of Al-alloy bipolar plate and pH value of K.H. Hou

2011 International Conference on Environment Science and Engineering
IPCBEE vol.8 (2011) © (2011) IACSIT Press, Singapore
Analysis on the corrosion behavior of Al-alloy bipolar plate and pH value of
water product for the PEMFC
K.H. Hou
M.D. Ger
Department of Power Vehicle and Systems Engineering
National Defense University
Taoyuan, 335 Taiwan
[email protected]
Department of Chemical and Materials Engineering
National Defense University
Taoyuan, 335 Taiwan
[email protected]
C.H. Lin*
S.W. Shiah, H.M. Chou
Graduate School of Defense Science
[email protected]
Department of Power Vehicle and Systems Engineering
National Defense University
cathode side and react with the oxygen gas. The reaction at
the positive electrode is:
Eo=1.229 V (2)
O2 + 4 H + + 4e − → 2 H 2O
The fuel cell performance considerably degraded with the
chemical catalyst change over the proton membrane. Cai et
al. [1] had proved platinum catalyst would react with water
and form free radical ●OH. The carbon dioxide (CO2) and
hydrofluoric acid (HF) form easily when the free radical
●
OH react with the perfluorosulfonic membrane at high
temperature. Above result involved the pH value change of
water that drained from the fuel cell as following reactions:
Abstract—The energy crisis and global warming are two
urgent problems in the world currently. Owing to these major
advantages of high efficiency and low emission, proton
exchange membrane fuel cell (PEMFC) has become a
promising candidate of power sources. The performance of Alalloy bipolar plate for the PEMFC system was investigated in
this paper. Due to the light weight and easy machining, the
5052 Al-alloy is considered as the good candidate for the
bipolar plate substrate. In the experimental analysis,
aluminum bipolar plates without coating were measured by
current-voltage curve and constant current test. Furthermore,
the chronological water acidity varied tardily and its pH value
ranged 4.2 to 5 initially. The more acid product would corrode
the aluminum bipolar plate severely and shorten the fuel cell
lifetime. After the experiment operated for 100 hours, the
bipolar plate corrosion behavior and membrane degradation
decrease the pH value and cell performance. It was available
that the fuel cell performance degradation can be quantified by
our method .These above results will be useful for the metallic
bipolar plate preliminary design.
2 R f − CF2COOH + •OH → R f − CF2 • + CO2 + H 2O
R f − CF2 • + •OH → R f − CF2OH → R f − COF + HF
R f − COF + H 2O → R f − CF2OH + HF
There were many researches focused on the degradation
membrane to increase the H2O2 formation rate [2-9] in the
fuel cell. It has been found that the deteriorated membrane
would enhance the water acidity at the cathode compartment.
Sethuraman et al. [7] measured the H2O2 concentration under
the constant temperature and relative humidity. It can be
predicted as a function of humidity and temperature by
studying the oxygen reduction reaction. Abundant H2O2
formed when the experiment set as higher temperature,
relative humidity and oxygen concentration. The free radical
formed when the H2O2 reacted with metal ions. Meanwhile,
another mechanism was discussed by the electrochemical
reaction potential in PEMFC as below:
H 2 O2 + M 2+ → M 3+ + OH • + OH −
Keywords-PEMFC, pH value, aluminum bipolar plate, water
I.
INTRODUCTION
Fuel cells are used for power generation with low
emission and high efficiency. The major advantages of
proton exchange membrane fuel cell (PEMFC) are lowtemperature operation, quick starting, high energy density,
clean, and simple design. Therefore, the PEMFC extensively
applied to power generation, portable electric equipment,
ship and hybrid vehicles. The fuel cell provided the electric
power when it was supplied with hydrogen and oxygen gas.
Electrons and H+ ions released from the hydrogen gas
consumption at the anode side. The reaction at the negative
electrode is:
2 H 2 → 4 H + + 4e −
(1)
O 2 + 2 H + + 2e − → H 2 O 2
Eo=0.695 V
(3)
Hence, the H2O2 formation rate may not too little owing
to the high electrochemical drive force. The
perfluorosulfonic membrane would be attacked when the
oxygen reduction reaction occurred. It could be inferred that
H2O2 formation rate involved the pH value change. Abdullah
et al. [10, 11] discussed the water character that drained from
the fuel cell at different relative humidity. The very acidic
The hydrogen gas electrochemical reaction was enhanced
by the catalysis. Electrons flow from the anode to cathode
side via the external circuit. Hydrogen ions transport to the
313
water drained from the cathode compartment easily under the
low relative humidity (35%). But, the opposite trend
occurred at high relative humidity (100%). Furthermore, the
pH value of water changed with the load owing to the
chemical reaction rate. The water approximated the
neutrality when the load operated at large current density
except for the low temperature and low relative humidity
condition. The pH value of water chronological change was
attributed to the ●OH production rate of the selectivity in the
oxygen reduction reaction. The lower temperatures would
enhance the stability of H2O2 molecules in the fuel cell.
