Carlo De Servi1

e-mail: [email protected]

Stefano Campanari
e-mail: [email protected]
Politecnico di Milano,
Department of Energy,
Via Lambruschini 4,
20156 Milano, Italy

Alessio Tizzanini
Enel Ingegneria e Ricerca SpA,
Via Mantova 24,
00198 Roma, Italy
e-mail: [email protected]

Claudio Pietra
Turboden s.r.l.,
Via Cernaia 10,
25124 Brescia, Italy

Enhancement of the Electrical

Efficiency of Commercial
Fuel Cell Units by Means
of an Organic Rankine Cycle:
A Case Study
Among the various fuel cell (FC) systems, molten carbonate fuel cells (MCFC) are nowadays one of the most promising technologies, thanks to the lower specific costs and a very
high electrical efficiency (net low heating value (LHV) electric efficiency in the range
45%50% at MWel scale using natural gas as fuel). Despite this high performance,
MCFC rejects to the ambient almost half of the fuel energy at about 350400  C. Waste
heat can be exploited in a recovery Rankine cycle unit, thereby enhancing the electric efficiency of the overall system. Due to the temperature of the heat source and the relatively
small power capacity of MCFC plants (from few hundred kWel to 10 MWel), steam Rankine cycle technology is uneconomical and less efficient compared to that of the organic
Rankine cycle (ORC). The objective of this work is to verify the practical feasibility of the
integration between a MCFC system (topping unit) and an ORC turbogenerator (bottoming unit). The potential benefits of the combined plant are assessed in relation to a commercial MCFC stack. In order to identify the most suitable working fluids for the ORC
system, organic substances are considered and compared. The figure of merit is the maximum net power of the overall system. Finally, the economical benefits of the integration
are determined by evaluating the levelized cost of electricity (LCOE) of the combined
plant, with respect to the standalone MCFC system. In order to assess the economic viability of the bottoming power unit, two cases are considered. In the first one, the ORC
power output is approximately 500 kWel; in the latter, about 1 MWel. Results show that
the proposed solution can increase the electrical power output and efficiency of the plant
by more than 10%, well exceeding 50% overall electrical efficiency. In addition, the
LCOE of the combined power plant is 8% lower than the standalone MCFC system.
[DOI: 10.1115/1.4023119]

Introduction

Fuel cells (FC) are a promising technology for distributed electricity production, especially for power applications in the few
hundred kWel to 10-MWel capacity range. They exhibit high electrical efficiency and low pollutants emissions, they can be applied
to combined heat and power generation (CHP), and they can use
natural gas as primary fuel as well as biogas (for instance, from
wastewater treatment) or fuel blends. Despite these figures of
merit, they have achieved limited penetration into the energy market, mainly due to their high specific costs compared to other conventional technologies. A possible way to improve power plant
economics consists of enhancing its electrical efficiency as much
as possible. Waste heat dissipated by the stack or the exhaust
gases can be exploited to generate additional electricity in an organic Rankine cycle (ORC) heat recovery system [1,2]. In this paper, the potential benefits of the integration between a fuel cell
power plant (topping unit) and an ORC genset (bottoming unit)
are assessed in relation to a specific commercial molten carbonate
fuel cell (MCFC) system [3]. This kind of fuel cell has been
selected due to its well-established performance, potential competitiveness, practical availability on the market, and an exhaust
gas temperature consistent with the use of a heat recovery cycle
[1,2,4].

The MCFC stack releases exhaust gases at a relatively low temperature, namely about 380  C. This heat source is particularly
suitable for recovery in a bottoming unit based on ORC technology. ORC power plants are nowadays more and more applied in
many fields as, for instance, in the exploitation of low enthalpy
geothermal sources [5] or biomass thermal conversion. In order to
identify the most suitable working fluids for the analyzed case
study, different candidate substances are considered and compared
in both subcritical and supercritical ORC configurations. The net
power of the bottoming system is maximized by optimizing its
thermodynamic parameters, the maximum pressure, and temperature of the cycle. The detailed simulations of the ORC are performed in a commercial process modeling tool [6], while an
optimization algorithm implemented in an external programming
environment [7] varies cycle parameters until the maximum value
of the objective function is reached.
Finally, the paper presents a preliminary economic analysis in
order to investigate the feasibility of the proposed solution.
Thanks to the modular features of the fuel cell system, two different power capacities of the combined plant are considered (500
kWel and 1 MWel, respectively), in order to assess the effect of
the bottoming unit capacity on the economics of the entire plant.

2
1

Corresponding author.
Contributed by the International Gas Turbine Institute of (IGTI) ASME for
publication in the JOURNAL OF ENGINEERING FOR GAS TURBINES AND POWER. Manuscript
received February 29, 2012; final manuscript received October 21, 2012; published
online March 18, 2013. Assoc. Editor: Piero Colonna.

MCFC Unit

In the frame of the various FC technologies, MCFCs feature

one of the lowest specific costs and a very high efficiency [8]. One
of the most experienced companies in this field claims that
MCFCs electricity costs are about 1.2 to 1.5 times the costs of

Journal of Engineering for Gas Turbines and Power

C 2013 by ASME
Copyright V

APRIL 2013, Vol. 135 / 042309-1

Downloaded From: http://gasturbinespower.asmedigitalcollection.asme.org/ on 08/08/2014 Terms of Use: http://asme.org/terms

Table 1

Main performance data of the MCFC unit [10]

MCFC system are equal to the average between the beginning of

life (BOL) and the end of life (EOL) conditions.

MCFC DFC3000 performance data

Power output @ BOL (beginning of life)
Power output @ EOL (end of life)
Efficiency (LHV)
Exhaust temperature
Exhaust flow
NOx emissions
SOx emissions
PM10
CO2

(kWel)
(kWel)
(%)
(  C)
(kg/h)
(g/MWh)
(g/MWh)
(g/MWh)
(kg/MWh)

