Microstructural Arrangement and Densification of 10GDC-xTZP Solid Solutions

Available online at www.sciencedirect.com

Energy Procedia 34 (2013) 506 – 516

10th Eco-Energy and Materials Science and Engineering (EMSES2012)

Microstructural Arrangement and Densification of 10GDC-xTZP Solid Solutions Suthee Wattanasiriwech*, Tanongsak Jindapetch, and Darunee Wattanasiriwech School of Science, Mae FahLuang University, Chiang Rai, Thailand 57100 Abstract Solid solutions between tetragonal zirconia polycrystals (TZP) and gadolinia doped ceria with a norminal composition Ce0.9G0.1O1.95(10GDC) powders were studied. The selected composition was 5-15 wt% TZP. The XRD results showed that a complete solid solution between TZP and 10GDC had occurred after sintering at 1500-1600 C, resulting in a progressive decrease of the cubic lattice parameters of the 10GDC. At 1500 C sintering, when the TZP content was increased, porosity was also increased while grain growth was suppressed. However, when the sintering temperature was changed to 1600 C, a rapid grain growth with ananisotropic grain morphologyfor the samples with 10 and 15wt% TZPwas obtained. Vickers hardness was slightly improved with increasing TZP contents when there was no effect of porosity. Fracture toughness was, however, slightly decreased with increasing TZP content. Unfavourable flexural strength was always obtained in all samples mainly due to the anomalous microstructure. The ionic conductivity was not only governed by microstructure but also by lattice deformation due to the solid solution.

2013 The Authors.Published by Elsevier ©©2013 The Authors. Published by Elsevier B.V. B.V. Selection peer-review under responsibility of COE ofEnergy Sustainable System, Rajamangala Selection andand/or peer-review under responsibility of COE of Sustainalble System,Energy Rajamangala University of Technology University(RMUTT) of Technology Thanyaburi(RMUTT) Thanyaburi Keywords: solid oxide fuel cells; tetragonal zirconia polycrystals; gadolinia doped ceria; ionic conductivity

* Corresponding author. E-mail address: [email protected]

1876-6102 © 2013 The Authors. Published by Elsevier B.V.

Selection and peer-review under responsibility of COE of Sustainalble Energy System, Rajamangala University of Technology Thanyaburi (RMUTT) doi:10.1016/j.egypro.2013.06.779

