Low-temperature sintering and microstructure evolution of Bi2O3-doped YSZ

Author’s Accepted Manuscript Low-temperature Sintering and Microstructure Evolution of Bi2O3-doped YSZ Jianxun Han, Jingde Zhang, Fei Li, Junpeng Luan, Baoxin Jia www.elsevier.com/locate/ceri

PII: DOI: Reference:

S0272-8842(17)32231-9 https://doi.org/10.1016/j.ceramint.2017.10.039 CERI16458

To appear in: Ceramics International Received date: 10 April 2017 Revised date: 1 October 2017 Accepted date: 9 October 2017 Cite this article as: Jianxun Han, Jingde Zhang, Fei Li, Junpeng Luan and Baoxin Jia, Low-temperature Sintering and Microstructure Evolution of Bi2O3-doped YSZ, Ceramics International, https://doi.org/10.1016/j.ceramint.2017.10.039 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Title Low-temperature Sintering and Microstructure Evolution of Bi2O3-doped YSZ Author names and affiliations Jianxun Han a, b, Jingde Zhang a, b, * , Fei Li a, b, Junpeng Luana, b, Baoxin Jia a, b a

Key Laboratory for Liquid-Solid Structural Evolution and Processing of Materials (Ministry of

Education), Shandong University, Jinan 250061, China

b

Key Laboratory of Special Functional Aggregated Materials, Ministry of Education, Shandong

University, Jinan 250100, China

*Corresponding author Jingde Zhang, Key Laboratory for Liquid-Solid Structure Evolution and Processing of Materials (Ministry of Education), Shandong University, Jinan 250061, China. Tel.: +86053188392914.

E-mail address: [email protected] Present/permanent address Shandong University, No. 17923, Jingshi Road, Jinan, Shandong Province, P. R.China. 1

Low-temperature Sintering and Microstructure Evolution of Bi2O3-doped YSZ Jianxun Han a, b, Jingde Zhang a, b, * , Fei Li a, b, Junpeng Luana, b, Baoxin Jia a, b a

Key Laboratory for Liquid-Solid Structural Evolution and Processing of Materials (Ministry of Education), Shandong University, Jinan 250061, China

b

Key Laboratory of Special Functional Aggregated Materials, Ministry of Education, Shandong University, Jinan 250100, China

Abstract Yttria-stabilized zirconia (YSZ) is one of the most common solid electrolyte materials for solid oxide fuel cells (SOFCs). However, the sintering temperature of commercial YSZ powders is usually above 1500℃. In this work, powders with variable amounts (from 1mol% to 12mol%) of bismuth oxide (Bi2O3) were separately added to commercially available yttria partially stabilized zirconia (3YSZ) powders as sintering additives, in order to lower the sintering temperature of the latter. After the addition, the powder mixtures were dry-pressed and sintered at different temperatures between 850℃ and 1200℃. Then, the relative densities, phases and microstructures of the sintered ceramic specimens were analyzed. The results indicated that the addition of Bi2O3 enhanced the sintering activity of 3YSZ, resulting in a 3YSZ material with 2

an extremely low sintering temperature. When the amount of Bi2O3 exceeded 7mol%, the ceramic specimens were densified up to 96.3% by a pressureless conventional sintering process at 850℃ for only 1.5 h. Furthermore, Bi2O3 can affect the degree of crystallinity of different phases, such as the tetragonal phase and monoclinic phase in 3YSZ powder.

Keywords:

Bi2O3;

YSZ;

low-temperature

sintering;

sintering

additives

1. Introduction Solid oxide fuel cells (SOFCs) are the next generation electrical energy conversion systems due to their high energy efficiency, environmental friendliness and fuel flexibility.[1] Yttria stabilized zirconia (YSZ) is used as a solid electrolyte for SOFC thanks to its excellent chemical and mechanical properties, thermal stability, and pure oxygen ion conduction. However, the sintering temperature of commercial 3

YSZ powders is usually above 1500℃, which has substantially constrained its application.[2,3] For example, in a cathode-supported cell, it is required that the YSZ and doped lanthanum strontium manganite (LSM) ceramics are co-fired at 1200-1250℃, which is well below the sintering temperature of YSZ. Additionally, it is also required to control the thickness of the YSZ film to less than 15 μm in the electrode-supported cell by reducing the sintering temperature.[4-7] Accordingly, reducing the sintering temperature of YSZ is very important for its application in SOFCs.