Oppositely, it was obvious that the pH value fluctuated
severely at high temperature. Above result indicated the
lower proton crossover reaction was caused by the higher
concentration hydrogen peroxide. Meanwhile, the hydrogen
peroxide affected the cell performance significantly. Hentall
et al. [12] considered the complex flow-field and adopted a
new process technique for making the bipolar plate. Hence,
they suggest the material fabricated into the bipolar plate
must be machined easily. The stainless steel was coated with
a thin gold that offered superior power performance than the
graphite. The gold film has the good corrosion resistance and
electricity conductivity property. The current-voltage (I-V）
curve had proven the metallic bipolar plate was really
feasible. There were also many researches focused on the
chronological change of the metallic bipolar plate [13-15] in
the fuel cell. This accomplishment would benefits to the
metallic bipolar plate preliminary design.
II.
Figure 1. The 5052 Al-alloy bipolar plate morphology
III.
RESULTS AND DISCUSSION
A. Assembly test
After the PEMFC operation environment test as Table I.
The chronological performance change of 5052 Al-alloy was
recorded as Fig 2. The performance decayed with the
operation time due to the variation on the sheet surface. The
single cell performance would decay as the thicker oxide
film and the corrosion over the sheet surface .The relation
between the single cell performances corresponds to
mathematical model as Eq (4).The model of PEMFC with an
empirical equation was induced by Kim [16].
E = Eo - blog(i) - Ri
（4）
The term E0 yields the electrode kinetic parameter for
oxygen reduction in the PEMFC. The term b is the Tafel
parameter for the oxygen reduction and R represents the
ohmic resistance. The 5052 Al-alloy substrate performance
data were fitted as Fig 3. The root mean square error value
approximate 9.47E-04 the Tafel parameter for the oxygen
reduction and R represents the ohmic resistance. The single
cell matched well with the mathematical model. These fitting
values of metallic bipolar plates were shown as Table I.
Basically, the R value increased with time as a whole and
displayed linearization property as Fig 4.
EXPERIMENT
The manufactured PEMFC used in this experiment has
been assembled with commercial membranes （ Nafion112）, aluminum bipolar plates and gas diffusion layers. The
active area of the Nafion 112 membrane set as 25cm2 (5cm
× 5 cm). The fuel gas was supplied with oxygen (140
c.c/Min) in the cathode side and hydrogen gas (210 c.c/Min)
in the anode side. The temperature of fully humid flow gas
and cell were set as 60oC. The range of electric load set as
0.85 to 0.2V (voltage scan rate was set as 0.1 V/hour) and
the limit current set as 33A.The water produced from
PEMFC cathode was measured by the pH meter (from
HACH Inc Benchtop series model number H260G).The
acidity of water droplet was recorded as per thirty seconds
and the average value was obtained for twenty points.
The single serpentine gas flow bipolar plate made from the
5052 Al-alloy was machined as Fig 1. The inset shows the
SEM topographies of the 5052 Al-alloy bipolar plate. Owing
to the light weight and ease of machining, the aluminum is
considered as the good candidate for the bipolar plate.
314
B. Corrosive behavior of the metallic bipolar plate
The SEM result under the fuel cell operation was shown
as Fig 5. Both bulks in the different environment have the
identical morphology over the sheet surface.
Figure 2.
Polarization curve chronological change under the experiment
setup
(a)Substrate in the Cathode side
(b)Substrate in the Anode side
Figure 3.
Fitting curve for 5052 Al-alloy polarization curve
Figure 5. Metallic bipolar plate morphology under the experiment setup
after 50 hours operation
In fact, the oxide film covered on aluminum plate surface
was thicker with time owing to the higher O2 concentration
in the cathode environment shown as Fig 5a. By the way, the
oxide film on the anode surface would not develop easily
under the H2 environment as Fig 5b. The analysis by
examined with EDAX, results stated that oxide in the
cathode side (54.5%) much higher than in anode side
(10.78%). Above results verified the acid electrolyte result in
serious corrosion in the fuel cell as Fig 5.
