Materials Chemistry and Physics 238 (2019) 121911

Contents lists available at ScienceDirect

Materials Chemistry and Physics

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

Improving performance in PEMFC by applying different coatings to metallic

bipolar plates
F. Madadi a, *, A. Rezaeian a, H. Edris a, M. Zhiani b
a
Department of Materials Engineering, Isfahan University of Technology, Isfahan, 84156-83111, Iran
b
Department of Chemistry, Isfahan University of Technology, Isfahan, 84156-83111, Iran

H I G H L I G H T S

� Effect of different coated material properties on performance improvement of PEM has been investigated.
� The NiCr, NiCrBSi and CoCrAlY were selected as a coating material because of their excellent corrosion resistance.
� High velocity oxy-fuel (HVOF) coating was applied for deposition on the Al plates to enhance bipolar performance in PEMFC.
� Corrosion resistance, interfacial contact resistance, current-voltage and durability tests conducted on coated plates.
� The experimental results have proven that NiCrBSi coating can be a candidate coating for PEMFC metallic bipolar plates.

A R T I C L E I N F O A B S T R A C T

Keywords: In the present study, aluminum plates were coated with three different materials using thermal spray in order to
HVOF coating be used as bipolar plate in proton exchange membrane fuel cells (PEMFC). NiCr, NiCrBSi, and (Co,Ni)CrAlY were
PEM fuel cell selected as coating materials due to their high corrosion resistance. High velocity oxy-fuel (HVOF) was the spray
Performance
coating method used in the present study. Accelerated corrosion resistance and interfacial contact resistance
Bipolar plate
(ICR) tests were conducted on the coatings under simulated PEMFC conditions. The corrosion current density of
the bipolar plates coated with NiCrBSi, NiCr, and (Co,Ni)CrAlY, on Al-substrate in the simulated cathodic
environment of a proton-exchange membrane fuel cell (PEMFC) were found to be 4.98 (1.03 E 5), 3.89
(0.00012), and 2.91 (0.0012) A/cm2, respectively. Also Ecorr for NiCr, NiCrBSi and (Co,Ni)CrAlY were 0.36,
0.31 and 0.39 (�0.01) volt. The electrochemical impedance spectroscopy (EIS) and polarization curves show
that the NiCrBSi coating has higher corrosion resistance than other coatings. The interfacial contact resistance
between the NiCrBSi and NiCr coated samples and carbon paper are the smallest on typical fuel cell stack
compression force of 140 N/cm2. Then, long-term durability tests were conducted on single cell with coated
bipolar plates. The results showed a considerably better performance and no decrease in cell voltage in the single
cell with NiCrBSi coated bipolar plates. It can be concluded that the performance of NiCrBSi alloy coated bipolar
plate developed using HVOF coating is acceptable in fuel cells as bipolar plate.

1. Introduction PEMFC system. This component is capable of uniform distribution of

reactive gases over active areas, collection of electrons generated
Fuel cells have recently been developed as devices, which convert through electrochemical reaction in the membrane electrode assembly
chemical energy into electric energy directly. For the time being, they (MEA), and removal of the heat and exhaust water from the fuel cell
can be a solution for serious concerns such as the shortage in fossil fuel [5–7]. In order to take advantage of the aforementioned capabilities,
resources and crises in environmental pollution. Proton exchange bipolar plates should possess specifications such as superior electrical
membrane fuel cell (PEMFC) is among the most promising types of fuel and thermal conductivity, high corrosion resistance and gas imperme­
cells [1–4]. ability, good mechanical performance, and low cost [8,9].
Bipolar plate is an important multifunctional component in the The plates can be made from different materials, the most common of

* Corresponding author.
E-mail addresses: [email protected] (F. Madadi), [email protected] (A. Rezaeian), [email protected] (H. Edris), [email protected] (M. Zhiani).

https://doi.org/10.1016/j.matchemphys.2019.121911
Received 21 March 2019; Received in revised form 22 July 2019; Accepted 26 July 2019
Available online 27 July 2019
0254-0584/© 2019 Elsevier B.V. All rights reserved.
F. Madadi et al. Materials Chemistry and Physics 238 (2019) 121911

