Journal of Power Sources 295 (2015) 175e181
Contents lists available at ScienceDirect
Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour
Nanoporous silver cathodes surface-treated by atomic layer deposition of Y:ZrO2 for high-performance low-temperature solid oxide fuel cells You Kai Li 1, Hyung Jong Choi 1, Ho Keun Kim, Neoh Ke Chean, Manjin Kim, Junmo Koo, Heon Jae Jeong, Dong Young Jang, Joon Hyung Shim* School of Mechanical Engineering, Korea University, Anam-dong, Seongbuk-gu, Seoul 136-713, South Korea
h i g h l i g h t s Performance of Ag electrodes is improved by ALD YSZ surface treatment. Power of SOFCs with Ag-ALD YSZ is similar to that of Pt fuel cells at LT regimes. ALD YSZ capping improved surface kinetics and long-term stability of Ag electrodes. Surface ALD YSZ prevents passivation of ORRs.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 29 March 2015 Received in revised form 21 June 2015 Accepted 25 June 2015 Available online 11 July 2015
We report high-performance solid-oxide fuel cells (SOFCs) with silver cathodes surface-treated using yttria-stabilized zirconia (YSZ) nano-particulates fabricated by atomic layer deposition (ALD). Fuel cell tests are conducted on gadolinia-doped ceria electrolyte pellets with a platinum anode at 250e450 C. In our tests, the fuel cell performance of the SOFCs with an optimized ALD YSZ surface treatment is close to that of SOFCs with porous Pt, which is known as the best performing catalyst in the low-temperature regime. Electrochemical impedance spectroscopy conﬁrms that the performance enhancement is due to improved electrode kinetics by the increase in charge transfer reaction sites between the surface of supporting silver and the ALDeYSZ particulates. Fuel cell durability tests shows that the ALD YSZ surface treatment improves the long-term stability. X-ray photoelectron spectroscopy also conﬁrms that the ALD YSZ capping prevents reduction of the surface silver oxide and destruction of the mesh morphology. © 2015 Elsevier B.V. All rights reserved.
Keywords: Solid oxide fuel cells Cathode Silver Atomic layer deposition Yttria-stabilized zirconia
1. Introduction Solid-oxide fuel cells (SOFCs) have attracted attention as nextgeneration generators because of their high conversion efﬁciency, thermal cycling capability, and fuel ﬂexibility. However, the high operation temperature of SOFCs (>800 C) has limited their practical use because of various problems, including long startup time, thermally induced stresses between components, chemical degradation, and mechanical degradation. A variety of efforts have been undertaken to use SOFCs below intermediate temperatures (IT) (~500 C) by using thin ﬁlm electrolytes [1e5] or novel electrode materials [6e8]. However, the key to develop high-performance
* Corresponding author. E-mail address: [email protected]
(J.H. Shim). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.jpowsour.2015.06.136 0378-7753/© 2015 Elsevier B.V. All rights reserved.