Figure 4.
TABLE I.
C. The time series analysis of the fuel cell
It takes 50 hours single cell operation to attain the stable
system before the time series was recorded. The cell on load
operated at constant current (100mA/cm2) and recorded as
chronological data (50-55 hours). The time series was
recorded per minute to analyze the cell voltage loss trend.
The chronological change stated the fuel cell could not keep
at the constant voltage as Fig 6. There are two reasons led to
the cell voltage loss. Firstly, the long term descend trend was
due to the ohmic resistance loss. The temperature set at 333K
which is fairly standard for the operation of PEM fuel cells.
Owing to the acid environment, the thicker oxide film
formed in the cathode side and corrosive phenomenon
occurred in the anode. The interfacial contact resistance and
dielectric constant would vary with time. Furthermore, the
Parameter R variation in the fitting model
THE FITTING PARAMETER OF CELLS UNDER THE
EXPERIMENT SETUP
No Time(hours)
1
2
3
4
5
0
15
30
50
100
EO（V）
b（V/dec）
R（mOhm）
0.927
0.895
0.924
0.934
0.937
0.03196
0.02729
0.02374
0.03579
0.02426
0.33
0.96
1.29
1.48
2.60
mean square
error
9.47E-04
8.53E-04
1.74E-03
1.21E-03
1.42E-03
315
ohmic loss led to the overall fuel cell performance descends
tardily.
Secondly, the short term variation was affected by the
operation condition. The cell voltage involved by the
electrochemical reaction, mass transfer phenomenon, catalyst
layer, temperature and pressure. The reaction stayed at
equilibrium when the forward and backward reaction
occurred simultaneously. Hence, the cell voltage perturbation
is considered as a random signal.
The descend trend occurred obviously in single cell. The
graph of the cell voltage data fitting were used by the linear
model as Eq (5-6):
(5)
Tr = β o + β1t + ε
Tˆr = β o + β1t
(6)
Figure 7. The performance and pH value change of water about the cell.
0.675
IV.
Variable
Actual
Fits
0.670
⑴ The aluminum bipolar plate possess weak corrosion
resistance in the high temperature and low pH value
environment. These results were valid by the time series
analysis and the mathematical model which induced by
Kim.
⑵The fuel cell environment led to the different corrosion
behavior in two sides (Anode and Cathode side). The
oxide film would increase the interfacial contact
resistance and descend the overall performance. The
thicker oxide film formed in the cathode side than the
anode.
⑶ It was obvious the bipolar plate corrosion behavior and
membrane degradation decrease the pH value and cell
performance. Furthermore, the pH value chronological
change of water could be an important index of the
membrane degradation. By the way, the metallic
components (end plates, collectors and fuel inlet) in the
fuel cell were corroded by the acid water easily. As for the
cell lifetime, it is important to manipulate the cell at
moderate thermal condition and decrease the ● OH
production. These above results will benefit to the metallic
bipolar plate preliminary design.
0.665
0.660
0.655
0.650
1
30
60
90
120
150
180
Time(Min)
210
240
270
CONCLUSIONS
300
Figure 6. The chronological trend of the fuel cell voltage
Where T and ε are known as the trend component and
error, respectively [17]. The over all performance durability
test was analyzed by the above model. A linear regression
fitted as voltage data at 100 mA/cm2 for the fuel cell
operating period of the experiment (from 50 to 55 hours)
voltage decay rates β1 of single cell was 35.4 μV / min .The
experiment initial value β O was 0.665 Volt. Above result
verified the aluminum bipolar plates possess weak corrosion
resistance in the high temperature and low pH value
environment.
D. pH character and cell performance
The chronological water acidity was stated by pH value
change and ranged from 4.2 to 5 initially as Fig 7.
Furthermore, abundant metal ions release rate would enhance
●
the water acidity owing to abundant free radical
OH
production. The more acid product would corrode the
metallic material severely and shorten the fuel cell lifetime.
Hence, the degradation of the cell performance could be
observed at Fig 7.
At current density 250 mA/cm2, the cell voltage was 0.52
(V) and the pH value of water was 4.7. After the Experiment
operated for 100 hours, the cell voltage and pH value
decreased to 0.2 (V) and 3.9, respectively. It was obvious the
bipolar plate corrosion behavior and membrane degradation
decrease the pH value and cell performance. Furthermore,
the pH value chronological change of water could be an
important index of the membrane degradation.
