Energy 76 (2014) 694e703

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Energetic analysis of a syngas-fueled chemical-looping combustion

combined cycle with integration of carbon dioxide sequestration


Angel Jime


nez Alvaro *
pez Paniagua, Celina Gonza
, Ignacio Lo
lez Ferna

ndez,
Rafael Nieto Carlier, Javier Rodríguez Martín
Department of Energy Engineering and Fluid Mechanics, Universidad Polit

ecnica de Madrid, Jos

e Guti

errez Abascal, 2, 28006 Madrid, Spain

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

Article history: Chemical-looping combustion for power generation has significant advantages over conventional com-
Received 20 December 2013 bustion. Mainly, it allows an integration of CO2 capture in the power plant without energy penalty;
Received in revised form secondly, a less exergy destruction in the combustion chemical transformation is achieved, leading to a
12 August 2014
greater overall thermal efficiency. Most efforts have been devoted to systems based on methane as a fuel,
Accepted 18 August 2014
Available online 16 September 2014
although other systems for alternative fuels have can be proposed. This paper focus on the study of the
energetic performance of this concept of combustion in a gas turbine combined cycle when synthesis gas
is used as fuel. After optimization of some thermodynamic parameters of the cycle, the power plant
Keywords:
Chemical-looping combustion
performance is evaluated under diverse working conditions and compared to a conventional gas turbine
CO2 capture and storage system. Energy savings related with CO2 capture and storage have been quantified. The overall efficiency
Gas turbine system increase is found to be significant, reaching values of around 5% (even more in some cases). In order to
Power generation efficiency analyze the influence of syngas composition on the results, different H2-content fuels are considered. In a
Synthesis gas context of real urgency to reduce green house gas emissions, this work is intended to contribute to the
conceptual development of highly efficient alternative power generation systems.
© 2014 Elsevier Ltd. All rights reserved.

1. Introduction combustion is possible, as in IGCC (integrated gasification com-

Since fossil fuels are widely used for power generation, the
effort required to achieve a real reduction of greenhouse gas
emissions is enormous. However, it is possible to attain a short and
medium term substantial reduction through sequestration of the
CO2 produced in fuels oxidation. The CCS (carbon capture and
storage) could thus facilitate the transition to the use of new
sources of clean energy.
There are basically three ways to preform CO2 capture in prac-
tice: a) “post-combustion” capture via amine chemical absorption,
e.g. described by Ref. [1]; b) “pre-combustion” capture, in the case
of a previous fuel decarbonization to a mixture of H2 and CO2 takes
place (several methods are assessed by Ref. [2]); c) “oxy-combus-
tion”, in which the fuel is burned into oxygen instead of into air, and
then nearly pure CO2 is obtained after condensation of water.
The first to strategies, pre- and post-combustion capture,
involve a high energy penalty, since they are based on conventional
separation methods such as separation by membrane, chemical
absorption or adsorption, and cryogenic separation. When pre-
* Corresponding author. Tel.: þ34 91 336 4262.

Jime
E-mail address: a.jimenez@upm.es (A.


nez Alvaro).
bined cycles), it is energetically cheaper due to the fact that CO2 is
less diluted in the fuel than it is in the flue gases; but energy con-
sumption is still significant. In the case of oxy-combustion,
although energy savings strategies have been recently proposed
to minimize the impact [3], the energy penalty occurs in the air
separation unit rather than in the separation of CO2.
The alternative technique of CLC (chemical-looping combus-
tion) was first proposed by Ref. [4], and afterwards several re-
searchers have contributed to the development of this technology.
Nevertheless, most efforts regarding CLC applied to gas turbine
systems have been dedicated to the study of methane as fuel,
although alternative fuels such as methanol have been proposed as
well [5]. This work focuses on the analysis of the energetic effi-
ciency of a CLC gas turbine system fueled by synthesis gas (syngas
hereinafter), in the base of a comparison with a similar gas turbine
system with conventional combustion. There is also prior work on
energetic analysis of a CLC gas turbine system with syngas as fuel
[6,7] but in both cases only the gas turbine cycle is analyzed instead
of a combined gasesteam cycle power plant, and the energy sav-
ings in the capture of CO2 are not quantified. In this paper the
overall energetic performance of a CLC combined cycled power
plant fueled by syngas and with integrated CO2 capture is analyzed
in detail.

http://dx.doi.org/10.1016/j.energy.2014.08.067
0360-5442/© 2014 Elsevier Ltd. All rights reserved.

