Energy Conversion and Management 107 (2016) 4351

Contents lists available at ScienceDirect

Energy Conversion and Management

journal homepage: www.elsevier.com/locate/enconman

Detailed exergetic evaluation of heavy-duty gas turbine systems running

on natural gas and syngas
Max Sorgenfrei , George Tsatsaronis
Institute for Energy Engineering, Technische Universitt Berlin, Berlin, Germany

a r t i c l e

i n f o

Article history:
Available online 11 April 2015
Keywords:
Gas turbine
Exergy analysis
Inefciencies
Syngas
IGCC

a b s t r a c t
Gas turbine systems are widely used for the production of electricity in a simple or combined-cycle mode
today. Based on their ability to allow a fast load change, gas turbine systems will become even more
important in the future since the volatile production of renewable energies will increase. In this study,
a state-of-the-art gas turbine running on natural gas, having an overall net efciency of approximately
40%, is modeled using Aspen Plus and characteristic parameters are identied. Based on these parameters, a gas turbine running on syngas was simulated. The emphasis here is on a very detailed evaluation of
the inefciencies. The models consider cooling and sealing ows. The syngas considered in this study is
typically used in IGCC processes with carbon capture resulting in a high concentration of hydrogen.
For both systems, twelve types of inefciencies were identied and rated. A comparison of the inefciencies within each system and between both systems represented by their exergy destruction ratios is
presented.
In case of the gas turbine running on natural gas, the most important results show that the stoichiometric combustion, followed by the addition of excess air represent the largest inefciencies. When just
applying an isentropic efciency, the exergy destructions associated with expansion and mixing at different temperatures and pressures of a gas turbine stage cannot be further sub-divided. Hence, this grouping
of inefciencies results to the third position. The effect of mixing at different compositions and the compression follows.
In the second case considered here (use of syngas instead of natural gas), the effects of mixing and adding excess air become more signicant due to a higher specic heat capacity of the combustion gas. In
both cases, the exergy destruction associated with mixing at different compositions can be neglected
except the one at the inlet of the pre-mixed combustor, which strongly depends on the particular conditions of the fuel gas. Inefciencies such as convective cooling of the vanes and blades, heat loss, losses
associated with the shaft and generator were found to represent a very small part of the overall exergy
destruction. The resulting exergy destructions and losses are shown in an exergy ow diagram.
2015 Elsevier Ltd. All rights reserved.

1. Introduction
Based on the scenario of global electricity production published
by the International Energy Agency (IEA), the relative consumption
of natural gas will increase only slightly by about 0.4% points from
the year 2011 to the year 2030. However, the relative consumption
of coal will decrease by 6.7% points. Based on the growth of the
Non-OECD nations, the demand for electricity will increase signicantly by 54%, leading to an increase of the absolute coal consumption by 29.7% [1]. Since both types of fuel will play an important

Corresponding author. Tel.: +49 30 314 21738.

E-mail addresses: sorgenfrei@iet.tu-berlin.de (M. Sorgenfrei), tsatsaronis@
iet.tu-berlin.de (G. Tsatsaronis).
http://dx.doi.org/10.1016/j.enconman.2015.03.111
0196-8904/ 2015 Elsevier Ltd. All rights reserved.

role in the future, a deep understanding of the energy conversion

systems is highly interesting. Usually gas turbines are used in a
simple or a combined-cycle mode when producing electricity. In
the combined-cycle mode, the overall net electrical efciency
reaches approximately 60% at its maximum [2], which is the highest value among all used fossil energy conversion systems. Another
advantage is the ability of fast load changes, which will become
even more important in the future because the volatile production
of renewable energies will increase. When using hard coal, a syngas has to be produced from it, to run a gas turbine system. One
of the most efcient technologies available is an IGCC (Integrated
Gasication Combined-Cycle) plant. In large scale operations the
syngas is produced by an entrained-ow gasier followed by a
CO shift reaction. Before the syngas enters the gas turbine system,

44

M. Sorgenfrei, G. Tsatsaronis / Energy Conversion and Management 107 (2016) 4351

Nomenclature
Abbreviations
AC
air compressor
CC
combustion chamber
CFD
computational uid dynamics
COT
combustion outlet temperature
GT
gas turbine
IGCC
integrated gasication combined-cycle
LHV
lower heating value
NGGT
case gas turbine system running on natural gas
SGT
case gas turbine system running on syngas
TIT
turbine inlet temperature
Variables
e
E_

