Efficiency Calculation and Configuration Design of
Efficiency Calculation and Configuration Design of
net/publication/266056306
CITATIONS READS
55 7,519
4 authors:
Some of the authors of this publication are also working on these related projects:
Solar cells; Fuel cells; Quantum heat engines; Thermoelectric devices, etc. View project
All content following this page was uploaded by Jincan Chen on 18 January 2015.
International Journal of
ELECTROCHEMICAL
SCIENCE
www.electrochemsci.org
A PEM electrolyzer system for hydrogen production is established and the corresponding efficiency is
derived. Based on semi-empirical equations, thermodynamic-electrochemical modeling of water
splitting reaction is systematically carried out. It is confirmed that the Joule heat resulting from the
irreversibilities inside the PEM electrolyzer is larger than that needed in the water splitting process in
the whole region of the electric current density. Some alternative configurations are designed to
improve the overall performance of the system and the corresponding expressions of the efficiency are
also derived. The curves of the efficiency varying with the electric current density are presented and
the efficiencies of the different configurations are compared. The optimally operating region of the
electric current density is determined. The effects of some of the important parameters on the
performance of the PEM electrolyzer system are analyzed in detail. Some significant results for the
optimum design strategies of a practical PEM electrolyzer system for hydrogen production are
obtained.
1. INTRODUCTION
Hydrogen has been considered as an ideal energy carrier to support sustainable energy
development, which can be effectively produced based on fossil-fuel-based processes, e.g. ethanol,
methane and natural gas reforming [1-5]. However, the development of renewable hydrogen
production technologies to replace fossil fuel-based hydrogen production methods is an important step
towards a sustainable hydrogen economy [6-9].
Water electrolysis integrated with clean energy sources such as nuclear power stations,
photovoltaics or wind turbines is a more practical technology for large-scale renewable hydrogen
Int. J. Electrochem. Sci., Vol. 7, 2012 4144
production. Up until now, most of the research and development on water electrolysis related to
renewable hydrogen production projects have focused on alkaline electrolysis systems and PEM
electrolyzers. Compared with traditional alkaline electrolysis systems, PEM electrolyzers are more
advantageous due to their ecological cleanness, easy maintenance, compactness, etc.[10, 11] Recently,
most studies on PEM electrolysis emphasize on the development of new catalysts [12-15], cell and
stack assembly for high efficiencies [16-19]. Comparatively, the detailed heat recovery strategies on
systematical efficiency and performance analysis are rarely investigated, which is worthwhile to be
further studied.
In the present paper, a PEM electrolyzer system is established and the efficiencies of the
system are presented for differently operating conditions. Based on semi-empirical equations,
thermodynamic-electrochemical modeling of the electrolysis process is systematically carried out,
where multi-irreversibilities inside the system are taken into account. In order to improve the
performance of the system, some alternative configurations for utilizing the redundant heat produced
in the PEM electrolyzer are put forward. The efficiencies of these configurations are derived and the
lower boundary of the operating current density is determined. The effects of some of the important
parameters such as the operating temperature, electrolyte membrane thickness, effectiveness of heat
exchangers, and inlet flow rate of H2O on the performance of the PEM electrolyzer system are
analyzed in detail.
Figure 1. The schematic diagram of a PEM electrolyzer system for hydrogen production.
Int. J. Electrochem. Sci., Vol. 7, 2012 4145
Figure 1 shows the schematic diagram of a PEM electrolyzer system for hydrogen production.
The system is mainly composed of a PEM electrolyzer, a separator, a pump and two heat exchangers.
In order to drive the water splitting reactions, water, electricity and heat, if necessary, are provided to
the PEM electrolyzer. The generated H2 flows out at the cathode and the generated O2 and residual
H2O flow out at the anode. The waste heat remained in the products at the outlet can be utilized
through the use of heat exchanger 1. After leaving heat exchanger 1, H2 is cooled down to reference
condition (298.15 K and 101,325 Pa), compressed and stored as fuel. The O2/H2O mixture flows into
the separator, O2 is cooled down and can be used as by-product; and the hot H2O is circulated in the
H2O supply stream at reference condition and pumped into heat exchanger 1 for the next H2 production
cycle. Due to the different thermodynamic parameters of reactants/products and the inefficiency of
heat exchanger 1, the feeding water should be further heated through heat exchanger 2 before reaching
the temperature of the PEM electrolyzer. Such a model is more general and reasonable than the
conventional ones [20] as it includes main energy consumption processes of hydrogen production and
the waste heat produced in the electrolyzer can be efficiently utilized.
