 PZCOO þ CO2 þ H2 O ! PZðCOOÞ2 þ 3.62  1010 33616.1706

2 þ H3 O
25
11 PZðCOOÞ2 þ 
þ H3 O ! PZCOO þ CO2 þ H2 O 5.56  10 76831.704
2

12 MDEA þ CO2 þ H2 O ! MDEAHþ þ HCO

3
2.22  107 37759.278
þ
13 MDEAH þ HCO
3 ! MDEA þ CO2 þ H2 O 1.06  1016 106323.168

(1) Reaction heat: The heat required for CO2 desorption. For 2.2.2. CO2 compression
MDEA, the reaction heat is 61 kJ/mol CO2, or 1.40 GJ/t CO2 The compression process was modeled in six stages under fixed
[25]. discharge conditions (15 MPa, 40 °C). We used three intercoolers at
(2) Sensible heat: The energy required to heat rich stream in stages 1, 3, and 5 in the CO2 compressor to remove moisture from
order to reach the decomposition temperature in the strip- the pressurized CO2 product [59].
per. This heat is provided to the stripper to increase the tem-
perature of the rich solution from Trich to Treb. Obviously, it is 2.2.3. Auxiliary equipment
proportional to the circulation rate of the absorbent (L). Auxiliary equipment such as heat exchangers, blowers, and
(3) Useless heat: The first term on the right-hand side corre- pumps were also integrated in the entire PCC process. The isen-
sponds to the evaporation heat of the steam which is tropic efficiency and mechanical efficiency of both the pumps
escaped out of the stripper and reached the reflux condenser and blowers were set at 80% and 95%, respectively [20]. The solvent
(E4). The second term is the energy required to heat up the pumping head was set at 0.21 MPa [62]. The cooling duty was con-
water refluxed back to the stripper from the condenser (E4) verted into electricity using a coefficient of performance of 5 [20].
with a temperature increase of Trich  TE4. This part of heat The electrical duties of the CO2 compressor, pumps, and blowers
won’t be able to get recovered. were directly derived from the Aspen simulation results.

Q reb ¼ nCO2 DHabs;CO2 Reaction Heat 3. Results and discussion

þLcp ðT reb  T rich Þ Sensible Heat
þnH2 O;vap DHvap;H2 O þ nH2 O;cond cp;H2 O ðT rich  T E4 Þ Useless Heat 3.1. Baseline case study
ð3Þ
A general regenerative absorption-desorption flow chart was
Low-pressure steam, which was assumed to be supplied by the adopted as the primary design for the MDEA/PZ-based process.
water/steam cycle of the power station, was used to provide The typical operation parameters used to define the baseline case
energy to the reboiler. This led to a subsequent electric power loss are listed in Table 3. The minimum Qreb of 3.70 GJ/t CO2 at 90%
of the power plant, which can be calculated by assuming that the CO2 removal is achieved with an alean of 0.04 for the baseline case.
extracted steam would be used in a Carnot cycle [59,60]. As our study deals with 540 t CO2/h (4 Mt CO2/a) from flue
The equivalent electric power loss (DPloss) could be described by gases, four parallel process trains are proposed, and each process
the following equation [61]: train has a CO2 removal capacity of 1 Mt CO2/a. Each process train
has an absorber column of £18 m  H6 m and a stripper column
DPloss ¼ heff  hCarnot  M CO2  Q reb ð4Þ
Parameter heff was calculated by:
Table 3
heff ¼ 0:6102 þ 0:00165  ðT sat; steam  273:15Þ ð5Þ Operation parameters of the MDEA/PZ-based PCC process.

CO2 removal ratio Pabs (MPa) Pstrip TE1 [13] (°C)

Parameter hCarnot was defined as:
[14] (%) (MPa)
T cond 90 0.11 0.13 10
hCarnot ¼ 1  ð6Þ
T sat; steam PZ concentration MDEA concentration Tlean[19] alean (mol CO2/mol
[24] (wt.%) [24] (wt.%) (°C) total amine)
Tcond was assumed to be 40 °C in this study [20], and Tsat,steam
5 45 40 0.04
was assumed to be 10 °C higher than Treb.
366 B. Zhao et al. / Applied Energy 185 (2017) 362–375

Table 4
Energy consumption of the MDEA/PZ-based PCC process integrated with an APC power plant.

Energy type Power penalty/train (kW) Power output (kW) Design specification
Power plant
Power output Electricity – +650,000 –
Net efficiency – – 38.9% –
Pretreatment unit
Blower Electricity 791 3164 3.05 kPa increase
Pump 1 Electricity 18.5 74 4  13 m3/min
CO2 capture unit
Stripper reboiler Steam 21,446 85,785 110.4 °C, 0.13 MPa
Pump 2 Electricity 5 20 4  0.8 m3/min
Pump 3 Electricity 229 916 4  47 m3/min
Pump 4 Electricity 228 912 4  48 m3/min
Pump 5 Electricity 0.2 0.8 4  0.09 m3/min
CO2 compression unit
Compressor Electricity 15014.6 60,058 15 MPa, 40 °C
Auxiliary
Pumps for cooling water Electricity 800 3200 4  50 m3/min
Others Electricity 1700 6800 –
Total power penalty Electricity 160,930
Overall power output +489,070
Overall net efficiency penalty 9.63%
Net power efficiency with PCC plant 29.27%

of £1 m  H8 m. Table 4 shows the energy penalty of the MDEA/

PZ-based PCC process integrated with an advanced pulverized coal
(APC) power plant. The MDEA/PZ-based process is energy-
intensive, with an energy penalty of 161 MW electricity. As a
result, the power output drops from 650 to 489 MW and the net
efficiency falls from 38.9% to 29.27%, leading to 24.76% (relative
term) and 9.63% (absolute term) decreases, respectively.

