1 Introduction

Fuel cells have attracted much attention for next-generation energy conversion system because of their high energy conversion efficiency, environmentally friendly, quiet operation and wide range of scalability. Recently, among various types of fuel cells, low temperature solid oxide fuel cells (LT-SOFCs), which operate at the relatively low temperature regime (300–500 °C) compared to conventional SOFCs (800–1000 °C), have been intensively studied since they have advantage of relatively low cost of system, long-term stability and fast start-up characteristic, which enable this system to extend its applications to residential application and auxiliary power unit of mobile application [1,2,3,4,5,6]. Despite of these potentials, however, achieving and maintaining high performance at the low temperature regime is still challenging issue because the ionic conductivity of solid oxide electrolyte is significantly reduced and the electrochemical reaction at the both anode and cathode side is sluggish as the operating temperature is reduced [7,8,9,10,11,12,13,14]. To overcome these issues and obtain high performance at the low operating temperature, the electrolyte thickness should be minimized to reduce the ohmic resistance and electrochemical activity at the electrode/electrolyte interface should be enhanced to reduce the polarization resistance.

In the past decades, with advances in thin-film fabrication technology such as physical vapor deposition (PVD) and atomic layer deposition (ALD), many researchers have succeeded in developing thin-film fuel cells, which have a very thin electrolyte layer and significantly reduced ohmic resistance, thus resulting in a high fuel cell performance [15,16,17,18,19,20,21]. For example, Huang et al. demonstrated that thin-film fuel cells with 100 nm-thick yttria-stabilized zirconia (YSZ)/GDC electrolyte deposited by sputtering method showed greatly reduced ohmic resistance and high fuel cell performance at the low operating temperature [17]. Further, Shim et al. reported that the thin-film fuel cells with only 60 nm-thick YSZ electrolyte, fabricated by using atomic layer deposition also showed reduced ohmic resistance and high cell performance [19].

Although the ohmic resistance could be effectively reduced by using a thin-film electrolyte structure, however, another issue of LT-SOFCs is the high polarization resistance at the electrode/electrolyte interface as mentioned above still remains. As the operating temperature is reduced, the charge transfer kinetics of conventional ceramic electrodes such as La1−xSrxMnO3 (LSM), La1−xSrxCoO3 (LSC) and La1−xSrxCo1−yFeyO3 (LSCF) are significantly slowed down leading to increased activation loss which significantly reduces the fuel cell performance [22]. For this reason, noble metal platinum (Pt) catalyst has been usually employed as catalytic electrode to obtain high electrochemical activity and to achieve high cell performance at the low operating temperature regime [23, 24]. However, extremely high material cost of platinum significantly increases the cost of entire membrane-electrode-assembly (MEA) and hinders commercialization of LT-SOFCs. Accordingly, it can be said that minimizing Pt catalyst loading while maintaining its exceptional catalytic activity is a crucial point in the developing LT-SOFCs.

In this regard, many researchers typically employed direct current (DC) magnetron sputtering method to fabricate a thin-film porous Pt catalytic electrode [23, 24]. and even the atomic layer deposition (ALD) has been recently demonstrated to fabricate ultra-thin Pt catalyst to reduce the Pt loading [25]. Although the thickness of Pt electrode could be reduced, further complementation for the design of catalytic electrode is still required. In other words, the Pt loading of sputter-deposited Pt films are still too high and the ALD-deposited films have concerns about sheet resistance and thermo-mechanical vulnerability due to an extremely low film thickness. Therefore, it can be considered that the optimal catalytic electrode design for the LT-SOFCs has appropriate film thickness, which is enough to have thermo-mechanical strength and electrical conductivity, and low Pt loading for the material cost reduction.