Suthee Wattanasiriwech et al. / Energy Procedia 34 (2013) 506 – 516

1. Introduction Solid Oxide Fuel Cells with cubic stabilized zirconia (CSZ) electrolyte are promising energy generation sources for the next generation due to their environmental friendliness and high efficiencies [1-4]. In this model, the cells must be operated at the temperature as high as 800-1000 C so the use of expensive high temperature alloy as an interconnector is required. In recent years, ceria based ceramics doped with aliovalentcations such as Gd3+, Sm3+, Y3+ have been received much attention according to their superior ionic conductivity to conventional CSZ electrolyte. The use of ceria based electrolytes could lead to the reduction of the operating temperatures of the cells to 500-700 C so conventional stainless steel could be used. Among these doping cations, gadolinium ion (Gd 3+) was reported to be one of the most interesting candidates due to its close lattice matching with Ce4+and thus the least lattice constraint. However, there were two significant problems associated with the use of ceria based electrolytes. The first problem was the high reducibility of ceria (Ce 4+ to Ce3+) which could result in lattice expansion and an electronic conduction causing an electronic leak current between anode and cathode[5,6]. The electronic leakage could lead to internal short-circuiting while the lattice expansion could contribute to mechanical stresses. In addition to the electrical problems, mechanical properties of ceria based electrolyte were significantly inferior to those of cubic zirconia. Bending strength of the sintered Ce0.9G0.1O1.95 (10GDC) fabricated from precipitated powders was found to be 143 10 MPa at room temperature and only 115 12 MPa at 800 C. Fracture toughness of this material was only 1.1 MPa.m1/2 [7]. The use of double layer thin film CSZ/GDC was proposed to solve the problem with an electronic conduction. However, the co-firing of such configuration was suffered from serious interdiffusion of the materials which led to delaminating of the films. The use of the graded composition Ce0.8Gd0.2O1.9/CSZ or (20GDC)x(CSZ)1-x in attempts to avoid extended interdiffusion between CSZ and 20GDC was found to give much lower ionic conductivity than both CSZ and 20GDC due to the large lattice deformation and scattering of oxygen ions[6]. The author thus suggested that the ceria-zirconia system with lower trivalent cation contents (Gd3+ and Y3+) would be an interesting alternative candidate. An effort to improve the mechanical properties of 20GDC by compositing it with 1-20 wt% TZP particles gave rise to poorer in both flexural strength and ionic conductivity[6]. It was considered that the TZP granules acted as regions where cracks could initiate. The porosity of up to 10% according to the presence of large voids in the centre of the zirconia granules indicated incomplete sintering. The above study had demonstrated that addition of the spray dried zirconia granules to the 20GDC matrix deteriorated both mechanical and electrical properties. From the suggestion to use lower trivalent cations in the ceria-zirconia system and from the drawback of incorporation of agglomerated TZP particles to 20GDC[6], the present study was thus aimed to explore the use of Ce0.9 Gd0.1O1.95 mixed with TZP of the same particle size range. Microstructure, phase, mechanical and electrical properties were addressed. 2. Experiments Ce0.9G0.1O1.95 (10GDC) was mixed with 0-15 wt% TZP using a mixed oxide method. The samples were coded as 10GDC-xTZP where x was the weight content of TZP. TZP powder was a commercial powder with 3 mole% yttria and a specific surface area of 16 3 m2/g (TZ-3Y; TOSOH, Japan). Particle size of the powder and the sintered samples were calculated from either the specific surface area (Eqn. 1) or the X-ray spectrum according to Scherrer(Eqn. 2):

507

508

Suthee Wattanasiriwech et al. / Energy Procedia 34 (2013) 506 – 516

d = 6/ .As (1) where As is the specific surface area (m2/g), (6.05 g/cm3) is the density value of TZP provided by the supplier and d is the particle diameter. (2) D = 0.9 /( cos ) where is the wavelength of the X-rays (1.541 Å), is the diffraction angle and is the calibrated width of the diffracted peak at half-maximum at the selected 2 . Particle size of the TZP powder calculated from its specific surface area was in the range of 50-80 nm [8].The ceria (CeO2) powder was supplied by Inframat (USA) and the particle size calculated from its specific surface area given by the manufacturer was also between 50-80 nm. The calculation from its XRD spectrum, however, gave the value 26 nm suggesting some degree of agglomerations of the original powder.Gadolinia powder (Gd2O3) was supplied by Metal Rare Earth (China) and the average crystallite size calculated from its XRD spectrum was 39 nm. The powders were mixed in a vibrational mill for 15 minutes using TZP balls. The mixed powder was then dried and sieved through a 200-mesh stainless steel sieve prior to a uniaxial press at 100 MPa. The compacted powder was subsequently sintered with a heating rate of 5 C/min at ambient to 1500-1600 C for 5 h. Phase identification was angle from 20 to 80 2 using step scan, with a 2 sampling interval of 0.01 . Microstructure was examined using SEM (LEO 1450 VP (Zeiss), Germany). Relative density was measured using an Archimedes method. Theoretical density of each set of samples was calculated according to Eqn. 3 [9]: Dth

4[0.666M Ce

0.074M Gd

0.256M Z r a 3 .N A

0.004M Y

1.833M O ]

(3)

whereMCe, MGd, MZr, MY, MO = atomic mass of cerium, gadolinium, zirconium, yttrium and oxygen a is lattice parameter and NA Grain size was measured by the line intercept method. The average value was obtained from 3 SEM micrographs in which each micrograph contained at least 100 grains. Ionic conductivity was measured using an impedance analyser (Solartron 1260, gain/phase analyzer). AC impedance measurements were carried out in air at temperatures in the range 250-600 C on specimens with silver paste electrodes. The selected frequency range was from 1Hz-1MHz at the voltage of 1 mV. Activation energy for conduction, Ea, was calculated by fitting data to the normal Arrhenius relation for thermally activation conduction: .e (-