2800
2520
47 6 2
379 6 10
16,600
4.5
0.045
0.009
445

conventional power systems, such as reciprocating engines or gas

turbines, depending upon the price of fuel [9]. However, in the
MWel capacity range, MCFCs are about 10 times cleaner in terms
of NOx emissions versus competitive conventional technologies,
and they are the most fuel-efficient (net LHV electric efficiency in
the range 45%50%) and lowest CO2 emission fossil-fuelled
power generation technology available on the market [9]. For
these reasons, MCFCs do qualify in many countries for government support grants and subsidies to encourage ultraclean power
generation [9].
MCFC units are already commercially available, even if they
have not yet achieved significant penetration into the energy market, mainly due to their specific costs (25002800 e/kWel for the
MWel scale [9]), substantially higher than conventional competitive technologies. In this study, the MCFC unit is modeled according to the performance of a commercial 2.8-MWel MCFC power
plant [10], whose data are summarized in Table 1.
Even if the MCFC system exhibits a high efficiency, almost
half of the fuel energy is rejected to the ambient. Generally, in
order to improve the fuel cell power plant economy, thermal
energy in the exhaust gases is recovered for CHP. However, from
an economical point of view, it can be more attractive to recover
waste heat in a bottoming power unit [1,2,4]. Due to the temperature of the exhaust gases (380 6 10  C) and the current limited
power capacity of the FC plant, this kind of heat source is particularly suitable for recovery in an ORC system. Moreover, the integration of the MCFC unit with the ORC turbogenerator requires
only simple modifications to the plant layout: the addition of a
counter-flow heat exchanger to heat the ORC working fluid. In
principle, the implementation of the ORC turbogenerator can be
designed also as a plant retrofit, which can also be technically and
economically attractive.
Temperature and composition of exhaust gases vary during the
useful life of the MCFC unit due to changes in efficiency of the
FC system. The power output of a MCFC stack typically suffers
from a progressive performance decay. This reduction is almost
linear over time and is periodically recovered with a maintenance
procedure based on stack replacement [3] (every five years, as
generally proposed by manufacturers). Therefore, for the purpose
of this study, it is possible to model the FC plant by referring to
mean values of composition, temperature, and mass flow rate of
the exhaust gases. As Table 2 shows, the assumed data for the
Table 2 Exhaust flow composition, mass flow rate, and temperature for the MCFC unit [9]
Components (mol %)
CO2
H2O
N2
O2
Total
Mass flow rate
Temperature (  C)

BOL

EOL

Assumed value

4.5%
18.8%
67.5%
9.1%
100.0%
4.61 kg/s
369  C

4.1%
17.3%
68.5%
10.2%
100.0%
5.07 kg/s
390  C

4.3%
18.1%
68.0%
9.6%
100.0%
4.84 kg/s
379  C

042309-2 / Vol. 135, APRIL 2013

3 Thermodynamic Design of the Heat Recovery Cycle

and Working Fluid Selection
One of the main steps in designing ORC processes is the selection of the working fluid. Since the exhaust gases are released
from the MCFC unit at about 380  C, the heat recovery cycle is
classified as a high temperature ORC [11]. Extensive surveys
about this kind of application have been already presented in literature: Angelino and Invernizzi [12] considered aromatics hydrocarbons, siloxanes, perfluorobenzene, and allied compounds for
space power cycles; Angelino and Colonna di Paliano studied, in
addition to the previous substances, the performance of alkanes
[1] and siloxanes mixtures [2], and Drescher and Bruggemann
[13] suggested the exploitation of alkylbenzenes in biomass CHP
plants.
3.1 Thermal Stability. Unlike water, organic substances
undergo chemical decomposition at high temperature [14]. For
this reason, fluids suitable for the analyzed application must feature a high level of thermal stability, and the maximum operating
temperature of the thermodynamic cycle is limited to preserve the
useful life of the ORC working medium. The temperature threshold, below which thermal degradation occurs at an acceptable
rate, should be assessed by experimental investigation. Testing
equipments should reproduce realistic operating conditions, since
fluid degradation is a complex irreversible phenomenon influenced by several aspects, such as the type of containing materials,
the residence time, and the nature and quantity of impurities
(moisture, lubricants, or air) [12,15]. However, such detailed data
only exist for a limited number of fluids. For the others, available
information is limited to laboratory tests, the results of which are
sometimes ambiguous [12,16]. For example, Andersen and Bruno
[17] found that toluene decomposes at 315  C with a rate higher
than that of n-pentane. On the contrary, it was successfully tested
in a dynamic test loop at 400  C [18], and it is currently used in a
power cycle with a turbine inlet temperature of 325  C [19]. These
discrepancies are generally due to differences in test conditions
that make the data retrieved in literature not strictly comparable
[12]. In general, from literature sources [12,16,17], it can be concluded that:

for similar substances (i.e., with the same chemical structure),

the thermal stability gets worse with fluids having higher molecular weight
branched molecules are less thermally stable than straight
chains, and ring structures are the most thermally stable (e.g.,
benzene)
thermal stability of organic fluids is negatively influenced by
the presence of air, moisture, or other impurities in the
system

For the considered case study, the constraint on the maximum

temperature limits the candidate working fluids to the list reported
in Table 3. Four classes of substances can be recognized: aromatic
hydrocarbons, alkanes, siloxanes, and hydrofluorocarbons. Most
of these fluids show good thermal stability at least up to 270  C.
The stability thresholds (Tstab) of Table 3 are assessed on the
basis of specific experimental investigation [14,18,2026], if
available, or of technical papers about real applications [19,27].
For thermal stability of n-butane, n-hexane, cyclobutane, and D3,
no reliable references are available in literature. Therefore, as a
first approximation, these substances are supposed to behave at
high temperature, as other fluids with similar molecular structure:
n-butane and n-hexane as n-pentane, cyclobutane as cyclopentane,
and D3 as the other siloxanes. It is also necessary to clarify that
the threshold temperature associated with cyclohexane has only a
qualitative validity. According to Ref. [24], at 290  C cyclohexane
Transactions of the ASME

Downloaded From: http://gasturbinespower.asmedigitalcollection.asme.org/ on 08/08/2014 Terms of Use: http://asme.org/terms

Table 3 Working fluids for high temperature ORC. For each fluid, the table shows the CAS number, the critical temperature, TC,
and pressure, PC, the normal boiling temperature, TB, and the molecular weight, MW, the thermal stability threshold, Tstab, and the
global warming potential, GWP.
CAS number

TC (  C)

PC (bar)

TB (  C)

MW (kg/kmol)

Tstab (  C)

GWP

Aromatic hydrocarbons
Benzene
Toluene
p-Xylene
o-Xylene
m-Xylene
Ethylbenzene
Propylbenzene
Buthylbenzene

71-43-2
108-88-3
106-42-3
95-47-6
108-38-3
100-41-4
103-65-1
104-51-8

288.87
318.60
343.04
357.28
343.70
343.97
365.14
387.33

49.06
41.26
35.26
37.45
35.40
36.16
32.01
28.87

80.07
110.60
138.31
144.38
139.07
136.17
159.20
183.26

78.11
92.14
106.17
106.17
106.17
106.17
120.19
134.22

> 350, [17,24]

> 350, [18,19,25]
300, [24]
300, [24]
300, [24]
< 290, [24]
n.a.
n.a.

Linear alkanes
n-Butane
n-Pentane
n-Hexane

106-97-8
109-66-0
110-54-3

151.98
196.55
234.67

37.96
33.70
30.34

0.49
36.60
68.71

58.12
72.15
86.18

270a
270, [27]
270a

20
20
n.a.

Cyclic alkanes
Cyclobutane
Cyclopentane

287-23-0
287-92-3

186.85
238.54

48.40
45.15

12.53
49.25

56.11
70.13

300b
300, [20]

n.a.
11

Substance

Cyclohexane
Linear siloxanes
MM
MDM
MD2M

110-82-7

280.49

40.75

80.74

84.16

< 290, [24]

n.a.