Thus, employing bi-layer structure which is composed of ultra-thin Pt catalytic active layer and electrical conducting layer using cheaper metal such as nickel, copper or silver can be a strategy to fabricate a thin-film electrode with high catalytic activity and reduced material cost. Among various candidate materials for electrical conducting layer, nickel (Ni) can be a promising candidate for its relatively high electrical conductivity, high melting point (~ 1455 °C and most of all, very cheap material cost (~ US$0.011/g) for abundancy. In addition to cost-effective perspective, another positive aspect of Ni/Pt bilayer structure is a microstructural stability. Because the electrochemical reaction occurs at the triple phase boundaries (TPBs) where the three phases (i.e., gas phase, electrolyte, and electrode), are in physical contact, the porosity and morphological stability of the metal electrodes is an important parameter to ensure the durability of fuel cell performance. However, when the Ni single-phase electrode is used as an electrode of LT-SOFCs, the Ni tends to severely agglomerate and causes rapid cell performance deterioration under operating conditions due to its high surface energy [26,27,28]. Therefore, the insertion of other metal layer, which has a relatively lower surface energy than Ni, between Ni and electrolyte can be an effective way to achieve microstructural stability in the Ni-based metal electrodes. However, despite these aspects, many previous researches on Ni/Pt bilayer electrode focused only on its material properties and applications in the proton-exchange membrane fuel cell (PEMFC) operating conditions [29,30,31,32]. In particular, to the best of our knowledge, there is no previous study demonstrating the electrochemical characteristics of Ni/Pt bilayer electrode as an anode for the LT-SOFCs.

In this study, we fabricated a Ni/Pt bilayer catalytic electrode via the sputtering method and evaluated its electrochemical activity as an anode for the LT-SOFCs. We further optimized the Pt loading amount in the Ni/Pt bilayer electrode and demonstrated that the optimized electrode structure has a cell performance similar to that of a Pt electrode but with significantly reduced Pt loading. We believe that the results presented herein may help to lower the membrane electrode assembly (MEA) fabrication cost for LT-SOFCs and facilitate the commercialization of LT-SOFCs.

2 Experimental

For the fabrication of Ni/Pt bilayer anode based fuel cells, the Pt catalytic layers were deposited with various thickness of 5, 10 and 20 nm on one side of polycrystalline YSZ substrate (1 cm × 1 cm, 500 μm thickness, MTI Corporation, USA) via DC sputtering using a commercial sputtering system (Daeki High-Tech Co., Ltd., Daejeon, Korea). Then, the Ni electric conducting layers were deposited on the deposited Pt thin film with thickness of 145, 140 and 130 nm respectively to make the total thickness constant to 150 nm. The total thickness of anode was determined to eliminate any sheet resistance effect [33]. A 150 nm-thick Pt anode was also prepared to compare the performance and Pt loading with the Ni/Pt bilayer anode. Both Ni and Pt thin films employed in this study were deposited under same working pressure of 75 mTorr with 30 sccm of Ar flow rate and a DC sputtering power of 100 W (0.25 A). The deposition rate of Ni and Pt thin films was confirmed to be 5.43 nm/min and 25.94 nm/min, by observing cross-sectional images of Field emission scanning electron microscopy (FE-SEM) analysis, and the results were used for the accurate thickness control of deposited films. During all the experiments, the target-to substrate distance was kept at 25 cm and the sample holder was rotated to ensure uniform film deposition and to prevent film growth in a specific direction. The high-purity Ni and Pt metal targets with a same diameter of 25.4 mm were used for both Ni and Pt sputtering respectively and the thickness of sputtering target was 2 mm for Pt target and 1 mm for Ni target due to magnetism characteristic of Ni metal. A 80 nm layer of porous platinum cathode was then deposited via DC sputtering on the opposite side of YSZ substrate under 75 mTorr of working pressure and 30 sccm of Ar gas flow with DC sputtering power of 100 W (0.25 A). The effective active cathode area was 1 × 1 mm2. The schematics of fabricated reference Pt and Ni/Pt bilayer anode SOFCs are shown in Fig. 1.

Fig. 1
figure 1

Schematics of reference Pt and Ni/Pt bilayer anode SOFCs: a Pt anode and b Ni/Pt bilayer anode

Full size image

The surface and interface morphologies of the “as-deposited” and “after-tested” anodes were examined via scanning electron microscopy (SEM) (Nova NanoSEM 450, FEI Corp.). To directly observe the Pt/YSZ interface morphology of the Ni/Pt bilayer anodes, pull-off delaminating process was also conducted. To prepare the specimens of the delaminating process, a 10 mm × 10 mm × 2 mm size of aluminum substrate was treated by O2 plasma using commercial plasma generation system (CUTE-MP, Femto Science, Inc., Korea) and coated with an epoxy adhesive (Epo-Tek® MA-5, Epoxy Technology, USA). The aluminum substrate was B-stage cured in a convection oven at 75 °C for 45 min and attached with top of the Ni/Pt bilayer anode. Another aluminum substrate with the same size was also coated with the other epoxy adhesive (DP420, 3M Corporation, USA) and attached with the other side of the specimen. After the specimen was cured in 120 °C for an hour, tabs were applied to both sides of the specimens by using the DP420 epoxy adhesive and the specimens were finally pull-offed by 10 μm s−1 using high precision micromechanical test system (Delaminator Adhesion Test system, DTS Company, USA).