h
n_
Dp

mole specic exergy

exergy rate
mole specic enthalpy
mole ow
pressure loss

CO2 can be captured by physical absorption resulting in a very high

amount of hydrogen [3]. This study focusses on the understanding
of the inefciencies within both gas turbine systems.
In several studies it has been reported about an exergy analysis
of a simple gas turbine model according to ISO 2314 to account for
the thermodynamic inefciencies within the compressor, combustion chamber and turbine [410]. Among others the authors of this
study [11] presented a conventional exergy analysis of several
IGCC concepts, using an enhanced gas turbine model that considers
air ows from the compressor to the turbine. El-Masri [12] presented the results of a conventional exergy analysis based on a
simplied three-stage gas turbine model focusing on the inefciencies associated with the cooling system. It was found that the overall efciency is inuenced by the tradeoff between decreased
combustion losses and increased turbine cooling losses. Khodak
and Romakhova [13] also performed a conventional exergy analysis of an air-cooled gas turbine system. The applied approach separates the system into a topping cycle producing electricity and a
bottoming cycle representing the process management of the cooling ows. The inefciencies within the cooling system were found
to be caused by heat transfer between the main gas and coolant,
the bottoming cycle and mixing at different compositions.
Staudacher and Zeller [14] performed a grouping of the inefciencies of an aircraft turbine, evaluating different setups of the secondary air system. Based on data supported by Rolls-Royce, the
inefciencies were rated using an exergy analysis focusing on the
secondary air system. The authors of this study published the
results obtained from an advanced exergy analysis of a gas turbine
system running on natural gas considering cooling ows [15]. By
splitting the exergy destruction into its avoidable and unavoidable
parts, the potential of improvement for each component is determined. It was found that a large amount of the exergy destruction
associated with combustion, addition of excess air, and mixing is
unavoidable. In comparison, this study focuses on the sources of
inefciencies based on an enhanced gas turbine system model presenting the results in more detail.
This paper focuses on the detailed modeling and evaluation of a
gas turbine system running on natural gas. To improve the understanding of inefciencies, the model is designed to calculate the
distribution of inefciencies among all components divided by its
sources. Furthermore, the bleed air of the compressor provided
to the gas turbine is further split into cooling and sealing parts.

R
s
T
x
y

universal gas constant

mole specic entropy
temperature
mole fraction
dimensionless exergy destruction

Subscripts and superscripts

ar
As-received
CC
combustion chamber
CH
chemical
D
destruction
F
fuel
i
index of components
j
index of streams
P
product
PH
physical
0
associated with the environment

Based on the developed model, any syngas can be used for combustion, provided that the heating value is sufcient to reach the
assumed ring temperature. In this study, the syngas is obtained
through gasication of bituminous coal and is conditioned using
also CO2 capture.
2. Modeling
The gas turbine is a highly complex system. Even at steady-state
conditions, CFD (Computational Fluid Dynamics) simulations solving differential equations, and other detailed simulations are necessary to approximate the real performance. In this study, the complex
system is approximated using global balances. The process simulations were undertaken using Aspen Plus Version 27 [16]. In this
software, the gas path was modeled using the RKS-BM (Redlich
KwongSoave with Boston Matthias Alpha function) [17] property
method. To manage all parameters, 19 calculators und six design
specications were used together with the Aspen software.
2.1. Determination of the inefciencies
In general, exergy destruction is caused by friction, mixing, heat
transfer and chemical reactions. In this paper, twelve processes are
considered that are associated with characteristic inefciencies in
a gas turbine system.













Compression.
Stoichiometric combustion.
Addition of excess air.
Convective cooling in vanes/blades.
Pressure drop (caused by the transport of working uids).
Expansion.
Mixing at different pressures.
Mixing at different temperatures.
Mixing at different compositions.
Heat loss.
Transport of shaft work.
Conversion of mechanical energy to electrical energy.

This approach is used to provide a comprehensible overview

among the gas turbine system components based on an exergy
analysis. Especially the mixing processes are subdivided into three

M. Sorgenfrei, G. Tsatsaronis / Energy Conversion and Management 107 (2016) 4351

major groups: First, the high-pressure stream is throttled by a

hypothetical throttling unit to the minimum inlet pressure of the
mixer. Then the temperature change represented by the difference
in physical exergies at isobaric conditions follows. Finally, the
effect of the difference in compositions represented by the difference in chemical exergies at isobaric and isothermal conditions is
calculated. The implementation of these inefciencies is presented
in the next subsection.
2.2. The gas turbine system
The cases evaluated here are further denoted by NGGT for the
gas turbine running on natural gas and SGT for the system running
on syngas. In both cases the ambient conditions are 15 C and
1.013 bar and the composition of the ambient air is 79% N2 and
21% O2.
The general model of the gas turbine system is presented in
Fig. 1. Generally, this system consists of an air compressor (AC), a
combustion chamber (CC) and a gas turbine (GT). In the following,
the model used in both cases is presented. Further assumptions are
presented in the next sections. The air compressor uses air at ambient conditions and the pressure ratio of 1.255 is kept constant for
all stages (13 stages). In this study, the polytropic efciency
includes the suction loss at the compressor inlet. The pressure of
the bleed air ows for cooling and sealing the gas turbine is
assumed to be at least 10% higher than the pressure at the inlet
of the gas turbine, but at least 1 bar [18]. The simulation considers
seven cooling and seven sealing ows. At the exit of the compressor, the air is accelerated to almost sonic speed and needs to be
throttled for combustion purposes. This pressure loss of the combustion chamber (6.5% of the inlet pressure) can be separated into
two parts: In general, the larger part should occur in the combustor
(T2) since this has a positive effect on the mixing and combustion
processes. The other part is the pressure loss in the diffuser (T1)
directly behind the compressor exit [19].
After the compressed air passes the diffuser, it is split into a
stoichiometric and an excess part. The stoichiometric air enters
the throttling unit (T2) representing the pressure losses caused
by the combustor swirler at the combustor inlet. The fuel gas is
throttled to the pressure of the combustor inlet and then