107-46-0
107-51-7
141-62-8

245.60
290.94
326.25

19.39
14.15
12.27

100.25
152.51
194.36

162.38
236.53
310.69

> 350, [22]

> 350, [23,26]
> 350, [23]

Cyclic siloxanes
D3
D4
D5

541-05-9
556-67-2
541-02-6

281.00
313.35
346.00

16.60
13.32
11.60

135.09
175.35
210.90

222.46
296.62
370.77

> 350c
> 350, [22,26]
> 350, [22]

Hydrofluorocarbons HFC
R125
354-33-6
R134a
811-97-2
R143a
420-46-2
R227a
431-89-0
R236fa
690-39-1
R245fa
460-73-1

66.02
101.06
72.71
101.75
124.92
154.10

36.18
40.59
37.61
29.25
32.00
36.51

48.09
26.07
47.24
16.34
1.44
15.14

120.02
102.03
84.04
170.03
152.04
134.05

> 350, [21]

> 350, [21]
350, [14]
> 350, [14]
> 350, [14]
260300, [14,15]

3420
1370
4180
3580
9820
1050

Since no detailed data are available in the literature regarding n-butane and n-hexane, for these fluids, the same thermal stability threshold of n-pentane
is assumed.
b
Since no detailed data are available in the literature for cyclobutane, the same thermal stability threshold of cyclopentane is assumed.
c
D3 thermal stability has been assumed equal to that of the other siloxanes.

exhibits only a marginal stability, not adequate for industrial

applications. Nevertheless, Tstab of cyclohexane is assumed equal
to 290  C, since (i) Ref. [24] does not describe how such a result
has been obtained and (ii) there are other laboratory tests proving
that a molecule of the same class, cyclopentane, features good
thermal stability up to 300  C [20], provided that no interaction
with air occurs.
Looking at Table 3, it is important to consider that some of the
reported values of Tstab are assessed in particular experimental
conditions, for instance, by considering pure fluids or test lengths
limited to few hours [14]. For this reason and also for precautionary considerations, these values can be higher than those applicable to a real application. Thermodynamic properties of the fluids
of Table 3 are retrieved from Refs. [28] and [29], while the global
warming potential (GWP) is taken from Ref. [30].
3.2 Environmental Impact and Safety Issues. Candidate
substances suitable for ORC power systems should also feature a
limited environmental impact. In order to evaluate this aspect,
some parameters are taken into account, such as ozone depletion
potential (ODP) and GWP, ASHRAE safety classification or
occupational exposure limit (OEL), and the flammability level
[3133]. In the present study, the following criteria are applied:

Fluids with ODP higher than zero have to be discarded, because

they will be phased out over the next few years. For this reason,
hydrochlorofluorocarbons (HCFC) are not examined.

Journal of Engineering for Gas Turbines and Power

Substances with a GWP greater than 1370 (i.e., the GWP of

R-134a, one of the most used fluids in refrigeration applications) feature an unacceptable environmental impact. Note
that R-134a will be phased out of automotive airconditioning applications in Europe due to its GWP by 2017.
All the substances listed in Table 3, except benzene, present
reasonably low toxicity characteristics.2
The flammability hazard posed by each fluid is controllable
by designing properly the ORC components, according to the
safety precautions prescribed in the operating environment
[15,34]. For instance, hydrocarbons as toluene and pentane
have been successfully used at relatively high temperature
[16,19,27] without posing any serious problem. However, the
hazard control requirements increase the cost of the ORC installation [15].

3.3 Screening of Working Fluids. In order to identify the

most suitable fluids for the investigated application, it is necessary, for each organic substance, to simulate the integration of the
ORC system with the fuel cell plant and to assess the maximum
1
The occupational exposure limit, the threshold below which a toxic substance
has no effect on the health of the workers for an exposure period of 8 h per day, is
1 ppm for benzene, against 100 ppm of toluene and xylene, and 300 ppm of R245fa.
Since toluene is currently employed in ORC systems and the toxicity level of the
other considered aromatic hydrocarbons, when available, is comparable to that one
of toluene (and also of some refrigerants), it is assumed that only benzene is affected
by a level of toxicity incompatible with ORC applications.

APRIL 2013, Vol. 135 / 042309-3

Downloaded From: http://gasturbinespower.asmedigitalcollection.asme.org/ on 08/08/2014 Terms of Use: http://asme.org/terms

Table 4 Condensation temperature of candidate fluids (fluids

which do not satisfy the environmental requirements discussed
in Sec. 3.2 are not considered)
Substance
R134a
Cyclopentane
Cyclobutane
R245fa
Cyclohexane
n-Hexane
n-Pentane
n-Butane
MM
Toluene
D3
Ethylbenzene
p-Xylene
m-Xylene
o-Xylene
MDM
D4
MD2M
D5

CAS number

Tcond (  C)

811-97-2
287-92-3
287-23-0
460-73-1
110-82-7
110-54-3
109-66-0
106-97-8
107-46-0
108-88-3
541-05-9
100-41-4
106-42-3
108-38-3
95-47-6
107-51-7
556-67-2
141-62-8
541-02-6

35.0
35.0
35.0
35.0
35.0
35.0
35.0
35.0
37.3
45.3
67.6
68.0
68.8
69.7
74.1
82.3
103.5
118.9
132.9

achievable power. Since this operation is time consuming, it is advisable to reduce the number of candidate fluids by discarding
those requiring a relatively high condensation temperature (Tcond).
At the condenser, two technical constraints are imposed:
(1) The minimum cycle temperature is 35  C, because cooling
water at 15  C is assumed.
(2) The minimum condensation pressure is 100 kPa, as a precaution against excessive inward leaking of incondensable
gases in the condenser, which would negatively affect conversion efficiency and the thermal stability of the working
fluid [12,16]. In addition, more costly requirements would
be needed in order to maintain stable vacuum conditions.
For each fluid, the ORC condensing temperature is thus prescribed by applying the more restrictive constraint (see Table 4).
In general, fluids with high Tcond achieve lower performance in
a waste heat recovery application, since they reject thermal energy
at a temperature which is higher than the prescribed minimum
(35  C). In fact, for the analyzed system, even in the case of a
CHP application, it is more convenient to maximize electricity
production in the ORC unit and exploit the residual thermal
energy of the flue gases for cogeneration purposes. Condensation
heat is not exploited for cogeneration, as usually done in biomassfired ORC power plants, but is dissipated to the environment. For
this reason, fluids demanding higher condensing temperature are

Fig. 1

042309-4 / Vol. 135, APRIL 2013

Fig. 2 Layout of the integrated plant

strongly penalized. With reference to the case study, some preliminary simulations have clearly shown that fluids characterized by
a condensation temperature higher than 60  C allow achieving, by
far, the worst conversion performance. These substances, highlighted in bold in Table 4, are not further considered in the
analysis.
3.4 Optimal Fluids. For each selected substance, whose saturation curves are shown in Fig. 1, detailed simulations of the
integrated power plant are carried out. Simulations require solving
the mass and energy balances only for the ORC system, since data
about MCFC exhaust gases (composition, temperature, and mass
flow rate) are provided by the manufacturer [9]. The layout of the
combined plant is shown in Fig. 2.
ORC simulations are performed using a commercial simulation
tool [6], while cycle parameters are optimized by an external code
implemented in a general purpose technical computing environment [7]. The optimization algorithm iteratively changes the maximum temperature (vapor temperature at turbine inlet) and the
evaporation pressure of the working medium in order to maximize
the objective function, namely, the net power of the ORC genset.
The main assumptions are summarized below:

Efficiency values of turbine and feed pumps are set to the

same values for all the considered fluids (values are suggested by the ORC manufacturer [35] and are reported in Table 5), even if the operating parameters, the fluid properties,
and the power size have a strong influence on component

Saturation curves of selected fluids

Transactions of the ASME

Downloaded From: http://gasturbinespower.asmedigitalcollection.asme.org/ on 08/08/2014 Terms of Use: http://asme.org/terms

Table 5

Component efficiency and operating parameters

Turbine efficiency (%)

Pump efficiency (%)
Cooling water temperature (  C)
Cooling water maximum temperature increase (  C)
Minimum temperature working fluid (  C)
Minimum pressure working fluid (kPa)
DTp.p. PHE (  C)
DTp.p. REG (  C)

78.7a
75.2
15
8
35
100
30
15

Overall turbine efficiency is estimated assuming 82% isentropic efficiency and 96% mechanical efficiency.

efficiencies. However, this approximation does not compromise the identification of most suitable organic substances for
the power plant [1].
Cooling water is assumed available at 15  C, allowing for a
condensation temperature of 35 if the resulting pressure is
above 100 kPa. The condensation temperature of each fluid is
reported in Table 4. The maximum temperature increase of
cooling water is limited to 8  C, in compliance with typical
environmental regulations.
The minimum temperature difference, DTp.p., in the primary
heat exchanger (PHE) and in the regenerator (REG) are not
taken as optimization variables, but they are fixed to 30  C
and 15  C, respectively. Moreover, the constraint on the
DTp.p. in the PHE univocally defines the mass flow rate of the
ORC working fluid.
The regenerator is included in the ORC plant layout for each
of the candidate fluids.

The last assumption is motivated by the fact that regeneration

reduces heat recovery (due to the higher inlet temperature of the
working fluid in PHE), but, depending on the working medium
and operating conditions, does not affect or allows for an increase
of the ORC power output [36]. The two following arguments can
explain this result. Suppose that the optimal value of turbine inlet
temperature and cycle pressure have been found, and consider the
composite temperature thermal power curve (T-Q curve) of the
exhaust gases and the working medium reported in Fig. 3. Once
the cycle parameters are fixed, the pinch point position in the PHE
occurs at a certain point along the T-Q curves, depending on the

Fig. 3 The composite temperature thermal power curve (T-Q

curve) of the exhaust gases and the working medium. Pinch
point does not occur at the ends of PHE.

Journal of Engineering for Gas Turbines and Power

exhaust gases inlet temperature and heat capacity. The mass flow
rate of the generated vapor is defined by the thermal power recoverable between the exhaust inlet and the pinch point of the PHE.
If fluid regeneration is not performed, the pinch point position and
mass flow rate of vapor do not change. Therefore, since the outlet
and inlet condition of turbine and feed pump are fixed (points 1, 2,
4, and 5 in Fig. 2), the net power of the cycle remains constant (if
variations in the pressure drops across the heat exchangers of the
ORC unit are neglected). In this case, it comes out that regeneration has no influence on ORC net power production.
Now consider the case of Fig. 4. The pinch point position is at
the inlet of PHE. At the same turbine inlet temperature and cycle
pressure, without regeneration, the pinch point moves at a lower
temperature and then the vapor mass flow rate decreases (it can be
deduced by observing the increase in the slope of the composite
curve of the working fluid in the T-Q diagram), penalizing the net
power output. In general, a new optimization of the cycle parameters does not improve the solution as much as obtained with fluid
regeneration. In this circumstance, the regenerator positively
affects the electricity production, since it enhances the matching
between the hot and cold composite curves in the PHE.
Moreover, for the analyzed case study, regeneration entails two
benefits:
(1) Due to the higher temperature of exhaust gases at the PHE
outlet, the remaining thermal energy can be exploited for a
cogeneration purpose.
(2) Since the maximum temperature increase of cooling water
is 8  C, regeneration permits reducing the power consumption of the condenser pump, thus increasing the net electricity production of the plant. On the contrary, this effect does
not occur when the DTp.p. of the condenser is instead kept
constant.
A crucial aspect for the validity of ORC simulations is the
selection of a reliable equation of state (EOS) for the estimation
of the thermodynamic properties of the working fluid. Reference
or technical equations of state (REOS) have been developed by
the National Institute of Standards and Technologies (NIST) for a
number of organic substances [28], among which most of the fluids are used in refrigeration. The adopted simulation tool provides the option of estimating fluid properties with REOS only
for a limited set of fluids. In case of MM, cyclopentane, and
cyclobutane, therefore, a simpler and less accurate thermodynamic model has been employed, namely the PengRobinson

Fig. 4 Composite temperature thermal power curve (T-Q curve)

of the exhaust gases and of the working medium. Pinch point
occurs at the inlet of the working fluid in PHE.

APRIL 2013, Vol. 135 / 042309-5

Downloaded From: http://gasturbinespower.asmedigitalcollection.asme.org/ on 08/08/2014 Terms of Use: http://asme.org/terms

cubic equation of state. This approximation does not compromise

the comparison among the various simulated heat recovery units
adopting different working fluids, because the error introduced
by PengRobinson EOS is expected to be within the total uncertainty of the overall model of the ORC system.
The solution found by the optimization procedure should also
satisfy physical and technical constraints typical of the ORC process. An arbitrary combination of the cycle parameters can be
unfeasible, for instance, yielding a two-phase fluid at the outlet of
the evaporator or a wet expansion in the turbine. Therefore, process constraints are embedded in the optimization algorithm,
returning an objective function equal to 1 if any constraint is
violated. This procedure has the following advantages:
(a)
(b)
(c)

explicit relationships to describe technical and physical

constraints of ORC are not needed
the original constrained problem is transformed into an
unconstrained one
the optimization procedure is robust with respect to simulation errors

On the other hand, this approach has the drawback that the
objective function is not continuously differentiable, so that tackling this optimization problem requires a robust derivative-free
method. For our problem, the complex method [37] is preferred
due to its robustness and relative simplicity.
The results of the optimization procedure are reported in
Table 6.
The organic substance allowing for the best performance is
cyclohexane: more than 80% of the MCFC exhaust energy can be
converted into electricity, and this simulated ORC system features
also the highest thermal efficiency, namely, 26%. This result is
achievable, thanks to the almost perfect matching between the hot
and cold composite curves in the primary heat exchanger, as
shown in Fig. 5. The regenerator improves the performance of the
ORC, because the pinch point position is located at the inlet of the
working fluid in the PHE. The cycle is supercritical, and this is the
case for all other working fluids, except toluene, which allows
achieving the maximum performance at relatively low pressure
(17 bar). In comparison to the cyclohexane ORC system, the adoption of toluene as the working fluid provides several advantages:
(a)

(b)

Fig. 5

Fluid regeneration is not necessary, unless cogeneration

is exploited, since the pinch point position is not affected
by the temperature of the working medium at the inlet of
the PHE (see Fig. 6).
The volume flow ratio (VFR) of the turbine is more
favorable (high VFR demands for a complex and large
turbine and, consequently, the cost of the system is
bound to be greater) [1].

