Micro-Tubular Solid Oxide Fuel Cell Polarization and Impedance Variation With Thin Porous Samarium-Doped Ceria and Gadolinium-Doped Ceria Buffer Layer Thickness

Solid oxide fuel cells (SOFCs) have been investigated for many years as an alternative energy conversion device due to their potential for high efficiency, low emissions, quiet operation, and fuel flexibility [1–4]. SOFCs are made of ceramic materials with two electrodes having mixed ionic and electronic conductivity surrounding an electrolyte layer. A thick yttria-stabilized zirconia (YSZ) electrolyte layer has been investigated at high temperatures (900–1000 °C) for many years, and a thinner (∼5–50 μm) YSZ electrolyte has achieved reduced ohmic resistance and improved performance at intermediate temperatures (600–800 °C) [5,6]. An anode-supported SOFC (AS-SOFC) is typically used for mechanical strength with the thin YSZ electrolyte and has achieved improved thermal shock resistance and performance compared with the thick electrolyte [7].

Operating at intermediate temperatures helps alleviate some of the current challenges in SOFC research including long startup time and corrosion and breakdown of components at high temperatures [8]. Unfortunately, Sr-doped LaMnO₃ (LSM) has low electrocatalytic activity and oxygen permeability at intermediate temperatures. Perovskite cathode materials such as La_0.60Sr_0.40Co_0.20Fe_0.80O_3-δ (LSCF) [9–11] and Ba_0.50Sr_0.50Co_0.80Fe_0.20O_3-δ (BSCF) [6,12–16] have superior electrocatalyic activity and oxygen permeability at these temperatures. However, these cathode materials are not directly compatible with a YSZ electrolyte because of interfacial reactions which create different materials, like BaZrO₃, La₂Zr₂O₇, and SrZrO₃ [6,10,13,17,18]. These intermediate layers form at high temperatures during sintering and are non-conductive which increases the resistance and decreases the fuel cell performance. For the case of LSCF-YSZ, the reactivity occurs at temperatures above 1000 °C due to the segregation of Sr and La leading to a destabilized LSCF structure [19] and Zr⁴⁺ cation diffusion [17].

To prevent these detrimental reactions, an interlayer or buffer layer has been proposed and investigated in many studies [14,18,20–33]. This buffer layer is often a doped-ceria material, such as Sm_xCe_1-xO_2-δ (SDC) or Gd_xCe_1-xO_2-δ (GDC), which is chemically compatible with the perovskite cathode [34]. These materials help decrease the reactivity between the LSCF and BSCF cathode material and YSZ [19] and have also been cited for improved ionic conductivity and surface exchange kinetics of the doped ceria compared with YSZ [35]. Many studies have investigated the proper sintering temperature because high-temperature sintering is needed to create a dense buffer layer for improved Sr²⁺ retention [36]. Unfortunately, if the sintering temperature is too high (>1300 °C) then interfacial reactions between YSZ and SDC or GDC are possible, which form a resistive (Ce,Sm)₂Zr₂O₇ phase that reduces ionic conductivity [15]. While this phase is less conductive than YSZ, it is not considered to be significantly lower [19]. In another study with a YSZ–GDC interface, a nominal composition of Ce_0.37Zr_0.38Gd_0.18Y_0.07O_1.87 was identified at the interface which had much lower ionic conductivity [34]. Using a SDC or GDC electrolyte is one alternative to preventing reactions with YSZ, but reduction of Ce⁴⁺ to Ce³⁺ at high temperatures in a reducing atmosphere results in electronic conductivity of the electrolyte [34].

Table 1 provides a summary of the thickness of the porous and dense buffer layer used in many studies. As shown in the table, the porous buffer layer is typically >2 μm thick (with most porous buffer layers ∼5 μm thick), while the dense buffer layer is typically <1 μm thick. While many studies utilizing different thicknesses of the porous buffer layer have been conducted [37], few studies have systematically investigated different thicknesses to understand how the polarization, impedance, and electrochemical performance vary with the thickness of the buffer layer [15]. A few studies have found that the dense buffer layer can be much thinner (∼0.2–0.4 μm) than a porous buffer layer (∼4 μm) to prevent interdiffusion of cations and improve performance [24,36]. Furthermore, when the buffer layer is dense, the thinnest buffer layer has the lowest resistance because the dense buffer layer can prevent reactivity between the cathode and electrolyte and a thin layer reduces ionic resistance [19]. However, the change in resistance with the thickness of the porous buffer layer is more complicated due to the possibility of interdiffusion through the porous layer, while a thick buffer layer increases ionic resistance. A recent study on a planar SOFC investigated a porous SDC buffer layer with the thickness varying between 0.4 and 2.3 μm [15]. An optimal thickness was observed at ∼1.5 μm due to low ohmic, activation, and concentration resistance attributed to reduced interfacial reactions and improved triple phase boundary. This optimal thickness of the porous buffer layer is less than the thickness used in most studies, which warrants further investigation. Further work is needed to investigate this optimal buffer layer thickness and to consider different buffer layer materials.