Duran et al., and Theunissen et al., used nanopowders to produce dense yttria partially stabilized zirconia ceramics sintered at 1200℃ and 1150℃, respectively, with a relative density exceeding 97%[8,9]. However, the high cost of nanopowders has limited the application of such a process. In order to lower the sintering temperature of regular YSZ powders, sintering aids can be added (“additives”) to mix with the YSZ powders, which can lead to an accelerated densification process by forming liquid phases at low temperatures[10]. The performance of the YSZ material can be improved if doped with one or two metallic oxides.[11-18] For instance, a previous study reported the achievement of 96% or higher theoretical density at 1175℃, after adding 2mol% transition metal oxides (e.g. cobalt, manganese or iron oxides)[19]. A similar result was achieved by doping 3YSZ with alumina and lithium nitride, showing that a near full density (~96%) was attainable by sintering the mixture at 1170℃ for 5 h [20]. The results of Honorato et al. revealed that with the addition of Ni into YSZ, sintering was accomplished at 1300℃, which was 100℃ lower than 4

before[21].

Bi2O3 has a lower melting point of 825℃ and thus is a good candidate material for use as a sintering aid of YSZ. De Marco V. et al. used Bi2O3 as the additive for Gadolinia-doped Ceria (GDC), which lowered its normal sintering temperature by more than 400℃ [22]. Moreover, it was also found that 3mol% Bi2O3 additive improved the shrinkage of barium zirconate from 10.4 to 19.0% at 1480℃, and reduced the volume of pores when the mixture was sintered at 1400℃ for 24 h [23]. In this work, we investigated the effect of Bi2O3 on the sintering characteristics of 3YSZ.

2. Experimental 2.1. Materials

Commercially available ZrO2-3mol% Y2O3 (3YSZ) powder mixed with 1mol%-12mol%

Bi2O3

powder

was

used

as

the

starting

raw

material.

Polyvinyl alcohol (PVA) served as the binder to prepare the green body. All the chemicals were analytical reagent and used without further purification.

2.2. Preparation of the Bi2O3-3YSZ samples

The dopant, in the form of bismuth oxide dissolved in ethanol, was mixed into the YSZ/ethanol suspension using the wet lapping method. First, the mixture was milled in a QM-SB planetary ball miller (Nanjing University Instrument Factory) at the speed of 320 rpm/min for 12 h. Next, the resulting suspension was dried in a 5

ZK-025 vacuum drying oven for 5 h at 100℃. Eventually, we obtained the green body powders by drying and subsequent filtering using a 100-mesh sieve.

The powders with added PVA binder were uniaxially pressed at 30 MPa into pellets with a 27-mm diameter and a 2-mm thickness. The sintering of the green bodies was performed using the pressureless conventional sintering (CS) method at 850 -1200℃ with a heating rate of 5℃/min and a holding time of 1.5 h at the sintering temperature. S1-S12 represented the samples doped with different amount of Bi2O3 from 1mol% to 12mol%, at 1mol% increments.

2.3. Characterization

To observe the polished and thermally etched surfaces of the fired samples, we performed scanning electron microscopy (SEM) characterization, on a scanning electron microscope from Opton Optical Technology Co., Ltd., (Beijing, China). The density of the samples was measured by the Archimedes’ method (ISO Standard 39231/1-1979 (E)). X-ray diffraction (XRD) analysis was performed on a Bruker X-ray diffractometer (Bruker Scientific Technology Co., Ltd., Beijing, China), θ-2θ scans were used to determine the phase structure of the starting materials and sintered samples with a scan speed of 0.002°s-1 in the 2θ range of [20°, 90°]. Additionally, we carried out the thermogravimetric analysis (TGA) using a TGA Q500 instrument in the temperature range of 25-1200℃ with a heating rate of 10 ℃/min and an air flow rate of 50 ml/min.

