Results in Physics 6 (2016) 730–736

Contents lists available at ScienceDirect

Results in Physics

journal homepage: www.journals .e lsevier .com/resul ts - in-physics

Corrosion kinetics of 316L stainless steel bipolar plate withchromiumcarbide coating in simulated PEMFC cathodic environment

http://dx.doi.org/10.1016/j.rinp.2016.10.0022211-3797/� 2016 The Authors. Published by Elsevier B.V.This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

⇑ Corresponding author.E-mail address: nbhuang@dlmu.edu.cn (N.B. Huang).

N.B. Huang a,⇑, H. Yu a, L.S. Xu b, S. Zhan a, M. Sun c, Donald W. Kirk d

aMaterial Property Test Center of Key Laboratory of Ship-Machinery Maintenance & Manufacture, Dalian Maritime University, Dalian 116026, PR Chinab Sunrise Power Co. LTD, Dalian 116085, PR ChinacDepartment of Physics, Dalian Maritime University, Dalian 116026, PR ChinadDepartment of Chemical Engineering and Applied Chemistry, University of Toronto, Toronto, Ontario M5S3E5, Canada

a r t i c l e i n f o a b s t r a c t

Article history:Received 22 August 2016Received in revised form 30 September2016Accepted 4 October 2016Available online 17 October 2016

Keywords:PEMFCMetal bipolar plateChromium carbide coatingCorrosion kineticsPitting corrosion

Stainless steel with chromium carbide coating is an ideal candidate for bipolar plates. However, the coat-ing still cannot resist the corrosion of a proton exchange membrane fuel cell (PEMFC) environment. Inthis work, the corrosion kinetics of 316L stainless steel with chromium carbide is investigated in simu-lated PEMFC cathodic environment by combining electrochemical tests with morphology andmicrostructure analysis. SEM results reveal that the steel’s surface is completely coated by Cr and chro-mium carbide but there are pinholes in the coating. After the coated 316L stainless steel is polarized, thediffraction peak of Fe oxide is found. EIS results indicate that the capacitive resistance and the reactionresistance first slowly decrease (2–32 h) and then increase. The potentiostatic transient curve declinessharply within 2000 s and then decreases slightly. The pinholes, which exist in the coating, result inpitting corrosion. The corrosion kinetics of the coated 316L stainless steel are modeled and accords thefollowing equation: i0 = 7.6341t�0.5, with the corrosion rate controlled by ion migration in the pinholes.

� 2016 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-NDlicense (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Introduction

Owing to its virtue of high efficiency, high power density, lowtemperature operation, fast and easy start-up, etc., protonexchange membrane fuel cells (PEMFC) have promising applica-tions in Electric Vehicles and backup power stations. AlthoughPEMFC is gaining worldwide interest, and attracting governmentsand the research institutes to spend huge human and materialresources on its research and development, and has achieved greatadvances in recent years, the following problems are still to besolved in its commercialization.

Lifetime [1,2] for different applications, the requirements forfuel cell lifetime vary significantly, ranging from 5000 h for carsto 40,000 h of continuous operation for stationary applications.The durability of the fuel cell has become one of the technicalchallenges restricting its commercialization.Cost High Cost is a major hurdle for commercialization. Onlywhen fuel cell costs are dramatically reduced to the US Depart-ment of Energy (DOE) target, will fuel cells be competitive.

Hydrogen Source the lack of infrastructure is also thought asone of the limiting factors of its large-scale application.

Basically, the commercialization process of PEMFC is deter-mined by the development of technology. As one of the key com-ponents of PEMFC, bipolar plates occupy not only 70�80% inweight but also about 30% in economic cost. Graphite bipolarplates have been a good start for the commercialization of PEMFC,but their poor mechanical strength and high processing costs makethem lack competitiveness in industrial applications. Althoughmetal bipolar plates possess so many advantages such as low-cost, excellent mechanical properties, good electrical and thermalconductivities, etc., they are subject to corrosion. The corrosiongenerated metallic ions may subsequentially poison the FC cata-lysts decreasing performance. Additionally, with corrosion, interfa-cial contact resistance increases because of the formation ofpassive surface layer films. Thus research on minimizing the corro-sion of metallic bipolar plates is needed. Although the coatings ofnoble metals such as gold or platinum would have excellent chem-ical stability and conductivity, their high price rules them out forachieving marketability. TiN coatings have received extensiveattention, but the pitting corrosion was very high and their fuel cellperformance has not been satisfactory in the presence of F� [3,4].Other coatings such Ni-Co-B can resist corrosion in PEMFC