For the fuel cell performance measurement and electrochemical analysis, each sample was mounted on a customized cell test station containing a temperature-controlled heating stage as described in our previous reports [33]. The current–voltage (I–V) behaviors were measured at temperatures of 350 °C, 400 °C and 450 °C with 20 sccm of H2 gas using a Gamry linear sweep voltameter (Gamry Instruments Inc., FAS2, USA). Electrochemical impedance spectroscopy (EIS) was conducted using a Gamry potentiostat under a frequency range from 1 MHz to 1 Hz with bias voltage conditions of 0.6 V and OCV.

3 Results and Discussion

Prior to detailed analysis, we first calculated the effect of Ni/Pt bilayer structure on the reduction of Pt loading amount for the anodes. Table 1 shows the calculated Pt loading amount of the 150 nm-thick Ni/Pt bilayer anode with various Pt thickness of 0 nm to 150 nm. The film porosity for the calculation was assumed to be 22.4% from the experimental result by converting SEM micrographs to binary images using Image J software.

Table 1 Calculation of Pt loading amount of the fabricated Ni/Pt bilayer anode with different Pt thickness of 0 nm, 5 nm, 10 nm, 20 nm and 150 nm
Full size table

As shown in Table 1, the calculated Pt loading amounts of each anodes were 0, 0.008, 0.00 [17], 0.033 and 0.250 mg/cm2 for the Ni150, Pt5/Ni145, Pt10/Ni140 Pt20/Ni130 and Pt150 anode respectively. It can be seen that the Pt150 anode that is composed of a Pt single phase has a high Pt loading of 0.129 mg/cm2, whereas the Ni/Pt bilayer anodes have a significantly reduced Pt loading. Especially, in the case of Ni150 and Pt5/Ni145 anodes, the entire Pt loading can be reduced to 0 and only 0.008 mg/cm2, which is significantly lower than the 2020 target value of the United States Department of Energy (DOE) for PEMFCs [34] and also lower that values reported in other previous studies using Pt catalyst for the SOFC electrodes [24, 25, 35].

Field emission scanning electron microscopy (FE-SEM) top view images of the deposited Ni/Pt bilayer anodes with different Pt thickness are shown in Fig. 2. All deposited films had very small size of Ni grain and showed nanoporous morphology which was sufficient to ensure penetration of the hydrogen fuels into the YSZ/Pt interfaces. This was also confirmed in the following current–voltage (I–V) and EIS measurements; any concentration losses were not observed in the I–V curves and EIS results. Meanwhile, there were no noticeable differences on the microstructures between deposited films because they had similar thickness of Ni layer on the top side of anode layer.

Fig. 2
figure 2

FE-SEM top-view images of the deposited Ni/Pt bilayer anode on YSZ substrate: a Ni150, b Pt5/Ni145, c Pt10/Ni140 and d Pt20/Ni130 anode

Full size image

The current–voltage (I–V) behaviors of fabricated fuel cells with different Ni/Pt bilayer anodes were measured at 450 °C and shown in Fig. 3a. All samples, except for the Pt20Ni130 sample, had an open circuit voltage (OCV) values in the range of 0.98–1.01 V which were close to the theoretical values. While the fuel cell with Pt0/Ni150 anode showed a maximum power density of 4.04 mW/cm2 at an operating temperature of 450 °C, the maximum power density values increased to 4.68 mW/cm2 as the Pt catalytic layer thickness increased by 10 nm. This is considered that the electrochemical reactivity on the anode side is improved by applying the Pt catalytic layer as we firstly designed, since Pt has higher catalytic activity for hydrogen oxidation reaction (HOR) than Ni [36].