which is stainless steel. Compared with stainless steel, bipolar plates 2. Experimental procedure
made of aluminum offer the advantages of lower density, lower bulk
electrical resistivity (about 25% of that of stainless steel), and higher In this study, Al plates (50 � 50 � 1.5 mm) were selected as sub­
thermal conductivity (8.5 times higher than that of stainless steel) [8, strate. NiCr, NiCrBSi, and (Co,Ni)CrAlY powders with an average grain
10]. In addition, it has been reported that the corrosion current density size of 25–55 μm were used as coating material. Prior to coating, Al-
of aluminum is 10 times higher than that of stainless steel 316 L [11]. substrates’ surface was blasted by sand followed by acetone cleaning to
Furthermore, aluminum bipolar plates can be fabricated more easily and increase adhesion. The sparing was carried out using HVOF thermal
at a lower cost [12]. Moreover, passive film (Al2O3) usually forms on the spray system (GTV/Top Gun). Kerosene and nitrogen gas were used as
surface of aluminum and causes a significant increase in interfacial fuels for combustion and powder carrier. The spray parameters are
contact resistance (ICR). Therefore, in order to benefit from the summarized in Table 1. The coating thickness was about
above-mentioned advantages, aluminum bipolar plates have to be sub­ 200 μm � 20 μm after about 5 gun passes.
jected to surface modification [13]. Furthermore, the performance of X-ray diffraction (XRD) with a monochromatic Cu-Kα radiation and
fuel cell declines due to corrosion. Corrosion may be observed in scanning electron microscopy (SEM) were used, respectively, for
metallic bipolar plates when used under acidic atmosphere in fuel cells, studying the different phases in the coating and the microstructure and
which leads to the dissolution of metallic ions. This phenomenon can fractographic evaluation of the coating. Coating porosity was measured
subsequently contaminate the membrane, catalyst, and electrode [14, using SEM. In this regard, 10 random backscattered images were taken
15]. Previous studies have shown that the performance of the catalyst from the coating cross section and the porosity level was determined
layer has significant impact on the overall performance of the cell in a accordingly using the Clemex TM image analysis software. Surface
PEMFC [16,17]. Carbon graphite is another well-known material used in roughness was measured using a surface profilometer (Mitutoyo Surftest
bipolar development. However, metallic materials (such as aluminum 402). A conventional three-electrode system was used to conduct the
bipolar) exhibit excellent electrical and thermal conductivity, mechan­ electrochemical experiment (corrosion test), using platinum net as a
ical properties, and workability as compared to carbon graphite. To counter electrode. For the simulation of the PEMFC environment, sam­
protect the aluminum from electrochemical corrosion while enhancing ple potential and saturated calomel electrode (SCE) in 1 M H2SO4 so­
surface electrical conductivity, the application of various coatings has lution at 348K with 5 ppm HF were measured and compared. It should
been investigated in simulated PEMFC environment. Lin et al. [18] be noted that in this experiment a platinum sheet was used as counter
deposited Ni–P coating through electroless deposition and indicated that electrode (CE), saturated calomel electrode (SCE) was employed as
Ni–P-coated aluminum bipolar plate exhibited higher hydrophobic reference electrode (Ref), and a test sample was considered as working
property and, in comparison with non-coated plates, it has lower electrode (WE). To prevent the passive film from forming on the surface
corrosion current density. In addition, Hung et al. [19,20] compared of the samples, first, cathodic polarization, and then, anodic polarization
carbide-based coated aluminum bipolar plate with that made by (at a potential sweep rate of 20 mV/min) were performed. The EIS tests
graphite composites bipolar plate. They found that the treated were carried out on coatings. The EIS results were modeled using ZView
aluminum bipolar plate provided at least a 22% reduction in hydrogen software; moreover, all electrochemical tests were repeated to confirm
consumption compared to the graphite bipolar plate, with no indication the results. An automatic fuel-cell test system (model 850e, Scribner,
of power degradation (due to metal corrosion) for at least 1000 h. Southern Pines, NC, USA) was used to assess bipolar plates with different
Although metallic bipolar plates have some unique features compared coatings and non-coated Al bipolar plates. The MEA area was deter­
with graphite bipolar plates, numerous studies are being conducted in mined as 5 cm2, the cell temperature as 353K, and the gas flow rate as
order to reduce these weaknesses. In this regard, coatings have been 250 ml/min.
applied using several methods, each one with its own advantages and
disadvantages Rendon-Belmonte et al. [21] have worked on the elec­ 3. Results and discussion
trochemical behavior of a Cr3C2 (NiCr) coating deposited on S4400 with
a high velocity oxygen-fuel (HVOF) thermal spraying process. The Cr3C2 3.1. Coating characterization
(NiCr) coating showed a good performance probably due to its high
content of chromium. According to the results obtained, if coating Fig. 1 shows the XRD patterns of the different coatings. The patterns
applied on the cathode side, fulfills the established criterion for corro­ belong to an as-sprayed top surface and a surface analyzed after grinding
sion resistance, being this as one of the most important criterion to for about 100 μ. As can be seen, the patterns associated with the top
consider in the case of flow plates. Nevertheless, despite the encouraging surface are identical to that belonging to 100 μ below the surface. This
results, when an analysis of the surface of the plate was made after the finding can be attributed to the lack of a major change in the phases due
evaluation, coating depletion was noticed, which in turn reduced the life to HVOF, such as oxidation.
expectancy of the fuel cell environment. Kim et al. [22] worked on Phase identification was conducted in the X-ray diffraction pattern of
metallic glass coated bipolar plates (Ni65Cr15P16B4) applied through the coatings before and after grinding the result of which can be
HVOF and found that it was easy to mount these bipolar plates in a single observed in Fig. 1. As can be seen in the XRD patterns, no difference was
fuel cell. They reported that high corrosion resistance and the single cell observed between the phases and compounds in terms of the surface and
employing the metallic glassy alloy-coated bipolar plates as well as that depth of the coating (after grinding). Furthermore, no noticeable change
with carbon graphite bipolar plates show good I–V performance. High in the peak or difference in their shape and appearance was observed,
velocity oxy-fuel (HVOF) can be an applicable method for which is due to the uniformity of the coating and lack of changes in
industrial-scale coating due to its distinct features such as formation of
dense coating with speed and accuracy in coating and cost effectiveness. Table 1
The present study was conducted with the aim to assess the func­ Spraying parameters.
tionality of coatings in highly corrosive fuel cell environment. In this
Spray parameter Amount
work, coatings with different corrosion resistant alloys were applied to
aluminum substrate in order to produce efficient, cost-effective, and Fuel flow rate (Lit/hr) 22.7
Powder feed rate (g/min) 58
durable metallic bipolar plates. ICR and accelerated corrosion resistance Oxygen flow rate (m3/hr) 52
tests were carried out for various bulk materials and different coatings. Oxygen pressure (Bar) 14
Nitrogen flow rate (m3/hr) 0.6 � 0.05
Nitrogen pressure (Bar) 4
Stand-off distance (mm) 350