N.B. Huang et al. / Results in Physics 6 (2016) 730–736 731

environment, but their performance degrades rapidly [5]. Thoughamorphous alloys such as Fe50Cr18Mo8Al2Y2C14B6 and Fe44Cr15-Mo14Y2C15B6N4 have better anti-corrosion performances comparedwith stainless steel, there is a challenge in making them in largeformat [6,7]. There is also the concern that recrystallization mayoccur during use or during cell construction and thus lead to corro-sion failure. Thus, attention is focused on coatings for protectingstainless steels. Nitride coatings have been shown to improve thecontact resistance of Ni-50Cr alloy [8] and the coating processhas been applied to lower Cr content stainless steels. Good resultswere found for the Ni-C rand AL29-4C� steel [9]. Through physicalvapor deposition, ion nitriding, forms an intermediate transitionlayer with Cr and its compounds [9–11], also had good corrosionresistance in PEMFC environment. Fu Yu et al. [12–17] fabricatedthe chromium nitride and carbon coating for 316L stainless steel,which basically meet the requirements of low contact resistance,high corrosion resistance, and relatively low cost. These beneficialresults created a new direction for the surface modification ofPEMFCmetal bipolar plates. For example, Hyundai Motor Companyprepared carbon film containing F on the surface of a metal bipolarplate by chemical vapor deposition. GM company (USA) exploited agraphite coating containing Fe and Cr on the metal bipolar platesurface. Feng, Larijani and Yi et al. deposited amorphous carbonfilms on the surface of 316L stainless steel and 304 stainless steelby magnetron sputtering method [18–20], respectively.

Carbide or nitride layers on stainless steel surface exhibit goodconductivity and good resistance to corrosion, and appear likely tobe used in commercial applications for PEMFC bipolar plates.Despite these developments few researchers focus on the corrosionkinetics of stainless steel bipolar plates with a chrome carbidecoating in PEMFC cathodic environment. Research on the corrosionprocess of stainless steel with coatings not only can evaluate itsanti-corrosion characteristic but also is in favor of knowing itseffect on membrane electrode assembly and the performancedegradation of PEMFC stack. In this paper, the corrosion perfor-mance of 316L stainless steel with a chrome carbide coating istested by electrochemical methods. Combined with the results ofmorphology and structure, its corrosion kinetics was determinedand modeled. Since 316L stainless steel with a chrome carbidecoating will be an alternative materials for PEMFC stacks, theseresults are essential for knowing the mechanism of corrosion ofstainless steel with chrome carbide coatings, and providing guid-ance for further improvement. Meantime, the results are also ben-eficial for monitoring the corrosion and evaluating the lifetime of316L stainless steel bipolar plates with chromiumcarbide coatingin PEMFC stacks.

Experimental details

Materials

The test materials were 316L stainless steel with chrome car-bide coating (20 mm � 20 mm � 1 mm) offered by Sunrise PowerCo. Ltd. The samples were ultrasonically cleaned for 15 min in ace-tone, and then dried. The corrosion solution was 0.5 mol�L�1 H2-SO4 + 5 mg�L�1 HF solution prepared with 18 MX ultrapure water(Beijing Xiangshun Yuan Co. Ltd.). In order to simulate PEMFCcathodic environment, the solution was continuously bubbled withair and was maintained at 80 �C. All the reagents used were analyt-ical reagents.

Electrochemical tests

The electrochemical tests were conducted on VMP3 electro-chemical workstation (EG&G company) using a three electrode

system. The cleaned sample with 1cm2 exposed area, used as theworking electrode. The reference electrode and the counter elec-trode were a saturated calomel electrode (SCE) and Pt net(20 mm � 20 mm), respectively. The cyclic potentiodynamic polar-ization was recorded from open circuit potential to more positivevoltages with a scan rate of 1 mV�s�1. When the current densityreached 1 mA�cm�2, the reverse sweep was initiated. The runwas stopped when the reverse curve intersected the forward curve.Electrochemical impedance spectroscopy (EIS) measurements wastested at open circuit potential when the sample was immersed inthe simulated solution for more than 30 min. During test, the fre-quency was swept from 100 kHz to 0.01 Hz while the ac amplitudewas 5 mV. EIS data were analyzed by ZsimpWin software. Unlessnoted otherwise, all the tests were performed at room temperatureand all the potentials refereed to SCE.

