International Journal of Greenhouse Gas Control 10 (2012) 148163 [628921]

International Journal of Greenhouse Gas Control 10 (2012) 148–163
Contents lists available at SciVerse ScienceDirect
International Journal of Greenhouse Gas Control
j our na l ho me p age: www.elsevier.com/locate/ijggc
CO 2capture in power plants: Minimization of the investment and operating cost
of
the post-combustion process using MEA aqueous solution
Patricia Moresa, Néstor Rodrígueza, Nicolas Scennaa,b, Sergio Mussatia,b,∗,1
aCAIMI Centro de Aplicaciones Informáticas y Modelado en Ingeniería, Universidad Teconologica Nacional Facultad Regional Rosario – Zeballos 1341 – S2000BQA – Rosario, Argentina
bINGAR (CONICET-UTN) Instituto de Desarrollo y Dise˜no, Avellaneda 3657 – 3000 – Santa Fe, Argentina
a r t i c l e i n f o
Article history:
Received
19 January 2012
Received
in revised form 30 May 2012
Accepted
4 June 2012
Available
online 6 July 2012
Keywords:CO
2capture cost
Mathematical
modeling
Optimization
NLP models
Greenhouse
gas emissions
Post-combustion
processa b s t r a c t
The post combustion process based on the CO 2absorption using amine aqueous solution is one of the more
attractive options to drastically reduce greenhouse gas emissions from electric power sector. However,
the solvent regeneration is highly energy intensive affecting the total operating cost significantly. The
CO 2removal target depends on the absorption and desorption processes where the main parameters
of both processes are strongly coupled. Consequently, the simultaneous optimization of the whole CO 2
capture process is essential to determine the best design and operating conditions in order to minimize
the total cost.
This paper presents and discusses different cost optimizations including both investments and oper-
ating costs. The impact of different CO 2emission reduction targets on the total annual cost, operating
conditions and dimensions of process units is investigated in detail. Optimized results are discussed
through different case studies.
© 2012 Elsevier Ltd. All rights reserved.
1. Introduction
Over the last decades, amine gas sweetening has become a
proven technology for the removal of CO 2from natural gas. Cer-
tainly, CO 2is efficiently captured in some large industrial plants
such as natural gas processing and ammonia production plants.
However, the CO 2capture from large power plants is still under
development.
There are three processes which have been shown to be techni-
cally feasible to CO 2capture from flue gas of fossil-fueled power
plants: post-combustion, pre-combustion and oxy-fuel combus-
tion.
The post-combustion process using amine, which will be stud-
ied in this paper, is the most promising technology due to its
capacity to treat large volumes of flue gas and it is well suited
for retrofitting existing plants since this process is an end of the
pipe treatment. However, it is still highly energy intensive due to
Abbreviations: ABS, absorber; BLOW, blower; COMP, compressor; COND, con-
denser;
ECO, economizer; I-COOL, inter stage cooler; LAC, lean amine cooler; MEA,
monoethanolamine;
REB, reboiler; REG, regenerator.
∗Corresponding
author at: Avellaneda 3657, 3000 Santa Fe, Argentina.
Tel.:
+54 342 4534451; fax: +54 342 4553439.
E-mail addresses: patricia mores@hotmail.com (P. Mores),
nestorhugo
r@yahoo.com (N. Rodríguez), nscenna@santafe-conicet.gov.ar
(N.
Scenna), mussati@santafe-conicet.gov.ar (S. Mussati).
1Tel.: +54 342 4534451; fax: +54 342 4553439.the thermal energy requirement needed to regenerate the amine
solution which increases the operating cost drastically. The loss of
amine in the regenerator unit is another drawback of this method.
During the last years many research activities in CO 2capture
have been done in different lines from experimental studies at
laboratory scale and pilot plants to the development and imple-
mentation of mathematical models in computers.
Much work on experimental studies at laboratory scale is being
carried out by various workers in order to identify and determine
the concentration of ionic species in the CO 2capture process using
amine aqueous solutions. One of the analytical methods used for
studying the species distribution in solutions of carbon dioxide in
aqueous monoethanolamine (MEA) and diethanolamine (DEA) is
the nuclear magnetic resonance (NMR) (Choi et al., 2012; Böttinger
et al., 2008; Yoon and Lee, 2003 ). The following are the major exper-
imental variables investigated: the absorbed amount of CO 2or CO 2
loading, ion concentrations and temperature. The results obtained
in this research area not only improve the understanding of solution
behavior associated with absorption and regeneration reactions but
also provide the information needed to develop a model capable of
describing the L–V equilibrium which is fundamental for simulation
and optimization of CO 2capture process using amines.
Particular emphasis is also placed on the study of reaction mech-
anisms between the different types of amines and CO 2(Vaidya and
Kenig, 2007, 2009; Xie et al., 2010 ). In addition, several researchers
are focusing on the identification of oxidative and thermal degrada-
tions of amine solvents (Islam et al., 2011; Lawal and Idem, 2006;
1750-5836/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.ijggc.2012.06.002

P. Mores et al. / International Journal of Greenhouse Gas Control 10 (2012) 148–163 149
Nomenclature
ACC annualized capital cost (million US$/year)
CRF capital recovery factor
DC direct cost (million US$)
G flue-gas flow-rate (mol/s)
H enthalpy (kJ)
I cost index
i interest rate
IC indirect cost (million US$)
L liquid flow-rate (mol/s)
n compounding period
OLC operating labor cost (million US$/year)
TO&MC total operating and maintenance cost (million
US$/year)
P pressure (kPa)
PC purchased cost (million US$)
SC start-up cost (million US$/year)
T temperature (K)
TAC total annual cost (million US$/year)
TCI total capital investment (million US$)
TPC total purchased equipment cost (million US$)
TOLC total operating labor cost (million US$/year)
TU&MUC total utility cost including heating and cooling util-
ities, electricity, MEA and water make-ups and R&D
costs (million US$/year)
U&MUC total utility cost including heating and cooling util-
ities, electricity, MEA and water make-ups (million
US$/year)
WC working capital (million US$/year)
X capacity: heat transfer area for heat exchangers
(m2); volume for absorber, stripper and tanks (m3);
real power for pumps, blower and compressors
(kW)
y mole fraction in vapor phase
KDC numerical constant used to compute the direct cost
KIC numerical constant used to compute the indirect
cost
KWC numerical constant used to compute the working
capital cost
KSC numerical constant used to compute start-up costs
˛CO2loading CO 2loading factor (CO 2mol/MEA mol)
Superscripts
0 reference value
1 current value
Subscripts
in inlet
out outlet
k piece of equipment (ABS, REG, BLOW, RAP, LAC, ECO,
MEA-TK, H2O-TK, COND, REB, COMP, I-COOL)
Lepaumier et al., 2010 ). Knowledge of degradation products and
main reactions allows a better understanding of amines chemical
stability for CO 2capture application.
In regards to the application of the mathematical modeling of
CO 2capture processes using amine, many articles focusing on the
simulations and parametric optimizations in order to maximize
the absorption efficiency have been published. A great number of
authors employed process simulators such as: Aspen Plus (Cozad
et al., 2010; Abu-Zahra et al., 2007a,b; Duan et al., 2012; Ali et al.,
2005 ), HYSYS (Khakdaman et al., 2008; Aliabad and Mirzaei, 2009 ),
and Aspen HYSYS (Oi, 2007 ), gProms (Kvamsdal et al., 2009; Lawalet
al., 2009 ). Other authors proposed and developed in-house
simulation algorithms (Rahimpour and Kashkooli, 2004; Sipöcz
et al., 2011; Aroonwilas and Veawab, 2007 ). Freguia and Rochelle
(2003) integrated a Fortran subroutine into Aspen Plus®to per-
form a rate-based calculation of CO 2absorption into MEA. In these
studies, key insights have been gained for the absorber and regen-
erator units. In most of these articles, the absorption column and
regeneration unit have been studied separately. The effect of the
pressure and temperature of absorber and different type of amines
(primary, secondary and tertiary amines) on the CO 2absorption
efficiency was studied. In addition, the reboiler heat duty and
operating conditions in the regenerator unit were also studied
in detail. More specifically, many authors analyzed the effect of
the different operating conditions for different types of amines in
order to minimize the heat duty in the regenerator unit because
it has a significant influence on the total operating cost (Oexmann
and Kather, 2010; Nuchitprasittichai and Cremaschi, 2010 ).
On the other hand, several authors performed techno-economic
evaluations of postcombustion processes (Karimi et al., 2011; Dave
et al., 2011; Ho et al., 2011; Huang et al., in press; Kuramochi et al.,
2010; Schach et al., 2010a,b; Oexmann et al., 2008; Peeters et al.,
2007; Romeo et al., 2008; Singh et al., 2003 , among others).
Karimi et al. (2011) investigated and compared five different
configurations for aqueous absorption/stripping analyzing their
energy requirements and capital costs and the best configurations
were defined based on CO 2avoided cost and total capture cost.
The following are the configurations analyzed: (a) conventional
process configuration, (b) split-stream configuration where the
rich amine is split into two streams going to two sections of the
stripper after preheating with two separate lean amine streams,
(c) multi-pressure stripper configuration, (d) vapor recompression
configuration and (e) compressor integration.
Unisim Design and ProTreat software were used for simulations,
while for the cost calculations, data from Turton et al. (2008) and
Sinnott et al. (2009) were considered. Results revealed that vapor
recompression configuration is the best configuration because it
has the lowest total capture cost and CO 2avoided cost and the
plant complexity does not increase very much compared to the
benchmark. The split-stream configuration with cooling of semi-
lean amine was the second best but this configuration increases
the investment cost and plant complexity significantly.
Dave et al. (2011) performed a techno-economic comparative
assessment for an amine based post-combustion capture (PCC) pro-
cess applied to representative coal-fired power plants in Australia
and China. The assessment is based on an in-depth analysis of the
cost of generation. Aspen Rate-Sep, Steam-Pro, Steam-Master and
PEACE software packages were used for process modeling and cost
estimation of the integrated power and capture plants. According
to the circumstances of both countries, the comparison reveals
that post-combustion capture in China can benefit significantly
from more energy efficient processes, whereas for the Australian
circumstances the focus should be on reduction of capital costs.
Development of more energy efficient PCC-processes will be most
beneficial in China, whereas in Australia lower capital costs should
be aimed for.
Kuramochi et al. (2010) compared the techno-economic perfor-
mance of post-combustion CO 2capture from industrial Natural Gas
Combined Cycle (NGCC) Combined Heat and Power plants (CHPs)
of scales from 50 MWe to 200 MWe with large-scale (400 MWe)
NGCC for short-term (2010) future. The results have shown that
the efficiency improvement by the better use of CHP capacity
for meeting CO 2capture energy demands and the long operation
time can potentially overweigh the disadvantage of higher cap-
ital costs. It should be mentioned, however, that the study was
based on a number of generalized relationships between the plant
scale and technical and economic performances of NGCC-CHP and

