International Journal of Minerals, Metallurgy and MaterialsVolume 21, Number 9, September 2014, Page 913DOI: 10.1007/s12613-01 4-0989-3Effects of environmental factors on corrosion behaviors of metal-fiber porouscomponents in a simulated direct methanol fuel cell environmentWei Yuan, Bo Zhou, Yong Tang, Zhao-chun Zhang, and Jun DengKey Laboratory of Surface Functional Structure Manufacturing of Guangdong Higher Education Insitutes, School of Mechanical and Automotive Engineering, SouthChina University of Technology, Guangzhou 510640, China(Received: 13 December 2013; revised: 11 March 2014; accepted: 20 March 2014)Abstract: To enable the use of metallic components in direct methanol fuel cells (DMFCs), issues related to corrosion resistance must beconsidered because of an acid environment induced by the solid electrolyte. In this study, we report the electrochemical behaviors ofmetal-fiber-based porous sintered components in a simulated corrosive environment of DMFCs. Three materials were evaluated: pure copper,AISI304, and AISI316L. The environmental factors and related mechanisms afcting the corrosion behaviors were analyzed. The resultsdemonstrated that AISI316L exhibits the best performance. A higher SO% concentration increases the risk of material corrosion, whereas anincrease in methanol concentration inhibits corrosion. The morphological features of the corroded samples were also charaterized in thisstudyKeywords: corrosion; metal fibers; porous; sintering; direct methanol fuel cells; environmental factorsHowever, the graphite plate is brittle, and therefore lacks1. Introductionmechanical resistance to external vibration; this brittlenessalso results in higher production costs related to the creationThe emergence of fuel-cell technology illuminates theof flow fields. With this background, numerous researchershave attempted to use metal as an alternative to graphite [8].secondary batteries by virtue of fuel cells' advantages suchIn contrast to graphite, metals are more machinable and ex-as higher energy efficiency and lower pollutant emissionshibit a greater mechanical strength. Thus, a wide variety of[1-2]. As a promising candidate, the direct methanol fuelshaping methods, such as cutting, stamping, etching, ancell (DMFC) has attracted considerable attention and iseven rapid prototyping, can be used to fabricate metallicstructurally similar to the popular hydrogen-based protoncomponents that can be made very thin to reduce both theexchange membrane (PEM) fuel cell. A DMFC uses amount of material used and the system volume. Despitemethanol directly as the fuel source without a reformingthese aluring properties, researchers who investigate the useprocess, which allows a more compact system structure andof metal components in DMFCs must consider the aggres-correspondingly higher energy density; DMFCs thereby ex-sive environment in such fuel cells, which are based on per-hibit great potential for use in portable applications [3- -4].fluorosulfonic acid membranes such as the Nafion° seriesAs a core component of the DMFC, the bipolar plate withmanufactured by DupontM. If the metal corrodes, the re-patterned flow fields not only helps distribute the reactantsleased elements inevitably poison the catalysts and hurt theand products but also serves as a current collector. In mostmembrane. Such effects must be avoided because they willof the early studies on DMFCs, non-porous graphite wassubstantially reduce the long-term performance and durabil-used in the fabrication of such components because of itsity of fuel cells. The open literature contains extensive re-good electrical conductivity and high chemical stability inports on the corrosion phenomena and anti-corrosion tech-the harsh environment of a PEM-based fuel cell [5- -7nology with regard to various metals and alloys [9- 16].Corresponding author: Wei YuanE-mail: mewyuan@scut.edu.cn◎University of Science and Technology Bejing and Springer-Verlag Berlin Heidelberg 2014中國煤化工李SpringerMYHCNM HG914Int. J. Miner. Metall. Mater, Vol. 21, No.9, Sep. 2014However, most previous reports have focused on the envi-2. Experimentalronment in hydrogen-fed PEM fuel cells. Detailed