6

The mass fraction of the monoclinic phase in the YSZ ceramics was estimated using the following formula:

X

m



I(111)m + I(111)m I(111)m + I(111)m  I(111)t

(1)

I(111)m where I(111)m , and I(111)t are the integrated intensities of the monoclinic

(111), monoclinic ( 111 ) and tetragonal (111) XRD peaks, respectively.

3. Results and discussion 3.1 Effect of Bi2O3 doping on the relative density of YSZ ceramics

The relative densities of the 3YSZ ceramics sintered at 850℃, 900℃ and 1,000℃ as a function of the doping amount of Bi2O3 (from 1mol% to 12mol%) are presented in Fig.1. These data reveal that, at the same sintering temperature, the relative density of the 3YSZ ceramics increased with the doping amount of Bi2O3. Moreover, the relative densities did not change significantly after the doping amount of Bi2O3 exceeded 7mol%. Noteworthy, the relative density of the ceramic sample doped with 7% Bi2O3 reached 96.3% when sintered at 850℃ for 1.5 h. Furthermore, if we consider the points in the individual curves where the relative density exceeded 95% as the critical points for densification of the YSZ ceramics, they are [7mol%, 96.3%], [6mol%, 96.6%] and [3mol%, 96.5%] for the curves of 850℃, 900℃ and 1,000℃, respectively.

7

The same sample sintered at different temperatures displayed different relative densities. When the amount of Bi2O3 was less than 7mol%, the relative density increased with the increase of the sintering temperature. In addition, if the doping amount of Bi2O3 reached 7mol% or higher, all samples sintered at temperatures at or above 850℃ showed a relative density higher than 96%. The SEM micrographs of the 3YSZ ceramics doped with 4mol% and 6mol% Bi2O3 (S4 and S6), and sintered at 1,000℃ are shown in Fig. 2. Clearly, the samples are very dense, without visible pores in the micron or sub-micron scale.

The melting point of Bi2O3 is 825℃, and it has been reported that Bi2O3 can be dissolved in ZrO2 at 710℃[24]. Accordingly, Bi2O3 liquid phase appeared when the process temperature exceeded 825℃. The densification of the 3YSZ samples was found to be dominated by a liquid phase-assisted sintering process[25]. The influence of the liquid phase-assisted sintering on the density of samples is dues to two different aspects. On the one hand, the mass-transport and diffusion rates of 3YSZ increase with the addition of the liquid phase, thus accelerating the densification process. On the other hand, the viscosity of the liquid phase decreases with an increasing temperature, resulting in better wetting properties at elevated temperatures. The large capillary force generated in this process would induce grain slipping and rearrangement, thus promoting the elimination of the pores. With an increasing Bi2O3 doping amount, more liquid phase would appear at the same sintering temperature, leading to an increase in the relative density.

8

As reported in the literature, the mass transport in a solid-liquid system is much faster than that in an all-solid system[26]. Accordingly, the samples of the same raw materials but sintered at different temperatures would show different densities due to the various viscosities of the liquid phase. Meanwhile, samples of diverse raw materials but sintered at the same temperature displayed different densities because of the various wetting properties of their liquid phases. However, when the amount of Bi2O3 exceeded 7mol%, the effect of the liquid phase would be at a saturation point, according to the result of the relative density. This phenomenon can be explained by the following: when the amount of Bi2O3 exceeds a threshold, it will locate between the YSZ grains. As a result, the elimination of the residual inter-grain pores via grain slipping and rearrangement becomes difficult. Instead, filling of the Bi2O3 liquid phase emerges as the dominant route to remove pores. Therefore, the grains of YSZ were bonded by the liquid phase formed by Bi2O3. At this time, the whole block was compact. Therefore, with the increase of Bi2O3 content, the densification process is no longer accelerated. Consequently, when the amount of Bi2O3 exceeded 7mol%, increasing the amount of Bi2O3 or the sintering temperature would have little influence on the densification of YSZ ceramics.