In order to evaluate the coating’s stability, potentiostatic tran-sient polarization was performed in simulated PEMFC cathodicenvironment. The applied potential was 0.6V. Based on the poten-tiostatic transient curve, the corrosion kinetics equation of thestainless steel having coating was obtained.

Surface morphology analysis

The topography and the cross-section of the coating wasinvestigated by scanning electron microscopy (Philips,XL-30TMP), X-ray diffraction meter (D/MAX-Ultima+, RigakuCorporation), X-ray photoelectron spectroscopy (XPS, SHIMADZU-AMICUS) system equipped with a water-cooled Mg-anode at atotal power dissipation of 120 W (12 kV, 10 mA).

Results and discussion

Cyclic potentiodynamic polarization

Fig. 1 showed the SEM cross-section and topography of thesample. From Fig. 1(a), the thickness of the coating was about240 nm. Before polarization, it was relatively rough and had pin-holes. It became smoother and the pinholes grew slightly afterpolarization. The spherical nodules and pinholes in the surfacewere seen to be active and displayed preferential loss during polar-ization. Thus, the polarized specimens tended to become smoothand the pinholes increased slightly.

Fig. 2 showed the cyclic potentiodynamic polarization curve of acoated sample in simulated PEMFC cathodic environment. FromFig. 2, when the polarization potential was below 0.6 V, the currentdensity maintained at 0.04 lA�cm�2. While the polarization poten-tial was above 0.6 V, the current density increased significantlywith increase in polarization potential. While in the reverse sweep,a small hysteresis loop was found. The formation of a hysteresisloop indicated that pitting had taken place, and the loop size canbe interpreted with the extent of pitting [21,22]. Conversely, thesmaller the hysteresis loop was, the less susceptible was the sur-face to pitting [23]. Since the coating prepared by pulsed bias arcion plating had pinholes, it is clear that the corrosive ion can pen-etrate through the coating and result in corrosion. Although thecoated sample had a very small hysteresis loop and showed bettercorrosion resistance, pitting corrosion would inevitably beobserved and be accompanied by general corrosion. The smoothersurface and the more enlarged the pinholes, the greater the corro-sion observed, thus supporting the above deduction.

Figs. 3 and 4 showed the surface phases of the coated samplebefore polarization and the surface analysis using X-ray photoelec-tron spectroscopy (XPS) after polarization. From Fig. 3, the coatingwas composed of Cr and Cr3C2, and there was no diffraction peak ofFe. The Cr signal in the analysis was due to the presence of a Cr

Fig. 1. SEM of 316L stainless steel with chromium carbide coating (a) cross section, topography before (b) and after polarization (c).

Fig. 2. Cyclic potentiodynamic polarization curve of 316L stainless steel withchromium carbide coating in si mulated PEMFC cathodic environment.

Fig. 3. XRD result of the coated sample before polarization.

732 N.B. Huang et al. / Results in Physics 6 (2016) 730–736

layer which was used to increase the adhesion between the Cr3C2

coating and the matrix. Combining the results of SEM and XRD, itis clear that the stainless steel matrix was completely covered bythe Cr and Cr3C2 except there were pinholes on the surface.

From Fig. 4a, XPS analysis shows Cr, Fe, C and O atoms werefound on the surface of the polarized sample. The peaks found at573.8 eV, 576.4 eV, 583.3 eV and 586.2 eV were identified asCr2p1/2 and Cr2p3/2, respectively. The peak of C1s was at 282.5 eV,