informa-tion concerning DMFCs in this field is scarce, although afew groups have used solid and porous metallic materials torials for corrosion evaluation (Table 1). The tested samplesfabricate DMFC components including the flow distributor,are shown in Fig. 1. The fibers constructed of pure copperdiffusion medium, and current collector [17-19].and AISI304 were produced by low-speed cutting with aTo this end, we recently reported the corrosion behaviorssuper-hard multitooth cutter on a horizontal lathe, as shownof several potential metallic materials in both simulated andin Fig. 2. A row of triangle microteeth were machined onactual DMFC environments [20] We intended to create thinthe cutting face of this tool so as to produce a string of con-fibers from these materials for fabricating a porous sinteredtinuous long fibers with diameters of 50100 um at one time.flow distributor, which serves as an anodic mass-transfer-The fibers were cleaned, segmented, loaded into a mold as-controlling medium in a passive DMFC to optimize thesembly, and then sintered in a resistance furnace. By adjust-management of multiphase flow. However, we reporteding the cutter geometry, cutting parameters, and sinteringonly preliminary results related to the electrochemical andconditions, we could control the structure and morphologymorphological characteristics of the candidate materials un-of the fibers and sintered felts. Furthermore, we could adap-der corrosive conditions. As an extension of our previous tively optimize their mechanical properties and functionalstudy, we presented additional information about the effectscharacteristics. In this experiment, the porosities of all theof environmental factors on the candidate materials in thissamples were 80%. More information about the processingpaper. .parameters is available elsewhere [21].Table 1. Three types of porous metallic materials investigated in this studySample No. Material Fiber production methodMajor element content/ w%Sintering conditions0.3 MPa (hydrogen atmosphere);l#CopperMultitooth cutingCu:≥99.9.5°Cmin+; 900C for 1 h; furnace cooling40 Pa (vacum); 5°Cmin ; 1200°C for#AISI304 Mulitooth cuttingCr: 18- 20; N: 8- -10; Mn:≤2; Si:≤11 h; furnace coolingAISI316LBundle drawingCr: 16-18; Ni: 10-14; Mo:2- 3; Mn:≤2; Si:≤1Purchased#i3#Fig. 1. Photos of three types of porous metallic samples.The effective size of the tested samples was 15 mmx 12air-breathing mode of a practical DMFC.mm x 2.5 mm. Before being tested, the samples wereThe electrochemical experiments were conducted using acleaned ultrasonically with acetone and then with deionizedthree-electrode system on an electrochemical workstationwater and were finally dried in air. A simulated solution(AUTOLAB PGSTAT302). The ambient temperature wascomposed of specified amounts of sulfuric acid (H2SO4),maintained at 25°C. Platinum foil (Pt≥99.95wt%, 20 mm Xhydrofluoric acid (HF), and methanol (CHzOH) was pre-15 mm x 0.2 mm) was used to prepare the auxiliary elec-pared. To evaluate the effects of environmental factors, wetrode, whereas a saturated calomel electrode (SCE) wasprescribed different solution compositions, as listed in Tableused as the refer中國煤化工wise seirie2. The cathode solution was bubbled to simulate theall electrode pot; SCE. For po-YHCNMH GW. Yuan et al, Effects of environmental factors on corrosion behaviors of metal-fiber porous components in a simulated... 915tentiodynamic polarization measurements, the scan rate wassolution in this case. The two polarization curves of samplemaintainedat 1 mV.s " ranging from -0.5 to 1.5 V. Before1# are almost coincident, showing obvious insensitivity toeach run, the sample was first stabilized under open-circuitthe SO 4 concentration. This behavior implies that the ef-conditions for 30 min. The surface morphology of eachfect of this acid ion can be neglected. In contrast, an increasesample was characterized by micrographic methods.in methanol concentration tended to slightly shift the curverightward, which indicates that the use of a higher methanolMili-ooth cuttoer后10Collected fibers10-3上Cu, 0.5 mol/L10-+F--_ Cu, 2.0 mol/LWorkpiece10-5 fAISI304, 0.5 mol/L---- AISI304, 2.0 mol/LAISI316L, 0.5 mol/L10-6 fAISI3 16L, 2.0 mol/LFig. 2. Multi-tooth cutting for metal fiber production.