In this paper, the porous buffer layer thickness is investigated with a micro-tubular SOFC (mT-SOFC) [1,2,42,43]. mT-SOFCs have advantages of improved thermal shock resistance, rapid startup, and thermal cycling compared with planar geometry [44–46]. An investigation of how the polarization, impedance, and power density vary when the buffer layer thickness changes is needed for mT-SOFCs. A SDC buffer layer is fabricated and compared with a GDC buffer layer that is sintered at the same temperature. The wet powder spray technique [6,15,47–49] is utilized to vary the buffer layer thickness between 0.3 and 2.0 μm. The mT-SOFCs are investigated with the current–voltage method, electrochemical impedance spectroscopy (EIS), scanning electron microscope (SEM), and energy-dispersive X-ray spectroscopy (EDS).

The anode-supported mT-SOFCs were fabricated using extrusion [50,51], dip coating [51], and wet powder spray techniques. The NiO (J.T. Baker) and YSZ (Tosoh) powders were mixed in a pug mill with water, pore former, steric acid, and polyethylene glycol for 2 h. The mixture was extruded using a ram extruder and dried for 48 h. The anode was pre-fired at 1100 °C for 2 h. The electrolyte was dip coated onto the anode in a slurry consisting of YSZ powder, polyvinyl butyral, phosphate ester, polyethylene glycol, and butanone that was ball milled for 12 h. The electrolyte was dried and sintered at 1400 °C for 4 h. The sintered YSZ electrolyte was ∼22 μm thick for all samples.

Sm_0.20Ce_0.80O_2-δ (fuel cell materials) and Gd_0.10Ce_0.90O_1.95 (fuel cell materials) slurries consisting of ethanol, ethylene glycol, glycerol, and the powder were ball milled for 24 h. A spray machine designed for depositing slurry on mT-SOFCs was utilized. Five different thicknesses of the SDC buffer layer were investigated with 15, 45, 75, 105, and 135 layers of SDC slurry sprayed. mT-SOFCs were also fabricated with a GDC buffer layer and the same number of layers of GDC slurry sprayed as the mT-SOFCs with the SDC buffer layer. The SDC and GDC buffer layers were sintered at 1350 °C for 4 h. Sintering temperatures above 1300 °C can result in YSZ/buffer layer interfacial reactions, as discussed in the introduction. However, in a previous study, sintering the SDC buffer layer at 1350 °C resulted in the optimal performance of the SOFC [6]. Despite the potential for interfacial reactions, this previous research found that sintering at 1350 °C resulted in improved densification of the buffer layer compared with a lower sintering temperature and prevented diffusion of cathode materials like Sr across the buffer layer. Sintering at higher temperatures (1400 °C) resulted in significant reactions between the YSZ and SDC or GDC buffer layer, which increased ohmic resistance and decreased power density. The sintered SDC buffer layer thickness was ∼0.4 μm, ∼0.9 μm, ∼1.3 μm, ∼1.7 μm, and ∼2.0 μm for the 15, 45, 75, 105, and 135 layers of SDC slurry sprayed. The sintered GDC buffer layer thickness was ∼0.3 μm, ∼1.0 μm, ∼1.3 μm, ∼1.6 μm, and ∼1.8 μm for the 15, 45, 75, 105, and 135 layers of GDC slurry sprayed. The sintered mT-SOFC had an internal diameter of 2.4 mm, an external diameter of 3.2 mm, and a cathode length of 13 mm.

Two cathode slurries were prepared with LSCF + SDC (7:3 w/w) and LSCF + GDC (7:3 w/w) powder (fuel cell materials) and a similar liquid composition as the electrolyte slurry. After ball milling for 2 h, the LSCF based cathode was dip coated [52] onto the buffer layer. The LSCF + SDC cathode was dip coated onto the mT-SOFCs with the SDC buffer layer, and the LSCF + GDC cathode was dip coated onto the mT-SOFCs with the GDC buffer layer. After drying, the cathodes of the mT-SOFCs with both types of materials were sintered at 1100 °C for 2 h.