Ea ) kT

(4)

whereEa is the activation energy for conduction, T the absolute temperature,k the Boltzmann constant and 0 the pre-exponential factor. To evaluate the biaxial flexural strength value, the ball-on-ring test was employed. The ball-on ring tests were achieved using a biaxial-flexure jig. Pellet shaped-samples with the thickness between 1.7 and 2.2 mm and the diameter between 11 and 15 mm were supported on a cylindrical ring with a diameter of 10 mm and centrally loaded with a steel ball. Three pieces of steel rods stand vertically to the base plate, equally away from the centre of the ring and equally apart from one another. All tests were performed at room temperature using an Instron hydraulic testing machine in the compressive mode. The crosshead speed of 0.5 mm/min was chosen for all samples. For each measurement, ten samples were tested to determine the average flexural strength and the maximum stress was calculated according to Kirstein[10].Hardness measurement was performed using Vickers micro-indentation (Micro-hardness tester, MMT3, Matsuzawa Co., LTD, Japan) under the load of 2903 mN for 10 seconds. At least 5 indentations were use for obtain the mean and standard deviation values of hardness and fracture

Suthee Wattanasiriwech et al. / Energy Procedia 34 (2013) 506 – 516

toughness. The fracture toughness (Kc) values were calculated from the measurement of radial cracks around the Vickers indentation, according to Eqn. 5 [11]. (5) KC 0.0016( E / HV )1 / 2 ( P / C1.5 ) Where C was the half crack length, P was the load (500 g) and E was the Young modulus which was 205 GPa for the entire toughness calculations. 3. Results and discussion 3.1. Properties of the starting materials Particle size of the TZP powder calculated from its specific surface area was in the range of 50-80 nm [9]. The ceria (CeO2) powder was supplied by Inframat (USA) and the particle size calculated from its specific surface area given by the manufacturer was also between 50-80 nm. The calculation from its XRD spectrum, however, gave the value 26 nm suggesting some degree of agglomerations of the original powder. 3.2 .Phase development and microstructural changes All samples were crystallised in a cubic fluorite structure with an upward shift to the higher 2 side when TZP was incorporated (Fig 1). A linear increase of lattice parameters as a function of TZP (Fig 2). content calculated

20TZP

Intensity (A.U.)

15TZP

10TZP 5TZP

0TZP

2 Fig 1. XRD spectrum for the 10GDC -xTZPsamples after sintering at 1500 C for 5 h.

Fig. 3 and 4 present SEM micrographs of the 10GDC samples with various TZP contents sintered at 1500 C-1600 C. At 1500 C, the 10GDC -0TZP sample had densified with grain growth and only small amount of porosity was observed. It was reported that nanocrystalline ceramics frequently suffered from grain growth even at low sintering temperatures[14]. However, in the 10GDC-5TZP samples, grain growth was slightly suppressed and higher degree of porosity could be observed. Increasing TZP contents to 10 wt% and finally to 15 wt% resulted in a highly porous microstructure of

509

510

Suthee Wattanasiriwech et al. / Energy Procedia 34 (2013) 506 – 516

smaller grain size. A similar trend was also observed when the sintering temperature was changed to 1550 C. When the temperature was increased to 1600 C, microstructure of the samples was totally different from that of the samples sintered at 1500 C and 1550 C. All samples were fully densified and grain growth was generally observed. It was noted that grain size of the 10GDC-10TZP and 10GDC-15TZP samples became considerably larger than those of 10GDC and GDC-5TZP and the unusual grain development was generally observable.

