international journal of hydrogen energy 35 (2010) 3758–3768

Available at www.sciencedirect.com

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

Performance evaluation of different configurations

of biogas-fuelled SOFC micro-CHP systems for
residential applications

Siamak Farhad a, Feridun Hamdullahpur b,*, Yeong Yoo c

a
Department of Mechanical and Aerospace Engineering, Carleton University, 1125 Colonel By Dr. Ottawa, ON K1S 5B6, Canada
b
Department of Mechanical and Mechatronics Engineering, University of Waterloo, 200 University Avenue, West Waterloo,
ON N2L 3G1, Canada
c
Institute for Chemical Process and Environmental Technology, National Research Council of Canada, 1200 Montreal Road. Ottawa,
ON K1A 0R6, Canada

article info abstract

Article history: Three configurations of solid oxide fuel cell (SOFC) micro-combined heat and power (micro-
Received 28 October 2009 CHP) systems are studied with a particular emphasis on the application for single-family
Received in revised form detached dwellings. Biogas is considered to be the primary fuel for the systems studied.
24 December 2009 In each system, a different method is used for processing the biogas fuel to prevent carbon
Accepted 15 January 2010 deposition over the anode of the cells used in the SOFC stack. The anode exit gas
Available online 9 February 2010 recirculation, steam reforming, and partial oxidation are the methods employed in systems
I–III, respectively. The results predicted through computer simulation of these systems
Keywords: confirm that the net AC electrical efficiency of around 42.4%, 41.7% and 33.9% are attain-
Solid oxide fuel cell able for systems I–III, respectively. Depending on the size, location and building type and
Micro-combined heat design, all the systems studied are suitable to provide the domestic hot water and electric
and power system power demands for residential dwellings. The effect of the cell operating voltage at
Biogas different fuel utilization ratios on the number of cells required for the SOFC stack to
Residential application generate around 1 kW net AC electric power, the thermal-to-electric ratio (TER), the net AC
Anode gas recirculation electrical and CHP efficiencies, the biogas fuel consumption, and the excess air required for
Steam reforming controlling the SOFC stack temperature is also studied through a detailed sensitivity
Partial oxidation analysis. The results point out that the cell design voltage is higher than the cell voltage at
which the minimum number of cells is obtained for the SOFC stack.
ª 2010 Professor T. Nejat Veziroglu. Published by Elsevier Ltd. All rights reserved.

1. Introduction has been recognized by the United Nations development

program as one of the important decentralized sources of
Biogas is a renewable and alternative source of fuel that can energy [1].
assist to reduce the consumption of fossil fuel and emission of Biogas, which contains mainly methane and carbon
greenhouse gases. This gas, which is mostly produced using dioxide, can be used as a fuel in a solid oxide fuel cell (SOFC) to
anaerobic digestion or fermentation of biodegradable mate- generate electric power with high efficiency and low envi-
rials such as biomass, manure, sewage, and municipal waste, ronmental impact [2–7]. An SOFC can achieve satisfactory

* Corresponding author. Tel.: þ1 519 888 4766; fax: þ1 519 885 5862.
E-mail addresses: siamak_farhad@yahoo.com (S. Farhad), fhamdull@uwaterloo.ca (F. Hamdullahpur).
0360-3199/$ – see front matter ª 2010 Professor T. Nejat Veziroglu. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijhydene.2010.01.052
 international journal of hydrogen energy 35 (2010) 3758–3768 3759