284.6 eV and 288.5 eV. The peaks at 710.4 eV, 717.0 eV and723.3 eV were identified as Fe2p1/2 and Fe2p3/2, respectively. Thepeak at 531.9 eV was from O1s. More detailed analysis based onthe standard energy spectrum data and literature [24,25] revealedthat the Crp3/2 peak in Cr2O3 was found at 576.4 eV, the Cr-O bondin O1s peak can be assigned to 531.5 eV. The peak position of Fe2p3/2(between 710 and 711.2 eV) had been investigated by manyresearchers [26,27]. The Fe2p3/2 peak for Fe2O3 had associated satel-lite peaks, which was located approximately 8 eV higher than themain Fe2p3/2 peak [28]. Fe2p3/2 for Fe3O4 hadn’t a satellite peak[28]. The binding energies of Fe2p3/2 and Fe2p1/2 obtained fromthe present study were 710.4 eV, 717.0 eV and 723.3 eV, respec-tively. The satellite peak obtained at 717.0 eV was clearly distin-guishable and did not overlap either the Fe2p3/2 or Fe2p1/2 peaks,which mean Fe2O3 existed in the oxides [28]. The other peaks man-ifested that there also had Fe3O4. The results indicated that therewere oxides of Cr and Fe on the surface of polarized sample. NoFe species was seen in XRD analysis (Fig. 3) prior to polarization,so that the XPS detection of iron oxide after polarization indicatedthat corrosion of the matrix must had taken place. The corrosionmechanism must have been from a corrosive ion penetrating thecoating, and migrating to the matrix where dissolution of the irontook place. The pathway for this corrosion was almost certainly thepinholes in the film. Along with the developing of corrosion pro-cess, the corroded species would have had difficulty in diffusinginto the bulk solution and hence would accumulate in the pinholes,and this process would hinder the corrosion rate. In addition, thestainless steel can passivate if there was sufficient oxygen avail-able. Enlarging pinholes would be expected to increase this oxygenavailability and slowed the corrosion process down.

EIS results

Fig. 5 showed the Nyquist diagram of the coated sampleimmersed in simulated PEMFC cathodic environment for differenttimes. The EIS spectra showed the characteristic of a single capac-itive resistance within the tested time. The capacitive resistancewas found to first decrease and then increase with the longer dura-tion in the electrolyte. The equivalent circuit (Fig. 6) was used to fitthese EIS data and the fitting results were shown in Table 1 where,Rs – solution resistance, CPE – Constant phase element, Rt – reac-tion resistance, W – Warburg impedance.

From Table 1, it is clear that CPE and Rt decreased graduallywith time elapsed between 2 to 32 h, which meant the corrosionrate increased gradually. At the initial stage, as seen in the SEMFig. 1a, there were many surface active sites which were preferen-tially dissolved in the electrolyte (Fig. 1b). In addition, the surfacefilm and the pinholes would also be susceptible to corrosion.Therefore it is not unexpected that the capacitive resistance and

Fig. 4. XPS survey of the corrosive bipolar plates (a) and XPS peaks of Cr2p (b), C1s (c), Fe2p (d), O1s (e).

N.B. Huang et al. / Results in Physics 6 (2016) 730–736 733

the reaction resistance slowly decreased. When the immersiontime exceeded 32 h, C and Rt began to increase and hence the bipo-lar plate’s corrosion rate decreased. At longer immersion times, pit-ting corrosion would become the dominant corrosion process andwould accelerate slowly. Evidence for this can be seen in theresults of SEM analysis for 20 and 80 h exposures. Fig. 7 showedthe surface morphology of the coated sample after immersion insimulated PEMFC environment for different times. The dissolutionof uneven active points on the surface was accompanied by pin-holes expanding (Fig. 7a) during immersion. When immersion timewas over 80 h, some of the coating had delaminated and becamedetached (Fig. 7b). At this stage surface roughness increased.Although the surface of 316L stainless steel was coved by the inter-mediate transition layer (having Cr and its compounds) and Cr3C2,

the corrosion of matrix under the coating still occured. The oxide ofFe found in XPS further confirmed that the stainless steel matrixwas corroded. With longer immersion times, some corrosion prod-ucts would deposited back to the surface and the pinholes, result-ing in the capacitive resistance decreased and the reactionresistance increased.

The potentiostatic transient polarization curve

The potentiostatic transient polarization curve of the coatedsample in simulated PEMFC cathodic environment was shown inFig. 8. From Fig. 8, the transient polarization curve declined sharplywithin 2000 s, and then decreased slightly. The average value wasabout 0.07 lA�cm�2. According to the results of SEM, corrosion

Fig. 5. The Nyquist plot of the coated sample immersed in simulated PEMFCcathodic environment.

Rsol

Rt

Q

Fig. 6. The equivalent circuit used for fitting EIS data.