3.2. Parameter sensitivity study

3.2.1. Effect of the absorption pressure

It is known that increasing pressure would benefit the absorp-
tion process. Mudhasakul et al. [24] found that more than 99%
CO2 could be removed at a Pabs of 4.16 MPa with an aqueous
MDEA/PZ solutions of 45 wt.% MDEA and 5 wt.% PZ. Fig. 3 shows
the effect of Pabs on Qreb, the compression work, and the net power
efficiency penalty. Increasing the pressure leads to a dramatic
increase in the compression work despite a slight decrease in Qreb, Fig. 3. Effect of absorption pressure (Pabs) on energy consumption and the net
even though slightly compressing the flue gas from 0.11 to power efficiency penalty.
0.20 MPa would greatly increase the net power efficiency penalty.
Hence, it can be suggested that the MDEA/PZ-based process under increase the pump work (pump 3) and decrease the CO2 compres-
high pressure is a highly energy-intensive operation for PCC pur- sion work. Treb rises with increasing Pstrip, which is likely to
pose. The Pabs should depend on the pressure of flue gases and decrease the net power efficiency since higher-quality steam
commercial absorption processes are normally operated in a Pabs would be extracted from the water/steam cycle of the power plant.
approximately equal to the atmospheric pressure. Therefore, a Pabs Fig. 5 illustrates the effects of Pstrip on Treb and the net power effi-
of 0.11 MPa was used in this work for simulation. ciency penalty. The minimum net efficiency penalties for PCC pro-
cesses with 5 and 10 wt.% PZ are achieved at Pstrip values of 0.15
and 0.19 MPa, respectively. Thereafter, Pstrip of PCC processes with
3.2.2. Effect of stripping pressure
5 and 10 wt.% PZ are set to 0.15 and 0.19 MPa, respectively. For PCC
Pstrip of 0.13 MPa is used in the baseline case. Fig. 4 illustrates
process with 15 and 20 wt.% PZ, the net efficiency penalty
the effect of Pstrip on Qreb and its three components of heat duties.
decreased on increasing Pstrip from 0.13 to 0.23 MPa. The boiling
As shown in Fig. 4, Qreb decreases with the increasing Pstrip. Gener-
temperatures of MDEA/PZ solutions are about 125 and 128 °C at
ally, lower Pstrip promotes CO2 desorption. However, higher Pstrip
0.21 and 0.23 MPa, respectively. The thermal stability of amine
can decrease the water content in the CO2 stream that left the
suggests that Treb should not exceed 125 °C [16]. Therefore, the
stripper, thereby decreasing the useless heat [61]. For a PCC pro-
Pstrip of PCC processes with 15 and 20 wt.% PZ are fixed to 0.21 MPa.
cess with 5 wt.% PZ, Qreb reduces from 3.7 to 3.4 GJ/t CO2 when
Pstrip is increased from 0.13 to 0.15 MPa; however, it does not
change when Pstrip is further increased to 0.23 MPa. Moreover, 3.2.3. Effect of CO2 loading of lean stream
increasing the PZ content from 5 to 20 wt.% also leads to lower Qreb. The CO2 loading of lean stream (alean ) directly influences the
The Qreb, the pump work, the subsequent compression work of decomposition temperature, refluxing rate and then Qreb. Fig. 6
CO2 product and the energy for the auxiliary units account for the shows that Qreb first decreases sharply with the increasing alean
total energy consumption of a PCC process. Higher Pstrip would due to lowering the refluxing rate and significant decrease of the
 B. Zhao et al. / Applied Energy 185 (2017) 362–375 367

Fig. 4. Effect of stripping pressure (Pstrip) on heat duties. (a) 5 wt.% PZ; (b) 10 wt.% PZ; (c) 15 wt.% PZ; (d) 20 wt.% PZ.