steam reforming, and partial oxidation are the methods

Nomenclature employed in these systems.
Cp heat capacity at constant pressure (kJ kg1 K1)
ex specific exergy (kJ kg1)
_
Ex exergy (W)
2. Systems description
h specific enthalpy (kJ kg1)
The configuration of the systems studied in this paper is
k heat capacity ratio (–)
shown in Fig. 1. These systems are mainly composed of an
LHV lower heating value (kJ kg1)
SOFC stack used to generate DC electric power and heat; an air
m_ mass flow rate (kg s1)
heater used to increase the temperature of the air before it
p pressure (Pa)
enters the SOFC stack; an air blower used to overcome the
T temperature (K)
pressure drop in the system; a burner used to convert the
Uf fuel utilization ratio (%)
_ chemical energy of the unutilized fuel in the SOFC stack to
W electric power (W)
heat; a boiler used to generate hot water for a residential
Greek letters dwelling; an inverter used to invert the DC electric power to
h efficiency (–) AC; and a reformer control volume that contains a biogas
clean-up system, heater(s) and/or reformer.
As shown in Fig. 1, three streams of biogas, air, and cold
water enter the systems. The composition of the biogas used
performance even if the methane content of the biogas is not in this study is the averaged composition of the biogas
high. Laboratory tests, performed by Jenne et al. (2002), produced in wastewater treatment plants in Ontario. This gas
confirmed that the efficiency of the SOFC drops by approxi- contains methane (60.8%), carbon dioxide (34.8%), nitrogen
mately 5% when the mole fraction of methane in the fuel (2.4%), oxygen (1.5%), and water vapor (0.01%). The range of
decreases from 70% to 30% [8]. Yi et al. (2005) showed that the hydrogen sulfide and silicon compounds in this gas is 2.5–3450
electrical efficiency of an integrated SOFC system drops and 0–2500 ppm, respectively. The other compounds such as
around 1.1% if biogas is used instead of natural gas [9]. toluene, benzene, methyl chloride, and chlorofluorocarbons
Biogas can be directly fed and internally reformed in an are present at levels below 10 ppm [35]. The composition of
SOFC stack; however, carbon deposition may occur and this gas is shown in the C–H–O ternary diagram in Fig. 2. As
gradually deactivate the SOFC anode catalysts [10–15]. shown in this figure, the biogas is located above the carbon
Another problem of internal reformation of biogas is a large deposition boundary (CDB) curves, indicating that carbon
temperature gradient in the SOFC stack due to a significant deposition over the anode catalyst in SOFCs is possible. As
cooling effect caused by the fast internal steam reforming explained in [36,37], the location of the biogas can be moved to
reaction [16]. To relieve these problems, biogas can be pro- below the CDB curves by adding sufficient anode exit gas,
cessed in an appropriate fuel processor before using in the water, or air to the biogas fuel in the reformer control volume.
SOFC stack [17]. The steam reforming [18–20], partial oxidation Of course, it is possible to move the location of the biogas to
[21,22], and auto-thermal reforming [23,24] are typical fuel below the CDB curves by adding CO2, but providing a suitable
processing methods which are usually suggested for SOFC source of CO2 for the operation of the SOFC system in resi-
systems. In addition to these fuel processing methods, anode dential dwellings is complicated. Hence, we do not consider
exit gas recirculation [25,26] can be applied to prevent carbon the effect of adding CO2 and dry reforming of the biogas in this
deposition over the anode catalyst. study.
Several researchers have suggested the application of SOFC In all systems studied, the biogas enters a gas clean-up
systems to generate electric power and thermal energy system where the contaminants in the biogas are reduced to
required for residential dwellings [27–32]. For this application levels that will not damage the anode and/or reformer
of SOFC systems, the thermal energy to electric power ratio catalysts. Depending on the operating temperature of the gas
(TER) generated in the SOFC system and the TER required for clean-up system, a heat exchanger may be required before
residential dwellings should match. The TER for residential cleaning the biogas. The most attractive and convenient
dwellings can be determined based on space heating, space method to remove hydrogen sulfide from the biogas in gas
cooling, or domestic hot water demands and its magnitude is clean-up systems is the use of an activated carbon bed. This
highly dependent on location, building type, design, usage method is highly effective at relatively low loadings of
pattern, time of day, and time of year [33]. The TER for hourly hydrogen sulfide (H2S < 200 ppm) [38]. In the case of high
averaged annual domestic hot water for a 200 m2 house can hydrogen sulfide content, additional removal technologies
range from 0.7 to 1.0 [34]. The TER for space heating can be are required to reduce the hydrogen sulfide content to below
substantially higher (more than ten times) than that for 200 ppm. A similar absorption bed can be used to remove
domestic hot water in cold climates [33]. silicon compounds that may cause a significant deactivation
In this paper, three biogas-fuelled SOFC micro-combined of the anode catalyst [35]. After cleaning the biogas, it should
heat and power (micro-CHP) systems for application in resi- be processed before entering the SOFC stack. Therefore, in
dential dwellings are evaluated through computer simulation. system I, after passing the cleaned biogas through a heat
In each system, a different method is used for processing the exchanger in the reforming control volume and increasing
biogas fuel to prevent carbon deposition over the anode of the its temperature, the high temperature cleaned biogas is
cells used in the SOFC stack. The anode exit gas recirculation, mixed with the anode exit gas (line 15 (I) in Fig. 1). In system
3760 international journal of hydrogen energy 35 (2010) 3758–3768

Fig. 1 – The configuration of biogas-fuelled SOFC micro-CHP systems (system I with anode exit gas recirculation, system II
with a stream reforming, and system III with a partial oxidation fuel processor).

II, an input water stream (line 15 (II) in Fig. 1) enters with the H2 þ O2- / H2O þ 2e (R1)
cleaned biogas to a steam reformer. In system III, the cleaned
biogas stream and an air stream, branched from the main CH4 þ H2O 4 CO þ 3H2 (R2)
inlet air stream (line 15 (III) in Fig. 1), enter a partial oxidation
reformer. After chemical reactions and increasing the CO þ H2O 4 CO2 þ H2 (R3)
temperature of the fuel to a certain value in the reforming
1/2O2 þ 2e / O2- (R4)
control volume, the processed fuel enters the SOFC stack.
The air also enters the SOFC stack after increasing its pres-
After the SOFC stack, the depleted fuel and air streams
sure and temperature in the blower and air heater to
enter a burner where the chemical energy of the unutilized
a certain value. Then, the H2 electrochemical reaction (R1)
fuel in the SOFC stack is converted to thermal energy. In the
and chemical reactions of steam reforming (R2) and water
case of system I, the required anode exit gas is branched from
gas shift reaction (R3) take place in the anode, and the elec-
the depleted fuel stream before entering the burner. The hot
trochemical reaction (R4) occurs in the cathode of cells and
gas produced in the burner then passes through the reforming
a DC electric power is generated in the SOFC stack. It should
control volume and air heater to supply their required heat.
be noted that the water gas shift reaction (R3) is assumed to
Finally, the thermal energy of this stream is used to increase
be in equilibrium in the anode. In fact, we assumed that the
the temperature of the inlet cold water to the boiler to supply
rate of this reaction at the operating temperature of the SOFC
the hot water required for residential dwellings.
stack with the presence of Ni catalyst in the anode is high
enough so that the water gas shift reaction can be considered
in thermodynamic equilibrium. In this condition, the H2
electrochemical reaction represents the total power gener- 3. Computer simulation
ated through the both H2 (R1) and CO (CO þ O2- / CO2 þ 2e)
electrochemical reactions in the anode. For residential To evaluate the performance of the systems shown in Fig. 1,
applications, the DC electric power generated should be we developed a computer code to simulate planar SOFCs at
inverted to AC using a DC to AC inverter. cell, stack and system levels. A detailed model of the cell,
 international journal of hydrogen energy 35 (2010) 3758–3768 3761

