Alfred University theses are copyright protected and may be used for education or personal research only. Reproduction or distribution in part or

Transcription

1

2 Alfred University theses are copyright protected and may be used for education or personal research only. Reproduction or distribution in part or whole is prohibited without written permission from the author.

3 A COMPARISON OF MICROSTRUCTURE AND ELECTRICAL PROPERTIES OF 8 MOL% YTTRIA STABILIZED ZIRCONIA PROCESSED UNDER CONVENTIONAL, MICROWAVE, AND FAST- FIRE SINTERING TECHNIQUES BY MICHAEL S. UGOREK B.S. ALFRED UNIVERSITY (22) SIGNATURE OF AUTHOR (Signature on file) APPROVED BY (Signature on file) DOREEN EDWARDS, ADVISOR (Signature on file) WALTER A. SCHULZE, ADVISORY COMMITTEE (Signature on file) CASPAR McCONVILLE, ADVISORY COMMITTEE (Signature on file) MATTHEW HALL, CHAIR, ORAL THESIS DEFENSE ACCEPTED BY (Signature on file) ALASTAIR CORMACK, DEAN, SCHOOL OF ENGINEERING

4 ACKNOWLEDGMENTS There are several people I would like to thank, for without their contributions this would never have been possible. Dr. Edwards, my thesis advisor, thank you for trusting in my ability as a student and your never ending patience and guidance for without you, this would have never been possible. Dr. Schulze, Dr. Mc Conville, and Dr. Holly Shulman, my thesis committee, thank you all for your guidance throughout my two years at college. To my family for without all their support and sacrifices, I would not have made it this far. To Ann Baldwin, whose hospitality and friendship has been like a second home to me (even for those times I was kicked out and remained on the couch). To Fran Williams, thank you for all your time, patience, and assistance through the last two years. Without you, things would not have moved as smoothly as they did. Finally to all my friends that I have meet here at Alfred, thank you for making these last four years, the most enjoyable ones! iii

5 TABLE OF CONTENTS Page Acknowledgments...iii Table of Contents...iv List of Tables...v List of Figures...vi abstract...ix I INTRODUCTION... 1 II LITERATURE REVIEW... 3 A. 8 mol% YSZ...3 B. Conventional Sintering...3 C. Microwave Sintering Microwave Material Interactions Hybrid Heating and Sintering of 8 mol% YSZ...6 D. Impedance Spectroscopy Determination of the Activation Energy Impedance Work on 8 mol% YSZ...11 III EXPERIMENTAL PROCEDURE A. Sample Sizes and Sintering Rates...13 B. Sintering Equipment...15 C. Characterization Density SEM / Grain Size Measurements Impedance Spectroscopy...17 IV RESULTS AND DISCUSSION A. Microstructure...18 B. Density...26 C. Impedance Data...3 V SUMMARY AND CONCLUSIONS VI FUTURE WORK REFERENCES APPENDIX iv

6 LIST OF TABLES Page Table I. Average Grain Size of YSZ Processed Using the Different Methods Table II. Average Density of YSZ Processed Using Three Sintering Techniques... 3 Table III. Summary of Electrical Conductivity Parameters for YSZ Sintered Using the Three Techniques v

7 LIST OF FIGURES Figure 1. Page Example of a typical Nyquist Plot showing grain, grain boundary, and electrode semicircles... 8 Figure 2. Equivalent circuit used to analyze the Nyquist plots... 9 Figure 3. Figure 4. Figure 5. Figure 6. Figure 7. A comparison of heating profiles for the microwave and conventionally sintered samples A comparison of the slower heating profiles for the microwave and fast-fired sintered samples A comparison of the two different microwave heating profiles used. Note s represents the shorter microwave run Diagram of the samples placed in between two SiC susceptors during the microwave sintering runs SEM micrograph of a conventionally sintered sample processed at 145 o C Figure 8. SEM micrograph of a microwave sintered sample processed at 145 o C Figure 9. SEM micrograph of a fast-fired sintered sample processed at 145 o C Figure 1. Microwave SEM micrograph using the microwave heating profile as shown in Figure 5, sintered at 14 o C... 2 Figure 11. Microwave SEM micrograph using the microwave s heating profile, as shown in Figure 5, sintered at 14 o C Figure 12. SEM micrograph of the cross section of a microwave sintered sample processed at 145 o C Figure 13. SEM micrograph of the cross section of a fast-fired sintered sample processed at 145 o C vi

8 Figure 14. Sketch of pellet cross section, showing approximate location for SEM imaging Figure 15. SEM micrograph of the microwave sintered sample processed at 145 o C without a dwell time Figure 16. SEM micrograph of the microwave sintered sample processed at 145 o C with a 15min dwell time Figure 17. Average grain size as a function of sintering temperature Figure 18. Density as a functin of sintering temperature Figure 19. Density as a function of average grain size for the sintered YSZ samples Figure 2. Representative impedance plots collected at a) 3 o C, b) 55 o C, c) 9 o C for 8 mol% YSZ Figure 21. Conductivity as a function of reciprocal temperature for samples processed at 135 o C with no dwell time Figure 22. Conductivity as a function of reciprocal temperature for samples processed at 14 o C with no dwell time Figure 23. Conductivity as a function of reciprocal temperature for samples processed at 145 o C with no dwell time Figure 24. Conductivity as a function of reciprocal temperature for samples processed at 14 o C. Note s indicates faster microwave heating rate profile, D indicates dwell Figure 25. Graph of the conductivity as a function of average grain size Figure 26 a-n. Impedance spectra for traditionally processed samples sintered at 14 o C, for temperatures ranging from 2 o C 1 o C Figure 27 a-m. Impedance spectra for microwave processed samples sintered at 14 o C, for temperatures ranging from 2 o C 1 o C Figure 28 a-n. Impedance spectra for fast-fire processed samples sintered at 14 o C, for temperatures ranging from 2 o C 1 o C vii

9 Figure 29 a-c. Representative SEM images of the traditional, fast-fire, and microwave samples sintered at 135 o C Figure 3 a-g. Representative SEM images of the traditional, fast-fire, and microwave samples, with and without the inclusion of a dwell time, sintered at 14 o C Figure 31 a-f. Representative SEM images of the traditional, fast-fire, and microwave samples, with and without the inclusion of a dwell time, sintered at 145 o C Figure 32 a-d. Representative SEM images of the fast-fire and microwave cross sections viii

10 ABSTRACT A comparative study of the microstructure and electrical properties was conducted on three different sintering techniques microwave, fast fire, and traditional were investigated as a means of sintering 8 mol% YSZ. YSZ powders were isostatically pressed into pellets and sintered at 135 o C 15 o C with rates ranging from 3 o C/min (traditional) to 6-2 o C/min (microwave and fast-fire) and dwell times ranging from 15min. Sintered pellets were characterized using scanning electron microscopy (SEM) to determine microstructure and impedance spectroscopy to determine high-temperature electrical properties. While the dependency of density and grain-size on sintering temperature was different for the three techniques, there were no significant differences in the relationships of density and electrical conductivity to grain size. By correlating the microstructure (grain size/ shape and the grain boundary) and high temperature electrical properties of sintered samples, a better understanding of the microwave sintering process and how this process affects final electrical properties of YSZ was determined ix