150 P. Mores et al. / International Journal of Greenhouse Gas Control 10 (2012) 148–163
CO 2capture systems. In reality, industrial NGCC-CHPs are “tailor-
made” for each industrial plant to meet plant-specific demands
and conditions. The obtained results therefore only provide general
indications about the techno-economic competitiveness of post-
combustion CO 2capture for medium scale industrial NGCC-CHPs.
Schach et al. (2010b) applied an exergoeconomic analysis for
the following three different configurations: (a) a standard absorp-
tion/stripping cycle which was used as the base case and represents
the benchmark, (b) absorber including an intercooler and (c)
split-stream configuration. Aspen Plus was used to perform the
simulations and RadFrac model was applied for the absorption and
stripper columns. Configurations (b) and (c) show better results
than the reference case. The matrix stripper configuration has
the best exergetic efficiency and also the lowest energy demand.
However, the cost of CO 2-avoided is higher than the cost of the con-
figuration with intercooler. This is due to the higher investment cost
for the matrix stripper. An additional stripper column and a sec-
ond cross heat exchanger are needed, whereas for the intercooler
configuration only an additional heat exchanger is required.
Oexmann et al. (2008) performed simulation campaigns in
Aspen Plus in order to find the process parameters of the full
CO 2-capture process for chemical absorption of CO 2by piperazine-
promoted potassium carbonate (K2CO 3/PZ) and the subsequent
CO 2-compression train that minimize the specific power loss.
Subsequently, absorber and desorber columns are dimensioned
to evaluate the corresponding investment costs. Regeneration
heat duty, net efficiency losses and column investment costs
are then compared to the reference case of CO 2-capture by
monoethanolamine (MEA). The results showed that, for the param-
eter values assumed, the investment costs of the absorption and
regeneration units using piperazine-promoted potassium carbon-
ate are lower than the reference process which uses MEA. The
enhanced reaction kinetics of the investigated K2CO 3/PZ solvent
lead to smaller column sizes.
Singh et al. (2003) conducted a techno-economic study of CO 2
capture from an existing coal-fired power plant adopting MEA
scrubbing (post-combustion capture) and O2/CO 2recycle combus-
tion. The results showed that both processes are expensive options
to capture CO 2from coal power plants. However, O2/CO 2recy-
cle combustion appears to be a more attractive retrofit than MEA
scrubbing due to a lower CO 2emission.
Most of the mentioned articles focus on the parametric opti-
mization of the process variables. To the knowledge of the authors,
only few articles dealing with the simultaneous optimization of the
whole post-combustion process involving detailed cost equations
and rigorous modeling of the process-equipments have been dis-
cussed in the specific literature (Lawal et al., 2009; Fashami et al.,
2009 ).
The development and implementation of detailed mathemat-
ical models of the post-combustion CO 2process using amines in
advanced optimization tools is often a complex task because it
involves developing accurate and rigorous models to describe all
pieces of equipments, including an absorber, stripper, liquid–vapor
equilibrium of the system, CO 2compression system and heat trans-
fer units, among others. Moreover, the development of a systematic
algorithmic procedure to find optimal solutions using realistic
cost functions with process simulators is difficult due to the recy-
cle structures contemplated in the flow-sheet (Oi, 2007; Vozniuk,
2010 ). In addition, the constraints needed to model the pieces of
equipments are usually non-linear and non-convex which may lead
to convergence problems.
This paper presents an optimization mathematical model con-
sisting in the minimization of the process total annual cost subject
to mass, energy and momentum balances where the process vari-
ables are optimized simultaneously. The proposed model used for
optimization is an extension of previous models presented in Moreset
al. (2011, 2012) . Certainly, the previous models were used to
study the absorption–desorption processes by solving several opti-
mization problems considering different efficiency criteria. These
models have now been extended to include the CO 2compression
stages and a complete cost model taking into account the invest-
ment and operating cost of each one of the pieces of equipments.
The resulting model is deterministic in nature and highly non-
linear. Temperature, flow-rate and composition profiles along the
absorber and regenerator units, dimensions (height and diameter)
and reboiler heat duty are some of the main variables which are
simultaneously optimized.
Flue-gas specification coming from a combined cycle power
plant, solvent concentration and CO 2removal target are considered
as model parameters (known values). General Algebraic Modeling
System (GAMS) which is a high-level algebraic modeling system for
large scale optimization is used for implementation and solving the
resulting mathematical model. In general, mathematical program-
ming environments such as GAMS, gPROMS, and AMPL have shown
to be powerful tools, especially when the optimization problem is
large, combinatorial and highly non-linear.
The paper is outlined as follows. Section 2 briefly describes
the post-combustion process. Section 3 introduces the problem
formulation. Section 4 summarizes the assumptions and the math-
ematical model. Sections 5 and 6 present and discuss the optimized
results obtained from the developed NLP model. Finally, Section 7
presents the conclusions and future work.
2. Process description
Fig. 1 shows a schematic overview of the post-combustion pro-
cess using MEA aqueous solution for CO 2capture from a flue gas
stream coming from a power plant.
As shown, the flue gas stream [Gin] mainly composed by CO 2,
O2, H2O, N2coming from a coal fired power plants (or gas-fired
combined cycle power stations) is fed into the bottom section of
the absorber [ABS] while the lean MEA aqueous solution [LLC] is fed
into the top section of the absorber.
The CO 2is chemically absorbed by the amine solvent as the flue-
gas stream passes upward through the packing material inside the
absorber. Then, the clean gas stream [Gout], with the most of the CO 2
removed, is emitted to the atmosphere. The CO 2rich solution leav-
ing the absorber [LRC] passes through the rich-lean heat exchanger
[ECO] where it increases temperature up to TRHand enters at the
top of the stripper [REG] for regeneration. The steam supplied in
the reboiler [REB] placed at the bottom of the regenerator is used
to heat the CO 2rich solution up to the boiling temperature [TREB].
Thus, the CO 2captured in the absorber is released from the CO 2
rich-amine solvent in the regenerator. As shown, from the top of
the stripper column, the stream mainly composed by CO 2product,
water vapor, and entrained amine enters at the reflux condenser
[COND] where a partial condensation of the stream is carried out
and the resulting stream accumulated in the reflux drum is pumped
back to the top of the stripper column. The final CO 2product is
delivered to the compression section where it is dried and com-
pressed through multi-stage compressors [COMP].
Finally, the lean solvent [LREB] is pumped from the regenerator
through the rich-lean heat exchanger [ECO] and the cooler [LAC]
and is again fed into the absorber top.
The selection of type of the amine aqueous solvent plays an
important role in the process efficiency. The following are some
of the aspects that must be considered when selecting the type of
solvent: corrosion, amine degradation, and energy needed to regen-
erate the amine solvent. In general, the absorption and desorption
processes are affected in opposite way by all types of amines; higher
absorption efficiency, higher energy requirement for regeneration.

P. Mores et al. / International Journal of Greenhouse Gas Control 10 (2012) 148–163 151
Fig. 1. Schematic diagram of the post-combustion CO 2capture using amines.
The overall efficiency of the absorption–desorption process
strongly depends on the type of amine used, steam temperature
and pressure, CO 2removal target, dimensions of the pieces of
equipment (heaters, reboiler, condenser, absorber, stripper, pumps,
compressors) and operating conditions (flow-rates, pressures, tem-
peratures).
According to the above mentioned, it is easy to conclude that
there exist many trade-offs between the process parameters and
variables. Therefore, the simultaneous optimization of the process
variables is essential to determine the best design of the entire CO 2
capture process.
3. Problem formulation
The optimization problem can be stated as follows. Given
the flue gas conditions (composition, temperature and flow-rate)
and the MEA solution concentration, the goal is to deter-
mine the optimal operating conditions and dimensions of the
absorber/regenerator columns, compressors, pumps and heat
exchangers in order to meet a given CO 2emission reduction target
at minimum total annual cost.
Mathematically, the numerical optimization problem stated in
this paper can be written in the following general form:
Minimize
F(x)
subject to : Hs(x) = 0 ∀s
Gt(x) ≤ 0 ∀t
Design specifications
where x is the vector of the model variables (heat duty in the
reboiler, condenser and heat exchangers, electricity consumed by
pumps and compressors, dimensions of absorber and regenera-
tor units, H2O and MEA make-up flow-rates). In addition, vector
x also includes the following continuous decisions: temperature,
composition and flow-rate profiles of aqueous solution and flue
gas streams along the columns.F(x)
refers to the objective function. In this model, the objective
function is the total annual cost (TAC) and is computed by Eq. (1).
Hs(x) refers to equality constraints (mass and energy balances, cor-
relations used to compute physical/chemical properties, pressure
drops, investment and operating expenditures, among others).
Basically, the equality constraints Hs(x) of the economic model are
given from Eqs. (2)–(17) . The remaining constraints concerning
the mass, energy and momentum balances can be found in Mores
et al. (2011, 2012) . On the other hand, Gt(x) refers to inequality
constraints which are used, for example, to avoid temperature
crossover and to impose lower and upper bounds in some of
the operating variables. Finally, design specifications refer to the
model parameters (known and fixed values), e.g. the CO 2removal
target, flue-gas conditions and specific costs of heating and cooling
utilities.
The investment and operating costs are computed in terms of
the main process variables of each one of the pieces of equipment
which are simultaneously optimized.
Finally, exploiting the robustness of the proposed model, the
influence of the CO 2removal target on the total annual cost is also
investigated.
4. Assumptions and mathematical model
The complete mathematical model involves mass, energy and
momentum balances (physical-thermodynamic model) and con-
straints used to compute investment and operating cost of each
one of the pieces of equipments (cost model). As follows, the main
assumptions and the mathematical model are presented.
•A
low CO 2concentration in flue gas stream is assumed. The main
reason of this is that the developed model will be coupled into
a combined cycle power plant in order to optimize the whole
integrated plant. The CO 2concentration in exhaust gases of such
power plants is low.
•A maximum gas flow-rate that can be treated by one absorp-
tion train recommended in the literature is assumed in this