sub-IT SOFCs is innovative cathode materials, because cathodic polarization dominates the overall energy loss in low-temperature operations, while impedance from ion transport across the electrolyte is the major cause of loss in high-temperature regimes. The most typical cathode materials for SOFCs are doped perovskite ceramics such as La1xSrxCo1yFeyO3d (LSCF) [9,10], La1xSrxMnO3d (LSM) [11,12], La1xSrxCoO3d (LSC) [13,14], Ba1xSrxCo1yFeyO3d (BSCF) [15,16], and Sm1xSrxCoO3d (SSC) [17,18]. However, activation energies of these materials for oxygen reduction reactions (ORRs) on the surface and ion transport through the electrode bulk are reported to be relatively high (1.5e2.0 eV) [19,20]. Hence, the overall cathodic impedance dramatically increases even with a slight decrease in the operation temperature. For this reason, perovskite-type cathodes are considered inappropriate for SOFCs that operate below IT regimes. Despite this, Pt or Pt alloys are regarded as the best performing material for low-temperature fuel cells [21,22]. However, the use of
Y.K. Li et al. / Journal of Power Sources 295 (2015) 175e181
Pt is not economical because of its high price (US$ 41.3 g1 as of Feb 20, 2015) [23,24]. Recent studies have reported Ag or Ag alloys as novel cathode materials for SOFCs because of their high oxygen solubility and diffusivity [25,26]. The price of Ag (US$ 0.58 g1 as of Feb 20, 2015) is lower than that of Pt [27,28], and the ORRs on the surface of Ag can be as active as those on the surface of Pt, or even more active in certain circumstances, in the sub-IT regime [29,30]. ORRs on Ag can occur either at the triple phase boundary (TPB) where the cathode, electrolyte, and oxygen gas are in physical contact (TPB-ORR) or over the entire surface of Ag followed by diffusion of O2 through the Ag bulk (Ag surface-ORR). Below the IT regime, the TPB-ORR process is more dominant than the Ag surface-ORR process. This is supported by both experiments and simulations. This implies that the increase in TPB sites between the electrode and electrolytes is essential for the effective use of Ag as a high-performance cathode material for sub-IT SOFCs. In this work, we propose a porous Ag nano-mesh, surfacecoated with yttria-stabilized zirconia (YSZ) nano-particulates, as a cathode for high-performance sub-IT SOFCs. The schematic of this approach is shown in Fig. 1. Because the YSZ electrolyte particles cover the Ag cathode, the TPB sites can be extended to the entire surface of Ag. However, there should be room for oxygen gas diffusion between the nanoparticle layers. Therefore, the amount of YSZ particulates should be optimized. Fabrication of the YSZ nano-particles is performed by atomic layer deposition (ALD), which is a type of chemical vapor deposition (CVD). In ALD, a vaporized source precursor is supplied to a reaction chamber and begins to deposit material onto the surface of substrates under appropriate conditions, i.e., the appropriate temperature and pressure and the presence of proper reaction groups on the substrate surface. Material deposition stops after the formation of one atom layer or less because the deposited layer is “blanketed” with the precursor ligands, preventing further chemisorption of more precursors onto the surface. The “blanket layer” is then removed after purging breaks the bond between the “blanket layer” and the deposited layer. A layer of a second material can be deposited after purging by introducing a second precursor into the reaction chamber. The second deposition will also be terminated by “ligand blanketing.” Through iteration of these depositionepurgeedepositionepurge processes, we can determine the thickness and composition of ﬁlms from the number of precursor cycles, independent of the precursor dose or the source injection direction, on a very precise scale, i.e., the scale of one atomic layer