useless heat [19,58,63]. The circulation rate of the solvent (L) rises take the increased marginal capital cost into account. So, the
with the increasing alean because the CO2 loadings in rich solution MDEA/PZ-based process using 20 wt.% PZ and 30 wt.% MDEA as
(arich ) almost keep constant at different alean . Then, Qreb increased the absorbent with CO2 removal ratio of 90% is recommended.
slightly with the increasing alean because of the larger circulation
rate of the solvent (L) and the higher sensible heat [19]. Fig. 6
shows that Qreb decreases with the PZ content increases from 5
3.2.5. Effect of the MDEA/PZ ratio
to 20 wt.% with a total amine concentration of 50 wt.% for the
The simulation demonstrates that the values of the minimum
MDEA/PZ-based process, a minimum Qreb exists for each specific
Qreb as well as their corresponding alean depend on the PZ contents
MDEA/PZ composition.
of the solvent. Fig. 8 summarizes the effects of PZ contents on the
minimum Qreb, corresponding alean , and net power efficiency pen-
3.2.4. Effect of CO2 removal ratio alty. The minimum Qreb decreases from 3.40 to 2.74 GJ/t CO2 when
The CO2 removal ratio influences the absorption efficiency and the PZ content is increased from 5 to 20 wt.% (Fig. 8b); alean repre-
affects the process energy consumption. Fig. 7 shows Qreb of the sents the regeneration degree of the solvent. The optimal alean cor-
MDEA/PZ-based processes with a total amine concentration of responds to the lowest Qreb. CO2 preferentially reacts with PZ to
50 wt.% and the PZ contents from 5 to 20 wt.% at different CO2 form a carbamate instead of reacting with MDEA to form a bicar-
removal ratios. Qreb is sensitive to the CO2 removal ratio at low bonate in aqueous MDEA/PZ solutions due to the higher stability
PZ contents. At a PZ concentration of 5 wt.% and an alean of 0.04, of the carbamate. Therefore, higher PZ contents lead to a higher
Qreb reduces by 3.21% when the CO2 removal ratio is decreased optimal alean , which shifts from 0.04 to 0.10 when the PZ content
from 99% to 90% and reduces by 9.43% when the removal ratio is is increased from 5 to 20 wt.%. This is consistent with the litera-
further reduced from 90% to 60%. The effect of the CO2 removal tures [14,41].
ratio on Qreb is negligible at high PZ contents. With a PZ content PZ is an effective promoter for aqueous MDEA solutions [23]. It
of 20 wt.% and an alean of 0.10, Qreb reduces by only 1.6% when quickly captures CO2 and then transfers it to MDEA molecules;
the CO2 removal ratio is decreased from 99% to 90% and reduces therefore, it accelerates the absorption rate and increases the CO2
by 3.43% when the ratio is further decreased from 90% to 60%. capacity of rich stream. As shown in Fig. 8a, adding PZ results in
However, increasing the CO2 removal ratio from 90% to 99% may an increase of arich and a reduction of the solvent circulation rate
increase the cost of CO2 capture ($/t CO2) by 1%; thus, removing (L). However, the addition of PZ also leads to higher reaction
more than 90% CO2 from the flue gases is not recommended in enthalpy (87 kJ/mol CO2) than that of a simple MDEA solution
industry [14]. Although MDEA/PZ-based processes using the absor- (61 kJ/mol CO2) [25]. Balancing the two factors, when 20 wt.% PZ
bent with a low PZ content and removing less than 90% CO2 could is added, Qreb reduces by about 18.1% and the net power efficiency
reduce the energy consumption, their economic feasibility should penalty decreases from 9.34% to 8.49%.
368 B. Zhao et al. / Applied Energy 185 (2017) 362–375

Fig. 5. Effect of stripping pressure (Pstrip) on Treb and the net power efficiency penalty. (a) 5 wt.% PZ; (b) 10 wt.% PZ; (c) 15 wt.% PZ; (d) 20 wt.% PZ.

Sensitivity analysis indicates that major energy savings can be the absorption equilibrium and lowers the CO2 capacity of the sor-
achieved by optimizing the lean loading (alean ) and the MDEA/PZ bent. Thus, minimizing the temperature rise of the absorber by
ratio of solvent, while the CO2 removal ratio exerts only a minor timely removing the absorption heat is desirable. Here, an inter-
effect. The optimal solvent regeneration pressure (Pstrip) highly cooler is proposed in the middle of the absorber to reduce the tem-
depends on the composition of the absorbent; increasing the pres- perature of the absorbent and to increase the CO2 loading of rich
sure of the massive flue gas is infeasible for the PCC purpose. Our stream (Fig. 9a). It could directly reduce the circulation rate of
work suggests that the minimum Qreb of about 2.74 GJ/t CO2 can the absorbent (L) and decrease the sensible heat duty in the reboi-
be achieved at a PZ content of 20 wt.% and an alean of 0.10. PZ concen- ler [13,14,18].
trations of higher than 20 wt.% are not recommended because of the The simulation results obtained with the intercooling process
resulting increase in the solvent viscosity. Accounting for the effect are listed in Table 5. The absorber includes 10 equilibrium stages
of viscosity in terms of viscosity normalized capacity, the maximum which are numbered from top to bottom. The installation position
CO2 capture capacity occurs when the molar ratio of MDEA/PZ is of intercooler is represented as the stage number. The absorbent is
between 7/2 (8.59 wt.% PZ/41.57 wt.% MDEA) and 5/5 (21.25 wt.% drawn out from a given stage, and pumped back at the same posi-
PZ, 29.40 wt.% MDEA) [14]. Further increasing the PZ content may tion after being cooled down to a given temperature (cooling tem-
cause volatile amine losses or solvent precipitation [14]. perature) in the intercooler.
Table 5 shows that lowering both the intercooler installation
3.3. Process optimization position and cooling temperature can save more energy in the
regeneration process. By cooling to 20 °C at the 9th stage, Qreb
Parameter optimization can decrease Qreb from 3.70 to 2.74 GJ/ drops to 2.40 GJ/t, which is corresponding to a 12.6% decrease com-
t CO2 and the net efficiency penalty from 9.63% to 8.49%. However, pared to that achieved in the conventional process without inter-
the reaction heat is only 1.6 GJ/t CO2, accounting for 58% of Qreb. cooling. Thus the overall net efficiency penalty is reduced from
Hence, Qreb can be further reduced by decreasing the sensible heat 8.49% to 7.92%. In the present study, cooling water at 10 °C is
and the useless heat. Based on the essential parameters presented assumed to be available in or around the power plant to cool the
above, several enhancement measures such as absorber intercool- absorbent to 20 °C. If 10 °C cooling water is not available, cooling
ing, rich-split, and stripper interheating as well as their combina- duty is supplied by the electricity and a subsequent net efficiency
tions have been purposed to minimize Qreb. The overall energetic penalty of 0.84% would be placed on the power system.
performance of PCC processes incorporating different process opti-
mizations were evaluated, and the corresponding process configu- 3.3.2. Simple rich-split
rations are shown in Fig. 9. As shown in Fig. 9b, the simple rich-split process is designed to
recover the potential heat of vapor in the upward gaseous stream
3.3.1. Absorber intercooling and to reduce the useless heat and then Qreb. No additional
CO2 absorption using aqueous amine is exothermic, and it energy-consuming units are required for the rich-split process.
increases the temperature inside the absorber. In turn, it shifts The stripper includes 15 equilibrium stages numbered from top
 B. Zhao et al. / Applied Energy 185 (2017) 362–375 369

