[go: up one dir, main page]

0% found this document useful (0 votes)
98 views21 pages

Systems: L Blum and E Riensche, Institute of Energy Research/Fuel Cells, Ju Lich, Germany

Download as pdf or txt
Download as pdf or txt
Download as pdf or txt
You are on page 1/ 21

Systems

L Blum and E Riensche, Institute of Energy Research/Fuel Cells, Julich, Germany


& 2009 Elsevier B.V. All rights reserved.

Introduction
The solid oxide fuel cell (SOFC) is described as a
technology that enables a very high electrical system
efficiency and provides, at the same time, high off-gas
temperatures, which enable a large range of heat applications. The first point is true if one succeeds in operating a cell at quite high cell voltages using a system
technology that enables high fuel utilization and low
internal consumption of the produced electricity (parasitic losses). The second point is only partly true, because
a large amount of heat is used internally in the system to
heat the cold incoming gases (especially air) and, in some
cases, to produce the steam needed for the reforming
process.
Various system concepts have previously been outlined for achieving efficient system operation. In the
following, the different basic plant arrangements and
their special features are described. For this purpose, first,
the different types of efficiencies are described in order
to explain the effect of single measures on the overall
electrical efficiency. Second, an overview on how to
perform energy balancing of SOFC plants is given, and in
particular a detailed description of the effects of internal
reforming and temperature differences across the stack
on parasitic power consumption is given. Finally, six
plant concepts are described to provide an insight into
various design options and the advantages and disadvantages that may be attributed to each of them.

Tasks of the System Control Loops


(Plant Requirements)
The various SOFC types have different requirements
regarding plant configuration, mainly based on the required operating temperature and on their capability of
allowing on-anode reforming.
In principle, the plant concepts for the SOFC are
much simpler than those for low-temperature fuel cells;
because no carbon monoxide poisoning occurs, a shift
reactor is not needed to convert most of the carbon
monoxide to carbon dioxide (as per phosphoric acid fuel
cell (PAFC) nor is it necessary to reduce the remaining
carbon monoxide to concentrations below 10 ppm (as per
proton-exchange membrane fuel cell (PEMFC)). The
possibility of internal (or integrated) reforming avoids the
necessity of a reformer only a small prereformer may
be needed (see SOFC Internal and External Reforming

Fuel Cells Solid Oxide Fuel Cells: Internal and External Reformation).
On the contrary, especially on the cathode side,
relatively large recuperators are needed to preheat the
incoming gas to the required stack inlet temperature.
Additionally, much more thermal insulation is required
to limit the heat losses at high stack temperatures.
To be able to operate an SOFC system, different
control loops and steering (open loop controls) must be
installed. These are as follows:
flow control proportional to the electric current,
fuel
air flow control proportional to the electric current
superposed
by the cooling requirements of the stack
(if it is air-cooled),

inlet temperature control into the stack,


air
inlet temperature control into the stack,
fuel
control of the fuel reformer,
temperature
control of water or steam,
flow
of product water,
removal
heating
sequence,
cooling sequence,
hot standby sequence, and
shutdown and safety shutdown.

Efficiencies Relevant for the System


Fuel cell systems promise high electrical efficiencies, but
on the way from cell to system many obstacles exist that
reduce the high efficiency values at the cell level. In
designing an SOFC system, one should be aware of the
different factors that can affect (and so reduce) the
electrical system efficiency. These factors can be assigned
to special efficiencies, the product of which will form the
total efficiency.
The definitions of these efficiencies are given in the
following.
Cell Efficiency
Cell efficiency ec, is the basic efficiency of a fuel cell and
it describes the relation between the specific energy,
delivered under load (h  zxFxVp(i )), and the reaction
enthalpy of the cell reaction, related to standard conditions (DH  zxFxVLHV):
ec

Vp
VLHV

1

99

100

Fuel Cells Solid Oxide Fuel Cells | Systems

where Vp is the (practical) cell voltage and VLHV the


LHV heading voltage. In the case of operation at a cell
voltage of 0.7 V using various gases, the cell efficiency
would amount to 56% for hydrogen, 67% for methane,
and 48% for carbon monoxide.

the direct current (DC) voltage level of the stack to the


alternating current (AC) voltage level of the consumer,
respectively the grid:

Fuel Utilization

Inverters of low power can have efficiencies as low as 85


90%, whereas the best ones with higher power output
and high input voltage can achieve 96%.

Owing to the formation of water on the anode side, the


water vapor partial pressure increases toward the gas
outlet of the cell. As a consequence, the Nernst voltage
drops markedly in the case of water content above 95%.
For this fundamental reason, it is not possible to electrochemically react all fuel in an SOFC. The relation
between the amount of fuel gas reacting in the cell m rea
and the amount of fuel gas m in entering the system (respectively the stack) is defined as fuel utilization. Because
it is not so easy to measure the amount of reacted fuel, it
is much more accurate to calculate it with Faradays law,
using the current produced by the stack, which is easy to
measure. As a stack normally consists of several layers,
the current has to be multiplied by the number of cells,
connected electrically in series, to obtain Itot:

uF

m rea
m in

M
 Itot
z
F
m in

Itot
Itot

Iin P m i;in  z  F
i
i
Mi

2

3

5

Parasitic Efficiency
Parasitic efficiency ep, describes the relation between
effective net power and produced gross power, because a
part of the produced power is needed for operating the
plant and is therefore not available for exploitation. As
the final consumers are normally supplied with AC, the
AC side of the inverter must be considered:
ep

PAC;gross  PAC;parasitic
PAC;net

PAC;gross
PAC;gross

6

The total electric plant efficiency, eel, can be derived by


combining the efficiencies described in eqns [1][6]. It
can also be described by the relation between the electrical energy, available for the consumer, and the energy
fed to the system by the fuel gas, but this provides less
information about the effect of the different plant components on the overall efficiency:
eel

PAC;net
es  ep  eInv ec  uF  ep  eInv 7
m  LHV

In power plant technology, it is general practice to use


the lower heating value (LHV); therefore the LHV is
generally used here in the calculations.

Stack Efficiency
Sometimes it is not very clear in publicized values which
efficiency is meant. Very often it is not the system efficiency but the stack efficiency (also sometimes called
effective efficiency). Stack efficiency, es, is the product of
cell efficiency ec and fuel utilization uF:
es ec  uF

PAC;gross
PDC;gross

Electric Plant Efficiency

where M is the molar mass and uF the utilization of fuel.


If a mixture of gases is supplied to the system, it is
sometimes more efficient to calculate how many electrons are supplied to the system by using Faradays law
for each gas component:
uF

eInv

4

In the case of operation at a cell voltage of 0.7 V and fuel


utilization of 80%, the stack efficiency, using methane as
the fuel, would amount to 53.6%.

Thermal Plant Efficiency


As the intention is often to use an SOFC system as a
combined heat and power plant (CHP), the thermal efficiency, eth, is of importance. It describes the relation
between the amount of waste heat (used by the consumer
for heating purposes) and the energy fed to the system by
the fuel gas:
eth

where Pth,

net

Pth;net
m  LHV

8

is the net used thermal power.