According to Fig. 1, the efficiency of the PEM electrolyzer system for hydrogen production can
be expressed as
N H 2 ,out LHV H 2
T , (1a)
T0 T0
E QPEM (1 ) QH 2O (1 )
Tx Ts
where N H 2 ,out is the outlet flow rate of H2; LHV is the lower heating value of H2; E is the
electric energy input; QPEM is the thermal energy provided to the PEM electrolyzer or released to the
environment, QH 2O is the thermal energy input to the second heat exchanger for further heating up
H2O, T0 and Ts are, respectively, the temperatures of the environment and external heat source
supplying heat for heating up H2O, Tx is the temperature of the external heat source, the PEM
electrolyzer, or the environment and depends on the sign of QPEM , H2 1 T0 / TH , and TH is the
combustion temperature of H2 at reference condition. Because TH is independent of the parameters of
the PEM electrolyzer system, one can simply give another definition of the efficiency as
N H 2 ,out LHV
T / H . (1b)
2
T0 T0
E QPEM (1 ) QH 2O (1 )
Tx Ts
Equation (1a) or (1b) may be used to calculate the efficiency of a PEM electrolyzer system for
hydrogen production. The results obtained only have a constant difference which is independent of the
parameters of the system. Below, we directly use Eq. (1b) to calculate the efficiency of the system for
simplicity.
When QPEM 0 , Tx=Ts and Eq. (1b) can be directly written as
Int. J. Electrochem. Sci., Vol. 7, 2012 4146
N H 2 ,out LHV
. (2)
T0 T0
E QPEM (1- ) QH 2O (1- )
Ts Ts
When QPEM 0 and the redundant heat | QPEM | resulting from the PEM electrolyzer is directly
released to environment, Tx T0 and Eq. (1b) can be simplified as
N H 2 ,out LHV
. (3)
T0
E QH 2O (1- )
Ts
It is worthwhile to note that if the redundant heat | QPEM | resulting from the PEM electrolyzer is
utilized, one can improve the performance of the PEM electrolyzer system. We will specially discuss
this problem in the next few sections.
It should be pointed out that for a practical electrolyzer system for hydrogen production; the
hydrogen storage and the operation of some auxiliaries usually consume some additional energy. For
the sake of calculative convenience, the additional energy is not considered in Eqs. (1)-(3) because it is
small compared with the input electric energy of the electrolyzer.
where△ G(T) is the electrical energy demand, i.e., the change in the Gibbs free energy,
Q(T)=T△ S(T) is the thermal energy demand, △ S(T) is the entropy change in the water splitting
reaction process, and T is the operating temperature of the PEM electrolyzer.
According to Refs. [24-27], the reversible voltage of a PEM water electrolysis process can be
determined by Nernst equation, i.e.,
PH PO
V0 1.229 8.5 10 4 (T T0 ) 4.3085 10 5 T ln 2 2 , (5)
PH O
2
Int. J. Electrochem. Sci., Vol. 7, 2012 4147
where PH 2 and PO2 are the partial pressures of hydrogen and oxygen, respectively.
For a practical PEM electrolyzer, the cell voltage required in the electrolytic hydrogen process
is always bigger than the reversible voltage owing to the existence of irreversibilities resulting mainly
from the following three overpotential losses.
(1) The activation overpotential losses [19, 24-28]
α A + αC RT j
Vact = ( ) ln( )
α AαC 2 F j0 , (6)
where j0 1.08 1017 exp(0.086T ) is the exchange current density [27], j is the current
density, R is the universal gas constant, F is Faraday’s constant, and A and C are the charge transfer
coefficients of the anode and cathode, respectively.
(2) The ohmic overpotential losses [19, 28-31]
tmem
Vohm j , (7)
mem
1 1
where tmem is the membrane thickness, mem (0.005139mem 0.003260) exp[1268( )] is
303 T
0.043 17.81a 39.85a 2 36a3 , 0 a 1,
the conductivity of a Nafion membrane [31], mem is the
14 1.4(a 1), 1 a 3.
membrane humidity [31, 32], a Pw / Psat is the membrane water activity, Pw is the partial pressure of
5
(T 273)2 1.4454107 (T 273)3
water vapor, and Psat 1.01325 102.82060.02953(T 273)9.183710 is the water saturation
pressure [26, 31, 33]. In a PEM electrolyzer, the membrane always operates under almost full
humidification, i.e. a 1 [33].
(3) The concentration overpotential losses [34]
2
j
Vcon j 1 , (8)
jL
E IV , (10)
and
where I jAp is the electric current through the PEM electrolyzer and Ap is the effective
surface area of the polar plate.
It is significant to note that the Joule heat produced in the PEM electrolyzer may be directly
used to supply to the water splitting reactions and hence the heat input of the external heat source can
be decreased. Thus, the thermal energy transferred from the external heat source to the electrolyzer
may be reduced from Q to
QPEM (Q QJ ) . (12)
According to Faraday’s law, the rate of electrochemical reaction depends on the operating
electric current [35, 36], i.e.,
dn I
, (13)
dt 2 F
where n is the number of moles and dn/dt is the molar consumption rate of H2O. Based on Eq.