Temperatureentropy diagram for cyclohexane

Fig. 6 Temperatureentropy diagram for toluene

Table 6 Simulated ORC performance for each candidate working fluid. The net power is scaled assuming 1 MW of recoverable
thermal energy from the fuel cell plant, which corresponds to an exhaust temperature at stack of 85  C. Table rows in bold characters represents the three cases considered in the economic analysis.

Substance
Cyclohexane
Cyclopentane
Cyclobutane
Toluene
n-Hexane
n-Pentane
R245fa
n-Butane
MM
R134a

EOS

ORC Net
powera (kWel)

Tin turbine
(  C)

Cycle
pressure (bar)

Vout/Vin
turbine

gel

gII

vb

Tout
exhaust (  C)

Cycle
type

Refprop
Peng Rob.
Peng Rob.
Refprop
Refprop
Refprop
Peng Rob.
Refprop
Peng Rob.
Peng Rob.

221.83
219.11
208.87
206.24
201.15
193.98
186.16
182.55
181.25
174.10

290.00
290.70
300.00
247.85
270.00
270.00
290.42
270.00
260.96
301.57

48.51
89.00
143.00
17.00
53.00
80.99
142.87
117.00
30.04
185.00

574
214
70
175
354
120
64
39
754
17

26.15%
25.01%
24.65%
24.86%
24.38%
23.84%
23.91%
22.89%
24.28%
23.35%

42.66%
42.14%
40.17%
39.66%
38.68%
37.31%
35.80%
35.11%
34.86%
33.48%

84.83%
87.61%
84.72%
82.95%
82.51%
81.36%
77.86%
79.76%
74.64%
74.56%

130.61
122.22
130.94
136.24
137.58
141.04
151.53
145.82
161.16
161.40

Supercritical
Supercritical
Supercritical
Saturated
Supercritical
Supercritical
Supercritical
Supercritical
Supercritical
Supercritical

ORC net power is computed as the difference between the electrical power of the turbogenerator and the consumption of the feed pump and the cooling
water pump.
b
Heat recovery factor.

042309-6 / Vol. 135, APRIL 2013

Transactions of the ASME

Downloaded From: http://gasturbinespower.asmedigitalcollection.asme.org/ on 08/08/2014 Terms of Use: http://asme.org/terms

(c)

The cycle pressure is lower, which implies a lower capital expenditure (CAPEX) for the bottoming unit, since
cycle pressure influences specific costs of piping and
heat exchangers, one of the major cost items of ORC
power plants [38].

For these reasons, the toluene ORC unit, as well as the cyclohexane one, is selected for the economic comparison with the
standalone MCFC power plant. In addition, an ORC system featuring a nonhydrocarbon fluid is also considered in order to extend
the analysis. Simulated heat recovery units employing MM and
R245fa as working fluids result in similar performance. The system with MM as working fluid is selected for inclusion in the economic analysis; even its performance is slightly lower, due to the
following attractive features:
(a)
(b)
(c)

the low cycle pressure, overall less than half of the

cyclohexane cycle pressure
the lower cost of the fluid compared to R245fa
the higher temperature of the exhaust gases at the PHE
outlet, an aspect which promotes the exploitation of the
remaining heat for cogeneration

Furthermore, the choice of MM is also supported by the fact

that state-of-the-art high-temperature ORC currently employs
mainly siloxanes as working fluids [11].
Finally, it is worth noting that, in each optimized cycle, the
vapor superheating is limited to the amount strictly required in
order to avoid that liquid droplets enter the turbine. As shown in
Ref. [33], superheating negatively affects cycle efficiency in case
the working fluid features a positive slope (dT/ds) of the saturation
vapor curve in the Ts diagram. In the MM and cyclohexane
cycles (see Figs. 5 and 7), it occurs that some thermodynamic
states in the expansion are very close to the saturation curve. As it
is well known that liquid droplets formation in the turbine can
affect negatively its efficiency and reliability, this aspect should
be further investigated.

Economic Analysis of the Combined Plant

For each of the selected optimized ORC systems, four different configurations are examined, in order to point out the influence of scale effects and of the cogeneration option on the
economics of the combined plant. As far as the plant scale is
concerned, two different power capacities of the bottoming system are analyzed:

Table 7
Case
1
2
3
4

Configurations examined in the economic analysis

ORC size

Cogeneration

Regenerator

MCFC modules

500700 kWel
500700 kWel
11.4 MWel
11.4 MWel

Yes

Yes

Yes

Yes

2  2.8 MWel
2  2.8 MWel
4  2.8 MWel
4  2.8 MWel

(1) the combined plant employs two MCFC modules, resulting

in an electrical power of the ORC unit between 500700
kWel, depending on the considered working fluid
(2) the combined plant consists of four MCFC modules and of
an ORC genset with a net power output above 1 MWel
In the following, these two plant configurations are referred to
as the 500 kWel and 1 MWel cases.
When the exploitation of the remaining heat in the flue gases
for cogeneration purposes is also investigated, a counter-flow heat
exchanger is added to the plant layout before the stack, such that
it can generate hot water for a heating network.
In contrast to the procedure followed for the selection of the
optimal fluids, for all cases without cogeneration, the regenerator
is not employed, even if this implies, especially for an MM ORC
system, a loss in electricity production. This choice puts into evidence the impact of the regenerator on the plant capital cost. The
analyzed configurations are summarized in Table 7.
4.1 Assumptions for the Computation of the LCOE. The
economic benefits due to the integration of the technologies are
assessed by evaluating the levelized cost of electricity (LCOE) of
the combined plant with respect to the standalone MCFC system.
MCFC Unit. In the case of the standalone MCFC unit, the
LCOE is equal to 11.5 ce per kWel installed, assuming 15 years of
service life of the plant and a cost of natural gas slightly higher
than 5 e/GJ [3,9]. In order to obtain conservative results, a cogenerative configuration is analyzed, considering the use of waste
thermal energy in the flue gases to supply a heating network.
Thus, the LCOE of the FC plant decreases about 2.3% (LCOE
DFC3000 with cogeneration 11.24 ec/kWh). The revenue
related to cogeneration is obtained on the basis of the following
hypotheses:
(a)
(b)
(c)

The minimum temperature of the exhaust gases at the

stack is 85  C (this value corresponds to a return temperature of the water of heating network equal to 70  C).
The thermal load of the heating system is 2000 equivalent hours per year (which corresponds to the typical
load of a heating system at midlatitudes).
A heat price equal to 25 e/MWh. This value is a conservative assumption according to the fuel cost stipulated for the power plant, although heat prices for civil
users can be remarkably higher, due to lower consumption volumes and different fuel taxation.