Jim
enez Alvaro
A.
Nomenclature
cp constant pressure specific heat, (J/(mol∙K))
cp zero-pressure specific heat, (J/(mol∙K))
CR conversion ratio, dimensionless
e flow exergy, (kJ/mol)
h specific enthalpy, (kJ/mol)
LHV fuel's lower heating value, (kJ/mol)
Ka equilibrium constant, dimensionless
n molar flow rate, (mol/s)
p pressure, (bar)
R ideal gas constant, (J/(mol∙K))
s specific entropy, (J/(mol∙K))
T temperature, (K)
v specific volume, (m3/mol)
W power, (kW)
x molar fraction, dimensionless
z compressibility factor, dimensionless
Symbols
h thermal efficiency/energetic efficiency,
dimensionless
Dh difference CLC  conventional thermal efficiencies,
dimensionless
z exergetic efficiency of steam cycle, dimensionless
u acentric factor, dimensionless
DG standard Gibb's function of reaction at 1 bar, (kJ/
mol)
DH standard enthalpy of reaction at 1 bar, (kJ/mol)
DfH standard enthalpy of formation at 1 bar, (kJ/mol)
2. The chemical-looping combustion concept
The idea of the CLC system is illustrated in Fig. 1. The gaseous
fuel is introduced into the reduction reactor and put in contact with
an oxygen carrier, typically a metal oxide. We denote it generically
by “MeO”. The fuel is then oxidized and the metal oxide is reduced,
reacting in accordance with both:
ð2n þ mÞMeO þ Cn H2m /ð2n þ mÞMe þ mH2 O þ nCO2
MeO þ CO/Me þ CO2
Thus, the output stream from reduction reactor contains a
gaseous mixture of CO2 and H2O, so that the only non-condensable
gas in that flow is CO2. The reduced metal oxide “Me” is then
transferred to the oxidation reactor where it is oxidized in presence
of air in accordance with:
1 fication” can be found in Ref. [8].
Me þ O /MeO
2 2
Fig. 1. Schematic diagram of the CLC concept.

et al. / Energy 76 (2014) 694e703 695
s standard entropy at 1 bar, (kJ/(mol∙K))
Subscripts
0, 1, 2, …, 22 state numbers in the CLC cycle
00 , 10, 20 , …, 70 state numbers in a standard Brayton cycle
298 standard thermochemical function evaluated a
25  C (298.15 K)
em electromechanical energetic efficiency
j generic chemical component
c thermodynamic properties at critical point
CC combined cycle
CC þ CCS combined cycle with carbon capture and storage
CCS carbon capture and storage
CLC working pressure of CLC reactors
CO2C1, CO2C2 CO2 compressors
GT gas turbine in a conventional Brayton cycle
GT1, GT2 gas turbines in the CLC cycle
r reduced temperature and pressure
red thermodynamic conditions at reduction reactor
(fuel reactor)
SC simple cycle
ST steam cycle
T standard thermochemical function evaluated at
temperature T
Superscripts
(0), (1) compressibility factor terms in LeeeKesler equation
eq at chemical equilibrium conditions
in initial conditions
M mixing thermodynamic function
As a result, at the outlet of this reactor oxygen-depleted air is
obtained, i.e. basically N2 and O2 in a certain proportion.
Depending on both, fuel and oxygen carrier, the reduction re-
actions (1), (2) can be endothermic or exothermic, while the
oxidation reaction (3) is always exothermic. If one or both of the
reduction reactions is/are endothermic and they take place at low/
medium temperature, then it would be possible to supply the
needed heat from a medium temperature source, e.g. the exhaust
gases in a gas turbine. If this is the case, since, according to Hess'
(1) Law, the overall amount of heat released in reduction and oxidation
reactions must be equal to the fuel's heat of combustion, the
oxidation reaction must have a heat of reaction higher than the
(2)
conventional combustion. As a result, more heat is released at high
temperature in comparison with a conventional combustion. CLC
acts as a “chemical heat pump” transforming energy with a lower
exergy content into energy with a higher exergy content. In other
words, the overall exergy destruction due to irreversibility is lower
with CLC than with conventional combustion. A very interesting
recent theoretical analysis of this kind of “chemical energy ampli-
(3)
On the other hand, the major advantage of CLC lies on the fact that
the carbon dioxide that results from the fuel oxidation is not diluted
with air or any other non-condensable gas, but contrarily it is obtained
in a relatively pure form. This removes the need to invest energy for
separation. The previous two aspects are the basis for this work.
Furthermore, it can also be mentioned that in the CLC combustion
process, the fuel and air pass through different reactors without flame,
what provides an opportunity to minimize NOx formation [9].
The CLC scheme in Fig. 1 can be implemented in practice in
different ways, depending on the oxide's physical characteristics, the
type of reactor and the operating conditions [10]. Typically fluidized-
bed reactors are used, in which the metal oxides “float” as fine
696
Jim
enez Alvaro
A.
et al. / Energy 76 (2014) 694e703
particles, guaranteeing a sufficient contact area for the chemical re-
actions with oxygen (oxidation) and with fuel (reduction) take place.
It may be added some inert material or any catalyst to improve the
physical properties and stability of the metal oxide particles and
chemical kinetics, although this is not investigated in this work.
3. Methodology
3.1. PATITUG: software for thermodynamic analysis
Simulations have been carried out by means of PATITUG, which
is a proprietary software developed at the Higher Technical School
of Industrial Engineering of the Technical University of Madrid
(UPM-ETSII). It is oriented to the design, analysis and optimization
of thermodynamic cycles and their elements. PATITUG is a library
that contains a number of modules that can be conveniently
assembled to calculate thermodynamic properties of all fluid
streams involved in a cycle. The main PATITUG's feature is the
rigorous thermodynamic treatment given in order to provide an
accurate thermodynamic characterization of the cycle. Different
models are included to handle pure substances, mixtures and
chemical transformations. A variety of equations of state, as ideal
gas, virial gas, LeeeKesler equation and IAPWS-IF97 for water, and
different expressions for the specific heat at nil pressure can be
selected. Regarding mixing models, currently IGM (ideal gas
mixture) and LRM (LewiseRandall model) are implemented.
Recent published works have been carried out by the Thermody-
namics group of professors at ETSII-UPM with this software. A
deeper description of PATITUG can be found in Refs. [11,12].
3.2. Cycle description and operating conditions
The thermodynamic cycle that has been analyzed is depicted in
Fig. 2. The main gas turbine is the one referred to as GT1, where
depleted air is expanded from the pressure at oxidation reactor to
around atmospheric pressure. In order to maximize the power
production, two ideas for an optimized components configuration
taken from previous works have been incorporated:
 Introduction of an air pre-heater in order to take advantage of
the solids heat capacity to increase the air mass flow through
GT1. This idea is proposed by Ref. [5].
Fig. 2. Representation of a combined cycle power plant with CLC and CO2 sequestration and storage.
 The oxidation and reduction reaction must be pressure-linked
due to the chemical looping followed by the oxygen carrier. As
suggested by Ref. [6], a second gas turbine (GT2) is then intro-
duced to convert into work the pressurized CO2 and H2O
mixture generated in the reduction reaction. This work is
maximized by heating this stream from the highest available
temperature heat source, the oxidation reactor.
The cycle parameters have been set as following. Ambient
conditions are temperature 15  C (288.15 K), pressure 1 atm
(101.325 kPa) and 60% of RH (relative humidity). Pressure drop at
the air filter is assumed 1 kPa. The value 0.845 is adopted as isen-
tropic efficiency for compressors and 0.895 for gas turbines. Syngas
fuel is taken at ambient temperature and pressure as well. In re-
actors, a pressure drop of 4% is assumed and heat loses of 0.5% and
0.2% are considered in the oxidation and reduction reactors,
respectively. Regarding pressure drops in the heat exchangers, 3.5%
is assumed for the HRSG (heat recovery steam generator) and 1% for
the rest of them. For the pinch point in heat exchanges, the value
10 K has been taken in gasegas exchanges and 50 K in the solidegas
exchange (air pre-heater). Temperature of flue gases at HRSG outlet
is reduced upto dew point (considering the high water content of
this stream) and never under 90  C (363.15 K). All the previous
values are reasonable and within their typical range in combined
cycle power plants [10,5,11]. Final pressure for CO2 storage has been
set to 85 bar, so that storage or transport as high-density super-
critical fluid is allowed. This value is used here since it was previ-
ously adopted by Ref. [10]; nevertheless, if higher pressures were
considered more suitable, the penalty for additional compression
power would not be excessive, as the compression ratio would not
increase significantly in any case.
A conventional Brayton cycle has also been simulated for com-
parison, taking the equivalent values of pressure drops and heat
loses and the same isentropic efficiency of compressors and turbines
than those of the CLC cycle. The Brayton cycle is shown in Fig. 3.
3.3. Definition of thermal efficiency parameters
For each of both cycles, the CLC and the conventional Brayton,
net power output and thermal efficiency are obtained by applying
the thermodynamic model described in Section 3.6. Three different
situations are analyzed:

Jim
enez Alvaro
A.
et al. / Energy 76 (2014) 694e703 697

Fig. 3. Conventional Brayton-steam combined cycle for power generation.

i) SC (simple cycle). Only gas turbines are considered with no conventional combined cycles is usually found in a narrow range
steam cycle downstream: between 0.67 and 0.71, which in addition agrees with our theo-
(a) Net power output: retical estimates. We have taken the value 0.7 in the present work.
(b) Thermal efficiency:
WSC ¼ hem ðWGT1 þ WGT2 Þ ðCLC cycleÞ WCC
WSC ¼ hem WGT ðconventional Brayton cycleÞ hCC ¼ ðCLC cycleÞ
n8 LHV
(4) (7)
WCC
hCC ¼ ðconventional Brayton cycleÞ
n60 LHV
Here, WGT1 ¼ n1(h1h2) þ n4(h4h5) is the difference between
the power produced by turbine GT1 and the air compressor iii) Combined cycle with carbon capture and storage (CC þ CCS).
consumption, WGT2 ¼ n8(h8h9) þ n11(h11h12) represents the The energy consumption by CO2 sequestration and
difference between the power produced by turbine GT2 and the compression upto the storage pressure, which is assumed to
syngas compressor consumption and be done in two compression steps with intermediate cooling,
WGT ¼ n10 ðh10  h20 Þ þ n30 ðh30  h40 Þ þ n60 ðh60  h70 Þ is the gas is added to the combined cycle energetic yield:
turbine production minus the consumption of air and fuel (a) Net power output:
compressors in the Brayton cycle. As usual, ni and hi denote the
molar flow rate and the molar enthalpy of stream i. The elec- WCCþCCS ¼ WCC þ WCO2C1 þ WCO2C2 ðCLC cycleÞ
tromechanical efficiency parameter hem represents the me- WCCþCCS ¼ WCC þ WCCS ðBrayton cycleÞ
chanical and electrical losses in bearings and electricity (8)
generator, and it has been assumed the value 0.98.
(b) Thermal efficiency:
WCO2 C1 ¼ n15(h15h16) is the first CO2 compressor power
WSC consumption, and WCO2 C2 ¼ n17(h17h18) is the consumption in
hSC ¼ ðCLC cycleÞ the second stage. Comparison with the conventional Brayton
n8 LHV
(5) cycle has been carried out, assuming a post-combustion CO2
WSC
hSC ¼ ðconventional Brayton cycleÞ sequestration. For that purpose, the power consumed for the
n60 LHV separation of CO2 from the flue gases stream has been estimated
in base of a series of post-combustion capture simulations
where LHV stands for the fuel's lower heating value. conducted with GT-Pro, which resulted in an electricity con-
sumption of approximately 380 kJ/kg of captured CO2 in addi-
ii) CC (combined cycle). Gas turbines as well as a downstream tion to the compression power upto the storage pressure.
steam cycle are both considered, while the carbon capture and (b) Thermal efficiency:
storage stage is not taken into account:
(a) Net power output. WCC ¼ WSC þ WST, where the power WCCþCCS
hCCþCCS ¼ ðCLC cycleÞ
production of the steam cycle WST is quantified as a fraction n8 LHV
of the physical exergy extracted from gas streams at the WCCþCCS
HRSG: hCCþCCS ¼ ðconventional Brayton cycleÞ
n60 LHV
(9)
WST ¼ z½n6 ðe6  e7 Þ þ n13 ðe13  e14 Þ ðCLC cycleÞ
WST ¼ zn40 ðe40  e50 Þ ðBrayton cycleÞ 3.4. Air and fuel composition
(6)
Dry air composition has been assumed as shown in Table 1,
where ei denotes the physical flow exergy of stream i, defined as neglecting other minor components of air. The imposition of 60%
ei ¼ (hih0)T0(sis0), si representing the i-stream’s molar entropy, RH results in air as a mixture of 1.009% water and 98.009% dry air
and the subscript 0 referring to the same stream's thermodynamic (in molar fraction).
property evaluated at ambient pressure and temperature. Simula- As fuel, a typical syngas from gasification of biomass is consid-
tions carried out with the Thermoflow's GT-Pro software [13] for a ered (after clean up and acid gas and moisture removal). This type
number of commercial gas turbines show that the value of z in of syngas is primarily a mixture of hydrogen and carbon monoxide,
698
Jim
enez Alvaro
A.
et al. / Energy 76 (2014) 694e703