2
F. Madadi et al. Materials Chemistry and Physics 238 (2019) 121911

Fig. 1. XRD patterns of the different coatings before and after grinding.

phase, transformative changes, and oxidation of elements. Moreover, a in Fig. 3, which, in turn, could affect the performance of the coating.
phase or compound resulting from the diffusion of Al-substrate with Presence of un-melted particles in the coating can reduce its co­
coating elements was not observed; this was confirmed in EDS results. herency; not only can it affect the mechanical strength of the coating,
Moreover, the JCDPS peaks related to the obtained phases can be but also reduce its physical properties such as electrical conductivity.
observed below the X-ray diffraction pattern of each coating in Fig. 1; a These sites can be considered as weak points, and mechanical pres­
good agreement was observed between the obtained peaks and JCDPS sure can lead to fractures in the coating at these points; this was also
peaks. Furthermore, the JCDPS number of the each phase can be observed in the fractographic study results that be discussed in the
observed in Fig. 1. Based on the determined phases, Ni (00-001-1258) following section.
and Cr (01-088-2323) were identified for the pattern obtained for NiCr, To determine the composition of the coating applied to Al, a local­
Ni3B (01-082-1699), Ni17Si3 (03-065-6491), Ni (00-001-1258) and ized composition analysis was conducted using EDS analysis across the
Cr1.12Ni2.88 (03-065-6491) were identified for NiCrBSi, and AlNi3 (03- coatings conducted. The results presented in Table 3 indicate that the
065-0144), Cr2Ni3 (03-065-6291) and Al0.96Co1.04 (00-044-1114) were application of the coating by HVOF technique was homogeneous.
identified for (Co, Ni) CrAlY, which had high compatibility with the According to the results, the possibility of oxygen element presence
obtained peaks. in coatings is very small; this is confirmed by the XRD analysis presented
The backscattered electron microscopy images of all HVOF spray in Fig. 1, which does not depict any oxide peak in patterns.
coating cross-sections are shown in Fig. 2. The images show the interface Elemental mapping was used to further evaluate the elemental dis­
between the coating and substrate. The substrate was not straight tribution and uniformity of the coating; the results are presented in
because it was sand blasted before being spray coated to increase the Fig. 4. The uniform elemental distribution in the coating can clearly be
adhesion of the film to the substrate. seen in this analysis, which results in the improvement of the perfor­
It should be noted that the observed roughness at the interface could mance of the coating and uniformity of the expected properties of the
be due to the high velocity impact of the splats on the soft CP aluminum. coating.
In addition, spraying under optimum conditions resulted in different Fractographic assessment of coated samples were studied by scan­
porosity levels depending on the coating chemical composition being ning electron cross sectional micrographs (Fig. 5). As can be seen in
sprayed. these images, NiCr had the highest hardness (Vickers hardness:
All cross sectional images of the coated samples showed a dense NiCr ¼ 900; NiCrBSi ¼ 750; MCrAlY ¼ 570) and had brittle fractures.
coating without fracture, open crack, or porosity. In this state, pene­ Moreover, at higher magnifications than Fig. 5-a, the presence of
tration of corrosive media into the coating and substrate will not be cleavage fractures was observed. However, the cross section of fractures
possible. Moreover, due to the presence of a cooling system during the in the NiCrBSi showed ductile fractures (Fig. 5-b). Coating toughness,
coating process, no fractures due to residual stresses and distortion of especially in thermal spray coatings, is of great importance and can
the coating and substrate were observed. demonstrate the necessity of suitable coatings for various applications.
Table 2 exhibits the porosity level of the coatings examined in this Some un-melted particles can be observed in Fig. 5-c, which is due to the
study. As can be seen, under optimized spray conditions, (Co,Ni)CrAlY high melting point of the coating material that have not had sufficient
and NiCrBSi showed the highest and lowest porosity levels, respectively. time to melt in the short duration of the thermal spray coating process
Comparing the three different coatings, one can conclude that (Co, and have not been splashed as a result of impact on the surface. Thus, the
Ni)CrAlY shows the least aspect ratio amount, mainly due to its higher coating was weak at these areas and fractures occur at these points.
melting point. This phenomenon resulted in un-melted particles shown These un-melted particles provide uniformity in some properties such as

