ARTICLE IN PRESS

Energy 33 (2008) 1508– 1517

Contents lists available at ScienceDirect

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

Analysis and integration of fuel cell combined cycles for development

of low-carbon energy technologies
Petar Varbanov a,, Jiřı́ Klemeš b
a
Marie Curie ERG ESCHAINS, Research Institute of Chemical and Process Engineering, University of Pannonia, Egyetem u. 10, Veszprém H-8200, Hungary
b
EC Marie Curie Chair (EXC) ‘‘INEMAGLOW’’, Research Institute of Chemical Technology and Process Engineering, FIT,
University of Pannonia, Egyetem u. 10, Veszprém H-8200, Hungary

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

Article history: Integrated and combined cycles (ICC, CC) traditionally involve gas and steam turbines only. The paper
Received 28 February 2008 analyses the further integration of high-temperature fuel cells (FC) having high electrical efficiency
reaching up to 60% compared with 30–35% for most gas turbines. The previous research on FC hybrids
Keywords: indicates achieving high efficiencies and economic viability is possible. The ICC of various FC
Energy efficiency types—their performance and the potential for utilisation of renewables—are analysed considering also
High-temperature fuel cells power generation capacity and site heat integration context. Further research and development with
Combined cycle industrial relevance are outlined focusing on CO2 emissions reduction.
Power cycle integration & 2008 Elsevier Ltd. All rights reserved.
Total site heat integration
CO2 minimisation

1. Introduction [2] compared the integration of molten carbonate fuel cells

The integrated combined cycles (ICCs) traditionally involve
only gas and steam turbines. They can be broadened to the
integration of high-temperature FC having high electrical effi-
ciency reaching 40–60%, compared to 30–35% for most gas
turbines [1]. There are three main CO2 pathways through fuel-
based energy systems, including FC, with regard to the ambient:
recycling, build-up and sequestration (Fig. 1). The extent of the
currently developing climate change is influenced by the volume
fraction of CO2 in the atmosphere. The presented diagram in Fig. 1
points to three major ways of limiting CO2 emissions—improving
energy conversion efficiency, increasing the CO2 recycling via
biofuels and CO2 sequestration. There is an extensive research
aiming at efficiency improvement of FC systems. Karvountzi et al.
Abbreviations: A, HEN ‘‘A’’; AB, Afterburner; ABC, Afterburner cooling; AP, Air
preheat; B, HEN ‘‘B’’; BD, Blow down; BFW, Boiler feed water; CC, Composite curve;
CC, Combined cycle; CFP, Carbon foot print; CHP, Combined heat-and-power
generation; CNDIN, Condenser inlet; CNDOUT, Condenser outlet; CR, Condensate
return; DAF, De-aerator feed; DAG, De-aeration steam at generation pressure; DAO,
De-aerator outlet; DAS, De-aeration steam; DW, Demineralised water; EC,
European community; EC, Exhaust cooling; ES, Expansion steam; FC, Fuel cell;
FCCC, Fuel cell combined cycle; FP, Fuel preheat; GCC, Grand composite curve; GS,
Generated steam; GT, Gas turbine; HEN, Heat exchanger network; HRSG, Heat
recovery steam generator; HTFC, High-temperature fuel cell; ICC, Integrated
combined cycle; LTFC, Low-temperature fuel cell; MCFC, Molten carbonate fuel
cell; PEM FC, Proton exchange fuel cell; SG, Steam generation; SOFC, Solid oxide
fuel cell; ST, Steam turbine; V, Vaporiser; mGT, Micro gas turbine (i.e. with capacity
up to 100–200 kW).
 Corresponding author. Tel.: +36 88 424 483; fax: +36 88 428 275.
E-mail address: varbanov@dcs.uni-pannon.hu (P. Varbanov).
(MCFC) and solid oxide fuel cells (SOFC) into hybrid systems. Kurz
[3] emphasises on the choice of appropriate gas turbines (GT) for
given fuel cells, and Massardo and Bosio [4] study the MCFC
combinations with gas and steam turbines. A very promising
option is to integrate the FC with bottoming cycles to design
dedicated power generation or combined heat-and-power (CHP)
applications [5].
2. Efficiency of FC and combined cycles
2.1. Operating temperature and fuel cell efficiency
Most FCs use H2. One exception is the direct-methanol FC
reported by Toshiba [6]. The primary fuel—mostly natural gas or
biogas, undergoes an endothermic conversion consisting of
reforming and shift reactions to generate the required H2. High-
temperature FCs (HTFCs) allow heat integration between the fuel
conversion and the power generation processes. In contrast, low-
temperature FCs (LTFCs) do not, and additional fuel is burnt to
supply the heat for the main fuel conversion [7], resulting in
electrical efficiencies around 35% for LTFC against 41% for HTFC.
Similar estimates result from MCFC integration [5] (see Table 1).
2.2. Combinations with bottoming cycles
After ensuring maximum energy recovery within the FC,
integrated systems can make even better use of the fuel. Steam
and the gas turbines can be used for increasing the power output,

0360-5442/$ - see front matter & 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.energy.2008.04.014
 ARTICLE IN PRESS