Inverter Efficiency
Because the inverter is the most important electronic
plant component, it is normally considered separately.
The inverter efficiency describes the loss when adapting

Total Efficiency
As there is competition between SOFC systems and
conventional CHP plants, the total efficiency, etot, is often

Fuel Cells Solid Oxide Fuel Cells | Systems

used for comparison. It describes the relation between the


sum of usable thermal power and electrical power and
the energy fed to the system by the fuel gas:
etot

PAC;net Pth;net
eel eth
m  LHV

9

Some of these efficiencies can be influenced by the system configuration and the layout of the single components. The main tasks of the system layout are to find a
solution for low internal consumption and to create a
robust configuration that allows high fuel utilization.

Energy Balance of an Solid Oxide Fuel


Cell Plant
The fuel cell plant is composed of the stack and plant
(BoP) components, of which the reformer plays a particularly important role in the energy balance.
The energy calculations are presented to demonstrate
the interdependencies between the different main plant
components. It is practical to use some simplifications:
reforming and shift reaction are completely on the
carbon
dioxide side, and no carbon monoxide is left

(no equilibrium calculations)


to simplify both explanation and understanding, the
entering heating energy (LHV) is set to 100 kW.

These simplifications will not replace detailed flow sheet


calculations, in most cases performed using commercial
codes, but they are useful to gain understanding of the
way in which parameter changes affect the system.
The following point must be borne in mind: when
doing an enthalpy balance of flows containing chemical

reactions, the amount of energy (reaction enthalpy) released by the reaction has to be taken into account as
another source of heat. The correct result will be obtained only if the enthalpy scale of all media involved is
related to the same standard conditions. Independent of
the temperature at which the reaction really takes place,
all reaction enthalpies have to be taken at standard
conditions (1013 mbar, 25 1C). This is because of the
definition of enthalpy, and reaction enthalpy in particular
(Kirchhoff s law), as a function of temperature.
Balancing the Changes in Reaction Enthalpy
To be in a position to determine the stack efficiency (the
product of cell efficiency and fuel utilization), one must
take into account which reaction enthalpy (heating value)
is entering the system and which reaction enthalpy and
electrical power are leaving the system. For this purpose,
a control volume is drawn around the reformer (including the shift reactor) and stack, as shown in Figure 1
The reaction enthalpy H in, which enters the system
with the fuel, is
H in m CH4 ;in  LHVoCH4 100 kW

Assuming a complete reforming and shift reaction, 4


mol of hydrogen are produced out of each mole of
methane.
Based on the LHVs of methane (803 kJ mol1) and
hydrogen (242 kJ mol1), there is an increase in reaction
enthalpy at the reformer outlet from 100 kW to
H ref ;out H S;in

4  LHVoH2
 H in 1:20  100 kW
LHVoCH4
120 kW

Q H2,rea

Balancing border

42.2 kW

Stack

Vp = 0.7 V
o

x LHV CH
4,in
4

mCH

100 kW

Reformer + Shift
Conversion rate 100%

Qref

20kW

mH

xLHV H
2,in
2

120 kW

101

uF = 80%
Hrea = 96 kW

PDC

53.8 kW

Figure 1 Control volume for balancing the changes in reaction enthalpy of a reformer with a stack.

mH

2,out

xLHVoH

24 kW

10

102

Fuel Cells Solid Oxide Fuel Cells | Systems

This difference of 20 kW (or increase by 20%) has to be


provided to the reformer and shift reactor as heating
energy Qref .
In the case of a fuel utilization uF of 80%, the enthalpy
H rea reacted in the stack is

One advantage of an SOFC system is that, because of the


high operating temperature, some of the heat contained
in the exhaust gases leaving the stack can be used to heat
the reformer. This, for example, happens during internal
reforming. This makes it possible to further increase
the fuel utilization in an SOFC stack. Only the electrochemical behavior of the cell prevents fuel utilization
exceeding 90%. In the case of low and mean temperature
fuel cells, fuel utilization is limited because the reformer
can be heated only by burning additional fuel. So
the maximal possible fuel utilization would be 100/
120 83%. Even this is only a theoretical value because
in reality there are heat losses and the gases and components have to be heated to the reaction temperature. A
realistic fuel utilization for these types of fuel cells is
around 70%.

H rea m H2 ;in  LHVoH2  uF 120 kW  0:8 96 kW

Operating at a mean cell voltage Vp of 0.7 V and based on


the heating voltage VLHV of hydrogen of 1.25 V, the
produced electrical power PDC is determined as
PDC H rea  ec H rea 

Vp
H rea  0:56 53:8 kW
VLHV

Relating this to the incoming hydrogen results in a stack


efficiency es of
es

PDC
53:8

44:8% uF  ec;H2
120
H S;in

Energy Balance of a Plant with External


Reforming
For balancing the whole plant, in addition to the energy
fluxes depicted in Figure 1, all incoming and outgoing
enthalpy flows have to be added. For this purpose, the
flow scheme has to be extended as shown in Figure 2.
The energy balance now has to be resolved according
to the material carrying the remaining waste heat out of
the system. In the case of an SOFC, this is normally the
air flow on the cathode side, which is also used as a
coolant.
For this purpose, the complete balance of stack and
reformer has to be performed first.

This value would also be valid including the reformer, if


the reformer had to be heated by burning additional
methane to provide the 20 kW heating energy.
As the enthalpy of the nonreacted fuel leaving the
stack is high enough, in our case no additional fuel is
necessary. So the balance, drawn from the whole control
volume (stack plus reformer/shift), results in a stack
efficiency of
es

PDC 53:8
53:8% uF  ec;CH4

100
H in
Pel (AN1+AN2)

Balancing border

AC1
Air for
cooling

mair,c x h(Tin)

mair,c x h(TS,in)

AN2
is

Air
stoichiom. mO2,st x h(Tin)
mN ,st x h(Tin)

AN1

mO ,st x h(TS,in)
2
mN ,st x h(TS,in)
2

AC2

mCH x LHVCH
Methane

Reformer + shift

mH

2,S,in

x LHVH

Stack

TS,in = 750 C
TS,out = 850 C
Vp = 0.7 V
uF =80%

Conversion rate 100%

mCH4 x h(TR,in)

mH ,S,in x h(TS,in)
2
mCO ,S,in x h(TS,in)
2
mH O,S,in x h(TS,in)

TR,in = TR, out = 750 C

2O,R

x h(TR,in)

Steam

Qref

mN x h(TN ,out)
2
2
mO x h(TO ,out)
2

Figure 2 Control volume of a plant with external reforming.

mN

2,st

x h(TS,out)

mH

2,S,out

x LHVH

mH ,S,out x h(TS,out)
2
mCO ,S,out x h(TS,out)
2
mH O,S, x h(TS,out)

mH

mair,c x h(TS,out)

PDC

out

Fuel Cells Solid Oxide Fuel Cells | Systems

103

Balance of reformer

Balance of stack

Before entering the reformer, the methane is preheated


in heat exchanger AC2 to the reforming temperature
(e.g., 750 1C) together with the water vapor using the hot
off-gas. As long as this can be done by internal heat
transfer, this is not relevant for the overall balance.
The energy input H CH4 by methane into the reformer
is defined as

The outgoing flows on the anode side of the stack can be


determined as follows.
The mass flow of nonreacted fuel can be determined
by inlet flow and fuel utilization:

H CH4 m CH4

h

i
 LHVoCH4 cp  Tref ;in  T0

where T0 is the standard temperature and cp the specific


heat capacity. The necessary amount of water is determined by the chosen steam to carbon ratio (S/C), which
is normally in the region of 2.5. This avoids the risk of
carbon formation in the piping and components and is
sufficient for steam reforming and shift reaction:
Reforming CH4 H2 O-CO 3H2
Shift CO H2 O-CO2 H2

m H2 ;S;out m H2 ;S;in  1  uF

The effluent carbon dioxide mass flow corresponds to the


incoming carbon dioxide mass flow (assuming that there
is no back reaction from carbon dioxide to carbon
monoxide).
m CO2 ;S;out m CO2 ;S;in

The outlet steam mass flow is the combination of the


inlet water and the amount formed in the electrochemical cell operation. The latter can be calculated
using Faradays law if the electrical current is known or
by the number of moles of reacted hydrogen:
n H2 ;rea

Based on this, the mass flow of water vapor can be calculated as

m H2 ;S;in  uF
MH 2

m H2 O;S;out m H2 O;S;in n H2 ;rea  MH2 O

Based on these, the outgoing enthalpies on the anode side


can be calculated:

m H2 O;ref S=C  n CH4  MH2 O



H H2 O;ref m H2 O;ref  cp  Tref ;in  T0

h

i
H H2 ;S;out m H2 ;S;out  LHVoH2 cp  TS;out  T0


H CO2 ;S;out m CO2 ;S;in  cp  TS;out  T0


H H2 O;S;out m H2 O;S;out  cp  TS;out  T0

The amounts of hydrogen and carbon dioxide that are


created by reforming and shift reaction and the remaining amount of steam result in (the indices now refer
to the stack inlet, which is equal to the reformer outlet):

Now consider the cathode side of the stack.


The stoichiometric amount of oxygen n O2 ;st necessary
to react with the consumed hydrogen can easily be calculated based on the oxy-hydrogen reaction:

The enthalpy H H2 O;ref that enters the reformer by the


water vapor flow is

n O2 ;st 0:5  n H2 ;rea 0:5 

m H2 ;S;in 4  n CH4  MH2


m CO2 ;S;in n CH4  MCO2



2
m H2 O;S;in S=C  2  n CH4  MH2 O m H2 O;ref  1 
S=C

This can be used to calculate the enthalpies leaving the


reformer:
h

i
H H2 ;S;in m H2 ;S;in  LHVoH2 cp  TS;in  T0


H CO2 ;S;in m CO2 ;S;in  cp  TS;in  T0


H H2 O;S;in m H2 O;S;in  cp  TS;in  T0

Qref H H2 ;S;in H CO2 ;S;in H H2 O;S;in  H CH4  H H2 O;ref

So the stoichiometric oxygen mass flow becomes


m O2 ;st n O2 ;st  MO2

Using the simplified assumption that air consists of only


oxygen (23.3 wt%) and nitrogen, the corresponding
amount of nitrogen is
m N2 ;st

Using these enthalpies, the energy Qref , necessary for


heating the reformer, can be determined:
11

These calculations also provide the input data for the


anode side of the stack.

m H2 ;S;in
 uF
MH2

1  0:233
 m O2 ;st 3:29  m O2 ;st
0:233

Now the inlet and outlet enthalpies on the cathode side


can be calculated:
H O2 ;stS;in
H N2 ;stS;in

H N2 ;stS;out
DH air;S;c



m O2 ;st  cp  TS;in  T0


m N2 ;st  cp  TS;in  T0


m N2 ;st  cp  TS;out  T0


m air;c  cp  TS;out  TS;in

12

104

Fuel Cells Solid Oxide Fuel Cells | Systems

The enthalpy DH air;S;c that has to be removed by the


cooling air results from the balanced equation. To be able
to do this, the electrical energy produced in the stack has
to be known. This results from the product of mean cell
voltage Vp and electric current I determined by the
reacted hydrogen using Faradays law.
I m H2 ;rea

To be able to do this, two assumptions are made:


there is no cooling taking place by the media flow on
the
anode side (all media having the same inlet and

outlet temperature) and


there is no heat loss to the environment.

With these assumptions, the calculated air flow will be


somewhat too high (in the range of 1020%), but the
values obtained will be useful to illustrate how cell efficiency, fuel utilization, temperature difference, and internal reforming interact.

zF
MH2

So the electric power is


PDC Vp  I

The balanced equation, resolved to the enthalpy DH air;S;c


(to be removed by the cooling air), is as follows:


DH air;S;c H H2 ;S;in  H H2 ;S;out H N2 ;stS;in  H N2 ;stS;out




H O2 ;stS;in H CO2 ;S;in  H CO2 ;S;out


13
H H2 O;S;in  H H2 O;S;out  PDC

Energy balance of the plant

Having now determined all enthalpy flows for both the


stack and the reformer, the energy balance for the whole
plant can be worked out. For this purpose, all electrical
consumers in the plant have to be known. The BoP
component consuming the most electric power is the air
compressor (AN1 AN2).
The compressor power consumption PC can be determined by


T
PC m air  cp  T2  T1 m air;c m air;st  cp 
eis
0
1
 k1
p2 k
@

1A
p1

14

This makes it clear that the air mass flow necessary for
cooling is the main driver for the compressor power. The
second key parameter is the pressure increase p2 necessary to overcome the flow resistance in the system
which is again influenced by the mass flow of air. Based
on fixed geometries, this is more than proportional to the
air flow.
There are two methods for minimizing the necessary
amount of air for cooling. One is to increase the temperature difference across the stack between the inlet
and outlet. The other is to use internal reforming.
Both methods carry some risks with regard to thermomechanical stress resulting from excessive temperature
gradients; however, these are issues for stack designers
and material scientists. In the following sections, how
these two methods will influence the amount of air necessary for cooling is analyzed.

Air flow as a function of cell efficiency and


temperature difference across the stack

First of all, the surplus air flow coefficient l needs to


be defined. The coefficient of surplus air flow, l, when
multiplied with the stoichiometric air flow (necessary
to react the consumed fuel) gives the total air flow;
l of 1 means no additional air for the cooling and l of 2
means the same amount for cooling as for the electrochemical reaction. In this case, the oxygen content at the
end of the electrochemical active cell would be half, that
is, about 10 vol%. To ensure the safe operation of the
cathode, the oxygen content should not be reduced below
this value.
As there is no general rule, some refer to l to use the
total fuel flow entering the system; so this l is a factor of
uF smaller than the one used here.
Coolant Flow without Internal Reforming
To show the dependency of l on cell efficiency and
temperature difference, one has to derive a relation between these parameters. This can be done as follows:
The energy to be removed on the cathode side is
DH air;S n air  cp;air  DTair l  n air;st  cp;air  DTair

15

Based on the above-mentioned assumptions, this is equal


to the waste heat of the electrochemical reaction QH2 ;rea :
DH air;S l  n air;st  cp;air  DTair QH2 ;rea 1  ec
 n H2 ;rea  LHVoH2

Therefore
l 1  ec

n H2 ;rea LHVoH2
1


DTair
n air;st
cp;air

16

Then, from a defined relation between fuel consumption


and stoichiometric air flow
n air;st

1
1
  n H2 ;rea
0:21 2

Fuel Cells Solid Oxide Fuel Cells | Systems

relation [16] can be described as follows:





LHVoH2
1

l 0:42 
1  ec
DTair
cp;air

17

Equation [17] describes the necessary surplus air flow


without internal reforming (no heat loss, no cooling via
anode gas flow).
As can be seen from Figure 3, the necessary air flow
increases markedly in the case of only 50 K temperature
difference. This would result in an unacceptably high
power consumption by the air compressor. Without internal reforming (or operating with a fuel gas containing
only hydrogen and carbon monoixde), the temperature
difference permitted across the stack should be at least
150 K or better 200 K, especially if one has to operate at a
lower cell voltage to increase the power density of the
stack.
Figure 4 illustrates the influence of compressor power
consumption on the system efficiency. If one assumes a
moderate flow resistance of the cathode side of the system (air preheater, stack, off-gas components, and piping)
of 100 mbar at a l of 4, a linear increase of flow resistance
with flow rate (all laminar, no deflections, etc.), and a
compressor efficiency of 50%, already at a l of 10 the
compressor would consume 20% of the stack power.
Coolant Flow with Internal Reforming
To assess the cooling effect created by internal reforming, one must first derive a correlation between the
energy to be removed on the cathode side (heat produced) QH2 ;rea , the energy consumed by the reforming
process Qref , and the fuel utilization uF:
QH2 ;rea 1  ec  H H2 ;S;rea 1  ec  uF  H H2 ;S;in

18

105

In the case of a system where hydrogen does not enter


the stack as a fuel but is formed from methane fuel as a
result of internal reforming, eqn [18] becomes


QH2 ;rea 1  ec  uF  H CH4 ;S;in Qref

19

Now H CH4 ;S;in can be replaced by Qref because of the


relation
H CH4 Qref H H2 1:2  H CH4 ) H CH4 5  Qref

This leads to
QH2 ;rea 1  ec  uF  6  Qref

20

So the part of reaction heat that can be removed by the


endothermic reforming process can be described by
1
Qref

6

u

1  ec
Q H2 ;rea
F

21

This means that low fuel utilization takes out a larger


proportion of the reaction heat and that a lower cell
voltage requires higher additional air flow for cooling.
These relations are pointed out in Figure 5. This
diagram shows that it will not be possible to operate a
system at a cell voltage of 0.8 V and fuel utilization of
50%, because nearly all the heat would be removed by
the reforming process, and gas flows and heat loss would
cool down the stack below the operating temperature. In
the case of high fuel utilization and low cell voltage, the
cooling effect by the reforming process falls below 50%.
Within a reasonable range of operating parameters
(because of the goal of high system efficiency), the
cooling rate by internal reforming should be in the range
of 4060%.

40

T = 50 K
T = 100 K

35

T = 150 K
Surplus air flow ()

30

T = 200 K
T = 300 K

25

Lower limit

20
15
10
5
0
0.4

0.5

0.6

0.7

0.8

0.9

1.1

Cell voltage (V)

Figure 3 Surplus air flow for cooling as a function of cell voltage in the case of different temperature difference values across the stack
(without internal reforming).

106

Fuel Cells Solid Oxide Fuel Cells | Systems


100%
Compressor efficiency 50%
90%
Flow resistance increases proportional to air flow (laminar)
Compressor power in relation to
stack power

80%

Flow resistance at  = 4 is set to 100 mbar

70%
60%
50%
40%
30%
20%
10%
0%
0

10

12

14

16

18

20

Stoichiometric air flow coefficient ()

Figure 4 Compressor power related to stack power as a function of stoichiometric air flow coefficient l.

120%

100%

Reasonable range of operation

60%

ref

/Q

H2,rea

80%

u = 50%
F

40%

u = 60%
F

u = 70%
F

20%

u = 80%
F

u = 90%
F

0%
0.4

0.5

0.6

0.7

0.8

0.9

1.1

Cell voltage (V)

Figure 5 Ratio of cooling by internal reforming and reaction heat as a function of cell voltage in the case of different fuel utilization
rates.

Using relation [21], eqn [17] can be expanded by integrating internal reforming:

100 K. In the case of low-rated cell voltage, it should be


even higher.





LHVoH2
1
1

l 0:42 
 1  ec 
DTair
6  uF
cp;air

System Concepts

22

Now the necessary air flow can be calculated in cases


where there is internal reforming for different fuel
utilizations.
The comparison of the necessary air flow with and
without internal reforming in the case of fuel utilization
of 60% (Figure 6) and 80% (Figure 7) reveals two
things: (1) with internal reforming, fuel utilization has a
significant impact on air flow rate and (2) the temperature difference across the stack should still be at least

There are various system concepts that can potentially


meet the operational requirements and control loops
described above. These vary in terms of complexity and
also provide different levels of efficiency.
In the following the most common concepts are presented in a very basic way, describing the specific features
and their pros and cons.
Looking at the block diagrams of concepts 1 to 6
which follow, one can see that a high-temperature fuel

Fuel Cells Solid Oxide Fuel Cells | Systems


40

T = 50K
T = 100K
T = 150K
Lower limit
T = 50K
T = 100K
T = 150K

Fuel utilization 60%


35
30
Surplus air flow ()

107

25
20
15

Without internal
reforming

10
5
0
0.4

0.5

0.6

0.7

0.8

0.9

1.1

Cell voltage (V)

Figure 6 Surplus air flow for cooling as a function of cell voltage in the case of different temperature difference values across the stack
(with and without internal reforming): fuel utilization 60%.

40

T = 50K
T = 100K
T = 150K
Lower limit
T = 50K
T =100K
T = 150K

Fuel utilization 80%


35

Surplus air flow ()

30
25
20
15

Without internal
reforming

10
5
0
0.4

0.5

0.6

0.7

0.8

0.9

1.1

Cell voltage (V)

Figure 7 Surplus air flow for cooling as a function of cell voltage in the case of different temperature difference values across the stack
(with and without internal reforming): fuel utilization 80%.

cell system does not necessarily imply that the off-gas


temperature is also very high. As seen above, this can
result if a large proportion of the energy content in the
stack or burner off-gas is needed to preheat the cold air
and/or to produce the steam needed on the fuel side.

Concept 1 Basic Arrangement with a Catalytic


Burner
The plant components necessary for fulfilling the requirements for safe stack operation are depicted in the
block diagram in Figure 8. This diagram represents one
possible basic arrangement using a catalytic waste gas
burner.

At the fuel side, there is a gas supply system, which


contains a blower or compressor (depending on the
available feed pressure and the required operation pressure in the system), flow control, valves for start up and
shutdown, and safety valves for emergencies. Depending
on the type of gas and its source, special cleaning steps
are necessary. In most cases, sulfur compounds will be
present. These will result from the natural source or (in
the case of natural gas) from the sulfur-based odorants,
which are added for safety reasons. Sulfur is liable to
poison or react with nickel in nickel-cermet-based anode
materials, and various claims have been made about
purely oxide anode materials, but for now it is prudent to
err on the side of caution. Thus, a desulfurization device
is needed to reduce the sulfur content to as low a value as

108

Fuel Cells Solid Oxide Fuel Cells | Systems


Interface to the central supply and environment:
System border line for energy balance

Hot water
steam
Waste heat
exploitation

Steam
generation

Waste
gas

Catalytic burner

Housing/insulation
Preheating

Preheating

SOFC Stack

Prereforming
Inverter
Gas cleaning/
humidification

Gas supply
system

Fuel

Internal
consumption
Signals
Measurement
& control

Air supply
system

Electric output AC

Air

Figure 8 Block diagram of a simple arrangement using a catalytic burner. SOFC, solid oxide fuel cell; AC, alternating current.