3
F. Madadi et al. Materials Chemistry and Physics 238 (2019) 121911

Fig. 2. Cross-sections of a) NiCr, b) NiCrBSi, and c) (Co,Ni)CrAlY deposited on Al plates.

leading to a decrease in ICR. In fact, higher pressure could increase

Table 2
interfacial contact points between the coatings and the GDL carbon
Porosity level of coatings.
papers, leading to the enhancement of both electrical and thermal
Porosity (%) Coated material Item conductivity [3].
4.3 NiCr 1 In addition, at a typical fuel cell stack compression force of
2.6 NiCrBSi 2 140–200 N/cm2, NiCrBSi and NiCr coatings exhibited the lowest ICR
5.9 (Co,Ni)CrAlY 3
value as compared with that of (Co,Ni)CrAlY coating and un-coated Al-
substrate. The surface roughness of the coatings was measured using a
electrical conductivity and mechanical strength. roughness profilometer. Results illustrate that NiCrBSi has the highest
surface roughness (Ra ¼ 7.632 μm), while (Co,Ni)CrAlY has the lowest
roughness (Ra ¼ 5.510 μm). It should be noted that for NiCr, Ra was
3.2. Interfacial contact resistance equal to 5.974 μm. The higher surface roughness of the coatings,
compared to the un-coated Al sample, enhanced ICR by increasing the
ICR between gas diffusion layer (GDL) and the above-mentioned actual contact area between the coating and the GDL. In fact, at higher
coated bipolar plates are presented in Fig. 6. This test was conducted pressures, as shown in Fig. 6, contact between GDL and bipolar plates
under various compression pressures ranging from 20 to 650 N/cm2. ICR increased resulting in increasing electrical current, which in turn
tests were performed on the basis of the two-probe technique. reduced ICR. In addition, un-coated Al-substrate presented the
As can be seen in Fig. 6, by increasing the applied pressure, the maximum ICR mainly due to the formation of Al2O3 layer on Al, which
conductivity level of all coating types at the interface was improved,

4
F. Madadi et al. Materials Chemistry and Physics 238 (2019) 121911

Fig. 3. Micrograph of un-melted particles on the top surface (Secondary Electron and Backscattered Electron images).

Table 3
EDS analysis across the coatings NiCr, NiCrBSi and (Co,Ni)CrAlY
NiCr

Elt Line (c/s) Intensity % Atomic Conc Unit

O Ka 0.00 0.00 0.00 wt.%

Cr Ka 574.44 21.26 19.30 wt.%
Ni Ka 1132.92 78.74 80.70 wt.%
Total 100.00 100.0 wt.%

NiCrBSi

Elt. Line (c/s) Intensity % Atomic Conc Units

O Ka 0.00 0.00 0.00 wt.%

Si Ka 371.70 8.45 4.32 wt.%
Cr Ka 1436.65 17.97 17.02 wt.%
Ni Ka 3145.08 73.58 78.66 wt.%
Total 100.00 100.00 wt.%

(Co,Ni)CrAlY

Elt. Line (c/s) Intensity % Atomic Conc Units

O Ka 0.00 0.00 0.00 wt.%

Al Ka 167.80 12.98 6.43 wt.%
Cr Ka 495.03 17.77 16.98 wt.%
Co Ka 428.12 23.73 25.69 wt.%
Ni Ka 660.51 42.22 45.52 wt.%
Y La 59.66 3.30 5.39 wt.%
Total 100.00 100.00 wt.%

deteriorates the conductivity. It should be noted that increasing porosity

level can result in reduced electrical conductivity; however, due to the
scattered porosity and low amount of porosity in the current work, ICR
was not significantly affected.

3.3. Corrosion resistivity of coating

In order to compare corrosion resistivity of the above-mentioned

coatings, the samples were tested under harsh corrosive solution (0.5
M H2SO4 þ 200 ppm HF) at 25 � C (Fig. 7). As previously mentioned, the
platinum sheet, saturated calomel electrode, and test sample were
employed as counter electrode (CE), reference electrode (Ref), and
working electrode (WE), respectively. In this regard, potentiodynamic
measurements were performed on the materials and the corrosion cur­
rent density of NiCrBSi, NiCr, (Co,Ni)CrAlY, and Al-substrate were found
to be 4.98 (1.03 E 5), 3.89 (0.00012), and 2.91 (0.0012) A/cm2,
respectively. Also Ecorr for NiCr, NiCrBSi and (Co,Ni)CrAlY were 0.36,
0.31 and 0.39 (�0.01) volt. Fig. 7 presents a passive zone between
0 and 0.8V. An essential property of bipolar plate materials is its high
corrosion resistance. As can be seen, the NiCrBSi and NiCr coatings
sprayed by HVOF technique exhibited higher corrosion resistance as
compared with (Co,Ni)CrAlY . NiCrBSi showed lower current density
and higher corrosion resistance than other coatings. Rp (polarization
resistance) for three coatings are 122, 131 and 98 kΩ cm2 fot NiC, Fig. 4. EDS mappings results of coatings cross sections and elemental distri­
bution of coated samples. a) NiCr b) NiCrBSi c) (Co,Ni)CrAlY.
NiCrBSi and (Co,Ni)CrAlY respectively. Lowest and highest Rp belong to
(Co,Ni)CrAlY and NiCrBSi that have largest and lowest corrosion rate