45

pre-mixed with the stoichiometric air. The following adiabatic

and stoichiometric combustion results in the adiabatic combustion
temperature at the combustor exit. In the next step, the combustion gas is mixed with the excess air. Prior to that, the excess air
is throttled to the pressure level of the combustion gas. A cooler
(HX) represents the total heat loss (0.5% of the LHVar) caused by
the high temperature level within the overall gas turbine system.
Finally, the cooling air used in the combustion chamber is mixed
into the main gas stream. The actual combustion process is highly
complex and strongly depends on the design used by the
manufacturer.
The ow diagram shown within dotted lines in the upper right
corner of Fig. 1 presents the detailed conguration of a turbine
stage according to Kail [18]. This design is used for all stages except
the last one, because no cooling is necessary due to the relatively
low temperature. A constant pressure ratio (about 0.5) among all
gas turbine stages is assumed. The pressure drop in the vane is
represented by a throttling process (the pressure loss is 3% of the
inlet pressure [18]). Since the rst three stages need to be cooled
by an air lm layer, to protect the materials against the high temperature of the main gas stream, the cooling of the vane follows.
Within every cooled vane or blade the heat transfer is separated
into two parts caused by convection and mixing.
To estimate the amount of convective heat transfer, it is
assumed that the cooling air of the rst and second stage exits
the vanes or blades at 600 C and in case of the third stage at
480 C [18]. Depending on the cooling design, this temperature
might be different. The temperature at the surface of the vanes
and blades is assumed to be 1040 C when TBC is used [20] and
in other cases the temperature is 950 C [21]. The sealing air is used
to prevent the main gas stream from passing the vane or blade
through the area between tip and housing. It is always mixed with
the main gas stream after the following rotor. This represents a
limited ability of the sealing air to produce work. Before the cooling or sealing air enters the main gas stream, it is throttled to the
lower pressure level of the main stream representing the mixing at
different pressures. The cooling air needed in the blades is then
mixed into the main stream. On the one hand, a part of this air produces work in the rotor but on the other hand, a part of the vane
cooling air does not produce work in the rotor. Following the

Fig. 1. Flow diagram of the gas turbine system.

46

M. Sorgenfrei, G. Tsatsaronis / Energy Conversion and Management 107 (2016) 4351

assumptions made by Kail [18], both effects approximately compensate each other. The sealing air of the blades is mixed into
the main gas stream after the throttle, which represents the pressure loss of the next stage vane. In case of the secondary air system,
it is difcult to separate the pressure loss into friction caused by
transport, and throttling caused by mixing.
A separate evaluation can be performed when developing a particular design and the pipe diameter, pipe length and air velocity
are known. Filters are not considered in this work. Further assumptions are shown in Table 1. The developed model potentially
enables engineers to perform a sensitivity analysis, for example,
on the pressure ratio or the working uid. Changing the ring temperature will need further modeling enhancements.
Table 1
Fixed and adjusted parameters from literature (case NGGT).
Parameter

Unit

Fixed
value

Air compressor
Number of stages [22]
Pressure ratio [2]
Polytropic efciency [23]

13
19.2
91.5

Combustion chamber
Pressure loss overall (DpCC) [24]
Pressure loss diffuser (average) [19]
Pressure loss swirler (average) [19]
Radiation loss [25]
Cooling air [18]

%
%
%
%
%

Gas turbine
Number of stages [22]
1. Stage cooling/sealing air [18]
2. Stage cooling/sealing air [18]
3. Stage cooling/sealing air [18]

% of inlet air
% of inlet air
% of inlet air

4
9.68/2.25
2.95/2.25
1.97/1.69

% of inlet air
%

1.12
52/48

56/44

44/56

950 (1040)

C
kg/s

625
820

612,9
815.4

% of LHVar
%
%

40
99.5
99

39.7

4. Stage sealing air [18]

1. Stage ratio vane/blade
cooling and sealing [24]
2. Stage ratio vane/blade
cooling and sealing [24]
3. Stage ratio vane/blade
cooling and sealing [24]
Surface temperature
vane/blade (TBC used) [20,21]
Exhaust temperature [2]
Exhaust mass ow [2]
Other
Overall efciency [2]
Mechanical efciency shaft [24]
Electrical efciency generator [24]