152 P. Mores et al. / International Journal of Greenhouse Gas Control 10 (2012) 148–163
paper. According to the literature, the largest economic sin-
gle train is approx. 4600 t per day from coal-fired flue gas or
2400 t per day from natural gas, based on a maximum col-
umn diameter of 12.8 m (Chapel et al., 1999 ). Therefore, it is
considered that the whole process involves one absorption col-
umn and one regenerator unit. For larger flue gas flow-rates
mentioned above, the number of absorbers and compressors
should be increased. Based on the two previous assumptions, it
is not necessary to specify the size of the reference power plant
(MWe).
•Packed columns are considered for the CO 2absorption and
amine regeneration. As first approximation, random packing was
assumed. The model will be further extended to also consider
structured packing.
•For mathematical modeling purpose, the total height of absorber
and regenerator columns is divided into N stages which allow to
compute profiles of temperatures, flow-rates and compositions
along the column. Ten stages (N = 10) are considered for absorp-
tion and regeneration columns. The first stage (Stage 1) refers to
the bottom of the columns and the last stage (Stage 10) denotes
the top of the columns.
•As first approximation, liquid and vapor phases are well-mixed.
Thus, there is no concentration and temperature gradients in sin-
gle liquid and vapor phases and point efficiency is equivalent to
Murphree efficiency (/DC1).
•Stage
efficiency may be computed similarly to the tray efficiency.
Dependence of stage efficiency with gas and liquid velocities and
enhancement factor, among others, is considered.
•Non-ideal behavior in the gas phase is assumed. Fugacity coeffi-
cients are computed by using Peng–Robinson equations of state
for a multi-component mixture (Peng and Robinson, 1976 ).
•As first approximation, ideal behavior in the liquid phase is
assumed.
•An enhancement factor is used to introduce the effect of the
chemical reaction on the CO 2transfer.
•Aqueous MEA amine solution is used as the solvent (30 wt.%).
•As
first approximation, thermal equilibrium is assumed between
the liquid and gas phases.
•Pressure drop along the absorber and regenerator units is con-
sidered by using Robbins correlation (Robbins, 1991 ).
•Dimensions of both columns are considered as optimization vari-
ables. Certainly, the heights and diameters of both columns and
the operating conditions are optimized simultaneously.
•CO 2and H2O are the only species transferred across the interface.
This assumption is widely accepted in the literature (deMontigny
et al., 2006 ).
•Dependence of densities, viscosities, diffusivities and enthalpies
with the temperature and composition are considered (Greer,
2008 ).
•Dependence of transfer coefficient in liquid and vapor phases
with the viscosity, density, nominal packing size and specific dry
area and effective interfacial area for mass transfer are taken into
account through Onda’s correlations (Onda et al., 1968 ).
•An upper bound (1.225 kPa/m of packing) for the maximum
allowable pressure drop is considered (Kister, 1992 ). Thus, the
total pressure drop may be lower than the upper bound.
•Lower and upper bounds for the superficial gas velocity are
also considered to avoid flooding problem and a bad gas–liquid
distribution. Values suggested in literature range from 70
to 80% of the flooding velocity (Kister, 1992; Seider et al.,
2009 ).
•It is assumed that absorber diameter should be ten times greater
than the nominal diameter of packing (Seider et al., 2009 ). A max-
imum absorber diameter is adopted (13.0 m). This upper bound
is suggested in the literature (Chapel et al., 1999 ).
•Six-tenths rule is used to compute the purchased equipment cost.•Direct,
indirect costs and working capital are considered and
calculated as percentages of the purchase and installation cost
(Abu-Zahra et al., 2007b ).
•Chemical reactions take place at the liquid and vapor interface.
The following reactions are considered:
2H 2O ↔ H3O++ OH−(R1)
2H 2O + CO 2↔ H3O++ HCO 3−(R2)
H2O + HCO 3−↔ H3O++ CO 32−(R3)
H2O + MEAH+↔ H3O++ MEA (R4)
MEACOO−+ H2O ↔ MEA + HCO 3−(R5)
MEA + CO 2+ H2O ↔ MEACOO−+ H3O+(R6)
CO 2+ OH−↔ HCO 3−(R7)
As mentioned earlier, the complete and detailed mathematical
model used in this paper have been previously presented in Mores
et al. (2011, 2012) . For this reason, the cost mathematical model is
only presented in the next section.
4.1. Cost mathematical model
4.1.1. Objective function
The objective function to be minimized is the total annual cost
[TAC]. It is computed as the sum of the annualized capital cost [ACC]
and the operating and maintenance cost [TO&MC]:
TAC = ACC + TO&MC (1)
4.1.2. Annualized capital cost [ACC]
The annualized capital cost is defined as the ratio between the
total capital investment [TCI] cost and the capital recovery factor
[CRF].
ACC= TCI CRF (2)
The total capital investment [TCI] depends on the direct cost
[DC], indirect cost [IC] and fixed capital investment [FCI]. Thus, TCI
is computed by Eq. (3):
TCI = DC + IC + FCI (3)
The direct cost DC is given by:
DC = KDCTPC (4)
where TPC refers to the total purchased equipment cost which is
computed by Eq. (10) . The constant factor KDC(2.688) is taken from
Abu-Zahra et al. (2007b) and includes the purchase and installation,
instrumentation and control, piping, electrical installation, build-
ing and building services, yard improvements, service facilities and
land. Table 1 lists the values assumed for each one of the items used
to compute KDC.
The indirect cost IC is given by:
IC = KICTPC (5)
where the constant factor KIC(1.0080) is also taken from Abu-Zahra
et al. (2007b) and includes the engineering, construction expenses,
contractors fees and contingency (Table 1).
The fixed capital investment [FCI] includes the working capital
[WC] and start up cost [SC] which are computed as follows:
FCI = WC + SC (6)
where WC and SC are computed as follows:
WC= KWCTPC (7)
SC= KSCTPC (8)

P. Mores et al. / International Journal of Greenhouse Gas Control 10 (2012) 148–163 153
Table
1
Parameter
values assumed to compute costs.
Percentage of total
purchased capital [TPC,
Eq.
(10) ]
Direct costs (DC), KDC 2.6880
1 Purchased
equipment (CP) 1.0000
2
Equipment installation 0.5280
3
Instrumentation and control 0.2000
4
Piping 0.4000
5 Electrical
installation 0.1100
6 Building
and building services 0.1000
7 Yard
improvements 0.1000
8 Services facilities 0.2000
9
Land 0.0500
Indirect
costs (IC), KIC 1.0080
10
Engineering 0.2688
11
Construction expenses 0.2688
12 Contractor’s
fee 0.0134
13
Contingency 0.4570
Fixed
capital investment (FCI) 1.2936
14
Working capital, KWC 0.9240
15 Start-up
cost plus MEA make-up cost, KSC 0.3696
Total capital investment (TCI) 4.9896
Taken from Abu-Zahra et al. (2007b) .
As indicated in Table 1, numerical values for KWC and KSCare,
respectively, 0.9249 and 0.3696 (Abu-Zahra et al., 2007b ).
The purchased equipment cost [PC] of the piece of equipment
“k” is computed using the six-tenths rule (m = 0.6) as follows:
PCk= PC0
k/parenleftbigg
X1
k
X0
k/parenrightbiggm/parenleftbigg
I1
k
I0
k/parenrightbigg
(9)
where
X and I are the capacity and cost index of the equipment
k (absorber, stripper, reboiler, condenser, heaters, blower, storage
tanks, pumps and compressors). The superscripts 0 and 1 refer to
the base and current values, respectively. The base cost of equip-
ment [PC0
k] depends on the X0and is computed by correlations
reported in Henao (2005) and Seider et al. (2009) and the basevalues
used for X0and I0in Eq. (9) are listed in Table 1a. Table 2 lists
the main investment items considered including the construction
material of each one of them.
Then, the following constraint is used to compute the total pur-
chased equipment cost (TPC):
TPC=/summationdisplay
kPCk, k = ABS, REG, BLOW , RAP, LAC, ECO, MEA-TK ,
H2O-TK , COND , REB, COMP , I-COOL (10)
The capital recovery factor is computed by Eq. (11)
CRF =i(1 + i)n
(1 + i)n− 1(11)
where n and i are, respectively, the compounding periods (eco-
nomic life of equipment) and the interest rate. It is assumed a
project life of 25 years and 8% of interest rate.
4.1.3. Total operating and maintenance cost (TO&MC)
The total operating and maintenance cost [TO&MC] is computed
as follows:
TO&MC = ı1TPC + TU&MUC + TOLC (12)
As shown, it depends on the following cost-items: (a) TPC (total
purchased equipment cost), (b) TU&MUC (heating and cooling
medium costs, electricity, MEA and water make-up and R&D costs)
and (c) TOLC (total operating labor cost). Factor ı1(0.3863) includes
the following items: (a) local taxes, (b) insurances, (c) maintenance,
(d) operating supplies, (e) plant overhead cost and (f) R&D cost. The
contributions of each one of the mentioned items to the factor ı1
are listed in Table 3 and were taken from Abu-Zahra et al. (2007b) .
4.1.3.1. Utility costs including electricity cost and MEA and H2O make-
up costs (TU&MUC). The utility cost “i” [UC i] is computed by Eq.
(13) where Y and P are the annual consumption computed from
the mass and energy balances and the specific price of the utility
Table 1a
Base
values used to compute PC0.
Piece of equipment X0I0Reference
Absorber/stripper packing: [ABS] P, [REG] P 182 m3
381.7 (1996) Henao, 2005Absorber/stripper vessel: [ABS] V, [REG] V 234 m2
Reboiler [REB] 100 m2
Heater [ECO] 900 m2
Coolers and condensers: [LAC], [COND], [I-COOL] 900 m2
Pump [RAP] 250 kW
Blower
[BLOW] 745 kW
499.6 (2006) Seider et al., 2009 Storage tank: [MEA-TK], [H 2O-TK] 100 m3
Compressor [COMP] 1400 kW
Table 2
Pieces
of equipments considered to compute the total capital investment including their construction material.
Equipment type Material of construction Capacity (X)
AbsorberVessel CS Superficial area
Packing–Intalox
Saddles, 5 in. Ceramic Packing volume
StripperVessel
CS Superficial area
Packing–Intalox
Saddles, 5 in. Ceramic Packing volume
Compressor
interstage coolers
Floating headSS (tubes)–CS (shell)
Exchange areaCondenser
Lean
amine cooler
Rich/lean
amine exchangerSS (tubes)–SS (shell)Reboiler Horizontal Kettle
Amine
storage tank Roof tankCS Volumetric capacityWater storage tank Roof tank
Flue
gas blower and driver Centrifugal (turbo)
SS Brake horsepower Compressor and drivers Centrifugal
Rich
amine pump and driver Centrifugal