Fig. 2 – The location of the averaged composition of the biogas produced in wastewater treatment plant in Ontario and CDB
curves at atmospheric pressure and temperatures of 873 K, 973 K, and 1073 K in the C–H–O ternary diagram.

including the electrochemical reactions (R1) and (R4) in the The balance of plant (BoP) components such as the air
anode and cathode, respectively, and the steam reforming heater, blower, boiler, reformer, and burner were thermody-
and water gas shift reactions (R2) and (R3) in the anode were namically modelled under steady state operating conditions.
considered in the computer code and the activation, ohmic, The properties, composition and flow rate of all streams in the
and concentration polarizations were determined. The detail three systems were determined after modeling the BoP
description of the modeling of polarizations used in the components. Then, the net AC electric power, heat produced,
computer code is presented in Refs. [39–42]. It should be noted electrical efficiency, CHP efficiency, and total exergy destruc-
that the inlet and outlet fuel streams from the anode were tion of the systems were determined from Eqs. (1) to (5),
assumed to be in thermodynamic equilibrium in the cell respectively. It should be noted that the effects of the boiler
modeling. feedwater pump, the biogas blower, and the heat transfer from
The SOFC stack was modelled by extending the cell the BoP components on the electrical and CHP efficiencies and
model. Since heat transfer from the SOFC stack affects cell exergy destruction of the systems were not taken into account
performance, an insulation system is required to control in this study.
this heat transfer. The insulation system considered in this
study consists of an insulation layer mechanically sup-
ported by two metal layers. To determine the rate of heat
transfer from the SOFC stack, the conductive heat transfer
in the insulation layer and the radiative and natural
convective heat transfer from the outer metal layer were
taken into account. The effect of the heat transfer from the
SOFC stack on the cell’s performance was considered in the
computer code. The power output obtained from an SOFC
stack is usually less than that obtained from a single cell
multiplied by the number of cells used in the SOFC stack
due to the gas and electron leakages and the effect of
bipolar plates in the stack. To reflect these effects on the
power output from the SOFC stack in the computer simu-
lation, we assumed a 5% voltage drop in Current-Voltage
(CV) characteristics of the SOFC stack in comparison with Fig. 3 – Results of the computer simulation and experiment
single cells. for a commercially available electrolyte supported cell (ESC 4).
3762 international journal of hydrogen energy 35 (2010) 3758–3768


_ destruction; total ¼ m
Ex _ net AC electric  Ex
_ biogas exbiogas  W _ hot water
Table 1 – Input data for computer simulation of the

biogas-fuelled SOFC micro-CHP systems for residential _
 Excold water (5)
applications.
In Eq. (1), the blower input power is determined from the
Parameter Value
following equation [50]:
Cell operating temperature 850  C
 kair 1 !
Cell operating pressure 1 atm
_ blower ¼ m p6 kair 1
Fuel inlet temperature to the SOFC stack 750  C W _ air Cpair T5 1 (6)
p5 hblower
Air inlet temperature to the SOFC stack 700  C
Fuel utilization ratio 60%, 70%, 80%
After determining the net electric power and the heat
Anode and cathode
produced in the system, the TER of the system can be obtained
Thickness 40 mm
Porosity (assumed) 0.33 (–) from Eq. (7).
Tortuosity (assumed) 4 (–)
Electrolyte thickness 90 mm Q_
TER ¼ (7)
Interconnect thickness 3000 mm _ net AC electric
W
Cell active length 10 cm
Cell active width 10 cm To prevent carbon deposition over the anode catalyst, the
Insulation system of the SOFC stack minimum required flow rate of the anode exit gas recircula-
Thickness 50 mm
tion for system I, water for system II, and air for system III was
Thermal conductivity 0.025 Wm1 K1
determined after obtaining the carbon deposition boundary.
Emissivity of the outer surface 0.8 (–)
of the SOFC stack The carbon deposition boundary is determined using the
Number of cells in the SOFC stack 27 three reactions of carbon decomposition (R5) [43], CO reduc-
Net AC electric power output from z1 kWe tion (R6), and the Boudouard reaction (R7) [36].
the SOFC system
Pressure drop in the system 0.3 bar CH4 4 C þ 2H2 (R5)
Air blower efficiency 62.5%
Inlet cold water temperature to the boiler 40  C CO þ H2 4 C þ H2O (R6)
Outlet hot water temperature from the boiler 90  C
Efficiency of DC to AC invertor 92%
2CO 4 C þ CO2 (R7)
Flue gas exhaust temperature TdewPoint þ 50  C
Pinch temperature of the boiler >20  C

4. Validation of the computer code

_ net AC electric ¼ W
W _ SOFC stack  hinverter  W
_ blower (1)

To validate the computer code at the cell level, the perfor-

Q_ ¼ m
_ cold water ðhhot water  hcold water Þ (2) mance of a commercially available electrolyte supported
cell (ESC 4) produced by H.C. Starck Company [44] was
W_ net AC electric simulated. This cell is composed of Ni/GDC (Gadolinia-
helectric ¼ (3)
_ biogas LHVbiogas
m doped Ceria) anode, dense YSZ electrolyte and YSZ/LSM
(lanthanum strontium manganese oxide) cathode. The Ni/
_ net AC electric þ Q GDC anode of this cell is more resistant against carbon
W
hCHP ¼ (4) deposition than the Ni/YSZ anode [45]. To simulate this cell,
m_ biogas LHVbiogas
it was assumed that the porosity and tortuosity of