-0.6- -0.4 -0.20.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6PotentialTable 2. Compositions of simulated environmentsig. 3. Efects of SO4 concentration on polarization behav-ElectrodeSimulated DMFC so2 1CH,OH4 1F/wt%iors in the simulated DMFC anode environment.environment (mol:L(mol:L-)High so,2.02x 100P rLow so42x 10-*01AnodeHigh CH2OH.02x 104Low CH;OH0.52x 10-403 t2.0 .0.01 (Methanol0-f--- Cu, 8.0 mol/Lcrossover)Cathode一AISI304, 0.5 mol/LLow so,. AISI3I6IO5molL0* tAISI3 16L, 8.0 mol/L3. Results and discussion-0.6-0.4-0.20.0 0.20.40.60.81.01.21.4 1.6Potential/ V vs. SCE3.1. Electrochemical corrosion behaviorFig. 4. Effects of CH3OH concentration on polarization be-Figs. 3 -5 compare the polarization behaviors of the three haviors in the simulated DMFC anode environment.samples in simulated anode and cathode environments of aO° rDMFC when solutions with different compositions wereused. With respect to sample 1#, no activation-to-passiva-tion transfer was apparent in its polarization curves. Th02 temergence of multiple corrosion potentials in the negative0-3↑region in Figs. 3 and 4 suggests that the system was still in-Cu, 0.5 molLchaos because the pure copper was easily oxidized. A rela----- Cu, 2.0 mol/L一- AISI304, 0.5 mol/Ltively stable stage was observed, indicating that a balanceAISI304, 2.0 mol/Lstatus \was formed in this section. After the potential was in-. AISI316L, 0.5 mol/L0°---- AISI3 16L, 2.0 mol/Lcreased to approximately -0.05 V, the current rapidly in-creased to a very high value with a magnitude of10 1 A. We-0.6-0.4-0.20.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6further noted that the current continued increasing at thepoint where the anode working potential was applied (0.2 V),Fig. 5. Effets中國煤化Irization behav-which suggests that the material underwent continuous dis- iors in the simula:YHCNMHGnt..916Int. J. Miner. Metall. Mater, Vol. 21, No.9, Sep. 2014concentration possibly enhances the corrosion resistance viasistance. Another noteworthy phenomenon is that the cur-the formation of a local protective area near the sample sur-rent of sample 3# exhibits a sharp decrease before reachingface. In the case of the cathode environment (see Fig. 5), thethe stable stage. This profile reveals that the material con-point of the applied cathode working potential (0.7 V) is lo-tinues to resist corrosion under the increasing potential untilcated in the high-current region, indicating that pure copperforms a stable passive film. The curves related to sample 3#cannot withstand corrosion in this environment.again confirm that a higher SO 4 concentration promotesFigs. 3-5 also ilustrate the corrosion behaviors of sam-corrosion, whereas a higher methanol concentration reducesples 2# and 3#. The anode and cathode polarization curvesthe degree of corrosion. Thus, we inferred that the use of aindicate an activation-to-passivation transfer process. Afterconcentrated fuel is favorable to resist corrosion of fuel cellthis transience, the curves returm to a relatively stable stage,components. The merit of this advantage lies in the high-reflecting the formation of a stable passive film on the mate-concentration operation, which will significantly enhance therial surface. The current at both the applied anode and cath-energy density and prolong the operating time of a DMFC.ode potentials resided in the stable passivation regions, sug-gesting that both AISI304 and AISI3 16L exhibit good cor-3.2. Morphological descriptionrosion resistance under the simulated conditions. We furtherFigs. 