5
F. Madadi et al. Materials Chemistry and Physics 238 (2019) 121911

Fig. 5. Scanning electron cross sectional micrographs of fractured specimens a) NiCr b) NiCrBSi c) (Co,Ni)CrAlY.

Fig. 6. Interfacial contact resistance of coated bipolar plate materials.

6
F. Madadi et al. Materials Chemistry and Physics 238 (2019) 121911

Fig. 7. Polarization curves of the NiCr, NiCrBSi, and (Co,Ni)CrAlY coatings.

among samples. The existing Ni, Cr, B, and Si improved corrosion ZCPE ¼ [C (jω) n] 1, where C is the CPE constant, ω is the angular
resistance of NiCrBSi. Ni shows excellent chemical stability and is one of frequency (rad/s), and j2 ¼ 1. The dispersive exponent n reflects
the proper anti-corrosion material in corrosion media. Alloying elements roughness and nonuniformity with 1 < n < 1; Non-uniform current
Cr, B, and Si could greatly improve corrosion resistances of the coatings distribution caused by surface roughness and porosity controls the n
[23–25]. value so that n varying between 0.5 and 1.
Increasing porosity level can be effective on corrosion resistivity. In Table 4 consists of the results of the EIS tests. Rct is closely related to
addition, internal porosities (as can be seen in the coating’s cross sec­ anodic dissolution, reflecting the change in the dissolution rate of metal.
tions in Fig. 2) in coatings have no effect on this parameter, as the The larger the Rct is, the higher the corrosion resistance will be. In
corrosive media cannot penetrate into this type of porosity. addition, it should be mentioned that the reduced Rct was caused by the
Figs. 8 and 9 show the EIS results of coated samples. The Nyquist accelerating rate of the charge-transfer reactions in the protective film.
plots (Fig. 8) have a semicircle indicates that materials possessed Also the elevated CPEf values may be related to the displacement of
capacitance characteristics; this arch did not follow by an inductive corrosion products (oxides and/or hydroxides), whereas increased
loop. In addition, these features illustrate the lack of local corrosion in CPEdl can be attributed to an enhancement in the dielectric constant or a
the protective layer and initiation of metal dissolution. According to reduction in the electric double layer [25,34–36]. The decrease in Rf
Mareci et al. [26], the inductive arch belongs to the metallic dissolution. indicates that compactness and uniformity of the protective film on the
Moreover, Zhang et al. [27] believed that the adsorption could cause the electrode surface became increasingly worse.
inductive loop. The semicircle at high frequencies is attributed to the The water generated during the cell operation was analyzed using
faradaic charge-transfer process, and the size of radius reflects the size of inductively coupled plasma-mass emission spectroscopy (ICP-OES)
the charge transfer resistance (Rct) [28]. technique by which the presence of small amounts of aluminum, chro­
The peak at high frequencies is associated with the behavior of the mium, nickel, platinum, and iron could be detected. The measured
protective film formed on the surface of NiCrBSi coating [25,29]. This concentrations were lower than the range of 5–10 ppm of metallic ions.
evolution suggests that a homogeneous and compact film was present on This range is usually found when there is a considerable contamination
the surface of NiCrBSi coating, strongly hindering corrosion processes. in the membrane. The absence of coating materials in large quantities
These results are consistent with the potentiodynamic polarization data. was noted as significant evidence that the coated aluminum plate had
The EIS results consisting of bode-phase and bode-magnitude anal­ the potential to operate for a longer time.
ysis for coated samples are shown in Fig. 9. The results for coated
samples represent similar behaviors in the EIS spectrums. 3.4. Fuel cell performance
The EIS results illustrate that the NiCrBSi coating has the highest
resistance among all samples, which results in better protection against The current-voltage (I–V) curves of the single fuel cell with bipolar
corrosive ions. plates coated with different material at 353K cell temperature and
Fig. 10 shows the best electrical circuit to model the impedance data 0.25 L min 1 gas flow rate are illustrated in Fig. 11. At the cell voltage of
in order for interpreting results; in fact, other researchers have used 0.5 V, the single cells with NiCrBSi, NiCr, (Co,Ni)CrAlY, and aluminum
similar circuits to model the EIS results [30–33]. In this model, Rs shows alloy bipolar plates, respectively, produced a current density of 885,
the solution’s resistance, Rct represents charge-transfer resistance of the 818, 665, and 360 mA cm 2.
electric double layer, and Rf is the film resistance. CPEdl and CPEf were As seen in this figure, compared to other bipolar plates, higher I–V
used to replace the electric double-layer and film capacitances, respec­ performance was observed in the single cell with the NiCrBSi coated
tively. The constant phase element or CPE shows non-ideal capacitive bipolar plates. Because of its lack of coating, the Al bipolar plate had
behavior. ZCPE is obtained through the following equation: higher corrosion current density; thus, it was corroded easily. This