P. Varbanov, J. Klemeš / Energy 33 (2008) 1508–1517 1509

Nomenclature R power-to-heat ratio () or (MW MW1) or

(kW kW1); R ¼ P/Q
A heat exchanger area (m2) T temperature (1C)
Cp specific heat capacity (kJ kg1 1C1) TBFW boiler feed water temperature (1C)
CP heat capacity flowrate (kW 1C1) TS supply temperature (1C)
ECHP CHP energy output of the FCCC system (kW) or (MW) TSat saturation temperature (1C)
m mass flowrate (kg s1) or (t h1) TSh superheat temperature (1C)
P pressure (bar), (atm) or (kPa) TT target temperature (1C)
Q heat flow (kW) or (MW) WSOFC power generated by a SOFC (kW) or (MW)
QEvap heat consumption rate for evaporating the BFW (kW) WST power generated by a steam turbine (kW) or (MW)
or (MW) WTotal total power generated by a FCCC system (kW) or
QFuel energy input to the system in the form of fuel (kW) or (MW)
(MW) DTMin minimum allowed temperature difference (C)
QPreheat heat consumption rate for preheating the BFW (kW)
or (MW) Greek symbols
QSH heat consumption rate for superheating the steam
(kW) or (MW) ZCHP CHP efficiency () or (%)
QTotal total heat consumption for generating steam (kW) or ZCHP,Max maximum possible CHP efficiency () or (%)
(MW) ZE electrical (power) efficiency () or (%)
QUH useful heat generated by the FCCC system (kW) or
(MW)

which can be supplemented by heat cogeneration. Only high-
temperature FCs are suitable for such integration, which combines
advantageously with the conclusion from the previous section.
Several integration options have been investigated in the
literature. A summary of the most interesting works is given in
Table 2. In the table, the rightmost column represents an estimate
of what would the CHP efficiency of the corresponding systems be
if the heat recovery from the main exhaust streams was
maximised.
2.2.2. Fuel cell–GT hybrids
The efficiency of FCCC with gas turbines has been investigated
elsewhere. Massardo and Bosio [4] studied the combination of
MCFC with gas turbines and the option of further adding steam
turbines. The systems achieve electrical efficiencies up to 69% at
power generation capacities of 10–15 MW. Uechi et al. [8]
investigated a SOFC+mGT (micro gas turbine) system with a total
power capacity of 30 kW (23.6 kW from the SOFC and 6.4 kW from
the mGT). Even this small system approaches 67% electrical
efficiency. Oyarzábal et al. [10], Lunghi and Ubertini [11]. Bedont

2.2.1. Fuel cell–steam cycle hybrids

Table 1
The simplest way for FC integration is with steam cycles [5]. MCFC properties from Varbanov et al. [5] for 2320 MW power generation
The sensitivity analysis in this work for a wide range of FC capital
costs indicates the potential benefits of fuel cell combined cycle Fuel for power generation (MW) 5002
(FCCC) systems. Using the year 2005 power prices and 1.255 $ h1 Additional fuel (no integration) (MW) 1610
MCFC efficiency (%) 35.09
they achieved prices as low as 40–47 $ MW1 h1
1 1 FCCC efficiency (%) 46.38
(32.65–38.37 h MW h ).

CO2
CO2 BUILD-UP

Fossil Fuels
Energy conversion
processes Power + Heat

Biofuels
CO2

CO2 recycling
Sunlight Sequestration

Fig. 1. CO2 pathways for energy systems.