Table 2 – The results obtained from the computer simulation for the SOFC micro-CHP systems at fuel utilization ratios of
80%, 70%, and 60%.
System I System II System III

Fuel utilization ratio (%) 80 70 60 80 70 60 80 70 60

1
Reforming agent to biogas ratio (kg kg ) 0.63 0.68 0.74 0.29 0.29 0.29 1.08 1.08 1.08
Biogas mass flow rate (kg h1) 0.432 0.443 0.475 0.436 0.454 0.508 0.536 0.59 0.662
AC electrical efficiency (%) 42.4 40.5 37.5 41.7 38.7 34.5 33.9 30.7 27.4
CHP efficiency (%) 76.9 78.4 80.4 72.9 75.1 77.6 80.5 82.5 84.5
TER (–) 0.81 0.94 1.15 0.75 0.94 1.25 1.38 1.68 2.09
Excess air (–) 5.2 4.2 3.3 4.9 3.9 3.1 5.1 4.2 3.4
Operating voltage of system (V) 16.3 16.9 17.3 16.5 17.1 17.5 16.3 16.7 17.1
Total exergy destruction of system (W) 1338 1401 1568 1369 1491 1751 1858 2122 2473
Blower input power (W) 156 132 119 141 118 105 158 145 132
Generated hot water (kg h1) 13.5 15.2 18.5 12.2 15.0 19.8 22.6 27.6 34.3
Exit flue gas temperature ( C) 79.1 82.1 85.0 85.5 88.9 92.5 82.4 85.2 88.6
 international journal of hydrogen energy 35 (2010) 3758–3768 3763

Table 3 – Reference species of the atmospheric air [46]. 5. Input data

Compound %mol
The input data used for evaluation of the systems studied for
Ar 0.912
CO2 0.0337
residential applications are listed in Table 1. The SOFC stack
D 2O 0.000344 used in these systems consists of twenty-seven ESC 4 cells
H 2O 2.215 that operate at the temperature, pressure and voltage of
He 0.000488 around 850  C, 1 atm, and 0.7 V, respectively. Each system
Kr 0.000098 generates around 1 kW net AC electric power after deducting
N2 76.305
the power required for the internal consumption of the
Ne 0.00178
system. It is also considered that the exhaust gases leave the
O2 20.531
Xe 0.0000088 systems at a temperature which is 50  C greater than the dew
point temperature of the exhaust gases. It should be noted
that the dew point temperature is calculated based on the
partial pressure of the water vapour in the exhaust gases.
electrodes are 0.33 and 4, respectively. The voltage and
power density of the ESC 4 cell obtained by the computer
simulation and the experiment at various current densities 6. Results and discussions
and the cell operating temperature of 850  C are shown in
Fig. 3. The computer code could predict the voltage and Based on the input data presented in Table 1, the results of the
power density of the ESC 4 cell with an average relative computer simulation for all three systems studied are
error of 1%. obtained and listed in Table 2.

Fig. 4 – The share of each component in the exergy destruction of the input biogas to (a) system I with anode exit gas
recirculation, (b) system II with steam reforming, and (c) system III with partial oxidation fuel processor, at the fuel
utilization ratio of 80%.
3764 international journal of hydrogen energy 35 (2010) 3758–3768

flow rate of the anode exit gas for system I depends on the fuel
utilization ratio, and it is equal to 0.63, 0.68, and 0.74 of the input
biogas mass flow rate at the fuel utilization ratios of 80%, 70%
and 60%, respectively. The minimum mass flow rate of the
input water to the reforming control volume of system II, and
the input air to the reforming control volume of system III is
0.29 and 1.08 of the input biogas mass flow rate, respectively,
and these values are independent of the fuel utilization ratio.
According to the computer simulation results, at the fuel
utilization ratio of 80%, the inlet biogas flow rate to systems I
to III is approximately 0.432 kg/h, 0.436 kg/h and 0.536 kg/h,
respectively. This result shows that the inlet biogas mass flow
rate to system I is the lowest among the systems studied.
System I also exhibits a net AC electrical efficiency of 42.4% at
the fuel utilization ratio of 80%, followed by system II with
41.7%, and system III with 33.9%. System III provides the
highest CHP efficiency among the systems studied, followed
by systems I and II. The CHP efficiency of systems I to III is
around 76.9%, 72.9%, and 80.5%, respectively, at the fuel
utilization ratio of 80%. The TER of systems I and II at the fuel
utilization ratio of 70% and 80% is well-matched with the TER
based on the domestic hot water demands for single-family
detached dwellings [34]. The studies show that the TER of
system III is suitable for this application at the fuel utilization
ratio of 90%. The excess air required for controlling the
temperature of the SOFC stack is the lowest for system II, and
so the size and initial investment cost of the blower, air
heater, and burner is the lowest for this system among the
systems studied. The operating voltage of system II is the
highest among the studied system. At the fuel utilization ratio
of 80%, the voltage level reaches around 16.5 V for system II
and 16.3 V for systems I and III. The total exergy destructions
in systems I and II are very close, especially at fuel utilization
ratios greater than 80%, and are substantially less than that in
system III. The studies also indicate that depending on the fuel
utilization ratio and the system configuration, 10% to 16% of
the electric power generated in the SOFC stack is consumed in
the air blower.