Fig. 6. Effect of alean on CO2 loading in rich solution (arich ) and heat duties. (a) 5 wt.% PZ; (b) 10 wt.% PZ; (c) 15 wt.% PZ; (d) 20 wt.% PZ.

to bottom (condenser is the 1st stage and reboiler is the 15th 3.3.3. Advanced rich-split
stage). The rich stream from the absorber is split into two streams. The direct mixing of the split stream with the vapor in the strip-
One is heated in E1 and then fed into the position at the 4th stage. per can lower the temperature of the CO2 stream that left the strip-
The other stream is directly fed into the stripper from the top (the per, lower Qreb and better CO2 desorption efficiency. However, the
2nd stage). The direct feeding stream can act as a quenching med- low temperature of the stripper top compromises the CO2 desorp-
ium to at least partly condense the upward vapor stream and tion efficiency of corresponding stages. A condenser is employed in
release part of CO2 in itself by utilizing the condensation heat of the advanced rich-split process for heat exchanging between one
the vapor. Thus, the rich-split modification reduces the water con- of the cold split streams and the hot lean stream from the stripper
tent in the CO2 stream and then the condenser duty. (Fig. 9c).
Fig. 10a shows that the split fractions affect Qreb and 0.2 seems Fig. 11a shows that increasing the split ratio decreases the use-
to be the optimal split ratio for this process. The temperature of less heat but increases the sensible heat simultaneously. An opti-
lean stream remains constant even with different split fractions. mal split ratio achieved is 0.3 (Fig. 11), slightly higher than that
Increasing the split fraction, the stream flow through the heat of simple rich-split. Same as in the simple rich-split process, the
exchanger E1 can achieve higher temperature due to lowering flow hot-side temperature difference of E1 is set to >8 °C (Fig. 11b).
rate. The temperature difference on the hot end (E1) could The advanced rich-split process decreases Qreb to a higher extent
decrease accordingly and is set to >8 °C (Fig. 10c) to maintain E1 by a maximum of 6.4% and reduces the condenser duty down to
efficient. Although the rich-split significantly decreases the useless 0.125 GJ/t CO2. The advanced process is also energy-competitive
heat, it also simultaneously increases the sensible heat (Fig. 10a). and does not require extra power consumption. The net power effi-
Thus an optimal split ratio of 20% leads to a Qreb of 2.61 GJ/t CO2 ciency penalty of the PCC process can be reduced from 8.49% to
corresponding to a decrease of 4.7%, and the condenser duty 8.20% by incorporating advanced rich-split modification.
decreases from 0.62 to 0.31 GJ/t CO2. This results in the net effi-
ciency penalty dropping from 8.49% to 8.28%.
Fig. 10b shows that the rich-split process has larger tempera- 3.3.4. Stripper interheating
ture gradient along the stripper and significantly higher CO2 pres- In order to make better use of the sensible heat of the hot lean
sure and lower H2O vapor pressure in the outgoing CO2 stream. stream, the stripper interheating process employs a heat exchanger
This is because that the higher temperature at the lower part of (E4) for the heat exchange between the semi-lean solutions and
the stripper favors the quick desorption of CO2 while the lower the lean stream (Fig. 9d). The semi-lean solution is extracted out
temperature at the top of the stripper benefits the condensation from the 5th stage of the stripper, and then pumped back to the
of water vapor and the recovery of the evaporation heat. same position after heat exchanging (Fig. 9d).
370 B. Zhao et al. / Applied Energy 185 (2017) 362–375

Fig. 7. Reboiler heat duty (Qreb) at different CO2 removal ratios. (a) 5 wt.% PZ; (b) 10 wt.% PZ; (c) 15 wt.% PZ; (d) 20 wt.% PZ.

Fig. 8. Effect of the MDEA/PZ ratio. (a) Effect of the MDEA/PZ ratio on CO2 loading and circulation rate of the solvent (L); (b) effect of the MDEA/PZ ratio on heat duties and the
net efficiency penalty.

