Effect of Li₂O additions upon the crystal structure, sinterability and electrical properties of yttria stabilized zirconia electrolyte

Abstract

The crystal structure, lattice constant, sinterability, microstructure, conductivity and cell performance of 8 mol% yttria stabilized zirconia (YSZ) with varying ratios of Li₂O (n mol%, n = 0, 0.25, 0.5, 1, 1.5, 1.7, 2, 2.5, 3), i.e., n mol% Li₂OYSZ, were investigated. The results show that the Li⁺ can dissolve into the ZrO₂ cubic lattice and cause lattice contraction due to the small radius of Li⁺. The YSZ retained its cubic lattice until Li₂O addition of 1.7 mol%, and the cubic-to-monoclinic transition was detected in cases of Li₂O content more than 1.7 mol%, leading to the increase of the lattice parameter. Li₂O addition of 0.25 and 0.5 mol% decreased the sintering temperature from the typical 1400 °C to 1265 °C and 1250 °C, and promoted both the densification and electrical conductivity of the YSZ samples. The conductivities of 0.25 mol% Li₂OYSZ and 0.5 mol% Li₂OYSZ samples at 800 °C were as high as 0.0275 S cm⁻¹ and 0.0313 S cm⁻¹, which are 1.26 and 1.43 times greater than that of pure YSZ, respectively. Then the densification of the YSZ samples reduced gradually at Li₂O content greater than 1 mol%. For n mol% Li₂OYSZ (n ≥ 1.7) samples, a dense and pure cubic phase ceramic could be achieved at sintering temperatures above 1400 °C due to the vaporization of Li₂O at such high temperatures resulting from its high vapor pressure. Solid oxide fuel cells (SOFCs) with 0.25 mol% Li₂OYSZ and 0.5 mol% Li₂OYSZ electrolyte displayed open circuit voltages higher than 1.0 V at 800 °C, indicating the absence of electronic conductivity, and meanwhile exhibited much increased maximum power densities than that of the cell with pure YSZ electrolyte, implying that inclusion of a small amount of Li₂O into YSZ electrolyte could be a desirable improvement approach for zirconia-based SOFC.

---

1. Introduction

Yttria stabilized zirconia (YSZ) is commonly used as an electrolyte for solid oxide fuel cells (SOFC) because of its pure ionic conductivity and chemical stability both in oxidizing and reducing atmospheres.^1–5 However, a major drawback of the use of zirconia-based materials is the high firing temperature required to achieve usable densification (usually exceeding 1400 °C for numerous hours^6,7). To reduce the sintering temperature, various alkaline-earth metal or transition metal oxides have been studied as sintering aids to improve sinterability of YSZ ceramics, such as metal oxides of Ca,⁸ Co,⁶ Fe,⁷ Cu,⁹ Mn¹⁰ and Ni.¹¹ Nevertheless, negative effects of these transition metal oxides on the properties of YSZ were also reported,^6,8,11–14 causing usually a decrease in conductivity, or even accompanied by the zirconia phase transformation sometimes.

As for lithium oxide or lithium salts, which have been used for the sintering aids of those ceramics that usually present poor densification, and encouraging results were reported.^15–20 In particular, Esposito et al.²¹ found that samaria doped ceria (SDC) could be fully densified at 900 °C with 5 mol% Li₂O as an additive, and Han et al.²² reported highly dense gadolinia doped ceria (GDC) sintered bodies with addition of 2.5 mol% Li₂O were obtained at a remarkably low temperature of 800 °C, even though SDC and GDC samples free of sintering additives were normally sintered at as high as 1550 °C to get a relative density of 95%.^23,24 To the best of the authors' knowledge, Li₂O has scarcely been reported to be applied as a sintering aid of YSZ, and the only application on promotion of YSZ conductivity was reported by Nagaeva et al.,²⁵ and consequently the present study was initiated to evaluate the potential of Li₂O additives to decrease the sintering temperature or improve other properties of YSZ. Besides, it is necessary to know how the properties will be changed by the possible interaction between YSZ and minor Li₂O additives.