Table 1 Table 2
Dry air composition (molar fraction in %). Composition (molar fraction) and LHV of fuels under study.

N2 O2 Ar CO2 Gasification CO H2 CO2 N2 Ar H2O LHV

substrate (%) (%) (%) (%) (%) (%) (kJ/mol)
78.045 20.990 0.935 0.030
Syngas A: Wood waste 46.90 26.02 18.45 8.09 0.52 0.02 195.64
Syngas B: Miscanthus 45.84 35.46 11.28 7.12 0.28 0.02 215.47

with certain amount of carbon dioxide, and some residual quanti-

ties of nitrogen and argon. We have taken two different syngas for
the simulations, with a significant difference in hydrogen content,
in order to determine the influence of the syngas composition on
results. Their composition and LHV are given in Table 2. Other
syngases from gasification of different biomass substrates or coal
usually present an hydrogen content between the two values
shown in the table.
of water and steam [14]. This equation of state has been used
only for gaseous water in the cycle points where temperature
exceeds the water boiling temperature at that pressure.
b) The virial gas equation of state truncated after the second
term:
pv ¼ RT þ BðTÞp (12)

where v is the molar volume and R the ideal gas constant. The
3.5. Oxygen carrier
correlation used for function B(T) is taken from Ref. [15]:

A few metal oxides have been proposed to act as oxygen carriers b g d ε

in CLC; typically nickel, iron, copper and manganese oxides, e.g. BðTÞ ¼ a þ þ þ þ (13)
T T3 T8 T9
NiO/Ni, Fe2O3/Fe3O4, CuO/Cu, CuO/Cu2O, Mn2O3/Mn3O4, etc. For the
case of methane as fuel, the first two options are the most
frequently used. However, for the case of the reaction with carbon This equation of state has been used for gaseous water in the
monoxide and hydrogen, only the pair F2O3/FeO from among those cycle points where temperature is below the water boiling
tested meets the requirement to provide an endothermic chemical temperature at that pressure (but exceeds the water boiling
transformation at the reduction reactor: temperature at water's partial pressure in that point, so we find
water in gaseous state) and for all non-condensable gases when
+


Fe2 O3 þ H2 /2FeO þ H2 O DH298 ¼ 39:611 kJ
mol their specific volume is at least twice its critical specific volume:
+ (10) v  vc .
Fe2 O3 þ CO/2FeO þ CO2 DH298 ¼ 1:564 kJ mol
c) LeeeKesler’s equation of state:
+ represents the standard enthalpy of reaction1 at 25  C
where DH298
pv
(298.15 K). Therefore, this is the MeO/Me pair selected for the ¼ zð0Þ þ uzð1Þ (14)
RT
simulations.
In the oxidation reactor takes place the needed reaction to close where u represents the gas' acentric factor, and z(0), z(1) are well
the chemical loop: known functions of Tr ¼ T/Tc and pr ¼ p/pc, given in Ref. [16]. Critical
temperature Tc and pressure pc and u have been taken from Ref.
1 + [15] as well. This equation of state has been used for the rest of the
2FeO þ O /Fe2 O3 DH298 ¼ 281:438 kJ=mol (11)
2 2 cases, i.e. non-condensable gases with v < vc and supercritical
Inert material has been reported necessary to achieve the fluids.
appropriate physical characteristics of the solid particles in relation
with their stability facing high changes of temperature. Among the 2. Specific heat at nil pressure:
several oxides that have been proposed, ZrO2 has been selected A temperature-exponential model for cp is needed in this
since [7] referred that YSZ (ZrO2 stabilized by yttria) seems to act as analysis, since polynomial expressions would lead to a loss of
a catalyst in reactions (10). The ratio between inert material and accuracy due to the very wide range of temperatures involved in
oxygen carrier has been assumed 0.27 mol ZrO2 per mol FeO, an the phenomenon. The following correlation, given by Ref. [15],
intermediate value between those given in Refs. [6] and [10] (0.2 has been used:
and 0.34). ZrO2 does not take part in any reaction and just acts as a !2 !2
c e
heat transporter between the two reactors. cp ðTÞ ¼aþb T
þd T
(15)
sinh Tc sinh Te