 ARTICLE IN PRESS
1510 P. Varbanov, J. Klemeš / Energy 33 (2008) 1508–1517
Table 2
Sources on cycle integration of fuel cells
Source System/notes
Uechi et al. [8] SOFC+mGT. Integrated GT compressor
Gunes and Ellis [9] PEM FC. Residential CHP
Oyarzábal et al. [10] PEM FC+GT. Considers CHP
Lunghi and Ubertini [11] MCFC+GT. No cogeneration
Bedont et al. [12] MCFC+GT. Integrated GT compressor
Massardo and Bosio [4] MCFC+GT+ST. 1- and 2-level HRSG
Campanari [13] SOFC+mGT
et al. [12] and Campanari [13] also investigated several different
FCCC arrangements where the main distinctions were in the
connection of the GT heat input and the placement of the
compressors. The options can be classified as:
(a) Systems with indirectly heated gas turbines. They have gas–gas
heat exchangers recovering the heat from the FC exhaust. The
FC and the GT components also have separate air compressors.
(b) Systems with an integrated air compressor. The GT compressor
is used to raise the pressure of the inlet to the FC cathode
compartment (the air steam). After that, the stream passes
through a post-combustor and through the GT expander,
where it generates torque.
Option (a) has the advantage that the working pressures in the
FC and the GT are independent, allowing more flexibility. For
option (b) the GT pressure must be lower than that in the FC,
which results in lower compression ratios and in lower GT
efficiencies. However, in this case the very large and costly
gas–gas heat exchanger is avoided, which can also reduce physical
space requirements.
2.2.3. Fuel cell–GT–steam cycle hybrids
These systems have not been much investigated due to their
relative complexity and the small marginal efficiency increase
compared to fuel cell–GT hybrids. From the sources in Table 2,
only Massardo and Bosio [4] investigate such a system with a
100 kW MCFC. These systems achieve maximum electrical
efficiency of the MCFC+GT+ST system of around 67.4% and 69.1%
for the cases of integrating single-level and two-level steam
cycles, respectively.
3. Fuel options and renewable energy
3.1. Major trade-offs
The fuels for FC-based systems influence the electrical
efficiencies, carbon emission levels and plant economics. With
regard to emissions, using H2-rich feedstocks such as natural
gas is more advantageous since they generate much less
carbon emissions. Biofuels lower the emissions too. The fossil
fuels are still priced relatively lower even with the rise of the
crude oil prices above 120 $ bbl1. Even with the further increase
of the oil price, the main bottleneck of using biofuels will be their
sufficient availability. This is an important though frequently
overlooked factor. This also calls for higher energy conversion
efficiency.
A study of CH4–CO2 fuel compositions for using in SOFCs [14]
suggests that maximum efficiency is achieved at CH4 volume
fraction around 0.45 which falls within the usual range of biogas
compositions. The main reason given in that paper is that H2 is
ZE (%) ZCHP,Max (%)
66.5 93.0
31.0 80.0
39.0 73.0
59.2 59.2
59.7 83.5
69.1 82.7
64.9 71.9
produced by dry reforming, where CO2 and CH4 are consumed in
equimolar quantities producing H2 and CO. Consequently, waste
treatment plants can be suitably equipped with SOFC-based units
to produce power and heat from biogas at top efficiency. Several
companies, including Siemens and General Electric [15], have also
started to develop coal-based FCs using coal synthesis gas as fuel.
Using syngas from combined biomass and coal gasification may
also be attractive.
3.2. Implications for carbon capture and sequestration
Burning biofuels is potentially carbon-neutral—see CO2 recy-
cling in Fig. 1. However, the overall carbon footprint analysis
reveals that actually some carbon is released into the ambient
[16]. The CFP is obviously considerably higher when using fossil
fuels and consequently CO2 capture and sequestration should be
considered.
FCs keep the path of the air stream apart from that of the fuel
and its products. Stoichiometrically the only anode-side products
are just CO2 and water. Certain residual amounts of the fuel are
also present in the FC anode exhaust. This prompts post-
combustion introducing a certain small amount of air into the
exhaust. Thus, FCs offer an opportunity for very efficient CO2
capture and subsequent sequestration. SOFC systems take this
advantage to the extreme since they can oxidise both H2 and CO
[8]. A very interesting research direction is the development of
cheaper SOFCs with maximum fuel utilisation, producing mix-
tures of water and CO2 only. This would potentially eliminate the
need for external CO2 capture units.
4. Application of FC-based energy conversion
4.1. Combined supply chains for energy, food and waste-
management
Increasing the share of renewables is highly dependent on the
ability to develop economically viable solutions. It is clear that
renewable energy production is in competition for primary
resources with food production. Food and other industries need
to manage large volumes of solid and liquid waste of organic
origin—e.g. manure, vegetable residues, black liquor, etc.
FCCC systems can be integrated into combined energy–food–
waste supply chains with other processes. See Fig. 2, as presented
by Beamon [17]. The benefits of the FC-based systems can come
from:
 Cost sharing between the biofuel and the other biomass-based
products.
 In some selected cases, waste streams can be used as fuel raw
materials which they cannot release directly into the environ-
ment. Using these resources for FC fuel generation would result
 ARTICLE IN PRESS

P. Varbanov, J. Klemeš / Energy 33 (2008) 1508–1517 1511

Agriculctural
Producers
Transportation
Food links
FCCC systems
Processing
Byproducts Treatment Fuels:
Waste Processes Biogas, H2,
Others Syngas, etc.
producing
organic waste

Fig. 2. Combined energy–food–waste supply chains.

in virtually negative fuel prices due to the fees to be charged 1200

for the waste treatment.
1000 Small driving forces
High capital cost
4.2. Types of applications and power-to-heat ratio 800

T (°C)
Another important indicator for evaluating the performance of 600
energy systems is the power-to-heat ratio (R). Generally, the
cogeneration efficiency decreases when the value of R is 400
increased. This is used to classify the demands and energy
200
conversion technologies. Energy users differ widely by the scale Hot Composite
Cold Composite
and the power-to-heat ratio (R) of the demands. Residential 0
applications feature daytime RDay410 and RNightE1. The R-values 0 10 20 30 40 50 60 70
of industrial energy demands also vary. An EC-project report [18]
Q (kW )
quotes values in the range 0.4–0.6. Grid supply power stations are
also a promising application, where district heating CHP variants Fig. 3. Composite curves for the heat integration of the 30 kW SOFC+GT system in
covering R 0.10–0.49 [19] are put at strong advantage by the Ref. [8].
legislation in most industrialised countries.
The cogeneration efficiencies for the reviewed systems 1000
are given in Table 2. They can serve a wide range of applications Larger driving forces
Lower capital cost
with essentially any practical value of R. For energy demands with CHP possible too
800
R41 (e.g. mechanical processing, grid-dedicated power plants), FC
hybrids can be applied directly. For smaller R some of the system
components such as the GT can be discarded. For very small values 600
T (°C)