2. Experimental

The different molar ratios of lithium oxide (i.e., n mol% Li₂O, where n = 0, 0.25, 0.5, 1, 1.5, 1.7, 2, 2.5, 3) to 8 mol% Y₂O₃ stabilized ZrO₂ (YSZ, Tosoh Corporation, Tokyo, Japan) were added into YSZ powders in the form of lithium nitrate (>99.99% pure, Baotou, China), then the powders were wet ground (in ethanol) in a polyurethane jar for 24 h. The resulting paste was then oven-dried at 70 °C and subsequently calcined at 600 °C for 2 h in air to decompose lithium nitrate into Li₂O. The derived powders were denoted as n mol% Li₂OYSZ, and they were characterized by differential thermal analysis and thermo-gravimetric analysis (DTA/TG, Netzsch STA 449C) in air at a heating rate of 10 °C min⁻¹ from ambient temperature to 1300 °C. Sintering behaviors of samples were performed using a dilatometer (Netzsch DIL 402C, Germany) at a heating rate of 5 °C min⁻¹ from room temperature to 1400 °C.

Slurries for tape casting were prepared as our previous study²⁶ by dispersing n mol% Li₂OYSZ powders in mixed ethanol–butanone solvent with caster oil as dispersant, dibutyl phthalate (DBP) as plasticizer, and polyvinyl butyral (PVB) as binder. The slurries were ball milled for 48 h before tape casting using a tabletop castor (DR-150, Japan). Thereafter, tapes were dried in air at room temperature for 12 h and punched to discs with a diameter of 20 mm, then fired in air at different temperatures, varied from 700–1400 °C for 8 h. The fired discs were characterized by X-ray diffraction (XRD, PANalytical X'Pert PRO, Netherlands) with Cu Kα₁ radiation over the range of 2θ from 20° to 80°. The relative densities of n mol% Li₂OYSZ pellets were determined by the Archimedes drainage method. Microstructures of fired ceramic were observed by a field emission scanning electron microscope (JSM-7800F, JEOL, Japan). Two electrode electrochemical impedance spectroscopy (IM6, ZAHNER, Germany) measurements were carried out in dry air in a temperature range between 300 °C and 800 °C and over a frequency range of 0.1 Hz to 4 MHz with 10 mV as the excitation ac amplitude.

The anode pastes comprising NiO : YSZ of 50 : 50 wt% were mixed throughly and ball milled with ethylcellulose and terpineol for 24 h, then the slurry was screen printed onto the fired electrolyte disks, followed by co-firing at 1250 °C for 5 h in air. Similarly, the (La_0.8Sr_0.2)_0.95Mn_3−δ (LSM) cathode paste comprising of LSM, YSZ, ethylcellulose and terpineol was coated on the other side of electrolyte. Afterwards the disks were sintered in air at 1000 °C for 2 h to get the NiO–YSZ‖n mol% Li₂OYSZ‖YSZ–LSM electrolyte-supported cells with about 150 μm thick electrolyte, and approximately 30 μm thick cathode, followed by sealing of the single cells with ceramic sealant (CERAMABOND 552, USA). And then, cell performances were evaluated at 800 °C in a humidified hydrogen with 3 vol% H₂O at a rate of 70 mL min⁻¹ on the anode side and ambient air on the cathode side. The I–V characterization was conducted on Arbin instrument (BT2000, USA).