3.6. Thermodynamic modeling

3.6.2. Mixing model for gas streams
Gaseous streams have been modeled by the LewiseRandall rule:
The thermodynamic properties of combustion gases can vary
significantly with respect to an ideal gas mixture. Therefore, a
vM ¼ 0
proper thermodynamic treatment is of great importance to obtain
hM ¼ 0 X (16)
realistic results.
sM ¼ R xj ln xj
j
3.6.1. Pure gases treatment
where the superscript M denotes the corresponding specific mixing
1. Selection of thermal equations of state: property and xj is the molar fraction of the j-th substance in the
a) IAPWS-IF97 equation of state, according to the last standard mixture.
on industrial formulation for the thermodynamic properties
3.6.3. Solids treatment
The functional dependence of molar enthalpy and entropy
1
Given for a standard pressure of 1 bar. with temperature for solids has been taken from the NIST-JANAF

Jim
enez Alvaro
A.
et al. / Energy 76 (2014) 694e703
Fig. 4. Conversion ratios of H2 and CO at 15 bar as a function of temperature.
Thermochemical Tables [17]. A slight correction to account the
variation of enthalpy with pressure has been applied, considering
ðvh=vpÞT zv. Variation of the specific Gibb's function with pressure
has also been considered for the chemical equilibrium
calculations.
3.6.4. Thermochemical data
+
Standard heats of formation Df H298 and standard molar en-
tropies s+298 have been taken from Ref. [15] for gases and from the
NIST Chemistry Webbook [18] for solids.
4. Results
4.1. Chemical equilibrium
One of the key parameters to be optimized in the cycle design is
the temperature of the reduction reactor. As the first of trans-
formations (10) is endothermic, the conversion ratio of this reaction
must increase with temperature, according to Le Ch^ atelier's Prin-
ciple. However, this temperature is limited as the reduction reactor
has to take the required heat from the flue gas streams at the gas
turbines outlets. In other words, the temperature of the reduction
reactor must be high enough for the reaction to occur in a high
percentage so that almost no fuel is lost, but at the same time it
should be sufficiently low compared with the gases temperature at
the outlet of the turbine in order to allow the recovery of the
necessary heat.
It is supposed that chemical equilibrium is achieved at the
reduction reactor, implicitly assuming that the reactants residence
time is high enough in comparison with the characteristic chemical
kinetics times. Catalysis may be needed. Thus, the chemical
composition of the gas stream at the reduction reactor's outlet, as a
function of temperature, is obtained from the chemical equilibrium
equation:
 
DG+T
Ka ðTÞ ¼ exp  (17)
RT
where DG+T is the standard Gibb's function of reaction at tempera-
ture T, and the equilibrium constant Ka can be related with the
molar fractions of gases at equilibrium. The dependence on pres-
sure is very slight due to the conservation of the number of gas
moles in both reactions (10). Results are shown in Fig. 4, where
conversion ratio is defined as
xeq
j
CRj ¼ 1  j≡H2 ; CO (18)
xin
j
699
eq
xin
j and xj being j's initial molar fraction in syngas and its molar
fraction at the chemical equilibrium.2
It can be seen that for a reaction temperature of around 800 K,
H2's conversion rate reaches 99% and CO's one is located at about
99.7%. For this reason, approximately 0.55% of the fuel's LHV would
be lost due to incomplete combustion. However, this effect is more
than offset by the lower exergy destruction in the combustion
process, as discussed below.
At the oxidation reactor, chemical equilibrium has been found to
never occur, as all FeO is oxidized before that point (equilibrium
would be attained for a really small oxygen's molar fraction, much
below the available oxygen in the reactor).
4.2. Cycle optimization and thermal efficiency
Some parameters dictate the thermodynamic conditions in all
streams of the cycle depicted in Fig. 2. The governing one is the TIT
(turbine inlet temperature), i.e. the temperature at which the
oxidation reaction takes place. As it is well known, TIT should be as
high as possible as long as the technical requirements in gas tur-
bines and reactors are satisfied, in order to increase efficiency. The
other key parameters to be determined by analysis and optimiza-
tion are temperature at the reduction reactor Tred and pressure at
both CLC reactors pCLC. In the case of the CC þ CCS cycle, there is
another degree of freedom, since pressure after the expansion in
GT2 is not a priori determined.
Simulations have been performed for a set of TIT in the range
going from 1400 K to 1550 K. From the gas turbine point of view,
higher temperature could be tried, since modern refrigerated gas
turbines allow TIT of the order of 1700 K, or even higher. Never-
theless, the required physical stability of the solid particles of metal
oxides would be undermined due to thermal stresses, as reported
by Ref. [10]. For each TIT value a Newton iteration procedure is
applied in order to find the optimal value of the other parameters
mentioned in the previous paragraph. The optimization is per-
formed with the three defined efficiency parameters (5), (7), (9) as
objective functions. The results for the found optimal values are
presented in Table 3.
It can be read from Table 3 that the reduction temperature
reaches values of around 800 K, which ensures a high ratio of fuel
oxidation, as discussed in Section 4.1. On the other hand, a higher
efficiency is attained by syngas B, which is the richest fuel in
hydrogen. This is expected, since the oxidation of fuel becomes
more endothermic, leading to a lower overall exergy destruction.
2
As xin
H2 O z0 for both considered syngas compositions, the couple of curves for
CRH2 result to be completely indistinguishable. Only one of them is printed.
700
Jim
enez Alvaro
A.
et al. / Energy 76 (2014) 694e703