11 I INTRODUCTION Microwave sintering is an innovative process that has several benefits in sintering low dielectric loss materials such as yttria stabilized zirconia (YSZ). Some of these benefits include rapid heating rates or shorter sintering times, which increases production. Microwave energy has the potential to produce the same microstructure compared to a conventionally sintered sample at reduced sintering temperatures. Microwave sintering is also an energy efficient process, which leads to reduced costs for production. Microwave sintering has also been shown to enhance the densification rate and to reduce the activation energy associated with sintering. Other potential benefits include improving YSZ s physical properties in comparison those obtained by conventional processing. YSZ is the most common electrolyte used in solid oxide fuel cells (SOFC). In SOFC s, YSZ is placed in between materials such as Ni or Ni-YSZ and LaSrMnO 3 for the anode and cathode, respectively. 1 YSZ s main function is the conduction of oxygen ions from the cathode to the anode, producing a voltage. For this application, YSZ must be fully dense, and have a high ionic conductivity at the operating temperatures of 8-1 o C (~.1 S cm -1 ). To determine the effectiveness of using a microwave sintering process, sintered samples of 8 mol% YSZ were prepared from conventional and fast-fire sintering techniques. Therefore, a comparative study on the microstructural evolution and the relationship to the high temperature electrical properties were preformed. The properties specifically being tested were the heating rate effect on the density vs. grain size relationship, ionic conductivity (activation energy), and the effects of microwave heating. Differences in the density vs. grain size have been reported when comparing microwave and conventionally sintered samples. For 3 mol% YSZ, there is a difference in the density vs. grain size relationship when using a faster heating rate. 2 This difference results from the difference in activation energy for sintering for the microwave and conventionally sintered samples. For 8 mol% YSZ, however, there have been only a few studies reported in the literature. 3-6 Therefore, it is of interest to determine what effect 1

12 microwave sintering has on the density/grain size relationship and the resulting ionic conductivity. In the literature, a so-called microwave effect is cited as the reason for the observed differences in sintering rate, and microstructure. However, it is unclear whether this effect results from inherent differences in the heating mechanism or simply the differences in the heating rate. In this study, samples were processed with both techniques having the same heating profile for a more effective comparison. Detailed work has been lacking in showing if there is any difference in the local electrical properties of 8 mol% YSZ when using a microwave sintering technique. The electrical properties of YSZ are dependent on the number of grain boundaries, especially in SOFC applications. Therefore, it is of interest to determine what effect the use of microwave sintering has on the local electrical properties of 8 mol% YSZ. To determine the effect of microwave sintering compared to conventional techniques, the samples were prepared in the following general manner. YSZ powders were isostatically pressed into pellets and sintered at 135 o C 15 o C with rates ranging from 3 o C/min (traditional) to 6-2 o C/min (microwave and fast-fire). Dwell times were included and ranged from 15 min (microwave and fast-fire) to 2hrs (traditional). The sintered pellets were characterized using scanning electron microscopy (SEM) to determine microstructure and grain size, using Archemedes method to determine density, and using impedance spectroscopy to determine the high-temperature electrical properties. Chapter 2 includes background information on microwave sintering theory and a brief overview of conventional sintering theory and impedance spectroscopy. Chapter 3 gives a detailed description of the processing and characterization technique. The rest of the chapter discusses the previous literature on 8 mol% YSZ. In chapter four, the experimental results of this study are given with a focus on the topics of interest outlined above. Conclusionary remarks and projections for future areas of research are given in chapters five and six respectively. 2

13 II LITERATURE REVIEW A. 8 mol% YSZ Pure zirconia has three different crystal structures, tetragonal, monoclinic, and cubic. At 115 o C pure zirconia experiences a phase transformation from tetragonal to monoclinic, which greatly increases the volume of the crystal structure. The increase in volume is accompanied by the formation of cracks in the material, which compromises the structural integrity. With the addition of a stabilizer (Y 2 O 3 ) into the system, the phase change is altered. Instead, both cubic and tetragonal phases are present above 115 o C. When cooling down to room temperature, the tetragonal monoclinic phase change does not occur, and the two initial phases (cubic and tetragonal) are present. Also, with the addition of 8 mol% Y 2 O 3, at room temperature the tetragonal phase does not exist and only the cubic phase is present. When this occurs the material is considered to be fully stabilized. 7 At room temperature, 8 mol% YSZ is a cubic cell, where the oxygen ions occur at the corners and the combination of yttria and zirconia occur at the faces. Since YSZ is in the form of AB 2, (Zr,Y)O 2, the crystal structure is similar to that found for CaF 2. 8 The main mechanism of transport in 8 mol% YSZ is the ionic conduction of oxygen via oxygen vacancies. The vacancies are increased by the substitution of Y 3+ for the Zr 2+ ions, which is shown by the following defect equation: 9,1 ZrO Y O Y ' + 3O 2 3 Zr O + V O (1) B. Conventional Sintering Sintering in a ceramic material occurs due to the reduction of free energy in the system. The main driving force for sintering is the reduction in the curvature of surfaces and/or grain boundaries of 8 mol% YSZ and is in direct relation to the reduction of the grain boundary surface energy. However, in order for sintering to occur in a reasonable period, matter must be transported by the diffusion of atoms or ions. This process occurs 3

14 due to defects within the material. In the case of YSZ, the defects are the vacancies on oxygen sites due to the addition of yttria. These defects also aid in determining the mechanisms of diffusion for which the sintering process can happen. 11 There are several diffusional paths which include i) surface diffusion ii) grain boundary diffusion iii) lattice diffusion (both surface and bulk diffusion), and iv) vapor transport. Out of all these mechanisms, only surface, grain boundary and surface lattice diffusion lead to the densification of the material. These mechanisms lead to the shrinkage and increased strength of the powder compact. Bulk lattice diffusion and vapor transport mechanisms aid in consolidation, the process of grain growth, and pore and grain boundary surface reduction. 11,12 True fast-fire sintering is associated with high heating rates where the stages of coarsening are passed and only densification occurs at the peak temperatures. For this experiment, the heating rates were below the typical rates observed for only densification to occur, therefore both coarsening and densification are observed in the fast-fire system. C. Microwave Sintering Microwaves are part of the electromagnetic spectrum with a frequency range from 3MHz to 3GHz. The most commonly used frequency for the sintering of materials (and used in this experiment) is 2.45 GHz (122nm wavelength). Microwaves are coherent and polarized and must follow the laws of optics, i.e. the microwaves can be reflected, transmitted, and absorbed. This latter feature is utilized in the sintering of materials. 13,14 Microwave sintering is a fundamentally different sintering process than conventional methods. In microwave sintering heat is produced internally rather than with an external heating source. This internal or volumetric heating gives rise to a reversed thermal gradient, compared to conventional processes. In microwave sintered materials, the core of the material is at a higher temperature than the surface. This form of heating, as mentioned before, enhances sintering and has the ability to improve the mechanical (strength), electrical (conductivity), and microstructural (grain size) properties. This phenomenon is described as the microwave effect, which is classified as the difference between temperatures of two sintering techniques that produce similar 4