(13), the outlet flow rates of H2, O2 and H2O are, respectively, given by
I
N H 2 ,out 2 N O2 ,out N H 2O,reacted (14)
2F
and
I
N H 2O,out N H 2O,in N H 2O,reacted N H 2O,in . (15)
2F
According to Fig. 1, the heat supplied by the external heat source to heat the compensatory
water is given by:
N H 2O,in N
T0 C p,H2O dT
T T T T
QH 2O
T0 2
C p , H O dT
2 T0
H , out C p , H 2
dT N O2 ,out T0 2
C p ,O dT N H 2O , out
I T I C p ,O2
N H2O,in 1/ 1
T
2 F T0
C p , H 2O dT
2F T0
(C p , H 2
2
)dT , (16)
Int. J. Electrochem. Sci., Vol. 7, 2012 4149
where is the effectiveness of the heat exchangers and C p ,i (i= H2, O2 or H2O) are the molar
heat capacities of species i. It may be easily proved from Eq. (16) and the data in Table 1 [37-39] that
QH 2O 0 .
Table 1. Thermodynamic parameters for the reactants/products at 101,325 Pa, where (g) and (l) refer
to gas and liquid phases, respectively.
Parameter Value
Faraday constant, F (C mol-1) 96,485
Universal gas constant, R (J mol-1 K-1) 8.314
Low heating value of H2, LHV (J mol-1) 241,830
50×10-6
Membrane thickness tmem (m)
Transfer coefficient of the anode, A 0.5 [15]
Transfer coefficient of the cathode, C
1 [15]
Using the above equations and the data in Tables 1 and 2 [10, 26, 31, 40-43], we can obtain the
curves of Qk (k=PEM and TE) varying with j, as shown in Fig. 2, where
Int. J. Electrochem. Sci., Vol. 7, 2012 4150
QTE QPEM (1 T0 / T ) QH2O (1 T0 / Ts ) and jQ is the current density when QTE 0 . It is clearly seen
from Fig. 2 that when the PEM electrolyzer works at the temperatures between 353 K and 373 K,
QPEM 0 , QTE 0 in the region of j jQ and QTE 0 in the region of j jQ . When the operating
temperature is increased, QPEM decreases.
Figure 2. The curves of QPEM and QTE varying with the current density for different temperatures.
A
Int. J. Electrochem. Sci., Vol. 7, 2012 4151
Figure 3. The part schematic diagrams of a PEM electrolyzer system for hydrogen production with
(A) QTE 0 and (B) QTE 0 . The other parts of the system are the same as those in Fig. 1
It is interesting to note that when QPEM 0 , besides Fig. 1, one can design some alternative
configurations for a PEM electrolyzer system, as shown in Fig. 3. It is seen from Fig. 3(A) that when
QPEM 0 and QTE 0 , the redundant heat QPEM generated in the PEM electrolyzer may be transferred
to heat the water through an additional heat exchanger, i.e., heat exchanger 3. In such a case, Tx T
and Eq. (1b) can be directly expressed as
N H 2 ,out LHV
. (17)
T T
E QPEM (1 0 ) QH 2O (1 0 )
T Ts
In Fig.3 (A), the heat QPEM (1 T0 / T )Ts / (Ts T0 ) , which has the same temperature level as the
external heat source connected to heat exchanger 2, is transferred to the water, so that the heat supplied
by the external heat source can be reduced from QH 2O to QH2O QPEM (1 T0 / T )Ts / (Ts T0 ) . It is seen
from Fig. 3(B) that when QPEM 0 and QTE 0 , one part of the redundant heat QPEM may be used to
replace the heat QH 2O supplied by the external heat source and to heat the water so that the temperature
of the water attains that of the PEM electrolyzer, and the other part of the redundant heat QPEM is
Int. J. Electrochem. Sci., Vol. 7, 2012 4152
released to the environment. In such a case, neither Eq. (1b) nor Eq. (17) may be directly used to
calculate the efficiency of the system, while the efficiency of the system should be expressed as
N H 2 ,out LHV
. (18)
E
In Fig. 3(B), the quantity of the heat transferred from the PEM electrolyzer to heat exchanger 2
is equal to QH2O QPEM (1 T0 / T )Ts / (Ts T0 ) and the quantity of the heat released to the environment
is QPEM QH2O QPEM 1 (1 T0 / T )Ts / (Ts T0 ) .
Unless otherwise described specifically, the following numerical calculations are performed
based on the data in Tables 1 and 2 and Eqs. (3), (17) and (18). According to Figs. 1 and 3, one can
generate some characteristic curves of the efficiency of the system varying with the current density
under differently given parameters, as shown in Figs. 4 and 5.