of R, e.g. smaller than 0.2, a CHP plant with R40.2 may be

designed and the excess power can be sold to the grid, provided 400
this is contractually and physically possible.
An interesting direction is the design and operation of FC-based
200
CHP systems serving large industrial sites. In oil refineries, there are Hot CC
often large amounts of chemically low-quality hydrocarbon feed- Cold CC
stocks (currently burned) suitable for reforming and use as FC fuels. 0
0 1000 2000 3000 4000
Q (kW )
4.3. Heat integration options
Fig. 4. Composite curves for the MCFC heat-and-power integration [5].
The heat integration has been a well-developed methodology
for minimising the energy consumption, increasing the efficiency
and minimising emissions [20]. It has been described in detail smaller driving forces, which would tend to increase the
elsewhere [21,22], offering various options to properly integrate capital costs.
the ICCs. The recovery of thermal energy between the components (ii) For the MCFC+ST [5] combination, higher efficiency is still
in FC systems has been analysed for different arrangements. possible, but the driving forces are much larger which
Figs. 3 and 4 show the composite curves (CCs) [21] for two indicates potentially smaller capital costs for the heat
representative cases—integration of a SOFC with GT and that of a recovery.
MCFC with a steam cycle, respectively. CCs are a very useful tool
for representation of heat exchange systems and can be beneficial
for selecting the integration of the component utilities into hybrid To illustrate the potential for generating steam the grand
systems. Comparing the two cases in Figs. 3 and 4 leads to a composite curves (GCC) corresponding to the same SOFC+GT
couple of interesting conclusions: and MCFC systems are shown in Figs. 5 and 6. Fig. 6 indicates
that the MCFC+ST arrangement allows significant gene-
(i) In the SOFC+GT arrangement [8], the components are more ration of any level steam for heating to be used on-site or sold
tightly integrated. This leads to high efficiency, but also to for profit.
 ARTICLE IN PRESS

1512 P. Varbanov, J. Klemeš / Energy 33 (2008) 1508–1517

1200 5.2. Case study description

1000 The case study is based on data derived from the example
described by Chan and Ding [23]. This represents a SOFC
800
generating around 3000 kW of electrical power at 50.90%
T (°C)

600 efficiency. The main FC system operating conditions include:

400  pressure 1 atm (1.01325 bar);

 70% fuel utilisation rate;
200  the fuel and the air streams enter the FC at 900 and 700 1C.
0
0 5 10 15 The system flowsheet is shown in Fig. 7.
Q (kW )
5.3. Heat integration data and targeting
Fig. 5. Grand composite curve for the heat integration of the 30 kW SOFC+GT
system in Ref. [8].
After an analysis of the flowsheet in Fig. 7, a number of process
streams which need thermal change have been identified and
extracted for the heat integration study (Table 3). The reformer
operation has been extracted as a two-segment stream as its Cp
1000 and consequently CP ¼ C P  m should be linearised by at least two
segments (Reformer-1 and Reformer-2).
800 The heat integration problem identified is a threshold one [22],
where under DTMin ¼ 32.2 1C there is no need for a hot utility.
600 Above this value, there is a need for some hot utility. The heat
T (°C)

recovery targets for DTMin ¼ 30 1C are shown in Figs. 8 and 9. The

400 minimum cold utility required for this case is 1197.08 kW.
By analysing Figs. 8 and 9, it is obvious that there is a
200 significant thermodynamic potential for either heat cogeneration
(stream to stream heat recovery) or increased power generation
0 from the referenced system. Cooling water or air can readily be
0 500 1000 1500 2000 2500 used to satisfy the process stream cooling requirements, but this
Q (kW ) would generate only a heat waste.

Fig. 6. Grand composite curve for the MCFC heat-and-power integration [5].
5.4. Generating hot water

5. Case study: energy integration of a SOFC The main purpose of this case study is to investigate the
options and scope for integrating a SOFC with other energy
5.1. Definitions technologies. A number of heat exchanger networks (HENs) have
been generated for recovering maximum heat from the various
This case study analyses the energy efficiency of several process streams (Table 3). The simplest option is to use the excess
possible integrated system options measured in several ways. 1197.1 kW waste heat as hot water.
Firstly, the total power produced by the system can be written as: The cogeneration is considered for feeding hot water to
households located close to the power generation plant. The
W Total ¼ W SOFC þ W ST . (1) usual temperature of the hot water used in households is between
Additionally, the CHP energy output of the system is defined as 45 and 50 1C. To cover the heat losses during transportation of the
the sum of the total power and the useful heat outputs: water from the power plant to the users, the generation
requirement for hot water is set at 65 1C. The resulting HEN
ECHP ¼ W Total þ Q UH . (2)
topology A is shown in Fig. 10. In this case all the waste heat from
Two following efficiency definitions are used in the study— the fuel cell system is used for heating the water and neither hot
electrical and CHP efficiency. nor cold utility is needed.
The network in Fig. 10 manages to utilise the whole 1197.1 kW
5.1.1. Electrical (or power) efficiency targeted by the heat integration pinch analysis (see Figs. 8 and 9).
This is the fraction of the generated electrical power relative to This requires 140.3 m2 total heat exchange area. As a result, the
the heat flowrate input to the system in the form of fuel: overall efficiency of the combined CHP system (SOFC+HEN)
becomes 70.8%.
W Total
ZE ¼ . (3)
Q Fuel
5.5. Increased power generation using a steam cycle
5.1.2. CHP efficiency
This is the fraction of the total useful energy output relative to For designing the steam cycle generating steam at various
the energy input of the fuel: pressure levels coupled with a condensing steam turbine cycle has
been evaluated. The simplified flowsheet of the steam cycle is
given in Fig. 11. This arrangement can be modified by slightly
ECHP W þ Q UH raising the exhaust pressure of the steam turbine and using it for
ZCHP ¼ ¼ Total . (4)
Q Fuel Q Fuel district heating again.
 ARTICLE IN PRESS