Table 6 shows both Qreb and the condenser duty (the useless 3.3.5. Combination processes
heat) in interheating process are lower than that in conventional Both the simple and advanced rich-split processes require no
process. Interheating can expand the high temperature zone in extra energy consumer and reduce the net efficiency penalty from
the stripper which enhances the desorption of CO2 (Fig. 12). With 8.49% to 8.28% and 8.20%, respectively. Although the stripper inter-
both E1 and E4, more heat can be recycled from the hot lean heating modification obtains the minimum Qreb, it reduces the net
stream. The interheating process can reduce Qreb to 2.56 GJ/t CO2 efficiency penalty only by 0.24% (absolute term) down to 8.25%
and the condenser duty to 0.56 GJ/t CO2 GJ/t CO2 (Table 4). because it requires an extra semi-lean pump and its condenser
Although the interheating operation requires a pump to deliver duty remains almost unchanged. The absorber intercooling modifi-
the semi-lean solutions, pump power is negligible comparing to cation seems to be the most energy-competitive process; however,
the total energy consumption. The interheating process can its energetic feasibility greatly depends on the availability of the
decrease the net efficiency penalty from 8.49% to 8.25%. corresponding cooling water. Various combinations of the afore-
 B. Zhao et al. / Applied Energy 185 (2017) 362–375 371

Fig. 9. Schematic flow sheets of process modifications: (a) absorber intercooling; (b) simple rich-split; (c) advanced rich-split; and (d) stripper interheating.

mentioned technics were evaluated in terms of Qreb and net power other under their optimal conditions. The optimal data for MEA-
efficiency penalty (Table 7). Any combination of the aforemen- based process were reported by Li et al. [13,18]. The basic assump-
tioned technics can lead to even lower Qreb and the net power effi- tions such as the availability of the 10 °C cooling water [59] and the
ciency penalty comparing to the employment of each single scale of integrated power plant (650 MW, 38.9% efficiency) [57] in
technic. The advanced process, which incorporates the intercool- both processes are identical [13,20]. This provides a sound base for
ing, rich-split, and inter-heating together, achieves the lowest Qreb the comparison of the two processes.
of 2.24 GJ/t CO2 among all the processes in this work. And the net Table 8 summarizes the detailed comparison of the two pro-
efficiency penalty can be dropped from 8.49% to 7.66%. Although cesses. The optimal MDEA/PZ-based process has an overall net effi-
the combinations can lead to higher energy efficiency, it requires ciency penalty of 7.66%, which is 16.1% (relative term) lower than
larger capital investment. This make the situations more that of the optimal MEA-based process (9.13%). This is correspond-
complicate. ing to an increase of 1.47% in the net efficiency of a power plant
which is a significant improvement if we keep in mind that the
3.3.6. Comparison with MEA-based process optimal net power efficiency of a power plant without PCC process
Is MDEA/PZ-based technology superior than MEA-based tech- is only 38.9%. The energy consumed by CO2 product compressor
nology for PCC? Here, the two processes were compared to each are almost identical since the CO2 capture capacity of both pro-
372 B. Zhao et al. / Applied Energy 185 (2017) 362–375

Table 5
Absorber intercooling process.a

arich (mol CO2/mol total amine) L (t absorbent/t CO2) Qreb (GJ/t CO2) Energy saving (%) Net power efficiency penalty (%)
Conventional process 0.5631 10.32 2.742 – 8.49
Intercooler position: stage numberb
5 0.6159 9.31 2.485 9.37 8.06
6 0.6225 9.20 2.453 10.54 8.01
7 0.6227 9.19 2.451 10.61 8.01
8 0.6271 9.12 2.431 11.34 7.97
9 0.6300 9.07 2.418 11.82 7.95
Cooling temperature (°C)c
40 0.5812 9.98 2.646 3.50 8.33
30 0.6050 9.51 2.529 7.76 8.13
20 0.6352 8.98 2.396 12.61 7.92
a
Total equilibrium stages number is 10.
b
Cooling to 25 °C.
c
Intercooling stage is the 9th stage.

Fig. 10. Effect of split fraction for simple rich-split process. (a) Effect of split fraction on heat duties; (b) profiles of temperature and partial pressures of H2O vapor/CO2 along
the stripper, 0.20 split ratio; (c) effect of split fraction on the net power efficiency penalty and the hot end temperature difference of heat exchanger (E1).

cesses are identical. The higher energy efficiency of MDEA/PZ- this solvent system which would definitely facilitate its commer-
based process is mainly attributed to the lower net efficiency pen- cialization. The process modifications employed in this work has
alty of CO2 regeneration resulting from the lower Qreb of the strip- proven to be energetically efficient in pilot plant trials [13,54] as
per. Developing alternative sorbents of lower reaction heat with well as in the commercial plants [55]. The 10 °C logarithmic mean
CO2 is a promising way to break the energy efficiency limitation temperature difference of heat exchanger was widely accepted in
of known aqueous amine sorbents. industry. The rich industrial experience in MDEA/PZ-based solvent
The MDEA/PZ-based solvent system and the standard regenera- system and these process configurations, and the higher energetic
tive absorption-adsorption technology has been widely applied in efficiency will really attract interest of the scientific and industrial
CO2 removal from ammonia plants and natural gas plants [24]. It communities. The commercialization of the MDEA/PZ-based PCC
has established solid engineering knowledge and experience for technology proposed in this work becomes increasingly promising.
 B. Zhao et al. / Applied Energy 185 (2017) 362–375 373

Fig. 11. Effect of split fraction for advanced rich-split process. (a) Effect of split fraction on heat duties; (b) effect of split fraction on the net power efficiency penalty and hot-
end temperature difference of E1.

Table 6
Comparison of energy consumption between conventional process and interheating process.a

Net power efficiency penalty (%) Qreb (GJ/t CO2) Reaction heat (GJ/t CO2) Sensible heat (GJ/t CO2) Useless heat (GJ/t CO2)
Conventional process 8.49 2.7417 1.5905 0.5336 0.6176
Interheating process 8.25 2.5613 1.5905 0.4136 0.5572
a
The interheating exchanger is positioned at 5th stage.

Table 8
Comparison of MDEA/PZ-based process with the MEA-based process.