is reasonably possible. Estimates as to the tolerable sulfur


content vary, but for long-term operation sub-ppm values
are probably more appropriate. At room temperature,
this can be zeolites or activated carbon impregnated with
metallic promoters or, at increased temperatures of about
300350 1C, zinc oxide can be used. Other critical components can be tars or silane, which might be present in
various biogases. These have to be removed or cracked
into less critical compounds by additional cleaning steps.
For carbon-containing fuel gases, a certain amount of
water vapor has to be added to avoid carbon formation
during heating up by disproportionation of carbon
monoxide via the so-called Boudouard reaction
2CO-C CO2

I

or by pyrolysis reactions such as


CH4 -C 2H2

II

Here the most critical temperature range is between 500


and 700 1C, at which methane has its highest carbon
activity, as can be seen from Figure 9. This requires that
the S/C ratio should be higher than the nominal value of

1.5. Normally, a value of 22.5 is chosen to be on the safe


side. A higher value would require more energy for vaporization and would further reduce the Nernst potential, which means lower cell efficiency.
After humidification, hydrocarbon gases should be fed
to a prereformer unit. This prereformer can be directly
heated or the gas can be preheated using the anode offgas. A prereformer is absolutely necessary in cases where
the gas contains higher hydrocarbons, as these have a
strong tendency to crack when they are heated above
500600 1C even at higher S/C ratios. This means that
soot is formed, which can block catalytic active surfaces
or even block piping or channels inside the stack. In
addition, cell tests at Siemens have shown that at least a
low amount of hydrogen is necessary at the stack inlet;
otherwise, the nickel anode tends to get destroyed in the
gas inlet area. A prereforming rate of 1020% was found
to be sufficient. This rate of reforming occurs at a temperature of about 450 1C, as can be seen from the equilibrium calculation presented in Figure 10. Owing to the
kinetic behavior of the reforming process, the actual reforming rate will always be somewhat lower than the one
calculated by thermodynamics.

Fuel Cells Solid Oxide Fuel Cells | Systems

109

1.60
p = 1 bar

1.40
1.20

S/C

1.00
0.80
Critical range for carbon formation
(carbon activity >1)

0.60
0.40
0.20
0.00
100

200

300

400

500

600

700

800

900

1000

1100

Temperature (C)

Figure 9 Critical steam to carbon ratio in the case of methane as a function of temperature and operating pressure.

0.7
0.6
CH4

Composition

0.5

H2O
CO2

0.4

H2
CO

0.3
0.2
0.1
0
350

400

450

500

550

600

650

700

750

800

Temperature (C)

Figure 10 Equilibrium composition of methane/steam mixture (S/C 2.5) as a function of temperature.

When leaving the reformer at about the reformed


operation temperature, whether or not an additional
anode gas heating stage is necessary is a question of stack
design and of the stacks mechanical properties. Owing
to the low heat capacity of the anode gas, it should be
possible in most cases to avoid an additional heat exchanger. After the prereformer heating stage, exhaust
gases can pass on to the catalytic after-burner to combust
any remaining fuel in the gas mixture.
At the air side of Figure 8, there is the air supply
system, which includes the input-blower/compressor
(depending on the required operation pressure in the
system), flow control, valves for start up and shutdown,
and safety valves. A standard air filter, normally used in
front of a fan, is sufficient for cleaning the ambient air. In

special environments, e.g., high salt content at the sea


side, additional measures may be required, but little work
has been done on this. One of the most bulky and costly
components in the SOFC system is the recuperative air
preheater. Using the cathode off-gas directly to preheat
the incoming air to the required inlet temperature requires a large heat exchanger, as only the temperature
difference across the stack is available to heat up the cold
air. An additional point to note is that the mass of cathode
off-gas available to transfer heat is smaller than the input
mass; as a proportion of the oxygen has been removed
from that gas inside the cell during cell functioning, this
smaller mass exacerbates the problem. These elements
will probably result in a large air preheater component,
parameterized by the permissible temperature gradient

110

Fuel Cells Solid Oxide Fuel Cells | Systems

across the stack. The advantage of this configuration is


that the highest temperature of the heat exchanger
is determined by the stack outlet temperature, which is
normally lower than the outlet temperature of the afterburner, as described in concept 3. This lower temperature
reduces the thermomechanical and corrosion stress to the
heat exchanger.
Cathode off-gases will probably leave the heat exchanger at a temperature below 200 1C (dependent on
DT across the stack) and are fed to the waste gas burner.
In managing the air stoichiometry in the after-burner to
extract as much energy as possible from the unspent fuel
in the anode gases, a certain amount of the cathode offgas can be bypassed and can directly fed into the heating
system. The relatively low temperature of the post heat
exchanger cathode gases and the low calorific value of the
depleted anode off-gas mean that a catalyst needs to be
employed in the after-burner to assist ignition. The advantage of this design is that the off-gas will always burn,
independent of conditions and composition. This catalytic solution, however, has to be weighed against the
costs of using a noble metal catalyst material. In some
cases, it might be advantageous to condense water out of
the anode off-gas before sending it to the after-burner.
This could increase the thermal efficiency of the system,
if the condensation enthalpy of the condensed water can
be transferred into the heating system. This will be
dependent on the actual temperature values in various
parts of the system.
The energy content of the off-gas leaving the burner
can be used to produce the steam necessary for reforming. The remaining energy can be transferred to the
district heating system, producing hot water and/or
steam.
The produced DC current has to be transferred into
grid-compatible AC current using an inverter. For control of the system and for data acquisition, appropriate
components are necessary.
Concept 2 Basic Arrangement with a Thermal
Burner
Figure 11 shows a modification of concept 1, using a
thermal burner instead of a catalytic burner. Thermal
burner here means that the gases are mixed at a temperature above the ignition point of hydrogen (560 1C),
methane (595 1C), and carbon monoxide (620 1C), so
there is no need for a catalyst. However, a catalyst may
still be employed to enable cold start up.
The main advantage of this configuration is that the
outlet temperature of the thermal burner can be higher
than the stack outlet temperature (especially in the case
of planar stack technology, which aims at reduced operating temperature). The burner outlet temperature can
vary between 850 and 1000 1C depending on the surplus

flow of air and fuel utilization. Combined with the increased mass flow, this results in a smaller size of air
recuperator. For example, the size can go down to onethird in the case of a hot temperature of 850 1C instead of
700 1C, heating the cold air to 600 1C. A disadvantage
may be the increased corrosion stress inside the heat
exchanger and the thermal burner itself.
Concept 3 Anode Gas Recycling
An important part of a systems complexity is the need
for steam production for the steam reforming reaction.
This requires additional components and consumes quite
a large amount of heat, most of which cannot be recovered by the waste heat exploitation system. Thus, it is
strongly advised that systems use water, which is already
available in the system, namely that in the anode off-gas.
This requires that a part of the anode off-gas is recycled
and mixed with the cold fuel supplied to the system. This
is illustrated in Figure 12.
As a second effect, this also means that a part of the
fuel leaving the stack is fed back to the reaction zone,
which increases the fuel utilization in the system. How
the recycling ratio influences the fuel utilization can be
derived from the following equation:
uF;sys

u
 F;S

1  R 1  uF;S

23

where the recycling rate R is defined as the ratio between


the mass flow of the recycled fuel m F;R and the mass flow
of the fuel m F;S;out leaving the stack:
R

m F;R
m F;S;out

24

Figure 13 shows the resulting system fuel utilization


uF,sys as a function of the recycling rate R in the case of
different stack fuel utilizations uF,S. For reasonable stack
fuel utilizations from 50 to 70% (which means a low risk
for too low fuel partial pressure somewhere inside the
stack due to inhomogeneous flow distribution), system
fuel utilization in the range of 8090% can be achieved
using an anode off-gas recycling rate between 65 and
75%. For example, the Siemens 125 kW system employed a recycling rate of 65% combined with a stack
fuel utilization of 65% resulting in a total utilization of
about 84%. This generates a significant improvement in
overall system efficiency.
It is not straightforward to source suitable components
for the recycling system due to the high temperature of
the fuel off-gases. One possibility is to use a jet pump,
driven by pressurized fuel entering the system. Siemens
employs this method, operating the jet pump at about
700 1C. The advantage of a jet pump is that there are no
moving parts, which makes it quite robust and suitable
for high temperature. The disadvantages are a poor