These assumptions, summarized in Table 8, are also used for

the cogenerative cases of the combined plant.
Table 8

MCFC LCOE assumptions

Life of the power plant

Fuel cost
O&M cost
Average power produced
Availability
Specific cost
Thermal power cogenerated for each module
Cogeneration time
Heat price
Fig. 7

(year)
(e/GJ)
(ce/kWh)
(kWel)
(hour/year)
(e/kWel)
(kWel)
(hour/year)
(e/MWh)

15
5.1
2.86
2660
8000
2450
1640
2000
25

Temperatureentropy diagram for MM

Journal of Engineering for Gas Turbines and Power

APRIL 2013, Vol. 135 / 042309-7

Downloaded From: http://gasturbinespower.asmedigitalcollection.asme.org/ on 08/08/2014 Terms of Use: http://asme.org/terms

It is worth noting that, in the evaluation of the LCOE, a low

cost of natural gas is considered; this is a conservative hypothesis,
because it reduces the abatement of LCOE achievable by integrating the two technologies.

ORC Unit. Costs of ORC systems are estimated on the basis

of the commercial experience of one of the major companies in
the field of ORC manufacturing [35], considering a turnkey supply. In particular, the cost evaluation procedure can be summarized as follows:
(a)

(b)

(c)

(d)

(e)

The turbine cost depends on its number of stages, mean

diameter, and maximum allowable operating pressure.3
The number of stages is three (for each considered substance) and the speed of revolution is 3000 rpm, in order
to allow direct coupling with the asynchronous generator. With these assumptions, the preliminary estimate of
the axial turbine size is performed according to the mean
line design procedure described by Macchi [39] and
Lozza et al. [40]. This method permits the optimization
of the basic turbine design variables (including mean diameter, isentropic degree of reaction of stages, all relevant blade geometric parameters, and pressure drop
subdivision among stages), using the Craig and Cox loss
correlation [41] to predict turbine performance.
The cost of heat exchangers is assumed proportional to
the weight of the materials (carbon steel), according to
an empirical correlation, developed, analyzing commercial proposals of different manufacturers. The design of
these components is carried out by an in-house tool, having set the heat exchanger type, the inlet conditions of
the streams, and the temperature difference at the pinch
point.
For the evaporator of the ORC unit, a kettle-type shell
and tube heat exchanger is considered for the subcritical
cycle, while, for the supercritical cycles, a once through
heat exchanger is designed.
The fluid cost is proportional to the mass (about 3 e/kg
for hydrocarbons, 5 e/kg for siloxanes) of the working
medium in the closed loop. The fluid quantity is estimated, considering the volume occupied by the liquid in
the heat exchangers, increased 20% to take into account
the capacity of the pipes.
Pumps are selected on the basis of the fluid type, mass
flow rate, and total head, considering an overhead of
10% with respect to the nominal condition as an operating margin. Prices are then retrieved from manufacturers lists.
The cost of other components, as piping, control system,
measuring devices, lubrications systems, wirings, electronic power conditioning systems, etc., represents a fixed
percentage (15%) of the total expenditure. This percentage decreases as the capacity of the ORC system increases.

Fig. 8

ORC capital cost (e/kW)

greater than the one of the toluene genset. It is also worth noting
that, doubling the ORC power, the specific capital cost is reduced
by 30%.
The ORC LCOE, shown in Fig. 9, is evaluated on the basis of
the assumptions summarized in Table 8 and considering an O&M
cost of 0.5 ce/kWh. It is also important to observe that, in the
computation of the ORC LCOE (i) the lifetime of the bottoming
unit is shorter than that usually considered for these power plants
(15 years versus about 20 years [42]) and (ii) the value of the interest rate (5%) is relatively high [43]. These are conservative
hypotheses, because they reduce the abatement of the LCOE
achievable by the integration of the two technologies.
4.2 Combined Plant Economy. Energy and economic performance of the combined plant are summarized in Table 9. Adding an ORC unit downstream from a MCFC system allows for an
increase of power and efficiency of more than 10%, except for the
MM cases without the regenerator. However, all the considered
configurations well exceed 50% of efficiency, which is the best
performance with respect to any competitive technology (like
large stationary or marine diesel engine) in this power capacity
range. On the contrary, the LCOE reduction (see Fig. 10) achieves

In general, the most expensive component of the cycle is the

turbine, which represents 35% of the total capital investment. The
shares of the other cost items are: 30% for the heat exchangers,
10% for the working fluid, and 25% for other components. In the
cost estimation procedure, the results of which are reported in Fig.
8, effects related to production volume are not taken into account.
The genset affected by the high specific cost is the one employing cyclohexane as the working fluid, even if the estimated power
output is the largest of all the considered ORC power plants. This
is due to the high cycle evaporation pressure, which is one of the
key variables influencing the specific cost: as it can be observed in
Fig. 8, after cyclohexane, the system with the highest capital costs
is the MM ORC unit, which is characterized by a cycle pressure
2
The maximum allowable operating pressure of the turbine is calculated as the
nominal operating pressure plus a safety margin of 3 bar.

042309-8 / Vol. 135, APRIL 2013

Fig. 9

ORC LCOE

Transactions of the ASME

Downloaded From: http://gasturbinespower.asmedigitalcollection.asme.org/ on 08/08/2014 Terms of Use: http://asme.org/terms

Table 9 Energy and economic performance of the integrated plant for the different analyzed configurations
Dgela (%)

Wel (MWel)

Qcogen. (MWth)

gel (%)

LCOE

Cyclohexane
500 kWel w/o cogen.
500 kWel w. cogen.
1 MWel w/o cogen.
1 MWel w. cogen.

5.99
6.00
12.06
12.07

0.67

1.24

52.93
53.00
53.27
53.34

10.53
10.49
10.39
10.35

12.62
12.77
13.35
13.49

6.32
6.66
7.60
7.96

Toluene
500 kWel w/o cogen.
500 kWel w. cogen.
1 MWel w/o cogen.
1 MWel w. cogen.

5.94
5.97
11.96
11.97

0.68

1.29

52.51
52.75
52.82
52.86

10.58
10.53
10.42
10.39

11.73%
12.24%
12.38%
12.48%

5.91%
6.34%
7.27%
7.56%

MM
500 kWel w/o cogen.
500 kWel w. cogen.
1 MWel w/o cogen.
1 MWel w. cogen.

5.82
5.89
11.69
11.83

0.85

1.67

51.39
52.01
51.65
52.27

10.73
10.60
10.61
10.48

9.34%
10.65%
9.89%
11.21%

4.53%
5.66%
5.59%
6.75%

Analyzed cases

DLCOE* (%)

With reference to the standalone MCFC plant.

Fig. 10

LCOE reduction obtained by implementing the ORC unit

lower improvements, approximately half of those related to the

energy performance of the combined plant. This is due to the fact
that ORC LCOEs (see Fig. 9) represent about 20% of the standalone MCFC LCOE, while the ORC electricity production is
about 9% of the total electricity production.
The best LCOE reduction is obtained in the case of cyclohexane systems, despite the highest capital cost of the ORC genset. In
fact, since the specific cost of the bottoming unit is significantly
lower than that of the MCFC system, the LCOE of the combined
plant is more influenced by the energy production of the ORC unit
than by its cost. For this reason, in the MM case, the quite high additional costs arising from the implementation of the regenerator are
more than offset by the increase in electricity production. Due to the
cogeneration and the regenerative configuration, the ORC power
plant employing MM as the working fluid approaches the economic
result of the ORC systems employing the two hydrocarbons.