Advanced Advanced MDEA/PZ-

MEA-based based technology
technology
[13,18,20]
Absorbent solutions 35 wt.% MEA 30 wt.% MDEA/20 wt.%
PZ
Pstrip (MPa) 0.20 0.21
Available cooling 10 10
water (°C)
CO2 removal 85 90
ratio (%)
Treb (°C) 123.7 125.56
Qreb (GJ/t CO2) 3.10 2.24
Net efficiency 4.72 3.67
penalty of CO2
regeneration (%)
Net efficiency penalty 3.15 3.16
Fig. 12. Profiles of temperature and partial pressures of H2O vapor/CO2 along the of CO2 compression
stripper in conventional process and interheating process. (%)
Net efficiency penalty of 1.26 0.83
auxiliary (%)
Overall net efficiency 9.13 7.66
4. Conclusion penalty (%)

An MDEA/PZ-based solvent system (20 wt.% PZ and 30 wt.%

MDEA) was proposed for post combustion CO2 capture in a
650 MW power plant. Process simulations using Aspen Plus were reaction heat with CO2 and the resultant lower energy penalty. A
conducted to evaluate the MDEA/PZ-based technology and to low Qreb value of 2.74 GJ/t CO2 was achieved in the conventional
understand the effects of process parameters and various process PCC process under the optimal operation conditions of 0.1 alean ,
modifications. The results showed that the MDEA/PZ-based solvent 90% CO2 removal, 0.11 MPa absorption pressure (Pabs), and
is a better aqueous amine system for CO2 capture due to its lower 0.21 MPa stripping pressure (Pstrip).

Table 7
Results of combination processes.

Conventional process Combined intercooling and rich-split Combined intercooling and interheating Advanced process
Qreb (GJ/t CO2) 2.7417 2.3187 2.2755 2.2409
Energy saving (%) – 15.4 17.0 18.3
Net power efficiency penalty (%) 8.49 7.79 7.73 7.66
374 B. Zhao et al. / Applied Energy 185 (2017) 362–375