or less. This unique characteristic of ALD has enabled deposition of ultrathin ﬁlms (<5 nm), laminating layers, and nanoparticles on the surface of complex-shaped substrates with high-aspect-ratio features [31,32]. In our work, we attempted uniform deposition of small ALD YSZ particulates on the surface of high-aspect-ratio Ag nano-meshes to be used as cathodes for SOFCs. Performance of the ALD YSZ-Ag mesh was evaluated in terms of morphology, electrochemical impedances, and fuel cell power. 2. Experimental 2.1. Preparation of electrolytes and electrodes The electrolytes were prepared by die-pressing Gd0.1Ce0.9O2x powder (GDC10, Rhodia, Korea) and sintering it at 1460 C for 8 h. The GDC pellets were polished to a thickness of 350 mm. Using DC magnetron sputtering (Korea Vacuum Tech., Ltd.), 150-nm-thick nano-porous Ag as the cathode and Pt as the anode were sputtered onto the GDC electrolyte pellet. Ag was deposited at a DC power of 260 W at 80 m Torr Ar pressure, while Pt was deposited at a DC power of 350 W at 90 m Torr Ar pressure. 2.2. ALD of YSZ A customized thermal ALD chamber (ICOT Inc.) was employed for the fabrication of YSZ. Commercial Tetrakis (ethlymethylamido) zirconium(IV) (LPN Inc.) and tris (methylcyclopentadienyl) yttrium(III) (Strem Chem.) were used as the cation precursors for ZrO2 and Y2O3. Distilled water was used as the oxidation source, and dry nitrogen (N2) was used as the carrier and purge gas at a ﬂow rate of 1.0 sccm. The deposition temperature was set at 250 C. The deposition cycle ratio of ZrO2 and Y2O3 was 4:1 (ZrO2:Y2O3 ¼ 4:1); the deposition cycle of Y2O3 was performed in the middle of the cycles. This ratio was proven to successfully produce 8e10 mol % YSZ [1,4,33,34]. We performed 7 and 15 deposition cycles of ALD YSZ on Ag, respectively. Hereafter, the Ag cathodes treated with 7 and 15 cycles of ALD YSZ are referred to as Ag-ALD YSZ 7 and AgALD YSZ 15, respectively. 2.3. SOFC characterization The microstructure and morphology of the fuel cells were observed and analyzed using scanning electron microscopy (SEM; Hitachi S-4300) and high-resolution transmission electron
Fig. 1. Procedure for sample preparation, i.e., ALD YSZ treatment on the sputtered Ag mesh used as a cathode for SOFCs operating in intermediate temperature ranges, and a schematic of the performance optimization of ALD YSZ surface treatment to improve electrode kinetics on Ag.
Y.K. Li et al. / Journal of Power Sources 295 (2015) 175e181
microscopy (HRTEM; JEM-2100 F). SOFCs were operated in a probe testing system with 20 sccm dry hydrogen gas ﬂow at the anode and ambient atmosphere at the cathode side. Pt wire was contacted to the cathode surface for current collection, and a gold ring was used for current collection and gas sealing at the anode side. More details are given in our previous publications [5,34,35]. The currentevoltage (IeV) curves and impedance data were investigated to evaluate the fuel cell performance during operation in a temperature range of 250e450 C using a potentiostat/galvanostat instrument (Gamry Reference 3000). In the IeV measurement, data were collected from open circuit voltage (OCV) to 0.2 V using a linear sweep, and the impedance frequency was measured over a range of 106e0.2 Hz using electrochemical impedance spectroscopy (EIS). The impedance data were ﬁtted and the area speciﬁc resistance (ASR) of each cell at the measurement temperature were calculated using a commercial ﬁtting software (Gamry PHS200). 3. Results and discussion Fig. 2 shows the result of the IeV measurement of SOFCs with bare Ag and Ag treated with ALD YSZ. For comparison, we also prepared an SOFC with 150-nm-thick porous Pt using the same method as that used for anode sputtering. As a result, the maximum power density from the sample with the bare Ag cathode was 10.6 mW cm2 at 450 C, while the peak power from the fuel cell with Ag optimally treated with ALD YSZ or the Ag-ALD YSZ 7 cathode was 14.7 mW cm2. In comparison, the sample with the porous Pt cathode, which is known as the best-performing lowtemperature catalyst, produced a maximum power of 16.5 mW cm2 at 450 C; thus the performance difference compared to Ag-ALD YSZ 7 was merely about 10%. In other temperature regimes, it was conﬁrmed that the performance difference between the samples with Pt and Ag-ALD YSZ 7 cathodes was 1.8e13.5% in terms of power output. As the amount of ALD YSZ on Ag increased, the fuel cell performance dropped signiﬁcantly compared with that of the Ag-ALD YSZ 7 and Ag-ALD YSZ 15 cathode samples. At 450 C, the maximum power of the Ag-ALD YSZ 15 