3. Results and discussion

3.1 XRD studies and thermal analysis

X-ray diffractograms of n mol% Li₂OYSZ (n = 0, 0.25, 0.5, 1, 1.5, 1.7, 2, 2.5, 3) discs fired at 1250 °C for 8 h are presented in Fig. 1. The same sample height and flatness were carefully preserved while performing the XRD tests to reduce the experimental error. From Fig. 1b, it's noted that the diffraction peaks of samples added with Li₂O shift to right (higher diffraction angles) with the increase of n in n mol% Li₂OYSZ (n ≤ 1.7 mol%) compared to that of pure YSZ, indicating that the Li⁺ was effectively incorporated into the zirconia crystal lattice and gradually reduced the lattice parameter as shown in Fig. 2, since the ionic radius of Li⁺ (0.68 Å)¹⁵ is smaller than that of Zr⁴⁺ (0.84 Å).^7,27 Besides, no detectable phase transition was observed for YSZ peaks when the Li₂O additive amounts were less than 1.7 mol%, suggesting that small Li₂O addition does not change the crystalline structure of YSZ. As the addition concentration of Li₂O is increased up to n = 1.7, monoclinic phase of zirconia appeared, and it became progressively intensified with the increasing Li₂O additive amounts, thus leading to the steady increase of the lattice constant. This indicates a possibility that the solid solution of cubic YSZ with Li₂O is limited to below 1.7 mol%. The lattice constant calculated on the basis of Bragg equation and crystal plane indices by adopting the MDI Jade software (version 5.0) gradually decreases first and then increases with the further increase of Li₂O content as indicated in Fig. 2, which is consistent with the observations that diffraction peak initially shifted toward higher angles from pure YSZ to 1.7 mol% Li₂OYSZ, followed by shifting back to lower angles for samples with the Li₂O addition more than 1.7 mol%.

Fig. 1 X-ray diffraction patterns of n mol% Li₂OYSZ (n = 0, 0.25, 0.5, 1, 1.5, 1.7, 2, 2.5, 3) pellets after sintering at 1250 °C for 8 h.

Fig. 2 Variation of lattice constant against Li₂O addition concentration.

XRD patterns of 1.7 mol% Li₂OYSZ pellets sintered at various temperatures from 700 to 1400 °C for 8 h are shown in Fig. 3. The monoclinic phase of zirconia was not observed in Fig. 3a until the temperature is increased up to 800 °C, and the intensities of the diffraction peaks of monoclinic zirconia increase steadily with increasing temperatures from 800 to 1100 °C, presumably indicating that the incorporation of lithium oxide into zirconia crystal lattice need to overcome certain energy barrier, and high temperature provides the possibility. Accordingly, the diffraction peaks shifted to lower angles as sintering temperature increased as illustrated in Fig. 3b due to the lattice expansion caused by increasing cubic-to-monoclinic phase transition. Then the diffraction peaks of monoclinic zirconia gradually abate or even disappear with the further increase of sintering temperatures over 1100 °C, accompanied by the diffraction peaks shifting toward higher angles as a result of crystal lattice contraction. When the 1.7 mol% Li₂OYSZ disc was sintered at 1400 °C, the X-ray diffraction pattern shows that the coexistence of cubic and monoclinic phases disappeared, only a cubic fluorite structure of zirconia is present on the diffractogram without a secondary phase of monoclinic zirconia. This could be attributed to the much higher vapor pressure of Li₂O²⁸ (at least three orders of magnitude higher) than that of common metal oxide additives such as Fe₂O₃, CoO and NiO at the same temperature. The higher the temperature is, the greater the vapor pressure of lithium oxide will be. Therefore, high sintering temperature will accelerate the vaporization of Li₂O, which is responsible for the diminution and disappearance of monoclinic zirconia above 1100 °C, similar to the report that Li species could mostly vaporize in the GDC with 2.5 mol% Li₂O samples sintered above 1100 °C mentioned elsewhere.²⁸

Fig. 3 X-ray diffractograms of 1.7 mol% Li₂OYSZ discs calcined at 700–1400 °C for 8 h.