of DpCC
of DpCC
of LHVar
of inlet air

6.5
35
65
0.5
12.4

Adjusted
value

9.92

8.71/2.03
2.66/2.03
1.77/
1.52
1.01

2.3. The case with natural gas (NGGT)

The gas turbine running on natural gas is based on data published by Siemens according to the state-of-the-art gas turbine
Siemens SGT5-8000H, which is the largest operating gas turbine
in the world. The world record test run, producing 578 MW and
having an overall net efciency of 60.75% was performed at the
combined-cycle power station Ulrich Hartmann in Germany in
the spring of 2011. In the combined-cycle conguration the turbine
is only scaled [2,22].
The rst and second stages of the gas turbine use thermal barrier coating (TBC) to protect the materials from the high temperature of the gas stream [2]. The mole composition of natural gas is
93.1% CH4, 3.2% C2H6, 1.6% N2, 1% CO2, 0.7% C3H8 and 0.4 C4H10
[26] and the calculated lower heating value LHVar is 47.19 MJ/kg.
The natural gas enters the system at 20 bar and ambient temperature. Table 1 presents the xed and adjusted parameters assumed
in the NGGT case. The demand for cooling and sealing air among
the gas turbine stages is taken from the modeling of the smaller
and older gas turbine SGT5-4000F [18]. These values are adjusted
by using 90% of the demand except the demand of cooling air used
in the combustion chamber, which is 80%. This adjustment is based
on publications by Siemens (see Table 1).
In order to provide a convenient overview, Fig. 2 presents the
amount of air ows entering the gas turbine related to the inlet
air ow of the compressor for both cases. The rst bar represents
the cooling and the second bar the sealing air used among the
gas turbine stages. Within stage four, no cooling air is needed. In
addition, the temperature of the main gas stream entering the
vanes and blades is shown on the secondary axis. The overall cooling and sealing demand amounts to 19.7% in case of the turbine
and 9.9% in case of the combustion chamber.
Fig. 3 presents the T,s-diagram for the NGGT case, the detailed
model presented in Fig. 1, and the three-component reference
model according to ISO 2314. The shown entropy is the entropy
at the given point multiplied with the ratio of the mass ow at
the given point and the one at the air compressor inlet. On the left
side, the adiabatic combustion temperature is presented at
2315 C. On the right side, the state variables of the rst turbine
stage are presented in detail. After the stoichiometric combustion,
the combustion gas is mixed adiabatically with the excess air. The
gas is then cooled at constant pressure and mixed with the cooling
air of the vane. Subsequently, the gas is expanded in the turbine
followed by a mixing process. Again, the gas is cooled at constant
pressure and the cooling air of the blade is then mixed into the
main stream. Finally this stream enters the next turbine stage.

1600
NGGT

1400

SGT

1200
temperature

1000

800
600

400
1

200

0
vane 1

blade 1

vane 2

blade 2

vane 3

blade 3

stage 4

Fig. 2. Temperature and cooling (left)/sealing (right) air of the turbine stages.

temperature [C]

cooling/sealing air to inlet air [%]

47

M. Sorgenfrei, G. Tsatsaronis / Energy Conversion and Management 107 (2016) 4351

1600
detail

temperature [C]

temperature [C]

1000
500

0.3

0.6

0.9

1.2

ISO

1500

1500

0
0.0

detail

ISO

2000

1.5

1400
1300
1200
1100
1.05

1.09

entropy [kJ/kgK]

1.13

1.17

1.21

1.25

entropy [kJ/kgK]

Fig. 3. T,s-diagram of the gas turbine system (NGGT case): (left) overall, (right) rst turbine stage.

The stoichiometric air-to-fuel ratio in the combustion chamber

is 1.79, and the one of the overall system amounts to 2.54. The COT
(Combustion Outlet Temperature) becomes 1490 C and the isentropic efciencies of the blades in each stage are respectively
90.5%, 91%, 91.5% and 92%. According to ISO 2314, the turbine inlet
temperature TITISO is calculated to be 1309 C and the isentropic
efciency of the compressor and turbine are 88.2% and 87.9%,
respectively.
2.4. The case with syngas (SGT)
Based on the characteristic parameters found in the simulation
of the NGGT case, a gas turbine system running on syngas was
simulated. Syngas is produced by common gasication and cleaning units of an IGCC plant. At rst, an oxygen-blown, entrainedow Shell gasier produces raw syngas through gasication of
Illinois #6 coal [26]. The raw gas is then cooled by a gas quench
and convective coolers producing steam. A two-stage water gas
shift reactor converts carbon monoxide to carbon dioxide by using
steam and producing hydrogen. The carbon dioxide and hydrogen
sulde are captured in a Selexol acid gas removal unit. Finally a
saturator dilutes and preheats the cleaned syngas. The mole
composition of the syngas amounts to 80.6% H2, 2.8% CO, 12.5%
H2O, 4.1% N2 and on the balance CH4 and CO2. The resulting temperature is 145.1 C und the pressure is 34.1 bar [3]. The LHVar is
calculated to 34.9 MJ/kg. Until the year 2010, the maximum content of hydrogen by volume used in F-class operation was 45%
and in industrial operation up to 95% [27]. Other IGCC concepts
integrate the gas turbine system with the air separation unit. In
this case, the compressor size increases to supply pressurized air.
The design of the SGT case is similar to that of the NGGT case
presented in Fig. 1. The air compressor uses the same pressure
ratios and isentropic efciencies. The assumptions for the combustion chamber are similar in both cases. In case of using high-hydrogen-content fuels, the high ame speed, the high ame
temperature and the wide ammability of hydrogen, along with
the low ignition energy and low density may cause blowout and
ashback. A lot of research focuses on modifying diffusion as well
as pre-mixed combustors for a low-emission operation ensuring
high reliability, availability and maintainability [28]. Regarding
the gas turbine, all temperatures of the main gas stream from
the COT to the nal exit temperature were kept constant; this corresponds to the same material limitations assumed in the NGGT
case. Hence, the isentropic efciencies of the rotor stages are kept
constant. To keep the temperature at the outlet of each mixing
process constant, the ow rates of the cooling and sealing air are
adjusted. The exit temperature of the cooling air after the
convective heat transfer in the blades or vanes is assumed to
remain constant at 600 C and 480 C respectively.