Table 3
Optimal cycle conditions.

TIT (K) Syngas SC CC CC þ CCS

Tred (K) pCLC (bar) hSC (%) Tred (K) pCLC (bar) hCC (%) Tred (K) pCLC (bar) pGT2 (bar) hCCþCCS (%)

1400 A 765.50 14.15 33.98 769.31 12.79 53.40 775.50 13.04 1.57 47.13
B 760.34 14.09 34.23 764.41 12.67 53.66 764.57 12.66 1.06 48.65
1425 A 774.40 14.57 34.45 777.90 13.23 53.92 783.98 13.48 1.54 47.65
B 769.16 14.50 34.71 772.90 13.10 54.19 772.79 13.08 1.04 49.17
1450 A 783.24 14.98 34.91 786.41 13.69 54.42 792.37 13.95 1.51 48.15
B 777.93 14.90 35.17 781.31 13.56 54.69 780.92 13.52 1.02 49.68
1475 A 792.02 15.40 35.36 794.87 14.16 54.91 800.72 14.42 1.48 48.63
B 786.63 15.31 35.63 789.69 14.01 55.18 789.01 13.97 1.00 50.17
1500 A 800.73 15.81 35.80 803.30 14.62 55.38 809.02 14.89 1.46 49.10
B 795.27 15.72 36.07 798.03 14.47 55.66 797.05 14.41 0.98 50.65
1525 A 809.38 16.22 36.22 811.69 15.10 55.84 817.27 15.36 1.43 49.55
B 803.85 16.12 36.50 806.33 14.94 56.12 805.05 14.86 0.97 51.12
1550 A 817.97 16.63 36.64 820.04 15.57 56.28 825.47 15.84 1.41 49.99
B 812.37 16.53 36.92 814.59 15.40 56.57 813.00 15.31 0.95 51.57

The optimal pressure in reactors is quite similar for CC and CC þ CCS
cycles, and somehow lower than that for SC.
As an example of the thermodynamic conditions along the cycle,
Table 4 shows temperature, pressure, composition and molar flow
(normalized to fuel molar flow) for every stream in the optimized
CC þ CCS cycle, in the case of syngas A and 1500 K at the oxidation
reactor (TIT).
The variation of the energetic performance with pCLC around the
optimal for the three analyzed cycles is given in Figs. 5e7.
Similarly to the behavior of conventional simple and com-
bined cycles, not only the thermal efficiency but also the optimal
pCLC increases with TIT. The difference in efficiency for both
synthesis gas maintains almost constant along the curves, in
values of 0.12e0.17% for SC, and 0.22e0.24% for CC. Another
interesting point to be remarked is that the optimal pressure in
SC is quite moderate. This is related to the fact that CLC attains its
best behavior when a relatively high temperature results at
turbines' outlets, as more energy is reintroduced in the CLC
“chemical heat pumping effect”. As mentioned previously, CC's
optimal pressure is found to be somehow lower than SC's one.
The optimal pCLC differs very slightly for both fuels in all studied
cases.
Regarding the thermal efficiency of the CC þ CCS cycle, it must
be kept in mind that it basically differs from that of the CC cycle
basically by a constant value that can be seen as proportional to the
amount of CO2 to be compressed for storage, i.e. the carbon initially
contained in the fuel. These figures are 65.35% for syngas A and
57.12% for syngas B, in molar fraction terms. For this reason,
together with the fact that B's LHV is greater than A's LHV efficiency
(LHV is in denominator of equation (9)), hCCþCCS results to be about
1.6% higher for syngas B, which is a significant amount. The position
of the optimal pCLC is nearly the same in these two cases.
Although in the case of CC þ CCS cycle the optimization pro-
cedure is design to also determine the optimal pGT2, the influence of
this parameter on the overall cycle thermal efficiency has found to
be minimal in the range of 0.5e2 bar.
4.3. Comparison with a conventional combustion cycle
A comparison between the thermodynamic behavior of CLC and
conventional combustion cycles, in equivalent conditions of TIT,
pressure ratio of compression, heat and pressure loses, and isen-
tropic efficiency of turbines and compressors, is reported in Fig. 8,
where Dh ¼ hCLChconventional.

Table 4
Thermodynamic conditions after cycle optimization (case: syngas A, TIT 1500 K).