Fig. 4 shows the TG/DTA results for 1.7 mol% Li₂OYSZ precursor particles. According to the TG curve, no significant weight loss was recognized. This result implies that the nitrates were decomposed completely, and there were no organic and nitrate group residues in the precursor powders derived from calcining the mixture of YSZ and lithium nitrate at 600 °C for 2 h as previously described in Experimental section. And the DTA curve presents four major peaks: two for endotherm and the other two for exotherm. The endothermic peak at about 845 °C was likely due to the cubic-to-monoclinic phase transformation as indicated in Fig. 3a. The latter endothermic peak around 1000 °C probably corresponded to the increase of monoclinic zirconia as depicted in Fig. 3a. The broad exothermic peaks at approximate 1100 °C and 1230 °C on the DTA curve may be ascribed to the diminishment and elimination of monoclinic zirconia as shown in Fig. 3a due to the vaporization of Li₂O at such high temperatures causing by its high vapor pressure. To avoid the occurring of frequent phase changes, the amounts of Li₂O addition should be less than the critical point of phase transition, i.e. the solid solution limit of 1.7 mol% Li₂O in YSZ.

Fig. 4 Differential thermal analysis (DTA) and thermo-gravimetric (TG) curves for 1.7 mol% Li₂OYSZ precursor particles.

3.2 Sintering behaviors and microstructural characterization

The plots in Fig. 5 show the linear shrinkage (ΔL/L₀) and shrinkage rate (d(ΔL/L₀)/dt) as a function of temperature for YSZ with various Li₂O levels. It is clear that the small content of Li₂O addition shifts the onset of sintering to low temperature range and decreases the maximum shrinkage rate temperature (T_max), which is in the following order: T_max^0.5% (∼1193 °C) > T_max^0.25% (∼1203 °C) > T_max^1% (∼1244 °C) > T_max^1.5% (∼1286 °C) > T_max^0% (∼1292 °C) > T_max^1.7% (∼1373 °C). Moreover, 0.25 mol% Li₂OYSZ and 0.5 mol% Li₂OYSZ samples nearly finished shrinkage at about 1265 °C and 1250 °C, respectively, while the YSZ sintering shrinkage was still going on even at 1350 °C. The results demonstrate that the Li₂O addition can remarkably reduce sintering temperature. Among these addition levels, 0.5 mol% Li₂O is the most effective additive amount for enhancing the densification of YSZ, and the difference of T_end (the shrinkage end temperature) values between the YSZ and 0.5 mol% Li₂OYSZ is about 150 °C.

Fig. 5 Sintering characteristics of n mol% Li₂OYSZ powders.

Fig. 6 depicts the cross-sectional microstructural morphology of n mol% Li₂OYSZ (n = 0, 0.25, 0.5, 1, 1.5, 1.7, 2, 2.5, 3) pellets sintered at 1250 °C. Porous structure with large holes were observed for the Li₂O free YSZ sample, and the density obviously improved as Li₂O content increased, only few pinholes were observed in 0.25 mol% Li₂OYSZ. A full dense microstructure was obtained in the 0.5 mol% Li₂OYSZ sample, which may be attributed to the liquid-phase sintering effect caused by lithium oxide that typically has a feature of achieving dense ceramics with small grains.²⁹ The grain size of 0.5 mol% Li₂OYSZ calcined at 1250 °C is about 100–200 nm, whereas YSZ sintered at the same temperature has an average size of around 300 nm as shown in Fig. 7, demonstrating that the liquid sintering mechanism dominates the process of 0.5 mol% Li₂OYSZ densification. The result coincides with the observation that the incorporation of lithium salts into ceria-based ceramics could suppress the grain growth and achieve dense ceramics with fine grains, which has been reported in other literature.³⁰ As the content of Li₂O was above 0.5 mol%, the porosity increased, and a sharp decrease of densification was observed as Li₂O increased over 1.7 mol% owing to the cubic-to-monoclinic phase transition. These observations are in good agreement with what we have illustrated in Fig. 5, which indicates that Li₂O can lower the sintering temperature of YSZ ceramic effectively from 1400 °C to 1250 °C.

Fig. 6 Cross-sectional SEM images of n mol% Li₂OYSZ pellets calcined at 1250 °C, (a) n = 0; (b) n = 0.25; (c) n = 0.5; (d) n = 1; (e) n = 1.5; (f) n = 1.7; (g) n = 2; (h) n = 2.5 and (i) n = 3.