Fig. 3
figure 3

a Current–voltage (I–V) curves of the fabricated fuel cells with Ni/Pt bilayer anode measured at 450 and b maximum powder density comparison of the samples measured at 350, 400, and 450

Full size image

In addition to this, the tendency of improving performance with the increasing thickness or amount of Pt catalytic layer can be explained by surface coverage of Pt on the YSZ electrolyte. In a previous study, O’Hayre et al. demonstrated that the 5 nm-thick Pt thin film fabricated by sputtering process showed isolated domains and did not cover the whole surface area of substrate, while the 15 nm-thick Pt thin film could cover the whole surface and formed entire layer [37]. Therefore, the 10 nm-thick Pt catalytic layer employed in this study may have a larger surface coverage area than the 5 nm-thick Pt layer and thus, the fuel cell performance was further enhanced with increasing thickness of Pt layer. Meanwhile, the Pt20/Ni130 sample showed relatively low OCV value of 0.91 V and low maximum power density of 3.08 mW/cm2. It was an unexpected trend because we first speculated that the maximum power density would be enhanced compared to the Pt0 sample and saturated at a particular value as the thickness of Pt layer was further increased. This will be further discussed and demonstrated in the following sections. Figure 3b shows the maximum power density comparison of the samples measured at the different operating temperature of 350, 400 and 450 °C. Although the performance difference between the samples slightly decreased with reducing operating temperature, the maximum power density trend was constant in all operating temperature regions.

To further investigate the cause of the difference in the cell performances, electrochemical impedance spectra (EIS) measurements were conducted at 450 °C on all samples. Figure 4a shows a representative EIS spectra of the fuel cell with Pt10/Ni140 anode under the OCV condition and an applied cell voltage of 0.6 V. In the impedance spectra, two arcs are visible in each applied voltage condition. The first high-frequency arc shows no noticeable changes with variation in the cell voltages, whereas the low-frequency arc shows significant change as the cell voltage varies. This indicates that the first high-frequency arc was associated to the ohmic resistance originated from ionic conduction through the solid electrolyte and electronic charge transport through the electric circuit. In general, because the ionic conduction tends to be more difficult than the electronic charge transport, the influence of electronic contribution on the ohmic resistance is negligible [38]. On the other hand, the low-frequency arc, which changes with the applied voltage conditions, was associated with polarization resistance at the electrode/electrolyte interfaces [39].

Fig. 4
figure 4

a Representative EIS spectra of the fabricated fuel cell (Pt10/Ni140) under different DC bias voltages of OCV and 0.6 V; b EIS spectra of the fuel cells with various Ni/Pt bilayer anode under a DC bias voltage of 0.6 V. All EIS measurements were conducted in the same temperature condition of 450 °C; c summarized EIS results based on the results of b; d Calculated capacitances of Ni/Pt bilayer anodes for the polarization resistance

Full size image

Figure 4b shows the Nyquist plot of the fuel cells with various Ni/Pt bilayer anode under same DC bias voltage of 0.6 V. The Nyquist plots of Ni150, Pt5/Ni145 and Pt10/Ni140 anode cells is fitted to the equivalent circuit model with a series resistance (R) and two parallel resistance-constant phase element (CPE) pairs as described in the inset of Fig. 4b. The Pt20/Ni130 anode cell was fitted with one series R and three R-CPE pairs. It can be clearly seen that the ohmic resistance of all samples is similar, while the polarization resistances of all samples are varied. The similar ohmic resistances can be explained by dominant ionic conduction contribution in the ohmic resistance as mentioned above. Because all fuel cell samples were made using the commercial YSZ substrates having same thickness, it is reasonable that they have similar ionic conducting behavior and ohmic resistance. On the other hand, the obvious difference in the polarization resistances between each sample means that the differences in the cell performance are mainly associated with the interfacial polarization resistance rather than ohmic resistance. Especially, because we made the conditions of Pt cathode of all samples constant by fabricating in the same batch process, it was reasonable that the difference in the polarization resistance originated from the anode variations.

The calculated polarization resistances from the fitted Nyquist plot were 14.94 Ω cm2, 13.02 Ω cm2, 10.79 Ω cm2 and 32.37 Ωcm2 for Ni150, Pt5/Ni145, Pt10/Ni140 and Pt20/Ni130 respectively. Consequentially, by applying a very thin 10 nm-thick Pt catalytic active layer, the polarization resistance of the Ni150 anode was effectively reduced by about 30%. As mentioned above, this is because the superior catalytic activity of the Pt catalyst for the HOR enhanced the electrochemical reaction at the anode/electrolyte interface. Also, it was remarkable that the optimal ratio of Pt/Ni bilayer structure was revealed to be Pt10/Ni140. At first, we speculated that the reduction of polarization resistance would be saturated at a certain point of Pt thickness due to complete coverage of Pt catalyst at the certain thickness. However, as can be seen in Fig. 4c, the tendency for the polarization resistance decreased with increasing Pt layer thickness until 10 nm as expected, but the Pt 20/Ni130 anode cell showed an unexpectedly high polarization resistance value. Moreover, an additional semicircle for the polarization resistance was observed on the Nyquist plot (Fig. 4b).