Absorber intercooling, simple rich-split, advanced rich-split, [21] And IK, Svendsen HF. Heat of absorption of carbon dioxide (CO2) in
monoethanolamine (MEA) and 2-(aminoethyl)ethanolamine (AEEA)
and stripper interheating were found to be efficient enhancement
Solutions. Ind Eng Chem Res 2007;46:5803–9.
measures in improving the process energetic efficiency. Combining [22] Li B-H, Zhang N, Smith R. Simulation and analysis of CO2 capture process with
these process modifications leads to a superior MDEA/PZ-based aqueous monoethanolamine solution. Appl Energy 2016;161:707–17.
PCC process with the highest energy efficiency. The reboiler heat [23] Ibrahim AY, Ashour FH, Ghallab AO, Ali M. Effects of piperazine on carbon
dioxide removal from natural gas using aqueous methyl diethanol amine. J Nat
duty (Qreb) of this process is as low as 2.24 GJ/t CO2. The resultant Gas Sci Eng 2014;21:894–9.
net power efficiency penalty of the MDEA/PZ-based process is [24] Mudhasakul S, Ku H-M, Douglas PL. A simulation model of a CO2 absorption
7.66% which is 16.1% (relative term) less than that of the optimal process with methyldiethanolamine solvent and piperazine as an activator. Int
J Greenhouse Gas Control 2013;15:134–41.
MEA-based process (9.13%). MDEA/PZ-based PCC process is more [25] Svensson H, Hulteberg C, Karlsson HT. Heat of absorption of CO2 in aqueous
competitive in industrial applications. solutions of N-methyldiethanolamine and piperazine. Int J Greenhouse Gas
Control 2013;17:89–98.
[26] Austgen DM, Rochelle GT, Peng X, Chen CC. Model of vapor-liquid equilibria for
aqueous acid gas-alkanolamine systems using the electrolyte-NRTL equation.
Acknowledgements Ind Eng Chem Res 1989;28:1060–73.
[27] Fang YJ, Carroll JJ, Mather AE, Otto FD. Solubility of mixtures of hydrogen
The authors are grateful to the financial support of the SINOPEC sulfide and carbon dioxide in aqueous N-methyldiethanolamine solutions. Ind
Eng Chem Process Des Dev 1982;21:539–44.
under the contract # 415023 and the key project of the National [28] Jou FY, Carroll JJ, Mather AE, Otto FD. The solubility of carbon dioxide and
Natural Science Foundation of China (NSFC 21336004). The authors hydrogen sulfide in a 35 wt.% aqueous solution of methyldiethanolamine. Can
would like to acknowledge China Chengda Engineering Co., Ltd. for J Chem Eng 1993;71:264–8.
[29] Jou FY, Mather AE, Otto FD, Carroll JJ. Experimental investigation of the phase
its software support in this work. equilibria in the carbon dioxide-propane-3 M MDEA system. Ind Eng Chem Res
1995;34:2526–9.
[30] Alghawas HA, Hagewiesche DP, Ruizibanez G, Sandall OC. Physicochemical
properties important for carbon dioxide absorption in aqueous
References
methyldiethanolamine. J Chem Eng Data 1989;34:385–91.
[31] Browning GJ, Weiland RH. Physical solubility of carbon dioxide in aqueous
[1] Metz B, Davidson OR, Bosch PR, Dave R, Meyer LA. Contribution of working alkanolamines via nitrous oxide analogy. J Chem Eng Data 1994;39:817–22.
group III to the fourth assessment report of the intergovernmental panel on [32] Haimour N, Sandall OC. Absorption of carbon dioxide into aqueous
climate change; 2007. methyldiethanolamine. Chem Eng Sci 1984;39:1791–6.
[2] Zhang Y, Chen H, Chen C-C, Plaza JM, Dugas R, Rochelle GT. Rate-based process [33] Li M-H, Lai M-D. Solubility and diffusivity of N2O and CO2 in
modeling study of CO2 capture with aqueous monoethanolamine solution. Ind (monoethanolamine+ N-methyldiethanolamine+water) and in
Eng Chem Res 2009;48:9233–46. (monoethanolamine+2-amino-2-methyl-1-propanol+water). J Chem Eng
[3] Dutcher B, Fan M, Russell AG. Amine-based CO2 capture technology Data 1995;40:486–92.
development from the beginning of 2013-a review. ACS Appl Mater Interf [34] Versteeg GF, Swaalj WV. Solubility and diffusivity of acid gases (carbon
2015;7:2137–48. dioxide, nitrous oxide) in aqueous alkanolamine solutions. J Chem Eng Data
[4] Middleton RS, Eccles JK. The complex future of CO2 capture and storage: variable 1988;33:29–34.
electricity generation and fossil fuel power. Appl Energy 2013;108:66–73. [35] Bishnoi S, Rochelle GT. Physical and chemical solubility of carbon dioxide in
[5] van den Broek M, Hoefnagels R, Rubin E, Turkenburg W, Faaij A. Effects of aqueous methyldiethanolamine. Fluid Phase Equilib 2000;168:241–58.
technological learning on future cost and performance of power plants with [36] Polasek JC, Inglesias-Silva G, Bullin JA. Using mixed amine solutions for gas
CO2 capture. Prog Energy Combust Sci 2009;35:457–80. sweetening. In: Proceedings of the annual convention-gas processors
[6] Kunze C, Spliethoff H. Assessment of oxy-fuel, pre- and post-combustion- association. p. 58–63.
based carbon capture for future IGCC plants. Appl Energy 2012;94:109–16. [37] Bishnoi S, Rochelle GT. Absorption of carbon dioxide into aqueous piperazine:
[7] Yan J. Carbon capture and storage (CCS). Appl Energy 2015;148:A1–6. reaction kinetics, mass transfer and solubility. Chem Eng Sci
[8] Abu-Zahra MRM, Niederer JPM, Feron PHM, Versteeg GF. CO2 capture from 2000;55:5531–43.
power plants: Part II. A parametric study of the economical performance based [38] Derks PWJ, Dijkstra HBS, Hogendoorn JA, Versteeg GF. Solubility of carbon
on mono-ethanolamine. Int J Greenhouse Gas Control 2007;1:135–42. dioxide in aqueous piperazine solutions. AIChE J 2005;51:2311–27.
[9] Gerbelová H, Versteeg P, Ioakimidis CS, Ferrão P. The effect of retrofitting [39] Hilliard MD. A predictive thermodynamic model for an aqueous blend of
Portuguese fossil fuel power plants with CCS. Appl Energy 2013;101:280–7. potassium carbonate, piperazine, and monoethanolamine for carbon dioxide
[10] Raynal L, Bouillon P-A, Gomez A, Broutin P. From MEA to demixing solvents capture from flue gas Dissertation. Austin (USA): University of Texas; 2008.
and future steps, a roadmap for lowering the cost of post-combustion carbon [40] Derks PWJ. Carbon dioxide absorption in piperazine activated N-
capture. Chem Eng J 2011;171:742–52. methyldiethanolamine Dissertation. The Netherlands: University of Twente;
[11] Wang M, Joel AS, Ramshaw C, Eimer D, Musa NM. Process intensification for 2006.
post-combustion CO2 capture with chemical absorption: a critical review. Appl [41] Ali BS, Aroua MK. Effect of piperazine on CO2 loading in aqueous solutions of
Energy 2015;158:275–91. MDEA at low pressure. Int J Thermophys 2004;25:1863–70.
[12] Steeneveldt R, Berger B, Torp TA. CO2 capture and storage: closing the [42] Xu G-W, Zhang C-F, Qin S-J, Gao W-H, Liu H-B. Gas–liquid equilibrium in a
knowing-doing gap. Chem Eng Res Des 2006;84:739–63. CO2–MDEA–H2O system and the effect of piperazine on it. Ind Eng Chem Res
[13] Li K, Leigh W, Feron P, Yu H, Tade M. Systematic study of aqueous 1998;37:1473–7.
monoethanolamine (MEA)-based CO2 capture process: techno-economic [43] Liu HB, Zhang Chengfang, Xu GW. A study on equilibrium solubility for carbon
assessment of the MEA process and its improvements. Appl Energy dioxide in methyldiethanolamine–piperazine–water solution. Ind Eng Chem
2016;165:648–59. Res 1999;38:4032–6.
[14] Frailie PT. Modeling of carbon dioxide absorption/stripping by aqueous [44] Bishnoi B, Rochelle GT. Absorption of carbon dioxide in aqueous piperazine/
methyldiethanolamine/piperazine Dissertation. Austin (USA): University of methyldiethanolamine. AIChE J 2002;48:2788–99.
Texas; 2014. [45] Oexmann J. Post-combustion CO2 capture: energetic evaluation of chemical
[15] Tobiesen FA, Svendsen HF, Mejdell T. Modeling of blast furnace CO2 capture absorption processes in coal-fired steam power plants
using amine absorbents. Ind Eng Chem Res 2007;46:7811–9. Dissertation. Germany: Hamburg University of Technology; 2011.
[16] Biliyok C, Lawal A, Wang M, Seibert F. Dynamic modelling, validation and [46] Knudsen JN, Andersen J, Jensen JN, Biede O. Evaluation of process upgrades and
analysis of post-combustion chemical absorption CO2 capture plant. Int J novel solvents for the post combustion CO2 capture process in pilot-scale.
Greenhouse Gas Control 2012;9:428–45. Energy Proc 2011;4:1558–65.
[17] Huang B, Xu S, Gao S, Liu L, Tao J, Niu H, et al. Industrial test and techno- [47] Cousins A, Wardhaugh LT, Feron PHM. Preliminary analysis of process flow
economic analysis of CO2 capture in Huaneng Beijing coal-fired power station. sheet modifications for energy efficient CO2 capture from flue gases using
Appl Energy 2010;87:3347–54. chemical absorption. Chem Eng Res Des 2011;89:1237–51.
[18] Li K, Cousins A, Yu H, Feron P, Tade M, Luo W, et al. Systematic study of [48] Cousins A, Wardhaugh LT, Feron PHM. A survey of process flow sheet
aqueous monoethanolamine-based CO2 capture process: model development modifications for energy efficient CO2 capture from flue gases using chemical
and process improvement. Energy Sci Eng 2016;4:23–39. absorption. Int J Greenhouse Gas Control 2011;5:605–19.
[19] Abu-Zahra MRM, Schneiders LHJ, Niederer JPM, Feron PHM, Versteeg GF. CO2 [49] Cousins A, Wardhaugh LT, Feron PHM. Analysis of combined process flow
capture from power plants: Part I. A parametric study of the technical sheet modifications for energy efficient CO2 capture from flue gases using
performance based on monoethanolamine. Int J Greenhouse Gas Control chemical absorption. Energy Proc 2011;4:1331–8.
2007;1:37–46. [50] Karimi M, Hillestad M, Svendsen HF. Capital costs and energy considerations of
[20] Li K, Yu H, Feron P, Tade M, Wardhaugh L. Technical and energy performance different alternative stripper configurations for post combustion CO2 capture.
of an advanced, aqueous ammonia-based CO2 capture technology for a Chem Eng Res Des 2011;89:1229–36.
500 MW coal-fired power station. Environ Sci Technol 2015;49:10243–52.
 B. Zhao et al. / Applied Energy 185 (2017) 362–375 375