154 P. Mores et al. / International Journal of Greenhouse Gas Control 10 (2012) 148–163
Table 3
Parameter
values used to compute the total operating and maintenance costs.
Percentage of total
purchased capital [TPC,
Eq.
(10) ]
ı1 0.3863
Local
taxes 0.0739
Insurance
0.0370
Maintenance
0.1478
Operating
supplies 0.0222
Plant
overhead cost 0.0887
R&D
cost 0.0185
Percentage
of utility
and
make-up costs
[U&MUC,
Eq. (14) ]
ı2 1.0500
Raw
material and utilities 1.0000
R&D
cost 0.0500
Percentage
of total
operating
labor cost
[OLC,
Eq. (16) ]
ı3 2.4466
Operating
labor 1.0000
Supervision
and support labor 0.3000
Laboratory
charges 0.1000
Administrative
cost 0.1500
Plant
overhead cost 0.7800
R&D
cost 0.1165
Taken from Abu-Zahra et al. (2007b) .
‘i’ (electricity, cooling water, low pressure steam, water make-up,
MEA make-up and MEA inhibitors) respectively.
UCi= YiP0
i/parenleftbigg
I1
i
I0
i/parenrightbigg
(13)
The total utility cost [U&MUC] is then computed as follows:
U&MUC =/summationdisplay
iUCi (14)
The following cost-item, hereafter named [TU&MUC], takes into
account the total utility cost [U&MUC] and R&D cost. Then, it is
computed as:
TU&MUC = ı2U&MUC (15)
The numerical value used for ı2and the corresponding reference
is detailed in Table 3.
Table 4
Stream
specifications assumed for optimization.
Flue gas stream
Gas
flow-rate [mol/s] 10,000
Inlet
temperature [K] 323.15
CO 2[mol fraction] 0.0422
H2O [mol fraction] 0.0845
O2[mol fraction] 0.1166
N2[mol fraction] 0.7567
Pressure
[kPa] 101.3
MEA
make-up stream
MEA
[mass fraction] 0.3
Temperature
[K] 298.15
H2O make-up stream
Temperature
[K] 298.15
Cooling
water stream
Inlet
cooling water temperature [K] 298.15
Outlet
cooling water temperature [K] 313.15
Reboiler
pressure [kPa] 202.6
Compression
pressure [kPa] 8600Table 5
Parameter
values assumed for the main pieces of equipment.
Heat exchangers
Minimum
temperature difference in cold and heat
side
[K]1
Global energy transfer coefficient [KW/m2K]
Condenser
[U COND ] 0.3202
Interstage
coolers [U I-COOL ] 0.2777
Lean
amine cooler [U LAC] 1.0050
Reboiler
[U REB] 1.3603
Economizer [U ECO] 0.7608
Pumps,
blower and compressor efficiency [%] 75
Hold
up MEA make-up tank [days] 1
Hold
up H2O make-up tank [days] 30
Absorber
and regenerator units
Column
type Packed
Stages
number 10
Packing
specifications
Type
of packed Ceramic Intalox Saddles
Specific
area [m2/m3] 118
Nominal
packing size [m] 0.05
Critical
surface tension [N/m] 0.061
Void
fraction 0.79
Dry
packing factor [m2/m3] 121.4
Taken from Fisher et al. (2005) , Henao (2005) , United Technologies Research Center
(1999) ,
Perry and Green (1997) and Blauwhoff et al. (1985) .
4.1.3.2. Total operating labor cost (TOLC). The operating labor cost
(OLC) is computed as follows:
OLC=/FSLS
LYopSop (16)
where Sopis the annual operator salary (US$/year). /FS refers to the
number of operating shift per year (1000 shift/year; 8000 operating
hours/year and 3 shifts/day have been considered). LSrefers to the
number of operators per shift (3.5). LYop is the amount of shifts to be
handled by one operator per year; in this paper 245 shift-op/year
is assumed. The total operating and labor cost (TOLC) includes the
following items: (a) supervision and support labor, (b) laboratory
charges, (c) administrative cost, (d) plant overhead cost and (f) R&D.
Each one of these items is computed as a percentage of the oper-
ating labor cost (OLC) as indicated in Table 3. Thus, total operating
and labor cost is given by:
TOLC = ı3OLC (17)
Thus, Eqs. (1)–(17) are basically the main constraints used
to compute investment and operating cost. In addition, several
“intermediate” variables and equations were also defined in order
to facilitate the model convergence. The complete mathemati-
cal model (physical-thermodynamic and cost models) involves
approximately 3200 variables and constraints. It was implemented
in General Algebraic Modeling System (GAMS; Brooke et al., 1996 ).
The generalized reduced gradient algorithm CONOPT 2.041 was
here used as NLP solver (Drud, 1992 ).
5. Results. Application of the NLP model
As mentioned earlier the models used in this paper to
describe the L–V equilibrium of the CO 2–MEA–H 2O system and
Table 6
Specific
costs of heating and cooling utilities, electricity.
Utility Specific prices [Pi]
used
in Eq. (7)
H2O make-up (Henao, 2005 ) 0.04 US$/t
MEA
make-up (Rao and Rubin, 2002 ) 1250 US$/t
Cooling
water 0.0329 US$/t
Steam
(Henao, 2005 ) 3.66 US$/GJ
Electricity
(Henao, 2005 ) 0.06 US$/kWh

P. Mores et al. / International Journal of Greenhouse Gas Control 10 (2012) 148–163 155
Table 7
Distribution
of the total operating and maintenance cost [TO&MC] for CO 2removal = 80 and 95%.
CO 2removal = 80%; TO&MC = 35.06
million
US$/yearCO 2removal = 95%; TO&MC = 42.19
million
US$/year
Heating and cooling utility costs, electricity cost, MEA and
H2O make-up costs, R&D cost [TU&MUC, %]73.36 74.89
Total purchased equipment cost [ı1TPC, %] 26.34 24.98
Total
operating labor cost [TOLC, %] 1.12 0.93
95 90 85 80 75 704042444648505254[million US$/y]Total operating cost
[TO&MC]
Total investment
[ACC]Total annual cost
[TAC]
CO2 removal target [%]10152025303540
[million US$/y]
Fig. 2. Total annual cost including investment capital operating cost vs. CO 2removal
target.
the absorber and regenerator units have been successfully ver-
ified with experimental data in previous publications (Mores
et al., 2011, 2012 ). In theses articles the models have been
also used to optimize the whole CO 2capture process from a
thermodynamic point of view. For this reason, the comparison
between predicted values by the proposed model (simula-
tion mode) with experimental data is not presented in this
paper.
According to Section 3, the goal of this paper is to simultaneously
optimize the operating conditions and dimensions of the pieces
of equipments in order to meet different CO 2removal targets at
minimum total costs.
As follows, the numerical results corresponding to the opti-
mal operating conditions of each one of the pieces of equipments
including their dimensions and heating/cooling utilities are pre-
sented and discussed separately in terms of CO 2removal targets.
Thus, the optimal solution set corresponding to each one of
95 90 85 80 75 700.7520.7540.7560.7580.7600.7620.7640.7660.7680.7700.772Total operating cost / total annual cost
[TO&MC]/[TAC]
Total investment / total annual cost
[ACC]/[TAC]
CO2 removal target [%]0.2280.2300.2320.2340.2360.2380.2400.2420.2440.2460.248
Fig. 3. Percentage of the investment capital operating cost vs. CO 2removal target.11 10 9 8 7 6 5 4 3 2 1 00.200.220.240.260.280.300.320.340.360.380.40α – 80 % of CO2 absorption
α – 95 % of CO2 absorption T – 80 % of CO2 absorption
T – 95 % of CO2 absorptionCO2 loading [CO2 mole / MEA mole]
Outlet temperature [K]
Stage from the bottom of the absorber322324326328330332334
Fig. 4. CO 2loading factor and temperature profiles along the absorber (80 and 95%
of
CO 2removed).
the fixed parameters is directly obtained from Figs. 2–5 and
Tables 7–14 .
Optimization problems have been solved using Intel Core 2 Quad
Extreme QX9650 3 GHz 1333 MHz processor and 4 GB RAM and the
parameters listed from Tables 4–6 . The data listed in Table 6 were
up-dated to 2011.
5.1.
Optimal total cost values vs. CO2removal target
Figs. 2 and 3 show the influence of the CO 2removal target on
the total annual cost and how it is distributed in total investment
and operating costs.
11 109 8 7 6 5 4 3 2 1 0 -10.180.200.220.240.260.280.300.320.340.360.380.40T – 80 % of CO2absorption
T – 95 % of CO2 absorption
Outlet temperature [K]CO2 loading [CO2 mole / MEA mole]
Stage from the bottom of the stripper378380382384386388390392394396 α – 80 % of CO2 absorption
α – 95 % of CO2 absorption

Fig. 5. CO 2loading factor and temperature profiles along the stripper (80 and 95%
of
CO 2removed).

156 P. Mores et al. / International Journal of Greenhouse Gas Control 10 (2012) 148–163
Table
8
Distribution
of the total utility costs [U&MUC] for CO 2removal target = 80 and 95%.
CO 2removal = 80%;
U&MUC
= 24.38 million
US$/yearCO 2removal = 95%;
U&MUC
= 29.95 million
US$/year
Steam [%] 57.36 58.05
Electricity
[%] 28.64 27.67
Cooling
water [%] 7.76 8.24
MEA
make-up [%] 5.87 5.67
H2O make-up [%] 0.38 0.36
5.2. Contribution of each one of the operating costs on the total
operating cost (TO&MC)
Tables 7 and 8 show how the total operating cost is influenced by
the CO 2removal target for CO 2removal target = 80 and 95%. Table 9
lists the operating costs of all cost-items for different CO 2removal
targets (70.0, 75.0, 80.0, 85.0, 87.5, 90.0, 92.5 and 95.0%).
5.3. Contribution of each one of the pieces of equipments on the
total annual capital cost (ACC)
Tables 10 and 11 show the contributions of the investments of
each one of the pieces of equipments on the total annual capital
cost for different CO 2removal targets (70.0, 75.0, 80.0, 85.0, 87.5,
90.0, 92.5 and 95.0%).
5.4. Optimal dimensions and operating conditions of each one of
the pieces of equipment
Tables 12 and 13 compare the optimal dimensions and operating
conditions of each one of the pieces of equipment for each one of
the CO 2removal targets.5.5.
Influence of the CO2removal target on the profiles of
temperature and amount of captured CO2along the height of
absorber and regenerator units
As mentioned in Section 1, the absorber and regenerator mod-
els are modeled by stages in order to obtain the optimal internal
profiles of composition, temperature and flow-rate. Figs. 4 and 5
compare the CO 2loading factor and temperature profiles along of
absorber and stripper for 80 and 95% of CO 2removal levels.
5.6.
Comparison of optimal designs involving different objective
functions
By exploiting the benefits of the proposed model, different opti-
mal designs obtained by considering different objective functions
are compared. Table 14 compares, the optimal solutions obtained
for the following three objective functions: (a) total annual cost
(min.), (b) total heat duty (min.) and (c) total heat duty/CO 2
recovery (min.) where the CO 2recovery is here considered as an
optimization variable.
5.7. Comparison between the output results predicted by the
model and a solution reported by other authors
In order to verify the proposed model, Tables 15–18 compare
the output results obtained by applying the proposed model with
other reported costs (investment and operating costs).
6. Discussion of results
6.1. Optimal total cost values
Fig. 2 illustrates the total optimal costs [TAC] in terms of the
CO 2removal targets. It can be observed that the total cost [million
Table 9
Optimal
values of TAC and TO&MC obtained for different CO 2removal targets.
Cost-item CO 2removal target (%)
70.0 75.0 80.0 85.0 87.5 90.0 92.5 95.0
Total annual cost [TAC, million US$/year] 42.1379 44.0565 46.1367 48.3788 49.5813 50.8814 52.4142 54.8305
Total operating and maintenance cost [TO&MC, million US$/year] 31.7724 33.3652 35.0640 36.9300 37.9323 38.9981 40.2918 42.1939
(1) Total utility and amine cost [TU&MUC, million US$/year] 22.9965 24.3287 25.7214 27.2867 28.1289 29.0068 30.1096 31.5975
Utility
and amine make-up cost [U&MUC, million US$/year] 21.7977 23.0604 24.3805 25.8642 26.6625 27.4946 28.5399 29.9502
Steam [UC Steam ] 12.9064 13.4371 13.9838 14.6868 15.0614 15.4232 16.0143 17.3865
Electricity
[UC Electricity ] 5.7771 6.3716 6.9829 7.5712 7.8945 8.2588 8.5441 8.2864
Cooling
water [UC Cwater ] 1.7815 1.8238 1.8908 1.9880 2.0409 2.0993 2.2205 2.4688
MEA
make-up [UC MEA mk] 1.2521 1.3416 1.4310 1.5205 1.5652 1.6099 1.6546 1.6993
H2O make-up [UC H2O mk] 0.0804 0.0862 0.0919 0.0977 0.1006 0.1034 0.1063 0.1092
(2)
Total purchase equipment cost [ı1TPC, million US$/year] 8.6471 8.9188 9.2370 9.5507 9.7177 9.9132 10.1127 10.5416
(3)
Total operating labor cost [TOLC, ı3OLC, million US$/year] 0.3913 0.3913 0.3913 0.3913 0.3913 0.3913 0.3913 0.3913
Table 10
Distribution
of the TPC for different CO 2removal targets.
CO 2removal target = 80%
(TPC
= 23.69 million US$)CO 2removal target = 85%
(TPC
= 24.49 million US$)CO 2removal target = 95%
(TPC
= 27.03 million US$)
Compressors [%] 39.57 41.17 38.97
Absorber
[%] 16.48 17.82 20.26
Lean-rich
amine heat exchanger [%] 17.00 16.15 12.87
Reboiler
[%] 7.00 7.13 7.41
Blower
[%] 5.94 6.57 6.50
Condenser
[%] 4.10 4.30 4.39
Desorber
[%] 3.93 3.94 4.10
Lean
amine cooler [%] 2.64 2.55 1.96
H2O tank [%] 1.14 1.24 1.30
Interstage
coolers [%] 1.19 1.23 1.17
MEA tank [%] 0.73 0.76 0.82
Pumps
[%] 0.30 0.29 0.25