7
F. Madadi et al. Materials Chemistry and Physics 238 (2019) 121911

Fig. 9. EIS results (bode-phase diagrams and bode-magnitude diagrams) for the
Fig. 8. Nyquist diagrams for the coated samples a) NiCr b) NiCrBSi C) MCrAlY.
coated samples a) NiCr b) NiCrBSi C) MCrAlY.

phenomenon resulted in an increase in contact with electrical resistance

improper I–V performance. In addition, similar I–V performance was
observed in the single cells with NiCrBSi coating and carbon graphite
bipolar plates. In fact, this coating can be utilized as a bipolar plate due
to its high corrosion resistance.
Long-term durability tests were performed on the single cell with
carbon graphite, Al, and NiCrBSi, NiCr, and (Co,Ni)CrAlY alloy bipolar
plates (constant current density: 200 mA cm 2; cell temperature: 353K;
RH: 30%; gas flow rate of H2 and O2: 250 ml min 1 at 25 psi) (Fig. 12). Fig. 10. Proposed equivalent circuit used for fitting the experimental data of
The cell voltage equipped with the uncoated-Al bipolar plates exhibited coated samples.
significant lower cell voltage due to well-known passive layer formation.
However, when using the coated-Al (NiCrBSi and NiCr alloy) bipolar not observed in cell voltage in the NiCrBSi and NiCr alloy bipolar plates.
plates, the cell voltage was comparable to the standard bipolar plate It should be mentioned that bipolar plates employed (Co,Ni)CrAlY
(graphite). Such behavior is attributed to both higher corrosion resis­ coated-Al showed medium cell voltage performance as compared to
tance and also lower interfacial contact resistance of the above coating other mentioned bipolar material, mainly due to its weaker corrosion
materials. It should be noted that in the NiCrBSi and NiCr alloy bipolar resistance. In addition, cell voltage was decreased gradually as the test
plates, initially the cell voltage decreased a small amount due to the was proceed for all bipolar with different materials, mainly due to the
formation of the passive film. Nevertheless, a significant decrease was increase in contact electrical resistance resulted from corrosion

8
F. Madadi et al. Materials Chemistry and Physics 238 (2019) 121911

Table 4 NiCrBSi, NiCr, and (Co,Ni)CrAlY coated bipolar plates. Potentiodynamic

EIS fitting results for coated samples. polarization scan was performed to determine the corrosion behavior of
Material/ Rs, Ω. CPEdl, nF. Rct, Ω. CPEf, μF. Rf, kΩ. the coatings and the ICR of material with a GDL has been studied. Then,
Model cm2 cm 2 cm2 cm 2 cm2 single cell with NiCrBSi, NiCr, and (Co,Ni)CrAlY coated bipolar plate
NiCr 21.02 265.5 172.1 1.076 80.6 underwent long-term durability tests. The conclusions obtained in this
NiCrBSi 29.06 268.6 213.6 0.6721 116.8 work are as follows:
(Co,Ni)CrAlY 16.11 680.7 114.2 1.6804 32.18 The experimental measurement has proven that NiCrBSi coating can
be an excellent candidate coating for PEMFC metallic bipolar plates
since its ICR value was the lowest among the studied materials and it is
products. Such gradual drop was less enhanced in coated-Al (NiCrBSi
comparable with graphite bipolar plate. In addition, corrosion resistance
and NiCr alloy) bipolar plates.
test indicated that (Co,Ni)CrAlY and NiCrBSi have the highest and
lowest corrosion current density. EIS results show that the total
4. Conclusion
impedance values of NiCrBSi coating is higher than two other coatings.
The high I–V performance of the NiCrBSi and graphite coated bipolar
In the present study, HVOF coating method was used to produce

Fig. 11. I–V curves of the single cell with graphite, Al, and NiCrBSi, NiCr, and (Co,Ni)CrAlY coated bipolar plates.

Fig. 12. Long-term durability tests of single cell with graphite, Aluminum, and NiCrBSi, NiCr, and (Co,Ni)CrAlY coated bipolar plates.