The cathode of the mT-SOFC was coated in silver paste (an active area of 1 cm²). A silver wire current collector was wound around the cathode. Gold paste was applied to a bare section on the exterior of the anode and wrapped in silver wire in a previously reported configuration, which reduces the complexity of the current collector for the anode [53]. The current–voltage method with the four-probe technique [54] was conducted with a Keithley 2420 source meter interfaced with a computer. The mT-SOFC’s open circuit voltage (OCV), polarization, and power density were monitored during testing. EIS was conducted with a Solartron 1260A frequency response analyzer and with a Solartron 1287 potentiostat under OCV conditions over a frequency range of 10⁶–0.1 Hz with a signal amplitude of 10 mV. The mT-SOFC cross section was investigated with a SEM using the backscattered electron detector to provide compositional data. An EDS analysis was conducted with a line scan from the cathode to the electrolyte to analyze composition distribution across the buffer layer.

One end of the mT-SOFC was sealed on a tube with ceramic paste. The mT-SOFCs were tested in a tubular furnace that was heated at a rate of 5 °C/min up to 750 °C. The fuel cell temperature was monitored with a K-type thermocouple. Hydrogen at a flowrate of 50 ml/min was regulated with mass flow controllers and labview software. Air was supplied via natural convection in the furnace. A representative image of the experimental setup showing a detailed view of the mT-SOFC configuration is shown in Fig. 1.

The mT-SOFCs with the SDC buffer layer were investigated using the current–voltage method and EIS techniques. Figure 2 shows the polarization and power density curves for the mT-SOFCs tested at 750 °C with 15, 45, 75, 105, and 135 layers of SDC slurry sprayed. As shown, the mT-SOFC shows high polarization losses and low power density with 15 layers of SDC sprayed. The buffer layer thickness was only ∼0.4 μm in this case. As the SDC powder utilized in these experiments had a particle diameter of 0.1–0.4 μm, the final sintered buffer layer thickness was on the order of the original particle size, which could result in leakage of the cathode material through the buffer layer when it was dip coated. This can occur as SDC is known for limited densification at the sintering temperatures utilized in these experiments [6]. The peak power density of the mT-SOFC at 750 °C with 15 layers of SDC sprayed was 212 mW/cm². As the number of layers of SDC slurry sprayed increases, the thickness of the buffer layer increases and the mT-SOFC polarization losses decreased significantly. This trend is counterintuitive as a thicker buffer layer can result in a longer transport path for the O²⁻ ions and increased ohmic resistance. With 45 and 75 layers of SDC slurry sprayed, the peak power density at 750 °C increased to 650 mW/cm² and 725 mW/cm², respectively. At 750 °C, a peak power density of 1004 mW/cm² and a power density of 802 mW/cm² at 0.6 V occurred with 105 layers of SDC slurry sprayed. The buffer layer thickness was ∼1.7 μm with 105 layers of SDC slurry sprayed. With additional layers of SDC slurry sprayed, the polarization increased slightly. For 135 layers of SDC slurry sprayed, the peak power density was 908 mW/cm² at 750 °C. From these results, it is evident that a change in the SDC buffer layer thickness from 0.4 to 1.7 μm (15 layers to 105 layers of SDC sprayed) results in a significant reduction in mT-SOFC polarization losses and an increase in power density.