Fuel Cells Solid Oxide Fuel Cells | Systems


Interface to the central supply and environment:
System border line for energy balance

111

Hot water/
steam

Waste heat
exploitation

Steam
generation

Waste
gas

Thermal burner

Housing/insulation
Preheating

Preheating

SOFC Stack

Prereforming
Inverter
Internal
consumption

Gas cleaning/
humidifications
Signals

Gas supply
system

Fuel

Measurement
& control

Air supply
system

Electric output AC

Air

Figure 11 Block diagram of a simple arrangement using a thermal burner. SOFC, solid oxide fuel cell.

efficiency and a poor controllability under part-load


conditions. Another option is to cool down the anode gas
for recycling to a level at which rotary-type devices can
be operated. However, the disadvantage with this option
is that an additional heat exchanger is necessary, implying extra cost and bulk.
Concept 4 Pressurized Hybrid
As the after-burner off-gas temperature is above 800 1C,
it is possible to use it to drive a turbine generating
additional electricity. The corresponding block diagram
is shown in Figure 14, where the off-gas of the afterburner is directly fed to the turbine that drives the
compressor for the air supply. The turbine off-gas is then
used to heat the incoming compressed air to the required
stack inlet temperature. A directly coupled SOFC/gas
turbine plant has to be operated at increased pressure to
ensure a sufficiently high SOFC off-gas pressure as it
enters the turbine stage. To suit the operating conditions
of microturbines, a pressure increase of about 22.5 bar

should be used. There is an additional reason for this


pressure level found in the overall thermodynamics. As
can be seen from Figure 15, starting from 835 1C, expansion in the turbine leads to an outlet temperature
between 660 and 730 1C, depending on the isentropic
efficiency eis, of the turbine. Because the off-gas is then
needed to heat up the cold incoming air to about 600
700 1C (depending on the stack design), a much higher
pressure is forbidden. The hot off-gas expands in the
microturbine and drives the compressor and generator,
which provides the additional electrical energy. Assuming a microturbine generator (MTG) efficiency of
30% (which is the upper limit for such small low-pressure machines), the increase in electrical system efficiency as a function of cell voltage is shown in Table 1
(based on rough estimations concerning parasitic consumption, which is comparably low, because the air
compressor is driven by the turbine).
A coupled SOFC/gas turbine system can yield an
increase in efficiency of between 15%-points (at 600 mV)
and 9%-points (at 850 mV), whereas the total efficiency

112

Fuel Cells Solid Oxide Fuel Cells | Systems


Interface to the central supply and environment:
System border line for energy balance

Hot water/
steam

Waste heat
exploitation

Waste
gas

Thermal burner

Housing/insulation
Preheating

Preheating

SOFC Stack

Prereforming
Inverter
Internal
consumption
Gas cleaning

Gas supply
system

Signals
Measurement
& control

Air supply
system

Electric output AC

Fuel

Air

Figure 12 Block diagram of an arrangement incorporating anode gas recycling.

100%

System fuel utilization u

F,sys

90%

80%
u

= 50%

= 60%

= 65%

= 70%

= 75%

= 80%

= 90%

= 100%

F,S

70%

F,S
F,S

60%

F,S
F,S
F,S

50%

F,S
F,S

40%
0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Recycling rate (R)

Figure 13 System fuel utilization uF,sys as a function of the recycling rate R in the case of different stack fuel utilizations uF,S.

decreases with a lower cell voltage (from 68 to 57%).


Also in this context, it is still an important research target
to further improve the cell behavior such that it can
generate a high power density at high cell voltage.
The same is true for fuel utilization. The ratio of
turbine power and SOFC power compared to the total
power is shown in Figure 16 in the case of fuel utilization

of 85% and in Figure 17 in the case of fuel utilization


of 70%.
The comparison of the total power output in the case
of different fuel utilizations shown in Figure 18 illustrates that, in a pressurized hybrid system, the total
power output decreases with lower fuel utilization,
although the power of the MTG increases.

Fuel Cells Solid Oxide Fuel Cells | Systems


Interface to the central supply and environment:
System border line for energy balance

113

Hot water/
steam

Waste heat
exploitation

Electric output AC

Microturbine
generator

Waste
gas

Thermal burner

Housing/insulation
Preheating

Preheating

SOFC Stack

Pressure vessel
Prereforming
Air compressor
Inverter
Internal
consumption

Gas cleaning
Signals
Gas supply
system

Air supply
system

Measurement
& control

Fuel

Electric output AC

Air

Figure 14 Block diagram of an arrangement using direct coupling with a microturbine (pressurized hybrid). SOFC, solid oxide fuel cell.

Temperature (turbine outlet) (C)

900

Turbine in: 850 C

800

Isentropic
turbine
efficiency

700
60%

Turbine out: min. 620 C !


600

Air preheater cold side out: 580 C

85%
100%

500
0.50

1.00

1.50

2.00

2.50

Pressure (turbine inlet) (bar)

Figure 15 Pressurized hybrid turbine requirements.

p_max

3.00

3.50

114
Table 1

Fuel Cells Solid Oxide Fuel Cells | Systems

Effect of coupling solid oxide fuel cell (SOFC) with microturbine generator (MTG) as a function of cell voltage

Cell
voltage
(V)

Cell
efficiency
(%)

Stack
efficiency
(%)

Electrical
efficiency
SOFC
(%)

Heating energy in offgas (behind burner)


related to system
input (%)

Electrical power
of MTG related
to system input
(%)

Total
electrical
efficiency
(%)

Ratio of
electrical
output MTG
to SOFC (%)

Power output
MTG in case of
1000 kW gas
input (kW)

0.600
0.650
0.700
0.750
0.800
0.850
0.900

57.7
62.5
67.3
72.1
76.9
81.7
86.5

49.0
53.1
57.2
61.3
65.4
69.5
73.6

44.3
47.9
51.6
55.3
59.0
62.7
66.4

51.0
46.9
42.8
38.7
34.6
30.5
26.4

15.3
14.1
12.8
11.6
10.4
9.2
7.9

59.5
62.0
64.5
66.9
69.4
71.9
74.3

34.5
29.3
24.9
21.0
17.6
14.6
11.9

153
141
128
116
104
92
79

Input data: efficiency of MTG 30%, fuel utilization of SOFC 85%, parasitic efficiency of SOFC 95%, inverter efficiency of SOFC 95%.