P. Varbanov, J. Klemeš / Energy 33 (2008) 1508–1517 1513

3055.0 kW
1664.6 kW Power
Air
AP Heat loss 1639.6 kW
25.0 °C 700.0 °C SOFC
2.28 kg/s
985.0 °C
0.12 kg/s 2.77 kg/s
Reformate 900.0 °C
CH4 25 °C
173.6 kW
Afterburner Loss
Fuel heat
6001.7 kW 2677.0 kW
0.49 kg/s 850.0 °C Cooling
FP Reformer
100.0 °C
1000.4 kW 932.2 °C
Water/steam 1227.9 kW reaction
P=1 atm 403.4 kW heating
V
EC 3119.1 kW
303.6 kW
0.36 kg/s 130.0 °C
SG
25 °C
Cogenerated heat
(water or steam) –
varied in the analysis

Water (demineralised) Exhaust

Fig. 7. SOFC flowsheet for the case study.

Table 3 1000
Heat integration data for the case study

Stream Type Ts (1C) Tt (1C) Q (kW) CP (kW 1C1) 800

Air preheat Cold 25.00 700.00 1664.58 0.2466 600

T (°C)

Fuel preheat Cold 99.97 850.00 1000.37 0.1334

Reformer-1 Cold 850.00 850.01 1227.90 12279
Reformer-2 Cold 850.01 900.00 403.38 0.8069 400
Vaporiser Cold 25.00 99.97 303.64 0.4050
Exhaust cooling Hot 932.22 130.00 3119.09 0.3889
200
Afterburner cooling Hot 932.22 932.00 2677.00 1205.8

0
0 500 1000 1500 2000 2500 3000
1000 Q (kW)

Fig. 9. Grand composite curve for the SOFC integration case study.
800

Several observations stem from the case study as discussed

600
T (°C)

below.

400
200
Hot CC
0 Fig. 13. It can be seen that, by increasing the steam generation
0 2000 4000
Q (kW )
Fig. 8. Composite curves for the SOFC integration case study.
The steam specifications are tracked by the corresponding
saturation temperatures (Table 4). All resulting HEN topologies for
the steam generation case are the same. As an example, the
network for the 120 bar steam (TSAT ¼ 324.68 1C) is shown in
Fig. 12, designated as Network B, where the main difference from
Network A is the size of the heat exchanger area. The performance
of all networks of type B (those with steam generation) is given in
Table 5.
5.5.1. Observation 1: efficiency trends
The electrical (power) and CHP efficiencies feature opposite trends
Cold CC with increasing the steam generation pressure. These are plotted in
6000 pressure, the electrical efficiency increases and the CHP efficiency
decreases simultaneously. This can be easily explained using
thermodynamic reasoning. At fixed steam exhaust pressure, by
increasing the inlet one, the driving force for energy conver-
sion—the pressure differential between the steam turbine inter-
faces, grows. It is clear that the larger this driving force is, the
larger is the potential to generate power and at the same time the
energy dissipation (losses) also increases due to the increasing
‘‘irreversibility’’ of the process.
The overall energy efficiencies of the system are bound by
the performance of two extreme designs. This is shown again in
Fig. 13 where the SOFC standalone efficiency is plotted as a
baseline serving as a lower bound on the efficiency. Similarly, the
CHP efficiency of the integrated system for hot water generation
 ARTICLE IN PRESS

1514 P. Varbanov, J. Klemeš / Energy 33 (2008) 1508–1517

Fig. 10. HEN topology A—for generating hot water.