When the doping amount of Bi2O3 was less than 1mol%, the ceramic sample (S1) could not be fully densified at a lower temperature (1,100℃). On the one hand, the sintering process is dominated by the conventional solid-state diffusion mechanism with little amount of the liquid phase. Indeed, a related research showed that YSZ with little amount of Bi2O3 can be sintered only when the sintering temperature 9

reached 1,500℃ [27]. On the other hand, when the samples were fired above 1,100℃, Bi2O3 volatilized heavily. The relative density of the three samples (S2, S3 and S4) as a function of the sintering temperature are shown in Fig. 3. A turning-point occurred at 1,100℃, and there after the density decreased with the increasing temperature. The TGA curves of the S1, S2, S3 and S4 samples are depicted in Fig. 4. Mass losses of the samples mainly comprised the moisture loss and the volatilization loss of Bi2O3. The curve of the samples in Fig. 4 (except for S2) can be divided into three segments: the first one spans the temperature range from 25℃ to 825℃, the second one spans from 825℃ to 1,100℃, and the third one is from 1,100℃ and above. In the first segment, the down-hill slope is mostly due to the moisture loss. The second segment is a plateau, indicating that, in this temperature range, the bismuth oxide was mostly in a liquid phase with little volatilization. When temperature reaches 1,100℃ and above, i.e., in the third segment of the TGA curves, a rapid mass loss occurs. To make the analysis more comprehensive, we plan to adjust the experimental parameters to prevent the irregularity of the curve of sample S2 to ensure the accuracy of the conclusions. However, it should be noted that the irregularity of the curve of sample S2 did not directly affect our conclusion that the effect of Bi2O3 doping on the relative density of 3YSZ ceramics is bad when the temperature is above 1,100℃. To reduce the volatilization of Bi2O3, the mixed YSZ and Bi2O3 powder can be used in the sintering furnace to control the atmosphere of Bi2O3, which can inhibit the volatilization of Bi2O3 in the sample.

3.2 Effect of Bi2O3 doping on the grain size 10

The grain size was calculated by the rectangle intercept method. All the samples had a mean grain size between 200 and 300 nm, with the largest observed grains being 1.2μm. The grain size increased slightly with the increase of the Bi2O3 concentration according to the SEM micrographs presented in Fig. 5 and Fig. 6. However, a clear relationship between grain size and the amount of Bi2O3 was not found, which is consistent with the result reported by Gil et al.[25] At 850℃, the largest grain size observed was 0.9 μm, while at 900℃ and 1,000℃, the maximum grain size was found to be 1 μm and 1.2 μm, respectively. Thus, the main factor determining the grain size of 3YSZ ceramics is the sintering temperature.

3.3 Composition of the phases

Bi2O3 changed the amount and crystallinity of the monoclinic and tetragonal phases in the 3YSZ ceramics. In other words, it altered the stability of tetragonal zirconia ceramics. The XRD pattern of the S1 sample sintered at 1150℃ with a high crystallinity of the tetragonal phase is depicted in Fig.7(a). According to equation (1), the mass fraction of the monoclinic and tetragonal phases in S1 was 11 wt% and 89 wt%, respectively. Compared with that of the tetragonal phase in the initial 3YSZ powder (64.9wt%), the ratio of the tetragonal phase increased, as inferred from the stronger and sharper XRD peak of the tetragonal phase. Some peaks of the monoclinic phase detected in the 3YSZ powders disappeared in the XRD pattern of the S1 ceramic, possibly due to a grain rearrangement during the sintering process. In particular, this was due to the substitution of the smaller Zr4+ ions (0.072 nm) by the 11

larger Bi3+ ions (0.117 nm) in ZrO2, which increased the average lattice constant, and as a result stabilized the tetragonal phase and improved its crystallinity at room temperature. Such a chemical substitution can be described by the following equation:

2 .. x Bi 2O3 ZrO   2 BiZr  Vo  3O o

(2)

However, when the amount of Bi2O3 was larger than or equal to 2mol%, the opposite trend occurred, as shown in Fig. 7(b). The relative intensity of the XRD peak of the tetragonal phase was reduced, indicating a decrease in the mass fraction of the tetragonal phase. This implies that the transformation from the tetragonal to the monoclinic phase began when the doping amount of Bi2O3 exceeded 2%. Also, the crystallinity of the monoclinic phase was improved with its increasing ratio; Kim et al. noticed the same phenomenon while they were studying the effect of the dopant on the property of the YSZ material.[28]