9
F. Madadi et al. Materials Chemistry and Physics 238 (2019) 121911

plates illustrated their suitability for PEMFC. Long-term durability test as a metal bipolar plate for a direct methanol fuel cell, Int. J. Hydrogen Energy 35
(2010) 4320–4328.
results did not show a significant decrease in cell voltage in the NiCrBSi
[17] M.P. Brady, M.A. Elhamid, G. Dadheech, J. Bradley, T.J. Toops, H.M. Meyer, P.
and NiCr alloy bipolar plates. However, a significant decrease was F. Tortorelli, Manufacturing and performance assessment of stamped, laser welded,
observed in cell voltage in the single cell with uncoated-Al bipolar and nitrided FeCrV stainless steel bipolar plates for proton exchange membrane
plates; this was attributed to the increase in contact electrical resistance fuel cells, Int. J. Hydrogen Energy 38 (2013) 4734–4739.
[18] C.-H. Lin, S.-Y. Tsai, An investigation of coated aluminium bipolar plates for
due to well-known passive layer formation. Thus, it can be concluded PEMFC, Appl. Energy 100 (2012) 87–92.
from the tests results that NiCrBSi coating can be a candidate for PEM [19] Y. Hung, K.M. El-Khatib, H. Tawfik, Testing and evaluation of aluminum coated
bipolar plates. bipolar plates of PEM fuel cells operating at 70 C, J. Power Sources 163 (2006)
509–513.
[20] Y. Hung, H. Tawfik, D. Mahajan, Durability and characterization studies of
References polymer electrolyte membrane fuel cell’s coated aluminum bipolar plates and
membrane electrode assembly, J. Power Sources 186 (2009) 123–127.
[1] B.-T. Tsai, C.-J. Tseng, Z.-S. Liu, C.-H. Wang, C.-I. Lee, C.-C. Yang, S.-K. Lo, Effects [21] M. Rend� on-Belmonte, J.T. P�erez-Quiroz, J. Ter�
an-Guill�en, J. Porcayo-Calder�
on,
of flow field design on the performance of a PEM fuel cell with metal foam as the A. Torres-Acosta, G. Orozco-Gamboa, Evaluation of a Cr3C2 (NiCr) coating
flow distributor, Int. J. Hydrogen Energy 37 (2012) 13060–13066, https://doi.org/ deposited on S4400 by means of an HVOF process and used for flow plates of PEM
10.1016/j.ijhydene.2012.05.008. fuel, Int. J. Electrochem. Sci. 7 (2012) 1079–1092.
[2] H. Heidary, M.J. Kermani, B. Dabir, Influences of bipolar plate channel blockages [22] S. Kim, S. Yamaura, Y. Shimizu, K. Nakashima, T. Igarashi, A. Makino, A. Inoue,
on PEM fuel cell performances, Energy Convers. Manag. 124 (2016) 51–60, Production of Ni65Cr15P16B4 metallic glass-coated bipolar plate for fuel cell by
https://doi.org/10.1016/j.enconman.2016.06.043. high velocity oxy-fuel (HVOF) spray coating method, Mater. Trans. 51 (2010)
[3] M. Zhiani, S. Kamali, S. Majidi, In-plane gas permeability and thought-plane 1609–1613.
resistivity of the gas diffusion layer influenced by homogenization technique and [23] M.T.M. Suarez, S. Bellayer, Corrosion behavior of Cr3C2-NiCr vacuum plasma
its effect on the proton exchange membrane fuel cell cathode performance, Int. J. sprayed coatings, Surf. Coat. Technol. 202 (2008) 4566–4571.
Hydrogen Energy 41 (2016) 1112–1119. https://doi.org/10.1016/j.ijhydene.201 [24] A.P.G.T. Liyanage, G. Fisher, Influence of alloy chemistry on microstructure and
5.10.052. properties in NiCrBSi overlay coatings deposited by plasma transferred arc welding
[4] M. Zhiani, S. Majidi, M.M. Taghiabadi, Comparative Study of On-Line Membrane (PTAW), Surf. Coat. Technol. 205 (2010) 759–765.
Electrode Assembly Activation Procedures in Proton Exchange Membrane Fuel [25] S.L.J.Q. Zeng, J. Sun, W. Emori, Corrosion behavior of thermally sprayed NiCrBSi
Cell, 2013, pp. 946–955, https://doi.org/10.1002/fuce.201200139. coating on 16MnR low-alloy steel in KOH solution, J. Mater. Eng. Perform. 25
[5] S.F. Husby H, O.E. Kongstein, A. Oedegaard, Carbonpolymer composite coatings (2016) 1773.
for PEM fuel cell bipolar plates, Int. J. Hydrogen Energy 2 (2014) 951–957. [26] I.A.D. Mareci, G. Bolat, J. Izquierdo, C. Crimu, C. Munteanu, R.M. Souto,
[6] Y.C. Park, S.H. Lee, S.K. Kim, S. Lim, D.H. Jung, S.Y. Choi, Effects of CrN/Cr Electrochemical characteristics of bioresorbable binary MgCa alloys in Ringer_s