HEN - steam
BD: Blowdown water
generation
Boiler (ZERO)
feedwater at
GS: Generated steam
generation
pressure ES: Steam for expansion
at generation pressure
DAG: Deaeration
steam
at generation Steam
BFW: Turbine CNDIN : Condenser inlet
pressure
Boiler P = 0.12 bara
Feed-
Water Condenser

DAS: CNDOUT: Condenser outlet

Deaeration P = 0.12 bara
DAO: DV: Deaerator vent
Deaerator steam
P = 1.2 bara P = 1.2 bara CR: Condensate
Outlet
return
P = 1.2 bara
P = 1 atm =
1.01325 bara
Deaerator DW:
DAF: Deaerator feed Demineralised
P = 1.2 bara water

Fig. 11. Steam cycle topology for increasing power generation.

forms an upper bound. While both bounds are quite obvious in
the light of the laws of thermodynamics, their significance for
future design of FCCC integrated systems needs to be outlined.
Identifying the bounds would allow establishing targets to
determine the thermodynamic capabilities of this kind of systems
and might provide the engineer a tool for judging the potential
technological, resource and economic viability of the systems.
There are two very important features of the system, both
considered to be resulting from the high temperature of the
SOFC—poor utilisation of the SOFC exhaust temperature potential
and the need for potentially expensive heat exchangers.
5.5.2. Observation 2: poor utilisation of the SOFC exhaust
temperature potential
Another important insight into the thermodynamics of the
CHP system can be gained from Fig. 14. There the grand composite
curves for thee of the investigated systems are superimposed to
 ARTICLE IN PRESS

P. Varbanov, J. Klemeš / Energy 33 (2008) 1508–1517 1515

Table 4
Parameters of the steam levels

TSAT (1C) P (bar) TSH (1C) TBFW (1C) QPreheat (kW) QEVAP (kW) QSH (kW) QTotal (kW)

110.00 1.43 310.00 110.00 - 1013.27 182.96 1196.23

120.00 1.99 320.00 110.00 19.12 993.58 183.53 1196.23
140.00 3.62 340.00 110.00 56.81 954.02 185.40 1196.23
160.00 6.18 360.00 110.00 93.89 913.93 188.41 1196.23
180.00 10.03 380.00 110.00 130.55 872.97 192.71 1196.23
200.00 15.55 400.00 110.00 167.03 830.72 198.47 1196.23
220.00 23.19 420.00 110.00 203.59 786.67 205.97 1196.23
240.00 33.47 440.00 110.00 240.55 740.13 215.54 1196.23
260.00 46.92 460.00 110.00 278.34 690.21 227.68 1196.23
300.00 85.88 500.00 110.00 358.79 574.56 262.88 1196.23
324.68 120.00 524.68 110.00 414.75 484.98 296.50 1196.23

Fig. 12. HEN topology B—for the case of steam generation.

Table 5
Parameters of the resulting HEN designs 70
CHP efficiency hot water

TSAT (1C) P (bar) A (m2) Steam generation (t h1)

Efficiency (%)

65
110.00 1.43 137.75 1.64
Power generation CHP
120.00 1.99 138.16 1.62
SOFC 51%
140.00 3.62 139.64 1.60 60 standalone
160.00 6.18 141.41 1.58 efficiency
180.00 10.03 142.64 1.56
200.00 15.55 143.80 1.54 55
220.00 23.19 144.90 1.52
240.00 33.47 145.94 1.51
260.00 46.92 146.46 1.50
300.00 85.88 158.86 1.47
50
324.68 120.00 149.59 1.46
0 5 10 15 20
Steam pressure (bar)

Fig. 13. Overall FCCC cycle efficiency.