80%
u = 85%
Electrical power related to gas input

70%

Total power

60%
SOFC power

50%
40%
30%
20%

MTG power

10%
0%
0.55

0.60

0.65

0.70

0.75

0.80

0.85

0.90

0.95

Cell voltage (V)

Figure 16 Pressurized hybrid electric power related to gas input as a function of cell voltage in the case of fuel utilization of 85%.
SOFC, solid oxide fuel cell.

70%
u = 70%

Total power

Electrical power related to gas input

60%
50%
SOFC power

40%

30%
20%
10%
0%
0.55

MTG power

0.60

0.65

0.70

0.75

0.80

0.85

0.90

0.95

Cell voltage (V)

Figure 17 Pressurized hybrid electric power related to gas input as a function of cell voltage in the case of fuel utilization of 70%.
SOFC, solid oxide fuel cell; MTG, microturbine generator.

Fuel Cells Solid Oxide Fuel Cells | Systems

115

80%

Electrical power related to gas input

75%

Total power with u = 85%


F

70%
65%
60%
Total power with u = 70%

55%

50%
45%
40%
35%
30%
0.55

0.60

0.65

0.70

0.75

0.80

0.85

0.90

0.95

Cell voltage (V)

Figure 18 Pressurized hybrid electric system power related to gas input as a function of cell voltage in the case of fuel utilization of
70% and 85%.

Interface to the central supply and environment:


System border line for energy balance

Electric output AC

Microturbine
generator

Hot water/
steam

Heat exchanger

Waste heat
exploitation

Waste
gas

Thermal burner

Housing/insulation
Preheating

Preheating

SOFC Stack

Prereforming
Inverter
Internal
consumption

Gas cleaning
Signals
Gas supply
system

Fuel

Measurement
& control

Electric output AC

Air supply
system
Air

Figure 19 Block diagram of an arrangement using indirect coupling with a microturbine (atmospheric hybrid). SOFC, solid oxide fuel
cell; AC, alternating current.

116

Fuel Cells Solid Oxide Fuel Cells | Systems

A further positive effect of pressurized operation is


the increase of power density of the SOFC up to about
20% (dependant on operating conditions). However, at
the same time, this pressurized hybrid operation requires
a higher effort/cost for the system components (e.g.,
pressure vessel with hot feedthroughs) and system control, because of coupling two systems with completely
different dynamic behaviors. Special measures have to be
taken to avoid a rapid pressure drop because of the risk of
turbine failure that this poses.
The concept described as concept 5 tries to avoid
these problems.

operation (which means a comparably simple system)


and there is no direct influence of the turbine behavior
on the fuel cell system. On the one hand, this gives more
flexibility in the layout of the turbine (e.g., the pressure
ratio can be higher) and, on the other hand, an additional
air fan (with additional power consumption) for the
SOFC is needed and there is some additional temperature loss from the heat exchanger. These effects lead to a
reduced gain in electrical efficiency, which is only half
that of the pressurized hybrid.

Concept 5 Atmospheric Hybrid

Concept 6 Pressurized Hybrid Combined with


a Steam Turbine

As illustrated in Figure 19, the gas turbine is not integrated into the gas loop of the fuel cell but is coupled
indirectly via a heat exchanger. The hot off-gas leaving
the after-burner heats the driving gas for the turbine via
this heat exchanger. This enables an atmospheric SOFC

For efficient utilization of the energy content of the


waste gas to produce electricity, the design concept
shown in Figure 20 can be adopted. In this case, waste
gas leaving the air preheater is not used for district
heating but to produce steam, which in turn drives an

Interface to the central supply and environment:


System border line for energy balance

Microturbine
generator

Electric output AC

water

Electric output AC

Steam turbine
Waste
gas

Thermal burner

Housing/insulation
Preheating

Preheating

SOFC Stack
Pressure vessel

Prereforming
Air compressor
Inverter
Internal
consumption

Gas cleaning
signals
Gas supply
system

Fuel

Measurement
& control

Electric output AC

Air supply
system

Air

Figure 20 Block diagram of an arrangement using direct coupling with a microturbine and a steam turbine (pressurized hybrid II).
SOFC, solid oxide fuel cell.

Fuel Cells Solid Oxide Fuel Cells | Systems

117

Table 2
Effect of coupling solid oxide fuel cell (SOFC) with a microturbine generator (MTG) and a steam turbine (ST) as a function
of cell voltage
Cell
voltage
(V)

Electrical
efficiency
SOFC
(%)

Heating energy
in off-gas
(behind
burner) related
to system input
(%)

Electrical
power of
MTG
related to
system
input (%)

Total
electrical
efficiency
(%)

Power
output MTG
in case of
1000 kW
gas input
(kW)

Heating
energy in offgas behind
MTG related
to system
input (%)

Electrical
power of
ST related
to system
input (%)

Total
electrical
efficiency
(%)

Power
output ST
in case of
1000 kW
gas input
(kW)

0.600
0.650
0.700
0.750
0.800
0.850
0.900

44.3
47.9
51.6
55.3
59.0
62.7
66.4

51.0
46.9
42.8
38.7
34.6
30.5
26.4

15.3
14.1
12.8
11.6
10.4
9.2
7.9

59.5
62.0
64.5
66.9
69.4
71.9
74.3

153
141
128
116
104
92
79

35.7
32.8
30.0
27.1
24.2
21.4
18.5

7.1
6.6
6.0
5.4
4.8
4.3
3.7

66.7
68.6
70.5
72.4
74.2
76.1
78.0

71
66
60
54
48
43
37

Input data: efficiency of ST 20%; efficiency of MTG 30%; fuel utilization of SOFC 85%; parasitic efficiency of SOFC 95%; inverter efficiency of SOFC 95%.

90%
u = 85%
Electrical power related to gas input

80%

Total power

70%
60%

SOFC power

50%
40%
30%
20%
MTG power
10%
0%
0.55

ST power

0.60

0.65

0.70

0.75

0.80

0.85

0.90

0.95

Cell voltage (V)

Figure 21 Pressurized hybrid combined with steam turbine (ST) electric power of the different plant components related to gas input
as a function of cell voltage in the case of fuel utilization of 85%. SOFC, solid oxide fuel cell; MTG, microturbine generator.

ST. Owing to the low waste gas temperature behind


the air preheater (probably less than 300 1C), even in the
case of a 30 1C condensation temperature the Carnot
efficiency is only around 40%. Being an optimistic, an ST
efficiency of 20% can be predicted. The simplified calculation giving the latter number only indicates the approximate potential of this system configuration. It is
possible to design more effective but complex arrangements; however, the resulting efficiency gains will be
somewhere in the range shown in Table 2. An increase of
about 47%-points is feasible, meaning a total electrical
system efficiency of about 7075% (see Figure 21). What
can also be seen from Table 2 is the size of the ST in the
case of 1 MW gas input. This would result in an ST
power of about 4060 kW. Realistically, the coupled
SOFC/MTG/ST system is viable only for plants with a
gas input above 10 MW.