Lattice parameter (Å Å) Lattice parameter(A)

5.43 5.42 5.41 5.40 5.39 5.38 5.37 5.36 5.35 -5

0

5

10

15

20

25

%wt TZP

Fig 2. Cubic lattice parameters as a function of TZP content for the 10GDC -xTZPsamples sintered at 1500 C for 5 h. 5%TZP 0%TZP 0TZP

5TZP

2 m 2 m

2 m 10TZP

2 m

20TZP

2 m

2 m

2 m

Fig 3. Effects of TZP content on microstructure after sintering at 1500 Cfor 5 h

Suthee Wattanasiriwech et al. / Energy Procedia 34 (2013) 506 – 516

5TZP

0TZP

2 m

2 m 10TZP

20TZP

2 m

2 m

Fig 4. Effects of TZP content on microstructure after sintering at 1600 Cfor 5 h

To interpret the result better, grain size and density of the samples with different TZP contents were plotted against the sintering temperature and the result was graphically shown in Fig. 5. Mobility and the driving force for grain growth are the decrease in free energy that accompanies the reduction in total grain boundary area[15]. When the system is a solid solution, apart from a few well developed one, the role of dopant is not yet understood very well. A dopant may change defect chemistry of the host lattice and thus change the diffusion coefficient or it may segregate and thus alter the structure and composition of surfaces and interfaces. The role of TZP to densification and grain growth of GDC is not yet clear at this stage. 100

Relative density (%)

96

50 0TZP 5TZP 10TZP 15TZP 20TZP

45 40 35

94

30

92

25 90

20

88

Grain size (μm)

98

15

86

10

84

5

82

0 1500

1550

1600

Temperture (°C)

Fig 5. Effects of sintering temperature on relative density and grain size of the samples with various TZP contents.

511

512

Suthee Wattanasiriwech et al. / Energy Procedia 34 (2013) 506 – 516

3.3. Mechanical properties Vickers indentation hardness, indentation fracture toughness and biaxial flexural strength of the samples with different TZP contents and sintering conditions are graphically plotted in Fig. 6. At 15001550 C, the hardness decreased with increasing TZP contents, possibly due to the increased porosity in these samples. When the sintering temperature was increased to 1600 C, where the samples were fully densified, the opposite trend was observed, i.e.Vickers hardness increased with increasing TZP contents and grain size. It has been acknowledged that hardness of materials can be in controlled by the type and strength of chemical bonding and also by microstructural arrangement. Materials with high percentage of strong covalent character are harder than materials with high ionicity. Hardness was found to increase with decreasing grain size when there was no effect of different compositions or bond strength[16]. In this study, 10GDC was alloyed with TZP in which Zr4+ was smaller in size but more electronegative so the ionic bond became shorter, the covalent character became more dominant and the bond strength thus became stronger. When the materials were densified, the harness was thus increased with increasing TZP content. Fracture toughness (Kc) measurement was performed only in the samples sintered at 1600 C since highly porous microstructure in the samples sintered at lower temperatures gave uncertainty in the measurement.Increasing x from 5 to 10, the fracture toughness deteriorated progressively before becoming unchanged at x = 20.Zhang et al. [17] reported the independence of fracture toughness of 10GDC upon grain size while Fu et al. reported the improvement of fracture toughness with the decrease of grain size in Y2O3 doped CeO2[11]. Since the presence of TZP promoted an unusualmicrostructure of 10GDC, Kc of the samples became poorer similar to the reported results [12]. It was shown that when TZP was dissolved in the host lattices, its high fracture toughness nature also disappeared. 3.4 .Ionic conductivity Table 1 shows the activation energy measured at 250-450 C range. Fig. 7 shows the variation of ionic conductivity, measured at 350 C, as a function of TZP content and sintering temperature. Both grain interior and grain boundary ionic conductivity dramatically decreased, with a greater decrease in the grain interior, when only 5 wt% TZP was introduced to 10GDC. A smaller decrease was observed with further increaseing of TZP contents. Badwal et al. [18] reported that the grain interior and grain boundary conductivity of the same TZP powder (TZ-3Y; TOSOH, Japan) were respectively 4.6 10-5 and 0.031 10-5 ohm-1 cm-1 and their respective activation energy were 89.9 kJ/mol or 0.93 eV and 103.6 kJ/mol or 1.07 eV. It was noted that the electrical properties obtained in our studies were even inferior to those of pure TZP. Increasing TZP content also resulted in increases of activation energies. When CSZ was doped to Ce0.8G0.2O1.90, an adverse effect on the ionic conductivity due to the large lattice deformation and scattering of oxygen ions was also observed[8]. Luo et al., however, reported that the solid solution between 10GDC and TZP when 10-50wt% 10GDC was doped to TZP could be the source of unfavorable ionic conductivity[19].