sample was 9.5 mW cm2, which was not only lower than the performance of the Ag-ALD YSZ 7 cell by about 35% but also lower than that of the untreated Ag cathode. This performance degradation is assumed to be due to the decrease of the TPBeORR sites with an increase in ALD YSZ on the surface of Ag. To verify this, we observed the microstructure of ALD YSZ on Ag with HRTEM. We prepared samples of sputtered Ag treated with YSZ for the characterization. Fig. 3 shows TEM images of bare Ag and YSZ-treated Ag samples for different numbers of ALD YSZ cycles. As shown in Fig. 3(b), the Ag-ALD YSZ 7 sample shows particulate YSZ on the Ag electrode, which can maximize the number of TPB sites for the ORR reaction. This might be the reason why Ag-
ALD YSZ 7 showed a higher power density than the untreated Ag electrode. In contrast, the Ag-ALD YSZ 15 sample forms a conformal, dense YSZ layer that can passivate the Ag electrode by capping the surface and lead to a lower power density than that of the bare Ag electrode. This result agrees with our assumptions described in the introduction and illustrated in Fig. 1. The decrease in performance with temperature change was severe in the sample with the bare Ag cathode. At 450 and 425 C, the SOFC with bare Ag exhibited a peak power of 10.6 and 2.8 mW cm2, respectively. However, the power of the bare Ag cell was less than 4 mW cm2 at 400 C, and the performance remained in the 0.3e1 mW cm2 range below 400 C. Compared to the power of the samples with Pt, Ag-ALD YSZ 7, and Ag-ALD YSZ 15, which were measured as 10.8, 9.9, and 4.2 mW cm2, respectively, at 400 C, the performance of the bare Ag cell was low. This phenomenon is suspected to be due to the formation of silver oxide that passivated the catalytic performance of the Ag cathode surface at temperatures lower than 400 C. It has been reported that silver oxide slabs, which favourably form on the surface of metallic silver at atmospheric conditions in the presence of oxygen, tend to thermally decompose over 600 K or about 325 C [36e39]. Our speculation is that silver oxide that formed on the surface of sputtered Ag began to decompose and open up pores that are essential for the cathodic ORR when heated either in air over 400 C (the bare Ag case) or in vacuum at 250 C (ALD condition, the AgALD YSZ 7 and 15 cases). We also speculate that the ALD YSZ surface layers prevented the formation of silver oxide when the AgALD YSZ samples were unloaded from the chamber and exposed to air, preserving the catalytic reaction sites. The EIS spectra can explain the good performance of the ALD YSZ 7 fuel cell, which was comparable to that of the Pt cell. Fig. 4 shows representative Nyquist impedance spectra of the cells with porous Pt, Ag-ALD YSZ 7, and Ag-ALD YSZ 15 cathodes measured at 450 C during fuel cell operation with varying cell voltage. The spectra ﬁtted well with the data obtained using circuits comprised of two parallel sets of a resistance and constant phase element with a serially connected resistance, as shown in Fig. 4. As a result, there was no signiﬁcant change in the intercept on the real axis when the arcs changed dramatically in response to the applied bias voltage. From this, it is clear that the real axis intercept values represent ohmic resistance majorly caused by ion conduction through the electrolyte, whereas the semicircular arcs indicate polarization impedance from electrode processes [40e42]. In the electrode process, it has been commonly reported that ORR at the cathode side is more sluggish than hydrogen oxidation reactions (HOR) at the anode side . For example, the exchange current density for HOR on a Pt surface in an acid electrolyte at a standard state is 103 A cm2, which is 106 times higher than the ORR under the same conditions . Therefore, it can be assumed that the size of
Fig. 2. Representative IeVeP curves for the GDC-based SOFCs with porous Pt, Ag-ALD YSZ 7, ALD YSZ 15, and bare porous Ag cathodes measured at (a) 425 C and (b) 450 C, (c) performance comparison with respect to peak power density measured at 250e450 C.
Y.K. Li et al. / Journal of Power Sources 295 (2015) 175e181
Fig. 3. HRTEM images of the (a) bare Ag, (b) Ag-ALD YSZ 7, and (c) Ag-ALD YSZ 15 cathodes.