15 microstructures. 12 The so called microwave effect and the changes in sintering behavior can be attributed to the difference in material interactions with the microwave energy. 1. Microwave Material Interactions Ceramic materials are heated by the absorption of microwave energy, and the material s dielectric loss. The dielectric loss can be described by two main mechanisms: conductive current (specifically ionic conduction) and dipolar re-orientation, 15 which can be expressed mathematically as an effective dielectric loss factor, ε e, 15 ε = ε" + " e σ ωε o (2) where ε represents the portion contributed by dipolar re-orientation, σ is the conductivity, ε o is the permittivity of free space, and ω is the angular frequency. For 8 mol % YSZ the dominant dielectric loss mechanism is via ionic conduction, therefore the ε is negligible. The amount of absorption of a ceramic material is based on the power deposited (P) or absorbed within the reflective cavity per unit volume. This is given as P V 2 " e 2 = σ E = ωε ε E (3) o where E is the magnitude of the electric field supplied by the microwave energy. This equation also assumes that ε, is negligible in YSZ. Combining Equation (1) and (2), and incorporating into tan δ, loss tangent, which is the common way of describing the energy " ε loss, where tan δ = e ε '. The power absorbed by a ceramic material is 1,15,16 r P V ' r 2 = ωε ε tanδ E (4) o As shown in Equation (4), the dielectric properties (ε, ε r and tan δ) play an important role in determining the extent of the power absorbed by a material. The amount of absorbed power that can be converted to heat in the sample can be expressed from Equation (4) as a function of temperature and time as: 16 5

16 dt dt " ωε oε e E = (5) ρc P 2 Where ρ is the density and Cp is the heat capacity of the material on a gram basis. This equation assumes that the structural parameters, such as cell dimensions, are included in the changes in dielectric properties (ε, ε r and tan δ). Due to the material s interactions, there are several problems that are associated with improper control when sintering with microwave energy. The main problem includes microwave sintering low loss ceramics such as 8 mol% YSZ without causing non-uniform microwave heating effects. The non-uniformity from sample preparation causes areas of the ceramic to heat at a faster rate than the other portions causing cracking from the induced stresses at the thermal boundaries. Another problem associated with the non-uniform heating is thermal runaway. By increasing the temperature, the dielectric properties (ε, ε r and tan δ) of the material increases, which causes portions of the material to heat exponentially. This portion of the material could potentially melt, deform, or fail. 13,14 Although these problems are significant, a method of sintering low loss ceramics by hybrid heating has been developed. 5,17,18 2. Hybrid Heating and Sintering of 8 mol% YSZ Hybrid (or indirect) sintering was developed to reduce the non-uniform heating effects of microwave energy for low loss ceramics, such as YSZ. Hybrid heating involves the addition of a high suscepting material to aid in thermally heating the ceramic until the ceramic can absorb microwaves directly. SiC rods are the most common material used for susceptors. At low temperatures, the SiC heats faster with the microwave energy than the YSZ. At low temperatures YSZ is heated by convection from the SiC avoiding the problems associated with non-uniform heating from microwave energy directly. 5 When the temperature increases above 5 o C, the ionic conduction of O 2- in YSZ couples with the microwave energy and heats faster than the SiC. Therefore, it has been shown experimentally that above 5 o C YSZ is being heated directly by the microwave energy. 5 The density / microstructure relationship in YSZ was studied previously, but the results are varied. Janey et al. showed that the microwave sintered samples had a finer 6

17 grain size when compared to conventionally sintered samples. 5,19 Nightingale et al. 2,2 and Goldstein et al., 21 however, showed that for samples sintered in a microwave field, there was no significant difference in the grain size or distribution. In all cases, however, the microwave-sintered samples did produce a higher density when compared to conventional sintering techniques at low temperatures. As sintering temperatures was increased, the density / microstructure relationship of the two techniques became similar. 3,15,19,2,22 The microwave temperature difference ranged from 5 o C to o C. Samuels and Brandon 23 even determined a temperature difference of 8 o C with their dilatometry measurements. All authors used the hybrid technique, with slightly modified SiC configuration. The heating rates of Nightingale et al. 2 ranged from a constant 2 o C/min to 2 o C/min and a maximum of 3 o C/min by Janey et al. 5 The soak times ranged from 1hr. D. Impedance Spectroscopy Impedance spectroscopy, pertaining to the study of ionic conduction in 8 mol % YSZ, is a relatively simple and straight forward process. This process constitutes measuring the impedance over predetermined frequency range by applying an AC voltage (1 volt) across the sample that has been electroded on both sides. When applying the AC voltage, the phase shift and amplitude of the current are measured as a function of frequency. The phase shift and amplitude are then converted into the real and imaginary portions of the impedance of the sample. The values are measured on a log scale, and a Nyquist plot is then graphed for analysis, 24 as shown in Figure 1. 7

18 Grain Grain boundary Electrode Figure 1. Example of a typical Nyquist Plot showing grain, grain boundary, and electrode semicircles. By plotting the imaginary portion of the impedance on the y-axis vs. the real impedance on the x-axis a characteristic plot, called a Nyquist plot, is produced as shown in Figure 1. In Figure 1, each point represents the impedance measured at a single frequency, where the lowest frequency is at the right end of the plot and the highest frequency is at the left. 24 The data shows a series of semicircles. Depending on the material used, each semicircle can be correlated to a physical process. In the case of 8 mol % YSZ, the first semicircles represent ionic conduction through the grain. The second and third semicircles represent the polarization at and conduction across the grain boundary and polarization at the electrode. Once the Nyquist plots are produced, valuable information can be extracted from the individual semicircles by using an equivalent circuit model. The information extracted consists of the individual resistance (grain, grain boundary and electrode), and from those values a sample resistivity can be determined. The details of the equivalent circuit used in this experiment will be discussed later, but the elements that are used to describe the semicircles are a resistor and constant phase element (CPE) in parallel. A 8

19 sample of the equivalent circuits used is shown in Figure 2. Altering these values will determine the degree of accuracy in modeling the Nyquist plots. 24 R1 R2 R3 CPE1 CPE2 CPE3 Figure 2. Equivalent circuit used to analyze the Nyquist plots. The CPE is an element that is used for materials that do not behave like an ideal capacitor. The difference can be shown by the following general equation for the impedance of a capacitor: ( ) α Z = A jω (6) where, ω is the angular frequency, j is equal to (-1) 1/2, A is a constant, and α is the exponential constant. For an ideal capacitor, A = 1/C (the inverse capacitance), and α is equal to one. For non-ideal conditions (CPE) α is less than one. 24 Conductivity and the activation energy of the conduction process are then determined from the resistance values extracted and the sample geometry over a temperature range. 1. Determination of the Activation Energy After the sample resistance (grain and grain boundary resistance combined) has been determined from the Nyquist plot, the conductivity of the sample at a given temperature is calculated using the following equation: 25 t σ = (7) AR S where σ is the conductivity of the sample, R S is the sample resistance, A and t are the area and thickness respectively. The conductivity as a function of temperature follows the following relationship: 25 σ E = A σ o exp T kt (8) 9