513

Suthee Wattanasiriwech et al. / Energy Procedia 34 (2013) 506 – 516

1/2

)

9

Fracture toughness (MPa m

Hardness (GPa)

(a)

7

5

1500 1550 1600

3 -5

0

5

10

15

20

TZP Content (%wt)

2.00

(b) 1.50

1.40

1.38

1.21

1.20 1.00

1600 0.50 0

5

10

25

15

20

TZP conent (%wt)

Flexural Strength (MPa)

190 (c)

170 150 130 110 1500 1550 1600

90 70 50 -5

0

5 10 TZP Content (%wt)

15

20

25

Fig 6.(a) Harness, (b), fracture toughness and flexural strength (c) of the samples sintered at 1500-1600 C for 5 h as a function TZP content Table 1. Activation energy values of the samples with different TZP contents measured at 250-450 C.

Sample/sintering temperature ( C) 0TZP/1500 5TZP/1500 10TZP/1500 15TZP/1500 0TZP/1600 5TZP/1600 10TZP/1600 20TZP/1600

(b )

Ea (eV) Bulk 0.93 0.96 1.04 1.10 0.81 0.93 0.98 1.10

GB 0.98 1.13 1.19 1.05 0.95 1.07 1.15 1.12

25

514

Suthee Wattanasiriwech et al. / Energy Procedia 34 (2013) 506 – 516

In the present study when 10GDC was sintered with 5-15% TZP, a complete solid solution also resulted. Lattice parameter became smaller indicating that 10GDC was under compression. This solid solution obviously affected the densification and grain growth of the host 10GDC. At low sintering temperatures, it hindered densification and grain growth. The reverse trend was observed when the sintering temperature was increased to 1600 C. Grain growth with abnormal grain shape was promoted in the samples with higher TZP contents. This result could also affect mechanical properties of 10GDC. Biaxial flexural strength decreased with increasing TZP contents suggesting that the effects of microstructure played an important role. Ionic conductivity of zirconia-ceria based electrolytes was mostly related to dopant, composition, microstructure and impurity[20]. AC impedance spectroscopy in the present study showed that addition of TZP to 10GDC had an adverse effect on both bulk and grain boundary conductivity. The greatest decrease was observed in the bulk conductivity especially in the 10GDC-5TZP samples sintered at 15001550 C ( 30 times lower) which did not exhibit much different microstructure from the reference 10GDC. So it was not likely that the conductivity was determined only by the microstructure related feature. The deterioration of both bulk and grain boundary suggested that dissolution of TZP into the 10GDC could adjust the feature of both parts. Thework based on HRTEM study of TiO2 and GeO2 doped TZP solid solution showed that segregation of the doping cations on grain boundaries actually occurred although SEM and XRD results showed only single phase solid solution. This segregation could retard the diffusion of oxygen anions and thus lowered ionic conductivity but increased activation energy in both grain and grain boundary parts [21]. Miyazaki, however, explained that the lowered ionic conductivity found in the TiO2 doped-TZP was due to the decrease of oxygen vacancy contents in TZP when solid solution was increased[22]. However, the above explanation was not able to explain the decrease of ionic conductivity in 10GDC doped TZP which should possess higher oxygen vacancy contents than the undoped TZP. Another study on the effect of localstructure of doped ceria on ionic conductivity using the density function theory (DFC) calculation showed that the local structure around each dopant and Ce 4+ as well as lattice deformation were important factors for ionic conduction in doped ceria [23]. Since our study showed the progressive decrease of lattice parameters (Fig. 2) and ionic conductivity (Fig. 7)with increasing TZP contents, it was thus possible that lattice deformation was another factor that deteriorated the ionic conduction process. This conclusion is in good agreement with the work done by F.-Y. Wang et al. who reported that the Ce0.85Gd0.1Sm0.5O1.925 had higher ionic conductivity than the Ce0.85Gd0.5Sm0.1O1.925 because the former had closer lattice constant to CeO2 than the latter[24].