Figure 4. The curves of the efficiency of the system varying with the current density for different
configurations. Curves I and II correspond to the configurations of Fig. 1 and Fig.3(A) and (B),
respectively.
Int. J. Electrochem. Sci., Vol. 7, 2012 4153
It is seen from Fig. 4 that when QPEM <0, the efficiency of the system (Curve II) shown in Fig.
3 is higher than that (Curve I) shown in Fig. 1. It is taken for granted because the redundant heat
QPEM generated in the PEM electrolyzer is utilized in Fig. 3. Thus, in Fig. 5, we only discuss the
performance of the system configurations shown in Fig. 3.
B
Int. J. Electrochem. Sci., Vol. 7, 2012 4154
Figure 5. The efficiency versus current density curves for different (A) operating temperatures, (B)
membrane thickness, (C) effectiveness of the heat exchangers, and (D) inlet flow rates of H2O.
Int. J. Electrochem. Sci., Vol. 7, 2012 4155
It is also seen from Fig. 4 that for a PEM electrolyzer system, there is a maximum efficiency
ηmax and a corresponding current density jm . The curves in Fig. 4 show clearly that in the region
of j < jm , the efficiency of the system decreases rapidly as the current density is decreased, as shown
by the left side of curve II which is generated by Eq. (17) corresponding to Fig. 3(A); while in the
region of j > jm , the efficiency of the system increases as the current density is decreased, as shown by
the right side of curve II which is generated by Eq. (18) corresponding to Fig. 3(B). When j = jm , Eq.
(17) is equal to Eq. (18) and jm is equal to jQ in Fig. 2. It means the fact that one should choose the
configuration shown in Fig. 3(B) so that the PEM electrolyzer system is operated in the region
of j ≥ jm . For the parameters given in Table 2, the value of jm is equal to 806 A/m2. It should be
pointed out that jm is the lowest bound of the optimized current density. In order to obtain a larger
hydrogen production rate, the current densities in the practical PEM electrolyzer systems [16, 44, 45]
are usually chosen to be much larger than jm , which is in agreement with the optimal condition of the
current density.
The influence of the operating temperature on the performance of the system is shown in Fig.
5(A). When j ≥ jm , the efficiency of the electrolysis system increases as the operating temperature is
increased. Moreover, the maximum efficiency and the current density corresponding to the maximum
efficiency increase monotonically with the increasing of the operating temperature.
The influence of the electrolyte membrane thickness on the performance of the system in the
optimal current density region is shown in Fig. 5(B). It shows that the efficiency of the electrolysis
system decreases as the electrolyte membrane thickness is increased. Furthermore, the influence of the
electrolyte membrane thickness on the efficiency of the system increases as the current density is
increased.
The influence of the effectiveness of the heat exchangers on the performance of the system is
shown in Fig. 5(C). The efficiency of the system increases as the effectiveness of the heat exchangers
is increased. However, the influence mainly happens in the region of j jm . When j>jm, the redundant
heat generated in the electrolyzer QPEM is larger than the thermal energy QH 2O needed from the
external heat sources, so that the influence of the effectiveness of the heat exchangers can be neglected.
The influence of the inlet flow rate of water on the performance of the electrolysis system is
shown in Fig. 5(D). It shows that the efficiency of the system decreases as the inlet flow rate of water
is increased. Similarly, the influence mainly happens in the region of j jm , and this influence can be
negligible in the other regions.
5. CONCLUSIONS
A PEM water electrolysis system for hydrogen production is established and the efficiencies of
the system are given for the cases of QPEM 0 and QPEM 0 . Based on semi-empirical equations, the
irreversible losses in a PEM water electrolysis process are calculated. It shows that in the whole
operating current density range, QPEM 0 . Some alternative configurations for utilizing the redundant
heat resulting from the PEM electrolyzer are put forward to improve the performance of the system. It
Int. J. Electrochem. Sci., Vol. 7, 2012 4156
is found that the efficiencies of the alternative configurations are higher than those that the redundant
heat is directly released into the environment. It is expounded that the operating current density for the
alternative configurations should be situated in the region j > jm . Moreover, it is revealed that the
efficiency of the system increases as the operating temperature and the effectiveness of the heat
exchangers are increased and decreases as the electrolyte membrane thickness and the inlet flow rate of
H2O are increased. The optimum design strategies obtained here may enrich thermodynamic and
electrochemical theories of a PEM electrolyzer and provide some guidance for the optimum design and
operation of practical PEM electrolysis systems for hydrogen production.
ACKNOWLEDGMENTS
This work was supported by the Fujian Natural Science Foundation and the Fundamental Research
Fund for the Central Universities (No. 201112G006), People’s Republic of China.
References