Fig. 4. Representative Nyquist impedance spectra measured at 450 C with varying cell voltages for cells with (a) porous Pt, (b) Ag-ALD YSZ 7, and (c) Ag-ALD YSZ 15 cathodes.
Fig. 5. Comparison of ohmic ASRs measured for cells with porous Pt, Ag-ALD YSZ 7, and ALD YSZ 15 cathodes at (a) OCV and (b) 0.3 V, (c) comparison of conductivity estimated from the ohmic ASRs with a reference GDC10 data . Comparison of polarization ASRs measured for cells with porous Pt, Ag-ALD YSZ 7, and ALD YSZ 15 cathodes at (d) OCV, (e) 0.3 V, and (f) 0.5 V.
Y.K. Li et al. / Journal of Power Sources 295 (2015) 175e181
Fig. 6. Potentiostatic currents measured for working fuel cells with the bare porous Ag, Ag-ALD YSZ 7, and Ag-ALD YSZ 15 cathodes at 0.5 V and 450 C.
the anode arc is much smaller than that of the cathode arc. In addition, as shown in Table S1, the ﬁtted values of the EIS spectra also show that, at the same operating temperature and cell voltage, the polarization resistance of the cells can depend on the type of cathode (Pt, Ag-ALD YSZ 7 and Ag-ALD YSZ 15). The variation of polarization resistance originates entirely from the type of cathode because the three cells consist of the same electrolyte (GDC) and anode (Pt). For these reasons, it can be concluded that the semicircular arcs represent the cathode kinetics. The calculated ohmic ASRs are presented in Fig. 5(a)e(c), and there is no variation of the voltage conditions, as shown in Fig. 5(a) and (b). The ionic conductivity computed using the ASR values and thickness of the GDC electrolyte pellets are close to the GDC10 reference value . A comparison of the polarization ASRs shows that the resistances decrease as the cell voltage decreases, as shown in Fig. 5(d)e(f), because the increase in the overpotential enhances the kinetics of the electrode process. This phenomena has frequently been reported in previous articles [46,47]. The impedance of Ag-ALD YSZ 15 is signiﬁcantly greater than those of the Pt or Ag-ALD YSZ 7 cathodes at low voltages of 0.5 and 0.3 V, as shown in Fig. 5(b) and (c), respectively. This is an indication that the charge transfer or the number of ORReTPB sites have decreased with a
larger amount of ALD YSZ on the Ag surface, as more current is needed at the low voltage levels during fuel cell operation. It is interesting that the polarization ASRs of the Pt and the Ag-ALD YSZ 7 electrodes are similar, although the areas of their charge transfer sites might be different; the ORR is expected to occur over the entire surface of the Ag-ALD YSZ 7 cathode while the reaction takes place mainly at the PteGDC boundary of the Pt cathode [48e50]. We have evaluated the stability of the electrodes over long-term fuel cell operation. Fig. 6 shows the potentiostatic currents measured from all four samples (fuel cells with porous bare silver, Ag-ALD YSZ 7, and Ag-ALD YSZ 15 cathodes and bare Pt electrodes) at 0.5 V and 450 C. The result conﬁrmed the excellent thermal and chemical stability of the Ag electrodes treated with ALD YSZ capping, which showed little variation over 4 h. In addition, although the Ag-ALD YSZ 7 sample had a current density comparable to that of bare Pt at ﬁrst, it showed an even higher current density than Pt after 4 h of operation. In contrast, the SOFC with a bare Ag cathode degraded and was almost dead within 1 h, after exhibiting abnormally high current. The SEM images provided in Fig. 7 clearly indicate that the Ag-ALD YSZ electrodes preserved the granular morphology after a long fuel cell operation with minor agglomeration of particles forming empty holes through the silver layers. However, the bare silver electrodes were scattered and completely disconnected after <1 h of operation at 450 C. Destruction of the bare silver layer is suspected to be due to the prolonged desorption of the surface silver oxide