The X-ray diffraction patterns of the t-ZrO2 phase in the 3YSZ powder and samples S1-S5 sintered at 1,000℃ are presented in Fig. 8. The XRD peaks of free Bi2O3 or a secondary phase were not present in Fig. 8(a), which implies that the Bi2O3 had completely dissolved in the ZrO2. The XRD peaks of the t-ZrO2 phase in sample S1 were slightly shifted to smaller angles compared with those of the 3YSZ powder, indicating an increase in the average lattice constant with the addition of Bi2O3. These observations are consistent with Eq. (2). The transition from the tetragonal to the monoclinic phase can be attributed to

12

two possible causes. One is the dissolution of Y2O3 in Bi2O3, according to the phase diagram of Y2O3-Bi2O3 [29]. The formation of a Bi2O3-Y2O3 solid solution will induce the segregation of Y2O3 from the ZrO2 [30], thus destabilizing the tetragonal phase.

An XRD peak of the solid solution of Bi2O3-Y2O3 emerged in the sample doped with 5% Bi2O3 (S5), as seen in Fig. 8(b), while the cubic Bi2O3 phase peak showed up in Fig. 9 for samples with more than 4mol% Bi2O3. When sintered at 1,000℃, the solid solution limit of Bi2O3 is about 2 mol%, which can be seen in Fig. 10. When the Bi2O3 content reaches 3mol%, the diffraction peak of β-Bi2O3 in the XRD pattern of sample S3 sintered at 1,000℃ emerged. This indicates that the Bi2O3 has not completely dissolved into the YSZ lattice, and a possible explanation for this phenomenon is that interfacial reactions lead to re-precipitation of Bi2O3 and then Bi2O3 is stabilized to room temperature by Y3+ in the cooling process.[31, 32] Therefore, when the content of Bi2O3 exceeded 2mol%, Bi2O3 could not be completely dissolved into ZrO2 lattice, which is consistent with the result shown in Fig. 8. When the amount of Bi2O3 exceeded 2mol%, Y2O3 dissolved in free Bi2O3 and formed a new solid solution. Specifically, the solid solutions in our samples include Y2O3-ZrO2, Bi2O3-ZrO2, Bi2O3-Y2O3 and Bi2O3-Y2O3-ZrO2.

An interface reaction between Bi2O3 and ZrO2 could be another possible reason for the destabilization of the tetragonal zirconia. Gulino et al. found that when the sintering temperature was above 800℃, Bi2O3 particles were precipitated near the grain boundaries and Bi3+ ions were segregated from the ZrO2 matrix.[33] 13

Consequently, the thermodynamically stable m-ZrO2 stayed steady until room temperature. In addition, in a study of Cr2O3 doped with ZrO2, Sohn et al. found that t-ZrO2 could transform into m-ZrO2 at a certain temperature. They attributed this transformation to the strong interfacial interaction between the Cr2O3 and ZrO2, which prevented the oxygen in the air from diffusing into the ZrO2 lattice. As a result, the transformation from t-ZrO2 to m-ZrO2 occurred during the cooling process. Here, we propose that the doped Bi2O3 played the same role as the “interface inhibitor” [34, 35].

The calculated mass fractions of the tetragonal zirconia in the samples sintered at 1,000℃ as a function of the Bi2O3 concentration are presented in Fig. 11. When the content of Bi2O3 was less than 1mol%, the ratio of the tetragonal phase increased in 3YSZ，as a result of the low amount of Bi2O3. which completely dissolved in ZrO2. At the same time, the smaller Zr4+ ions were replaced by the larger Bi3+ ions, thus increasing the average lattice constant and improving the crystallinity of the tetragonal phase at room temperature. The mass fraction of the tetragonal phase showed negligible changes (remaining low at ~30%) after the concentration of Bi2O3 exceeded 2mol%. Again, this can be explained by the shift of the stability regions in the Bi2O3-Y2O3-ZrO2 ternary solid solution system, which destabilized the tetragonal zirconia. With the gradual increase of the bismuth oxide content, the amount is still very small, and thus the content of yttrium oxide is very little, which causes the amount of yttria dissolved in bismuth oxide and zirconium oxide to remain basically unchanged. Accordingly, the mass fraction of the ceramic remains basically stable as 14