6.1. Exergy analysis

Excluding nuclear, magnetic, electrical, and interfacial effects,

the exergy of a stream can be divided into four components of
physical, chemical, kinetic, and potential exergies. In this
study, it is assumed that the changes of the inlet and outlet
Fig. 5 – Effect of the cell operating voltage at fuel utilization kinetic and potential exergies of a stream are negligible in
ratios of 80%, 70%, and 60%, on the number of cells in the comparison with the physical and chemical exergies. To
SOFC stack (dashed lines) and TER (solid lines) in (a) system calculate the physical and chemical exergies, the reference
I with anode exit gas recirculation, (b) system II with steam atmospheric air species listed in Table 3, and the reference
reforming, and (c) system III with partial oxidation fuel temperature, pressure, and relative humidity of 298.15 K,
processor (the yellow column indicates the range of cell 101325 Pa, and 70%, respectively, are taken into account. After
voltages at which the minimum number of cells is required finding the exergy of the streams, the exergy destruction of
in the SOFC stack operated with fuel utilization ratios a component is obtained from the difference between the
exergy of the input and output streams from the component
between 60% and 80%).
with the consideration of the exergy of the heat and work to or
from the component.
The minimum mass flow rate of the reforming agent To determine the share of each component of the systems
required for each system, to prevent carbon deposition over the studied in the exergy destruction of the input biogas, an
anode catalyst of the ESC 4 cell, is determined using the extensive exergy analysis at the cell fuel utilization ratio of
computer simulation. As shown in Table 2, the minimum mass 80% was performed and the results are illustrated in Fig. 4. As
 international journal of hydrogen energy 35 (2010) 3758–3768 3765

Table 4 – The cell operating voltage at which the minimum number of cells is obtained for the systems studied, at fuel
utilization ratios of 60%, 70%, and 80%.
System I System II System III

Fuel utilization 60 70 80 60 70 80 60 70 80
ratio (%)

Cell operating 0.62 0.58 0.55 0.57 0.61 0.62 0.58 0.60 0.61
voltage (V)
Minimum number 21 22 23 21 22 23 22 23 24
of cells (–)

shown in this figure, the exergy destruction in the SOFC stack shown in this table, for fuel utilization ratios between 60% and
is not as significant as in the air heater, because the heat 80%, the range of mCVs is obtained between 0.55 V and 0.62 V
generated due to polarizations in the high temperature ESC 4 for system I, 0.57 V and 0.62 V for system II, and 0.58 V and
cells can still generate additional electricity in other power 0.61 V for system III. These voltage ranges are shown with
generation devices. The air heater has the largest share in the yellow columns in Figs. 5 and 6.
exergy destruction of the input biogas in all the studied Based on the mCV, The operating voltages of a cell can be
systems, followed by the SOFC stack and burner/boiler for divided into two groups; those which are less than mCV
system I, the SOFC stack and burner for system II, and the (Group I) and those larger (Group II). Considering this classi-
boiler and SOFC stack for system III. There is the potential to fication, we will show that the optimum operating voltage of
generate additional electric power in these systems, especially a cell always falls in Group II of the cell category.
in system III, if they are combined with other power genera- As shown in Fig. 5, with decreasing the cell voltage, the
tion devices and appropriately optimized using pinch tech- number of cells required for the SOFC stack increases for
nology and exergy analysis [47,48]. Group I and decreases for Group II of the cell voltages. For both
groups of voltages, the number of cells also decreases with
6.2. Sensitivity analysis decreasing the fuel utilization ratio.
As shown in Fig. 5, the TER of all the systems studied
The results listed in Table 2 were obtained for the SOFC stack increases progressively with decreasing the cell operating
with twenty-seven ESC 4 cells. Since the SOFC stack repre- voltage. The TER also increases with decreasing the fuel
sents 30–50% of the initial investment cost of an SOFC system utilization ratio, because the fuel unutilized in the SOFC stack
[49], the number of cells has an important role to economically can be used in the burner to generate additional heat for the
optimize the SOFC system. Two key parameters that affect the boiler to produce more hot water for residential dwellings. For
number of cells in the SOFC stack are the cell operating Group I of the cell voltages, the TER of all the systems studied
voltage, which is the average of the single cell voltages, and is greater than unity. It means, the heat produced in these
the fuel utilization ratio [50]. In this study, a correlation systems is more than the heat required for the domestic hot
between the cell operating voltage at different fuel utilization water demand for single-family detached dwellings.
ratios and the number of cells required for the SOFC stack to
generate around 1 kW net AC electric power for residential 6.2.2. Effect of the cell operating voltage and fuel utilization
dwellings is established through a detailed sensitivity anal- ratio on electrical and CHP efficiencies
ysis. Since the cell operating voltage and fuel utilization ratio Fig. 6 illustrates the effect of the cell operating voltage at
affect the TER, excess air, and electrical and CHP efficiencies, different fuel utilization ratios of 60%, 70%, and 80%, on the
these parameters are also monitored during the sensitivity electrical and CHP efficiencies of the systems. As shown in
analysis. It should be noted that our study does not cover the this figure, an increase in the operating voltage of the cell
influence of the cell operating voltage and fuel utilization ratio leads to a linear increase in the net AC electrical efficiency of
on the durability and thermo-mechanical reliability of the cell. all the systems studied. In the range of the fuel utilization
ratios investigated, the net AC electrical efficiency is higher at
6.2.1. Effect of the cell operating voltage and fuel utilization elevated fuel utilization ratios.
ratio on the number of cells and TER The CHP efficiency of all the systems studied decreases
As shown in Fig. 5, the number of cells required for the SOFC linearly with a slight slope with increasing the cell operating
stack to generate around 1 kW net AC electric power for voltage to around 0.7 V. If the cell operating voltage exceeds
residential dwellings changes approximately as a concave this value the CHP efficiency of systems I and II decreases as
upward parabolic curve with the cell operating voltage for all a convex upward parabolic curve; however, depending on the
the systems studied. Indeed, the number of cells first reduces fuel utilization ratio, the CHP efficiency of system I may line-
with decreasing the cell operating voltage and then increases. arly increase at a certain cell operating voltage.
The cell operating voltage at which the minimum number of Although the CHP efficiency of the systems obtained at the
cells required in the SOFC stack is obtained is called ‘‘mCV’’ in cell voltages of Group II is slightly smaller than that obtained
this study, which is a function of the fuel utilization ratio. at the cell voltages of Group I, the electrical efficiency of these
The mCV for each system at fuel utilization ratios of 60%, systems obtained at the cell voltages of Group II is notably
70%, and 80% has been determined and listed in Table 4. As higher than that obtained at the cell voltages of Group I.
3766 international journal of hydrogen energy 35 (2010) 3758–3768