0.002 -0.722

-0.084

1500

-1.689 -3.166

-3.026

-2.429

-2.457

-1.556

-1.547 -3.099

-3

-1

-1

ohm .cm )

0.50 0.00 -0.50 -1.00 -1.50 -2.00 -2.50 -3.00 -3.50

-0.052

Suthee Wattanasiriwech et al. / Energy Procedia 34 (2013) 506 – 516

1550

0 TZP 5 TZP 10 TZP 15 TZP

(a)

1600

Sintering Temperature (°C)

0.50 log (x10 -3ohm-1.cm-1)

0.00

0 TZP 5 TZP 10 TZP 15 TZP

-2.50

-1.911 -2.037 -2.105 -2.263

-2.00

-1.617 -1.975 -2.287 -2.430

-1.50

-2.080 -2.449 -2.732

-1.00

-1.482

-0.50

(b) (b)

-3.00 -3.50 1500

1550

1600

Sintering Temperature (°C)

Fig 7. Ionic conductivity for (a) grain interior and (b) grain boundary of the 10GDC samples with various TZP content sintered at different temperatures for 5 h

4. Conclusion A complete solid solution of TZP in 10GDC lattices was observed in the present study. Such solid solution resulted in progressive decrease of lattice parameters of 10GDC with increasing TZP contents. It also retarded densification at low sintering temperature (1500 C) but promoted unusual grain development at higher sintering temperature (1600 C). Both features were proved to be harmful to fracture toughness, flexural strength as well as ionic conductivity of the 10GDC. However, the microstructural changes alone were not enough to explain the great decrease of ionic conductivity in the 10GDC-5TZP samples whose microstructure was least different from the undoped10GDC. Lattice