To further understand the reason for the additional circle and high polarization resistance value of Pt20/Ni130 anode, the capacitance values were calculated from the fitted EIS data using the formula:

$${\text{C}} = (R^{{1 - {\text{n}}}} {\text{Q}})^{1/n} ,$$

where C is the capacitance, R is the polarization resistance, Q is the pseudocapacitance value and n is the exponent value extracted from the fitted EIS data. The calculated capacitances are summarized in Fig. 4d. As each parallel R-CPE corresponded to one of the processes occurring in the complex electrode reactions, the analyzation of calculated capacitances helped to identify the corresponding process. In general, the hydrogen oxidation reaction at the metal electrode/electrolyte interface was mostly influenced by two major electrode process in series: the charge-transfer reaction at the electrode/electrolyte interface and hydrogen dissociative adsorption/diffusion on metal surface [40, 41]. In this regard, the capacitance values of all samples for the first semicircle were in the range of 2–5 μF cm−2, which was in the same order of magnitude for a typical double-layer capacitance related to the charge-transfer reaction at the electrode/electrolyte interface [36, 43]. On the other hand, the capacitance value of the Pt20/Ni130 anode cell for the second semicircle was calculated to 0.2 mF cm−2. According to previous studies, the adsorption of a dissociative hydrogen on a Ni surface would give a capacitance in the order of ~ 10 mF cm−2 [43, 44]. Although our calculated capacitance value is lower than the reported value, it is acceptable considering the relatively low film thickness of the anode fabricated by the thin-film fabrication process. Further, it was known that the dissociative adsorption/diffusion reactions of hydrogen occur more favorably on Ni than on Pt via a spillover mechanism which furthermore frees up metal sites for dissociative adsorption of hydrogen. This is because the Ni has a sufficiently lower Gibbs energy for the formation of adsorbed oxygen species on the surface to generate spillover mechanism than Pt electrode [42, 43]. From this aspect, we speculated that the additional polarization semicircle and high polarization resistance may be attributed to the difference in the reactivity for the hydrogen dissociate adsorption/diffusion process between Ni and Pt layer at the Ni/Pt interface once they make complete layer-by-layer structure. This trend was also confirmed in our further experimentation about Pt30/Ni120 and Pt75/Ni75 bilayer anode structure.

Thin-film porous metal electrodes usually tend to easily agglomerate even under LT-SOFC operating conditions and cause rapid cell performance deterioration. Therefore, the morphological stability of metal electrodes is an important parameter. Figure 5a–d shows FE-SEM top-view surface morphologies of the Ni/Pt bilayer anode with various thickness of Pt catalytic layer after fuel cell operation. The Ni150 anode showed severe agglomeration and significantly reduced porosity after fuel cell operation at elevated temperatures. It is well known that porous Ni film tends to easily agglomerate at elevated temperature conditions compared to other metallic electrode material due to its higher surface energy [26,27,28]. Therefore, it can be expected that the morphological instability of porous Ni anode causes dramatic cell performance degradation. On the other hand, the Ni/Pt bilayer anode showed very rough surface with flake-shape morphology (Fig. 5b–d). The difference in the surface morphology between Ni and Ni/Pt anode can be expected to influence of the different surface energy caused by the inserted Pt layer between the Ni and YSZ electrolyte. Although the morphological information of the top Ni layer can be obtained from the FE-SEM images, the morphology of the Pt/YSZ interface where the actual electrochemical reaction occurs cannot be observed. Thus, we conducted pull-off delaminating process after the fuel cell operation to directly observe the interface morphology between the Pt catalytic layer and YSZ electrolyte. In a previous study, Yu et al. demonstrated that a delamination method based on double cantilever beam (DCB) delamination technique can be employed to observe the Pt/YSZ interface of a single-phase Pt electrode [44]. The pull-off delaminating process employed in this study was similar to the DCB process, but was a more convenient process for the observation purposes. Figure 5e–h shows the interface morphology of the Ni and Ni/Pt bilayer anode near the YSZ electrolyte. It can be clearly seen that the interface morphology of the Ni150 anode (Fig. 5e) is very similar with the top-view image (Fig. 5a), and the interface morphology of the Ni/Pt bilayer anodes shows relatively porous morphology which cannot be visible from the top-view FE-SEM images. Although the porosity of the Ni/Pt bilayer anodes was reduced compared to the porosity before the operating test, they showed a relatively porous microstructure than the Ni single-phase anode. This relative microstructural stability is a result of reducing the high surface energy of Ni by inserting thin Pt catalytic active layer between the Ni and YSZ electrolyte. Interestingly, the only 5 nm-thick Pt layer inserted between Ni and YSZ greatly affects to the surface energy difference and their agglomeration mitigation.