Stream n (mol/s) T (K) p (bar) xN2 (%) xO2 (%) xAr (%) xCO2 (%) xH2 O (%) xCO (%) xH2 (%)

0 6.71 288.15 1.01 77.26 20.78 0.93 0.03 1.01 0.00 0.00
1 6.71 288.15 1.00 77.26 20.78 0.93 0.03 1.01 0.00 0.00
2 6.71 674.96 15.04 77.26 20.78 0.93 0.03 1.01 0.00 0.00
3 6.71 1174.80 14.89 77.26 20.78 0.93 0.03 1.01 0.00 0.00
4 6.35 1500.00 14.29 81.67 16.25 0.98 0.03 1.07 0.00 0.00
5 6.35 864.63 1.06 81.67 16.25 0.98 0.03 1.07 0.00 0.00
6 6.35 819.02 1.05 81.67 16.25 0.98 0.03 1.07 0.00 0.00
7 6.35 363.15 1.01 81.67 16.25 0.98 0.03 1.07 0.00 0.00
8 1.00 288.15 1.01 8.09 0.00 0.52 18.45 0.02 46.90 26.02
9 1.00 636.45 14.89 8.09 0.00 0.52 18.45 0.02 46.90 26.02
10 1.00 809.02 14.29 8.09 0.00 0.52 65.19 25.79 0.16 0.25
11 1.00 1490.00 14.15 8.09 0.00 0.52 65.19 25.79 0.16 0.25
12 1.00 1082.50 1.47 8.09 0.00 0.52 65.19 25.79 0.16 0.25
13 1.00 819.02 1.46 8.09 0.00 0.52 65.19 25.79 0.16 0.25
14 1.00 363.15 1.41 8.09 0.00 0.52 65.19 25.79 0.16 0.25
15 0.76 298.15 1.41 10.66 0.00 0.70 85.87 2.25 0.21 0.32
16 0.76 490.98 10.93 10.66 0.00 0.70 85.87 2.25 0.21 0.32
17 0.74 298.15 10.93 10.87 0.00 0.71 87.60 0.29 0.21 0.33
18 0.74 497.66 85.00 10.87 0.00 0.71 87.60 0.29 0.21 0.33
19 0.74 298.15 85.00 10.90 0.00 0.71 87.82 0.04 0.21 0.33
20 1.12 724.96 14.89 e e e e e e e
21 1.84 809.02 14.89 e e e e e e e
22 1.12 1500.00 14.89 e e e e e e e

Jim
enez Alvaro
A.
et al. / Energy 76 (2014) 694e703 701

Fig. 5. Thermal efficiency of the simple cycle (SC) for different TIT. The intermediate marks in the optimal curves are given for TIT values 1425 K, 1450 K, 1500 K and 1525 K.

Fig. 6. Thermal efficiency of the combined cycle (CC) for different TIT. The intermediate marks in the optimal curves are given for TIT values 1425 K, 1450 K, 1500 K and 1525 K.
702
Jim
enez Alvaro
A.
et al. / Energy 76 (2014) 694e703

Fig. 7. Thermal efficiency of the combined cycle with CO2 sequestration (CC þ CCS). The intermediate marks in the optimal curves are given for TIT values 1425 K, 1450 K, 1500 K and
1525 K.

It can be observed that the values of SC thermal efficiency are
between 2.1 and 2.9 percentage points higher for CLC. This result is
notable, since CLC cycle requires additional heat exchanges and
extra pressure losses, implying exergy destruction. Furthermore
about 0.5% of fuel's LHV is lost due to incomplete combustion. That
clearly highlights that the lowest overall exergy destruction ach-
ieved by the “chemical heat pumping effect” overrides fairly all
other unfavorable effects.
However, the CC thermal efficiency is found to be very similar in
CLC and conventional cycles. This could be expected, since the
available energy in the exhaust gases is greater in the case of a
conventional Brayton cycle, so then more exergy is available for
recovery at the HRSG.
Nevertheless, the full potential of the CLC technique is stated on
view of the CC þ CCS cycle thermal efficiency increase, which
reaches 4.5e5.9 percentage points. The figures are higher for the
syngas A because this one is the richest in carbon content, and so a
further energy penalty in CO2 capture with chemical conventional
methods is attained. The differences are significant in all cases,
what finally justifies (and quantifies) the interest of this technology.
5. Conclusions and final remarks
In this paper, an analysis on the energetic performance of gas
turbine-based cycles with chemical-looping combustion and a
comparison with similar systems with conventional combustion,
when syngas from gasification of biomass is used as fuel and iron
oxides are used as oxygen carrier, is presented. Thermal efficiencies
of simple cycle, combined cycle and combined cycle with CO2
sequestration and storage have been calculated, after determining

Fig. 8. Difference of efficiency between CLC and conventional combustion cycles with equivalent thermodynamic conditions.