the content of bismuth oxide increases. It should be noted however that the phase transition of t-ZrO2 to m-ZrO2 occurred at about 1,100℃. In other words, the transformation from the tetragonal to monoclinic phase in our samples was triggered by the addition of Bi2O3, as well as the ZrO2 own thermal characteristics . Another problem that needs to be pointed out is that the mechanical strength of the ceramic will be reduced to some extent with the decrease of the mass fraction of the tetragonal phase.[36,37] Fine grain and dispersion strengthening might be used to address this situation. However, these approaches require further experimentation to validate.

4.Conclusion This work showed that doping of 3YSZ with Bi2O3 can effectively reduce the sintering temperature of 3YSZ ceramics. When the doping amount of Bi2O3 was 3mol%, the 3YSZ ceramics can be sintered at a temperature of 1,000℃ with a relative density up to 96.5%. The main factor determining the grain size of the 3YSZ ceramics is the sintering temperature. Furthermore, the addition of Bi2O3 changed the stability and crystallinity of the YSZ phases. When the doping amount was 1mol%, the concentration of the tetragonal zirconia increased by more than 20% compared with the raw 3YSZ powder. When the content of Bi2O3 was greater than or equal to 2mol%, the ratio and crystallinity of the tetragonal phase was decreased. This can be attributed to the segregation of Y2O3 from the zirconia matrix (to be dissolved in the excessive Bi2O3), leading to the destabilization of the tetragonal zirconia.

Acknowledgements 15

The work described in this paper is supported by The Major Program of the National Natural Science Foundation of China (No.50942024). The authors would like to express their deep gratitude to Prof. Jun Ouyang (School of Materials Science and Engineering, Shandong University) and Yofre C. (Professional Editor, University of Portsmouth, MATERIALS SCIENCE) for proofreading the revised manuscript.

Reference [1]Panthi D, Tsutsumi A. Micro-tubular solid oxide fuel cell based on a porous yttria-stabilized zirconia support[J]. Scientific reports, , 4(2014)5754. [2]. Guo H, Bayer T J M, Guo J, et al. Cold sintering process for 8mol% Y2O3-stabilized ZrO2 ceramics[J]. Journal of the European Ceramic Society, 37(2017)2303-2308. [3]. Timurkutluk B, Timurkutluk C, Mat M D, et al. A review on cell/stack designs for high performance solid oxide fuel cells[J]. Renewable and Sustainable Energy Reviews, 56(2016)1101-1121. [4]. Zhou L, Cheng M, Yi B, et al. Performance of an anode-supported tubular solid oxide fuel cell (SOFC) under pressurized conditions[J]. Electrochimica Acta, 16

53(2008)5195-5198. [5]. Chen Y, Zhang Y, Baker J, et al. Hierarchically oriented macroporous anode-supported solid oxide fuel cell with thin ceria electrolyte film[J]. ACS applied materials & interfaces, 6(2014)5130-5136. [6]. Zhang Lan, et al. "NiO/YSZ, anode-supported, thin-electrolyte, solid oxide fuel cells fabricated by gel casting." Journal of Power Sources,170 (2007)55-60. [7]. Han Min fang, Tang Xiu ling，Che Ling zhi, Low temperature sintering process of dense YSZ electrolyte films. Journal of Materials Science and Engineering ， 1(2007)34-39. [8]. Duran P., Vilegas M., Capel F. and Moure C., Low-temperature sintering and microstructural development of nanocrystalline Y-TZP powder. Journal of European Ceramic Society, 16(1996)945–952. [9]. Theunissen G.S.A.M.,Winnubst A.J.A., Burggraaf A.J. , Sintering kinetics and microstructure development of nanoscale Y-TZP ceramics. Journal of the European Ceramic Society, 11(1993)315-324. [10]. Du J, Sanders A P, Jindal V, et al. Rapid in situ formation and densification of titanium boride (TiB) nano-ceramic via transient liquid phase in electric field activated sintering[J]. Scripta Materialia, 123(2016)95-99. [11]. Ivers-Tiff é e Ellen, Andr é Weber, and Dirk Herbstritt. "Materials and technologies for SOFC-components." Journal of the European Ceramic Society, 21