Fig. 7 SEM surface graphs of (a) YSZ and (b) 0.5 mol% Li₂OYSZ pellets sintered at 1250 °C.

Variation of full density temperatures of n mol% Li₂OYSZ (n = 0, 0.25, 0.5, 1, 1.5, 1.7, 2, 2.5, 3) pellets as a function of Li₂O dopant levels are illustrated in Fig. 8. When 0.25 mol% Li₂O was added to YSZ, the full density temperatures decreased considerably from 1400 °C to 1265 °C, and 0.5 mol% Li₂OYSZ seems to be the most preferred alternative with the lowest densification temperature required around 1250 °C among all the doping levels. As the concentration of Li₂O increased from 0.5 to 3 mol%, the full density temperatures increased rapidly from 1250 °C to 1550 °C. Afterwards, the phase inspection was carried out for these dense ceramic discs with Li₂O content more than 2 mol%, and the XRD patterns are exhibited in Fig. 9. What is different from the diffractogram of n mol% Li₂OYSZ (n ≥ 2) calcined at 1250 °C as shown in Fig. 1 is that only a pure cubic zirconia phase remained without the occurring of a secondary phase of monoclinic zirconia for n mol% Li₂OYSZ (n ≥ 2) samples sintered above 1500 °C as a result of Li₂O vaporization at elevated temperatures. The results match what we have illustrated in Fig. 3 for 1.7 mol% Li₂OYSZ sintered at 1400 °C. So the high vapor pressure of Li₂O ensures that Li₂O can be developed as a non-residual sintering additive.

Fig. 8 Variation of full density temperatures of n mol% Li₂OYSZ (n = 0, 0.25, 0.5, 1, 1.5, 1.7, 2, 2.5, 3) pellets as a function of Li₂O dopant levels.

Fig. 9 X-ray diffractograms of n mol% Li₂OYSZ (n = 2, 2.5, 3) discs calcined at 1500–1550 °C for 8 h.

3.3 Electrical properties and cell performance

Fig. 10 shows the bulk, grain boundary and total electrical conductivity of fully dense doped YSZ as a function of the amount of Li₂O at temperatures of 300–800 °C. Fig. 10a presents that the grain boundary conductivities of n mol% Li₂OYSZ are always lower than that of the bulk below 400 °C, and then the bulk and grain boundary conductivity tend to be equally matched at 450 °C. Subsequently, the bulk and grain boundary conductivity followed a reverse pattern with the further increase of temperature above 500 °C, i.e. the grain boundary conductivity of n mol% Li₂OYSZ is higher than that of the bulk. The space charge model of acceptor-doped zirconia proposed by Guo et al.³¹ may account for the phenomenon. According to the model, a grain boundary consists of a grain-boundary core and two adjacent space-charge layers. The grain-boundary core of acceptor-doped zirconia is positively charged, oxygen vacancies are therefore depleted in the space-charge layer as a result of the strong Coulomb repulsive force between positively charged oxygen vacancies and grain-boundary core, leading to the lower grain boundary conductivities.³² However, the block effect of grain-boundary core became nondominant above 450 °C owing to increasing oxygen vacancies were sufficiently mobile to transport across the grain boundaries at elevated temperatures, resulting in the enhancement of contribution of grain boundary conductivities. Moreover, the bulk, grain boundary and total electrical conductivity firstly increased from pure YSZ to 0.25 mol% Li₂OYSZ and 0.5 mol% Li₂OYSZ. However, further increase in the Li₂O addition levels led to a decrease in the conductivities for 1 mol% Li₂OYSZ, 1.5 mol% Li₂OYSZ and 1.7 mol% Li₂OYSZ. This may be explained using the equation:where the symbol denotes the usual meanings of the Kroger–Vink notations. Accordingly, each pair of Li⁺ cations creates three defect for the charge compensation.²⁵ On one hand, the negatively charged will be attracted to the positively charged grain-boundary core, which caused the enrichment of around the grain boundary and neutralized the positive electricity field of grain-boundary core,^28,31,33 thus weakened the repulsive force of grain-boundary core to oxygen vacancies, therefore the grain boundary conductivity increased after Li₂O doping. On the other, the negatively charged and positively charged point defects tend to cluster due to the electrostatic attraction, and this tendency increases at the higher addition level leading to the formation of defect complex.^34–36 In 0.25 mol% Li₂OYSZ and 0.5 mol% Li₂OYSZ samples, the concentration and transition probability of oxygen vacancy increased, which leads to the gradual increase in the bulk and grain boundary conductivity, and accordingly promoted the total conductivity. However, on further increase of the Li⁺ doping (n > 0.5), more and more oxygen vacancies were bounded in the defect complex and difficult to migrate, thereby the effective concentration and transition probability of oxygen vacancy decreased, consequently resulting in low bulk and grain boundary conductivity, so does the total electrical conductivity as shown in Fig. 10b. It's also noted that the total electrical conductivity of YSZ was 0.0219 S cm⁻¹ at 800 °C, which is comparable to the report in other literature of 0.0212 S cm⁻¹.³⁷ The conductivities of 0.25 mol% Li₂OYSZ and 0.5 mol% Li₂OYSZ samples at 800 °C were improved significantly, reaching as high as 0.0275 S cm⁻¹ and 0.0313 S cm⁻¹, which are 1.26 and 1.43 times greater than that of lithium free YSZ, respectively, suggesting that small contents of Li₂O doped YSZ could be a potential and excellent electrolyte material for SOFCs.