20 where σ is total ionic conductivity, σ o is a pre-exponential factor, E A is the activation energy (in Joules), k is Boltzman s constant, and T is the temperature in Kelvins. Rearranging the following equation and taking the natural logarithm the conductivity equation can be expressed as: E A 1 ln( σ T ) = lnσ o (9) k T Plotting ln (σt) vs. 1/T an Arrhenius plot is produced. The data is then fit as a straight line and the activation energy can be determined from the slope of the graph (E A /k). In Equation (8) there is an extra temperature dependence (σ o /T) that is not present in the more common conductivity equation for solid state. This extra temperature dependence is related to the diffusion of the charged ions through the material that is highly thermally dependent. This difference is shown by the following brief derivation of the Nernst-Einstein equation for ionic conduction in solids. With an applied electric field, the net jump frequency for the movement of an ion from one lattice position to another is given as: f net qεaυ = exp kt ( Ea ) where ε is the applied field, q is the charge of the ion, a is the lattice jump distance, υ is the lattice vibration frequency, Ea is the activation energy associated for conduction, k is Boltzman s constant, and T is the temperature in Kelvin. kt (1) The drift velocity, V D, is given as fnet*a, which describes the movement of the ions through the material over the lattice distance. Given the net flux of ions as: J = nqvd = σε (11) where n is the number of charge carriers per volume. Combining equations (1) and (11) J ( Ea ) = σε = nq * ε υ q a exp (12) kt kt Re-arranging Equation (12) in terms of the conductivity and combining the temperatureindependent terms as σ o : 1

21 ( Ea o ) = ( Ea ) 2 nq a υ σ = exp exp (13) kt kt T kt 2 σ For solid state electronic conduction, the drift velocity is not largely a function of temperature. For electronic conduction V D = eτ/m, where e is the charge of the conducting species, m is the mass of the conducting species, and τ is the relaxation time or the time between collisions. Therefore, for ionic conduction there is an extra temperature term, which correlates to the motion of the charged ions through the material. 2. Impedance Work on 8 mol% YSZ Chen et al. 26 conducted impedance spectroscopy on nanoscale YSZ ranging from 2 o C to 1 o C, but mainly focused the analysis on the upper temperature regions. The highest conductivity value was.15 S/cm for samples sintered at 135 o C for 4 hrs in a conventional furnace. The authors were able to separate the grain and grain boundary contributions and analyzed the individual contributions as a function of sintering temperature. Chen et al. 26 noted that the contribution of the grain conduction increased rapidly initially but leveled out above 13 o C. The grain boundary contribution initially increased, but above 135 o C, the conductivity decreased. This decrease brought the overall conductivity of the sample, which this trend Chen et al. contributes to the increase in exaggerated grain growth of the sample. Ciacchi et al. 3 conducted DC conductivity on YSZ over the same temperature region, and AC impedance over the lower temperature regions (<5 o C). The AC data was used to separate out the grain and grain boundary contributions to the overall conductivity. They concluded that the grain boundary contribution was minor and were unable to separate out the individual contributions without a large amount of error. Ciacchi et al. also concluded that based on the DC conductivity data, there was no significant difference between the microwave and conventionally sintered samples when comparing the overall ionic conductivity. 3 Gong et al. conducted impedance spectroscopy on YSZ and showed that when plotting the natural log of the conductivity multiplied by temperature as a function of 1/T, a non-arrhenius behavior was observed. This non-arrhenius behavior was in the form of 11

22 two distinct slopes in the ln(σt) vs. 1/T curve. The ln(σt) vs. 1/T curve above 85K showed a deviation from the initial low temperature trend line. This change in slope is attributed to two different conduction mechanisms present in YSZ. In the lower temperature regions, the oxygen vacancies formed clusters, where as the oxygen vacancies were free to conduct the upper temperature regions. 25 However, there are other cases (those mentioned above) in which the analysis showed there was only one mechanism of conduction for YSZ. 2,3,26 12

23 III EXPERIMENTAL PROCEDURE YSZ powders with no binder additions were isostatically pressed into pellets and sintered at 135 o C 15 o C with rates ranging from 3 o C/min (traditional) to 6-2 o C/min (microwave and fast-fire). Dwell times for microwave and fast-fire runs ranged from to 15 minutes. For conventional runs, dwell times ranged from to 2 hours were used. The sintered pellets were characterized using scanning electron microscopy (SEM) to determine microstructure and grain size, Archemedes method to determine density, and impedance spectroscopy to determine the high-temperature electrical properties. A. Sample Sizes and Sintering Rates One-gram, ½-inch diameter pellets were prepared from Tosoh TZ-8YS powder by unaxial pressing at 8 MPa followed by cold isostatic pressing at 24 MPa without the addition of a binder or other materials. The pellets were sintered to o C via traditional (3 o C/min), microwave (6-2 o C/min), and fast-fire (6-2 o C/min) techniques. Dwell times ranged from to 15min (microwave and fast-fire) to to 2hrs (traditional) at peak temperatures. Figures 3-5 show the heating profiles conducted in this experiment. Figure 3 shows a comparison of the heating rates of the traditional (constant heating rate) to the microwave sintered samples. Figure 4 shows the comparison of the slower microwave sintering profile and the programmed fast-fire heating rate. Figure 5 shows a comparison between the two different microwave heating profiles. For each sintering technique, sacrificial YSZ powder was placed above and below the pellets to reduce surface contamination during sintering. 13

24 Temp ( o C) Microwave Traditional Time (min) Figure 3. samples. A comparison of heating profiles for the microwave and conventionally sintered Temp ( o C) Microwave Fast-Fire Time (min) Figure 4. A comparison of the slower heating profiles for the microwave and fast-fired sintered samples. 14

25 Temp ( o C) Microwave Microwavve S Time (min) Figure 5. A comparison of the two different microwave heating profiles used. Note s represents the shorter microwave run. B. Sintering Equipment Hybrid microwave sintering was conducted using a 1.3 kw, 2.45 GHz variable-power system (ThermWAVE, Research Microwave Systems, USA). Samples were placed between two circular SiC susceptors inside a refractory-board container as shown in Figure 6. The samples were heated using 5% power output until 5 o C, followed by and maintained at 1% power output to the desired sintering temperature and dwell times. Fast-fire sintering was conducted in a vertical tube furnace with computercontrolled sample feed, which allowed the heating profiles to be programmed to mimic the heating profile measured during the microwave heating runs. Traditional sintering was conducted in a standard tube furnace with a programmed heating rate of 3 o C/min. For each technique, the temperature was measured using a thermocouple placed near, but not touching, the surface of the sample. The thermocouple used during the microwave heating runs was shielded in metal to reduce arcing effects. 15

26 Sample Holder SiC Samples Figure 6. Diagram of the samples placed in between two SiC susceptors during the microwave sintering runs. C. Characterization 1. Density Green density was geometrically measured using the thickness, radius, and weight of the pressed pellet. Sintered pellet density was measured using Archemedes method in water with a boiling period of 3 ½hrs and cooling period to room temperature of 1hr. The dry, saturated, and suspended weight of each sample were measured, and the density calculated by the following equation: D Density = (14) ( W S ) where D is the dry weight, W is the saturated weight, and S is the suspended weight. 2. SEM / Grain Size Measurements SEM images were taken on the pellet as fired surface and cross sections after polishing to 1 µm. Average grain size was determined using a circular-intercept method. 27 A circle of 79.5mm was drawn on printed 117mm x 8mm SEM micrographs that were calibrated for correct magnification. The grains (within the circle) and partial grains (intercepted by the circle) were tabulated and the average grain size was calculated assuming spherical particles. Multiple SEM micrographs of the surface and cross section where incorporated and averaged together in determining the grain size of each material. 16