The overall cooling and sealing demand of the turbine amounts

to 20.4% and in the case of the combustion chamber to 10.6%
related to the compressor inlet air. The overall demand increases
compared to the NGGT case due to a higher specic heat capacity
of the combustion gas generated by the combustion of syngas. The
adiabatic combustion temperature increases to 2366 C due to the
higher mass-based heating value of hydrogen as well as the higher
temperature of the fuel gas. The new stoichiometric air-to-fuel
ratio is 2.1 in the combustion chamber and 2.9 for the overall system. Furthermore, the TITISO increases to 1313 C. The overall net
efciency increases by 2.1% points to 41.8%.
3. System analysis
An exergy analysis was conducted to quantify the thermodynamic inefciencies [29]. Exergy is the maximum theoretical useful work obtainable as the system is brought into complete
thermodynamic equilibrium with the thermodynamic environment while the system interacts with this environment only [30].
The exergy rate E_ j of a stream of matter is given by the physical,
chemical, kinetic and potential exergy rate. The contribution of
kinetic and potential exergies is neglected in the following.

_ CH
E_ j E_ PH
j Ej

  

_
 
E_ PH
j n  h  h0  T 0 s  s0

E_ CH
n_ 
j

X
X
xi eCH
xi lnxi
i RT 0 
i

!
3

The calculation of the chemical exergy rate E_ CH

according to Eq.
j
(3) is only valid for a mixture of ideal gases. The chemical exergy of
the gas is weighted with the chemical exergy of liquid water by its
fractions if condensation occurs at ambient conditions. The model
of Szargut [31] is used as reference environment. In Aspen Plus,
the exergy rates were calculated by incorporating our own
FORTRAN subroutines into the ASPEN software. These routines
use procedures developed in connection with the THESIS software
[32] and the material properties from the ASPEN software. The
exergy destruction rate E_ D;k within the k-th component represents
inefciencies caused by chemical reaction, heat transfer, friction,
and mixing. It is calculated as the difference of the exergy rate of
fuel E_ F;k and the exergy rate of product E_ P;k .

E_ D;k E_ F;k  E_ P;k

yD;k E_ D;k =E_ F;tot

48

M. Sorgenfrei, G. Tsatsaronis / Energy Conversion and Management 107 (2016) 4351

The exergy destruction ratio yD,k represents a dimensionless

variable, in which the exergy destruction rate within the k-th component is related to the exergy rate of the total plant fuel E_ F;tot The
fuel for the total system used in the calculation of exergetic
efciencies is represented by the fossil fuel and air entering the
gas turbine system.

4. Results and discussion

In Section 2.1, the thermodynamic inefciencies of a gas turbine
system are presented. Based on the developed model, it is not possible to sub-divide the inefciencies expansion, mixing at different
temperatures and mixing at different pressures of a gas turbine
stage. The model does not consider the reaction level of the stages,
which is needed when applying an exergy analysis for the evaluation of these inefciencies. However, the energetic analysis is not
affected when using isentropic efciencies. For further investigations a more complex model is needed, since the reaction level
has an effect on the demand of cooling air. The exergy of the
exhaust gas represents the exergy loss of the overall gas turbine
system, and is not further discussed. In the following, some values
of the NGGT case are presented.
Fig. 4 presents the hierarchy of inefciencies illustrated in
descending order for the NGGT case exergy destruction ratio. The
values are presented in Tables 2 and 3. The largest inefciency in
both cases is caused by the stoichiometric combustion (18.4%)
since chemical reactions usually cause the highest exergy destruction among all inefciencies. In the second position comes the
addition of excess air (4.8%) required because of the temperature
limitation of the gas turbine. The temperature decreases by about
825 C in the NGGT case to protect the rst vane material. The
grouping of expansion and mixing at different temperatures and
pressures (4.2%) takes the third position. First improvements can
be realized by (a) optimizing the vane and blade shape (isentropic
efciency), (b) increasing the convective heat transfer within the
vanes and blades by using enhanced TBC materials, and (c) improving the extraction points of the compressor. Mixing at different
compositions (2.3%) directly results from the combustion of the
particular fuel gas as well as the ow rates entering the ctitious
mixing unit. The rst one is almost unavoidable. Using enhanced
TBC materials reduces the cooling ow. The inefciency associated
with compression (1.8%) is in the fth position of importance,
whereas the exergy destruction caused by uid transport pressure
drop (0.8%) has a small value. In this study, the pressure drop is

generated by the combustor and the transport through the vanes.