Fig. 10 (a) Bulk (open symbols), grain boundary (filled symbols) conductivity and (b) total electrical conductivity of doped YSZ as a function of Li₂O contents at temperatures of 300–800 °C.

Fig. 11 shows the electrical conductivity of YSZ with different amounts of Li₂O as a function of temperature. From the Arrhenius plots, the activation energies of n mol% Li₂OYSZ both in the medium-low temperature range and the medium-high temperature range can be achieved, and the detailed values are summarized in Table 1. It is noted that the activation energies of n mol% Li₂OYSZ are ranked as follows: 0.5 mol% Li₂OYSZ < 0.25 mol% Li₂OYSZ < YSZ < 1 mol% Li₂OYSZ < 1.5 mol% Li₂OYSZ < 1.7 mol% Li₂OYSZ, revealing exactly the reversed change trend with the conductivity, which indicates content of Li₂O dopant plays an important role in determination of total conductivity.

Fig. 11 Temperature dependence of the electrical conductivity of YSZ with various Li₂O addition.

Table 1 Activation energies of n mol% Li₂OYSZ (n = 0, 0.25, 0.5, 1, 1.5, 1.7) electrolytes

Temperature range	Activation energies of n mol% Li2OYSZ (n = 0, 0.25, 0.5, 1, 1.5, 1.7)/eV
	n = 0	n = 0.25	n = 0.5	n = 1	n = 1.5	n = 1.7
Medium-low	1.188	1.171	1.169	1.189	1.190	1.245
Medium-high	0.860	0.855	0.852	0.864	0.872	0.877

---

The electrochemical performance of n mol% Li₂OYSZ (n = 0, 0.25, 0.5, 1, 1.5, 1.7) as electrolyte materials was evaluated in NiO–YSZ‖n mol% Li₂OYSZ‖YSZ–LSM electrolyte-supported cells. Characteristics of cells with different composite electrolytes measured in 3% H₂O humidified H₂ fuel and ambient air oxidant at 800 °C is given in Fig. 12. The measured open circuit voltage (OCVs) of the cells with Li₂O doped YSZ electrolytes were all above 1.0 V, indicating the absence of unfavorable electronic conduction in the Li₂O doped YSZ electrolytes under the temperature and oxygen pressure employed in this study. The 0.5 mol% Li₂OYSZ demonstrated the highest output power among all the compositions studied, and as expected the maximum power densities are ranked in the following order: 0.5 mol% Li₂OYSZ > 0.25 mol% Li₂OYSZ > YSZ > 1 mol% Li₂OYSZ > 1.5 mol% Li₂OYSZ > 1.7 mol% Li₂OYSZ, revealing that the cell performance has a lot to do with the observed electrical conductivities. The peak power densities of cells equipped with 0.25 mol% Li₂OYSZ and 0.5 mol% Li₂OYSZ reached as high as 0.106 and 0.122 W cm⁻², respectively, exhibiting much better performance than that of the cell using lithium oxide free YSZ electrolyte, thereby demonstrating that a small amount of Li₂O addition into YSZ electrolyte could be a desirable improvement approach for zirconia-based SOFC.