Conclusions

This work investigates the major aspects of the practical feasibility of the integration between a MCFC power plant and an organic Rankine cycle system by assessing the influence of all main
parameters affecting the integration, both from the technical and
Journal of Engineering for Gas Turbines and Power

economical point of view. The adopted method is as follows: (i)

selection of the more promising working fluids and of the more
profitable plant layouts for the ORC unit; (ii) optimization of thermodynamics parameters of the heat recovery system in order to
maximize the electricity production of the combined plant; and
(iii) economic analysis of the most promising configurations.
In particular, after a preliminary screening of the organic substance suitable for high-temperature ORC systems, ten candidate
fluids are selected. For each of them, detailed simulations of the
ORC are performed using a commercial process modeling tool,
while the optimization of the cycle parameters (cycle maximum
temperature and evaporation pressure) is obtained by an optimization algorithm implemented in an external programming environment. The ORC layout considered in the optimization procedure
includes a regenerator, since simple considerations demonstrates
that regeneration, even if it reduces heat recovery, impacts positively on the ORC net power.
Results of the optimization procedure show that the best working fluids for the analyzed application are cycloalkanes, especially
cyclohexane and cyclopentane. Even if the simulations of systems
using MM and toluene as the working fluids result in a lower
power output, these working media have some favorable features,
namely, a lower cycle pressure and turbine VFR, which reduce
the capital costs of the ORC unit. For this reason, the following
economic analysis considers these two working fluids as well as
cyclohexane. For each selected substance, four different configurations are examined, in order to point out the influence of the
ORC power capacity and the cogeneration option on the economy
of the combined plant. The benefits of the integration are assessed
on the basis of the LCOE.
The study results show that the addition of an ORC unit to the
MCFC system increases the electrical efficiency of the FC plant
from 47% to more than 53%. Using conservative assumptions
with commercial data provided by experienced companies in the
fields of FC and ORC systems, it is demonstrated that this efficiency enhancement allows reducing the LCOE by a value of
about 6%8%. The highest performance is obtained with cyclohexane as the ORC systems working fluid, which, in the best case
(with cogeneration and an ORC power size greater than 1 MWel),
allows reducing the MCFC LCOE by 8%. However, similar
results are obtained for toluene and the linear siloxane MM.
The economic feasibility of the combined plant is also demonstrated for relatively smaller capacity (500 kWel) of the ORC
unit, and it becomes particularly attractive for multi-MW MCFC
plants, implementing at least two modules of 2.8 MWel each,
where both efficiency and LCOE reach their best values. These
performance improvements can be achieved, implementing
APRIL 2013, Vol. 135 / 042309-9

Downloaded From: http://gasturbinespower.asmedigitalcollection.asme.org/ on 08/08/2014 Terms of Use: http://asme.org/terms

already well-established technology with clearly predictable costs,

and without penalizing the reliability of the plant. This may result
attractive with respect to more complex solutions, such as the
hybrid MCFC-gas turbine cycles, which offer higher performances, but involve a much more critical development of dedicated
turbomachinery, and a more difficult plant integration [4].

Nomenclature
Acronyms
BOL
CAPEX
EOL
EOS
FC
GWP
HFC
LCOE
LHV
MCFC
MW
ODP
ORC
PHE
REG
REOS
VFR

beginning of life of MCFC stack

capital expenditure
end of life of MCFC stack
equation of state
fuel cell
global warming potential
hydrofluorocarbons
levelized cost of electricity
low heating value
molten carbonate fuel cell
molecular weight
ozone depletion potential
organic Rankine cycle
primary heat exchanger
regenerator
reference equation of state
volume flow ratio

Symbols
PC
Q
Tb
TC
Tcond
Tstab
V
W
s

critical pressure
heat power
boiling temperature
critical temperature
condensing temperature
thermal stability threshold of the organic fluid
volume flow
electrical power
entropy

Subscripts
el
in
out
p.p.

electrical
inlet condition
outlet condition
pinch point

Greek Letters
g efficiency
v recovery factor

References
[1] Angelino, G., and Colonna di Paliano, P., 2000, Organic Rankine Cycles
(ORCs) for Energy Recovery From Molten Carbonate Fuel Cells, Proceedings
of the 35th Intersociety Energy Conversion Engineering Conference, Las
Vegas, NV, July 2428.
[2] Angelino, G., and Colonna di Paliano, P., 2000, Air Cooled Siloxane Bottoming Cycle for Molten Carbonate Fuel Cells, Fuel Cell Seminar, Portland, OR,
October 30November 2.
[3] Remick, R. J., Wheeler, D., and Singh, P., MCFC and PAFC R&D Workshop
Summary Report, U. S. DOE, January 2010, http://www1.eere. energy.gov/
[4] Campanari, S., Iora, P., Macchi, E., and Silva, P., 2007, Thermodynamic Analysis of Integrated MCFC/Gas Turbine Cycles for Sub-MW and Multi-MW
Scale Power Generation, ASME J. Fuel Cell Sci. and Technol., 4, pp.
308316.