The inefciencies associated with the transport of the cooling
and sealing air are included in the mixing processes. With respect
to convective cooling (0.4%), Fig. 4 shows that this represents a
minor inefciency because the major cooling part is provided by
the lm layer. The potential of applying steam cooling seems to
be relatively low if the absolute convective heat remains constant
and a lm layer is still used (hybrid cooling). Inefciencies caused
by heat loss (0.4%), transport of shaft work (0.2%) and conversion of
mechanical into electrical energy (0.4%) can be neglected compared to the others.
However, in the SGT case the hierarchy slightly changes. The
effect of expansion and mixing at different temperatures and pressures (5.6%) becomes more important than the addition of excess
air (5.3%). Thus, the amount of cooling and sealing air increases
to achieve the same mixing-unit outlet temperature compared to
the NGGT case. This is caused by the higher specic heat capacity
of the combustion gas in the SGT case. A general comparison of
both cases shows that the exergy destruction is usually higher in
the SGT case with the exception of the stoichiometric combustion
(12.3%) caused by the differences in the enthalpy and entropy of
reaction. The overall exergetic efciency of the NGGT case results
to 37.7% and the SGT case to 40.8%. Based on the approach used
in this study, the exergy destruction slightly shifts from the combustion chamber to the gas turbine (80%/14% to 78%/16%) when
using syngas.
For the NGGT case, the distribution among the combustion
chamber and the gas turbine stages is presented in Fig. 5. As presented on the left side, the largest part of exergy destruction
associated with mixing at different temperatures and pressures
occurs when the fuel gas is mixed into the stoichiometric air at
the combustor inlet (1.1%). The corresponding temperature difference is about 350 C. This strongly depends on the conditions of the
natural gas grid or rather on the boost compressor used for natural
gas handling. Since the difference in temperature and pressure
decreases in the turbine stages toward lower pressures, the associated exergy destruction decreases too. This behavior corresponds
to the effect of mixing at different compositions since the composition of the main gas and cooling ow converges after each mixing
process. Hence, the largest amount of exergy destruction occurs
within the cooling system of the combustion chamber (1%). The
effect of pressure drop caused by uid transfer through the vanes
increases at lower temperatures.
As introduced in Section 2.1, in general the exergy destruction is
caused by four sources. In the NGGT case, the amount of exergy

20
NGGT

exergy destruction ratio [%]

18
16
14
12
10

SGT

stoichiometric
combustion

addition of excess air

expansion
+ mixing at different temperatures
+ mixing at different pressures

convective cooling in
vane/blade
heat loss

8
6
4

mixing at different
compositions
compression
pressure drop

2
0

conversion of
mechanical to
electrical energy
transport of
shaft work

inefficiencies
Fig. 4. Hierarchy of inefciencies ordered by the exergy destruction of the NGGT case.

49

M. Sorgenfrei, G. Tsatsaronis / Energy Conversion and Management 107 (2016) 4351

exergy destruction ratio [%]

2.0
pressure drop
convecve cooling
mixing at dierent composions
expansion
+ mixing at dierent pressures
+ mixing at dierent temperatures

1.6

1.2

0.8

0.4

0.0
combustor

cooling
system

stage 1

stage 2

CC

stage 3

stage 4

GT

Fig. 5. Exergy destruction of some inefciencies among the combustion chamber (CC) and gas turbine (GT) of the NGGT case.

natural gas

35

99,5

turbine

100,9

combustion
chamber

generator

38,3
28,6

,5

compressor

air

net work

37,7

,4 conversion

27,8

flue gas

5,8 secondary air

1,8

,2

compression

,4

,2

pressure drop
heat loss
,2

,4
,5

mixing at diff. pres.

1,1

mixing at diff. temp.

1,8

pressure drop

expansion

1,4

4,8

shaft

convective cooling

mixing at diff. comp.

addition of excess air

,1
,9

18,4
1

stoichiometric combustion

mixing at diff. temp.

mixing at diff. comp.
mixing at diff. pres.

Fig. 6. Exergy ow diagram of the NGGT case (values related to the total fuel exergy).

Table 2
Exergy destruction ratios of the NGGT case in (%).
AC
Stage 1

CC
Stage 2

Stage 3

1.85
18.39
4.75
0.39
0.76
4.17
2.30
0.40
0.19
0.38

Shaft

Stage 1

Sum
Compression
Stoichiometric combustion
Addition of excess air
Convective cooling
Pressure drop
Expansion
Mixing at different pressures
Mixing at different temperatures
Mixing at different compositions
Heat loss
Transport of shaft work
Conversion of mechanical into
electrical energy

GT

Vane
1.17

0.42

Stage 2
Blade

Vane

Stage 3
Blade

Vane

Generator

Stage 4
Blade

0.26
18.39
4.75
0.10
0.23
0.17
1.15
1.39
0.40

0.34

0.09
0.12
0.87

0.06

0.21

0.12

0.05
0.13
0.74

0.04

0.07

0.07

0.05
0.14
0.67

0.07

0.14
0.57

0.04
0.19
0.38

50

M. Sorgenfrei, G. Tsatsaronis / Energy Conversion and Management 107 (2016) 4351

Table 3
Exergy destruction ratios of the SGT case in (%).
AC

CC

Stage 1

Stage 2

GT

Stage 3

Shaft

Stage 1
Vane

Compression
Stoichiometric combustion
Addition of excess air
Convective cooling
Pressure drop
Expansion
Mixing at different pressures
Mixing at different temperatures
Mixing at different compositions
Heat loss
Transport of shaft work
Conversion of mechanical into
electrical energy

1.79
12.34
5.33
0.40
0.75

1.13

0.41

Stage 2
Blade

Vane

Stage 3
Blade

Vane

Generator

Stage 4
Blade

0.25
12.34
5.33
0.10
0.21

5.61

0.90
1.90
1.58
0.41

2.61
0.41
0.21
0.41

0.10
0.12

0.06

0.86
0.38

0.25

0.05
0.13

0.04

0.74
0.13

0.09

0.05
0.14

0.15

0.66

0.55

0.07

0.08

0.03
0.21
0.41

syngas

33,9

99,5

combustion
chamber

turbine

103,6

generator

41,4
29,3

,5

air

compressor

,4

26,7

conversion
flue gas

5,8 secondary air

1,8

net work

40,8

,2

compression

,2

pressure drop
,4

,4 heat loss
,9
1,6
1,9
5,3

,5

mixing at diff. pres.