27 3. Impedance Spectroscopy Impedance spectroscopy was conducted on samples electroded with Engelhard 6926 platinum paste. The paste was fired on with a 5 o C/min heating and cooling rate with a dwell time of 2 minutes. Measurements were conducted from 1 o C to 1 o C in air, using a Solatron 126 equipped with custom control software. The samples were heated at 1 o C/min, with a dwell time of 1 min at each measurement temperature in order for the sample to reach equilibrium. The data was calibrated for lead corrections, and an excitation potential of 1 V was used. The frequency was scanned from 1MHz to 5Hz on a logarithmic scale. The resulting spectra were analyzed using the Zview software package (Scribner and Associates, USA). 17

28 IV RESULTS AND DISCUSSION A. Microstructure The polished surface microstructures that were produced using all three sintering techniques showed the same general shape, i.e. equiaxed grains in a granular aggregation. This morphology is typical of that reported for 8 mol% YSZ. 2,3,5,1,2,26,28 The microstructures for the microwave and conventionally sintered samples showed a bimodal distribution for the two relative sizes of the grains, however the fast-fired samples were more uniformly mixed. The bimodal distribution was segregated in localized areas of larger and smaller grain size approximately 1 µm and.3 µm respectively. In other words, the microstructures were not uniformly mixed with respect to the two different grain sizes. This shows that there is an effect in the microwave processed samples that produces a larger amount of grain growth when compared to the fast-fired samples, even though the microwave and fast-fired samples were heated at the same rate. Typical microstructures of the conventional, microwave, and fast-fire sintered samples processed at 145 o C are shown in Figures 7 9. When comparing the two different heating rates for the microwave samples, there was no significant difference in the two microstructures at 145 o C. At 14 o C, the slower heating rate produced a slightly larger grain size than the faster heating rate. These microstructures are shown in Figures 1 and 11, where the former is the slower heating profile. In all cases, the porosity was located at the grain boundary regions, with little inter-granular porosity. 18

29 1 µm Figure 7. SEM micrograph of a conventionally sintered sample processed at 145 o C. 1 µm Figure 8. SEM micrograph of a microwave sintered sample processed at 145 o C. 19

30 1 µm Figure 9. SEM micrograph of a fast-fired sintered sample processed at 145 o C. 1 µm Figure 1. Microwave SEM micrograph using the microwave heating profile as shown in Figure 5, sintered at 14 o C. 2

31 1 µm Figure 11. Microwave SEM micrograph using the microwave s heating profile, as shown in Figure 5, sintered at 14 o C. Cross sectional micrographs were taken of the microwave and fast-fired samples to show any effects of the volumetric heating mechanism caused by the use of microwave energy. Figures 12 and 13 show micrographs of the cross sectional microstructures of the microwave and fast-fired samples, respectively. The micrographs were taken at various positions along the cross sections of a sample, marked as an X, as indicated in Figure µm Figure o C. SEM micrograph of the cross section of a microwave sintered sample processed at 21

32 1 µm Figure o C. SEM micrograph of the cross section of a fast-fired sintered sample processed at X X X X X X X X Figure 14. Sketch of pellet cross section, showing approximate location for SEM imaging. With the microwave samples, it is noted that there is a slight, but not statistically significant variation in average grain size through the cross sectional area. As the center of the sample is reached, the grains show a slight increase in average diameter. When comparing the cross section of the samples to the surface of the microwave processed samples, the internal average grain size was significantly larger, producing an average size of 1.1 µm vs..7 µm for the center and surface, respectively. Nightingale et al. 2,2 have shown that there was no difference in the cross sectional area compared to the rest of the sample, which is contrary to what was observed. From this observation, it is suggested that the interior of the sample was at a higher temperature than what was observed at the surface. The fast-fire samples, however, did not show any significant change through the cross-section or when compared to the surface, suggesting the surface and interior were at similar temperatures. 22

33 The addition of dwell times for the microwave and fast-fire resulted in some increase in the average grain size, while that for the traditional samples did not. The grain sizes for the traditionally fired samples at 14 o C were similar to the microstructures produced at 145 o C without any dwell times. However, comparing the 14 o C dwell time samples to the samples fired at 14 o C without a dwell there is an increase in the average grain size. For the fast-fired samples with the addition of a dwell, time the grain size increased as well, which is to be expected. The microwave sintered samples showed the greatest increase in grain size with the addition of a dwell time. The average grain size increased from about 1 µm to 1.7 µm. Figures 15 and 16 show the before and after dwell microstructures of the 145 o C microwave samples. 1 µm Figure 15. dwell time. SEM micrograph of the microwave sintered sample processed at 145 o C without a 23

34 1 µm Figure 16. SEM micrograph of the microwave sintered sample processed at 145 o C with a 15min dwell time. The evolution of grain size with respect to the sintering temperature (no dwell) is shown in Figure 17. At the lowest sintering temperatures (135 o C), the grain sizes of all three samples are about the same around.4µm. Since this temperature region is well below the manufacturer s recommended sintering temperature of 14 o C, the samples have just begun the initial stages of consolidation and sintering. As sintering temperatures are increased, the conventionally sintered samples, first show a slight increase in the average grain size, and then show a more dramatic change above 145 o C. 24

35 3 2.5 Average Grain Size (µm) Traditional Fast-Fire Microwave Temperature ( o C) Figure 17. Average grain size as a function of sintering temperature. Table I lists the average grain size for each sintering temperature and technique. The trend in the average grain size with respect to the sintering temperature is attributed to the fact that the conventionally sintered samples had a slower heating rate, providing a longer time for the sample to sinter around the target temperature. The trend in average grain size as a function of sintering temperature is consistent to what has been observed before. 2,5,2,21 What is also of interest is that the microstructures at o C of the conventionally and microwave sintered samples are similar. This comparison shows that using microwave energy to sinter 8 mol% YSZ can achieve comparable microstructures however, in a shorter time. 25

36 Table I. Average Grain Size of YSZ Processed Using the Different Methods. Technique Temperature ( o C) Dia. Grain Size (µm) Dia. Grain Size (µm) (No Dwell) (No Dwell)* (Dwell)* Fast-Fire NA NA NA NA 1.1 Conventional NA NA NA NA NA NA Microwave NA NA * microwave s heating profile was used see Figure 5 for reference. B. Density Figure 18 shows the density as a function of sintering temperature. Density increases with increasing sintering temperature, as expected. The conventionally sintered samples achieved a higher density at lower temperatures than both the microwave and fast-fire samples. This is contrary to what has been shown before, where the microwave sintered samples showed a higher density then conventionally sintered samples. 2,5,2 The conventionally sintered density of the samples is higher than the microwave or fast-fire due to the longer heating times. The conventionally sintered sample was heated at a slower rate; therefore, there is more time for diffusion, and sintering, to occur. 26