P. Mores et al. / International Journal of Greenhouse Gas Control 10 (2012) 148–163 157
Table
11
Optimal
values of total direct and indirect costs and fixed capital investment for different CO 2removal targets.
Cost-item CO 2removal target [%]
70.0 75.0 80.0 85.0 87.5 90.0 92.5 95.0
Total annual cost [TAC, million US$/year] 42.1379 44.0565 46.1367 48.3788 49.5813 50.8814 52.4142 54.8305
Annualized
capital cost [ACC, million US$/year] 10.3655 10.6913 11.0727 11.4487 11.6489 11.8832 12.1225 12.6366
Capital
recovery factor [CRF]a0.093676 0.093676 0.093676 0.093676 0.093676 0.093676 0.093676 0.093676
Total
purchased equipment cost [TPC, million US$] 22.1760 22.8729 23.6890 24.4935 24.9218 25.4231 25.9349 27.0347
Compressors [COMP] 8.7749 9.1295 9.4841 9.8340 10.0057 10.1769 10.3526 10.5365
Absorber [ABS] 3.6543 3.9528 4.2668 4.5568 4.7282 4.9392 5.0739 5.4761
Lean-rich
amine exchanger [ECO] 3.7691 3.5817 3.4711 3.3247 3.2326 3.1632 3.1178 3.4800
Reboiler
[REB] 1.5517 1.5805 1.6225 1.7020 1.7479 1.7968 1.8885 2.0035
Blower
[BLOW] 1.3171 1.4564 1.5911 1.7089 1.7763 1.8563 1.9049 1.7566
Condenser
[COND] 0.9098 0.9525 0.9942 1.0424 1.0680 1.0936 1.1352 1.1866
Desorber [REG] 0.8708 0.8734 0.8904 0.9289 0.9531 0.9732 1.0167 1.1083
Lean amine cooler [LAC] 0.5851 0.5648 0.5486 0.5287 0.5169 0.5059 0.4957 0.5310
H2O tank [H 2O-TK] 0.2518 0.2751 0.2960 0.3201 0.3332 0.3454 0.3584 0.3515
Interstage
coolers [I-COOL] 0.2634 0.2722 0.2820 0.2921 0.2971 0.3021 0.3080 0.3153
MEA
tank [MEA-TK] 0.1622 0.1687 0.1772 0.1906 0.1985 0.2066 0.2194 0.2210
Pumps
[RAP] 0.0658 0.0654 0.0649 0.0645 0.0642 0.0639 0.0637 0.0683
Direct
cost [DC, million US$] 59.6092 61.4824 63.6760 65.8385 66.9897 68.3372 69.7130 72.6693
Indirect
cost [IC, millionUS$] 22.3534 23.0559 23.8785 24.6894 25.1211 25.6265 26.1424 27.2510
Fixed
capital investment [FCI, million US$] 28.6878 29.5893 30.6450 31.6857 32.2397 32.8883 33.5504 34.9732
Total Working Capital [WC] 20.4913 21.1352 21.8893 22.6326 23.0284 23.4916 23.9646 24.9808
Start-up
and initial solvent costs [SC] 8.1965 8.4541 8.7557 9.0531 9.2114 9.3966 9.5858 9.9923
aConstant value.
US$/year] varies linearly with the CO 2removal targets from 70.0
to 80.0% of CO 2removal target and then from 80.0% the total cost
increases a little more rapidly as the CO 2removal target increases.
Certainly, the total cost increases in 4.7 and 7.8% when the CO 2
removal target increases, respectively, from 75.0 to 80.0% and from
90.0 to 95.0%.
Fig. 3 shows how the total cost is distributed in operating cost
[TO&MC] and investment [ACC] for each one of the CO 2removal
targets. The reported results clearly indicate that the total cost
distribution in operating cost and investment is not significantly
affected by the CO 2recovery target. It can be seen that the ratio
of [TO&MC]/[TAC] slowly decreases as the CO 2removal target
increases. Naturally, the ratio of [ACC]/[TAC] increases at the same
rate that [TO&MC]/[TAC] decreases.
The average contributions of the operating cost and investment
on the total cost for all CO 2recovery targets are 76.2 and 23.8%
respectively.6.2.
Contribution of each one of the operating costs on the total
operating cost (TO&MC)
Table 7 illustrates the total contributions of each one of the
cost-items (TU&MUC, purchased equipment and TOLC) on the total
operating cost (TO&MC) for two CO 2removal targets (80 and 95%).
In addition, Table 8 shows how the total utility costs reported in
Table 7 are distributed in heating and cooling utilities, total electric-
ity consumed by pumps, blower and compressors and MEA and H2O
make-up costs. Table 9 presents a detailed description of the cost-
items for the all CO 2removal targets considered in Figs. 2 and 3.
As expected, it can be seen from Table 7 that, independently of
CO 2recoveries, the total operating and maintenance cost [TO&MC]
is strongly influenced by [TU&MUC]. More specifically, it accounts
for more than 73.0% of the [TO&MC]. The contributions of the total
purchased equipment cost [ı1TPC] and the total labor cost (TOLC)
are, respectively, 26.0 and 1.0%.
Table 12
Optimal
dimensions and operating conditions of absorption and regeneration units for different CO 2removal targets.
CO 2removal target
70.0% 75.0% 80.0% 85.0% 87.5% 90.0% 92.5% 95.0%
Absorber [ABS]
Packing
height [HABS, m] 12.79 15.16 17.63 19.91 21.27 22.94 23.98 25.18
Diameter
[DABS, m] 12.14 12.13 12.13 12.13 12.13 12.13 12.13 12.68
Total
packing volume 1479.82 1752.01 2037.72 2301.62 2457.81 2650.24 2773.58 3181.13
Solvent
inlet flow-rate [mol/s] 20,609.5 20,606.4 20,612.8 20,630.8 20,636.9 20,639.1 20,663.5 22,719.6
Solvent
inlet temperature [TLC, K] 315.47 316.48 317.36 318.57 319.34 320.09 320.85 320.48
Solvent
outlet temperature [TRC, K] 322.44 322.90 323.42 324.03 324.41 324.86 325.37 325.74
Inlet
gas temperature [TGin, K]a323.15 323.15 323.15 323.15 323.15 323.15 323.15 323.15
Outlet
gas temperature [TGout, K] 324.53 325.40 326.18 327.07 327.56 328.01 328.47 328.22
Pressure
drop [/Delta1PABS, Pa/m] 466.46 467.24 467.31 467.77 468.20 468.39 468.44 387.81
Inlet
CO 2loading [˛LC] 0.2616 0.2610 0.2551 0.2413 0.2332 0.2246 0.2062 0.1989
Regenerator
[REG]
Packing
height [HREG, m] 4.91 4.96 5.21 5.49 5.64 5.71 5.84 6.31
Diameter [DREG, m] 4.45 4.45 4.47 4.59 4.67 4.74 4.91 5.20
Total
packing volume 76.54 77.40 82.23 91.20 96.84 101.22 110.83 134.52
Inlet
liquid temperature [TRH, K] 383.29 382.77 382.74 382.89 382.91 383.01 383.52 384.53
Outlet
liquid temperature [TREB, K] 391.42 391.45 391.79 392.54 392.94 393.34 394.10 394.37
Outlet
gas temperature [TC, K] 322.40 320.48 319.71 319.44 319.29 319.31 319.95 321.56
Pressure
drop [/Delta1PREG, Pa/m] 119.10 125.72 137.12 149.60 155.09 161.60 175.53 179.30
Inlet
CO 2loading [˛RC] 0.3883 0.3972 0.4007 0.3964 0.3931 0.3893 0.3756 0.3566
aFixed value.