Fig. 7 – Effect of the cell operating voltage at fuel utilization

ratios of 80%, 70%, and 60%, on the inlet biogas mass flow
rates in (a) system I with anode exit gas recirculation
(green curves), (b) system II with steam reforming (red
curves), and (c) system III with partial oxidation (blue
curves) fuel processor.

with decreasing the cell voltage, the biogas mass flow rate
increases progressively for all the systems studied. The higher
the fuel utilization ratio, the lower the biogas mass flow rate
required for the systems studied to generate 1 kW AC electric
power. If the amount of the biogas fuel generated for a resi-
dential dwelling is limited to a certain value, this figure would
be important to select an appropriate cell voltage and fuel
utilization ratio for the systems studied.

6.2.4. Effect of the cell operating voltage and fuel utilization

ratio on excess air
Fig. 8 shows the effect of the cell operating voltage at fuel
utilization ratios of 60%, 70%, and 80% on the excess air
required for controlling the temperature of the SOFC stack for

Fig. 6 – Effect of the cell operating voltage at fuel utilization

ratios of 80%, 70%, and 60%, on the net AC electrical
efficiency (solid lines) and CHP efficiency (dashed lines) in
(a) system I with anode exit gas recirculation, (b) system II
with steam reforming, and (c) system III with partial
oxidation fuel processor (the yellow column indicates the
range of cell voltages at which the minimum number of
cells is required in the SOFC stack operated with fuel
utilization ratios between 60% and 80%).

6.2.3. Effect of the cell operating voltage and fuel utilization Fig. 8 – Effect of the cell operating voltage at fuel utilization
ratio on biogas fuel consumption ratios of 80%, 70%, and 60%, on the excess air in (a) system I
The effect of the cell operating voltage at fuel utilization ratios with anode exit gas recirculation (green curves), (b) system
of 60%, 70%, and 80% on the biogas mass flow rate for all three II with steam reforming (red curves), and (c) system III with
systems studied are shown in Fig. 7. As shown in this figure, partial oxidation (blue curves) fuel processor.
 international journal of hydrogen energy 35 (2010) 3758–3768 3767