Fig. 8. Potentiostatic transient curve in simulated cathodic environment.

734 N.B. Huang et al. / Results in Physics 6 (2016) 730–736

occured mainly from the pinholes on the surface once the surfaceroughness was lost. With longer times, some of the pinholes wereenlarged and some corrosion extended from these pinholes underthe coating to cause a trench or groove. Since the destruction ofbipolar plate was mainly contributed by the corrosion through

Table 1The fitting results of EIS data.

Exposuretime/h

RsX.cm2

CCPE

Yo/S.secn.cm�2

2 4.52 2.48E�054 4.50 2.32E�056 4.52 2.26E�058 4.95 2.24E�0520 4.19 2.22E�0532 4.04 2.24E�0544 4.01 2.27E�0556 4.04 2.25E�0568 3.99 2.26E�0580 3.93 2.31E�05

Fig. 7. SEM of the coated sampled after immersed 20 h a

pinholes, it can be treated as pitting corrosion. Due to the complex-ity of the process, some simplification was necessary before thekinetic equation for the pitting corrosion of bipolar plate couldbe established [29–33]. Firstly, the pinholes were treated as oneintegral whole. Secondly, the corrosion current was mainly fromthe corrosion occurring within the pinholes and the contributionfrom general corrosion was neglected; Finally, the pitting corrosionwas controlled by ion migration. Based on these hypothesises, therelationship between pitting depth (h) and current density (i0) wasobtained according to Faraday’s law [32,33]:

n ¼ Qz � F ¼ i � t

z � F ð1Þ

i0 ¼ iS

ð2Þ

RtX.cm2

WS.sec0.5.cm�2

n

0.95 7.59E+05 4.05E�060.96 9.26E+05 2.65E�060.96 1.37E+06 4.19E�060.96 1.75E+06 6.98E�060.96 1.22E+06 8.48E�060.96 7.42E+05 4.72E�060.96 7.88E+05 3.35E�060.96 1.06E+06 3.20E�060.96 1.27E+06 3.70E�060.96 1.37E+06 3.87E�06

nd 80 h in simulated PEMFC cathodic environment.

N.B. Huang et al. / Results in Physics 6 (2016) 730–736 735

h ¼ VS¼ mq � S ¼ n �M

q � S ¼ Q �Mz � F � q � S ¼ i � t �M

z � F � q � S ¼ i0 � t �Mz � F � q ð3Þ

where n – the mole number of metal reacted;Q – the consumed electric quantity during the reaction, C;z – the transferring electrical charge number;F – faraday constant;I – the current during the corrosion, A;i0 – the current density during the corrosion, A�cm�2

S – the pinhole’s area, cm2;m– the dissolved metal, g;q – the metal density, g�cm�3;M – the mole mass, g�mol�1.

The derivative of pitting depth with respect to time iscalculated:

dhdt

¼ i0 �Mz � F � q ð4Þ

i0 ¼ z � F � qM

� dhdt

ð5Þ

The process of the corroded metal ions diffusing from pinhole tobulk solution was in accord with Fick’s law [29] (Eq. (6)).

n ¼ D � S � ðcs � cbÞh

� t ð6Þ

where D – the diffusion coefficient, m2�s�1;

cs – the concentration of metal ions at the bottom of pit,mol�L�1;cb – the concentration of metal ions in bulk solution, mol�L�1.

Since the development of pitting corrosion was supposed to becontrolled by the oxygen reduction at the film surface, cs was con-stant because the metal ions at the bottom of pit were saturated. cbcan be regarded as zero because the metal ions’ concentration inbulk solution was very low. So these equations, Fick’s law and Fara-day’s law) can be simplified as:

n ¼ D � S � csh

� t ð7Þ

i0 ¼ z � F � D � csh

ð8Þ

Calculating the integral of Eq. (5), and substituting it into Eq.(8),

h ¼ 2D �M � cq

� �1=2

t1=2 ð9Þ

i0 ¼ z � F � 2D �M � cq

� �1=2

t1=2 ð10Þ

Supposing z � F � q�D�c2M

� �1=2¼ k, Eq. (10) was simplified as:

i0 ¼ k � t�1=2 ð11ÞThe data was fitted to a curve which was also shown in Fig. 8

(Be consistent) Fig. 8. The fitted curve had the following equation,

i0 ¼ 7:6341 � t�0:5 R ¼ 0:996 ð12ÞComparing Eq. (12) with Eq. (11), it is found that they had the

same power index and the correlation coefficient was 0.996, whichsupported the claim that the development of the corrosion throughthe pit was controlled by the ion migration.