6 8 describe the morphological features of the threenoted that the SO 4 concentration of 2.0 molL-1 promotedsamples before and after corrosion tests. As evident froma higher corrosion current than the So4concentration ofresults in Fig.6, the color of the copper fiber changed from0.5 molL". This result agrees well with the rationale ofbright yellow to dark brown with burr-like structures formedelectrochemical corrosion, which indicates that a higherand distributed on its surface. Such obvious changes in bothsO 4 concentration leads to a higher corrosion rate. Whena higher concentration of So2 is used, more acid ions of copper in the simulated DMFC environment [20]. In viewcontact the passive film. In this case, the electrical conduc-of the AISI304 fibers, a few distinct yellow spots appearedtivity between the tested sample and the solution must beon its surface after corrosion, as shown in Fig. 7. Austeniticenhanced so as to lower the corrosion resistance of the ma-stainless steel is known to exhibit a higher corrosionterial. Fig. 4 shows that the current corresponding to;methanol concentration of 8.0 molL' is lower than thatcorresponding to a concentration of 0.5 mol'L' in the pas-sivation region. This lower current can be attributed to thedecrease in electrical conductivity of the working solution.Another possible reason for this phenomenon is the fact thatthe addition of CH3OH facilitates the formation of a shieldlayer that protects the passive film from extreme corrosion.Notably, the polarization curve of sample 3# exhibited twominimum values of current density in the negative-voltageregion when 8.0 mol'L' methanol was used. This phe-nomenon can be ascribed to the use of a higher methanolconcentration and also to the corrosion resistance of(bAISI316L. These two factors combine to facilitate theemergence of an altermant process of formation and dissolu-tion of the passive film with the increase in load potential.Although the polarization curve for sample 3# exhibits aprofile similar to that of sample 2#, it indicates a muchlower magnitude of corrosion current compared to that ofsample 2#. Meanwhile, the corrosion potential of sample 3#becomes more positive than that of sample 2#. These resultsconfirm that AISI3 16L performs better than AISI304 in thesimulated DMFC environment. This better performance oAISI316L is attributed to the inclusion of molybdenum andFig. 6. Morphol中國煤化Iple 1# before (a)to its higher nickel content, which enhances its corrosion re-and after (b) corrYHCNMHGW. Yuan et al, Effects of environmental factors on corrosion behaviors of metal-fiber porous components in a simulated... 917resistance when a dense chromium-rich oxide film, whichprevents continuous penetration of the corrosive medium, isformed on its surface. When tested in the acid environment,the protective film might be oxidized by the oxidizing solu-tion to produce a yellow-colored chromate solution. Fig. 8shows the appearance of the AISI316L fibers produced bythe bundl-drawing technique. Compared with the fibersprepared by cutting, such fibers have a relatively smoothersurface. After undergoing corrosion, the fiber surface be-comes very coarse because of the formation of numerousmicropores/cavities on its surface, which can be verified bythe magnified view of fibers (see bottom-right SEM imagesin Fig. 8).4. Conclusions(1) Among the three tested samples, pure copper per-formed the worst in the simulated DMFC environment. Thestainless steel AISI304 and AISI3 16L samples both exhib-ited good corrosion resistance because of the formation of apassive film at the applied potentials. Further comparisonrevealed that AISI3 16L performed better than AISI304.(2) Copper exhibited a less sensitivity to the change inFig 7. Morphological characterization of sample 2# before (a) so 2- concentration, which strongly affected stainless steel.and after (b) corrosion.A higher concentration of so 4- tended to promote corro-sion. These results suggest that the methanol concentrationused in DMFCs should be increased to mitigate corrosion,consistent with the aim of realizing high-concentration op-eration of a practical DMFC to enhance its energy densityand operating time.(3) Microscopic characterization revealed that non-vis-ualized destroyed traces were present on the surface ofAISI316L. From the perspective of long-term applications,the exploration of effective anticorrosion methods is impor-tant to protect AISI316L from corrosion in a DMFC envi-ronment.20020 DW 103年20 SE 10200181418，... wuAcknowledgementsThis work was financially supported by the Natural Sci-ence Foundation of Guangdong Province, China (No.S2013040016899)，the Fundamental Research Funds forCentral Universities of China (No. 2013ZM0003), the Na-tional Natural Science Foundation of China (No. 51275180),and the Open Fund of Shanghai Key Laboratory of DigitalManufacture for Thin-walled Structures (No. 2013001).ReferencesFig. 