all three studied systems. As shown in this figure, the increase Overall, the biogas-fuelled SOFC systems studied show an
of the cell operating voltage leads to a decrease in the excess appropriate performance to generate electric power and hot
air; and so the size and initial investment cost for the air water demands for single-family detached dwellings.
heater, blower, and burner decreases for all the systems
studied. In the range of the fuel utilization ratio investigated,
the excess air also decreases with decreasing the fuel utili-
zation ratio.
Acknowledgments
Overall, the results of the sensitivity analysis indicate that
The authors gratefully acknowledge financial support
with decreasing the cell operating voltage in Group I, the
provided by NSERC of Canada, EcoEnergy Technology Initia-
number of cells required for the SOFC stack increases, and
tive Program, and AAFC-NRC Bioproducts Program.
the excess air required for controlling the SOFC stack
temperature increases. Hence, the size and initial invest-
ment cost of the systems studied increase. The decrease in
references
the cell operating voltage in this group also leads to
a decrease in the electrical efficiency and an increase in the
biogas consumption; however, the CHP efficiency does not
[1] Reddy AKN, Williams RH, Johansson TB. Energy after Rio:
change considerably. The TER of the systems studied also
prospects and challenges. New York, NY: United Nations
exceeds the TER required for the domestic hot water Development Program; 1997.
demands for single-family detached dwellings. Conse- [2] Athanasiou C, Coutelieris F, Vakouftsi E, Skoulou V,
quently, the optimum cell operating voltage is not in this Antonakou E, Marnellos G, et al. From biomass to
group of the cell voltages and should be found in Group II. In electricity through integrated gasification/SOFC system-
fact, the lowest cell design voltage is limited to the cell optimization and energy balance. Int J Hydrogen Energy
2007;32:337–42.
voltage at which the minimum number of cells for the SOFC
[3] Van herle J, Membrez Y, Bucheli O. Biogas as a fuel source for
stack is obtained (mCV). SOFC co-generators. J. Power Sources 2004;127:300–12.
In Group II of the cell voltages, with increasing the cell [4] Shiratori Y, Oshima T, Sasaki K. Feasibility of direct-biogas
operating voltage, the number of cells for the SOFC stack SOFC. Int J Hydrogen Energy 2008;33:6316–21.
increases, whereas, the biogas consumption and excess air [5] Farhad S, Yoo Y, Hamdullahpur F. Effects of fuel processing
decrease. Therefore, a detailed economic analysis is required methods on industrial scale biogas-fuelled solid oxide fuel
to find the cell design voltage in this group. cell system for operating in wastewater treatment plants.
J Power Sources 2010;195(5):1446–53.
[6] Yentekakis IV, Papadam T, Goula G. Electricity production
from wastewater treatment via a novel biogas-SOFC aided
7. Conclusions process. Solid State Ionics 2008;179:1521–5.
[7] Van herle J, Maréchal F, Leuenberger S, Membrez Y,
The results of this study indicate that biogas is a suitable fuel Bucheli O, Favrat D. Process flow model of solid oxide fuel
for residential applications of the SOFC systems studied. Due cell system supplied with sewage biogas. J Power Sources
2004;131:127–41.
to the high carbon dioxide content of the biogas fuel, the
[8] Jenne M, Dörk T, Schuler A. Proceedings of the Fifth
amount of the reforming agent needed to prevent carbon
European Solid Oxide Fuel Cell Forum. Lucerne, Switzerland;
deposition over the anode of ESC 4 cells decreases. For an 2002:460–6.
SOFC stack with 27 ESC 4 cells operating at a fuel utilization [9] Yi Y, Rao AD, Brouwer J, Samuelsen GS. Fuel flexibility study
ratio of 80%, the net AC electrical efficiency of 42.4%, 41.7%, of an integrated 25kW SOFC reformer system. J Power
and 33.9% is predicted for systems I to III, respectively. In Sources 2005;144:67–76.
these conditions, system III shows the highest CHP efficiency [10] Assabumrungrat S, Laosiripojana N, Pavarajarn V,
Sangtongkitcharoen W, Tangjitmatee A, Praserthdam P.
with 80.5%, followed by system I with 76.9%, and system II
Thermodynamic analysis of carbon formation in solid oxide
with 72.9%. The TER of systems I and II at the fuel utilization fuel cells with a direct internal reformer fueled by methanol.
ratios of 70% and 80%, and system III at the fuel utilization J Power Sources 2005;139:55–60.
ratio of 90%, is suitable to produce the electric power and hot [11] Sangtongkitcharoen W, Assabumrungrat S, Pavarajarn V,
water demands for single-family detached dwellings. Laosiripojana N, Praserthdam P. Comparison of carbon
The exergy analysis shows that there is a considerable formation boundary for different types of solid oxide fuel
potential to generate additional electric power from the cells with methane feed. J Power Sources 2005;142:75–80.
[12] Park EW, Moon H, Park M, Hyun SH. Fabrication and
systems studied, especially from system III, if they are
characterization of Cu–Ni–YSZ SOFC anodes for direct use of
combined with other power generation devices and appro- methane via Cu-electroplating. Int J Hydrogen Energy 2009;
priately optimized. 34:5537–45.
The sensitivity analysis indicates that the optimum cell [13] Hofmann Ph, Panopoulos KD, Fryda LE, Schweiger A,
voltage is greater than the cell voltage at which the minimum Ouweltjes JP, Karl J. Integrating biomass gasification with
number of cells for the SOFC stack is obtained. The minimum solid oxide fuel cells: effect of real product gas tars,
fluctuations and particulates on Ni-GDC anode. Int J
number of cells at the fuel utilization ratio of 80% is obtained
Hydrogen Energy 2008;33:2834–44.
at the cell voltage of 0.55 V for system I, 0.62 V for system II,
[14] Wang Y, Yoshiba F, Kawase M, Watanabe T. Performance
and 0.61 V for system III. In these conditions, the number of and effective kinetic models of methane steam reforming
ESC 4 cells required for the SOFC stack is 23 cells for systems I over Ni/YSZ anode of planar SOFC. Int J Hydrogen Energy
and II, and 24 cells for system III. 2009;34:3885–93.
3768 international journal of hydrogen energy 35 (2010) 3758–3768