Conclusions

By electrochemical techniques, the corrosion kinetics of 316Lstainless steel with chromium carbide coating was investigatedin PEMFC cathodic environment. The EIS results indicated thatthe capacitive resistance and the reaction resistance first slowlydecrease (2–32 h) and then increase, which mean the corrosionrate increased first and then decreased. Although the chromiumcarbide coating had high corrosion resistance and acted as a barrierlayer inhibiting the substrate from corrosion, some pinholes foundin the coating, ultimately lead to failure of the coating by a pittingcorrosion mechanism. The corrosion kinetics equation of thecoated 316L stainless steel was i0 = 7.6341t�0.5 in simulated PEMFCcathodic environment, which meant the corrosion was controlledby the ion migration in the pinholes.

Acknowledgements

This work was financially supported by the National KeyResearch and Development Program of China (Grant No.2016YFB0101206), National Natural Science Foundation of China(21276036, 21676040), the Liaoning Provincial Nature ScienceFoundation of China (2014025018), and the Fundamental ResearchFunds for the Central Universities (3132016341), the Foundation ofLiaoning Educational Committee (L2014199), the Shanghai KeyLaboratory of Digitalized Manufacturing of Complex Sheet MetalStructures (No. 2014004).

References

[1] Liu J, Chen F, Chen YG, Zhang DM. Plasma nitrided titanium as a bipolar platefor proton exchange membrane fuel cell. J Power Sources 2009;187:500–4.

[2] Knights SD, Colbow KM, St-Pierre J, Wilkinson DP. Aging mechanisms andlifetime of PEFC and DMFC. J Power Sources 2004;127:127–34.

[3] Schmittinger W, Vahidi A. A review of the main parameters influencing long-term performance and durability of PEM fuel cells. J Power Sources2008;180:1–14.

[4] Cho EA, Jeon US, Hong SA, Oh IH, Kang SG. Performance of a 1kW-class PEMFCstack using TiN-coated 316 stainless steel bipolar plates. J Power Sources2005;142:177–83.

[5] Gamboa SA, Gonzalez-Rodriguez JG, Valenzuela E, Campillo B, Sebastian PJ,Reyes-Rojas A. Evaluation of the corrosion resistance of Ni-Co-B coatings insimulated PEMFC environment. Electrochim Acta 2006;51:4045–51.

[6] Fleury E, Jayaraj J, Kim YC, Seok HK, Kim KY, Kim KB. Fe-based amorphousalloys as bipolar plates for PEM fuel cell. J Power Sources 2006;159:34–7.

[7] Jayaraj J, Kim YC, Kim KB, Seok HK, Fleury E. Corrosion studies on Fe-basedamorphous alloys in simulated PEM fuel cell environment. Sci Technol AdvMater 2005;6:282–9.

[8] Wang H, Brady MP, Teeter G, Turner JA. Thermally nitrided stainless steels forpolymer electrolyte membrane fuel cell bipolar plates: Part 1: Model Ni–50Crand austenitic 349TM alloys. J Power Sources 2004;138:86–93.

[9] Brady MP, Wang H, Yang B, Turner JA, Bordignon M, Molins R, et al. Growth ofCr-nitrides on commercial Ni-Cr and Fe-Cr base alloys to protect PEMFCbipolar plates. Int J Hydrogen Energy 2007;32:3778–88.

[10] Yang B, Brady MP, Wang H, Turner JA, More KL, Young DJ, et al. Protectivenitride formation on stainless steel alloys for proton exchange membrane fuelcell bipolar plates. J Power Sources 2007;174:228–36.

[11] Brady MP, Yang B, Wang H, Turner JA, More KL, Wilson M, et al. The formationof protective nitride surfaces for PEM fuel cell metallic bipolar plates. JOM-US2006;58:50–7.

[12] Zhang HB, Lin GQ, Hou M, Hu L, Han ZY, Fu Y, et al. CrN/Cr multilayer coatingon 316L stainless steel as bipolar plates for proton exchange membrane fuelcells. J Power Sources 2012;198:176–81.