Fig. 5
figure 5

FE-SEM images of the Ni/Pt bilayer anode with different Pt catalytic layer thickness of a, e 0 nm, b, f 5 nm, c, g 10 nm, and d, h 20 nm after fuel cell operating test; a–d viewed from the top, and e–h viewed from Pt/YSZ interfacial perspectives processed by pull-off delaminating process

Full size image

To compare the fuel cell performance of the optimized Pt10/Ni140 anode with reference single-phase Pt anode, the I–V behavior of the fuel cell with Pt150 anode was measured and compared with that of the Pt10/Ni140 anode cell. Figure 6a shows the measured I–V curves of the fuel cells with Pt10/Ni140 and Pt150 anode. Interestingly, the Pt150 anode and the Pt10/Ni140 anode cell showed almost same maximum power density value. The measured maximum power density values were 4.70 mW/cm2 and 4.68 mW/cm2 for the Pt150 and Pt10/Ni140 anode respectively. The fuel cell performance in terms of the peak power density value may look small, however, it is because of the commercially available thick electrolyte (500 μm) and the fabrication of high performance fuel cell is not the main objective of this study. This result implied that almost the same catalytic activity of Pt150 anode and corresponding actual Pt/YSZ TPBs can be implemented with only 10 nm-thick Pt catalytic layer of the Pt10/Ni140 bilayer anode and the Ni electrical conducting layer exhibits an anode performance comparable to that of the single-phase Pt anode as we first expected. Moreover, it was also noteworthy that the employed Pt loading in the developed Pt10/Ni140 bilayer anode was only about 0.009 mg/cm2, which was only 7% of the Pt150 anode conventionally fabricated with the Pt single phase. The measured EIS spectra of both Pt10/Ni140 and Pt150 anode cell also well supported the tendency for the I–V behaviors of the fuel cells; both ohmic resistance and polarization resistance of cells were very similar. This means that the electrical characteristic and electrochemical reactivity of both Pt10/Ni140 and Pt150 anode are similar as mentioned above. The results of the study can provide significant implication on designing catalytic metal electrode structures for less amount of precious expensive metals not only for fuel cells but also various other energy conversion systems.

Fig. 6
figure 6

a Current–voltage (I–V) curves of the fuel cells with Pt10Ni140 bilayer anode and Pt150 anode, and b EIS spectra of the cells under a bias voltage of 0.6 V. Both the I–V curves and EIS spectra are measured at 450 °C

Full size image

4 Conclusion

In this study, we fabricated the Ni/Pt bilayer catalytic electrode by using the DC sputtering technique and evaluated its performance as an anode for the LT-SOFCs. The thickness of the Pt catalytic active layer was varied and optimized to enhance the cell performance and reduce the Pt loading. Among various Pt/Ni bilayer anodes, the Pt10/Ni140 bilayer anode showed the highest cell performance which is attributed to the superior catalytic activity of Pt and sufficient coverage on the YSZ electrolyte. EIS analyses also indicated that the ohmic resistance of all samples were similar while the polarization resistance of the Pt10/Ni140 anode cell showed the lowest value. The morphological analysis employing pull-off delamination process effectively showed the interface morphology of the bilayer anode at the Pt/YSZ interface. From the interface analysis, it can be observed that the Ni/Pt bilayer anode showed a more stable microstructure in the fuel cell operating condition than the Ni single-phase anode did, owing to the difference in the free energy of Pt and Ni. Meanwhile, the optimized Pt10/Ni140 anode having only 7% of Pt loading in the conventional single-phase Pt150 anode exhibits very similar cell performance and polarization resistance with the Pt150 anode. The Ni/Pt bilayer catalytic electrode demonstrated herein may help to reduce the fabrication cost of the LT-SOFCs and facilitate the commercialization of the systems.