12,3

stoichiometric combustion

mixing at diff. temp.

,9
mixing at diff. pres.
1

mixing at diff. comp.

addition of excess air

pressure drop
,1

mixing at diff. temp.

shaft

convective cooling

mixing at diff. comp.

1,8

expansion

Fig. 7. Exergy ow diagram of the SGT case (values related to the total fuel exergy).

destruction associated with chemical reactions is calculated to

55.1%, heat transfer 2.4%, and the grouping of mixing and friction
42.5% excluding the electrical generator. Fig. 6 presents the exergy
ow diagram of the NGGT case to give an overview on the inefciencies of the gas turbine system. The same diagram for the SGT
case is presented in Fig. 7.
When comparing the ISO model to the detailed model considering cooling ows, we notice that the exergy destruction shifts
partly from the combustion chamber to the gas turbine. El-Masri
[12] presented the exergy destruction among the GT stages using
a three-stage model making a comparison difcult. Furthermore,
the ring temperature and pressure ratio of the compressor is signicantly lower (1277 C/14 compared to 1490 C/19.2).
Staudacher and Zeller [14] calculated the so-called irreversible
combustion, nozzle heat and chemical losses to add the most
to the exergy loss of an aircraft engine.
A comparison is not possible, because none of the assumptions
are presented. Tsatsaronis et al. [15] showed that the chemical
reactions are the most signicant sources of exergy destruction,

too. Preheating the reactants or decreasing the oxidant-to-fuel

ratio improves the combustion limited by the thermal properties
of the utilized materials. Splitting the exergy destruction into an
avoidable and an unavoidable part, reveals that about 95% of the
exergy destruction associated with combustion and about 76% of
the exergy destruction associated with mixing are unavoidable.
5. Conclusions
Based on an appropriate model, the exergy analysis is a suitable
tool to produce a detailed account of inefciencies of a gas turbine
system. Characteristic parameters based on the state-of-the-art gas
turbine SGT5-8000H were identied and used in the SGT case. As a
result of this study, it was found that the stoichiometric combustion and the addition of excess air of a gas turbine system running
on natural gas are the most important inefciencies. When the gas
turbine stages are modeled using the isentropic efciency, it is not
possible to sub-divide the exergy destruction associated with
expansion, mixing at different temperatures and mixing at

M. Sorgenfrei, G. Tsatsaronis / Energy Conversion and Management 107 (2016) 4351

different pressures since the reaction level is not included. This

grouping of inefciencies was calculated to be in the third position
directly prior to mixing at different compositions, which strongly
depends on the fuel gas conditions. The inefciencies associated
with compression and pressure drop caused by uid transport follow subsequently. The exergy destruction ratios are presented for a
gas turbine system running on natural gas and syngas. In the SGT
case, the specic heat capacity of the combustion gas stream
increases compared to the NGGT case. Assuming that the mixer
outlet temperature remains constant, the effect of mixing becomes
more important.
This study addresses the detailed evaluation of a gas turbine
system. Further investigations on improving this system could be
performed by a parameter variation or an advanced exergy analysis [33]. An economic analysis can be combined with this exergy
analysis in the so-called exergoeconomic analysis [34], rating the
cost of the exergy destructions. The exergy analysis presented in
this study can also be used at part-load conditions to help understanding the off-design behavior of the gas turbine system.
References
[1] World Energy Outlook 2013. International Energy Agency, Paris; 2013.
[2] Siemens AG. The SGT5-8000H proven in commercial operation. <http://www.
energy.siemens.com/mx/pool/hq/power-generation/gas-turbines/SGT5-8000H/
downloads/SGT5-8000H%20proven%20in%20commercial%20operation.pdf>;
2012.
[3] Sorgenfrei M, Tsatsaronis G. Exergetic assessment of a syngas-redox-based
IGCC plant for generating electricity. J Eng Gas Turbines Power 2013;136(3).
http://dx.doi.org/10.1115/1.402588.
[4] Ertesvag IS, Kvamsdal HM, Bolland O. Exergy analysis of a gas-turbine
combined-cycle power plant with precombustion CO2 capture. Energy
2005;30:539. http://dx.doi.org/10.1016/j.energy.2004.05.029.
[5] Cassetti G, Rocco MV, Colombo E. Exergy based methods for economic and risk
design optimization of energy systems: application to a gas turbine. Energy
2014;74:26979. http://dx.doi.org/10.1016/j.energy.2014.07.043.
[6] Acikkalp E, Aras H, Hepbasli A. Advanced exergy analysis of an electricitygenerating facility using natural gas. Energy Convers Manage 2014;82:14653.
http://dx.doi.org/10.1016/j.enconman.2014.03.006.
[7] Mamaghani A, Naja B, Shirazi A, Rinaldi F. Exergetic, economic, and
environmental evaluations and multi-objective optimization of a combined
molten carbonate fuel cell-gas turbine system. Appl Therm Eng 2015;77:111.
[8] Ghazikhani M, Khazaee I, Abdekhodaie E. Exergy analysis of gas turbine with
air bottoming cycle. Energy 2014;72:599607.
[9] Soltani S, Yari M, Mahmoudi SMS, Morosuk T, Rosen MA. Advanced exergy
analysis applied to an externally-red combined-cycle power plant integrated
with a biomass gasication unit. Energy 2013;59:77580.
[10] Ahmadi P, Dincer I, Rosen MA. Exergy, exergoeconomic and environmental
analyses and evolutionary algorithm based multi-objective optimization of
combined cycle power plants. Energy 2011;36:588698.
[11] Erlach B, Schmidt M, Tsatsaronis G. Comparison of carbon capture IGCC with
pre-combustion decarbonisation and with chemical-looping combustion.
Energy 2011;36:380415. http://dx.doi.org/10.1016/j.energy.2010.08.038.