042309-10 / Vol. 135, APRIL 2013

[5] Di Pippo, R., 2005, Geothermal Power Plants: Principles, Applications and
Case Studies, Elsevier, New York.
[6] Aspen Technology Inc., 2011, ASPEN PLUS v. 7.3, Burlington, MA.
[7] The MathWorks Inc., 2011, MATLAB 7.12, Natick, MA.
[8] Moreno, A., McPhail, S., and Bove, R., 2008, International Status of Molten
Carbonate Fuel Cell (MCFC) Technology, Joint Research CentreInstitute for
Energy, JRC Scientific and Technical Report EUR 23373 EN, European Commission, Luxembourg.
[9] FuelCell Energy, 2011, private communication.
[10] FuelCell Energy, 2010, DFC3000: Direct FuelCell Power Plant Applications
Guide, FuelCell Energy, Danbury, CT.
[11] Lai, N. A., Wendland, M., and Fischer, J., 2011, Working Fluids for HighTemperature Organic Rankine Cycles, Energy, 36, pp. 199211.
[12] Angelino, G., and Invernizzi, C. M., 1993, Cyclic Methylsiloxanes as Working
Fluids for Space Power Cycles, ASME J. Sol. Energy Eng., 115, pp. 130137.
[13] Drescher, U., and Bruggemann, D., 2007, Fluid Selection for the Organic Rankine Cycle (ORC) in Biomass Power and Heat Plants, Appl. Therm. Eng., 27,
pp. 223228.
[14] Angelino, G., and Invernizzi, C., 2003, Experimental Investigation on the
Thermal Stability of Some New Zero ODP Refrigerants, Int. J. Refrig., 26(1),
pp. 5158.
[15] Zyhowski, G. J., Honeywell Refrigerants Improving the Uptake of Heat Recovery Technologies, http://www.honeywell-orc.com/wp-content/uploads/2011/09/
Honeywell-Refrigerants-Improve-Uptake-Heat-Recovery-Technologies.pdf
[16] Schroeder, D. J., and Leslie, N., 2010, Organic Rankine Cycle Working Fluid
Considerations for Waste Heat to Power Applications, ASHRAE Trans.,
116(Part 1), pp. 525533.
[17] Andersen, A., and Bruno, T., 2005, Rapid Screening of Fluids for Chemical
Stability in Organic Rankine Cycle Applications, Ind. Eng. Chem. Res., 44,
pp. 55605566.
[18] Havens, V. N., Ragaller, D. R., Silbert, L., and Miller, D., 1987, Toluene Stability Space Station Rankine Power Systems, Proceedings of the 22nd Intersociety Energy Conversion Engineering Conference (IECEC), Philadelphia, PA,
August 1014.
[19] van Buijtenen, J. P., 2009, The Tri-O-Gen Organic Rankine Cycle: Development and Perspectives, Power Eng., 13(1), pp. 412.
[20] Ginosar, D. M., Petkovic, L. M., and Guillen, D. P., 2011, Thermal Stability of
Cyclopentane as an Organic Rankine Cycle Working Fluid, Energy Fuels,
25(9), pp. 41384144.
[21] Calderazzi, L., and Colonna, P., 1997, Thermal Stability of R-134a, R-13I1,
R-7146, R-125 Associated With Stainless Steel as a Containing Material, Int.
J. Refrig., 20, pp. 381389.
[22] Colonna, P., Nannan, N. R., Gurdone, A., and Lemmon, E. W., 2006,
Multiparameter Equations of State for Selected Siloxanes, Fluid Phase Equilib.,
244, pp. 193211.
[23] Colonna, P., Nannan, N. R., and Gurdone, A., 2008, Multiparameter Equations
of State for Siloxanes: [(CH3)3-Si-O1/2]2-[O-Si-(CH3)2]i1,,3, and [O-Si(CH3)2]6, Fluid Phase Equilib., 263(2), pp. 115130.
[24] Prabhu, E., 2006, Solar Trough Organic Rankine Electricity System. Stage 1:
Power Plant Optimization and Economics, Subcontract Report No. NREL/SR550-39433.
[25] Angelino, G., Invernizzi, C., and Macchi, E., 1991, Organic Working Fluid
Optimization for Space Power Cycles, Modern Research Topics in Aerospace
Propulsion, Springer-Verlag, New York.
[26] Colonna di Paliano, P., 1996, Fluidi di Lavoro Multi Componenti Per Cicli
Termodinamici di Potenza (Multicomponent Working Fluids for Power
Cycles), Ph.D. thesis, Politecnico di Milano, Milano, Italy.
[27] Leslie, N. P., Zimron, O., Sweetser, R. S., and Stovall, T. K., 2009, Recovered
Energy Generation Using an Organic Rankine Cycle System, ASHRAE
Trans., 115(Part I), pp. 220230.
[28] Lemmon, E. W., Huber, M. L., and McLinden, M. O., 2010, NIST Standard
Reference Database 23: Reference Fluid Thermodynamic and Transport
Properties-REFPROP, Version 9.0, National Institute of Standards and Technology, Standard Reference Data Program, Gaithersburg, MD.
[29] NIST Chemistry WebBook, NIST, http://webbook.nist.gov/chemistry/
[30] Calm, J. M., and Hourahan, G. C., 2011, Physical, Safety and Environmental
Data for Current and Alternative Refrigerants, Proceedings of 23rd International
Congress of Refrigeration (ICR2011), Prague, Czech Republic, August 2126.
[31] Papadopoulos, A. I., Stijepovic, M., and Linke, P., 2010, On the Systematic
Design and Selection of Optimal Working Fluids for Organic Rankine Cycles,
Appl. Therm. Eng., 30, pp. 760769.
[32] Maizza, V., and Maizza, A., 1996, Working Fluids in Non-Steady Flows
for Waste Energy Recovery Systems, Appl. Therm. Eng., 16(7), pp. 579590.
[33] Chen, H., Goswami, D. Y., and Stefanakos, E. K., 2010, A Review of Thermodynamic Cycles and Working Fluids for the Conversion of Low-Grade Heat,
Renewable Sustainable Energy Rev., 14, pp. 30593067.
[34] Marciniak, T. J., Krazinski, J. L., Bratis, J. C., Bushby, H. M., and Buyco, E.
H., Comparison of Rankine-Cycle Power Systems: Effects of Seven
Working Fluids, Argonne National Laboratory Report No. ANL/CNSVTM87.
[35] Turboden s.r.l., http://www.turboden.eu
[36] Martelli, E., Nord, L. O., and Bolland, O., 2012, Design Criteria and
Optimization of Heat Recovery Steam Cycles for Integrated
Reforming Combined Cycles With CO2 Capture, Appl. Energy, 92, pp.
255268.
[37] Box, M. J., 1965, A New Method for Constraint Optimization and a Comparison With Other Methods, Comput. J., 8, pp. 4252.

Transactions of the ASME

Downloaded From: http://gasturbinespower.asmedigitalcollection.asme.org/ on 08/08/2014 Terms of Use: http://asme.org/terms

[38] Stijepovic, M. Z., Linke, P., Papadopoulos, A. I., and Grujic A. S., 2012, On
the Role of Working Fluid Properties in Organic Rankine Cycle Performance,
Appl. Therm. Eng., 36, pp. 406413.
[39] Macchi, E., 1977, Design Criteria for Turbines Operating With Fluids Having a
Low Speed of Sound in Closed Cycle Gas Turbines, Lecture Series 100 on
Closed Cycle Gas Tubines, Von Karman Institute for Fluid-Dynamics, Bruxelles.
[40] Lozza, G., Macchi, E., and Perdichizzi, A., 1982,On the Influence of the Number of Stages on the Efficiency of Axial-Flow Turbines, 27th International Gas
Turbine Conference and Exhibition, London, April 1822.

Journal of Engineering for Gas Turbines and Power

[41] Craig, H. R. M., and Cox, H. J. A., 1970, Performance Estimation of Axial
Flow Turbines, Proc. Inst. Mech. Eng., 185, pp. 407424.
[42] Bronicki, L. Y., 1999, Organic Rankine Cycle Power Plant, for Waste Heat
Recovery, 13th Symposium on Industrial Applications of Gas Turbines, Banff,
Alberta, Canada, October.
[43] Costi di Produzione di Energia Elettrica da Fonti Rinnovabili, Rapporto
Commissionato da AEEG al Politecnico di Milano - Dipartimento di
Energia, Dicembre 2010, http://www.autorita.energia.it/allegati/docs/11/
103-11arg_rtalla.pdf

APRIL 2013, Vol. 135 / 042309-11

Downloaded From: http://gasturbinespower.asmedigitalcollection.asme.org/ on 08/08/2014 Terms of Use: http://asme.org/terms