slabs [36e39]. We have discussed the initiation of fuel cell above 425 C with the bare silver electrode by reduction of the surface-passivating silver oxide. The increasing current appearing initially during the long-term test of the bare silver cell seems to arise from this reduction process, which involves electrons and oxide-ion ﬂux from the silver oxide. In comparison, the silver oxide seems to have already been removed from the Ag-ALD YSZ electrodes before deposition of ALD YSZ in the vacuum environment during the ALD process, which may have resulted in the signiﬁcantly improved stability. We investigated the chemical binding energies of bare Ag and ALD YSZ-treated Ag cathodes using XPS before and after the longterm fuel cell operation at 450 C. In the XPS analysis, as shown in Fig. 8, Ag3d3/2 and Ag3d5/2 peaks were identiﬁed at 368.2 eV and 374.2 eV, respectively. As a result, there was a negative shift of the
Fig. 7. SEM images of the bare Ag cathode (a) before and (b) after the fuel cell potentiostatic measurement at 0.5 V, 450 C for less than 1 h; the Ag-ALD YSZ 7 cathode (c) before and (d) after the measurement at 0.5 V, 450 C for over 4 h; the Ag-ALD YSZ cathode 15 (e) before and (f) after the measurement at 0.5 V, 450 C for over 4 h.
Y.K. Li et al. / Journal of Power Sources 295 (2015) 175e181
Acknowledgements This research was supported by the International Research & Development Program of the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (Grant No. NRF-2014K1A3A1A47067412). The Brain Korea 21 Plus program (21A20131712520) is also acknowledged for their support. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jpowsour.2015.06.136. References Fig. 8. XPS spectra measured on the surface of bare Ag, Ag-ALD YSZ 7, and Ag-ALD YSZ 15 cathodes before and after fuel cell potentiostatic measurement at 0.5 V, 450 C.
3d peaks after operation, indicating a reduction of metal oxides. In contrast, the peak shift of the Ag-ALD YSZ samples was less signiﬁcant. It should be noted that the Ag3d peaks of the Ag-ALD YSZ 15 cathode appeared to be more positive that those of the Ag-ALD YSZ 7 cathode both before and after operation, implying that there are more silver-oxide bonds on the surface for higher ALD YSZ content [51,52].
4. Conclusions In this work, we investigated the use of porous silver cathodes that were surface-treated with ALD YSZ for high-performance SOFCs in the IT regime. The samples were fabricated on GDC electrolyte pellets and tested as fuel cells with porous Pt anodes. ALD YSZ was conducted for 7 and 15 cycles on porous sputtered Ag, and, for comparison, bare porous Ag and Pt cathodes were also prepared by sputtering. The overall thickness of the electrodes was 150 nm, and the performance evaluation was conducted at 250e450 C. As a result, the Ag with optimized ALD YSZ surface treatment, i.e., the Ag-ALD YSZ 7 sample, produced a peak power of 14.7 mW cm2 at 450 C, which is similar to the performance of the porous Pt cathode, the best performing low-temperature catalyst, which produced a peak power of 16.5 mW cm2 in our test. At all test temperatures, the performance difference between the sample with the Ag-ALD YSZ 7 cathode and those with the porous Pt cathode was 1.8e13.5% in terms of power. EIS conﬁrmed that this enhanced performance was due to improved electrode kinetics, presumably the increase in ORReTPB sites between the surface of the supporting silver and the ALD YSZ particulates. The potentiostatic current measurement conﬁrmed that the Ag-ALD YSZ cathodes are thermally and chemically stable for a long period of time, 4 h in this test, while the bare silver cathodes degraded to almost being dead within 1 h. The XPS data also conﬁrmed that the ALD YSZ capping prevented reduction of the surface silver oxide and destruction of the mesh morphology.