Fig. 12 Cell voltage (hollow symbols) and power density (solid symbols) against current density curves for cells running on 3% H₂O humidified H₂ at 800 °C with different Li₂O content YSZ electrolytes.

4. Conclusions

In the present study, the effects of small amounts (≤3 mol%) of Li₂O addition on the crystal characteristics, sintering behaviours, micro morphology, electrical properties and electrochemical performance of 8 mol% Y₂O₃ stabilized ZrO₂ were carried out. The increased Li₂O addition levels induced the gradual lattice contraction because of the smaller ionic radius of Li⁺ compared with that of Zr⁴⁺, and the solid solution limit of Li₂O in 8 mol% Y₂O₃ stabilized ZrO₂ is 1.7 mol%, above which phase transition from cubic to monoclinic occurred, accompanied by the lattice expansion. By adding 0.25 and 0.5 mol% Li₂O, the densification temperature of YSZ was reduced from the conventional temperature of exceeding 1400 °C to 1265 °C and 1250 °C, respectively. High electrical conductivities of 0.0275 S cm⁻¹ and 0.0313 S cm⁻¹ at 800 °C were obtained for fully dense 0.25 mol% Li₂OYSZ and 0.5 mol% Li₂OYSZ samples, respectively. SOFCs using 0.25 mol% Li₂OYSZ and 0.5 mol% Li₂OYSZ as electrolytes displayed excellent output power densities. These results suggest that small amounts of lithium oxide addition are promising for use in low temperature preparation of zirconia-based electrolytes for solid oxide fuel cells with excellent electrochemical performance.

Acknowledgements

We are greatly grateful to the financial supports from the State Key Development Program for Basic Research of China (2012CB215404), the National Natural Science Foundation of China (No. 21233006, 21473164, 21503197 and 51261120378) and Fundamental Research Funds for the Central University, China University of Geosciences (Wuhan) (No. CUG150603, CUG150615, CUG150627 and CUGL150802).