158 P. Mores et al. / International Journal of Greenhouse Gas Control 10 (2012) 148–163
Table 13
Optimal
dimensions and operating conditions of reboiler, condenser, heat exchanger units, compressors and pumps for different CO 2removal targets.
CO 2removal target
70.0% 75.0% 80.0% 85.0% 87.5% 90.0% 92.5% 95.0%
Reboiler [REB]
Temperature
[TR, K] 391.42 391.45 391.79 392.54 392.94 393.34 394.10 394.37
Area
[AREB, m2] 2303.79 2375.53 2481.65 2687.62 2809.38 2941.81 3196.23 3526.97
Heat
load [QR, kJ/s] 79,768.47 83,048.54 86,427.39 90,771.89 93,087.24 95,323.25 98,976.77 107,457.61
Driven force [DMLT R] 25.45 25.70 25.60 24.83 24.36 23.82 22.76 22.40
Condenser [COND]
Liquid
flow-rate to stripper [mol/s] 404.402 404.604 422.717 463.317 486.937 514.767 590.019 713.834
Temperature
[TCOND , K] 322.397 320.481 319.708 319.436 319.292 319.307 319.949 321.564
Heat
transfer area [ACOND , m2] 3217.010 3472.604 3729.924 4035.886 4202.738 4371.590 4652.617 5008.980
Heat
load [QCOND , kJ/s] 43,568.1 45,117.3 47,645.2 51,440.4 535,266 558,431 60,797.7 68,392.5
Driven force [K] 42.296 40.576 39.893 39.806 39.776 39.894 40.810 42.642
Lean-rich amine exchanger [ECO]
Driven
force [/Delta1tml, K] 8.098 8.655 9.025 9.616 10.007 10.306 10.562 9.812
Heat
transfer area [AECO, m2] 18,545.50 17,034.89 16,167.27 15,046.27 14,358.65 13,848.36 13,518.53 16,236.34
Heat
load [QECO, kJ/s] 114,252.7 112,165.6 111,008.9 110,072.6 109,317.7 108,581.8 108,631.0 121,203.6
Lean
amine cooler [LAC]
Driven
force [/Delta1tml, K] 17.34 18.36 19.24 20.44 21.21 21.96 22.73 22.36
Heat
transfer area [ALAC, m2] 1541.54 1453.33 1384.75 1301.96 1253.70 1209.74 1169.18 1311.36
Heat
load [QLAC, kJ/s] 26,865.8 26,814.2 26,775.6 26,746.5 26,727.1 26,704.4 26,707.6 29,463.9
Interstage coolers [I-COOL]
Average
driven force [/Delta1tml, K] 56.76 56.57 56.50 56.48 56.47 56.47 56.54 56.71
Total heat transfer area [AI-COOL , m2] 407.69 430.52 456.67 484.30 498.06 512.37 529.11 550.07
Total
heat load [QT I-COOL , kJ/s] 6439.5 6767.0 7169.2 7599.3 7813.0 8038.0 8313.3 8675.5
Compressors
(4) [COMP]
Total
compression power [PT COMP , kW] 4542.51 4852.33 5170.35 5492.13 5652.89 5814.97 5983.34 6161.79
Pump
[RAP]
Power
[PRAP, kW] 57.33 56.62 55.96 55.36 54.98 54.51 54.29 60.91
Blower
[BLOW]
Power
[PBLOW , kW] 2065.01 2441.71 2829.57 3187.10 3399.65 3658.37 3819.37 3336.97
Total
electricity [kW] 6664.84 7350.65 8055.88 8734.58 9107.52 9527.86 9857.01 9559.67
Table 8 clearly shows that, independently of CO 2recoveries, the
total utility cost [U&MUC] is strongly influenced by the steam used
in the reboiler [UC Steam ] and by the total electricity consumed by
pumps, blower and compressors [UC Electricity ] and the correspond-
ing contributions are about 57.7 and 28.15%, respectively. The costs
of MEA make-up and cooling medium (water) represent approx.
13.5%. The cost of H2O make-up is insignificant (0.38%).
As illustrated in Table 9, the ranking order of the cost distribution
does not vary with the CO 2removal target.
6.3. Contribution of each one of the pieces of equipments (PCj) on
the total purchased equipment cost (TPC)
Tables 10 and 11 clearly show the contributions of the invest-
ments of each one of the pieces of equipments on the total
equipment purchased cost (TPC). As expected, TPC is significantly
influenced by the compressors, the absorption unit and the lean-
rich amine exchanger [ECO]. Indeed, these pieces of equipments
represent more than 73.00% of the total capital investment. As is
also shown in Table 10, the contributions of the blower (BLOW)
and reboiler (REB) for CO 2removal target = 85.0% are quite similar
(approx. 7.0–8.0% each) and thus the ranking order may change as
the CO 2removal target and cost parameters vary. This also applies
for the following groups of pieces of equipments: (a) [ABS] and
[ECO]; (b) [COND] and [REG] and (c) [I-COOL] and [H 2O-TK].
A detailed comparison of the distribution of the each one of the
cost-items on the TPC for the different CO 2capture levels is shown
in Table 11. The comparison shows that the ranking order from
CO 2removal target = 75.0–80% varies with respect to 70%. In addi-
tion, the ranking order from CO 2removal target = 85.0–92.5% varies
with respect to 95.0%. Certainly, from 75.0 to 80.0% of CO 2removal
the contribution of the water make-up tank [H 2O-TK] is greater
than the inter-stage coolers [I-COOL], in contrast to that observed
for 70.0%. Then, from CO 2removal target = 85.0–92.5% the order
between the blower [BLOW] and reboiler [REB] is also modified;the
contribution of [BLOWER] becomes progressively more impor-
tant than [REB], in contrast to that observed for 70.0% to 80.0% and
95.0%. Thus, it is possible to conclude that the contribution per-
centage of each one of pieces of equipments is slightly influenced
by the CO 2removal target but the effect is strong enough to vary
the corresponding ranking order.
6.4. Optimal dimensions and operating conditions of each one of
the pieces of equipment
Tables 12 and 13 compare the heat transferred driving force
and dimensions of the main process-units to reach different CO 2
removal targets to minimize the total annual cost. The results listed
in both tables correspond to those optimized results analyzed in the
previous sections.
From the reported results, the following conclusions can be
drawn:
– The volume of the packing material of the absorber for CO 2
removal target 95.0% (3181.13 m3) is 115.0% greater than that
required for 70.0%. Despite that the absorber diameter is an opti-
mization variable, it remains almost constant as the CO 2removal
target increases (12.2 m). Thus, the absorber height and therefore
the packing material volume considerably increase as the CO 2
removal target increases.
– The outlet gas temperature and the solvent inlet and outlet tem-
peratures in the absorber slightly vary with the CO 2removal
targets.
– In order to meet the CO 2emission target, the inlet CO 2load-
ing in the absorber slightly decreases as the CO 2removal target
increases.
– As expected, the higher CO 2removal targets, the higher amount
of CO 2removed from the flue-gas stream and therefore the higher
values of CO 2loading leaving the absorber (˛RC).

P. Mores et al. / International Journal of Greenhouse Gas Control 10 (2012) 148–163 159
Table
14
Optimal
solutions for different optimization problems.
P1. Minimizing TAC
(CO 2removal = 85%)P2. Minimizing the
total
heat duty (CO 2
removal = 85%)P3. Minimizing the total
heat
duty/CO 2recovery
(CO 2removal “free”)
CO 2removal % 56,818.08 56,818.08 59,529.44
Amount of CO 2captured [kg/h] 48.379 49.393 51.029
TAC [million US$/year] 10.366 11.533 11.915
TCI [million US$/year] 31.772 37.860 39.114
TO&MC [million US$/year] 56,818.08 56,818.08 59,529.44
Total purchased equipment cost [TPC, million US$]
Compressors
[COMP] 9.834 10.192 10.485
Absorber [ABS] 4.557 4.908 5.052
Desorber [REG] 0.929 0.896 1.229
Reboiler
[REB] 1.702 1.640 1.708
Lean-rich
amine exchanger [ECO] 3.325 2.874 2.817
Blower
[BLOW] 1.709 1.951 1.910
Condenser [COND] 1.042 0.760 0.804
Interstage
coolers [I-COOL] 0.292 0.345 0.355
Lean
amine cooler [LAC] 0.529 0.494 0.484
H2O Tank [H 2O-TK] 0.320 0.362 0.379
MEA tank [MEA-TK] 0.191 0.189 0.205
Total
utility cost [U&MUC, million US$/year]
Cooling
water [UC water ] 1.988 1.856 1.975
H2O make-up [UC H2O mk] 0.098 0.098 0.102
MEA
make-up [UC MEA mk] 1.520 1.520 1.593
Steam
[UC Steam ] 14.687 14.649 15.225
Electricity
[UC Electricity ] 7.571 8.560 8.684
AbsorberPacking height [HABS, m] 19.91 23.34 23.92
Diameter [DABS, m] 12.13 11.94 12.10
Total packing volume [m3] 2301.62 2614.08 2752.40
Solvent inlet temperature [TLC, K] 318.57 335.88 336.71
Solvent outlet temperature [TRC, K] 324.03 324.43 324.78
Outlet gas temperature [TGout, K] 323.15 327.83 328.47
Pressure drop [/Delta1PABS, Pa/m] 467.77 501.93 471.81
Inlet CO 2loading [˛LC] 0.241 0.260 0.238
DesorberPacking height [HREG, m] 5.49 5.07 9.71
Diameter
[DREG, m] 4.59 4.53 4.93
Total packing volume [m3] 91.20 82.00 185.43
Inlet liquid temperature [TRH, K] 382.89 380.02 380.73
Outlet liquid temperature [TREB, K] 392.54 391.52 392.71
Pressure
drop [/Delta1PREG, Pa/m] 149.60 127.09 107.83
Inlet CO 2loading [˛RC] 0.396 0.416 0.402
Reboiler, condenser and heat exchangers
Heat
load in reboiler [kJ/s] 90,771.89 90,537.31 94,101.44
Specific heat load in reboiler [MJ/kg of CO 2] 5.751 5.736 5.691
Heat
transfer area in reboiler [m2] 2687.6 2526.7 2703.2
Heat load in condenser [kJ/s] 51,440.4 42,597.8 47,203.0
Heat
transfer area in condenser [m2] 4035.9 2384.6 2616.2
Heat
load in lean-rich amine heat exchanger [kJ/s] 110,072.6 103,013.0 103,812.7
Heat
transfer area in lean-rich amine heat exchanger [m2] 15,046.3 11,799.9 11,413.5
Heat
load in lean amine cooler [kJ/s] 26,746.5 26,551.4 26,565.0
Heat
transfer area in lean amine cooler [m2] 1302.0 1163.7 1123.1
– In regards to the dimensions of the MEA regeneration unit, it
is possible to conclude that the optimal volume of the pack-
ing material increases as the CO 2removal target increases.
The height and diameter increase, respectively, 29.5% and
16.8% from 70.0 to 95.0 CO 2capture % resulted in an increase
of the volume of the packing material in about 75.8%
(58.0 m3).- The CO 2removal target does not affect the outlet temperatures
of liquid and vapor in the regenerator.
– By comparing the optimal values for the reboiler, it is concluded
that the reboiler temperature and the driving force do not vary
with the CO 2removal target.
– As expected, the heat transfer areas and their heat loads in
the reboiler, condenser, lean-rich amine heat exchanger and
Table 15
Comparison
of the dimensions for absorption and regeneration units obtained from the proposed model and reported data.
Column dimensions
Fisher et al. (2005) This work
Absorber Regenerator Absorber Regenerator
H [m] 15.00 10.00 17.68 11.03
D
[m] 9.70 5.50 8.84 5.66
Vol [m3] 1108.47 237.58 1085.78 277.75
dP
[kPa] 10.30 10.30 8.83 2.42

160 P. Mores et al. / International Journal of Greenhouse Gas Control 10 (2012) 148–163
Table
16
Comparison
of the capacities of the main piece of equipments obtained from the
proposed
model and reported data.
Piece of equipment Equipment sizing
Fisher et al. (2005) This work (2011)
Flue gas blower [kW] 9200.0 7128.0
Absorber (4) [m3] 4433.9 4343.1
Rich amine pump [kW] 1732.0 278.0
Stripper
(4) [m3] 950.3 1111.0
Condenser
[m2] 15,272.0 21,809.2
Reboiler
[m2] 30,436.0 48,751.9
Rich/lean
amine exchanger
[m2]101,136.0 87,686.5
Lean amine cooler [m2] 35,972.0 31,023.5
Amine
storage tank [m2] 276.0 1259.6
Water
storage tank [m2] 628.3 3198.1
CO 2compresors [kW] 34,845.0 40,306.7
CO 2compresors interstage
coolers
[m2]Not included 3933.2
lean amine cooler increase significantly as CO 2removal target
increases. However, the corresponding driving forces are not
affected.
– The CO 2removal target has a strong influence on the total elec-
tricity consumed by compressors and pumps. Certainly, the total
electricity necessary to remove 95.0% of CO 2from flue-gas stream
is 43.0% higher than that required to remove 70.0%.
6.5. Influence of the CO2removal target on the profiles of
temperature and amount of CO2captured along the height of
absorber and regenerator units.
Figs. 4 and 5 illustrate the outlet temperature and CO 2loading
factor in each one of stages of the absorber and stripper.
Fig. 4 clearly shows that, the profiles of temperature along the
stages of the absorber for CO 2removal target = 95% reach greater
values than those predicted for 80%. Both temperature profiles fol-
low qualitatively similar trends and exhibit maximum values indifferent
parts of the absorber. Certainly, for 95% of removal the
temperature increases from 328.4 (inlet temperature as indicated
in Table 9) to 333.3 K in Stage 7 (maximum value) and then it grad-
ually decreases to 325.7 at the bottom of the absorber (Stage 1).
The temperature drop at the bottom of the absorber results from
the cold gas entering to the bottom and contacting the hot liq-
uid flowing downwards. In regards to the temperature drop, it is
important to mention that the proposed model assumes, as a first
approximation, thermal equilibrium between the liquid and vapor
phases. Then, the temperature profiles obtained under non-thermal
equilibrium condition may be similar or slightly different to those
illustrated in Figs. 4 and 5. This aspect will be further investigated
in detail.
As is also shown in Fig. 4, the CO 2loading factor (CO 2mol/MEA
mol) monotonically increases from the top to the bottom of the
column, as expected.
Finally, Fig. 5 shows the corresponding temperature and CO 2
loading profiles along the stripper. It can be seen how the CO 2load-
ing decreases from the top to the bottom of the stripper. The optimal
values corresponding to the leaving CO 2loading for 80% and 95%
of CO 2recovery are 0.26 and 0.20, respectively. Moreover, the opti-
mal values of reboiler temperature (Stage 0) are 391.8 and 394.3 K,
respectively.
Again, as in the case of the absorber, the corresponding temper-
ature and CO 2loading profiles for both CO 2removal targets follow
similar trends. In addition, these trends have been also obtained for
the remaining CO 2removal targets.
6.6. Comparison of optimal designs involving different objective
functions
As mentioned earlier, by exploiting the benefits of the proposed
model, different optimal designs obtained by considering different
objective functions are compared. Thus, it is interesting to com-
pare how much the design of the capture plant is affected when
considering cost in the objective function, as opposed to the heat
duty and specific heat duty as well. Precisely, the optimal solutions
obtained for the following three objective functions are minimized
Table 17
Comparison
of the purchasing equipment costs obtained from the proposed model and reported data.
Piece of equipment Purchased equipment cost (million US$)
Fisher et al. (2005) Fisher et al. (2005) updated to 2011 This work (2011)
Flue gas blower 2.040 2.553 1.005
Absorber
16.320 20.423 12.193
Rich
amine pump 0.272 0.340 0.120
Stripper
3.760 4.705 6.015
Condenser
1.880 2.353 3.071
Reboiler
4.600 5.756 8.049
Rich/lean
amine exchanger 11.200 14.016 13.561
Lean
amine cooler 4.012 5.021 4.828
Amine
storage tank 0.058 0.073 0.338
Water
storage tank 0.097 0.121 0.591
CO 2compresors 16.048 20.082 23.348
Interstage
coolers (CO 2compression) 0.749 0.937 1.099
Others:
Cooling tower, filtration, CO 2compressor
separator,
reflux accumulator and pumps7.797 16.688 Not included
Total 74.371 93.067 74.217
Table 18
Comparison
of the operating costs obtained from the proposed model and reported data.
Fisher et al. (2005) Fisher et al. (2005) updated to 2011 This work (2011)
MEA make-up [million US$/year] 6.95 8.70 8.79
H2O make-up [million US$/year] 2.57 3.21 3.46
Solid
waste disposal [million US$/year] 0.05 0.06 Not included
Cooling
water [million US$/year] Not included – 3.42
Total
utility cost [TU&MU, million US$/year] 9.57 11.98 15.68