5. Glossary Ag, Silver; Pt, Platinum; YSZ, Yttria-stabilized zirconia; GDC, Gadolinia-doped ceria; SOFCs, Solid oxide fuel cells; IT, Intermediate temperature; ALD, Atomic layer deposition; CVD, Chemical vapor deposition; OCV, Open circuit voltage; ASR, Area-speciﬁc resistance; TPB, Triple-phase-boundary; SEM, Scanning electron microscope; EIS, Electrochemical impedance spectroscopy; AC, Alternating current; DC, Direct current; ORR, Oxygen reduction reaction; XPS, X-ray photoelectron spectroscopy.
 J.H. Shim, J.S. Park, J. An, T.M. Gur, S. Kang, F.B. Prinz, Chem. Mater. 21 (2009) 3290e3296.  H.S. Noh, J.W. Son, H. Lee, H.S. Song, H.W. Lee, J.H. Lee, J. Electrochem. Soc. 156 (2009) B1484eB1490.  S. Ha, P.C. Su, S.W. Cha, J. Mater. Chem. A 1 (2013) 9645e9649.  J.H. Shim, S. Kang, S.-W. Cha, W. Lee, Y.B. Kim, J.S. Park, T.M. Gür, F.B. Prinz, C.C. Chao, J. An, J. Mater. Chem. A 1 (2013) 12695e12705.  K. Bae, D.Y. Jang, H.J. Jung, J.W. Kim, J.-W. Son, J.H. Shim, J. Power Sources 248 (2014) 1163e1169.  C. Xia, M. Liu, Adv. Mater. 14 (2002) 521.  Z. Shao, S.M. Haile, Nature 431 (2004) 170e173.  Q. Zhou, Y. Shi, T. Wei, Z. Li, D. An, J. Hu, W. Zhao, W. Zhang, Z. Ji, J. Wang, Int. J. Hydrogen Energy 39 (2014) 10710e10717.  F.S. Baumann, J. Fleig, M. Konuma, U. Starke, H.-U. Habermeier, J. Maier, J. Electrochem. Soc. 152 (2005) A2074eA2079.  W.-H. Kim, H.-S. Song, J. Moon, H.-W. Lee, Solid State Ion. 177 (2006) 3211e3216.  S.P. Jiang, J. Mater. Sci. 43 (2008) 6799e6833.  W. Wang, S.P. Jiang, Solid State Ion. 177 (2006) 1361e1369.  M. Kubicek, Z. Cai, W. Ma, B. Yildiz, H. Hutter, J. Fleig, ACS Nano 7 (2013) 3276e3286. e, J. Electrochem. Soc. 155 (2008) B730eB737.  C. Peters, A. Weber, E. Ivers-Tiffe  B. Wei, Z. Lü, S. Li, Y. Liu, K. Liu, W. Su, Electrochem. Solid State Lett. 8 (2005) A428eA431.  W. Zhou, R. Ran, Z. Shao, J. Power Sources 192 (2009) 231e246.  C. Xia, W. Rauch, F. Chen, M. Liu, Solid State Ion. 149 (2002) 11e19.  H. Fukunaga, M. Koyama, N. Takahashi, C. Wen, K. Yamada, Solid State Ion. 132 (2000) 279e285.  S.B. Adler, Chem. Rev. 104 (2004) 4791e4844.  S. Wang, T. Kato, S. Nagata, T. Honda, T. Kaneko, N. Iwashita, M. Dokiya, Solid State Ion. 146 (2002) 203e210.  M.K. Debe, Nature 486 (2012) 43e51.  T. Ralph, M. Hogarth, Platin. Met. Rev. 46 (2002) 117e135. , A. Neitzel, T. Ska la, N. Tsud, M. Vorokhta,  A. Bruix, Y. Lykhach, I. Matolínova , J. Myslive re, V. Potin, V. Stetsovych, K. Sev cíkova cek, K.C. Prince, S. Bruye F. Illas, V. Matolín, J. Libuda, K.M. Neyman, Angew. Chem. Int. Ed. 53 (2014) 10525e10530.  