References

1. K. Keizer, A. J. Burggraaf and G. With, J. Mater. Sci., 1982, 17, 1095.
2. B. Li, J. Zhang, T. Kaspar, V. Shutthanandan, R. C. Ewing and J. Lian, Phys. Chem. Chem. Phys., 2013, 15, 1296.
3. T. Tatte, M. Part, R. Talviste, K. Hanschmidt, K. Utt, U. Maeorg, I. Jogi, V. Kiisk, H. Mandar, G. Nurk and P. Rauwel, RSC Adv., 2014, 4, 17413.
4. M. F. Han, X. L. Tang, H. Y. Yin and S. P. Peng, J. Power Sources, 2007, 165, 757.
5. E. Drozdz, RSC Adv., 2016, 6, 84752.
6. G. S. Lewis, A. Atkinson and B. C. H. Steele, J. Mater. Sci. Lett., 2001, 20, 1155.
7. H. B. Gao, J. Liu, H. Y. Chen, S. Li, T. M. He, Y. Ji and J. D. Zhang, Solid State Ionics, 2008, 179, 1620.
8. Z. G. Lv, R. S. Guo, P. Yao and F. Y. Dai, Mater. Des., 2007, 28, 1399.
9. J. L. Shi, T. S. Yen and H. Schubert, J. Mater. Sci., 1997, 32, 1341.
10. T. S. Zhang, S. H. Chan, W. Wang, K. Hbaieb, L. B. Kong and J. Ma, Solid State Ionics, 2009, 180, 82.
11. T. S. Zhang, Z. H. Du, S. Li, L. B. Kong, X. C. Song, J. Lu and J. Ma, Solid State Ionics, 2009, 180, 1311.
12. C. C. Appel, N. Bonanos, A. Horsewell and S. Linderoth, J. Mater. Sci., 2001, 36, 4493.
13. S. Linderoth, N. Bonanos, K. V. Jensen and J. B. Bilde-Sørensen, J. Am. Ceram. Soc., 2001, 84, 2652.
14. C. Bowen, S. Ramesh, C. Gill and S. Lawson, J. Mater. Sci., 1998, 33, 5103.
15. D. W. Yuan, S. F. Wang, W. Huebner and G. Simkovich, J. Mater. Res., 1993, 8, 1675.
16. J. D. Nicholas and L. C. De Jonghe, Solid State Ionics, 2007, 178, 1187.
17. C. L. Tsai, M. Kopczyk, R. J. Smith and V. H. Schmidt, Solid State Ionics, 2010, 181, 1083.
18. H. Naghib-zadeh, C. Glitzky, I. Dörfel and T. Rabe, J. Eur. Ceram. Soc., 2010, 30, 81.
19. A. Sayyadi-Shahraki, E. Taheri-Nassaj, S. A. Hassanzadeh-Tabrizi and H. Barzegar-Bafrooei, J. Alloys Compd., 2014, 597, 161.
20. S. M. Jamil, M. H. Dzarfan Othman, M. A. Rahman, J. Jaafar, A. F. Ismail and M. A. Mohamed, RSC Adv., 2015, 5, 58154.
21. V. Esposito, M. Zunic and E. Traversa, Solid State Ionics, 2009, 180, 1069.
22. M. F. Han, Z. Liu, S. Zhou and L. Yu, J. Mater. Sci. Technol., 2011, 27, 460.
23. K. Maca, J. Cihlar, K. Castkova, O. Zmeskal and H. Hadraba, J. Eur. Ceram. Soc., 2007, 27, 4345.
24. K. Maca, V. Pouchly and P. Zalud, J. Eur. Ceram. Soc., 2010, 30, 583.
25. N. Y. Nagaeva, A. A. Surin, L. A. Blaginina and V. P. Obrosov, Glass Ceram., 2008, 65, 199.
26. J. Xiong, C. R. Jiao, M. F. Han, W. T. Yi, J. Ma and C. Y. Yan, RSC Adv., 2015, 5, 87477.
27. A. J. Feighery and J. T. S. Irvine, Solid State Ionics, 1999, 121, 209.
28. T. L. Zhu, Y. Lin, Z. B. Yang, D. Su, S. G. Ma, M. F. Han and F. L. Chen, J. Power Sources, 2014, 261, 255.
29. X. H. Zhao, M. J. Wang, Q. L. Zhang, H. Yang, L. Hu and D. Yu, Mater. Lett., 2014, 122, 9.
30. M. Chen, H. Zhang, L. Fan, C. Wang and B. Zhu, Int. J. Hydrogen Energy, 2014, 39, 12309.
31. X. Guo and R. Waser, Prog. Mater. Sci., 2006, 51, 151.
32. X. Guo, W. Sigle and J. Maier, J. Am. Ceram. Soc., 2003, 86, 77.
33. X. Guo and R. Waser, Solid State Ionics, 2004, 173, 63.
34. P. P. Dholabhai, J. B. Adams, P. Crozier and R. Sharma, Phys. Chem. Chem. Phys., 2010, 12, 7904.
35. Y. Lin, S. M. Fang, D. Su, K. S. Brinkman and F. L. Chen, Nat. Commun., 2015, 6, 6824.
36. M. P. Delmo, S. Yamamoto, S. Kasai, T. Ono and K. Kobayashi, Nature, 2009, 457, 1112.
37. X. D. Xu, C. A. Tian, Q. Y. Yin and J. H. Cheng, Bull. Chin. Silic. Soc., 2011, 30, 593.

---