37 6 Density (g/cm 3 ) 5 4 Traditional Fast-Fire Microwave Microwave s Temperature ( o C) Figure 18. Density as a functin of sintering temperature. Above 145 o C there is a leveling out in the density, which indicates that the maximum density has been reached for the samples prepared in the method outlined in this thesis. Achieving high density (>98% of theoretical) was not achieved due to the low starting green density (~5% of theoretical) and due to the large amount of internal porosity in the sample. At the higher sintering temperatures, in Figure 18, the density of the microwave sintered samples approaches that of the conventionally processed samples, which shows that the microwave processing technique can produce the same density as conventionally sintered samples, but at a significant shorter time. The microwave effect, the difference between temperatures of microwave and conventionally sintered materials with the same microstructure, was calculated to be around 3 +/- 2 o for this experiment. This value thus far is the smallest reported in literature. Janey et al. reported a difference of o C. 5 However, this difference could be caused by the longer dwell times of 1hr rather than the 15min in this experiment. The large difference could also be contributed to the fact that Janey et al. sintered the material in Ar/N 2 atmosphere, and not in air. Nightingale et al. determined a microwave effect of 5 o C 2,2, which is similar to what was determined here. The 27

38 difference between their work was the longer dwell times at peak temperatures. Another cause for the differences could result from how accurately the temperature was measured both in this thesis as well in the other experiments. In microwave sintering there is always a question of whether or not the thermocouple is reading the actual temperature. Since microwave sintering involves a volumetric heating mechanism, the interior of the samples could potentially be at a higher temperature than the surface of the samples. Also, the introduction of a shielded thermocouple could potentially cause the local area around the thermocouple to be at a higher temperature due to the microwave energy being focused by the metal. To correct for this error, the thermocouple was raised a few millimeters to cause the concentration of microwave energy to be reduced, giving the cavity within the microwave box a more uniform heating profile. By raising the thermocouple above the sample, the temperature that is observed will be less than the actual sample temperature, since the sample is the dominant heating sources. Density was plotted as a function of average grain size. Figure 19 shows this relationship for all three sintering techniques. All three techniques exhibit a similar density-vs.-grain-size relationship, which suggests that the relative contribution of lattice, grain-boundary, and surface diffusion in 8 mol% YSZ does not change significantly with heating rate or with the sintering method used. 28

39 6 Density (g/cm 3 ) Traditional Fast-Fire Microwave Average Grain Size (µm) Figure 19. Density as a function of average grain size for the sintered YSZ samples. The two different heating profiles for the microwave sintered samples and the inclusion of a dwell time greatly affected the final density of the sintered samples. Table II gives the average density for each microwave heating profile and the densities of the samples with the 15-minute dwell times. The density of the samples processed using the faster heating profile was significantly lower compared to that of the samples processed using the slower microwave heating profile. The lower density is caused by the reduced amount of time for sintering to occur for the faster heating profile. Incorporating the dwell times, the densities of the samples processed using the microwave s profile increased and were comparable to the densities achieved without a dwell time using the microwave profile from Figure 5. 29

40 Table II. Average Density of YSZ Processed Using Three Sintering Techniques Technique Temperature ( o C) Average Density (g/cm 3 ) No Dwell Dwell * Microwave Fast-Fire Traditional * Dwell times were 15 minutes for microwave and fast-fire; 2 hours for conventional C. Impedance Data Figure 2 shows representative impedance data collected at three different temperatures. The spectrum collected at 3 o C (Figure 2a) shows three arcs, which correspond to grain, grain-boundary, and electrode contributions. For the spectrum collected at 55 o C (Figure 2b), only two arcs are present, corresponding to grainboundary and electrode contributions. Because the frequency response of the grain interior lies outside the measurement range, the grain contribution was modeled as an offset resistance. For the spectrum collected at 9 o C (Figure 2c), only one arc, corresponding to the electrode interface contribution, is observed. In this case, the sample resistance (grain plus grain-boundary) is taken as the offset-resistance. The sample resistance (determined by combining the contributions of the grain and grain boundary resistance) were extracted from the spectra and used to calculate the sample conductivity. A full set of impedance plots for the conventional, microwave, and fast-fire samples sintered at 145 o C are given in the Appendix. 3

41 (Ohms) -5-2 a) Collected at 3 o C grain b) Collected at 55 o C R g + R gb R g + R gb grain boundary electrode electrode -1 c) Collected at 9 o C R g + R gb electrode (Ohms) Figure 2. Representative impedance plots collected at a) 3 o C, b) 55 o C, c) 9 o C for 8 mol% YSZ. Figure are plots of the conductivity data for a set of YSZ samples sintered from 135 o C o C using the three different sintering methods. Figure 24 shows the conductivity data for the 14 o C samples that included the faster heating profile for the microwave sintered samples and the samples that included a dwell time. For ionic conductors, the conductivity is expected to depend on temperature as E A 1 ln( σ T ) = lnσ o (15) k T where σ is conductivity, σ o is a constant, T is measurement temperature in Kelvin, E a is activation energy, and k is Boltmann s constant. Values for the activation energies and σ o were extracted from the plots of ln (σt) vs. 1/T and are summarized in Table II, along with the statistics of the linear least-squares fit to the data. 31

42 ln (σ T) Fast-Fire Traditional Microwave /T Figure 21. Conductivity as a function of reciprocal temperature for samples processed at 135 o C with no dwell time ln (σ T) Fast-Fire Traditional Microwave /T Figure 22. Conductivity as a function of reciprocal temperature for samples processed at 14 o C with no dwell time. 32

43 ln(σ T) Fast-Fire Traditional Microwave /T Figure 23. Conductivity as a function of reciprocal temperature for samples processed at 145 o C with no dwell time ln (σ T) Trad 14D MW 14s MW 14sD FF 14s /T Figure 24. Conductivity as a function of reciprocal temperature for samples processed at 14 o C. Note s indicates faster microwave heating rate profile, D indicates dwell. 33

44 For samples processed with all three processing temperatures, there was a linear relationship of ln (σt) vs. 1/T, indicating that the mechanism of conduction does not change over the measured temperature range. The only significant difference was that the conductivity of the fast-fire samples produced at 135 o C was slightly lower than that for both the microwave and conventional sintered samples. This difference is due to the larger number of grain boundaries, which results from the fast heating rate and the temperature being under the manufacturer s recommended sintering temperature. This did not occur for the microwave processed samples because the grain size in the center of the samples were larger than the surface, therefor the amount of grain boundaries were reduced. For each sintering temperature and technique, the activation energy was determined from equation (12). The calculated activation energies range from.86 to.95 ev and are given in Table III. For a given sintering method, the activation energy increases with decreasing sintering temperature, which is attributed to the larger number of grain boundaries in the samples processed at lower sintering temperatures. Also, the microwave and conventionally sintered samples showed similar activation energies, both less than the fast-fired samples, showing that there is no significant difference in the mechanism of conduction for the conventional and microwave processed samples. Table III shows the comparison of the activation energy for the samples processed using dwell times. The activation energy is significantly higher for the faster microwave heating profile, which can be explained by the higher number of grain boundaries and lower density. With the addition of a dwell time, the activation energy decreases as is expected. However, the activation energies of the microwave dwell time samples are still higher than the slower heating profile samples of the same temperature. The conventional sample showed an increase in activation energy with the addition of the dwell time, which was unexpected. The activation energy for this sample is similar to the lowest sintering temperature without the inclusion of a dwell. Nightingale et al. showed that testing the conductivity via DC four-probe method showed the microwave sintered samples showed slightly lower, but comparable, electrical activation energy when analyzed against conventional methods. 3 34