Jim
enez Alvaro
A.
the governing parameters of the CLC cycle through an optimization
algorithm. Two different H2-content syngases have been selected as
fuel input in order to quantify the influence of fuel composition in
the results.
The main conclusions of this work can be summarized as
follows:
 According to chemical equilibrium calculations, a temperature
in the reduction reactor in a range 760e830 K ensures a high
degree of conversion ratio in the chemical transformation, but
also allowing the possibility of taking all the required heat from
the flue gases streams at the gas turbines' outlets. The more
hydrogen content in the fuel, the more the reduction tempera-
ture should be lowered in order to achieve the energy balance
(see Fig. 4 and Table 3).
 The optimal pressure in CLC reactors has found to be lower than
that of the combustion chamber in a conventional Brayton cycle,
since the CLC benefits in terms of exergy savings increase for
higher temperatures of the flue gases streams.3
 The increase of thermal efficiency in the case of a simple gas
turbine cycle is between 2.1 and 2.9 percentage points, being
higher for the fuel with larger amount of hydrogen (see SC case
comparison given in Fig. 8).
 However, this figure does not justify by itself the use of this
technique for power generation, since the difference in the flow
physical exergy contained in the flue gases streams compen-
sates this effect, leading to attain very similar thermal effi-
ciencies when a downstream steam cycle is implemented in a
combined cycle power plant (see CC case comparison given in
Fig. 8).
 The true potential of a CLC system for power generation lies on
the energy savings with CO2 sequestration and storage. When
the energy cost of CO2 capture through chemical processes is
taken into account, the overall thermal efficiency increase rea-
ches significant values, in the range between 4.5 and 5.9 per-
centage points, depending on the total amount of carbon
contained in the synthesis gas (see CC þ CCS comparison given
in Fig. 8).
Results show that the CLC cycle offers an interesting potential
for efficient power generation and nearly zero CO2 emissions.
Whereas biomass can be considered a CO2-neutral energy source,
actual emissions would be de facto negative when the sequestration
of CO2 from biomass syngas combustion is conducted.
As future work, we find of interest to carry out an overall en-
ergetic performance analysis of a fully IGCC power plant with CLC.
3
The analysis of the optimal pressure ratio for a conventional Brayton cycle is out
of the scope of this paper. This analysis can be seen in Ref. [11] for methane as fuel.
Pressure ratios from 15 to 25 are found as optimal values in the TIT range under
study here.

et al. / Energy 76 (2014) 694e703 703
The present work can be seen as the starting point for that objec-
tive. Until now, it has not been possible to do such an analysis due to
the fact that, as far as we know, commercial thermal engineering
software for power generation does not have implemented a CLC
system module. Currently we are working on the development of
an interface to integrate our CLC thermodynamic model with
Thermoflow Inc.'s software in order to go after this goal.
References
[1] Chiesa P, Consonni S. Natural gas fired combined cycles with low co2 emis-
sions. J Eng Gas Turb Power 2000;122(3):429e36.
[2] Urech J, Tock L, Harkin T, Hoadley A, Mare
chal F. An assessment of different
solvent-based capture technologies within an IGCC-CCS power plant. Energy
2014;64(1):268e76.
[3] Hagi H, Le Moullec Y, Nemer M, Bouallou C. Performance assessment of first
generation oxy-coal power plants through an exergy-based process integra-
tion methodology. Energy 2014;69(6):272e84.
[4] Ishida M, Jin H. A new advanced power-generation system using chemical-
looping combustion. Energy 1994;19(4):415e22.
[5] Zhang X, Han W, Hong H, Jin H. A chemical intercooling gas turbine cycle with
chemical-looping combustion. Energy 2009;34(12):2131e6.
[6] Anheden M, Svedberg G. Exergy analysis of chemical-looping combustion
systems. Energy Convers Manage 1998;39(16e18):1967e80.
[7] Anheden M. Analysis of gas turbine systems for sustainable energy conver-
sion. Ph.D. thesis. Stockholm, Sweden: Royal Institute of Technology; 2000.
ISRN KTH/KET/e11eSE.
[8] Zheng D, Jing X. Chemical amplifier and energy utilization principles of heat
conversion cycle systems. Energy 2013;63(15):180e8.
[9] Ishida M, Jin H. A novel chemical-looping combustor without NOx formation.
J Ind Eng Chem 1996;35(7):2469e72.
[10] Consonni S, Lozza G, Pelliccia G, Rossini S, Saviano F. Chemical-looping
combustion for combined cycles with co2 capture. J Eng Gas Turb Power
2006;128(3):525e34.
[11] Nieto R, Gonz

pez I, Jime
alez C, Lo
nez A. Efficiency of a standard gas-turbine
power generation cycle running on different fuels. Int J Exergy 2011;9(1):
112e26.
[12] Escudero M, Jime
nez A, Gonza
lez C, Nieto R, Lo

pez I. Analysis of the behavior
of biofuel-fired gas turbine power plants. Therm Sci 2012;16(3):849e64.
[13] GTPRO-Win Software for turbine power and cogeneration system design.
Subdury, MD, USA: Thermoflow Inc.; 2001.
[14] IAPWS. Revised release on the IAPWS industrial formulation 1997 for the
thermodynamic properties of water and steam. Tech. rep. Lucerne,
Switzerland: The International Association for the Properties of Water and
Steam; 2007.
[15] Daubert TE, Danner RP. Physical and thermodynamic properties of pure
chemicals: data compilation. New York, NY, USA: Hemisphere Publishing
Corporation; 1989.
[16] Lee BI, Kesler MG. A generalized thermodynamic correlation based on three-
parameter corresponding states. AIChE J 1975;21(3):510e27.
[17] http://kinetics.nist.gov/janaf/, online [accessed 30.07.14], National Institute of
Standards and Technology, U.S. Department of Commerce.
[18] http://webbook.nist.gov/chemistry/form-ser.html, online [accessed 30.07.14],
National Institute of Standards and Technology, U.S. Department of
Commerce.