[13] Wu B, Lin GQ, Fu Y, Hou M, Yi BL. Chromium-containing carbon film onstainless steel as bipolar plates for proton exchange membrane fuel cells. Int JHydrogen Energy 2010;35:13255–61.

[14] Wu B, Li HK, Lin GQ, Fu Y, Hou M, Yi BL. Research of modification of stainlesssteel bipolar plates with Cr1-Xnx films deposited by arc ion plating. ActaMetall Sin 2009;45:1125–9.

[15] Wu B, Fu Y, Xu J, Lin GQ, Hou M. Chromium nitride films on stainless steel asbipolar plate for proton exchange membrane fuel cell. J Power Sources2009;194:976–80.

[16] Fu Y, Lin GQ, Hou M, Wu B, Shao ZG, Yi BL. Carbon-based films coated 316Lstainless steel as bipolar plate for proton exchange membrane fuel cells. Int JHydrogen Energy 2009;34:405–9.

736 N.B. Huang et al. / Results in Physics 6 (2016) 730–736

[17] Fu Y, Hou M, Lin GQ, Hou JB, Shao ZG, Yi BL. Coated 316L stainless steel withCrxN film as bipolar plate for PEMFC prepared by pulsed bias arc ion plating. JPower Sources 2008;176:282–6.

[18] Feng K, Shen Y, Sun HL, Liu DL, An QZ, Cai X, et al. Conductive amorphouscarbon-coated 316L stainless steel as bipolar plates in polymer electrolytemembrane fuel cells. Int J Hydrogen Energy 2009;34:6771–7.

[19] Yi PY, Peng LF, Feng LZ, Gan P, Lai XM. Performance of a proton exchangemembrane fuel cell stack using conductive amorphous carbon-coated 304stainless steel bipolar plates. J Power Sources 2010;195:7061–6.

[20] Larijani MM, Yari M, Afshar A, Jafarian M, Eshghabadi M. A comparison ofcarbon coated and uncoated 316L stainless steel for using as bipolar plates inPEMFCs. J Alloy Compd 2011;509:7400–4.

[21] Foulkes FR, McGrath P. A rapid cyclic voltammetric method for studyingcement factors affecting the corrosion of reinforced concrete. Cem Concr Res1999;29:873–83.

[22] Bentur A. Steel corrosion in concrete: spons architecture price book, 1998.1998-03-24.

[23] Saremi M, Mahallati E. A study on chloride-induced depassivation of mild steelin simulated concrete pore solution. Cem Concr Res 2002;32:1915–21.

[24] Wang F, Watanabe T. Preparation and characterization of the electrodepositedFe-Cr alloy film. Mater Sci Eng A Struct 2003;349:183–90.

[25] Duan SLH. Hard chromium plating from a trivalent plating bath. Plat SurfFinish 1995;6:84–9.

[26] Yamashita T, Hayes P. Analysis of XPS spectra of Fe2+ and Fe3+ ions in oxidematerials. Appl Surf Sci 2008;254:2441–9.

[27] Roosendaal SJ, van Asselen B, Elsenaar JW, Vredenberg AM, Habraken FHPM.The oxidation state of Fe(1 0 0) after initial oxidation in O2. Surf Sci1999;442:329–37.

[28] Muhler M, Schlögl R, Ertl G. The nature of the iron oxide-based catalyst fordehydrogenation of ethylbenzene to styrene 2. Surface chemistry of the activephase. J Catal 1992;138:413–44.

[29] Walton JC, Cragnolino G, Kalandros SK. A numerical model of crevice corrosionfor passive and active metals. Corros Sci 1996;38:1–18.

[30] Walton JC. Mathematical modeling of mass transport and chemical reaction increvice and pitting corrosion. Corros Sci 1990;30:915–28.

[31] Sharland SM, Jackson CP, Diver AJ. A finite-element model of the propagationof corrosion crevices and pits. Corros Sci 1989;29:1149–66.

[32] Sharland SM, Tasker PW. A mathematical model of crevice and pittingcorrosion—I. The physical model. Corros Sci 1988;28:603–20.

[33] Sharland SM. A mathematical model of crevice and pitting corrosion—II. Themathematical solution. Corros Sci 1988;28:621–30.