www.platinumprice.org, (Last accessed on 20.02.2015).  J. Van Herle, A. McEvoy, J. Phys. Chem. Solids 55 (1994) 339e347.  T. Kenjo, H. Takiyama, Electrochim. Acta 39 (1994) 2685e2692.  R. Liu, X. Yu, G. Zhang, S. Zhang, H. Cao, A. Dolbecq, P. Mialane, B. Keita, L. Zhi, J. Mater. Chem. A 1 (2013) 11961e11969.  www.silverprice.org, (Last accessed on 20.02.2015).  T.-J. Huang, X.-D. Shen, C.-L. Chou, J. Power Sources 187 (2009) 348e355. € dickemeier, K. Sasaki, L. Gauckler, I. Riess, J. Electrochem. Soc. 144  M. Go (1997) 1635e1646.  S.M. George, Chem. Rev. 110 (2009) 111e131. € , Adv. Mater. 19 (2007) 3425e3438.  M. Knez, K. Nielsch, L. Niinisto  K.S. Son, K. Bae, J.W. Kim, J.S. Ha, J.H. Shim, J. Vac. Sci. Technol. A 31 (2013) 01A107.  D.Y. Jang, H.K. Kim, J.W. Kim, K. Bae, M.V.F. Schlupp, S.W. Park, M. Prestat, J.H. Shim, J. Power Sources 274 (2015) 611e618.  H.J. Jeong, J.W. Kim, K. Bae, H. Jung, J.H. Shim, ACS Catal. 5 (2015) 1914e1921.  M. Bowker, Surf. Sci. 155 (1985) L276eL280.  L. Tjeng, M. Meinders, J. Van Elp, J. Ghijsen, G. Sawatzky, R. Johnson, Phys. Rev. B 41 (1990) 3190.  J.H. Shim, X. Jiang, S.F. Bent, F.B. Prinz, J. Electrochem. Soc. 157 (2010) B793eB797.  J.H. Shim, Y.B. Kim, J.S. Park, J. An, T.M. Gür, F.B. Prinz, J. Electrochem. Soc. 159 (2012) B541.  J. Larminie, A. Dicks, M.S. McDonald, Fuel Cell Systems Explained, Wiley, New York, 2003.  W. Zhu, S. Deevi, Mater. Sci. Eng. A 362 (2003) 228e239.  H. Jung, K. Bae, D.Y. Jang, Y.H. Lee, S.W. Cha, J.H. Shim, Int. J. Hydrogen Energy
Y.K. Li et al. / Journal of Power Sources 295 (2015) 175e181 39 (2014) 17828e17835.  S. de Souza, S.J. Visco, L.C. De Jonghe, Solid State Ion. 98 (1997) 57e61.  R. O'Hayre, S.-W. Cha, W. Colella, F.B. Prinz, Fuel Cell Fundamentals, John Wiley & Sons, New York, 2005, p. 409.  B.C.H. Steele, A. Heinzel, Nature 414 (2001) 345e352.  H. Huang, M. Nakamura, P. Su, R. Fasching, Y. Saito, F.B. Prinz, J. Electrochem. Soc. 154 (2007) B20eB24.  Y.B. Kim, J.H. Shim, T.M. Gür, F.B. Prinz, J. Electrochem. Soc. 158 (2011) 021515.
 M. Mosiałek, P. Nowak, M. Dudek, G. Mordarski, Electrochim. Acta 120 (2014) 248e257.  R. Radhakrishnan, A.V. Virkar, S.C. Singhal, J. Electrochem. Soc. 152 (2005) A210eA218.  Y. Lin, R. Ran, Z. Shao, Int. J. Hydrogen Energy 35 (2010) 8281e8288.  F.K. Bock, T.M. Christensen, S.B. Rivers, L.D. Doucette, R.J. Lad, Thin Solid Films 468 (2004) 57e64.  M.F. Al-Kuhaili, J. Phys. D. Appl. Phys. 40 (2007) 2847e2853.