track the change of the temperature–heat–flowrate profile of
the system with the evolving heat recovery temperatures. The 15.55 bar (200 1C saturation temperature), and the topmost
bottommost curve depicts the case of hot water generation, the one—generation of 120 bar (absolute pressure, saturation tem-
middle one shows steam generation at absolute pressure of perature 324.68 1C) steam. It is obvious that, when generating hot
 ARTICLE IN PRESS
1516 P. Varbanov, J. Klemeš / Energy 33 (2008) 1508–1517
1000
800
600
T (°°C)
Steam at
400 120 bar(a)
Steam at
200
15.55 bar(a)
Hot water 65 °C
0 0
0 1000 2000
Q (kW)
Fig. 14. Superimposed GCCs for different SOFC integration options.
water, the temperature potential of the SOFC exhaust is extremely
poorly utilised despite the excellent CHP efficiency. Even generat-
ing steam at 120 bar results in only modest utilisation.
The main conclusion from this observation is that energy
recovery from an SOFC or any other HTFC using HENs is suitable
mainly for a specific range of power-to-heat ratios (R). In the
current case study, the highest R ¼ 4.751 for generating 15.55 bar
steam, while the lowest is R ¼ 2.552 for generating hot water.
5.5.3. Observation 3: potentially expensive heat exchangers
Another consequence of the high FC exhaust temperature is
that the heat exchangers may need to be built of special materials
and/or with advanced designs. This may contribute to higher
capital costs and thus to be economically suitable only to
applications with high profit margins and/or large scales, where
the returns on the investment can be faster realised.
Two possible solutions can be suggested. One is to exploit the
high FC exhaust temperature directly by adding gas turbines as
discussed previously [2–4,10–13]. These machines already use
special materials and are intended for power generation, where
the profit margins are generally much higher than for heating or
CHP. Moreover, such an addition also fits nicely to the problem
with efficiently utilising the exhaust temperature potential from
the thermodynamic point of view.
5.5.4. Observation 4: trends in the steam cycle when varying the
steam pressure
The last observation concerns the performance of the steam
cycle alone (Fig. 15). The power generation from steam first
asymptotically increases, then features a peak and slowly
decreases, with increasing the steam pressure. In parallel, the
dryness of the turbine exhaust decreases monotonically and the
highest-pressure steam cycle, which is feasible, is the one with
15.55 bar (see Fig. 15). These trends can be explained with the
monotonic increase in the total energy extraction from the steam,
consuming increasing fractions of the steam superheat and part of
the steam condensation heat.
This trend reveals an important design limitation of systems
integrating HTFCs (or any other source of high-temperature waste
heat) with steam turbines—there is an upper limit on the pressure
differential of a single turbine stage. One possible solution, if even
larger power generation is required, is to either employ a multi-
level steam cycle with reheat/letdown and another is to switch to
HTFC+GT based hybrid cycles.
The former option (multistage steam power cycle) would
require additional steam turbine machines or a single multi-stage
turbine, which would be more expensive and of much higher scale
250 1.00
200 0.95
0.90
Power (MW)
Dryness, -
150 Minimum feasible dryness
0.85
100
0.80
50 0.75
Power generation Dryness Fraction
0 0.70
50 100 150
3000 Steam pressure (bar)
Fig. 15. Steam cycle performance.
than the SOFC considered here. Many combinations are possible.
For instance, a larger steam turbine plant can be combined with
more HTFCs [4] (e.g. 10–20 units), or possibly in the future HTFCs
will be capable of generating 20–30 MW of power, which would fit
the scale of larger industrial steam turbines.
The latter option (HTFC+GT) seems more attractive since mGTs
are already available commercially and they fit the present HTFC
prototypes by both the thermodynamic profile and the scale of the
capacity.
6. Conclusions and future work
Studies of the trends and the potential benefits of fuel cells
heat integration based on the Heat Integration (Pinch) Methodol-
ogy have been performed. It has been found that the focus should
be on high-temperature FC. Combining FC with either GT or ST is
very efficient. Integration with both bottoming cycles provides no
significant benefits in terms of overall efficiency.
An obvious observation is that lowering the FC cost while
preserving their high efficiency is needed. The emphasis should be
put on the CHP rather than electrical efficiency, which has the
potential to greatly increase the overall fuel utilisation efficiency.
Waste treatment and biogas plants can be suitable fuel
suppliers for FC-based CHP systems. Gasified biomass or coal
can be attractive too. With regard to fuel supply, the need for
combined energy–food–waste supply chains has been revealed.
Clean coal power plants should be based on SOFC with CO2
sequestration.
In terms of designing systems which both maximise the fuel
utilisation rate and are economically viable, several directions for
further research and development can be outlined:
 The considered energy systems are complex and heteroge-
neous process systems. Designing them would require tools
with a significant strength for solving combinatorially inten-
sive problems with heavy optimisation loads. A method
efficiently handling the combinatorial nature of such pro-
blems is the P-graph framework developed by Friedler
and co-workers for heat exchange network synthesis [24],
district energy system [25] and combined heat exchange and
separation [26].
 There is continuously a need for novel advanced design for
heat transfer and power generation equipment reducing their
cost. Currently special alloys and ceramic materials are used in
constructing the high-temperature devices—gas turbines and
SOFCs. A possible cost reduction solution can be to lower
the SOFC working temperatures down to 800 1C [27] and
 ARTICLE IN PRESS
P. Varbanov, J. Klemeš / Energy 33 (2008) 1508–1517
eventually lower, which would allow the use of less expensive
alloys for the heat exchangers, which defines a trade-off with