45 Table III. Summary of Electrical Conductivity Parameters for YSZ Sintered Using the Three Techniques Sintering Sintering Temp. σ o (S/cm) E A (ev) R 2 Method ( o C) Traditional x (no dwell) x x x Microwave x (Profile A) x (no dwell) x Fast-fire x (Profile A) x (no dwell) x Microwave x (Profile B) x (no dwell) Traditional x (15 m dwell) Microwave x (Profile B) x (15 m dwell) Fast Fire (Profile B) (15 m dwell) x *Note: Conductivity can be calculated as σ = (σ o /T) exp (E A /kt) 35

46 Figure 25 shows the conductivity of each sample prepared by the different sintering techniques as a function of the average grain size. The conductivity of 8 mol% YSZ increases with increasing grain size, reflecting the significant contribution of grain boundaries to sample resistance. The conductivity-vs.-grain-size relationship is similar for samples processed using all three methods, which suggests that the local electrical properties of the grain and grain-boundary regions do not depend on the sintering method used. The microwave and conventionally sintered samples produced by Li et al. shows the same general relationship for ZrO 2 co-doped with CaO and Y 2 O Conductivity Average Grain Size (µm) Traditional Fast-Fire Microwave Figure 25. Graph of the conductivity as a function of average grain size. 36

47 V SUMMARY AND CONCLUSIONS The microstructure and electrical properties of 8 mol% YSZ sintered using three different techniques were compared. The density and grain size was highly dependent on the sintering method and sintering temperature used. At higher sintering temperatures, however, the relationship between the density and grain size as a function of sintering temperature of the microwave and conventional processes were similar. This indicates that the microwave-processed samples can achieve comparable density / grain size relationships, but the time required for the microwave samples is significantly less. The relationship between the density and electrical properties to the average grain size for all three sintering techniques were similar. This can be inferred that the local electrical properties of 8 mol% YSZ is not dependent on the sintering method used but only on the microstructures produced. This also implies that any of the three techniques can be used to achieve 8 mol% YSZ with desired characteristics, but that the time required to achieve the desired microstructure will be different. 37

48 VI FUTURE WORK Recent work has shown that the properties of less pure YSZ can be significantly improved using different heating profiles. 29,3 The current study was conducted using relatively high-purity samples. Future work, involving less pure YSZ powders, should be conducted to determine whether microwave sintering can provide any benefits for sintering 8 mol% YSZ. The addition of a binder and other processing aids for producing a higher starting green density will aid in achieving a higher final density material. The ability to co-fire 8 mol% YSZ with cathode and anode materials would be beneficial. Studies should be conducted to determine the interaction of different materials to microwave energy, and the interface between the electrolyte and the electrodes. 38

49 REFERENCES 1. C. Song, "Fuel Processing for Low-Temperature and High-Temperature Fuel Cells: Challenges, and Opportunities for Sustainable Development in the 21st Century," Catalysis Today, 77 [1-2] (22). 2. S.A. Nightingale, D.P. Dunne, and H.K. Worner, "Sintering and Grain Growth of 3 mol% Yttria Zirconia in a Microwave Field," J. Mater. Sci., 31 [19] (1996). 3. F.T. Ciacchi, S.A. Nightingale, and S.P.S. Badwal, "Microwave Sintering of Zirconia - Yttria Electrolytes and Measurement of Their Ionic Conductivity," Solid State Ionics, [2] (1996). 4. D.Z. de Florio and R. Muccillo, "Sintering of Zirconia - Yttria Ceramics Studied by Impedance Spectroscopy," Solid State Ionics, 123 [1-4] 31-5 (1999). 5. M.A. Janney, C.L. Calhoun, and H.D. Kimrey, "Microwave Sintering of Solid Oxide Fuel Cell Materials: I, Zirconia - 8 mol% Yttria," J. Am. Ceram. Soc., 75 [2] (1992). 6. Y.-S. Jung, J.-H. Lee, J.H. Lee, and D.-Y. Kim, "Liquid-Phase Redistribution During Sintering of 8 mol% Yttria - Stabilized Zirconia," J. Eur. Ceram. Soc., (in press). 7. Ceramic Materials for Electronics: Processing, Properties, and Applications, 2nd ed.; pp Edited by R.C. Buchanan. Marcel Dekker, New York, J. William D. Callister, Materials Science and Engineering an Introduction, 4th ed.; pp John Wiley & Sons, New York, F.C. Fonseca, E.N.S. Muccillo, and R. Muccillo, "Analysis of the Formation of ZrO2:Y2O3 Solid Solution by the Electrochemical Impedance Spectroscopy Technique," Solid State Ionics, 149 [3-4] (22). 1. D.D. Upadhyaya, A. Ghosh, K.R. Gurumurthy, and R. Prasad, "Microwave Sintering of Cubic Zirconia," Ceram. Int., (21). 11. M.N. Rahaman, Ceramic Processing and Sintering; pp Marcel Dekker, New York, P. Boch and N. Lequeux, "Do Microwaves Increase the Sinterability of Ceramics?" Solid State Ionics, (1997). 39

50 13. W.H. Sutton, "Microwave Processing of Ceramics - An Overview." Mater. Res. Symp. Pro., 19, (1992). 14. W.H. Sutton, "Microwave Processing of Ceramic Materials," Am. Ceram. Soc. Bull., 68 [2] (1989). 15. Z. Xie, J. Yang, X. Huang, and Y. Huang, "Microwave Processing and Properties of Ceramics with Different Dielectric Loss," J. Eur. Ceram. Soc., 19 [3] (1999). 16. D.E. Clark, D.C. Folz, and J.K. West, "Processing Materials with Microwave Energy," Mater. Sci. Eng., A (2). 17. J. Wilson and S.M. Kunz, "Microwave Sintering of Partially Stabilized Zirconia," J. Am. Ceram. Soc., 71 [1] C4-1 (1988). 18. C. Zhao, J. Vleugels, C. Groffils, P.J. Luypaert, and O.V.D. Biest, "Hybrid Sintering with a Tubular Susceptor in a Cylindrical Single-Mode Microwave Furnace," Acta Mater., 48 [14] (2). 19. M.A. Janney and H.D. Kimrey, "Diffusion-Controlled Process in Microwave-Fired Oxide Ceramics," Mater. Res. Symp. Pro., 189, (1991). 2. S.A. Nightingale, H.K. Worner, and D.P. Dunne, "Microstructural Development during the Microwave Sintering of Yttria - Zirconia Ceramics," J. Am. Ceram. Soc., 8 [2] (1997). 21. A. Goldstein, N. Travitzky, A. Singurindy, and M. Kravchik, "Direct Microwave Sintering of Yttria - Stabilized Zirconia at 2.45GHz," J. Eur. Ceram. Soc., 19 [12] (1999). 22. A. Pawlowski, M.M. Bucko, and Z. Pedzich, "Microstructure Evolution and Electrical Properties of Yttria and Magnesia Stabilized Zirconia," Mater. Res. Bull., 37 [3] (22). 23. J. Samuels and J.R. Brandon, "Effect of Composition on the Enhanced Microwave Sintering of Alumina-Based Ceramic Composites," J. Mater. Sci., 27 [12] (1992). 24. Impedance Spectroscopy: Emphasizing Solid Materials and Systems; pp Edited by J.R. Macdonald. John Wiley & Sons, New York, J. Gong, Y. Li, Z. Tang, Y. Xie, and Z. Zhang, "Temperature - Dependence of the Lattice Conductivity of Mixed Calcia/Yttria - Stabilized Zirconia," Mater. Chem. Phys., 76 [2] (22). 4