51

[12] El-Masri MA. Exergy analysis of combined cycles: Part 1 air-cooled Braytoncycle gas turbines. J Eng Gas Turbines Power 1987;109:22836.
[13] Khodak EA, Romakhova GA. Thermodynamic analysis of air-cooled gas turbine
plants. J Eng Gas Turbines Power 2001;123:26570.
[14] Staudacher S, Zeller PW. Exergy analysis for the performance evaluation of
different setups of the secondary air system of aircraft gas turbines. ASME
turbo expo 2007: power for land, sea and air; 2007.
[15] Tsatsaronis G, Morosuk T, Koch D, Sorgenfrei M. Understanding the
thermodynamic
inefciencies
in
combustion
processes.
Energy
2013;62:311. http://dx.doi.org/10.1016/j.energy.2013.04.075.

[16] Aspen Plus . <http://www.aspentech.com/>.

[17] Soave G. Equilibrium constants for modied RedlichKwong equation-ofstate. Chem Eng Sci 1972;27:1196203.
[18] Kail C. Analyse von Kraftwerksprozessen mit Gasturbinen unter energetischen,
exergetischen
und
konomischen
Aspekten
(german).
Mnchen,
Germany: Technische Universitt Mnchen; 1998.
[19] Brunling
WJG.
Flugzeugtriebwerke
(German).
3rd
ed.
Berlin
Heidelberg: Springer; 2009. http://dx.doi.org/10.1007/978-3-540-76370-3.
[20] Dennis R. The gas turbine handbook. National Energy Technology Laboratory,
U.S. Department of Energy; 2006.
[21] Lechner C, Seume J. Stationre Gasturbinen (German). Berlin/
Heidelberg: Springer; 2010. http://dx.doi.org/10.1007/978-3-540-92788-4.
[22] Fischer WJ, Nag P. H-Class high performance siemens gas turbine (SGT-8000H
series). <http://www.energy.siemens.com/co/pool/hq/power-generation/gasturbines/SGT5-8000H/downloads/H%20class%20high%20performance.pdf>;
2011.
[23] Jonsson M, Bolland O, Bcker D, Rost M. Gas turbine cooling model for
evaluation of novel cycles. In: ECOS 2005: international conference on
efciency, cost, optimization, simulation and environmental impact of
energy systems. Trondheim, Norway; 2025 Jun, 2005.
[24] Cerri G, Botta F, Chennaoui L, Giovannelli A, Salvini C, Miglioli M, et al.
Description of the models adapted or developed ad hoc for the IGCC&CCS
plants. Low emission gas turbine technology for hydrogen-rich syngas (H2IGCC); 2011.
[25] Boyce MP. Gas turbine engineering handbook. 3rd ed. Gulf Professional
Publishing; 2006. ISBN 13: 978-0-7506-7846-9.
[26] Black J. Cost and performance baseline for fossil energy plants volume 1:
bituminous coal and natural gas to electricity. National Energy Technology
Laboratory; 2013. DOE/NETL-2010/1397.
[27] Renzenbrink W, Scholz MH. H2 gas turbine a stepping stone to CCS. In: World
hydrogen energy conference (WHEC); 2010.
[28] Smith I. Gas turbine technology for syngas/hydrogen in coal-based IGCC. IEA
Clean Coal Centre; 2009.
[29] Bejan A, Tsatsaronis G, Moran M. Thermal design and optimization. New
York: JohnWiley & Sons, Inc.; 1996.
[30] Tsatsaronis G. Denitions and nomenclature in exergy analysis and
exergoeconomics. Energy 2007;32:24953.
[31] Szargut J, Morrison D, Steward F. Exergy analysis of thermal, chemical and
metallurgical processes. Berlin: Springer; 1988.
[32] Eisermann W, Hasberg W, Tsatsaronis G. THESIS - Ein Rechenprogramm zur
Simulation und Entwicklung von Energieumwandlungsanlagen (german).
Brennstoff-Wrme-Kraft 1984;36:4551.
[33] Morozyuk T, Tsatsaronis G. Advanced exergy analysis for chemically reacting
systems application to a simple open gas-turbine system. IjoT 2009;12:
10511.
[34] Tsatsaronis G, Cziesla F. Thermoeconomics: encyclopedia of physical science
and technology. 16th ed. Academic Press; 2002. 659680.