8. Morphological characterization of sample 3# before (a)中國煤化工[1] G. Hoogers,MYHCN MH G Hinsberger, M.918Int. J. Miner. Metall. Mater, Vol. 21, No.9, Sep.2014Hogarth, R. Stone, and D. Thompsett, Fuel Cell Technologymembrane fuel cells, J. Power Sources, 230(2013), p. 25.Handbook, CRC Press, Boca Raton, 2003, p.1.[13] C.H. Liang, C.H. Cao, and N.B. Huang, Electrochemical be-2] CK. Dyer, Fuel cells for portable applications, J. Powerhavior of 304 stainless steel with electrodeposited niobium asSources, 106(2002), No.1-2, p.31.PEMFC bipolar plates, Int. J. Miner. Metall. Mater,3] W.M. Qian, D.P. Wilkinson, J. Shen, HJ. Wang, and JJ.19(2012), No. 4, p.328.Zhang, Architecture for portable direct liquid fuel cells, J.[14] Y. Yang, LJ. Guo, and HT. Liu, Factors affecting corrosionPower Sources, 154(2006), No. l,p. 202.behavior of SS316L as bipolar plate material in PEMFC4] A.S. Arico, S. Srinivasan, and V. Antonucci, DMFCs: fromcathode envronments, Int. J. Hydrogen Energy, 37(2012),fundamental aspects to technology development, Fuel Cells,No. 18, p. 13822.1(2001), No.2, p.133.[15] R. Wlodarczyk and A. Wronska, Effect of pH on corrosion of5] J.S. Cooper, Design analysis of PEMFC bipolar plates con-sintered stainless steels used for bipolar plates in polymersidering stack manufacturing and environment impact, J.exchange membranefuel cells, Arch. Metall. Mater,Power Sources, 129(2004), No. 2, p. 152.58(2013), No.1,p.89.6] W. Schmittinger and A. Vahidi, A review of the main pa- [16] S. Laedre, O.E. Kongstein, A. Oedegaard, F. Seland, and H.rameters influencing long-term performance and durability ofKaroliussen, The effect of pH and halides on the corrosionPEM fuel cells, J. Power Sources, 180(2008), No.1,p. 1.process of stainless steel bipolar plates for proton exchange7] JF. Wu, X.Z. Yuan, JJ. Martin, HJ. Wang, JJ. Zhang, J.membrane fuel cells, Int. J. Hydrogen Energy, 37(2012), No.Shen, S.H. Wu, and W. Merida, A review of PEM fuel cell23, p. 18537.durability: degradation mechanisms and mitigation strategies,[17] R. Chen and T.S. Zhao, Porous current collectors for passiveJ. Power Sources, 184(2008), No. 1, p. 104.direct methanol fuel cll, Electrochim. Acta, 52(2007), No.8] H. Tawfik, Y. Hung, and D. Mahajan, Metal bipolar plates13, p. 4317.for PEM fuel cell: a review, J. Power Sources, 163(2007), No.[18] T. Shudo and K. Suzuki, Performance improvement in direct2, p.755.methanol fuel cells using a highly porous corrosion-resisting9] RA. Antunes, M.CL. Oliveira, G. Ett, and V. Ett, Corrosionstainless steel flow field, Int. J. Hydrogen Energy, 33(2008),of metal bipolar plates for PEM fuel cells: a review, Int. J.No.11, p. 2850.Hydrogen Energy, 35(2010), No. 8, p. 3632.[19] J.G. Liu, G.Q. Sun, FL. Zhao, G.X. Wang, G. Zhao, LK.[10] HL. Wang, M.A. Sweikart, and J.A. Turmer, Stainless steelChen, B.L. Yi, and Q. Xin, Study of sintered stainless steelas bipolar plate material for polymer electrolyte membranefiber felt as gas diffusion backing in air-reathing DMFC,J.fuel cells, J. Power Sources, 115(2003), No. 2, p. 243.Power Sources, 133(2004), No.2, p.175.[1] P.Y. Yi, LF. Peng, T. Zhou, J.Q. Huang, and X.M. Lai, [20] W. Yuan, Y. Tang, X. Yang, L. Bu, and Z. Wan, CorrosionComposition optimization of multilayered chromium-nitride-behavior of porous metal fiber-sintered felt in both simulatedcarbon film on 316L stainless steel as bipolar plates for pro-and practical environments of a direct methanol fuel cell,ton exchange membrane fuel cells, J. Power Sources,Corrosion, 69(2013), No. 1,p. 25. .236(2013), p.47.[21] W. Yuan, Y. Tang, XJ. Yang, and ZP. Wan, Toward using[12] P.Y. Yi, LF. Peng, T. Zhou, H. Wu, and X.M. Lai, Devel-porous metal-fiber sintered plate as anodic methanol barrieropment and characterization of multilayered Cr -C/a- C:Cr flmin a passive direct methanol fuel cell, Int. J. Hydrogen Energy,on 316L stainless steel as bipolar plates for proton exchange37(2012), No.18, p. 13510.中國煤化工MYHCNMH G