17

(2001)1805-1811. [12]. Fergus, Jeffrey, et al., eds. Solid oxide fuel cells: materials properties and performance. CRC press, 2016. [13]. Verkerk M J, Winnubst A J A, Burggraaf A J. Effect of impurities on sintering and conductivity of yttria-stabilized zirconia[J]. Journal of Materials Science, 17(1982) 3113-3122. [14]. Nawa M, Nakamoto S, Sekino T, Niihara K. Tough and strength Ce-TZP/alumina nanocomposites doped with titania. Ceram Int, 24(1998)497–506. [15]. Schneider J, Begand S, Kriegel R, Kaps C, Glien W, Oberbach T. Low temperature aging behavior of alumina–toughened zirconia. J Am Ceram Soc, 91(2008)3613–3618. [16]. Shah K, Holloway AJ, Denry LI. Effect of coloring with various metal oxides on the microstructure, color, and flexural strength of 3Y-TZP. J Biomed Mater Res Part B Appl Biomater, 87(2008)329-337. [17]. Nakajima H, Itoh K, Kaneko H, Tamaura Y. Effect of Fe doping on the crystalline and optical properties of yttria-stabilized zirconia. J Phys Chem Solids, 68(2007)1946-1950. [18]. Matsui K, Yamakawa T, Uehara M, Enomoto N, Hojo J. Mechanism of alumina-enhanced sintering of fine zirconia powder: influence of alumina concentration on the initial stage sintering. J Am Ceram Soc ,91(2008)1888-1897.

18

[19]. Dey T, Sharma A D, Dutta A, et al. Transition metal-doped yttria stabilized zirconia for low temperature processing of planar anode-supported solid oxide fuel cell[J]. Journal of Alloys and Compounds, 604(2014)151-156. [20]. Hbaieb K. Reducing sintering temperature of yttria stabilized zirconia through addition

of

lithium

nitrate

and

alumina

[J].

Ceramics

International,

38(2012)4159-4164. [21]. López-Honorato E, Dessoliers M, Shapiro I P, et al. Improvements to the sintering of yttria-stabilized zirconia by the addition of Ni [J]. Ceramics International, 38(2012)6777-6782. [22]. De Marco V, Sglavo V M. Effect of Bismuth Oxide as Sintering Aid for Gadolinia-doped Ceria at 1050°C[J]. ECS Transactions, 68(2015)413-420. [23]. Le S, Zhang J, Zhu X, et al. Sintering and electrochemical performance of Y2O3-doped barium zirconate with Bi2O3 as sintering aids[J]. Journal of Power Sources,232(2013)219-223. [24]. Levin E M, McMurdie H F. Phase diagrams for ceramists, 1975 supplement[J]. 1975. [25]. Gil. V, Tartaj. J, Moure. C, Duran. P., Effect of Bi2O3 addition on the sintering and microstructural development of gadolinia-doped ceria ceramics. Journal of the European Ceramic Society, 27(2007)801–805. [26]. Wang Y. P., Zhou. L, Zhang, M. F, etc., Room-temperature saturated ferroelectric

19

polarization in BiFeO3 ceramics synthesized by rapid liquid phase sintering. Applied physics letters, 84(2004)1731-1733. [27]. Kohama K, Ito K. Direct solid-state diffusion bonding of zirconium carbide using a spark plasma sintering system[J]. Materials & Design,110(2016)888-894. [28]. Kim YH，Kim H G．The effect of Bi203 on the microstructure and electrical conductivity of ZrO2-Y2O3 ceramics[J]. Journal of Materials Science: Materials in Electronics, 5(1994)260-266． [29]. Akiteru Watanbae. Phase equilibria in the system Bi2O3-Y2O3: no possibility of δ-Bi2O3 stabilization[J]. Solid State Ionics, 86-88(1996)1427-1430． [30]. Watanabe A, Kikuchi T. Cubic-Hexagonal Transformation of Yttria-Stabilized δ-Bismuth

Sesquixide

Bi2-2XY2XO3(x=0.215-0.235)[J].