515

516

Suthee Wattanasiriwech et al. / Energy Procedia 34 (2013) 506 – 516

deformation due to the solid solution could be another key factor that governed the ionic conduction in the TZP doped-10GDC electrolyte. Acknowledgement The authors are highly appreciated the financial support from Mae FahLuang University, Thailand. References [1] Kudo T. and Obayashi, H.J. Oxygen Ion Conduction of the Fluorite-Type Ce1 xLnxO2 x/2 (Ln = Lanthanoid Element, Electrochem. Soc 1975; 122(1):142-47. [2] Pérez-Coll D, Nùñez P, Frade J.R, Abrantes, J.C.C. Conductivity of CGO and CSO ceramics obtained from freeze-dried precursors, Electrochimia Acta 2003; 48(11): 1551-57. [3] Hatchewell C, Sammes N.M, Brown, I.W.M. Fabrication and properties of Ce0.8Gd0.2O1.9 electrolyte-based tubular solid oxide fuel cells, Solid State Ionics 1999; 126(3-4): 201-08. [4] Mogensen M, Sammes N.M, Tompsett , G.A. Physical chemical and electrochemical properties of pure and doped ceria, Solid State Ionics 2000; 12: 63-94. [5] Yahiro H, Baba Y, Eguchi K. and Arai H. High Temperature Fuel Cell with Ceria-Yttria Solid Electrolyte Electrochem. Soc 1988; 135: 2077-80. [6] Ball R.J. and Stevens R. Characterisation of Ce0.8Gd0.2O1.9/3Y-TZP composite electrolytes effects of weight % 3Y-TZP particles, J. Mat. Sci 2003; 38 :1413-23. [7] Trunec M. Fabrication of zirconia- and ceria-based thin-wall tubes by thermoplastic extrusion, J. Euro. Ceram. Soc 2004; 24(4): 645-51. [8] Wattanasiriwech D and Wattanasiriwech S, Preparation and Phase Development of Yttria-Doped Ceria Coated TZP Powder, J. Mater. Sci. 2008; 43(19): 6473-79. [9] Ji-Guang L, Takayasu I. and Toshiyuki M. Low temperature processing of dense samarium-doped CeO2 ceramics: sintering and grain growth behaviours, ActaMaterialia 2004; 52(8): 2221-28. [10] Kirstein A. F. and Wooley R.M. Symmetrical bending of thin circular elastic plates on equally spaced point supports, J. Res. Natl. Bur. Stan. Sect. C 1967; 71(1): 1-10. [11] Fu Y.P. Ionic conductivity and mechanical properties of Y2O3-doped CeO2 ceramics synthesis by microwave-induced combustion,Ceram. Inter 2009 ;35 : 653-59. [12] Cullity B.D. Elements of X-ray Diffraction: 2nd ed., Addison-Wesley Publishing Company, INC., pp 1978; 327-30. [13] Mazaheri M, Simchi A, Dourandish M, Golestani-Fard F, Densification and Grain Growth of Nanocrystalline 3Y-TZP during two-step sintering, J. Euro. Ceram. Soc 2008; 28(15) :2933-39. [14] Rahaman N.M. Sintering of Ceramics, CRC Press, USA, pp 2008; 130. [15] Tadokoro S. K. and Muccillo, E. N. S. Physical characteristics and sintering behavior of ultrafine zirconia ceria powders, J. Euro. Ceram. Soc 2002; 22(9-10): 1723-28. [16] Zhang T.S, Ma J, Kong L. B., Hing P., Kilner J. A. Effect of Mn addition on the densification, grain growth and ionic conductivity of pure and SiO2-containing 8YSZ electrolytes, Solid State Ionics 2004;167(1-2) :191 96. [17] Badwal P.S.S, Ciacchi T.F, Giampietro K. M. Analysis of the conductivity of commercial easy sintering grade 3 mol% Y2O3 ZrO2 materials, Solid State Ionics 2005;176(1-2):169-78. [18] Tsoga A, Naoimidis A, Jungen W. and Stöver D. Processing and characterisation of fine crystalline ceria gadolinia yttria stabilized zirconia powders, J. Euro. Ceram. Soc 1999;19 : 907-12. [19] Luo J. Ball R.J. and Stevens R. Gadolinia doped ceria/yttria stabilised zirconia electrolytes for solid oxide fuel cell applications, J. Mater. Sci 2004; 39: 235-40. [20] Kang Y. J, Park H. J. and Choi, G. M. The effect of grain size on the low-temperature electrical conductivity of doped CeO2, Solid State Ionics 2008; 179(27-32): 1602-05. [21] Yoshida H, Morita K, Kim B.-N. andHiraga, K. Ionic conductivity of tetragonal ZrO2polycrystal doped with TiO2 and GeO2, J. Euro.Ceram. Soc 2009; 29(3) 411-18. [22] Miyazaki H, Influence of TiO2 solid solution on the thermal property and ionic conductivity of partially stabilized zirconia, Int. J. Appl. Ceram. Tec 2008: 5(5), 490-98. [23] Yoshida H, Inagaki T, Miura K, Inaba M. and Ogumi Z. Density functional theory calculation on the effect of local structure of doped ceria on ionic conductivity, Solid State Ionics 2003:160(1-2) 109-16 [24] Wang F.-Y, Chen S. and Cheng S, Gd3+ and Sm3+ Co-doped Ceria Based Electrolytes for Intermediate Temperature Solid Oxide Fuel Cells, Electrochem. Commun 2004: 6, 743-46.