coating layer on durability of metal bipolar plates under a fuel recirculation system solution: revealing the impact of local pH distributions during in-vitro dissolution,
of direct methanol fuel cells, Int. J. Hydrogen Energy 38 (2013) 10567–10576. Mater. Sci. Eng. C 60 (2016) 402–410.
[7] M.D. Tawfik H, Y. Hung, Metal bipolar plates for PEM fuel cell a review, J. Power [27] W.L.Y. Zhang, C. Yan, F. Wang, Electrochemical behavior of anodized Mg alloy
Sources 2 (2007) 755–767. AZ91D in chloride containing aqueous solution, Corros. Sci. 47 (2005) 2816–2831.
[8] K.C. Kwon HJ, Y.P. Jeon, Die design of aluminum bipolar plate fabrication by [28] M. Mouanga, P. Bercot, Comparison of corrosion behaviour of zinc in NaCl and in
stamping process and its investigation, Adv. Mater. Res. 445 (2012) 108–113. NaOH solutions, electrochemical analyses, Corros. Sci. 52 (2010) 3993–4000.
[9] S.H. Lee, V.E. Pukha, V.E. Vinogradov, N. Kakati, S.H. Jee, S.B. Cho, [29] E.H.H.X.H. Li, J.Q. Wang, Corrosion behavior for alloy 690 and alloy 800 tubes in
Nanocomposite-carbon coated at low-temperature: a new coating material for simulated primary water, Corros. Sci. 67 (2013) 169–178.
metallic bipolar plates of polymerelectrolyte membrane fuel cells, Int. J. Hydrogen [30] Y.G.Y. Gu, C.-F. Chen, S. Bandopadhyay, C. Ning, Y. Zhang, Corrosion mechanism
Energy 38 (2013) 14284–14294. and model of pulsed DC microarc oxidation treated AZ31 alloy in simulated body
[10] K.C. Jin CK, Fabrication by vacuum die casting and simulation of aluminum bipolar fluid, Appl. Surf. Sci. 258 (2012) 6116–6126.
plates with micro-channels on both sides for proton exchange membrane (PEM) [31] K.V.D. Sreekanth, N. Rameshbabu, Effect of various additives on morphology and
fuel cells, Int. J. Hydrogen Energy 37 (2012) 1661–1676. corrosion behavior of ceramic coatings developed on AZ31 magnesium alloy by
[11] C.Y. Lee SJ, C.H. Huang, Investigation of PVD coating on corrosion resistance of plasma electrolytic oxidation, Ceram. Int. 38 (2012) 4607–4615.
metallic bipolar plates in PEM fuel cell, J. Mater. Process. Technol. 140 (2003) [32] L.L. Guo-Hua, C. Huan, W. Xing-Quan, P. Hua, Z. Gu-Ling, Z. Bin, Y.S.-Z. Heon-Ju,
688–693. Characteristics of sealed plasma electrolytic oxidation coatings with
[12] J.R. Mawdsley, J.D. Carter, X. Wang, S. Niyogi, C.Q. Fan, R. Koc, Composite-coated electrochemical impedance spectroscopy, Chin. Phys. B 19 (2010) 085202.
aluminum bipolar plates for PEM fuel cells, J. Power Sources 231 (2013) 106–112. [33] A.H.E. Ziyaei, M. Atapour, H. Edris, Corrosion behavior of PEO coatings formed on
[13] K.-H. Hou, C.-H. Lin, M.-D. Ger, S.-W. Shiah, H.-M. Chou, Analysis of the corrosion AZ31 alloy in phosphate- based electrolytes with calcium acetate additive,
behavior of Al alloy bipolar plate for proton exchange membrane fuel cell (Pemfc) J. Mater. Eng. Perform. (2017), https://doi.org/10.1007/s11665-017-2765-9.
under operating thermal conditions, Int. J. Green Energy 9 (2012) 71–83. [34] C.A.M. Sanchez, J. Gregori, Electrochemical impedance spectroscopy for studying
[14] M.P. Brady, H. Wang, B. Yang, J.A. Turner, M. Bordignon, R. Molins, M.A. Elhamid, passive layers on steel rebars immersed in alkaline solutions simulating concrete
L. Lipp, L.R. Walker, Growth of Cr-Nitrides on commercial Ni–Cr and Fe–Cr base pores, Electrochim. Acta 52 (2007) 7634–7641.
alloys to protect PEMFC bipolar plates, Int. J. Hydrogen Energy 32 (2007) [35] F.H.W.X.M. Wang, H.Y. Yang, A cationic gemini-surfactant as effective inhibitor for
3778–3788. mild steel in HCl solutions, Corros. Sci. 52 (2010) 1268–1276.
[15] A. Pozio, F. Zaza, A. Masci, R.F. Silva, Bipolar plate materials for PEMFCs: a [36] M. Li, S. Luo, C. Zeng, J. Shen, H. Lin, C. Cao, Corrosion behavior of TiN coated
conductivity and stability study, J. Power Sources 179 (2008) 631–639. type 316 stainless steel in simulated PEMFC environments, Corros. Sci. 46 (2004)
[16] Y.-C. Park, S.-H. Lee, S.-K. Kim, S. Lim, D.-H. Jung, D.-Y. Lee, S.-Y. Choi, D.- 1369–1380.
H. Peck, others, Performance and long-term stability of Ti metal and stainless steels

10