Solid

State

Ionics,

21(1986)287-291. [31]. Poleti D, Karanovic L, Zdujic M, Jovalekic C. Phase composition of Bi2O3 specimens doped with Ti, Zr and Hf[J]. Journal of the Serbian Chemical Society,77(2012)1091-1096. [32]. Šulcová P, Trojan M. Thermal analysis of the Bi2O3-Y2O3-ZrO2 pigments[J]. J Therm Anal Calorim, 293(2008)795-798. [33]. Gulino A, La Delfa S, Fragala I, Egdell RG. Low-temperature stabilization of tetragonal zirconia by bismuth[J]. Chemistry of Materials, 8(1996)1287-1291. [34]. Sohn J, Ryu S, Park M, Pae Y. Preparation and characterization of chromium 20

oxide supported on zirconia[J]. journal of materials science, 28(1993)4651-4659. [35]. Štefanić G, Musić S, Gajović A. Thermal behavior of the amorphous precursors in the ZrO2–CrO1.5 system[J]. Journal of Molecular Structure, 744(2005)541-549. [36]. Bocanegra-Bernal, M. H., & De La Torre, S. D. Phase transitions in zirconium dioxide and related materials for high performance engineering ceramics. Journal of materials science, 37(2002)4947-4971. [37]. Hannink, Richard HJ, Patrick M. Kelly, and Barry C. Muddle. "Transformation toughening in zirconia‐ containing ceramics." Journal of the American Ceramic Society 83 (2000) 461-487.

Figure captions Figure 1. The effect of the Bi2O3 content on the relative density of 3YSZ ceramics at different temperature. (a) the global tendency; (b) local magnification. Figure 2. SEM micrographs of the 3YSZ ceramics sintered at 1,000℃. (a) doped with 4mol%; (b) doped with 6mol% Bi2O3 (S4 and S6) Figure 3. Relative density of the ceramic samples with 2mol%, 3mol% and 4mol% Bi2O3 versus the sintering temperature. Figure 4. The TGA curves for samples doped with 1mol%, 2mol%, 3mol% and 4mol% Bi2O3 from 25 -1200℃. Figure 5. SEM micrographs of the polished and thermally etched surfaces of Bi2O3-3YSZ composites sintered at 1,000℃. (a) samples with 2mol% Bi2O3; (b) 21

samples with 4mol% Bi2O3; (c) samples with 6mol% Bi2O3; (d) samples with 8mol% Bi2O3. Figure 6. SEM micrographs of the polished and thermally etched surfaces of Bi2O3-3YSZ composites sintered at 850℃. (a) samples with 9% Bi2O3; (b) samples with 10% Bi2O3. Figure 7. X-ray diffraction patterns of (a) the 3YSZ powder and Sample with 1mol% Bi2O3; and (b) the 3YSZ powder and sample with 2mol% Bi2O3 both sintered at 1,150℃. Figure 8. X-ray diffraction patterns of: (a) 3YSZ powder and S1 (Samples with 1mol% Bi2O3) sintered at 1,000℃; (b), S2, S3, S4 and S5 (Samples with 2, 3, 4 and 5mol% Bi2O3, respectively), sintered at 1,000℃ from 28.5° to 31.0°. Figure 9. X-ray diffraction patterns of S4-S7 (Samples with 4, 5, 6 and 7mol% Bi2O3, respectively) composites sintered at 1,000℃ from 31.0° to 33.5° Figure 10. X-ray diffraction patterns of S2 (Samples with 2mol% Bi2O3) and S3 (Samples with 3mol% Bi2O3) after both were sintered at 1,000℃,. Figure 11. The mass fraction of the tetragonal phase on 3YSZ versus the amount of Bi2O3 doped (0-7mol%)

22

fig 1

23

fig 2

24

fig 3

25

fig 4

26

fig 5

27

fig 6

28

fig 7

29

fig 8

30

fig 9

31

fig 10

32

fig 11

33