REFERENCES
[1]
[2]
[3]
[4]
[5]
316
M. Cai, M. S. Ruthkosky, B. Merzougui, S. Swathirajan, M. P.
Balogh and S. H. Oh,“Investigation of thermal and electrochemical
degradation of fuel cell catalysts,” J Power Sources , vol. 160, pp.
977–986, 2006.
A. Collier, H. Wang, X. Z. Yuan, J. Zhang and D. P. Wilkinson,
“Degradation of polymer electrolyte membranes,” J Hydrogen
Energy , vol. 31, pp. 1838–1854, 2006.
M. Aoki, H. Uchida and M. Watanabe, “New Evaluation Method for
Degradation Rate of Polymer Electrolytes,” J Electrochemical
Society , vol. 3(1), pp. 485-492, 2006.
M. Schulze, N. Wagner, T. Kaz and K. A. Friedrich, “Combined
electrochemical and surface analysis investigation of degradation
processes in polymer electrolyte membrane fuel cells,” J
Electrochimica Acta , vol. 52, pp. 2328–2336, 2007.
V. A. Sethuraman, J. W. Weidner, A. T. Haug, S. Motupally, L. V.
Protsailo, “Hydrogen peroxide formation rates in a PEMFC anode
and cathode effect of humidity and temperature,” J Electrochemical
Society, vol. 155(1), pp. B50-B57, 2008.
[6]
A. M. Abdullah., T. Okajima, F. Kitamura and T. Ohsaka, “A simple
in situ characterization technique for the onset of the chemical
degradation of PEM fuel cells’ fluorinated membranes,” J
Electrochemistry Communications, vol. 10, pp. 1732-1735, 2008.
[7] V. A. Sethuraman., J. W. Weidner, A. T. Haug and Protsailo L.V.
“Durability of perfluorosulfonic acid and hydrocarbon membranes:
effect of humidity and temperature,” J Electrochemical Society, vol.
155(2), pp. B119-124, 2008.
[8] M. Inaba, M. Sugishita, J. Wada, K. Matsuzawa, H. Yamada and A.
Tasaka “Impacts of air bleeding on membrane degradation in polymer
electrolyte fuel cells,” J Power Sources, vol. 178, pp. 699-705, 2008.
[9] S. Hommura, K. Kawahara, T. Shimohira and Y. Teraoka
“Development of a method for clarifying the perfluorosulfonated
membrane degradation mechanism in a fuel cell environment,” J
Electrochemical Society, vol. 155(1), pp. A29-33, 2008.
[10] A. M. Abdullah, T. Okajima, F. Kitamura and T. Ohsaka “Effect of
operating conditions on the acidity of H2/Air PEM fuel cells’ water,”
J Electrochemical Society, vol. 16(2), pp. 543-550, 2008.
[11] A. M. Abdullah, A. M. Mohammad, T. Okajima, F. Kitamura, T.
Ohsaka “Effect of load, temperature and humidity on the pH of the
[12]
[13]
[14]
[15]
[16]
[17]
317
water drained out from H2/air polymer electrolyte membrane fuel
cells,” J Power Sources, vol. 190, pp. 264-270, 2009.
P. L. Hentall, J. B. Lakeman, G. O. Mepsted, P. L. Adcock and J. M.
Moore “New materials for polymer electrolyte membrane fuel cell
current collectors,”J. Power Sources, vol. 80, pp. 235-241, 1999.
D. R. Hodgson, B. May, P. L. Adock and D. P. Davies “New
lightweight bipolar plate system for polymer electrolyte membrane
fuel cells,” J. Power Sources, vol. 96, pp. 233-235, 2001.
J. Wind, R. Spah, W. Kaiser and G. Bohm “Metallic bipolar plates
for PEM fuel cells,” J. Power Sources, vol. 105, pp. 256-260, 2002.
H. Wang, M. A. Sweikart and J. A. Turner “Stainless steel as bipolar
plate material for polymer electrolyte membrane fuel cells,” J. Power
Sources, vol. 115, pp. 243-251, 2003.
J. Kim, S. M. Lee, and S. Srinivasan “Modeling of proton exchange
membrane fuel cell performance with an empirical equation,”
Electrochemical Society, vol. 142(8), pp. 2670-2674, 1995.
P. J. Brockwell, R. A Davis, “Introduction to time series and
forecasing,” pp. 9-13, Springer, New York, 1996.