Conclusions
In this review, some basic arrangements of SOFC systems
are described, starting with atmospheric systems using a
catalytic burner or a thermal burner and anode gas recycling. For illustrating the potential electrical efficiency
of SOFC systems, the combinations with a gas turbine
and also with an ST are described. To be able to evaluate
the potential of the different systems, first the essential
efficiencies relevant for fuel cell systems are defined and
then the basics for calculating energy balance are illustrated. Equations are given to describe, for example, the
effect of fuel recycling on system fuel utilization or the
effect of internal reforming on the necessary air flow for
cooling the stack.
It is obvious that electrical efficiency strongly depends
on cell voltage and fuel utilization. If cells that operate

118

Fuel Cells Solid Oxide Fuel Cells | Systems

with a high fuel utilization at cell voltages of 800 mV are


available, a net electrical efficiency above 55% can be
achieved. The combination in a pressurized system
with a gas turbine enables efficiencies of up to 70%
and combining this system with an additional ST allows
efficiencies of up to 75%. However, as an investigation
of the size of these STs shows, such combined systems
(SOFC/MTG/ST) make sense only above a gas input
of 10 MW.

Nomenclature
Symbols and Units
cp
F
h
H
H
H CH4
H in
H rea
H ref;out
H S;in
IDC
Iin
Itot
LHV1
M
Mi

m
F;R
m
F;S;out
m
i;in
m
in
m
rea
m
n
p
PAC,gross
PAC,net
PAC,parasite
PC
PDC

specific heat capacity at constant


pressure (J mol  1 K)
Faraday constant: 96 485 As mol  1
specific enthalpy (J mol  1 or J g  1)
enthalpy (J)
enthalpy of mass flow (J s  1)
enthalpy flow entering the reformer
(J s  1)
enthalpy flow entering the system
(J s  1)
enthalpy flow reacted in the stack
(J s  1)
enthalpy flow leaving the reformer
(J s  1)
enthalpy flow entering the stack (J s  1)
current produced by the stack (A)
current that could be produced by the
gas entering the system (A)
current produced by the reaction
(current per cell  number of cells) (A)
lower heating value at standard
conditions (J mol  1)
molar mass
molar mass of component i (g mol  1)
amount mass flow of fuel gas
mass flow of recycled fuel
mass flow of fuel leaving the stack
mass flow of fuel component i entering
the system (kg s  1)
mass flow amount of fuel entering the
system (kg kg s  1)
mass flow amount of fuel reacted in the
cell (kg s  1)
molar flow (mol s  1)
pressure (bar)
gross power behind the inverter (W)
net power behind the inverter (W)
electric power consumed by the system
(feeder behind the inverter) (W)
compressor power consumption
DC electric power

PDC,gross
Pel
Pth,net
Q
Q ref
Q H2 ;rea

R
T
T0
uF
uF,S
VLHV
Vp
zi
DH
DT
ec
eel
eInv
eis
ep
es
eth
etot
k

gross power in front of the inverter


(stack power) (W)
electric power
net used thermal power (W)
heat (J)
heat flow to be supplied to the reformer
(J mol  1)
heat flow produced by the
electrochemical reaction of H2
(J mol  1)
recycling rate
temperature (K or 1C)
standard temperature (25 1C)
utilization of fuel (%)
stack fuel utilization
LHV heating voltage (V)
(practical) cell voltage (V)
equivalent number of component i
reaction enthalpy (J mol  1)
temperature difference (K)
cell efficiency
electrical efficiency
inverter efficiency
isentropic efficiency
parasitic efficiency
effective or stack efficiency
thermal efficiency
total efficiency
coefficient of surplus air flow

Abbreviations and Acronyms


AC
AN
BoP
CHP
DC
LHV
MTG
PAFC
PEMFC
SOFC
ST

alternating current
compressor No. in flow scheme
Balance of Plant
combined heat and power plant
direct current
lower heating value
microturbine generator
phosphoric acid fuel cell
proton-exchange membrane fuel cell
solid oxide fuel cell
steam turbine

See also: Fuel Cells Solid Oxide Fuel Cells: Internal


and External Reformation.

Further Reading
Blum L, Peters R, David P, Au SF, and Deja R (2004) Integrated stack
module development for a 20 kW system. In: Mogensen M (ed.)
Proceedings of Sixth European Solid Oxide Fuel Cell Forum, pp.
173182. Lucerne: European Fuel Cell Forum.
Finkenrath M (2005) Simulation und Analyse des dynamischen
Verhaltens von Kraftwerken mit oxidkeramischer Brennstoffzelle

Fuel Cells Solid Oxide Fuel Cells | Systems

(SOFC), vol. 44, ISBN 3-89336-414-5. Julich, Germany:


Forschungszentrum.
Finkenrath M, Lokurlu A, Blum L, and Stolten D (2005) Modelling the
dynamic behaviour of a planar SOFC CHP system. In: Mogensen M
(ed.) Proceedings of Sixth European Solid Oxide Fuel Cell Forum, pp.
569578. Lucerne.
Gubner A (1996) Modelling of High Temperature Fuel Cells: The
Thermal, Chemical, Electrochemical and Fluidmechanical Behaviour
of Solid Oxide Fuel Cells Operating with Internal Reforming of
Methane. Thesis, University of Portsmouth, UK.
Gubner A, Landes H, Metzger J, Seeg H, and Stubner R (1997)
Investigations into the degradation of the cermet anode of a solid
oxide fuel cell. In: Stimming U, Singhal SC, Tagawa H, and Lehnert
W (eds.) Proceedings of the Fifth International Symposium on Solid
Oxide Fuel Cells (SOFC-V), pp. 844850. Aachen: The
Electrochemical Society, Inc.
Hartvigsen JJ, Elangovan S, and Khandkar AC (2003) System design.
In: Vielstich W, Lamm A, and Gasteiger HA (eds.) Handbook of Fuel
Cells Fundamentals, Technology and Applications, ISBN 0-47149926-9, pp. 1070--1085. John Wiley & Sons, Ltd.
Larminie J and Dicks A (2000) Fuel Cell Systems Explained. ISBN
0-471-49026-1. John Wiley & Sons Ltd.
Lundberg WL and Veyo SE (2001) Conceptual design and performance
analysis of a 300 MWE LNG-fueled pressurized SOFC/gas turbine

119

power plant. In: Yokokawa H and Singhal SC (eds.) Proceedings of


Seventh International Symposium on Solid Oxide Fuel Cells, 38
June 2001, Tsukuba, Japan.
Lundberg WL, Veyo SE, and Moeckel MD (2001) A High-Efficiency
SOFC Hybrid Power System Using the Mercury 50 ATS Gas Turbine.
ASME Paper 2001-GT-521.
Palsson J (2002) Thermodynamic Modelling and Performance of
Combined Solid Oxide Fuel Cell and Gas Turbine Systems. ISBN 91628-5247-7. Lund Institute of Technology.
Sriramulu S, Teagan P, Stratonova M, and Thijssen J (2003) MW-class
hybrid power system based on planar solid oxide stack technology.
Proceedings of Seventh International Symposium on Solid Oxide
Fuel Cells, 38 June 2001, Tsukuba, Japan.
Unverzagt G (1995) Energetische und wirtschaftliche Optimierung von
Festoxid-Brennstoffzellen-Kraftwerken; Jul-3159, ISBN 0944-2952.
Veyo, SE, Litzinger KP, Vora SD, and Lundberg WL (2002) Status of
Pressurized SOFC/Gas Turbine Power System Development at
Siemens Westinghouse. ASME Paper 2002-GT-30670.
Veyo SE, Lundberg WL, Vora SD, and Litzinger KP (2003) Tubular Sofc
hybrid power system status. Proceedings of ASME Turbo Expo
2003: Power for Land, Sea, and Air. Atlanta, Georgia, USA. ASME
Paper 2003-GT-38943.

You might also like