Electrochemical impedance spectroscopy was conducted on the mT-SOFCs at 750 °C immediately after the current–voltage test. Figure 3 shows the EIS data on a Nyquist plot for 15, 45, 75, 105, and 135 layers of SDC sprayed. With 15 layers of SDC sprayed, the total resistance was much larger than any of the other mT-SOFCs with thicker buffer layers. The total resistance tended to decreases with a minimum occurring for the mT-SOFC with 105 layers of SDC sprayed. Figure 3 inset shows the lower intercept of the real axis. The trend in ohmic resistance was identical to the changes in polarization observed with the current–voltage method. The ohmic resistance of the mT-SOFC decreased as the number of layers of SDC spray deposited increased from 15 up to 105. The ohmic resistance increased gradually as the number of layers of SDC sprayed increased above 105, but the ohmic resistance remained lower than the results with 15, 45, and 75 layers of SDC spray deposited. The initial decrease in ohmic resistance with less layers of SDC sprayed is attributed to a decrease in reactions between the LSCF cathode and YSZ electrolyte. A thinner, porous buffer layer is less effective at preventing Sr and Zr diffusion across the buffer layer [36]. This possibility was noted in a previous study on the porous SDC buffer layer thickness of a planar SOFC [15] and will be investigated further with EDS later in this work. With 105 layers of SDC sprayed, the reactions appear to have been minimized due to a thicker SDC layer. Further increase in the buffer layer thickness increases the ionic diffusion path for O²⁻ ions through the SDC buffer layer and YSZ electrolyte and results in increased ionic resistance. The changes in ohmic resistance appear to be primarily responsible for the trend in mT-SOFC polarization and power density as the SDC buffer layer thickness changes. This is supported by the lower polarization of the mT-SOFC with 135 layers sprayed (Fig. 2), which had lower ohmic resistance and higher polarization resistance compared with the mT-SOFC with 75 layers sprayed. The polarization resistance also showed a slight increase in polarization resistance as the buffer layer thickness exceeds the optimal ∼1.7 μm (105 layers of SDC sprayed). Differences in the cathode activation polarization are typically observed at high frequency (∼100 Hz) [55]. The mT-SOFC with 15 layers of SDC sprayed has higher impedance compared with 45 layers of SDC sprayed near 100 Hz, which indicates that a change in cathode activation polarization is also occurring as a result of increasing the buffer layer thickness.

The mT-SOFCs with different thicknesses of the GDC buffer layer were investigated with the current–voltage method. Figure 4 shows the results for 15, 45, 75, 105, and 135 layers of GDC slurry sprayed. Overall, changing the GDC buffer layer thickness resulted in very similar trends as those observed for the mT-SOFCs with the SDC buffer layer. With 15 layers of GDC slurry sprayed, the thinnest buffer layer of ∼0.3 μm resulted. However, the mT-SOFC performance resulting from 15 layers of GDC slurry sprayed was much higher than the mT-SOFC with 15 layers of SDC slurry sprayed. At 750 °C, a power density of 382 mW/cm² occurred at 0.6 V with 15 layers of GDC slurry sprayed, which was almost double the peak power density with 15 layers of SDC slurry sprayed. Higher performance also occurred with 45 and 75 layers of GDC slurry sprayed (compared with the SDC buffer layer) with peak power densities at 750 °C of 721 mW/cm² and 772 mW/cm², respectively. A peak power density and high power density of 617 mW/cm² at 0.6 V occurred with 75 layers of GDC slurry sprayed. The performance at 750 °C with 105 layers of GDC slurry sprayed was only slightly lower with a power density of 610 mW/cm² at 0.6 V. The optimal condition observed was 75 layers of GDC sprayed for a thickness of ∼1.3 μm. However, at high current densities, there was a more apparent difference between the mT-SOFC performance with 75 and 105 layers of GDC sprayed. This will be discussed further when the EIS data are analyzed. Comparing the results of the mT-SOFCs with SDC and GDC buffer layers indicates that there was a much larger change in polarization losses and power density for the mT-SOFC with the SDC buffer layer compared with those with the GDC buffer layer. For example, with the same number of layers of SDC and GDC sprayed in each case (15 and 105 layers sprayed), the mT-SOFC power density at 0.6 V increased from 199 to 802 mW/cm² with a SDC buffer layer and from 382 to 610 mW/cm² with a GDC buffer layer.

Figure 5 shows the EIS data for the mT-SOFCs with 15, 45, 75, 105, and 135 layers of GDC slurry sprayed. Figure 5 inset shows the real axis intercept. The ohmic resistance of the mT-SOFC at 750 °C was high with 15 layers of GDC slurry sprayed and decreased to a minimum with 45 layers of GDC slurry sprayed. This decrease is attributed to a decrease in reactions between the YSZ electrolyte and LSCF cathode as the buffer layer thickness increases. However, the ohmic resistance increased gradually as the thickness of the GDC buffer layer increased with more than 45 layers sprayed. That trend differs from the mT-SOFCs with the SDC buffer layer as GDC appears to have been either: (1) better at preventing the interfacial reactions and decreasing the ohmic resistance with fewer layers of powder sprayed or (2) the lower ohmic resistance is due to higher ionic conductivity of GDC compared with SDC. The optimal condition of 75 layers of GDC sprayed showed higher total resistance than 105 layers of GDC sprayed, but it was due to higher polarization resistance, which only effected the mT-SOFC at higher current densities (Fig. 4). Those higher current density conditions are well beyond the operating voltage of a SOFC and are likely the result of a manufacturing defect rather than the 75 layer case actually having higher polarization resistance. An increase in polarization and a decrease in power density with more than 75 layers of GDC slurry sprayed are primarily attributed to an increase in ohmic resistance, as the total polarization resistance was similar for the mT-SOFCs with 45, 105, and 135 layers of GDC sprayed. The case of 45 layers of GDC slurry sprayed showed the lowest ohmic resistance but higher polarization at low current densities (Fig. 4). The higher polarization at low current density was linked to the higher polarization resistance observed in EIS compared with 105 and 135 layers of GDC sprayed. When comparing the mT-SOFCs with SDC and GDC buffer layers, the EIS data indicate that the GDC buffer layer resulted in lower ohmic resistance for the case of 45 and 75 layers of spray deposited, while the mT-SOFCs with the SDC buffer layer had lower ohmic resistance for 105 and 135 layers of spray deposited.