the system efficiency. On the one hand, the lower the SOFC
working temperature, the cheaper are the materials in general.
On the other hand, the lower the temperature, the smaller is
the potential for efficient integration of gas turbines. In this
case too, heat integration can offer solutions which optimise
the energy gain under the modified conditions.
 In parallel with the energy and the greenhouse effect, the
industrial significance of other resources—most notably
water—rapidly increases. The list of resources to simulta-
neously account for usually also includes fuel, chemicals,
labour, finances and, of course—time. All these resources are
available in limited amounts and they cannot be considered in
isolation. This calls for more sophisticated frameworks for
design and optimisation of all process systems, including
energy conversion facilities. A recent study by Zhelev and
Ridolfi [28] suggested one potential way to perform this task
by introducing the combination of the pinch analysis and a
unified representation of the various resources—the emergy.
Acknowledgements
The financial support from the EC projects Marie Curie Chair
(EXC) MEXC-CT-2003-042618 INEMAGLOW and Marie Curie
Reintegration Grant MERG-CT-2007/46579 ESCHAINS is gratefully
acknowledged.
References
[1] Gas Turbine World. Handbook 2000–2001, vol. 21. Southport, USA: Pequot
Publishing; 2001.
[2] Karvountzi GC, Price CM, Duby PF. Comparison of molten carbonate and solid
oxide fuel cells for integration in a hybrid system for cogeneration or tri-
generation. ASME, Adv Energy Syst Div AES 2004;44:139–50.
[3] Kurz R. Parameter optimization on combined gas turbine–fuel cell power
plants. ASME J Fuel Cell Sci Technol 2005;2:268–73.
[4] Massardo AL, Bosio B. Assessment of molten carbonate fuel cell models and
integration with gas and steam cycles. J Eng Gas Turbines Power 2002;124:
103–9.
[5] Varbanov P, Klemeš J, Shah RK, Shihn H. Power cycle integration and
efficiency increase of molten carbonate fuel cell systems. J Fuel Cell Sci
Technol 2006;3(4):375–83.
[6] Toshiba’s direct methanol fuel cell officially certified as world’s smallest by
Guinness World Records. TOSHIBA. /uk.computers.toshiba-europe.com/
cgi-bin/ToshibaCSG/news_article.jsp?service=UK&ID=0000005758S; 2006
[24.12.2007].
[7] Wendt H. Electrical efficiencies of methane fired, high- and low-temperature
fuel cell power plants. J Appl Electrochem 2005;35:1277–82.
1517
[8] Uechi H, Kimijima S, Kasagi N. Cycle analysis of gas turbine–fuel cell cycle
hybrid micro generation system. J Eng Gas Turbines Power (Trans ASME)
2004;126:755–62.
[9] Gunes MB, Ellis MW. Evaluation of energy, environmental, and economic
characteristics of fuel cell combined heat and power systems for residential
applications. J Energy Resour Technol (Trans ASME) 2003;125:208–20.
[10] Oyarzábal B, Ellis MW, Von Spakovsky MR. Development of thermodynamic,
geometric, and economic models for use in the optimal synthesis/design of a
PEM fuel cell cogeneration system for multi-unit residential applications.
J Energy Resour Technol (Trans ASME) 2004;126:21–9.
[11] Lunghi P, Ubertini S. Efficiency upgrading of an ambient pressure molten
carbonate fuel cell plant through the introduction of an indirect heated gas
turbine. J Eng Gas Turbines Power (Trans ASME) 2002;124:858–66.
[12] Bedont P, Grillo O, Massardo AF. Off-design performance analysis of a hybrid
system based on an existing molten fuel cell stack. J Eng Gas Turbines Power
(Trans ASME) 2003;125:986–93.
[13] Campanari S. Full load and part-load performance prediction for integrated
SOFC and microturbine systems. J Eng Gas Turbines Power (Trans ASME)
2000;122:239–46.
[14] Stainforth J, Omerod RM. Implications for using biogas as a fuel source for
solid oxide fuel cells: internal dry reforming in a small tubular solid oxide fuel
cell. Catal Lett 2002;81(1/2):19–23.
[15] Future Fuel Cells R&D. (US–DOE). /www.fossil.energy.gov/programs/
powersystems/fuelcells/S; 2006 [24.01.2008].
[16] Perry S, Klemeš J, Bulatov I. Integrating waste and renewable energy to reduce
the carbon footprint of locally integrated energy sectors. Energy, on line
doi:10.1016/j.energy.2008.03.008.
[17] Beamon BM. Designing the green supply chain. Logistics Inf Manage
1999;12(4):332–42.
[18] Neelis ML, Patel MK, Bach PW, Haije WG. Analysis of energy use and carbon
losses in the chemical and refinery industries. Project report (ECN-I-05-008),
Utrecht University, 2005.
[19] Sipilä K, Pursiheimo E. Small-scale biomass CHP plant and district heating.
/www.vtt.fi/inf/pdf/tiedotteet/2005/T2301.pdfS; 2005 [28.02.2008].
[20] Klemeš J, Dhole VR, Raissi K, Perry SJ, Puigjaner L. Targeting and design
methodology for reduction of fuel, power and CO2 on total sites. Appl
Thermal Eng 1997;7:993–1003.
[21] Linnhoff B, Townsend DW, Boland D, Hewitt GF, Thomas BEA, Guy AR, et al.
User guide on process integration for the efficient use of energy 1988. 1st ed.
Rugby: IChemE; 1994.
[22] Smith R. Chemical process design and integration. Wiley; 2005. 685pp.
[23] Chan SH, Ding OL. Simulation of a solid oxide fuel cell power system fed by
methane. Int J Hydrogen Energy 2005;30:167–79.
[24] Nagy AB, Adonyi R, Halasz L, Friedler F, Fan LT. Integrated synthesis of process
and heat exchanger networks: Algorithmic approach. Appl Thermal Eng
2001;21:1407–27.
[25] Weber C, Heckl I, Friedler F, Marechal F, Favrat D. Network synthesis for a
district energy system: a step towards sustainability. In: Marquardt W,
Pantelides C, editors. Computer aided chemical engineering, Elsevier,
16th European symposium on computer aided process engineering and
9th international symposium on process systems engineering, 2006.
p. 1869–74.
[26] Heckl I, Friedler F, Fan LT. Integrated synthesis of optimal separation and heat
exchanger networks involving separations based on various properties. Heat
Transfer Eng 2005;26(5):25–41.
[27] Fukiu T, Ohara S, Murata K, Yoshida H, Miura K, Inagaki T. Performance of
intermediate temperature solid oxide fuel cells with La(Sr)Ga(Mg)O3
electrolyte film. J Power Sources 2002;106(1/2):142–5.
[28] Zhelev TK, Ridolfi R. Energy recovery and environmental concerns addressed
through emergy-pinch analysis. Energy 2006;31(13):2486–98.