P. Mores et al. / International Journal of Greenhouse Gas Control 10 (2012) 148–163 161
and compared: (1) total annual cost (hereafter named P1), (2) total
heat duty (hereafter named P2) and (3) total heat duty/CO 2recovery
(hereafter named P3) where the CO 2recovery is here considered as
an optimization variable. The parameters used to solve the three
problems are listed Tables 4 and 5.
As it is clearly shown in Table 14, similar optimal solutions
are obtained by solving P1, P2 and P3 problems. The total annual
costs predicted by the three objective functions are fairly similar.
Therefore, the detailed comparison in Table 14 reveals that the
objective functions involved by P2 and P3 problems can be used
as alternative optimization criteria for the cost-optimal design of
post-combustion CO 2capture processes using amines. This aspect
is valuable from a solution strategy point of view, because the opti-
mal solutions obtained by solving P2 or P3 can be efficiently used
as initialization to solve P1. Certainly, according to the comparison
of the values predicted for P1 and P2, it is easily to conclude that
solutions of P2 or P3 problems are not only initial feasible points
but also they are in the neighborhood of the optimal solutions for
P1. Therefore, a feasible solution can be easily obtained for P1 after
a few iterations if the solution corresponding to P2 or P3 is used as
starting point.
6.7. A comparison between the output results predicted by the
model and data reported by other authors
It is also interesting to compare the output results obtained by
applying the proposed model with costs reported by other authors
(operating cost and investment). The reference case is taken from
Fisher et al. (2005) and, for a valid comparison, the reported costs
were properly updated to 2011. It should be mentioned, however,
that the process configuration of the reference case is slightly dif-
ferent in comparison to the configuration considered in this article.
For instance, in contrast to this article, Fisher et al. (2005) included
a cooling tower, filtration, CO 2compressor separator and reflux
accumulator and are considered to compute the total purchased
equipment cost. In this article, the cost of cooling water used as
cooling medium is considered to compute the total operating cost.
For a more detailed comparison, the complete cost solutions are
compared and discussed from Tables 15–18 . Therefore, the results
presented in these tables correspond to an unique optimal solution.
Tables 15–17 compare the dimensions, capacities and pur-
chased equipment costs of the main piece of equipments. For the
dimensions of the absorber and regenerator columns, Table 15
clearly shows that the predicted values are in good agreement with
the published data.
From the results presented in Table 16, it can be seen that the
capacities of some of the piece of equipments predicted by the
model are close to the values reported by Fisher et al. but the
capacities predicted for some of the process-units are different.
The differences could be explained by several factors not taken into
account in Fisher et al. but considered in this work and/or vice versa.
For example, the assumptions considered to derive the mathemat-
ical models (mass, energy and momentum balances) may influence
the solutions. Other aspect to be mentioned is that, as opposed to
this work, the solution reported by Fisher et al. was obtained by
simulations instead of optimizations and consequently the results
may be different.
Table 17 compares the predicted and the published values of
the purchased equipment cost of each one of the piece of equip-
ments. Despite the fact that the cost equations used in this work
are different to those used in Fisher et al., the predicted values are in
good agreement. It should be noted that the total purchased equip-
ment cost reported by Fisher et al. is 76.379 million US$ if the costs
involved by the cooling tower, filtration, CO 2compressor separator,
reflux accumulator and pumps are not considered. Also, Table 17clearly
shows that the purchased costs estimated by Fisher et al.
increased 25.0% from 2005 to 2011.
Table 18 compares the distribution of the total utility cost. It
should be stressed here that the cooling water cost was not included
by Fisher et al. (2005) in contrats to this work. Certainly, the values
listed in Table 18 corresponds to a cooling water cost of 0.01 US$/t.
In addition, it should also be mentioned that, for a valid comparison,
the costs of electricity to run compressors and pumps and steam
used in the reboiler of the regeneration section are not included in
Table 18.
Basically, if the cooling water cost is not considered the total
variable O&M cost predicted by the proposed cost model is quite
similar to that reported by Fisher et al. (2005) .
7. Conclusions. Future works
A non-linear and predictive mathematical model for the simul-
taneous optimization of the reactive CO 2absorption into aqueous
MEA solutions has been presented. The proposed model is a valu-
able tool not only to optimize the process but also to simulate the
absorption process if the degree of freedom of the equation sys-
tem is zero. The model variables are simultaneously optimized.
A detailed objective function including the total annualized cap-
ital cost and operating and maintenance costs has been used for
minimization. Output results provide the optimal profiles of tem-
perature, composition and flow-rate of liquid and gas streams along
the height of the absorption and regeneration units. In addition, the
capacities of compressors, pumps, heat transfer areas including the
heat loads and consumption of heating and cooling utilities are also
obtained.
The effect of the CO 2removal target on the total annual cost
and the contributions of each one of the pieces of equipments have
been investigated. From the obtained results, it is concluded that,
from CO 2removal target = 70.0–80.0% the total cost (investment
and operating costs) varies linearly and then, from 80.0 to 95.0%, it
increases exponentially. The total operating cost, which is strongly
influenced by the utilities and electricity, represents more than
75% of the total annual cost. The contribution of the steam used
in the reboiler and the electricity consumed by pumps and com-
pressors on the total operating cost are approx. 57.0% and 28.0%,
respectively.
The ranking order of the contributions of the steam, electricity,
cooling water, MEA and H2O make-ups on the total operating and
maintenance cost is not affected by the CO 2removal target.
In regards to the total purchased equipment cost, it was found
that the compressors and the absorption unit are the pieces of
equipments with the first and second largest impact, respectively,
on the total annual cost for all CO 2removal targets analyzed. From
the obtained results, it is possible to conclude that the contribution
percentage of each one of pieces of equipments is slightly influ-
enced by the CO 2removal target but the effect is strong enough
to vary the corresponding ranking order of some of them (blower
[BLOW], reboiler [REB], inter-stage coolers [I-COOL] and water tank
[H 2O-TK]). From 75.0 to 80% of CO 2removal the contribution of the
water make-up tank [H 2O-TK] is greater than the inter-stage cool-
ers [I-COOL], in contrast to that observed for 70.0%. Then, from CO 2
removal target = 85–92.5% the order between the blower [BLOW]
and reboiler [REB] is also modified; the contribution of [BLOWER]
becomes progressively more important than [REB], in contrast to
that observed for 70.0% to 80% and 95%.
It is interesting and essential to investigate the best alternative
integration between the CO 2capture processes and power-
electricity generation plants in order to reduce the total annual
cost. In fact, part of the steam generated in the power plant can be
used in the reboiler of the regeneration section. Also, the electricity