The mT-SOFCs with the SDC buffer layer were analyzed with the SEM. Cross-sectional morphologies of the mT-SOFCs are shown in Fig. 6 for the 15 and 135 layers of SDC sprayed. Figure 7 shows cross-sectional morphologies for the mT-SOFCs with 15 and 135 layers of GDC sprayed. As shown, the sintered YSZ electrolyte thickness was ∼22 μm and was relatively consistent across all samples, as expected, due to the common fabrication technique. As was mentioned previously, the sintered SDC buffer layer thickness measured from the SEM images was ∼0.4 μm, ∼0.9 μm, ∼1.3 μm, ∼1.7 μm, and ∼2.0 μm for the 15, 45, 75, 105, and 135 layers of SDC slurry sprayed, respectively. The sintered GDC buffer layer thickness was ∼0.3 μm, ∼1.0 μm, ∼1.3 μm, ∼1.6 μm, and ∼1.8 μm for the 15, 45, 75, 105, and 135 layers of GDC slurry sprayed, respectively. One of the unexpected results was the difference in cathode thickness across samples. For the mT-SOFC with 15 layers of SDC sprayed, the LSCF + SDC cathode was ∼6 μm thick, while all the other samples had a cathode thickness of ∼9 μm. Similarly, for the mT-SOFC with 15 layers of GDC sprayed, the LSCF + GDC cathode was ∼2 μm thick, while all the other samples had a cathode thickness of ∼6 μm. During the dip coating fabrication process, it was evident that the cathode slurry adhered better to the thicker buffer layers than it did to the thin buffer layer that had only 15 layers of material spray deposited. The buffer layer, being porous, allows for greater penetration of the cathode slurry (during dip coating) when it is thicker, which improves the adhesion. Additional cells were prepared with no buffer layer, and it was evident that the cathode slurry did not adhere well to the dense YSZ electrolyte with no buffer layer. The result was a thinner cathode when the buffer layer was thin. This effect was not as evident when the buffer layer was thicker as the cathode thickness remained relatively constant across the other samples. As a result, the higher activation polarization observed with EIS for the mT-SOFC with 15 layers of SDC sprayed (Fig. 3) and for the mT-SOFC with 15 layers of GDC sprayed (Fig. 5) can partially be attributed to a thin cathode layer. As LSCF has mixed ionic and electronic conductivity, a thinner cathode can limit the area for the oxygen reduction reaction.

Energy-dispersive X-ray spectroscopy was utilized to analyze the effect of the buffer layer thickness on the interdiffusion of species from the cathode to the electrolyte. A line scan was conducted on the mT-SOFC cross section between the cathode and electrolyte with the relative concentration of species monitored throughout. Figure 8 shows the results of the EDS analysis for the mT-SOFC with 15 and 45 layers of SDC sprayed. With 15 layers of SDC deposited on the electrolyte, some separation of the cathode and electrolyte materials occurs as the Zr and Y have a clear drop in concentration across the buffer layer. Similarly, the La, Sr, Co, and Fe of the cathode had a clear drop in concentration across the buffer layer with only 15 layers of SDC sprayed. However, the relative concentration of Zr in the cathode decreases when the buffer layer thickness increased as a result of depositing 45 layers of SDC. Similarly, the relative concentrations of Sr, Fe, and Co in the electrolyte decrease when the buffer layer thickness increased.