[51] Le Moullec Y, Kanniche M. Screening of flowsheet modifications for an efficient [58] Oexmann J, Kather A. Minimising the regeneration heat duty of post-
monoethanolamine (MEA) based post-combustion CO2 capture. Int J combustion CO2 capture by wet chemical absorption: the misguided focus
Greenhouse Gas Control 2011;5:727–40. on low heat of absorption solvents. Int J Greenhouse Gas Control
[52] Le Moullec Y, Neveux T, Al Azki A, Chikukwa A, Hoff KA. Process modifications 2010;4:36–43.
for solvent-based post-combustion CO2 capture. Int J Greenhouse Gas Control [59] Linnenberg S, Darde V, Oexmann J, Kather A, Well WJMV, Thomsen K.
2014;31:96–112. Evaluating the impact of an ammonia-based post-combustion CO2 capture
[53] Li K, Yu H, Yan S, Feron P, Wardhaugh L, Tade M. Technoeconomic assessment process on a steam power plant with different cooling water temperatures. Int
of an advanced aqueous ammonia-based postcombustion capture process J Greenhouse Gas Control 2012;10:1–14.
integrated with a 650-MW coal-fired power station. Environ Sci Technol [60] Goto K, Yogo K, Higashii T. A review of efficiency penalty in a coal-fired power
2016;50:10746–55. plant with post-combustion CO2 capture. Appl Energy 2013;111:710–20.
[54] Oh S-Y, Binns M, Cho H, Kim J-K. Energy minimization of MEA-based CO2 [61] Oexmann J, Hensel C, Kather A. Post-combustion CO2-capture from coal-fired
capture process. Appl Energy 2016;169:353–62. power plants: Preliminary evaluation of an integrated chemical absorption
[55] Cousins A, Cottrell A, Lawson A, Huang S, Feron PHM. Model verification and process with piperazine-promoted potassium carbonate. Int J Greenhouse Gas
evaluation of the rich-split process modification at an Australian-based post Control 2008;2:539–52.
combustion CO2 capture pilot plant. Greenhouse Gases Sci Technol [62] And ABR, Rubin ES. A technical, economic, and environmental assessment of
2012;2:329–45. amine-based CO2 capture technology for power plant greenhouse gas control.
[56] Wang M, Lawal A, Stephenson P, Sidders J, Ramshaw C. Post-combustion CO2 Environ Sci Technol 2002;36:4467–75.
capture with chemical absorption: a state-of-the-art review. Chem Eng Res [63] Alhajaj A, Mac Dowell N, Shah N. A techno-economic analysis of post-
Des 2011;89:1609–24. combustion CO2 capture and compression applied to a combined cycle gas
[57] EIA. Updated capital cost estimates for utility scale electricity generating turbine: Part I. A parametric study of the key technical performance indicators.
plants. U.S. Energy Information Administration; 2013. Int J Greenhouse Gas Control 2016;44:26–41.