162 P. Mores et al. / International Journal of Greenhouse Gas Control 10 (2012) 148–163
generated can be efficiently used to drive pumps and compressors.
Thus, the mathematical model presented in this paper will be
coupled to a previous one which has been developed to study
a combined cycle power plant (CCPP). The resulting model will
also consider the entire configuration of the process (number and
arrangement of pieces of equipments) as optimization variable.
Discrete decisions will be used to model different alternative
integrations between the CO 2capture process and the combined
cycle power plant. Pressure levels, number and type of steam
turbines (back-pressure, extraction–condensation turbines) and
the design of the heat recovery steam generator including number
of tubes in economizer, evaporator and super-heater are some of
the optimization variables for the CCPP.
Finally, the extension of the proposed steady state model to a
dynamic model in order to optimize the start-up and shut-down
phases of the process study start-up operation is also another
interesting point to be addressed in future works. Non-thermal
equilibrium between liquid and vapor phases will be also consid-
ered.
Acknowledgements
The following two organizations are gratefully acknowledged
for their financial support: Consejo Nacional de Investigaciones
Científicas y Técnicas (CONICET, Argentina) and the Universi-
dad Tecnológica Nacional Facultad Regional Rosario (UTN-FRRo,
Argentina).
References
Abu-Zahra, M.R.M., Schneiders, L.H.J., Niederer, J.P.M., Feron, P.H.M., Versteeg, G.F.,
2007a. CO 2capture from power plants. Part I. A parametric study of the technical
performance based on monoethanolamine. International Journal of Greenhouse
gas control 1, 37–46.
Abu-Zahra, M.R.M., Schneiders, L.H.J., Niederer, J.P.M., Feron, P.H.M., Versteeg, G.F.,
2007b. CO 2capture from power plants. Part II. A parametric study of the eco-
nomical performance based on mono-ethanolamine. International Journal of
Greenhouse gas control 1, 135–142.
Aliabad, Z., Mirzaei, S., 2009. Removal of CO 2and H2S using aqueous alkanolamine
solutions world academy of science. Engineering and Technology 49, 194–203.
Ali, C., Backham, L., Croiset, E., Douglas, P., 2005. Simulation of CO 2capture using
MEA scrubbing: a flowsheet decomposition method. Energy Conversion and
Management 46, 475–487.
Aroonwilas, A., Veawab, A., 2007. Integration of CO 2capture unit using single-
and blended-amines into supercritical coal-fired power plants: implications
for emission and energy management. International Journal of Greenhouse Gas
Control 1 (2), 143–150.
Blauwhoff, P.M.M., Kamphuis, B., Van swaaij, W.P.M., Westerterp, K.R., 1985.
Absorber design in sour natural gas treatment plants: impact of process vari-
ables. Chemical Engineering and Processing 19, 1–25.
Böttinger, W., Maiwald, M., Hasse, H., 2008. Online NMR spectroscopic study of
species distribution in MEA–H 2O–CO 2and DEA–H 2O–CO 2. Fluid Phase Equilibria
263 (2), 131–143.
Brooke, A., Kendrick, D., Meeraus, A., 1996. GAMS – A User’s Guide (Release 2.25).
The Scientific Press, San Francisco, CA.
Chapel, D., Ernst, J., Mariz, C., 1999. Recovery of CO 2from flue gases: commercial
trends. Originally presented at the Canadian Society Chemical Engineers Annual
Meeting, October 4–6, Saskatoon, Saskatchewan, Canada.
Choi, J.H., Oh, S.G., Kim, Y.E., Yoon, Y., Nam, S.C., 2012. Carbon dioxide absorp-
tion characteristics of aqueous alkanolamine using nuclear magnetic resonance
spectroscopy. Environmental Engineering Science 29 (5), 328–334.
Cozad, A., Sahinidis, N., Miller, D.C., 2010. Optimization of power plant simulations
with integrated carbon capture systems using black-box algorithms. In: 2010
Annual Meeting Conference Proceedings, November 7–12, Salt Lake City, UT.
Dave, N., Do, T., Palfreyman, D., Feron, P.H.M., Xu, S., Gao, S., Liu, L., 2011. Post-
combustion capture of CO 2from coal-fired power plants in China and Australia:
an experience based cost comparison. Energy Procedia 4, 1869–1877.
deMontigny, D., Aboudheir, A., Tontiwachwuthikul, P., Chakma, A., 2006. Modelling
the performance of a CO 2absorber containing structured packing. Industrial &
Engineering Chemistry Research 45, 2594–2600.
Drud, A.S., 1992. CONOPT, A GRG Code for Large Scale Nonlinear Optimization. Ref-
erence Manual. ARKI Consulting and Development A/S, Bagsvaerd, Denmark.
Duan, L., Zhao, M., Yang, Y., 2012. Integration and optimization study on
the coal-fired power plant with CO 2capture using MEA. Energy 3,
http://dx.doi.org/10.1016/j.energy.2011.12.014 .
Fashami, S.Z., Edgar, T.F., Rochelle, G.T., 2009. Texas Wisconsin California Control
Consortium Meeting , February 3rd, 2009.Fisher, K.S., Beitler, C., Rueter, C., Searcy, K., Rochelle, G., Jassim, M., Figueroa, J.,
2005. Integrating MEA regeneration with CO 2compression to reduce CO 2cap-
ture costs. In: Fourth Annual Conference on Carbon Capture and Sequestration
DOE/NETL, May 2–5.
Freguia, S., Rochelle, G.T., 2003. Modeling of CO 2capture by aqueous
monoethanolamine. AIChE Journal 49, 1676–1686.
Greer, T., 2008. Modeling and simulation of post combustion CO 2capturing. PhD
Thesis. Telemark University College, Faculty of Technology, Porsgrunn, Norway.
Henao, C., 2005. Simulación y evaluación de procesos químicos, 1st edition. Univer-
sidad Pontificia Bolivariana, Medellín.
Ho, M.T., Allinson, G.W., Wiley, D.E., 2011. Comparison of MEA capture cost for low
CO 2emissions sources in Australia. International Journal of Greenhouse Gas
Control 5 (1), 49–60.
Huang, Y., Wang, M.H., Stephenson, P., Rezvani, S., McIlveen-Wright, D., Minchener,
A.J., Hewitt, N., Dave, A., Fleche, A. Hybrid coal-fired power plants with CO 2cap-
ture: a technical and economic evaluation based on computational simulations.
Fuel, in press .
Islam, M.S., Yusoff, R., Ali, B.S., Islam, M.N., Chakrabarti, M.H., 2011. Degradation
studies of amines and alkanolamines during sour gas treatment process. Inter-
national Journal of the Physical Sciences 6 (25), 5877–5890.
Karimi, M., Svendsen, H.F., Hillestad, M., 2011. Capital costs and energy consider-
ations of different alternative stripper configurations for post combustion CO 2
capture. Chemical Engineering Research and Design 89 (8), 1229–1236.
Khakdaman, H.R., Zoghi, A.T., Abedinzadegan, M., Ghadirian, H.A., 2008. Revamping
of gas refineries using amine blends. IUST International Journal of Engineering
Science 19 (3), 27–32.
Kister, H.Z., 1992. Destillation Design. McGraw Hill, USA.
Kuramochi, T., Faaij, A., Ramírez, A., Turkenburg, W., 2010. Prospects for cost-
effective post-combustion CO 2capture from industrial CHPs. International
Journal of Greenhouse Gas Control 4 (3), 511–524.
Kvamsdal, H.M., Jakobsen, J.P., Hoff, K.A., 2009. Dynamic modeling and simulation of
a CO 2absorber column for post-combustion CO 2capture. Chemical Engineering
and Processing: Process Intensification 48, 135–144.
Lawal, O.A., Idem, R.O., 2006. Kinetics of the oxidative degradation of CO 2loaded
and concentrated aqueous MEA–MDEA blends during CO 2absorption from flue
gas streams. Industrial & Engineering Chemistry Research 45, 2602–2607.
Lawal, A., Wang, M., Stephenson, P., Yeung, H., 2009. Dynamic modelling of CO 2
absorption for post combustion capture in coal-fired power plants. Fuel 88 (12),
2455–2462.
Lepaumier, H., Martin, S., Picq, D., Delfort, B., Carrette, P., 2010. New amines for CO 2
capture. III. Effect of alkyl chain length between amine functions on polyamines
degradation. Industrial & Engineering Chemistry Research 49, 4553–4560.
Mores, P., Scenna, N., Mussati, S., 2011. Post-combustion CO 2capture process:
equilibrium stage mathematical model of the chemical absorption of CO 2into
monoethanolamine (MEA) aqueous solution. Chemical Engineering Research
and Design 89 (9), 1587–1599.
Mores, P., Scenna, N., Mussati, S., 2012. A rate based model of a packed column for CO 2
absorption using aqueous monoethanolamine solution. International Journal of
Greenhouse Gas Control 6, 21–36.
Nuchitprasittichai, A., Cremaschi, S., 2010. Optimization of CO 2capture process with
aqueous amine using response surface methodology. In: AIChE Annual Meeting,
Conference Proceedings.
Oexmann, J., Hensel, C., Kather, A., 2008. Post-combustion CO 2-capture from coal-
fired power plants: preliminary evaluation of an integrated chemical absorption
process with piperazine-promoted potassium carbonate. International Journal
of Greenhouse Gas Control 2 (4), 539–552.
Oexmann, J., Kather, A., 2010. Minimising the regeneration heat duty of post-
combustion CO 2capture by wet chemical absorption: the misguided focus on
low heat of absorption solvents. International Journal of Greenhouse Gas Control
4 (1), 36–43.
Oi, L.E., 2007. Aspen HYSYS simulation of CO 2removal by amine absorption from a
gas based power plant. In: SIMS 2007 Conference, Gosteborg.
Onda, K., Takeuchi, H., Okumoto, Y., 1968. Mass transfer coefficients between gas
and liquid phases in packed columns. Journal of Chemical Engineering of Japan
1, 56.
Peeters, A.N.M., Faaij, A.P.C., Turkenburg, W.C., 2007. Techno-economic analysis of
natural gas combined cycles with post-combustion CO 2absorption, including
a detailed evaluation of the development potential. International Journal of
Greenhouse Gas Control 1 (4), 396–417.
Peng, D.Y., Robinson, D.B., 1976. A new two constant equation of state. Industrial
Engineering Chemical Fundamentals 15, 59–64.
Perry, R.H., Green, D.W., 1997. Perry’s Chemical Engineers’ Handbook, 7th edition.
McGraw Hill.
Rahimpour, M.R., Kashkooli, A.Z., 2004. Modeling and simulation of industrial car-
bon dioxide absorber using amine-promoted potash solution. Iranian Journal of
Science & Technology, Transaction B 28 (B6), 653–666.
Rao, A.B., Rubin, E.S., 2002. A technical, economic, and environmental assessment of
amine-based CO 2capture technology for power plant greenhouse gas control.
Environmental Science and Technology 36, 4467–4475.
Robbins, L.A., 1991. Improve pressure-drop prediction with a new correlation.
Chemical Engineering Progress 87 (5), 87–91.
Romeo, L.M., Bolea, I., Escosa, J.M., 2008. Integration of power plant and amine
scrubbing to reduce CO 2capture costs. Applied Thermal Engineering 28 (8–9),
1039–1046.
Schach, M.O., Schneider, R., Schramm, H., Repke, J.U., 2010a. Techno-economic
analysis of postcombustion processes for the capture of carbon dioxide from

P. Mores et al. / International Journal of Greenhouse Gas Control 10 (2012) 148–163 163
power plant flue gas. Industrial & Engineering Chemistry Research 49 (5),
2363–2370.
Schach, M.O., Schneider, R., Schramm, H., Repke, J.U., 2010b. Exergoeconomic anal-
ysis of post-combustion CO 2capture processes. Computer Aided Chemical
Engineering 28, 997–1002.
Seider, W.D., Seader, J.D., Lewin, D.R., Widagdo, S., 2009. Product and Process Design
Principles: Synthesis, Analysis and Evaluation, 3rd edition. John Wiley & Sons
Inc.
Singh, D., Croiset, E., Douglas, P.L., Douglas, M.A., 2003. Techno-economic study
of CO 2capture from an existing coal-fired power plant: MEA scrubbing vs.
O2/CO 2recycle combustion. Energy Conversion and Management 44 (19),
3073–3309.
Sinnott, R.K., Towler, G., Coulson, J.M., Richardson, J.F., 2009. Chemical Engineering
Design, 5th edition. Elsevier.
Sipöcz, N., Tobiesen, A., Assadi, M., 2011. Integrated modelling and simulation of a
400 MW NGCC power plant with CO 2capture. Energy Procedia 4, 1941–1948.Turton, R., Bailie, R.C., Whiting, W.B., Shaeiwitz, J.A., 2008. Analysis, Synthesis and
Design of Chemical Processes, 3rd edition. Prentice hall.
United Technologies Research Center, 1999. Combustion, 2000 Phase II, Progress
report DE-AC22-95PC95144—25.
Vaidya, P.D., Kenig, E.Y., 2007. CO 2–alkanolamine reaction kinetics: a review of
recent studies. Chemical Engineering & Technology 30 (11), 1467–1474.
Vaidya, P.D., Kenig, E.Y., 2009. Kinetics of carbonyl sulfide reaction with alka-
nolamines: a review. Chemical Engineering Journal 148 (2–3), 207–211.
Vozniuk, I.O., 2010. Aspen HYSYS process simulation and AspenICARUS cost estima-
tion of CO 2removal plant. FMH606 Master’s Thesis. http://teora.hit.no/dspace/
bitstream/282/975/1/Master%20ThesisIevgeniia%20Vozniuk.pdf .
Xie, H.B., Zhou, Y., Zhang, Y., Johnson, J.K., 2010. Reaction mechanism of
monoethanolamine with CO 2in aqueous solution from molecular modeling.
Journal of Physical Chemistry A 114 (43), 11844–11852.
Yoon, S.J., Lee, H., 2003. Substituent effect in amine–CO 2interaction investigated by
NMR and IR spectroscopies. Chemistry Letters 32 (4), 344.