The results of the EDS analysis for the mT-SOFC with 15 and 45 layers of GDC sprayed are shown in Fig. 9. Similar trends occurred as was observed from the mT-SOFCs with the SDC buffer layer, but the change in concentration across the buffer layer was less pronounced. For the case of the mT-SOFC with 15 layers of GDC spray deposited, there was no noticeable change in concentration of many of the species (Zr, Y, Sr) across the buffer layer. The presence of these elements, in particular, can lead to the formation of La₂Zr₂O₇ and SrZrO₃, which were previously reported as non-conductive layers formed during sintering that decrease the SOFC electrochemical performance [17]. This interdiffusion and formation of these non-conductive layers explains the high ohmic resistance when only 15 layers of SDC or GDC were spray deposited. However, in the case of 45 layers of GDC sprayed, there was much better separation between the elements in the cathode and electrolyte. Despite the similarities between the mT-SOFCs with SDC and GDC buffer layers, the EDS results indicate that 15 layers of GDC were less effective at preventing interdiffusion of the electrolyte and cathode species than the 15 layers of SDC. However, the polarization curves and EIS results indicate that GDC was better at preventing the interdiffusion due to lower ohmic resistance. As a result, the differences in the EDS results appear to represent specific parts of the cross section only. The thicker GDC buffer layer (e.g., 105 layers of GDC spray deposited) resulted in much less interdiffusion of the cathode and electrolyte materials, which resulted in less interfacial reactions that form non-conductive layers and improved SOFC power density.

The mT-SOFC with 105 layers of SDC sprayed was utilized in another experiment [11] to assess the fuel cell stability due to the potential diffusion of Sr from the cathode to the electrolyte at operating temperatures. No degradation of the mT-SOFC voltage or power density was observed over a 60 min test at a constant current density of 350 mA/cm² at 800 °C [11]. Further research is needed to analyze long-term stability of mT-SOFC with the thin porous buffer layer.

Micro-tubular SOFCs were prepared using the same fabrication techniques, but with different buffer layer materials (i.e., SDC and GDC) and with the buffer layer thickness varying between 0.3 and 2.0 μm. The comparison of the samples using the current–voltage method and EIS indicates that the thickness of the porous buffer layer has a significant effect on the SOFC polarization losses, power density, and impedance. The thinnest SDC and GDC buffer layers resulted in high polarization, low power density, high ohmic resistance, and high cathode activation polarization. Increasing the buffer layer thickness resulted in lower polarization and higher power density. However, the mT-SOFCs with the GDC buffer layer had the lowest ohmic resistance with 45 and 75 layers of material spray deposited, while the mT-SOFCs with the SDC buffer layer had the lowest ohmic resistance with 105 and 135 layers of material spray deposited. With the same number of layers of SDC and GDC sprayed in each case (15 and 105 layers sprayed), the mT-SOFC power density at 0.6 V increased from 199 to 802 mW/cm² with a SDC buffer layer and from 382 to 610 mW/cm² with a GDC buffer layer. The EDS line scan indicates that GDC was less effective at preventing species interdiffusion than SDC with the same number of layers of spray deposited. When the buffer layer is thin, interdiffusion of the cathode and electrolyte materials results in the formation of non-conductive layers and is attributed as the cause of the higher ohmic resistance and high cathode activation polarization. The thinnest buffer layer also had less adhesion to the cathode during fabrication with dip coating, which resulted in a thinner cathode when sintered and is also attributed as a reason for the higher activation polarization. Based on these results, it is evident that even small changes (∼1 μm) in the thickness of the porous buffer layer can result in significant changes in polarization, power density, impedance and interdiffusion of cathode and electrolyte materials.

This material is based upon work supported by an Agreement with Syracuse University awarded by its Syracuse Center of Excellence for Environmental and Energy Systems with funding under prime award number DE-EE0006031 from the US Department of Energy and matching funding under award number 53367 from the New York State Energy Research and Development Authority (NYSERDA) and under NYSERDA contract 61736. This material is also based upon work supported by the National Science Foundation Graduate Research Fellowship Program under Grant No. 1746928.

There are no conflicts of interest.

The datasets generated and supporting the findings of this paper are obtainable from the corresponding author upon reasonable request. The authors attest that all data for this study are included in the paper.

DC-SOFC=

dual chamber solid oxide fuel cell

GDC=

gadolinium-doped ceria

LSCF=

lanthanum strontium cobalt ferrite

LSM=

lanthanum strontium manganite

SDC=

samaria-doped ceria

SC-SOFC=

single chamber solid oxide fuel cell