51 26. X.J. Chen, K.A. Khor, S.H. Chan, and L.G. Yu, "Influence of Microstructure on the Ionic Conductivity of Yttria - Stabilized Zirconia Electrolyte," Mater. Sci. Eng., A335 [1-2] (22). 27. Z. Jefferies, E. Met, A.H. Kline, and E.B. Zimmer, "The Determination of Grain Size in Metals," Trans. Am. Inst. Mining & Metal. Eng., 54, (1917). 28. Y. Li, J. Gong, Y. Xie, Z. Tang, and Z. Zhang, "Microstructure Dependence of Electrical Conductivity of (ZrO).9 - (Y2O3).4 - (CaO).6 Solid Electrolyte," Mater. Sci. Eng., B278 [56] 14-4 (2). 29. L. J-H, T. Mori, L. J-G, T. Ikegami, M. Komatsu, and H. Haneda, "Improvement of Grain-Boundary Conductivity of 8 mol% Yttria-Stabilised Zirconia by Precursor Scavenging of Siliceous Phase," J. Electro. Soc., 147 [7] (2). 3. C.C. Appel and N. Bonanos, "Structural and Electrical Characterisation of Silica- Containing Yttria-Stabilised Zirconia," J. Eur. Ceram. Soc., 19 [6-7] (1999). 41

52 APPENDIX Figure 26 a-n: Impedance spectra for traditionally processed samples sintered at 14 o C, for temperatures ranging from 2 o C 1 o C. -4.E+6-3.E+6-2.E+6-1.E+6.E a) Impedance spectra taken at 2 o C. 42

53 b) Impedance spectra taken at 3 o C c) Impedance spectra taken at 4 o C. 43

54 d) Impedance spectra taken at 5 o C e) Impedance spectra taken at 55 o C. 44

55 f) Impedance spectra taken at 6 o C g) Impedance spectra taken at 65 o C. 45

56 h) Impedance spectra taken at 7 o C i) Impedance spectra taken at 75 o C. 46

57 j) Impedance spectra taken at 8 o C k) Impedance spectra taken at 85 o C. 47

58 l) Impedance spectra taken at 9 o C m) Impedance spectra taken at 95 o C. 48

59 n) Impedance spectra taken at 1 o C. Figure 27 a-m: Impedance spectra for microwave processed samples sintered at 14 o C, for temperatures ranging from 2 o C 1 o C a) Impedance spectra taken at 2 o C. 49

60 b) Impedance spectra taken at 3 o C c) Impedance spectra taken at 4 o C. 5

61 d) Impedance spectra taken at 5 o C e) Impedance spectra taken at 55 o C. 51

62 f) Impedance spectra taken at 6 o C g) Impedance spectra taken at 7 o C. 52

63 h) Impedance spectra taken at 75 o C i) Impedance spectra taken at 8 o C. 53

64 j) Impedance spectra taken at 85 o C k) Impedance spectra taken at 9 o C. 54

65 l) Impedance spectra taken at 95 o C m) Impedance spectra taken at 1 o C. 55

66 Figure 28 a-n: Impedance spectra for fast-fire processed samples sintered at 14 o C, for temperatures ranging from 2 o C 1 o C a) Impedance spectra taken at 2 o C b) Impedance spectra taken at 3 o C. 56

67 c) Impedance spectra taken at 4 o C d) Impedance spectra taken at 5 o C. 57

68 e) Impedance spectra taken at 55 o C f) Impedance spectra taken at 6 o C. 58

69 g) Impedance spectra taken at 65 o C h) Impedance spectra taken at 7 o C. 59

70 i) Impedance spectra taken at 75 o C j) Impedance spectra taken at 8 o C. 6

71 k) Impedance spectra taken at 85 o C l) Impedance spectra taken at 9 o C. 61

72 m) Impedance spectra taken at 95 o C n) Impedance spectra taken at 1 o C. 62

73 Figure 29 a-c: Representative SEM images of the traditional, fast-fire, and microwave samples sintered at 135 o C. 1 µm 29 a) SEM image of a traditionally sintered sample. 1 µm 29 b) SEM image of a fast-fired sintered sample. 63

74 1 µm 29 c) SEM image of a microwave sintered sample. Figure 3 a-g: Representative SEM images of the traditional, fast-fire, and microwave samples, with and without the inclusion of a dwell time, sintered at 14 o C. 1 µm 3 a) SEM image of a traditionally sintered sample without a dwell time. 64

75 1 µm 3 b) SEM image of a traditionally sintered sample with a dwell time. 1 µm 3 c) SEM image of a fast-fired sintered sample without a dwell time. 65

76 1 µm 3 d) SEM image of a fast-fire sintered sample with a dwell time. 1 µm 3 e) SEM image of a microwave sintered sample without a dwell using the microwave heating profile. 66

77 1 µm 3 f) SEM image of a microwave sintered sample without a dwell time using the microwave s heating profile. 1 µm 3 g) SEM image of a microwave sintered sample with a dwell time using the microwave s heating profile. 67

78 Figure 31 a-f: Representative SEM images of the traditional, fast-fire, and microwave samples, with and without the inclusion of a dwell time, sintered at 145 o C. 1 µm 31 a) SEM image of a traditionally sintered sample. 1 µm 31 b) SEM image of a fast-fire sintered sample without a dwell time. 68

79 1 µm 31 c) SEM image of a fast-fire sintered sample with a dwell time. 1 µm 31 d) SEM image of a microwave sintered sample without a dwell using the microwave heating profile. 69

80 1 µm 31 e) SEM image of a microwave sintered sample without a dwell time using the microwave s heating profile. 1 µm 31 f) SEM image of a microwave sintered sample with a dwell time using the microwave s heating profile. 7

81 Figure 32 a-d: Representative SEM images of the fast-fire and microwave cross sections. 1 µm 32 a) SEM image of a fast-fire sintered sample at the edge of the cross section. 1 µm 32 b) SEM image of a fast-fire sintered sample at the center of the cross section. 71

82 1 µm 32 c) SEM image of a microwave sintered sample at the edge of the cross section using the microwave heating profile. 1 µm 32 d) SEM image of a microwave sintered sample at the center of the cross section using the microwave heating profile. 72