[15] Hofmann Ph, Panopoulos KD, Aravind PV, Siedlecki M, [33] Braun RJ, Kleina SA, Reindla DT. Evaluation of system
Schweiger A, Karl J, et al. Operation of solid oxide fuel cell on configurations for solid oxide fuel cell-based micro-
biomass product gas with tar levels >10 g Nm3. Int J combined heat and power generators in residential
Hydrogen Energy 2009;34(22):9203–12. applications. J Power Sources 2006;158(2):1290–305.
[16] Achenbach E, Riensche E. Methane/steam reforming kinetics [34] Braun RJ, Klein SA, Reindl DT. Considerations in the design
for solid oxide fuel cells. J Power Sources 1994;52:283–8. and application of solid oxide fuel cell energy systems in
[17] Piroonlerkgula P, Laosiripojanab N, Adesinac AA, residential markets. ASHRAE Trans 2004;110(1):14–24.
Assabumrungrat S. Performance of biogas-fed solid oxide [35] Wheeldon I, Caners C, Karan K, Peppley B. Utilization of
fuel cell systems integrated with membrane module for CO2 biogas generated from ontario wastewater treatment plants
removal. Chem Eng Process 2009;48:672–82. in solid oxide fuel cell systems: a process modeling study. Int
[18] Arteaga LE, Peralta LM, Kafarov V, Casas Y, Gonzales E. J Green Energy 2007;4:221–31.
Bioethanol steam reforming for ecological syngas and [36] Farhad S, Hamdullahpur F. Developing fuel map to predict
electricity production using a fuel cell SOFC system. Chem the effect of fuel composition on the maximum voltage of
Eng J 2008;136:256–66. solid oxide fuel cells. J Power Sources 2009;191(2):407–16.
[19] Peksen M, Peters R, Blum L, Stolten D. Numerical modelling [37] Farhad S, Hamdullahpur F. Developing fuel map to predict
and experimental validation of a planar type pre-reformer in the effect of fuel composition on the maximum efficiency of
SOFC technology. Int J Hydrogen Energy 2009;34:6425–36. solid oxide fuel cells. J Power Sources 2009;193(2):632–8.
[20] Chan SH, Ding OL. Simulation of a solid oxide fuel cell power [38] Yan R, Liang DT, Tsen L, Tay JH. Kinetics and mechanism of
system fed by methane. Int J Hydrogen Energy 2005;30: H2S adsorption by alkaline activated carbon. Environ Sci
167–79. Technol 2002;36:4460–6.
[21] Piroonlerkgul P, Assabumrungrat S, Laosiripojana N, [39] Chan SH, Khor KA, Xia ZT. A complete polarization model of
Adesina AA. Selection of appropriate fuel processor for a solid oxide fuel cell and its sensitivity to the change of cell
biogas-fuelled SOFC system. Chem Eng J 2008;140:341–51. component thickness. J Power Sources 2001;93:130–40.
[22] Hotz N, Senn SM, Poulikakos D. Exergy analysis of a solid [40] Virkar AV, Chen J, Tanner CW, Kim J. The role of electrode
oxide fuel cell micropowerplant. J Power Sources 2006;158: microstructure on activation and concentration
333–47. polarizations in solid oxide fuel cells. Solid State Ionics 2000;
[23] Franzoni A, Magistri L, Traverso A, Massardo AF. 131:189–98.
Thermoeconomic analysis of pressurized hybrid SOFC [41] Kim JW, Virkar AV, Fung KZ, Mehta K, Singhal SC. Low
systems with CO2 separation. Energy 2008;33:311–20. temperature, high performance anode-supported solid oxide
[24] Modafferi V, Panzera G, Baglio V, Frusteri F, Antonucci PL. fuel cells. J Electrochem Soc 1999;146(1):69–78.
Propane reforming on Ni–Ru/GDC catalyst: H2 production for [42] Colpan CO, Hamdullahpur F, Dincer I, Yoo Y. Effect of
IT-SOFCs under SR and ATR conditions. Appl Catal A Gen gasification agent on the performance of solid oxide fuel cell
2008;334:1–9. and biomass gasification systems. Int J Hydrogen Energy;
[25] Akkaya AV, Sahin B, Erdem HH. An analysis of SOFC/GT CHP 2009;. doi:10.1016/j.ijhydene.2009.08.083.
system based on exergetic performance criteria. Int J [43] Younessi-Sinaki M, Matida EA, Hamdullahpur F. Kinetic
Hydrogen Energy 2008;33:2566–77. model of homogeneous thermal decomposition of methane
[26] Peters R, Riensche E, Cremer P. Pre-reforming of natural gas and ethane. Int J Hydrogen Energy 2009;34(9):3710–6.
in solid oxide fuel-cell systems. J Power Sources 2000;86: [44] H.C. Starck Company, www.hcstarck-ceramics.com/
432–41. (accessed April 2009).
[27] Kazempoor P, Dorer V, Ommi F. Evaluation of hydrogen and [45] Hofmann Ph, Schweiger A, Fryda L, Panopoulos KD,
methane-fuelled solid oxide fuel cell systems for residential Hohenwarter U, Bentzen JD, et al. High temperature
applications: system design alternative and parameter study. electrolyte supported Ni-GDC/YSZ/LSM SOFC operation on
Int J Hydrogen Energy 2009;34:8630–44. two-stage Viking gasifier product gas. J Power Sources 2007;
[28] Sammes NM, Boersma R. Small-scale fuel cells for residential 173:357–66.
applications. J Power Sources 2000;86:98–110. [46] Szargut J, Morris DR, Steward FR. Exergy analysis of thermal,
[29] Lee KH, Strand RK. SOFC cogeneration system for building chemical, and metallurgical processes. Hemisphere
applications, part 1: development of SOFC system-level Publishing Corporation; 1988.
model and the parametric study. Renewable Energy 2009;34: [47] Farhad S, Saffar-Avval M, Younessi-Sinaki M. Efficient design
2831–8. of feedwater heaters network in steam power plants using
[30] Barrera R, De Biase S, Ginocchio S, Bedogni S, Montelatici L. pinch technology and exergy analysis. Int J Energy Research
Performance and life time test on a 5kW SOFC system for 2007;32(1):1–11.
distributed cogeneration. Int J Hydrogen Energy 2008;33: [48] Farhad S, Younessi-Sinaki M, Golriz MR, Hamdullahpur F.
3193–6. Exergy analysis and performance evaluation of CNG to LNG
[31] Bompard E, Napoli R, Wan B, Orsellob G. Economics converting process. Int J Exergy 2008;5(2):164–76.
evaluation of a 5kW SOFC power system for residential use. [49] Nietsch T, Clark J. Market Oriented Design Studies for SOFC
Int J Hydrogen Energy 2008;33:3243–7. Based Systems. ETSU Report F/0.1/00129/REP; 1999.
[32] Alanne K, Saari A, Ugursal VI, Goodc J. The financial viability [50] Farhad S, Hamdullahpur F. Conceptual design of a novel
of an SOFC cogeneration system in single-family dwellings. ammonia-fuelled portable solid oxide fuel cell system.
J Power Sources 2006;158:403–16. J Power Sources 2010;195:3084–90.